Full text of "Journal of the Society of Motion Picture Engineers"

See other formats

From the collection of the

Prelinger
v Jjibrary

San Francisco, California
2007

JOURNAL OF THE SOCIETY OF
MOTION PICTURE ENGINEERS

Vol 47 JULY, 1946 No. 1

CONTENTS

PAGE
Report of the Subcommittee on 16-Mm Film Splices 1

A Complete Motion Picture Production Plant for Metro-
politan New York R. B. AUSTRIAN 12

Aluminum and Chromium as Gelatin Hardeners

H. L. BAUMBACH AND H. E. GAUSMAN 22

•

The Application of Pure Mathematics to the Solution
of Geneva Ratios R. W. JONES 55

A National Film Library — The Problem of Selection

J. G. BRADLEY 63

The Waller Flexible Gunnery Trainer F. WALLER 73

60th Semi Annual Technical Conference 88

Society Announcements 92

Copyrighted, 1946, by the Society of Motion Picture Engineers, Inc. Permission to republish
material from the JOURNAL must be obtained in writing from the General Office of the Society.
The Society is not responsible for statements of authors or contributors.

Indexes to the semi-annual volumes of the JOURNAL are published in the June and December
issues. The contents are also indexed in the Industrial Arts Index available in public libraries.

JOURNAL

OF THE

SOCItTY of MOTION PICTURE ENGINEERS

MOTCL PENNSYLVANIA • N£W YORKi. N-V • Tfit. P6NN. 6 O62O

HARRY SMITH, JR., EDITOR

Board of Editors
ARTHUR C. DOWNES, Chairman

JOHN I. CRABTREE ALFRED N. GOLDSMITH EDWARD W. KELLOGG

CLYDE R. KEITH ALAN M. GUNDELFINGER CHARLES W. HANDLE Y

ARTHUR C. HARDY

Officers of the Society

* President: DONALD E. HYNDMAN,

350 Madison Ave., New York 17.
*Past-President: HERBERT GRIFFIN,

133 E. Santa Anita Ave., Burbank, Calif.
^Executive Vice-President: LOREN L. RYDER,

5451 Marathon St., Hollywood 38.
** 'Engineering Vice-President: JOHN A. MAURER,

37-01 31st St., Long Island City 1, N. Y.
^Editorial Vice-President: ARTHUR C. DOWNES,

Box 6087, Cleveland 1, Ohio.
^Financial Vice-President: M. R. BOYER,

350 Fifth Ave., New York 1.

* Convention Vice-President: WILLIAM C. KUNZMANN,

Box 6087, Cleveland 1, Ohio.
*Secretary: CLYDE R. KEITH,

233 Broadway, New York 7.
*Tr*easurer: EARL I. SPONABLE,

460 West 54th St., New York 19.

Governors

*fFRANK E. CAHILL, JR., 321 West 44th St., New York 18.
**FRANK E. CARLSON, Nela Park, Cleveland 12, Ohio.
**ALAN W. COOK, Binghamton, N. Y.

*JOHN I. CRABTREE, Kodak Park, Rochester 4, N. Y.

*CHARLES R. DAILY, 5451 Marathon St., Hollywood 38.
**JOHN G. FRAYNE, 6601 Romaine St., Hollywood 38.
**PAUL J. LARSEN, 1401 Sheridan St., Washington 11, D. C.
** WESLEY C. MILLER, Culver City, Calif.

*PETER MOLE, 941 N. Sycamore Ave., Hollywood.
"jHoLLis W. MOYSE, 6656 Santa Monica Blvd., Hollywood.

*WILLIAM A. MUELLER, 4000 W. Olive Ave., Burbank, Calif.
*°A. SHAPIRO, 2836 N. Western Ave., Chicago 18, 111.

*REEVE O. STROCK, 111 Eighth Ave., New York 11.

*Term expires December 31, 1946. tChairraan, Atlantic Coast Section.
**Term expires December 31, 1947. tChairman, Pacific Coast Section.
*°Chairman, Midwest Section.

Subscription to nonmembers, $8.00 per annum; to members, $5.00 per annum, included in
their annual membership dues; single copies, $1.00. A discount on subscription or single copies
of 15 per cent is allowed to accredited agencies. Order from the Society at address above.
Published monthly at Easton, Pa., by the Society of Motion Picture Engineers, Inc.

Publication Office, 20th & Northampton Sts., Easton, Pa.

General and Editorial Office, Hotel Pennsylvania, New York 1, N. Y.

Entered as second-class matter January 15, 1930, at the Post Office at Easton,

Pa., under the Act of March 3. 1879.

JOURNAL OF THE SOCIETY OF
MOTION PICTURE ENGINEERS

Vol 47 July, 1946 No. 1

REPORT OF THE SUBCOMMITTEE ON
16-MM FILM SPLICES*

Introduction. — A splice is a little thing; and being little, it has
been given but little attention. If we are to take heed of the advice
of our Scotch friends, we must remember that "many a mickle
makes a muckle — " and pay more attention to our splices.

Before World War II, little was published upon the subject of
16-mm splices. They were discussed in the Standards Committee
of our Society, but the subject matter was then, as now, considered
rather dull and of interest to only a very small group that was faced
with splicing problems and was forced through circumstances to do
something about them. But the volume of 16 mm has grown from
a mere trickle of release prints to an imposing volume of some 400
million linear feet more or less, manufactured in the last year.
This volume is now too large to be ignored ; the problem of splicing
16 mm is now going to affect too many people with many diverse
interests.

Definition. — What is a splice, and how does a splice come into
being? John Andreas, that patient man who spent most of his odd
moments during 2 years compiling a "Glossary of Terms Dealing
with the Motion Picture Art"1 denned a splice as "Any type of ce-
ment or mechanical fastening by which two separate lengths of film
are united end-to-end so as to function as a .single piece of film when
passing through a camera, film processing machine, or projector."
The Glossary of Technical Terms2 did not define a splice but did
define splicing as "Joining the ends of film by cementing." Although
Subcommittee C of Z52, the War Committee on Photography and
Cinematography of the American Standards Association did not
write a definition of a splice for the War Standard "Nomenclature
for Motion Picture Film Used in Studios and Processing Labora-

*Presented May 10. 1946, at the Technical Conference in New York.

2 REPORT OF THE SUBCOMMITTEE Vol 47, No. 1

lories" (Z52.14-1944), it did spend some time preparing a new
standard for 16-Mm Sound Splices, Z52.20. But that is getting
ahead of the story.

Splices in Release Prints. — We must remember that splices have
a number of functions ; one of the most talked about is that of re-
pairing a torn film. If a film becomes torn in use, it is either worn
out, or it has been subjected to carelessness in handling or run on a
poor machine. Despite the proverbially poor operating condition
of 16-mm machines, film damage seems surprisingly low for the
amount of film projected if one judges by the insurance rates in force
in most circulating film libraries. Possibly repair can be considered
one of the lesser functions of a splice. Repair splices are customar-
ily made by the film user; less frequently by a circulating library.

A new print — one just out of the laboratory — starts its life with
a minimum of 2 splices — one that attaches the head leader to the
print and the other that attaches the tail leader to the print. Most
prints have at least one more splice per 400-ft roll; this one addi-
tional splice was permitted in American War Standard Z52.3.
There was much discussion about this extra splice. Those in favor
of it felt that film life was not seriously reduced thereby and that the
life disadvantage was more than offset, as the "short ends" accumu-
lated in printing would be better utilized. (Short ends accumulate
when the film to be printed is an odd length, not a multiple of 400 ft,
the unit length for film supplied as raw stock.) By the way, it is
well to remember at this point that most 16-mm films used during
the war by the Services were not run to the wear-out point.

A new print obtained from a laboratory today may contain more
than the number of splices just mentioned; poor grade prints are
likely to have many more. It should be noted that an unspliced
print will normally show an appreciably longer life than a spliced
print particularly on machines with sharp bends in the film path.
An unspliced print will avoid that distressing phenomenon known
as the splice jump which occurs when a splice passes through the
projector movement. For practical purposes, a print may be con-
sidered optional with respect to splices when sufficient film is pro-
vided on the head end and on the tail end of the picture proper to
permit replacement of the leaders some 5 or 6 times — and when the
print has no splices within the picture proper.

All through this discussion of splices in release prints, we have
assumed that the picture proper will appear as a positive ; negative-

July, 1946

REPORT OF THE SUBCOMMITTEE

type images are not customarily used in release prints and for the
purpose of this discussion, will be considered unusual.

Splices in the Original. — Let us pass over for the time being what
happens in the laboratory and consider the original film in rela-
tion to splices. For the purposes of this paper, we need not con-
sider 35-mm film. Customarily, a 35-mm splice is made in the
negative and is of such width that no portion of it appears within the
16-mm projector aperture when a reduction print is projected. We
cannot discuss 8-mm splices at this point as the 8-mm situation will
be reviewed after the 16-mm solutions are under way. In consider-
ing 16-mm original material, it is well to consider what original

0.070-INCH SPUCE 0.100-INCH SPLICE

FIG. 1. Relative 16-mm splice encroachments on picture area.

.picture material may be and what 16-mm original sound material
may be.

Although 35-mm picture originals almost invariably are nega-
tives, 16-mm picture originals are almost irivariably reversals or
direct positives. Good examples are Kodachrome, Ansco Color, and
black-and-white reversal original. It is only in special cases that
negative is used as original material.

If we examine a splice made with any present-day commercial
splicing machine we invariably find that the splice encroaches upon
the picture image appearing in the projector aperture. Fig. 1 shows
the amount of encroachment involved with splices of 2 different
widths, namely, 0.070 in. and 0.100 in. Our 16-mm splice does en-
croach with either dimension — quite a different situation from that
encountered with 35 mm. Needless to say, the diagonal splice is
located diagonally across the spliced frame in the picture.

4 REPORT OF THE SUBCOMMITTEE Vol 47, No. l

Even if we are extremely careful in making splices, either splice
appearing in the original will appear in every release print made be-
cause of the encroachment. As present-day 16-mm subjects of
commercial origin may have as many as 150 splices in a single 400-ft
roll (and it is not unusual to find 80 as a typical average), the impor-
tance of making every splice a good unobtrusive one can hardly be
overemphasized. Fortunately, splices made in original reversal and
in color reversal show up much less objectionably than like splices
made in original negative material. If 0.070-in. straight splices are
neatly and cleanly made, they will be almost invisible in the release
print particularly if the edges of the splice are carefully painted out
or "blooped" in the assembled original film. (Blooping may be the
incorrect term as we refer here to treatment oLthe picture and not
of the sound.) Needless to say, the 0.100-in. straight splice and the
0.070-in. diagonal are not capable of a neat and workmanlike result
when compared with the 0.070-in. straight splice.

The situation with regard to splicing the sound original is different
from that of the picture original. In the past, most 16-mm sound
originals were recorded as nonpush-pull negatives. We may expect
a very material increase in the number of direct sound positives in
the years to come. Most 16-mm sound originals fortunately are
not recorded simultaneously with the taking of the picture but are
scored afterward (with off-stage voice) in accordance with the timing
established by means of a "shot list" (cue sheet) made from the pic-
ture.8 With a competent staff and with suitable recording facilities,
there is little reason for more than 2 or 3 splices. One of these is
used to attach the head leader to the original and another to attach
the tail leader to the original. If more than these 2 splices are re-
quired, a sound bloop will be needed for each additional splice.

There is no point in discussing the splicing of the sound original
further; your Subcommittee needs data on practices both present
and contemplated about sound splicing in 16 mm, and also 16-mm
sound blooping. It seems difficult to consider the two separately
if we are to be logical about our work.

For the present we may say that it is customary in most cases to
use the same kind of splice for original sound as for original picture.
The procedure, however, must be recognized as an arbitrary one
since the diagonal splice has certain advantages in splicing sound
film.

Splices Made at Other Stages Intermediate Between the Original

July, 1946

REPORT OF THE SUBCOMMITTEE

i 1-1 CO CO CO CN N

• dddddd

00 00 CO CO
!§; CSI »T5 00 00 CO co'

t&issss

gdddddd

>C5«OCOO

-lit

r^*O

• 000000

1 I

or- O co co coco
iyOW<Ni-.0^

gdddddd

REPORT OF THE SUBCOMMITTEE

Vol 47, No. 1

STRAIGHT SPLICE

efe

Curved

Diag

'////Y.

§zl

l+l

r
I

onal

•-H

CURVED

SPLICE
Straight

Mm In. Equiv. Mm In. Equiv. Mm In. Equiv.
A (MAX) 1.78 0.070 1.78 0.070 1.78 0.070
B 7.62 0.300 15.24 0.600 15.24 0.600
C (MAX) 5.97 0.235 8.51 0.335 8.51 0.335
D (MAX) 3.53 0.139 8.51 0.335 8.51 0.335
E(MIN) 1.65 0.065 6.74 0.265 6.73 0.265
F (MIN) 4.09 0.161 6.74 0.265- 6.73 0.265
R 105.0 4.125

FIG. 3, Dimensions and dimensioning practice used in Z52.20-1944.

July, 1946 REPORT OF THE SUBCOMMITTEE 7

and the Release Print. — It must be recognized that splices are made
on intermediate films used in the laboratory and in editing. For the
moment, we shall not discuss such splices as their specialized natures
usually dictate how they are made. Your committee will appreciate
receiving data on the splices used tor this purpose and the reasons
governing their choice.

The Standardization History of Splices. — Having mentioned
briefly where splices are used, let us now examine what has been
done in considering the standardization of splices.

Before World War II, two standards were approved through the
ASA; one for silent film Z22. 24-1941 "16-Mm Film Splices— Nega-
tive and Positive" and one for sound film, Z22. 25-1941 similarly
titled. Both are shown in Fig. 2. In both cases the standard width
for a diagonal splice was 0.070 in. and the standard width for straight
splices was 0.100 in. regardless of whether the film is silent or sound,
or whether the film is negative or positive, and regardless of whether
the film is an original, a release print, or any other intermediate film.

When Z52 studied the splice question, the American War Stand-
ard Z52.20-1944 "Positive and Negative Splices for Processed 16-
Mm Sound Motion Picture Film" was approved and issued. Di-
mensions and dimensioning practice used are shown in Fig. 3. The
width called for in the straight splice was reduced to a maximum of
0.070 in. in all cases. Please note that this width had previously
been specified only for the diagonal splice. Please note further that
a new type of splice appeared in the Z52 standard — the curved splice.

As War Standards are not valid beyond the end of the war, the
ASA Sectional Committee on Motion Pictures Z22 required a review
of the existing War Standards. During the war, conflict between
the older Z22.25 and the newer Z52.20 was automatically reconciled
in favor of the War Standard. With the end of the war, the subject
was called for review to determine what would be desirable as a regu-
lar American Standard. Thus the subject was referred by Commit-
tee Z22 back to the SMPE. The Standards Committee of the
vSMPE set up the present Subcommittee. To get the earliest action
possible, the Subcommittee was authorized to study and recommend
16-mm splices; the purpose was to resolve the conflict between the
Z22.25 and the Z52.20 Standards.

The present Subcommittee was appointed with the author as
Chairman and film manufacturers, splicer manufacturers, and splicer
users represented. The first meeting was held January 23, 1946,

REPORT OF THE SUBCOMMITTEE

Vol 47, No. 1

CURVED SPLIC*

Diagonal

Straight

Curved

A (MAX)

In.
0.070
0.300

Mm
1.78
7.62

In.
0.070
0.600

1.78

15.24

In.
0.070
0.600
4.125

Mm
1.78
15.24
104.78

Note 1. Dimension A is maximum and a narrower width splice of adequate strength is desirable.

Note 2. The diagonal splice is symmetrical about the center of the included frameline.

Note 3. The straight splice is symmetrical about the included perforation.

Note 4. The curved splice is symmetrical about the included perforation.

Note 5. The center of radius R is on the film centerline.

FIG. 4. Dimensions and dimensioning practice for 16-mm sound splices as pro-
posed for a one-year trial.

July, 1946 REPORT OF THE SUBCOMMITTEE 9

The minutes of that meeting showed rather clearly that the splice
problem is in need of considerable study. The views of many of
those present showed points of common thought but a number of
unsolved problems were uncovered. One very significant point of
difference was the fact that with most existing splicing equipment,
the quality of the splice made depends to a very great degree on the
skill and dexterity of the operator. Thus the questions arose.
Shall we consider the amateur as a typical user to be governed by
our standard, or shall we consider only a professional specialist?
Those questions have not been completely resolved as yet. The
Subcommittee did agree that a standard was needed and that con-
tinuing work would be required to obtain a satisfactory answer.
Possibly the best solution is to eliminate this human variable en-
tirely with an automatic splicing machine.

The new proposal for silent film is a simplification of the Z22.24-
1941 Standard brought into line with the 0.070-in. maximum recom-
mendation. Fig. 2 shows dimensioning practice of Z22. 24-1941 on
the left and, although not shown, the new silent splice proposal has
the same dimensions and is presented in the same manner as the
new sound splice proposal shown in Fig. 4, except that the curved
splice is not used with silent film.

Fig. 2 shows also the dimensioning practice of Z22. 25-1941, Fig. 3
shows the Z52.20 War Standard, and Fig. 4 shows the new proposal.

This new proposal for sound film likewise represents a simplifica-
tion of the earlier standards in line with the maximum width con-
cept of the Z52.20 War Standard.

At the meeting of the Standards Committee held on February 20,
1946, the recommendations of the Subcommittee were reviewed. It
was agreed that consideration of the new proposal as a standard
would be premature because the War Standard had not been in ef-
fect sufficiently long to test its value satisfactorily. It was agreed
that—

(a) The American Standards Z22.24 and Z22.25 should be rescinded,

(b) The SMPE shall adopt the Subcommittee recommendation for a one-year
trial period before final approval,

( d) The new proposal shall be withheld from submittal to the United Nations
Standards Coordinating Committee at this time.

Present Status. — As matters now stand, the new proposal, Fig.
4, is published here for trial. 411 concerned are certain, however,

10 REPORT OF THE SUBCOMMITTEE Vol 47, No. 1

that the trial period will bring out many pertinent facts about splices
that have not been previously submitted for standards consideration.
One point under discussion is the relative desirability of the 0.070-in.
splice compared with the 0.100-in. splice. Fig. 1, as mentioned
previously, shows the relative encroachment of these splices upon
the projector aperture. As you will note in the illustration, not
even the 0.070-in. splice made under "ideal conditions" with sym-
metrical overlap in the lap joint and with that much-desired but
rarely-realized symmetrical placement of the picture image with
respect to the sprocket holes on the film will provide zero encroach-
ment upon the projector aperture area. And note, too, that the
"ideal condition" is the best, not the worst, to be expected within
present standard limits. There is plenty of room for an enterprising
designer to make a splicing machine that will provide a strong splice
that does not encroach upon the picture aperture area.

In reviewing the splice problem, let us remember that all splices
previously discussed are of the lap-joint type. Before the Society a
short time ago came a suggestion for a butt joint with scotch tape
overlay for 35-mm work prints — for the purpose of simplified edit-
ing. You will observe that your author regarded this use of splices
as one of specialized nature. Sixteen-millimeter films are customar-
ily edited in a manner different from 35-mm films; this splicing
method may not be applicable.

In Subcommittee discussion, the question was raised as to whether
emulsion position of the leading and trailing film edges should be
considered or specified. As the whole subject of emulsion position
is still open, it was considered pointless to indicate it in connection
with splices until the more general problem of emulsion position
has been considered further. The broad question of 16-mm emul-
sion position was discussed by the author in 1942.4 Despite the
absence of any mention of emulsion position, it is considered good
practice to make a lap joint with the emulsion side of one piece of
film cemented to the base side of the other piece of film to be joined.
In this manner the same side is up on both pieces of film.

The location of the leading and trailing edges of the film has not
been specified. Many projectors will run film either in the forward
direction or in the backward direction. Although some machines for
sound will run in only the forward direction, manufacturers have so
far indicated no preference.

We have suggested a symmetrical splice as the illustration of the

July, 1946 REPORT OF THE SUBCOMMITTEE 1 1

new proposal shows. Some laboratories that make UK (iriswold
nonsymmetrical negative splice in release prints do not feel that the
symmetrical splice is most desirable. Your Subcommittee is anxious
to obtain for the record the whys and wherefores that explain their
position.

The lowly splice is one of the biggest little things in motion pic-
tures. Now is the time to give it the attention it has long deserved.
Four-hundred million feet of 16-mm film a ye#r is too big an item to
be ignored even though we may choose to ignore the 2 million or
more splices that appear in that footage. We need facts upon which
to base our decisions. Many in the Society have the facts. Let us
have them available to the Subcommittee where they may be con-
sidered instead of remaining buried in unused files. And if we do not
have all the facts that we need, let us collect them. The user, the
manufacturer, and the laboratory are all waiting for the result.
Your Subcommittee is anxious to do its job of establishing the most
practicable standards for splices. To paraphrase Al Smith, " — let's
put it in the record."

WM. H. OFFENHAUSER JR.
Chairman

[Ed. Note: Several comments have already been received from industry users,
and readers are encouraged to send in their comments and criticisms direct to
W. H. Offenhauser, Jr., Columbia Broadcasting System, 485 Madison Avenue,
New York 22, N. Y.

At the end of a year's trial, another report covering the Society's specific recom-
mendations for American Standards will also be published in the JOURNAL, j

REFERENCES

1 ANDREAS, J.: "Glossary of Terms Dealing with the Motion Picture Art,"
unpublished (Nov., 1942).

2 "Glossary of Technical Terms Used in the Motion Picture Industry," Trans.
Soc. Mot. Pic. Eng., XIII, 37 (May, 1929), p. 48.

3 CLEMENGER, J. F., AND WOOD, F. C. : "Sixteen-Mm Equipment and Practice
in Commercial Film Production," /. Soc. Mot. Pic. En?.., XXXTV, 6 (June, 1940),
p. 555.

4 OFFENHAUSER, W. H., JR.: "A Review of the Question of 16-Mm Emulsion
Position," /. Soc. Mot. Pic. Eng.. XXXIX, 2 (Aug., 1942), p. 123.

A COMPLETE MOTION PICTURE PRODUCTION PLANT
FOR METROPOLITAN NEW YORK*

RALPH B. AUSTRIAN**

Summary. — A brief description of the plot, buildings, and equipment of New
York's newest motion picture studio operated by the RKO-Pathe Corporation.

For quite some time RKO-Pathe, Inc., has keenly felt the need for its
own studio located in metropolitan New York. Such a studio would
be used by it for the production of the several series of theater
shorts it is. currently making, for the increased production of com-
mercial pictures, for the production of documentary pictures, and
for the production of other specialized and varied subjects it has long
contemplated making.

During the war period the restriction on building or alterations to
existing buildings naturally precluded any action toward acquiring
either a building or a site. RKO-Pathe' s problem was further ag-
gravated by the fact that production demands on it in the above-
mentioned fields were greatly expanded. RKO-Pathe's own facilities
and such rental facilities as were from time to time available to it
were sorely taxed to keep abreast of its production schedule.

The great success the Armed Forces experienced with training films
has clearly indicated a tremendous new field for the motion picture.
Industry is now calling for films with which to train new workers,
demonstrate new products, and sell new products. Another field
which is due for an immediate and wide expansion is the production of
educational films for use primarily in institutions of learning to supple-
ment the usual pedagogical methods. Finally, the just awakening
field of television will require film producing facilities of some mag-
nitude.

All of these factors justified the decision to open and equip a
modern studio complete in every detail including a commercial -size

* Presented Oct. 17, 1945, at the Technical Conference in New York.
** President, RKO Television Corporation, New York.
12

MOTION PICTURE PLANT FOR NEW YORK

laboratory in metropolitan New York close to the heart of the city.
These conditions were hard to meet. The chances of obtaining such a
location in New York where realty values are high seemed remote
indeed. In addition to wanting the studio to be in the metropolitan
district, consideration was given to the need of future expansion,
parking facilities, and good transportation. The possibility of find-
ing an existing building available and suitable for such a studio
seemed slim, yet a thorough
survey was justified in order
to save the time required to
erect a new building which
would require at least one or
possibily ll/2 years.

After a considerable search
in the desirable areas, RKO-
Pathe was fortunate in locat-
ing a building which, with a
minimum amount of altera-
tions, exactly suited its re-
quirements, and furthermore
the building was vacant and
immediately available. It
stands on a plot of ground
large enough— 180 X 200 ft
(36,000 sq ft)— to afford room
for the necessary supplemen-
tary buildings and future ex-
tensions to the main building
(Fig. 1A and IB). It is 11
stories high, full brick, re-
inforced concrete and steel fireproof construction, with double
thick walls with air space in between. There are no windows to any
of the stages. Each stage has its own individually controlled air-
conditioning system. It is 100 per cent sprinkler equipped.

It is located in a district where zoning laws permit the operation
of a motion picture studio, the northeast corner of Park Avenue and
106th Street, New York City.

Pathe Film Industries, Inc., whose New York City laboratories are
presently located on the premises of Pathe News, 625 Madison Avenue
New York, was also satisfied with the facilities the new building

FIG. 1A.

R. B. AUSTRIAN"

Vol 47, No. 1

offered and have agreed to expand and incorporate the service
laboratory into the new studio.

Main Studio. — A brief description of each floor and floor plans of
the main studio building follow.

Basement. — Here will be located the usual heating plant, air-
conditioning compressor units, etc. All chemicals used in the process-
ing of film will be mixed here.

FIG. IB.

•
First or Street Level Floor. — Pathe Laboratories is setting up a

modern compact service laboratory on the first floor (Fig. 2). The
entire laboratory will be finished with salt-glazed tile, and the air-
conditioning equipment will be of the latest type using electrostatic
filters and automatic controls to achieve maximum cleanliness and
uniform conditions throughout the entire area. All of the equipment
will be of modern design, with most of it being designed and con-
structed by Pathe Laboratories, Inc.

Second and Third Floors, Main Studio. — This is the largest shoot-
ing stage, measuring approximately 97 X 74 X 32 ft high. This

July, 1940 MOTION PICTURE PLANT FOR f^EW YORK 1£>

shooting stage has a wooden floor. All 4 walls and ceiling will be
covered with 2 in. rock wool. This stage will be used only for speech
recording and, of course, production shooting (Figs. 3 and 4).

Fourth Floor Laboratory, Offices, Vaults, Cutting Rooms, etc. —
The entire fourth floor is being devoted to cutting rooms, office space,
and screening room for the trade (Fig. 5).

SAl.TGLX.ZrD TILt

PAT-HE INDUSTRIES INC.
FIG. 2. First floor.

Fifth and Sixth Floors. — Here are located stages C and E each of
which will be 63 ft long X 30 ft wide X 22 ft high (Figs. 6 and 7).
Between stages C and E is a room 51 ft long X 22 ft wide X 22 ft
high opening onto both stages, which will be used for the storing of
set dressing material, props, electrical equipment, etc. These stages
will also have wooden floors and will have all walls and ceilings covered
with 2-in. rock wool. These 2 studios also will only be used for speech
recording and production shooting:.

R. B. AUSTRIAN

Vol 47, No. 1

FIG. 3. Second floor.

PASS ELEV NO

FIG. 4. Third floor.

, 1946 MOTION PICTURE PLANT FOR NEW YORK 1?

Seventh and Eighth Floors.— These 2 floors (Figs. 8 and 9) house
studio A, which is for music scoring and is 63 ft X 52 X 22 ft, and
studio B which contains a theater seating 70 and this is 50 X 25 X
22 ft. It will also be used for dubbing.

Studio A will have the floor covered with linoleum, underneath
which is a layer of cork. The south and west walls will be treated

YA«D

~— y

FILM FILM FILM

cv/rr>«i m. et/rrwK, KM.

L_LJ

^ ••—

t« oc^.ce <

Souno *>*

UBCTKM

If"" —

Cwoarr

P»OJ«cno* ^__^__

«>VMO mon

•l*V»TW,.

ST^KI

" """

O»-FIC«

. _J v-

A « oonorTtoMmc

ti
t

«U>.T conjoin CUTT

LOI:

p."

r
D

' o
o

with 1-in. rock wool blanket. Then there will be applied convex
plaster panels varying in width from 25 to 45 in. X 2iy2 ft long.
These will be applied to the walls in a vertical position, care being
taken to scatter the various widths at random to prevent any stand-
ing wave characteristics or undesirable reinforcement. Every other
panel is removable for acoustic adjustments.

The east and north walls of this studio are also treated with a
layer of 1-in. rock wool blanket, and convex plaster panels measuring
from 25 to 45 in. X 2V /t ft long, but on these walls the panels are

R. B. AUSTRIAN

Vol 47, No. 1

FIG. 6. Fifth floor.

or JTUOIO t

FIG. 7. Sixth floor.

July, 1946

MOTION PICTURE PLANT FOR NEW YORK

attached in a horizontal position and are not removable. Here again
the various widths are applied at random. Adjustable extra full
drapes will be hung on the east and north walls over the plaster panels
to allow for acoustic corrections. Fifty per cent of the ceiling area
will have convex panels applied. Balance of the ceiling area will be
exposed rock wool.

Studio B is laid out as a combination theater and dubbing room.
The ceiling and walls are treated with 1-in. rock wool bats. Over

FIG. 8. Seventh floor.

these bats are convex curved plaster panels similar to those provided
in Studio A. This ceiling will have convex plaster panels covering
75 per cent of the area. The remaining 25 per cent will be exposed
rock wool. At the projection screen end of the room adjustable drapes
will be hung to provide for acoustic, adjustments. The monitor booth
for dubbing and rerecording will also be located here. The ninth to
eleventh floors are not being equipped at the present time, but they
are held in reserve and offer considerably more production space.

Sound Equipment. — RCA sound equipment is being installed
throughout. There will be 2 complete studio sound channels. The 2
recorders of these channels may be used to record 2 productions si-

R. B. AUSTRIAN

Vol 47, No. 1

multaneously, or both recorders may be fed in parallel. These 2 re-
corders are Selsyn driven. There are 6 film phonographs, 2 of which
may also be synchronously driven. This is accomplished by having
synchronous 3-phase motors as well as Selsyn motors, on the same
shaft.

All dialogue mixing will be accomplished by "tea- wagon" mixers
which may be plugged into outlets provided on all stages. When
using tea-wagons, monitoring will be accomplished by ear plug type

FIG. 9. Eighthfloor.

headphones. Monitoring in the scoring and dubbing control rooms
will be by means of 2-way standard monitor speakers supplemented if
necessary by earphones. There will be 2 projectors located on the
Upper or eighth floor level. One of these supplies projection for the
theater-dubbing stage, The other will be used to project a picture
to the scoring stage screen where both the conductor of the orchestra
and the scoring mixer can view it through a triple glass window
located between the music scoring stage and the rerecording room.
These projectors can be driven by either synchronous or Selsyn
motors, again by means of having both types of motors mounted on a
common shaft. If necessary, these projectors may l?e interlocked

July, 1946 MOTION PICTURE PLANT FOR NEW YORK 21

with each other and with other component parts of the recording
system.

On each of the 3 main dialogue shooting stages outlets will be pro-
vided for rear screen process projection. The recording console has
8 mixing positions each with rerecording compensators. Compression
and variable high-pass filters may be inserted in the voice channel.
The 8 mixer positions may be separated and 4 mixers run to each of 2
film recorders.

An interesting constructional detail is the fact that a chorus may
be recorded in studio B whereas the orchestra is in studio A . Each
will be acoustically insulated from the other. Each will record
on its own channel. Proper balance can easily be obtained by the re-
recording process. Both orchestra and singers naturally will be in
sight of each other and of the recordist. He will view them through
the triple glass window. A monitor loudspeaker will feed the neces-
sary music to the singers.

There are 2 disk recorders driven by synchronous motors. Disk
playback is available in either or both music scoring stages and re-
cording room simultaneously. The recording equipment is located
in the central plant with a patch field to all studios. Below this
"recording central" and connected by a spiral stairwell are located
the 6 film phonographs. The recording channels are extremely flexible
and will permit disk to film, film to film, film to disk, and disk to disk
recording. One portable truck-mounted recording channel will also
be available for location shooting. This plant of course will provide
RKO Television Corporation with complete eastern production
facilities.

ALUMINUM AND CHROMIUM AS GELATIN HARDENERS*

H. L. BAUMBACH AND H. E. GAUSMAN**

Summary. — The hardening action of aluminum and chromium upon gelatin is
explained on the basis of the formation of chemical compounds between the metal ionr
and gelatin protein molecules. The metal ions must possess a positive charge in ordes
that they may combine with the negatively charged carboxyl groups. The factors
which affect the degree of combination are discussed in detail for both aluminum and
chromium, the most important being the pH values of the solutions and the presence of
competing anions.

Aluminum fixing baths harden at pH values between 3.5 and 6.0, with the pH of
maximum hardening dependent upon the complexing anions that are present. Data
are presented that permit selection of the most suitable formulas that can be used under
given operating conditions.

Ion migration experiments and other tests are detailed which indicate that chromium
fixing baths lose their active hardening properties when the charge on the complex
chromium- molecules becomes zero or negative. Combination of positively charged
chromium and gelatin carboxyl groups is relatively independent of chromium con-
centration. Complex chromium molecules of zero charge are lightly held by the gela-
atin and cause hardening only during and following the drying of the gel.

Chromium fixing baths harden gelatin at pH values from 3 to 6, with the pH of
maximum hardening depending upon the sulfite-to-chromium-molal ratio and the age
of the solution. Bisulfite ions form complex ions with chromium and increase the
rate of hydrolysis of the chromium ions, thereby tending to cause loss of the hardening
properties of the solutions with age. By limiting the pH to 4.0 and using a molal
ratio of sulfite to chromium less than 2:1, the active hardening properties of a chro-
mium fixing bath are at least partially retained. Other conditions are presented under
which the hardening properties are also maintained.

Hypo and silver are retained in washed films by both aluminum and chromium if
the complex metal molecules still possess a positive cJiarge after combination with the
gelatin occurs. The pH values of the solutions and the presence of complex-forming
anions are factors which influence the formation of the desired monovalent complex
metal ions.

Introduction. — Chromium and aluminum are commonly used in
photographic fixing baths to permit safe handling of the unhardened
photographic film during and following processing. The water-
absorptive qualities of gelatin are decreased as a result of the harden-

* Presented Oct. 16, 1945, at the Technical Conference in New York.
** West Coast Laboratory, Paramount Pictures, Inc., Hollywood,

GELATIN HARDENERS 23

ing process and the emulsion is toughened to such an extent that it is
less readily scratched or torn while it is wet. After the hardened film
has been dried, it is less apt to show handling marks and it is some-
what less susceptible to scratching.

The literature is obscure and often contradictory regarding the fac-
tors which affect the hardening qualities of aluminum and chromium,
and the chemistry of the processes that are involved are but little
understood. The baths that have been recommended are primarily
designed to maintain their hardening properties during the normal
fixing life of the bath, with a minimum amount of chemical control.
The use of continuous electrolytic silver recovery and the necessity
for stopping the developing action uniformly have extended the useful
life of fixing baths considerably, by offering control over the silver
concentration and the pYL of the bath. It has become possible to
operate a fixing bath with a measure of continuous replenishment and
to maintain its important functions at constant values ; consequently,
it has become important that the factors affecting the behavior of the
hardening agents within the solution be determined.

In recent years the tendency has been toward the use of aluminum
rather than chromium as a hardener for motion picture work, because
of the greater stability of aluminum fixing baths and less need for care-
ful control. Chrome-alum fixing baths have tended to lose their
hardening characteristics with age and to precipitate an objectionable
sludge which deposits on the film. Moreover, they must be subjected
to more rigid chemical control in order to function satisfactorily.
Probably the preference for aluminum has been influenced somewhat
by the better understanding of the use of this substance as contrasted
to the rather erratic and complicated behavior of chromium, but more
information is needed on the chemistry of both aluminum and chro-
mium before their relative merits can be evaluated.

The single factor that affects the action of the hardening agents to
the greatest degree is the pH of the solution, and since, with chemical
control, it is possible to maintain the fixing bath at almost any de-
sired pH value, it is important to determine the influence of the pH
upon the hardening properties of aluminum and chromium. If the
pH is controlled in a bath, the quantity and nature of developer that
is carried into the fixing bath lose their significance in regard to indi-
cating the condition of the bath, and chemical and physical methods
must be used to determine the ingredients that should be added, and
the replenishment procedure.

24 H. L. BAUMBACH AND H. E. GAUSMAN Vol 47, No. 1

The conception that film hardening is caused by the simple precipi-
tation of basic salts of aluminum or chromium within the cellular
structure of the gelatin leaves much to be desired in terms of explain-
ing the action of these substances, since gelatin is frequently hardened
under conditions where precipitation will not take place. Moreover,
if this explanation were correct, hardening should depend primarily
upon the quantity of alkali carried by the film as it enters the fixing
bath, and hardening should be at a maximum when the fixing bath is
closest to the pH of precipitation. Actually, neither of these factors
is significant so far as the degree of hardening is concerned. It is
quite improbable that the formation of an insoluble hydroxide of
aluminum or chromium bears any more than incidental relationship
to the hardening process.

The solution chemistry of aluminum and chromium is complex, but
it is only by means of a discussion of the solution reactions of these
substances in line with the modern principles of atomic structure that
any logical explanation of their hardening properties can be made.
With the knowledge of the chemical reactions that are involved, the
hardening action of solutions of these substances can be controlled.

Aluminum Fixing Baths. — Aluminum is an amphoteric element
capable of existing in solution in both acid and basic forms. In
solutions of low pH, aluminum possesses a positive charge and enters
into reactions typical of these ions, with the formation of such com-
pounds as aluminum chloride, AlCl3-6H2O, and aluminum sulfate,
A12(SO4)3'9H2O. In solutions of high £H, aluminum is present as a
negative ion with the formation of compounds such as sodium alu-
minate, NaAlO2.

The aluminate form has been used as an addition agent to photo-
graphic developers of special properties, but aluminum is commonly
used in fixing baths in its acid form. Aluminum forms soluble com-
pounds with the ingredients that are needed in fixing baths and hence
its use as a hardener does not restrict the normal fixing action of the
bath. Its thiosulfate, bisulfite, sulfate, bromide, iodide, chloride,
acetate, and citrate are all sufficiently soluble and do not cause pre-
cipitation within the bath. However, certain basic salts of many of
the above ions or of the ortho phosphate ion are not appreciably solu-
ble, and the precipitation of these compounds limits the usefulness
of the fixing bath both by removing aluminum ions from solution,
thereby reducing the hardening action, and by acting as a source of
scum and dirt that collects on the surface of the film. Basic alumi-

July, 1946

GELATIN HARDENERS

num salts tend to form at pH values as low as :\5, so that it is advis-
able to modify the fixing bath constituents by the addition of com-
pounds that tend to form more soluble basic salts in order to permit
operation of fixing baths at higher pH values.

Many substances will act in the capacity to form more soluble
basic salts, and most of them constitute organic ions that form slightly
dissociated complex ions with aluminum. Some of the substances
that have been suggested as complexing agents are shown in Fig. 1,
which is a plot of the pH of precipitation (sludging) of basic aluminum

0 5 10 15 20 25 30

GRAMS PER LITER OF ADDED ACID

FIG. 1. The influence of concentration of various or-
ganic acids or their salts upon the pH at which an alumi-
num fixing bath sludges. Each liter of the bath contains
350 grams of hypo, 15 grams of anhydrous sodium sulfite,
and 25 grams of potassium aluminum alum, in addition to
the added complexing acid.

compounds from a typical fixing bath, against weights of complexing
agents. Thus, a fixing bath which contains no complexing agents
sludges at a pH of 4.2, the increase from the value of 3.5 being the
result of the slight complexing action of thiosulfate and bisulfite.
The pH of precipitation is not appreciably raised by further addition
of these ingredients. Of the complexing agents that are shown, it is
evident that citric acid is the most effective material, by weight, that
can be used for this purpose; addition of as little as 4 grams per liter
of this substance to the given formula prevents immediate precipita-
tion of aluminum at any acid pH. Acetic acid is much less effective
as an addition agent, and it is apparent that as much as 15 grams per
liter of the acid must be present to prevent precipitation up to a pH of

H. L. BAUMBACH AND H. E. GAUSMAN Vol 47, No. l

5.0. The addition of boric acid, as was suggested by Russell and
Crabtree,1 increases the complexing action of acetic acid but it is rela-
tively ineffective if used alone ; its effect is shown for 2 concentrations
of acetic acid, 6 grams per liter and 15 grams per liter.

When it is possible to control the pH of a fixing bath, the buffering
capacity of the bath becomes relatively unimportant and it is feasible

5.0

6.0

310 4.0

pH OF FIXING BATH

FIG. 2. The melting point values of the gelatin layer of
picture negative film that is developed (10 min in negative
developer), rinsed (15 sec), and fixed (10 min) in aluminum
baths containing various complexing acids. Separate
baths are regulated at pH values from 3 to 6. The baths
contain 350 grams per liter of hypo, 15 grams per liter of
sodium sulfite, 25 grams per liter of potassium aluminum
alum and the acid. Only enough of the complexing acid
is added to prevent sludging to pH 6 (Fig. 1), which is 3.4
grams per liter of citric acid, 9 grams per liter of lactic
acid, and 15 grams per liter of acetic acid with 7.5 grams
per liter of boric acid.

to use any substance with suitable complexing action which will form
a compound with aluminum that is soluble at the desired pH value.

The most important single factor that affects the hardening action
of aluminum is the />H of the fixing bath. Fig. 2 presents data of the
hardening action of aluminum upon the gelatin of typical negative
film in fixing baths containing various acids as complexing agents.
The quantity of the acid used was not selected on a weight basis, but
rather upon the amount that would complex the aluminum to the

July, 1946 GELATIN HARDENERS 27

same pH of precipitation ; the amounts were selected such that sludg-
ing would occur at an approximate pH of 6. As the pH increases
from a value of 3, the hardening action is independent of the nature
of the complexing ion until the pH exceeds 4.5 when citrate ion inter-
feres with the hardening action. Lactate ion does not interfere until
the pR exceeds 5.0, while, if acetic and boric acids are the complexing
agents, hardening continues to be good to the pH of precipitation.
Increasing the concentration of any of the complexing agents raises
the pH of sludging and also decreases the pH at which hardening is
impaired. In these experiments the degree of hardening was deter-
mined by measuring the melting point (solution temperature) of un-
hardened negative film as suggested by Crab tree and Hartt.2

If operation of a fixing bath at a pH of 4.0 is desired and if the al-
kali that is introduced by developer is neutralized by the addition of
any noncomplexing acid, such as sulfuric or bisulfite, equal hardening
will be produced with any of the acids tested, provided that complex-
ing of the aluminum is carried to the same extent.

The Solution Chemistry of Aluminum and Gelatin Hardening.—
When salts are dissolved in water, there should be no greater tend-
ency for solution to occur than there is in the air, if it were not for
the attractive forces possessed by the water molecules for the ions
that form the salt. The water molecules are polar particles that pos-
sess free electron pairs by means of which attachment can be made
to ions or ionic groups. The surrounding of the ions by water mole-
cules changes the identity of the ions sufficiently to permit them to
overcome the electrostatic forces which hold them within the crystal
lattice. The energy that is required to separate such a large charge
as that on Al+++ from negative ions is so great that the solubility of
this ion in water can only be accounted for on the basis of ion hydra-
tion.

Thus, it has been established that aluminum exists in water solu-
tion in the form of a hydrated ion where the number of associated
water molecules is at least equal to the maximum coordination num-
ber of aluminum or, diagrammatically, in the form Al(H2O)e+++. At
least 6 water molecules surround each aluminum ion in a definite di-
rected manner, and the molecules are said to be coordinated through
the sharing of electron pairs between the oxygen atoms and the vacant
energy levels of the aluminum atom. The strong positive charge on
the central aluminum atom tends to repel protons from the coordi-
nated water molecules and the protons are in turn coordinated with

28 H. L. BAUMBACH AND H. E. GAUSMAN Vol 47, No. l

other water molecules; hence, the hydrated aluminum ion possesses
acid properties and becomes reduced in net charge upon neutraliza-
tion.
The type of reaction that is involved is diagrammed in reaction (1).

A1(H20)6+++ H- H2O = A1(H2O)5(OH)++ + H3O+. (1)

The net charge on the aluminum ion is reduced by one and the hy-
dronium ion which is formed accounts for the acid properties of the
aluminum.

Aluminum ion first shows its acid properties at a pU. of 3 and, in
all, functions as a tribasic acid, since each mole of aluminum requires
3 moles of sodium hydroxide for neutralization. As the £H of the
solution increases from 3, additional release of protons from coordi-
nated water molecules occurs, leading to the reactions shown in (2) and
(5).

A1(H2O)5(OH)++ + H2O = A1(H2O)4(OH)2 + -f H3O+. (2)

A1(H2O)4(OH)2 + + H20 = A1(H20)3(OH)3 + H8O+. (3)

After the addition of 3 equivalents of alkali, the reactions have
practically gone to completion with the formation of uncharged
Al(HaO)3(OH)3, which is insoluble and precipitates from solution.

The tit ration curve for aluminum with alkali (Fig. 3) has no breaks
in it, which indicates that all 3 reactions are in rapid equilibrium with
each other and that the ionization constants of the 3 acid ions are
close together. The tendency is toward the formation of a mixture
of the ions, with the pH determining the ion that is present in the
greatest concentration. While the first evidence of acid properties
(reaction (1)) occurs at a £H of 3, the first precipitation of a portion
of the A1(H2O)3(OH)3 is observed at a pH of 3.5 and occurs before the
first equivalent of alkali has been added.

With aluminum, chromium, zinc, and other amphoteric elements,
this type of reaction progresses with the formation of soluble nega-
tively charged ions called aluminates, chromiates, zincates, etc.,
in a manner diagrammed by reaction (4).

A1(H2O)3(OH)3 + H2O = A1(H20)2(OH)4- + H3O+. (4}

The net positive charge on the aluminum ion decreases as the pH in-
creases, becoming negative in alkaline solutions.

The nature of the aluminum ion is altered by the loss of protons or,
in other words, by substitution of hydroxyl groups for water mole-
cules. Greater modification occurs when certain other groups become

July, 1946

GELATIN HARDENERS

substituted; the only necessary qualification for coordination being
that the groups must possess at least one free pair of electrons that
can assume energy levels in the aluminum atom. If the pH is such
that free acetate ions, for example, can exist in solution with hy-
drated aluminum ions, the acetate ions will tend to replace coordi-
nated groups as the concentration of acetate ions increases, giving the
new aluminum ion different properties (*. e., greater solubility).
Hence, such ions as Al(H2O)4(OH)(Ac)+ or A1(H2O)4(OH)(HSO8) +
exist in a fixing bath.

0 123456

MLS. 10 N NoOH

FIG. 3. The titration of 12.5 grams of potassium
aluminum alum in 500 milliliters of water with approxi-
mately 10 N sodium hydroxide. The 3 acid equivalents of
aluminum are nearly equal and are active between the pH
values of 3.5 and 7.

Citrate ions are even more striking in their ability to coordinate
with aluminum, as evidenced by the complete solubility of aluminum
at all pH values when the molal ratio of citrate to aluminum exceeds
1:3 (Fig. 1). Citric acid is a tribasic acid, and one mole of the acid
can accommodate 3 moles of aluminum.

Tartaric acid is a dibasic acid and, similarly, forms a soluble coor-
dination complex with aluminum ion when present in a molal ratio
greater than 1:2. The evidence is that coordination occurs with the
aluminum ion through the supplying of an electron pair by the oxy-
gen atom of the carboxyl group.

30 H. L. BAUMBACH AND H. E. GAUSMAN Vol 47, No. 1

Fresh acid solutions of citric acid and aluminum in a molal ratio of
1 : 3 that are quickly made alkaline show complete solubility at all pH
values. However, if the solution is allowed to stand at a pH near 7,
slow precipitation of some of the aluminum takes place as hydroxyl
groups replace carboxyl groups; the solution slowly becomes more
acid. In this respect the behavior of aluminum closely parallels that
of chromium ^hich will be discussed later.

The nature of the organic anion determines the completeness with
which hydroxyl groups can be replaced and coordination can occur,
so that, while citrates and tartrates coordinate readily, relatively large
concentrations of ions such as formate, acetate and propionate must
be present to convert aluminum ions completely and prevent precipi-
tation of aluminum at any pH.

The protein molecules that comprise gelatin contain carboxyl
groups which possess free electron pairs as the acid groups dissociate,
and it is reasonable that such groups can coordinate with aluminum
ions under favorable conditions. Since the carboxyl groups of the
gelatin are fixed in position, the attached aluminum ions may migrate
only within the gel.

The requirements for favorable coordination of aluminum and gela-
tin carboxyl groups involve, first, the absence of other groups within
the solution with greater coordination tendencies, which explains the
loss of hardening properties which occurs when citrate or tartrate
ions are present in the solution. If other carboxyl groups are present
(i. e., acetate), they should be mild enough in their action and low
enough in concentration that the gelatin carboxyl groups can compete
with them for the aluminum.

The second requirement for coordination involves the presence of
the free electron pairs on the oxygen atoms of the gelatin carboxyl
groups. The acid properties of gelatin are mild, which is to say that
the ionization constants are low. The isoelectric point of photo-
graphic gelatin occurs at a pH of 4.9, which is the pH at which the
acid and basic properties are equal in magnitude ; it is not the mini-
mum £H at which carboxyl groups can ionize, and some gelatin car-
boxyl groups can probably exist in ionic form at pm values below 3.
As the gelatin is made more alkaline from a pH of 3, greater numbers
of carboxyl groups will become ionized with practically complete ioni-
zation occurring at pH values above 7 ; thus the conditions for coor-
dinating carboxyl groups become increasingly favorable as the pH
increases from 3 to 7.

July, 1946 GELATIN HARDENERS 31

The third requirement for coordination involves the nature of the
charge on the complex aluminum ion. If the aluminum ion is nega-
tively charged (aluminate), it will tend to be repelled by the similar
charge of the gelatin carboxyl ion and coordination cannot be ex-
pected to occur. Similarly, a net charge on the aluminum ion of
zero will permit no more than accidental coordination. In order for
aluminum ions to be attracted to the negatively charged gelatin ions
it is necessary that the aluminum be positive in charge, which can
only be the case at pH values somewhat less than 7. The net result
of the second and third requirements is the type of hardening curve
with respect to pH that appears for acetic and boric acids in Fig. 2.
Evidence of the first requirement not being met is shown by the de-
crease in hardening that occurs at the higher pH values in solutions
containing citrate and lactate ions.

As the pH decreases, the net positive charge on the aluminum ion in-
creases, and if coordination does occur at pH values as low as 4, for
example, the net charge on the coordinated aluminum ion will still be
positive ajid the complex molecule will still be free to accept other
negatively charged ions such as acetate, hydroxyl or silver thiosulfate
groups, the last of which is of great importance in the retention of
hypo and silver during washing and will be discussed later.

When gelatin becomes hardened, it is characterized by a lower de-
gree of water absorption at a given temperature, or, to express the
same thing in a different way, there is an increase in the temperature
at which gelatin absorbs sufficient water to cause disruption of the
attractive forces that permit an oriented structure to the gel. Ordi-
nary unhardened gelatin swells when placed in water, to an extent that
depends primarily upon the pH of the solution. The swelling tend-
ency increases at pH values on both sides of the isoelectric point and
follows, in a general way, the increased ionization. Since water mole-
cules are polar, they can readily become attached only to other polar
groups, hence the swelling of gelatin arises largely out of the presence
of ionic groups on the protein molecules; when these groups are made
neutral by coordination with aluminum ions, the tendency for the
gelatin to absorb water is reduced.

The problem of preventing the excessive swelling of gelatin during
photographic processing is frequently approached in another way;
the addition of soluble ionic salts, such as sodium sulfate, to the photo-
graphic solutions provides competition for the water molecules that
limits the quantity of water that the ionic protein groups may retain.

32 H. L. BAUMBACH AND H. E. GAUSMAN Vol 47, No. l

This type of hardening is described as "temporary," since it disap-
pears when the film is subsequently washed in water. The salts that
are formed between aluminum, chromium, iron, and other metal ions
and the protein molecules are insoluble in water and such hardening
is called "permanent."

Practical Operation of an Aluminum Fixing Bath. — While potas-
sium aluminum alum is usually suggested as the source of aluminum
for fixing baths, some economy can be mads- by using equivalent
amounts of other aluminum salts, such as hydrated aluminum chlo-
ride or aluminum sulf ate, since moderate quantities of ions like potas-
sium, chloride or sulf ate are without harmful effect upon the action of
the bath.

If the £H of the fixing bath is to be held constant, and such must
be the case in a motion picture laboratory if the developing action is
to be stopped uniformly, choice of a fixing bath formula will depend
somewhat upon the pH value that is selected for plant operation.
From the data of Fig. 1 and for an aluminum alum concentration of
25 grams per liter, it is evident that 15 grams per liter of acetic acid
(or its equivalent of sodium acetate) will prevent sludging to a pH
of 5.0. Since it is not advisable to operate within less than 0.5 ^>H
unit of the precipitation point, because of the danger of scum forma-
tion in the wash water, the limiting pH at which this formula should
be operated is probably 4.5. If the pH of operation is as low as 4.0,
as little as 5 grams per liter of acetic acid should be sufficient. Crab-
tree, Eaton, and Muehler3 have demonstrated the advantages of oper-
ating an aluminum fixing bath at pH values as great as 5.0 to mini-
mize hypo and silver retention; if this is done, it is advisable to add
enough boric acid to prevent sludging to a pH of 6.

If complexing with boric acid is not carried beyond a pH of 6, such
a formula will harden well at all usable pH values, but it represents
a needless use of chemicals for operation at the lower pH values.
The boric acid-acetic acid fixing bath hardens gelatin poorly at a pH
of 3 and quite well at p~H. values between 4 and 6 (Fig. 2), with maxi-
mum hardening occurring at a £H of about 5.

The preparation of aluminum fixing baths is quite simple and the
chemicals may be added dry to the solution as it is prepared, provided
the agitation is adequate and the ingredients are added in the order
of hypo, sodium sulfite, acetic acid (boric acid), and alum. The
temperature of the water should be high enough at the start so that
the addition of the hypo will result in a normal temperature of use.

July, 1946 GELATIN HARDENERS 33

The concentration of bisulfite ion will become lower with use, owing
to loss of sulfur dioxide, and some means of analysis and replenish-
ment is desired, although the concentration of this substance is not
critical in an aluminum bath. It is only necessary to maintain the
sulfite concentration at a level high enough to prevent sulfurization
under the conditions of use. Replenishment of the bisulfite can
equally well be made by addition of sodium bisulfite or sodium sulfite
and sulfuric acid. Probably regulation of the pH should not be at-
tempted with acetic acid, especially if there is excessive alkali carry-
over, because too much acetate ion is harmful to the hardening prop-
erties of the bath, and because relatively large quantities are needed
to produce a small change in />H. Sulfate ion is not harmful in the
fixing bath and sulfuric acid can be safely added in its concentrated
form to the fixing bath, provided agitation is good, the acid is added
slowly, and the sulfite concentration is maintained. Plants using
electrolytic silver recovery systems usually need little acid to neu-
tralize developer alkali, since acid is liberated at the anodes while
silver is formed at the cathodes.

Within their limitations, other acids may be used in aluminum fix-
ing baths with satisfactory results. Formic acid, citric acid, and lac-
tic acid used in the proper quantities as outlined above, and doubtless
many other acids not investigated here, can be made to function where
control over pH is practiced, but there are few advantages to be
gained over the use of the economical acetic acid. Citric acid is of
particular interest in connection with its influence upon hypo and
silver retention and will be mentioned later.

The hardening properties of the aluminum-acetic acid formula do
not change with age, as is usually the case with chromium; sulfur is
formed in the bath if the pH is too low or the sulfite concentration is
too low. If the sulfur is filtered off and the condition corrected, use
of the bath may be continued. Basic aluminum salts precipitate if
complexing is not adequate or the pH is too high, but these substances
will usually redissolve if the pJi is lowered. Aluminum fixing baths
are easy to prepare and to maintain; they harden well under the
proper conditions and are open only to the criticism that they tend
to prevent complete removal of hypo and silver during the washing
process.

Chromium Fixing Baths. — Chrome-alum fixing baths have been
suggested where unusual hardening of the gelatin is desired, such as
might be needed under tropical conditions. Since the hardening

H. L. BAUMBACH AND H. E. GAUSMAN Vol 47, No. 1

and sludging properties of chrome-alum fixing baths become inferior
with age, it has been suggested that the solutions should be freshly
prepared and discarded when they lose their hardening properties.
Such statements as "chrome-alum fixing baths tend to harden the
film excessively or not at all" appear in the literature and have tended
to discourage the use of chromium for this purpose. Nevertheless,

y so

sro

g 60

SOq/l

0 2 4 6 8 10 12 14 16

AGE OF THE SOLUTION IN DAYS

FIG. 4. The melting point values of the gelatin layer of pic-
ture negative film that is developed (10 min in negative de-
veloper), rinsed (15 sec), and fixed (10 min) in chromium fixing
baths containing 350 grams per liter of hypo, 15 grams per
liter of sodium sulfite, and various concentrations of chrome
alum. The baths are all regulated to pH 4.0 by means of so-
dium hydroxide or sulfuric acid. Loss of hardening properties
with age occurs, unless the concentration of chrome alum is
high.

chrome-alum fixing baths are used successfully as continuously re-
plenished solutions, producing a moderate degree of hardening and
satisfactory general operation.

Like aluminum, chromium does not form insoluble compounds with
fixing bath ingredients. Unlike aluminum, chromium does not im-
mediately reach equilibrium with the components of the solutions in
which it is placed, and it is quite possible to prepare such solutions in
a manner which will require days or even months for an equilibrium
to be reached. The typical chrome-alum fixing bath is representative

July, 1946

GELATIN HARDENERS

of just such a condition, and the loss of hardening properties with age
represents the reaching of an equilibrium between the added chro-
mium and the ingredients of the fixing bath. Because of the slow
reactions involved with chromium, changes that are made within the
bath are usually not immediately effective or may not persist on
standing.

While the organic carboxylic acids such as acetic and citric acids
can be added to chrome-alum fixing baths in moderate concentrations
without immediate change in
the function of the bath, the
loss of hardening proceeds at a
rapid rate to reach the new equi-
librium condition. Consequently,
the use of any organic acid is
not advisable in a chrome-alum
fixing bath.

Sulfuric acid is usually recom-
mended for use with chrome-
alum fixing baths because the
baths prepared with this acid
have the greatest stability. It
is a strong acid that has little
of the desired buffering action
characterized by the weak or-
ganic acids, and consequently it
must be added repeatedly • to
the bath to neutralize the alkali
that is introduced as developer.
Sodium sulfite or bisulfite is
necessary to prevent sulfuriza-
tion of the fixing bath and pro-
vides most of what little buffer-
ing capacity is possessed by the bath. However, as will be shown
later, much of the bisulfite is not free to act either as a buffer or to
prevent sulfurization.

In a study of the effect of chromium concentration upon film hard-
ening, a series of chrome-alum fixing baths was prepared and con-
trolled to a pH of 4.0. The degree of gelatin hardening was deter-
mined for each of the samples over a period of time, producing the
results that are shown in Fig. 4. It is evident that, as the concentra-

t 7

0 5 10 15

GRAMS PER LITER OF SODIUM SULFITE

FIG. 5. The effect of concentration
of sodium sulfite upon the pH of sludg-
ing of a fresh chrome-alum solution
(25 grams per liter). Precipitation is
prevented in acid solution if the molal
ratio of sulfite to chromium exceeds
2:1.

H. L. BAUMBACH AND H. E. GAUSMAN Vol 47, No. l

tion of chromium is increased in this solution, the rate at which the
solution loses its hardening properties decreases until, at a concen-
tration of 50 grams per liter, no loss of hardening occurred in 17 days.
Gelatin solution temperatures are not shown in excess of 100 C, so
that this part of the curve is missing. However, the lack of aging
cannot be accounted for on the basis of concentration alone, since a
drop in the hardening curve for the 50 grams per liter concentration

i'°

s.o

6.0

3£ 4JO

pH OT FIXING BATH

FIG. 6. The melting point values of the gelatin layer
of picture negative film that is developed (10 min in nega-
tive developer), rinsed (15 sec), and fixed (10 min) in
chromium fixing baths containing 350 grams per liter of
hypo, 17.5 grams per liter of sodium sulfite, and 25 grams
per liter of chrome alum. Separate baths are controlled to
pH values from 3 to 6. The molal ratio of sulfite to chro-
mium is 3:1.

would have been expected in a few days. In addition to maintaining
its hardening properties, the sample with 50 grams of chrome alum
showed sulfurization after a week of storage at a pH of 4.0, in spite
of the high concentration of sulfite in the solution.

The influence of the concentration of chromium upon the loss of
hardening properties with age and upon sulfurization suggests that
the relative balance between chromium and bisulfite is important.
Accordingly, the effect of sulfite concentration upon the £H of pre-
cipitation of chromium was investigated for a solution of thiosulfate
and chrome alum; the results of these experiments are shown in Fig.

July, 1946

GELATIN HARDENERS

5. Here it is apparent that as the concentration of sulfite is increased
the solubility of the basic chromium salt also increases, and when the
molal ratio of sulfite to chromium exceeds 2:1, no precipitation of
chromium occurs at any acid pH.

Since the molal ratio of sulfite to chromium is thus of importance in
a chromium fixing bath, the degree of hardening as a function of pH
was studied at various molal ratios from 1 : 1 to 3 : 1 with the results

100

3.0 4.0 5.0 6.0

pH OF FIXING BATH

FIG. 7. The melting point values of the gelatin layer of picture
negative film that is developed (10 min in negative developer),
rinsed (15 sec), and fixed (10 min) in chromium fixing baths con-
taining 350 grams per liter of hypo, 9 grams per liter of sodium sul-
fite, and 25 grams per liter of chrome alum. Separate baths are con-
trolled to pH values from 3 to 6. The molal ratio of sulfite to chro-
mium is 1.5:1.

shown in Figs. 6 to 8. The loss of hardening properties with age and
the pH values at which hardening occurs are very definitely related
to the molal ratio of sulfite to chromium. The chrome-alum fixing
bath shown in Fig. 6 had a sulfite- to-chromium-molal ratio of 3:1
and is representative of the chrome-alum baths that have been rec-
ommended for use. The fresh bath shows moderately good harden-
ing at a pH of 3.0 with decreasing hardening as the pH increases,
practically no hardening being produced at a pH as high as 6. Upon
standing for a period of time, the bath loses its hardening properties

H. L. BAUMBACH AND H. E. GAUSMAN Vol 47, No. 1

more rapidly as the pU. is increased, the greatest stability being ob-
served at a £H of 3.0. This bath does not sulfurize or sludge upon
standing, but evolves considerable sulfur dioxide.

When the molal ratio of sulfite to chromium is reduced to 3:2, and
the same series of tests run (Fig. 7), it is observed that the fresh bath
hardens somewhat better and does not show a drop in hardening
properties until the pH is in excess of 5. As the solutions aged, loss
of hardening properties occurred at all pH values except 4.0 and was

100

3.0 4jO

pH OF FIXING BATH

5.0

6.0

FIG. 8. The melting point values of the gelatin layer
of picture negative film that is developed (10 min),
rinsed (15 sec), and fixed (10 min) in chromium fixing baths
containing 350 grams per liter of hypo, 6 grams per liter
of sodium sulfite, and 25 grams per liter of chrome alum.
Separate baths are controlled to pH values from 3 to 6.
The molal ratio of sulfite to chromium is 1:1.

most rapid at higher pH values. The bath at pH 3.0 sulfurized and
that at £H 6.0 precipitated chromium in a few hours, but those at £H
values of 4.0 and 5.0 remained free of precipitate for the duration of
the tests.

A fresh chrome-alum bath with the ratio of sulfite to chromium of
1 : 1 hardens gelatin well at all pH values between 3 and 6 (Fig. 8)
and loss of hardening properties occurs rapidly at pTL values over 5.
Sulfurization occurs extremely rapidly at a pH of 3.0 and in a few
hours at pH 4.0. Chromium precipitates in a few hours from the

July, 1946

GELATIN HARDENERS

bath at pH 6.0 and only the bath at pH 5.0 remained clear for the
duration of the tests.

This series of tests shows definitely that the molal ratio of sulfite
to chromium is of the greatest importance in the satisfactory opera-
tion of a chrome-alum fixing bath. It is unfortunate that chromium
and sulfite form complex compounds, since the necessity of adding
sulfite to prevent sulfurization must be weighed against the resulting
loss in hardening properties.

100° C

01 234 5678

MLS. ON NoOH

FIG. 9. Titration of 12.5 grams of chrome alum in 500
milliliters of water with 10 N sodium hydroxide. The full
acid properties of chromium are not immediately evident
at room temperature and there is greater separation be-
tween the 3 acid equivalents of chromium than there is for
aluminum.

The aged chrome-alum fixing bath with a sulfite-to-chromium ratio
of 3 : 1 shows a reduction in the degree of hardening with increased
washing after development, while with the bath of 1 : 1 ratio and the
aluminum baths the amount of washing after development has no
effect upon the degree of hardening.

Chromium is a colored ion and when combination with gelatin oc-
curs, the film is stained slightly, by an amount which is proportional
to the degree of hardening.

The Solution Chemistry of Chromium and Gelatin Hardening.—
Like aluminum, trivalent chromium exists in water solution as a

40 H. L. BAUMBACH AND H. E. GAUSMAN Vol 47, No. 1

coordinated ion of the type Cr(H2O)6+++, which has properties that
are very similar to those of the corresponding aluminum ion. Thus,
the hydrated ion has marked acid properties and it enters into a series
of reactions that is identical with reactions (1) to (4) that were dis-
cussed under aluminum. Somewhat higher pH values are needed,
however, to form the corresponding chromiate ion.

Perhaps the most striking difference between the behavior of tri-
valent chromium and that of aluminum lies in the comparison of the
reaction rates of the 2 substances. The coordination reactions of
aluminum are practically instantaneous at room temperature, while
those of chromium are so slow that days or even months may be re-
quired for the chromium ions to reach equilibrium in the solutions.
The type of sluggish reaction that is typical with chromium is illus-
trated by the simple, rapid titration of a chrome-alum solution with
standard sodium hydroxide (Fig. 9). In this titration at room tem-
• perature, an equilibrium condition is never permitted to occur and
the complete acid properties of the chromium are not evident. Dur-
ing the course of such a titration, the tendency is always present for
the slow loss of protons from the coordinated water molecules, with
the resulting drift to lower pH values. If the titration is performed
stepwise, with 24-hr lapses after every addition of sodium hydroxide,
or if the titration is performed at high temperature (80-100 C), there
is opportunity for the equilibrium condition to be approached and
the acid properties of the chromium are much more pronounced, with
a break in the titration curve appearing at the first equivalence point.
(The hot titration shown in Fig. 9 was performed by adding the hy-
droxide to the chrome alum solution, which was heated to 100 C.
Small samples of the solution were cooled to 25 deg for pH measure-
ment and returned to the bulk solution.) It is apparent that the
equilibrium acid properties of chromium are greater than those of
aluminum and that the first step in the hydrolysis is especially strong.
The conductimetric measurements of Kuntzel, Riess, and Konigfeld4
also indicate this spread in ionization constants. Even quite acid
solutions of chromium (pH 3-4) will thus contain the hydrolyzed
chromium ion, when an equilibrium has been reached.

The sluggish nature of chromium in reaching an equilibrium with
its solutions greatly complicates the study of the properties of such
solutions, since they undergo a change with time and the end products
are usually quite different from the starting materials. The fact that
chromium fixing baths tend to lose their hardening properties with

July, 1946 GELATIN HARDENERS 41

age is suggestive that the solutions are not initially compounded to
form a satisfactory equilibrium condition.

Chromium differs also from aluminum in its much greater tendency
to form coordination complexes with the simple anions, such as sul-
fate, sulfite, and the halides. Definite chromium chloride complexes
exist, for example, where various amounts of the chloride are directly
coordinated to the chromium and cannot be precipitated with silver
nitrate. Thus chromium chloride can be prepared in forms ranging
from Na3(CrCl6) to Cr(H2O)6Cl3, where the charge on the complex
chromium ion ranges from —3 to +3. Even a water solution of
chrome alum may contain a part of the sulfate coordinated with the
chromium that cannot be precipitated with barium ion. Such coor-
dination complex ions are relatively common with chromium and
they may persist for some time when added to solutions that are un-
favorable for their formation. The coordination chemistry of
chromium has been discussed in considerable detail by Friedman.5

Chrome alum is a double sulfate of chromium and potassium ; the
violet colored crystals dissolve in water to give a violet solution that is
colored by the Cr(H2O)6+++ ion. Upon standing, the acid properties
become increasingly evident as the green Cr(H2O)5(OH)++ ion forms
and the solution becomes more gray in color. The equilibrium con-
centrations of the 2 ions are influenced by the pH and the tempera-
ture of the solution. If the violet solution is heated, it becomes
green ; upon cooling, as long as a month may be required for a return
to the original equilibrium condition. Hence, a solution of chrome
alum that has stood for some time will contain a considerable variety
of ions, including Cr(H2O)6+++, Cr(H2O)6(OH)++, Cr(H2O)6(SO4) +,
H3O+, K+ and SO4~.

It was pointed out in the previous section that bisulfite forms a co-
ordination complex with chromium. A chrome-alum fixing bath
sulfurized rapidly unless the molal ratio of sulfite to chromium ex-
ceeded 1:1, while the presence of a sulfite- to-chromium ratio of 2:1
conveyed soluble properties upon the chromium ion at all acid £H
values. Thus, coordination of bisulfite by the chromium removes
the bisulfite ion from solution, and it is no longer free to act as an
inhibitor of sulfurization and cannot be detected by the aldehyde
procedure for the analysis of fixing baths. Since bisulfite is a nega-
tively charged ion, coordination with the positively charged chromium
ion will result in a reduction of the net positive valency of the complex
ion. It was previously shown that the presence of excess sulfite in a

H. L. BAUMBACH AND H. E. GAUSMAN Vol 47, No. 1

fixing bath caused more rapid deterioration of the hardening proper-
ties of the bath, but, except for the expected reduction in valency of
the chromium, it could not definitely be stated in which forms the
chromium would harden satisfactorily or what happened to the ionic
form when the bath had lost its hardening properties.

In order to establish the ionic forms of chromium that are responsi-
ble for the hardening action, it is necessary to know the number of

0 20 40

AGE OF THE SOLUTION IN HOURS

100

FIG. 10. The influence of sulfite concentration upon the rate of
coordination of hydroxide with chromium at />H 4.0. The sulfite
concentration is varied from 0:1 to 6:1 molal ratio in solutions of
25 grams per liter of chrome alum that were adjusted to pH 4.0 be-
fore the addition of the chromium and kept at pU. 4.0 by frequent
addition of measured sodium hydroxide. Two hydroxyl groups are
associated with each chromium atom when equilibrium is reached at
pH 4.0. The rate at which equilibrium is attained is directly pro-
portional to the sulfite concentration.

hydroxyl groups associated with each chromium ion and the net
charge on the complex ion. Consequently, various chromium solu-
tions were prepared and their properties studied as the solutions aged.
One rather simple way in which the number of hydroxyl groups as-
sociated with each chromium atom can be determined, at any time
interval, is to measure the amount of standard sodium hydroxide that
is required to keep the pH of the solution at a constant value. The
results of a series of such determinations are plotted in Fig. 10 as a
function of time. A pure solution of chrome alum undergoes hydrol-
ysis rather slowly as indicated by the curve, some 40 hr being neces-

July,

GELATIN H ARDENT kS

sary for the liberation of one equivalent of acid at a pH of 4.0 and a
temperature of 25 C. The series of aging curves at various sulfite-
to-chromium-molal ratios reveals the great effect of sulfite concentra-
tion upon the rate of hydrolysis of the chromium. Sulfite does not
appear to alter the equilibrium number of hydroxyl groups on each
chromium at a pH of 4.0, but it very definitely affects the rate at
which the equilibrium is reached. Thus, in the solution containing
a sulfite-to-chromium-molal ratio of (,> : 1 , a degree of hydrolysis is ob-

+ 90 VOLTS

CATHODE
ELECTROLYTE

ANODE
ELECTROLYTE

FIG. 11. Apparatus for determining the ionic charge on
the chromium complex molecules and the effect of the
charge upon gelatin hardening. Chromium ions with a
positive charge harden gelatin irreversibly, while chro-
mium with a zero or negative charge produces a mild
hardening that is reversed by washing.

tained in a few minutes that requires a week for the plain chrome-
alum solution. A similar set of curves, determined at pH 5.0 (not
shown), reveals the same relationship, with an equilibrium value
involving a greater number of hydroxyl groups.

The information that is obtained from the foregoing experiments
suggests the cause for the loss of hardening properties of chrome-alum
baths, and shows the influence of sulfite concentration upon the rate
of loss but it alone is not enough to indicate the reactions which occur.
In order to obtain additional information upon the nature of the hard-
ening process, these same solutions were subjected, at definite time

44 H. L. BAUMBACH AND H. E. GAUSMAN Vol 47, No. l

intervals, to qualitative ion transfer experiments and motion picture
film hardening tests.

When a potential difference exists between electrodes that are im-
mersed in a solution, ions are attracted to the electrodes of opposite
charge and the migration of the ions within the solution conducts the
current. Since the ions of chromium are highly colored, migration of
the complex ions can be determined by visual examination of their
travel into the salt bridges of the apparatus shown in Fig. 11. It has
been demonstrated that a gelatin gel offers practically no resistance to
ion migration, unless the ions form chemical compounds with the gela-
tin. Hence the gelatin plugs in the solution ends of the salt bridges
provide free ion migration from the solution without mechanical mix-
ing and, in addition, enable observation of any chemical reaction or
physical hardening. The salt bridges contain a solution of sodium
sulfate of the same molality as the chrome-alum test solutions and
are adjusted to the same pH with sulfuric acid. The results of some
of these tests are presented in Table 1 and are so important to the
interpretation of chromium hardening that they will be discussed in
some detail.

A freshly prepared solution of chrome alum contains most of the
chromium ions in the trivalent form, Cr(H2O)6+++, which is violet
in color. When the potential is applied across the gelatin plugs, the
violet ions enter only the cathode compartment, indicating that the
chromium carries only a positive charge. With continued electroly-
sis, most of the chromium ions quickly travel the length of the plug,
staining the gel as they travel and gathering in greatest concentration
at the negative end of the plug; none of the chromium leaves the gel.
This experiment indicates that the attraction of gelatin for Cr-
(H2O)6+++ ions is sufficient to overcome the potential gradient within
the solution and that chemical combination must occur. The fact
that the violet ions tend to migrate to the negative end of the gel
shows that the positive valency of the chromium has not been satis-
fied by the combination with the gelatin. The gelatin layer that con-
tains the chromium is completely insoluble in boiling water, while the
corresponding anode section is unhardened.

The chrome-alum solution aged at pU 4.0, which had hydrolyzed
to the ion Cr(H2O)6(OH)++, was tested for ion migration. Chro-
mium entered the cathode compartment only, and in this case the
cathode gel, after an hour of electrolysis, had become stained violet
in color near the positive end of the gel and green in color at the nega-

July, 1946

GELATIN HARDENERS

(/I

n
c
D

•H
1

•g

•3 S

•S

*S w

• •.

• •*

• A

• •.

* •>

rt c

, ^

a m

a
o

-i_>

§1
*i

'S
u
•d

.s
>,

~ -S ^
bo ^6 bo

C be

f [H ^J

•^ t-i

^ bo

^a
S

ing on dr

§ g

•^

II!

t/2 *§

c-» J3

S •§ _

•a

§°

i§

sis

1 2

J5 .§ bo

"§

x "E

tn J:

4J f p-l

§ i

+ „

^-v <^v

ffi

o 2.

v*— -^ «4

•si

£^

o o

t/3 CO

o o

** •«

K W

«
<

v_^

o o

dfi

*~
c

c
o

ts
O

d •

rt
O

ffi

-0.

-t

o o

^ o

00 0

N **

g*i<0

(M i

1C 00

3 N i

iO CO CO

CO CO
l_

^^22

uS*'

S x

! s

iCj

5 S

^ j.

tf) *^ O o

fii'ii

Ti ^ aj

< 0 1

3^o*S

< o 8 -j

i jEJ

5 <J

o °£ "g

l§8

Z;|K
•g < o

46 H. L. BAUMBACH AND H. E. GAUSMAN Vol 47, No. 1

tive end. The green chromium ions retained a net positive charge
within the gel, while the violet ions did not and were deposited as
they entered the gel and combined with the gelatin groups. The
green form is the ion Cr(H2O)6(OH) ++, while the violet ion is probably
Cr(H20)5(S04)+.

Other chromium solutions were tested in this apparatus, with the
results that are shown in the table. It was found that, whenever
any positively charged chromium ion entered the cathode gel and
stained it, the gel was hardened so that it was insoluble in boiling
water. The solutions that contained chromium with a net zero
charge gave no ion migration, but the slow diffusion of these ions into
both the cathode and anode gels formed stained skins that were
hardened to a mild degree. The chromium from these tests readily
washed from the gels and the gels swelled in w'ater, melting at rela-
tively low temperatures. The anode gels from the tests involving
transfer of chromium ions of negative charge were also hardened to a
moderate degree and had the properties of the gels that were hardened
with chromium of zero charge. Hence the gelatin plugs that figured
in the ion migration experiments showed 2 types of hardening with
chromium. The hardening produced by any of the positively charged
chromium ions is complete and permanent; chromium cannot be
washed from the gels or removed by reversal of the emf , and definite
chemical combination must occur between the positive chromium
ions and the negatively charged gelatin carboxyl groups, with the for-
mation of coordination-type chromium molecules that are restricted
to the gel. This type of hardening is relatively independent of the
concentration of chromium within the bath, since the gel will tend to
react with chromium until the negatively charged carboxyl groups
are satisfied. Combination of the gel with chromium molecules of
zero charge and even negatively charged chromium ions does occur,
but in this case the combination is weak and is reversed by washing ;
combination of this type is thus of the nature of an adsorption process.
It is probable that the molecular or negatively charged chromium
unites with gelatin by the same mechanism that permits combination
with positively charged chromium, namely by coordinating with
gelatin carboxyl groups, but there is no favorable charge to make the
bond tenacious and therefore the reaction is reversed merely by dilu-
tion.

Motion picture negative film samples that were developed and
hardened in these aged chromium sample solutions proved to be

July, 1946

GELATIN HARDENERS

hardened only by those solutions which contained positively charged
chromium ions. Solutions in which the chromium had a charge of
zero or negative were entirely without hardening action. Combina-
tion of gelatin carboxyl groups and negative chromium would not
be expected to take place, except under the unusual conditions that
are encountered in the ion migration experiments, where unfavorable
charge is overcome by an applied emf. Uncharged chromium, how-
ever, is unaffected by the charge on the gel and such ions are free to
enter a gel by diffusion.

'00

024 6 8 10 12 14 16 18 20

WASHING TIME IN* MINUTES

FIG. 12. The effect of washing film after fixation upon
the melting point of dried negative film that is fixed in a
bath of chromium possessing a net zero charge. No hard-
ening is produced until the film is dried and the adsorbed
chromium tends to be removed during the washing process.

While film is unhardened by molecular chromium during normal
film fixation, the chromium that is left in the film combines actively
with the gelatin during the drying process ; even in the dry state the
combination of molecular chromium with gelatin carboxyl groups
continues to take place and the film becomes increasingly insoluble
upon standing. Fig. 12 shows the effect of washing time upon the
hardening produced after drying for film that was bathed in a chrome-
alum bath that contained chromium of zero charge. Oxidation. of
coordinated bisulfite groups or loss as sulfur dioxide could cause a

48 H. L. BAUMBACH AND H. E. GAUSMAN Vol 47, No. 1

return of positive charge that would favor the combination with gela-
tin that occurs during and following the drying process. Negatively
charged chromium ions have no hardening action upon the gelatin of
motion picture films, either during processing or following the drying
operation.

The chrome-alum solutions that contained bisulfite, and that were
shown to reach equilibrium with respect to hydroxide at different
rates (Fig. 10), proved to lose their positive charge in a manner that
is directly related to the presence of hydroxyl groups. The solutions
at pH 4.0 that involved 2 hydroxyl groups for each chromium atom
were found to contain only chromium of zero charge. The same con-
ditions (pH 5.0) that involved more than 2 hydroxyl groups produced
chromium of negative charge by the amount in excess of the 2 hy-
droxyl groups. Hence it is quite certain that chromium coordinates
with one bisulfite group (when at least this much bisulfite is present)
and that additional bisulfite catalyzes the attainment of the equilib-
rium condition involving the basicity of the chromium. Thus, at £H
4.0, the molecule Cr(H2O)3(HSO3)(OH)2 is formed as an end product,
and the rate of formation is proportional to the bisulfite concentra-
tion. At higher pH values, negatively charged ions are formed that
can be diagrammed as Cr(H2O)2(HSO3)(OH)3- or Cr(H2O)2(HSO3)2-
(OH)2~. Consequently, it is possible to explain the loss of hardening
properties which occur during the aging of chrome-alum fixing solu-
tions. The chrome-alum baths that are low in sulfite (especially
those with less than the 1 : 1 molal ratio) will retain positively charged
chromium ions of the form of Cr(H2O)4(OH)2+ or Cr(H2O)4 (HSO«)-
(OH) + for a considerable length of time and they will, therefore, re-
tain their active hardening properties.

Chromium solutions, both with and without thiosulfate, give iden-
tical results in the experiments that are outlined above ; it is reason-
able to assume that thiosulfate ions do not form coordination type
complexes with chromium ions, in spite of the high concentration of
this substance in photographic fixing baths. It is evident that the
presence of any additional negatively charged ions that will coordi-
nate with chromium (especially the salts of organic carboxylic acids)
will tend to eliminate the rather small number of positively charged
chromium ions and will hasten the loss of hardening within the solu-
tion. It is for this reason that Crabtree and Russell6 found that the
organic carboxylic acids, such as acetic acid, were unsuitable for use
with chromium fixing baths.

July, 1946 GELATIN HARDENERS 49

The evidence strongly indicates that the hardening of gelatin by
chromium and by aluminum follows the same mechanism, consisting
of the formation of coordination complexes between gelatin carboxyl
groups and the metal ion of positive charge. This view is supported
by the spectrophotometric measurements of Kuntzel and Riess,7 who
obtained very similar curves for chromium in the presence of gelatin,
glycocoll, and acetic acid, and of Kuntzel and Droscher,8 who gave
evidence for the formation of complex salts between chromium and
gelatin. Conditions are most favorable for coordination (hardening)
when the maximum number of metal ions have a positive charge and
when the maximum number of gelatin carboxyl groups have a nega-
tive charge. Positive charge for the metal ions is favored by a low
pH, while negative charge for the gelatin carboxyl groups is favored
by high />H ; consequently the pH range that is favorable to the hard-
ening process is rather narrow and the pH of maximum hardening is
dependent upon the conditions within the solution and the isoelectric
point of the gelatin.

Other concepts of the hardening process have been advanced; it
has been suggested that hardening is the result of the precipitation of
basic chromium salts within the gel and that coordination of chro-
mium occurs with gelatin amino groups or with the polypeptide link-
ages of the protein molecules. These concepts are not necessary in ex-
plaining the hardening action of chromium and, indeed, are not sup-
ported by the evidence, since combination of chromium and gelatin
occurs actively with any positively charged chromium ion, even the
Cr(H2O)6+++ ion of no basicity, and combination between positively
charged gelatin and negatively charged chromium ions does not tend
to take place.

Since the maximum positive valency of chromium in a photographic
fixing bath is not likely to exceed one, even at a pH value of 4, the
valency of the chromium-protein complex is satisfied and there is no
appreciable tendency to retain silver thiosulfate. Consequently,
chrome-alum fixing baths have silver and hypo retention properties at
pH 4.0 that are comparable to the properties of aluminum baths at
pPL 6.0, where a similar situation exists with respect to ionic charge.

Practical Operation of a Chromium Fixing Bath. — In the previous
section it was pointed out that chromium produces 2 types of gela-
tin hardening, depending upon the charge on the chromium mole-
cule. Chromium molecules of positive charge harden gelatin very
well, usually rendering it insoluble in boiling water as it leaves the

50 H. L. BAUMBACH AND H. E. GAUSMAN Vol 47, No. 1

hardening bath; moreover, the hardening process is relatively inde-
pendent of the chromium concentration. The other type of harden-
ing does not occur until the film has been dried and is caused by the
retention of uncharged molecules of chromium which combined chemi-
cally with the gelatin during and following the drying of the film.
This type of hardening contributes nothing to the safety of handling
during processing but produces satisfactory projection and handling
properties as the film stands after being dried.

The type of hardening that is produced by positively charged chro-
mium ions is the more desirable, since it is during the processing of
the film that it is most easily damaged. In order to obtain this type
of hardening consistently, it is necessary to maintain careful chemical
control over the hardening bath as it is being used. Loss of positive
charge on the chromium complex results from the increased pH in the
solution and the presence of too much bisulfite. The hardening prop-
erties can be retained most readily in a chromium bath by keeping the
pH of the bath at 4.0, or somewhat below, and by maintaining a sul-
fite-to-chromium-molal ratio that is less than 2:1. While the hard-
ening qualities of such a bath are excellent with sulfite-to-chromium
ratios of 1 : 1 or less, so little free sulfite exists in the solution that it
sulfurizes very rapidly. Consequently, satisfactory operational
qualities are maintained only if the ratios of sulfite to chromium lie
between 1 : 1 and 2:1.

Control over the pH of the fixing bath is necessary in plant opera-
tion in order that the action of development will be stopped uniformly ;
although the actual pH value that is selected for control may vary
from 3 to 5, it is probably most satisfactory to regulate the pH to a
value of about 3.8. The lower the pH value that is selected for con-
trol, the higher may be the sulfite-to-chromium ratio, without losing
the hardening properties of the bath. Russell and Crabtree9 have
shown that fixing baths should not be used at pH values much less
than 4.0, if the silver image is not to be "reduced" by the solvent ac-
tion of the bath. At pH values as high as 5, so little sulfite may be
used (only a 1 : 1 ratio) that control is very difficult. At pH 3.8 the
silver density reduction properties of the bath are low, the sulfunza-
tion properties are not critical, and the hardening properties are satis-
factorily retained at sulfite-to-chromium ratios up to 2:1. Fig. 13
illustrates the manner in which a chromium fixing bath of 2 : 1 sulfite-
to-chromium-molal ratio maintains its hardening properties with age
at H 4.0.

July, 1940)

GELATIN HARDENERS

The preparation of a chromium fixing bath should follow the pro-
cedure that is designed to delay loss of positive charge on the chro-
mium as long as possible. Under no circumstances should any of the
organic acids, such as acetic, citric, or formic, be used in its prepara-
tion or control. The desired quantity of sodium thiosulfate should
be dissolved first in water that is high enough in temperature to pro-

100

:90

I 70

gso

O O

4 6 8 10 12 14

AGE OF FIXING BATH IN DAYS

FIG. 13. The melting point values of negative film
developed, rinsed, and fixed in chrome-alum fixing baths
containing 350 grams per liter of hypo and a 2:1 molal
ratio of sulfite to chromium for various chrome-alum
concentrations from 10 grams per liter. Baths are con-
trolled to pH. 4.0 by frequent addition of sodium hydrox-
ide. If the molal ratio of sulfite to chromium does not ex-
ceed 2:1, the hardening properties tend to be retained at
pH4.

duce the hypo solution at room temperature or below. The anhy-
drous sodium sulfite should next be added to the hypo solution in an
amount that is related to the chrome alum that is to be added later.
Enough sulfuric acid should be added to bring the pH of the sulfite
and hypo bath down to about 6.0; in this form the bath has excellent
keeping qualities and may be stored indefinitely. Just before use,
the dry chrome alum should be added to the solution, with good agita-
tion, until the chromium salt has dissolved. If the agitation is not
adequate, complete solution is not possible; hence, under certain

52 H. L. BAUMBACH AND H. E. GAUSMAN Vol 47, No. l

conditions, it may be preferable to use a solution of chrome alum
which has been freshly prepared with the minimum amount of cool
water. The pH of the solution after the addition of the alum will be
close to 4 and it will become lower upon standing. This bath will not
lose its hardening properties completely upon standing, but for satis-
factory operation it is necessary that the £H be controlled by addi-
tion of sulfuric acid or sodium hydroxide.

Replenishment of a chromium fixing bath should be based on proper
maintenance of pH, sulfite concentration, hypo concentration, and
concentration of positively charged chromium. It is most satisfac-
tory to add the sulfite and hypo in the dry form and to replenish the
chromium with a freshly prepared saturated solution of chrome alum.
As little as 0.5 gram per liter of chrome alum per day is required to
maintain the hardening action of a bath that is in constant use.

Chromium fixing baths are not nearly so simple to operate as are
aluminum baths, but they possess certain advantages. In addition
to a somewhat greater degree of hardening, the use of a chromium
fixing bath, under the conditions outlined above, permits easy removal
of silver and hypo from the finished product, with resulting perma-
nence of the film.

The Retention of Hypo and Silver by Aluminum and Chromium.—
Crabtree, Eaton, and Muehler3 have made a comprehensive study
of the factors that influence the retention of hypo and silver by film
that is fixed in various solutions. Their researches have indicated
the difficulty that is experienced in washing hypo and silver from film
that is not processed under favorable conditions. An aluminum fixing
bath was found to cause retention of much greater quantities of the
thiosulfate complex than was caused by a chromium bath, and it was
found that the aluminum bath had much better properties when oper-
ated at pH values above the isoelectric point of gelatin.

Limited experiments in this laboratory have confirmed the obser-
vations of the above authors and, in addition, have shown that
chromium may retain hypo and silver and that aluminum may not
retain these substances, depending upon the complexing anions that
are present. Since aluminum and chromium normally differ de-
cidedly in their retention properties, it is not likely that the isoelectric
point of gelatin bears any more than incidental relationship to the
retention problem. It was found that the aluminum-citric acid
formula discussed earlier had retention properties at pH 4.2 that were
the equal of the properties of the aluminum-acetic acid-boric acid

July, 1946 GELATIN HARDENERS 53

bath at pH 5.0, even though practically equal hardening was pro-
duced. Also, the properties of the chromium fixing baths, with a 3 : 1
molal ratio of sulfite to chromium, were better at low pH values than
the bath with the 1 : 1 ratio.

The above authors found that hypo retention generally parallels
silver retention; this suggests that the mordanting action involves
the coordination of a complex silver thiosulfate ion, the simplest of
which is the ion, AgS2Os~. Since this ion is negatively charged, it
will tend to coordinate with positively charged complex metal ions
but it will not readily coordinate with neutral molecules or negatively
charged ions. Aluminum retains the thiosulfate complex when its
coordination with gelatin carboxyl groups still leaves the group with a
positive charge. Reduction of the net positive charge by complexing
with citrate ions or by increasing the pH precludes the possibility of
coordinated AgS2Os~ groups. If the thiosulfate complex is coordi-
nated with aluminum in the fixing bath and such groups become fixed
to protein carboxyl groups, they can be removed in the washing proc-
ess by metathesis with hydroxyl groups, which explains the help-
ful action resulting from the addition of alkali to the wash water.

Divalent positively charged chromium ions are rarely present in
chrome-alum fixing baths because coordination of one bisulfite group
and one hydroxyl group occurs almost at once. Divalent chromium
exists in fixing baths of low sulfite concentration and low pH ; such
solutions cause retention of silver and hypo as is the case of aluminum.

REFERENCES

1 RUSSELL, H. D., AND CRABTREE, J. I.: "An Improved Potassium Alum Fix-
ing Bath Containing Boric Acid," /. Soc. Mot. Pict. Eng., XXI, 2 (Aug., 1933),
p. 137.

2 CRABTREE, J. I., AND HARTT, H. A.: "Some Properties of Fixing Baths,"
Trans. Soc. Mot. Pict. Eng., XIII (May, 1929), p. 364.

8 CRABTREE, J. I., EATON, G. T.f AND MUEHLER, L. E.: "The Removal of
Hypo and Silver Salts from Photographic Materials as Affected by the Composi-
tion of the Processing Solutions," /. Soc. Mot. Pict. Eng., 41, 1 (July, 1943), p. 9.

4 KUNTZEL, A., RIESS, C., AND KONIGFELD, G. I "Mineral Tanning III — The
Formation of Masked Complexes in Normal and Basic Solutions of Chromium
and Aluminum Salts," Collegium (1935), p. 484.

6 FRIEDMAN, J. F.: "Photographic Reviews," Amer. Phot., 36 (Dec., 1942),
p. 34; 37 (Feb., 1943), p. 42.

6 CRABTREE, J. I., AND RUSSELL, H. D.: "Some Propeities of Chrome- Alum
Stop Baths and Fixing Baths — Pt. I," /. Soc. Mot. Pict. Eng., XIV (May. 1930).
p. 483; "— Pt. II," p. 667.

54 H. L. BAUMBACH AND H. E. GAUSMAN

7 KUNTZBL, A., AND Rmss, C. : "Mineral Tanning IV — The Nature of the
Combination of Basic Chrome Salts with Hide Substance," Collegium (1936), p.
138.

8 KUNTZEL, A., 'AND DROSCHER, K. T. : "Mineral Tanning XII — The Reaction
' of Chrome Salts with Gelatin," Collegium (1940), p. 106.

9 RUSSELL, H. D., AND CRABTREE, J. I.: "The Reducing Action of Fixing
Baths on the Silver Image," /. Soc. Mot. Pict. Eng., XVIII, 3 (Mar., 1932), p.
371.

THE APPLICATION OF PURE MATHEMATICS TO
THE SOLUTION OF GENEVA RATIOS *

RON W. JONES'

Summary. — A method is described and formulas given to determine mathe-
matically the relative angular displacement and film velocity ratios in the regular
Geneva movement as used in 3 5 -mm motion picture projectors.

The subject of Geneva movement analysis has been ably dealt
with in the literature. The author, however, feels the urge to con-
tribute his small share and to this end sets out herein a system of
direct ratio solution by means of pure trigonometry.

The method described was evolved during an analysis of relative
film velocities with a view to investigating the possibilities of optical
compensation for intermittent film motion. Experimental work in
this direction has indicated the necessity of near sine wave motion
with the possible expansion of the pull-down period to 180 deg. A
knowledge of Geneva movement curve diagrams was essential and it
was deemed desirable, in view of other aspects of the work, to resort,
if possible, to pure mathematics.

It is not the purpose of this paper to set forth other than the
methods and formulas employed to obtain the relative angular dis-
placement and velocity ratio values for the orthodox Geneva move-
ment as used in 35-mm motion picture projection equipment.

Fig. 1 displays a four- to-one movement at rest in a position wherein
the cam pin has entered the starwheel slot to a depth equal to half
its own diameter. This may be termed "start of pull-down period"
and all ratio characteristics may be observed during the 45 deg of
cam rotation immediately following this point. Obviously this
supplies all necessary data for plotting the half curves. The full
90-deg curves are obtained by continuation to the right of the 45-
deg abscissa since the second half represents a reversal of the first.

* Submitted Jan. 11, 1946.
** Western Electric Company (Aust.) Pty., Ltd., Brisbane, Australia.

R. W. JONES

Vol 47, No. 1

Relative dimensional ratios with reference to Fig. 1 are as fol-
lows:

77 - 7 = 45°

M =• 90°
a — c = b cos 7 = b sin y

FIG. 1.

The first curve to be plotted is that for relative angular displace-
ment for cam and starwheel. Fig. 2 represents, schematically, a 5-
deg rotation of the cam from original zero position and solution for 7
may be obtained from

tan

— 7

For logarithmic computation

L tan ^-? = log (b - c] - log (b + c) + L tan ^90° - |

July, 1946

APPLICATION OF MATHEMATICS

At zero position taking c as unity then

b = — ^— =» 1.414

COS 7J

so that, solving for a 5-deg cam displacement, we have

L tan t—^2 = log 0.414 - log 2.414 + L tan ^90
" = T.2343 + 10.4389 = 9.6732

whence

Now

— y

FIG. 2.

- 25°14'orM - y - 50°28'

90° - J - 90° - 20" - 70

so that M + T = 140°.

R. W. JONES

Vol 47, No. 1

0 5 10 15 20 25 30 35 40 45 50 55
ANGULAR DISPLACEMENT OF CAM ( DEGREES )

FIG. 3.

FIG. 4.

July, 1946 APPLICATION OF MATHEMATICS

By subtraction

M + 7

140 °00'
50°28'

.'. Starwheel displacement = 45° - 44°46' •» 14'

Repeated solution for successive 5-deg steps of cam rotation gives
the corresponding increments of starwheel displacement and from
this we obtain the curve as in Fig. 3.

FIG. 5.

The value log (b — c) -- log (b + c) is a constant, remaining
1.2343 for all applications.

Consideration may now be given to the linear velocity of the
cam pin. Fig. 4 shows the cam only and PI represents any position
of the pin between P and PZ. If we take PE = R, linear velocity =
V, and rpm — S, then obviously

V, = 6*2 TT R

and

R. W. JONES

sin <f> S2 TT R

Vol 47, No. 1

From this we may obtain the linear velocity of the pin in any
positive direction and thus the value Vx in Fig. 5 in which PI repre-
sents the pin at its intersection with the starwheel slot DM, and x is
normal to DM.

Now since y is parallel to PE, sin <£ = sin co, and since co + (3 = 90°,
/? is complimentary to co, so that cos /? = sin co = sin <f>.

10 15 20 25 30 35 40 45 50 55
ANGULAR DISPLACEMENT OF CAM ( DEGREES )

FIG. 6.

At zero position 6 = 90° and as the cam and starwheel revolve in
engagement 6 diminishes until finally, at 45 deg cam position, 6 = 0.

It is evident then that at any point we may obtain Vx from Vx =
cos e S2 TT R.

The value Vq (i.e., linear velocity of starwheel tip) may thus be
found from

Vq = COSd S2Tr

Our Fig. 3 curve gives values of a for corresponding values of </> so
that 6 may be found for any position from 6 = 90° — (0 + «).

July, 1946 APPLICATION OF MATHEMATICS 01

From the foregoing equations we may, by progressive solutions,
plot the curve (Fig. 6) showing the velocity ratio between cam and
starwheel and the instant starwheel velocity, at any angular position,
in revolutions per given time period.

The following facts which are revealed are of interest :

(1) From zero to 7-deg cam rotation there is very little movement of the
starwheel, which contributes in no small measure to the ability of a 90-deg shutter
blade adequately to obscure the normal pull-down period, providing the shutter
is working at a reasonable beam diameter.

(2) From zero to 25.5 deg the starwheel is at a lower velocity than the cam.
Beyond the 25.5 deg the starwheel velocity is in excess of the cam velocity.

(5) From zero to approximately 30 deg the rate of acceleration of the starwheel
increases. Between 30 and 35 deg it remains almost constant. Beyond 35 deg
it decreases.

^••eaiBiiiaiiiiiie

FIG. 7.

(4) Top speed of starwheel: 3470 rpm (58 rps.)

Top speed of film: 867. 5 ft per min (14.5 ft per sec)

As an interesting practical check on Fig. 3, this curve was also
recorded photographically. The method employed involves the
exposure of a moving strip of motion picture positive film to the
projected image of a small hole drilled in a steel film running through
the projector.

The unexposed film is enclosed in a lightproof container mounted
directly in front of the projector mechanism. A 4-frame standard
sprocket, mounted internally and driven continuously at one-to-one
directly from the projector camshaft, is arranged to drive the film
horizontally across a narrow scanning aperture in the front of the
container and in line with the optical axis of the projector.

The standard perforated steel film is threaded through the pro-
jector gate and driven by the intermittent sprocket in the normal

62 R. W. JONES

manner, the projector shutter being removed so that the complete
cycle is unobscured. An extension barrel fitted to the objective lens
permits of the lens being worked far enough forward to focus a one-to-
one image at the surface of the unexposed film. At single-frame
intervals the steel film is drilled 1/64 in. in the center of the normal pic-
ture area.

The developed film carries a photographic trace of the spot image
and this is, in effect, a recorded diagram of the relative angular posi-
tions of cam and starwheel throughout the complete pull-down
period. The remaining 270 deg, during which the starwheel is at
rest, is recorded as a straight line since the spot image is stationary.

Fig. 7 is a print from the original record.

Projection of the negative onto a sheet of co-ordinate paper en-
ables us to obtain values for the curve, which in fact, are almost iden-
tical with our calculated values.

The author wishes to express his appreciation of the assistance
rendered him by D. Urquhart and W. Kersley in the construction and
operation of the apparatus for obtaining the above-mentioned photo-
graphic diagram.

A NATIONAL FILM LIBRARY— THE PROBLEM OF
SELECTION*

JOHN G. BRADLEY**

Summary. — Some pictorial evidence of all motion pictures produced should be
preserved. Bulkiness of material, however, and expense of handling makes preserva-
tion of all motion pictures in their entirety impracticable. A solution may be found
in the use of film strip which would preserve one frame of each important scene.

The Library of Congress proposes to preserve a considerable part of the motion pic-
tures produced. Considerations in selection policy include possible uses to be made of
such a collection, the avoidance of repetitious matter, and the public acclaim given such
material in the form of both critical opinion and box office returns.

The collection will be world-wide in scope and will include both 35-mm and 16-mm
film. The motion picture industry is invited to cooperate in creating a national film
library on the theory that such a library will be mutually advantageous to all concerned
— government, education, research personnel, procedures, and others.

If N money, personnel, and related facilities were available the
simplest method of assembling a motion picture library collection
would be to include all available material for permanent preservation.
This method has the virtue of requiring little if any exercise of judg-
ment but carries its own penalties as will be noted later in this dis-
cussion.

A second method would be to include all available material as
before but control its volume and subject matter content through
periodic diminutions.

A third method, the one that has been used by the Library during
the last 2 years, is based on selections through preaudits. Both the
second and third methods require the exercise of judgment, one in
advance of the selections and the other following a period of reflec-
tion.

A fourth method, one that is presently recommended, is also
based on selection as opposed to total collection and contains some

* Presented Oct. 15, 1945, at the Technical Conference in New York.
** Director of the Motion Picture Project, The Library of Congress, Wash-
ington, D. C.

64 J. G. BRADLEY Vol 47, No. 1

elements of preauditing. It differs principally from the method pre-
viously used in that it admits a greater segment of the public in the
determination of the items selected.

If precedent has any value in this connection it should be pointed
out that neither The Library of Congress, nor the National Archives
has found it expedient to preserve everything available. For example,
if the National Archives undertook such a course it would find itself
encumbered with a million and one items such as carbon copies of the
originals, work sheets, receipts and invoices, punch cards, cancelled
money orders and vouchers, and miscellaneous forms that had served
their legal and administrative functions and that contained little
historical value or permanent interest. In other instances it would
find such information to be highly repetitious with only occasional
variants having significance. In still other instances it would find
that much of the essential information found in such material had
been siphoned off into less bulky form such as statistical tables. In
any case, the reality of the situation in terms of handling and storage
costs as they relate to Government records became acute and led to
the enactment of special disposal legislation in an effort to resolve the
problem.

A similar reality exists for The Library of Congress in the case of
motion pictures, a reality that is intensified by the fact that the
volume of such material has already reached major proportions and
that this type of material is relatively more bulky and expensive to
handle than paper material. Whereas a manuscript or a printed book
covering a particular subject might occupy only a fraction of a cubic
foot of space and involve a nominal charge to reproduce, a motion
picture film on the same subject might require nearly 2 cu ft (six 35
mm reels to the cubic foot) and cost $200 or more to reproduce. The
cost of screening and consultation is likewise more expensive, requir-
ing a projection or workroom with special equipment and the services
of one or more technicians.

Continuing the comparison with paper records it may be said that
motion pictures also have much repetitious matter. For example,
The Three Stooges, a series produced by Columbia Pictures, follows
the same general pattern throughout the years, the variants con-
sisting chiefly of minor changes in plot, setting, and clothing. The
same can be said of the Laurel and Hardy pictures, most of the West-
erns, and dozens of others. As in the case of paper records, the siphon-
ing-off process is applicable to motion pictures in the execution of film

July, 1946 SELECTING A NATIONAL FILM LIBRARY 65

strips which will be discussed later. It appears, therefore, thai both
the volume and nature of the film under consideration as well as the
precedents established in archival and library practices would justify
only a partial inclusion of the total material available.

Indeed no alternative plan seems feasible at the moment. Respon-
sible judgment will be required, however, in formulating a selection
technique that will adequately serve the ends sought, that will be
sufficiently basic to serve as a major premise and at the same time
sufficiently flexible to accommodate itself to changing circumstances.
Upon what factors will judgment be exercised and who will partici-
pate in exercising it? What are the service implications of such a
collection? What kinds of films should be considered and from what
sources should they be acquired ? How much film should be included ?
These are a few of the considerations that should be taken into ac-
count in developing a selection formula for a film library.

Heretofore judgment has been exercised, in making selections for
The Library of Congress, by a small group of analysts on the basis of
diagnosis (preaudits) in terms of certain rather fugitive preconceptions
such as good or bad, true or false, appropriate, and the like. On the
whole the canons of selection, previously used by The Library of
Congress, cover the general direction in which the movement should
go. But no small group of analysts, regardless of its qualifications,
could make selections in terms of diagnosis that would satisfy another
group similarly qualified; the factors of judgment are too tenuous.
Nor could such a group satisfy the public as a whole ; the elements of
controversy are too pronounced. This brings us, therefore, to a con-
sideration of public participation — at least, participation by those
segments of the public most concerned. While an individual member
of the public might make his own selections in terms of the good or
bad concept, his judgment would be counterbalanced by the judgment
of other individuals; likewise the judgments of special groups would
counterbalance each other. The result would be a general acceptance
of what is rather than a search for what should be, which is the Li-
brary's present attitude toward the printed page. If such participa-
tion is allowed and proves successful the collection would represent
a broad horizontal foundation upon which each person in his own time
could erect his own vertical structure in terms of his own individual
interests.

More specifically what segments of the public should be asked to
participate? The answer to this question can be suggested by a con-

66 J. G. BRADLEY Vol 47, No. 1

sideration of the consumer pattern as, Who saw the pictures? How
many saw them? Where were they seen? etc. Again, what do the
critics, steeped in the tradition of the movies, think of them? How
do the producers themselves evaluate their own product? There are
also many reviewing groups throughout the country that reflect
critical opinion in terms of special interests such as education, religion,
and the like; what evaluation do they place on the different pictures?
These are all valid, elemental, and determinable factors. They are
widespread in their significance and democratic in character. The
composite judgment of such segments, being self-imposed, should
prove generally acceptable to all concerned. The deficiencies, if any,
found in a collection resulting from such participation could be cor-
rected by The Library of Congress.

Perhaps one of the most important considerations in terms of this
discussion is the service implications of such a collection. In other
words, who will use it and how will it be used? The statement has
been made in this connection that such a collection should serve the
needs of future historians. No objection is offered to such a sugges-
tion. Certainly no medium records our comings and goings quite so
graphically as does the motion picture nor offers the historian quite
so rich a warehouse of source material with which to evaluate the past.
Even the producer may turn historian on occasion with profit to
himself both in terms of box-office returns and pride in his craft.
But no priority should be given to the historian at the expense of
many others having an equal interest in such material. These others
would include students of the manifold arts as reflected in motion
pictures such as music, the dance, make-up, costuming, speech, and
drama in general; scientists and technicians having an interest in
optics, electronics, and photochemistry; sociologists and psychia-
trists interested in human behavior; business men contemplating
investments; public leaders and public officials who may want to
mobilize our national resources through the use of this medium in
some great crisis; producers looking for research material as well as
actual footage on nonrestricted films; and taxpayers in general who
may be motivated solely by a curiosity in life as mirrored on the
screen.

Sources may be divided into several overlapping categories such as
domestic and foreign, professional and amateur, government and
private, etc. Government sources, for example, will yield both orig-
inal negatives and service prints and the obligation for selecting and

July, 1946 SELECTING A NATIONAL FILM LIBRARY 67

preserving the originals rests with others. Duplicate copies of such
films originating with the government, however, along with all non-
government films remain eligible for acquisition by The Library of
Congress. Perhaps all kinds of films and all available sources should
be considered without prejudice. The philosophy of the untouchables
would not seem to apply here and no film or source in itself should be
regarded as taboo whether the film be an entertainment or an educa-
tional film; whether it be a training, propaganda, medical, musical,
gangster, or slapstick film ; whether, in a broader sense, it be a factual
or fictional film; or whether the source be domestic or foreign.

The volume of the collection should be liberal, sufficient to give an
accurate index of production and consumption and adequate to
serve research needs on a wide front; diminutions can be made later
if necessary. The chief limitations with respect to volume would seem
to be (7) that the collection itself should not become repetitious and
(2) that it should not seriously duplicate evidence found elsewhere
as in the printed page. Thus a film library might not, for example,
want all of the Three Stooges nor all available newsreels covering a
particular event. Such a situation would represent repetitious subject
matter. Neither would it want a full motion picture coverage on an
extended conference at which no significant action takes place and
which could be more economically recorded by other means.* In this
case the motion picture of the conference would represent needless
duplication or expensive substitution. In general the volume of the
collection will be determined by quotas on the basis of available
funds and the funds will be determined largely on the basis of service
rendered. Factual film depicting people, things, and events should
be selected on a liberal basis and other films should be selected on a
representative basis. It should not be forgotten, however, that in the
so-called entertainment film are found many of the basic arts and
sciences and that a fictional motion picture itself and in its own right
is a thing, while those participating in it are certainly people.

The burden of this discussion so far has been a consideration of
selection versus total collection. There is one aspect of total collec-
tion, however, that merits comment; I refer to the possible use of the
film strip. This device, in brief, represents a pictorial summary
of the complete release which may be consulted through the use of a
microfilm reading machine. In other words, it is a strip of film con-
taining a series of still pictures selected from the original release
copy in the same sequence as found in the original. In this connec-

68 J. G. BRADLEY Vol 47, No. 1

tion it should be remembered that each frame in a scene represents a
progressive repetition of the subject matter depicted. Thus in one
frame a man's hand may appear on a level with his eyes while in
succeeding frames it may progress downward until it appears on a
level with his waist. The loss sustained in discarding all the frames in
a scene except the one or two retained in the film strip would represent
primarily only a loss of motion. For most of the potential users of
such a collection, especially those interested in documentary studies,
such a loss would not comprise a serious handicap. A more serious loss
would be that of the sound track; but even this might partially be
overcome be rerecording through some microphonographic process
and, as far as speech is concerned, through the preservation of the
dialogue in script form. In any case the film strip technique would
need be applied at first and in terms of this discussion only to those
subjects not otherwise selected for preservation. Its possibilities
should be explored further.

Another aspect of total collection is found in the literature related
to film: production schedules, published reviews, scenarios, cutting
continuities, dialogues, and the like. From such material and from
kindred sources a union catalogue could be evolved that should
prove beneficial to producers and scholars alike. Already The
Library- of Congress has a priceless collection of material covering all
copyrighted film from the beginning of the motion picture industry.
The organization of such material on a service basis awaits only the
availability of additional funds and personnel.

In applying the proposed formula the following outline of general
sources is submitted together with a suggested quota and pertinent
comments for each source group listed.

Group 1 — American Newsreels. — Acquire the complete edited
output of one of the major American newsreel companies for one
year and examine the output of the other newsreel companies for
possible supplemental material. Alternate this plan from year to
year among the companies concerned. The selection of the supple-
mentary material can be made for the most part through an ex-
amination of data sheets. Acquisitions to be made currently.
Estimated annual yield, approximately 200,000 ft.*

Group 2 — American Citations. — Acquire each edited title listed

* Subject to revision in terms of greater post-war divergence of subject matter
among newsreel companies.

July, 1946 SELECTING A NATIONAL FILM LIBRARY 69

one or more times in the following categories: (a) Academy cita-
tions, (b) Film Daily citations including the "ten best" as well as
those listed in the so-called honor roll, (c) citations by leading news-
papers, (d) citations by miscellaneous reviewing organizations repre-
senting special interests, and (e) the so-called box office champions.
No screening required; acquisitions to be made at end of the year.
Estimated annual yield, 100 titles, or 500,000 ft.

Group 3 — Producer Selections. — This group would include cita-
tions by the producers of their own pictures not covered in Groups 1
and 2. In other words, a producer, having invested his best thought,
his time, his money, and other resources in the production of a pic-
ture, has a right to be heard in the matter of selection and preserva-
tion by a public institution. Each producer should be free to set up
his own standards of selection; in brief, his selections would repre-
sent the pictures he wants the government to preserve. The maxi-
mum quota from each producer should be set at approximately
50,000 ft. Acquisitions to be made at the end of the year. Esti-
mated annual net* yield, 100 titles, or 500,000 ft.

Group 4 — American Miscellaneous. — Acquire, on a selection and
quota basis, other edited subjects from American sources that are
not covered in Groups 1, 2, and 3 and that will represent a well-
balanced cross section of the industry's output. These subjects
should include many films that the critics have not acclaimed and
the producers have not selected or that may have been box-office
failures but that, nevertheless, are a part of the movie production
and consumption pattern. Most of these selections could be made
on the basis of reviews and data sheets; screenings could be ar-
ranged for doubtful cases. Acquisitions to be made at the end of
the year. Estimated annual yield, 100 titles, or 500,000 ft.

Group 5 — American Government. — This group should include
nonrecord film and library copies of record film produced or spon-
sored by the government. Details covering plans of operation
should be worked out jointly with the National Archives. Esti-
mated annual yield, 500,000 ft.

Group 6 — American Nontheatrlcal. — This group should consist
principally of factual expository film used for teaching, training,
and documenting purposes. It is sometimes referred to as the 16-
mm field although many of the negatives are and presently should

* Total selected less duplications found in Group 2.

70 J. G. BRADLEY Vol 47, No. 1

be on the 35-mm size. A substantial part of such film is available
through copyright channels but the scope should be materially
expanded. No screening implied. Acquisitions to be made cur-
rently. Estimated annual yield, 200 titles, or 150,000 ft.

Group 7 — Foreign Miscellaneous. — This group should include
a well-balanced cross section of foreign productions, world wide in
scope, that would consist of newsreels, shorts, features, and exposi-
tories. Selections for the most part could be made on the basis of
critical opinion. Acquisitions to be made currently. Estimated
annual yield, 100 titles, or 500,000 ft.

Group 8 — Unedited Footage. — This group may be divided into
2 subgroups: (a) film that has been exposed in connection with
production work but not used in the final release and that has been
set aside as "library shots," and (b) related sequences taken from
edited or released pictures and compiled into series on such topics
as geography, transportation, industry, agriculture, music, speech,
sports, and the like. The possibilities of this second subgroup have
been discussed with key men in government, industry, and educa-
tion with favorable reaction but remain to be explored further.
Acquisitions to be made currently. Estimated annual yield,
150,000 ft.

The total annual yield under the foregoing plan would be approxi-
mately 3,000,000 subject ft, or 900 titles. If both a preservation
copy and a service copy could be included for each item acquired,
the storage load would be 6,000,000 ft (6000 reels), or approximately
1000 cu ft. Note: It should be observed that groups 1, 2, 3, and 4
represent the domestic 35-mm or theatrical field on released material.
These 4 groups would yield less than 2,000,000 subject ft a year in
newsreels, shorts, and features. The amount of such film thus pro-
posed for selection and preservation represents a little less than 25
per cent of the total output of these 4 source groups, a quota that
appears to be entirely reasonable.

In conclusion I wish to say that the plan outlined herein has been
discussed widely with government people handling film. It is pro-
posed to discuss it also with leaders in the motion picture industry
in an effort to perfect it on a practical basis. It is believed that
through such discussions and through other exchanges of opinions and
benefits, acquiescence to a plan will be turned into active support of
a movement. The benefits which the American people and the
government will derive from such a collection through the years to

July, 194G SELECTING A NATIONAL FILM LIBRARY 71

come are valid and obvious. The use of motion pictures during the
recent international struggle has demonstrated the power of this
relatively new medium.

The benefits which the producers will enjoy are equally valid, and
equally obvious when all the possibilities are considered. For example,
the vast resources of the Copyright Office of The Library of Congress
in terms of its literature related to motion pictures remain practically
unexplored. It is proposed to organize this material and make it
easily accessible to everyone interested. The usefulness of a union
catalogue on extant film, whether the film itself is in The Library or
not, suggests a service possibility that has considerable merit. The
screening facilities contemplated in connection with the collection
should attract a large patronage. Footage enjoying a public domain
status should prove particularly attractive to producers if their rela-
tionship with the National Archives and The Library of Congress over
the last 2 or 3 years can be taken as a guide.

Finally, there is the consideration of pride of*craft. For example,
The Library of Congress is proposing to give the screen a comparable
recognition to that heretofore given the printed word. Such recogni-
tion has been sought by leaders in the motion picture industry for
years and Will Hays, Terry Ramsaye, Sol Bloom, and others have
been pioneers in the movement. It is also understood that members
of the Academy of Motion Picture Arts and Sciences have given
serious consideration to such a movement. It is not inconceivable
that the proposed national film library should, with the cooperation
of all concerned, some day become equal in size and usefulness to the
largest general library now in existence. No objection can be lodged
against such a possibility if such a library earns its way in terms of
service.

DISCUSSION

QUESTION: -What is the scientific significance of the proposed collection?

MR. BRADLEY : We expect to build up a great film library containing important
scientific data. I would say also that we do not intend to build up any collection
competitively; we are going to serve as a central facility and as such we hope to
implement what other film libraries are doing in Chicago, Los Angeles, New York,
and elsewhere. As a central film library becomes strong it follows that other
libraries will be correspondingly strong. In other words, we might regard our-
selves as wholesalers and the other libraries of the country as the jobbers.

There are 2 things we do not propose to do: we do not intend to get into film
production, nor do we plan to distribute film on a retail basis. We hope, however,
to accept some of the burdens of film distribution on a jobber basis corresponding

72 J. G. BRADLEY

closely to our interlibrary loan policy in respect to books. In other words, both
in the matter of film production and film distribution, we do not intend to invade
either the creative or administrative functions of others. After the agency of
origin has completed its primary or administrative distribution and after it has
turned its film over to The Library of Congress then we will exploit additional
values that may be found in the film through supplemental distribution, but on a
jobber basis. For the reason that we are not a lending library to individuals in
respect to books, neither can we lend films to individuals except, of course, in
rare cases.

QUESTION: How will pictures be selected for the Library?

Mr. BRADLEY: This is covered in some detail in the text. However, I wish to
say that the film will be selected on a referendum basis. For example, if the
Academy of Motion Picture Arts and Sciences or the various newspapers of the
country or any other interested segments of the public select film as representing
the 10 best, the 50 best, or box-office champions, then The Library of Congress will
accept such film on the basis of public acclaim. In general, we will select photo-
plays on a representative basis and factual films on a rather complete basis.

QUESTION : How does this program relate to the film program of the National
Archives?

MR. BRADLEY: The National Archives is primarily interested in the records
of the government, while The Library of Congress is interested in library material.
That is, the National Archives would have a legal interest in the negative and,
perhaps, the master positive of a given film subject, while The Library of Con-
gress would be interested in extra copies of the same film. There is no conflict
between the 2 programs.

QUESTION: How do you determine the date of selection?

MR. BRADLEY: The selection, for the most part, will be made at the end of the
year, but on newsreels and other factual films as well as factual expository films,
many of the selections will be made currently.

THE WALLER FLEXIBLE GUNNERY TRAINER*

•

FRED WALLER **

Summary. — A description is given of the equipment devised to train gunners to
hit fast-moving targets. The more important and novel features are discussed. The
trainer not only reproduces for the observer any desired environment and target, but
also correctly simulates conditions of fire in a way that otherwise could only be found
in actual combat.

A humorous slant or gag often conveys an idea better than a serious
description. James Reddig, one of the Eastman engineers, was
asked by another friend of mine how the gunnery trainer had changed
from the experimental model he had seen and what it looked like.
Jim replied, "Oh, that's easy. You take the end off the Triborough
Bridge, put four men on it with their feet dangling in the air, a con-
sole like a church organ, and behind that photocells, amplifiers, levers,
scanners, and a lot of other things that I cannot understand. Then,
take the Perisphere from the World's Fair, cut it into 4 pieces, push
the end of the Triborough Bridge into one of the pieces and you have a
Waller Gunnery Trainer. It's just as simple as that. ' '

As this description and Fig. 1 give you an idea of the size and com-
plexity of 'the machine, it is obvious that a complete analysis and
description of the apparatus cannot be given in one paper, so the
following covers the more important and novel features.

The purpose of developing this machine was to train gunners, under
realistic conditions, to estimate quickly and accurately the range of a
target, to track it, and to estimate the correct point of aim when using
noncomputing sights. To accomplish this purpose, the Waller flex-
ible gunnery trainer uses a special spherical screen process. This
process was conceived by Ralph Walker, a well-known architect, and
myself in 1938, and several years were spent in developing the ap-
paratus and overcoming the problems involved.

In June, 1940, H. Martyn Baker, an old friend of mine who is a

* Presented Oct. 15, 1945, at the Technical Conference in New York.
** Vitarama Corporation, Huntington Station, N. Y.

F. WALLER

Vol 47, No. 1

graduate of the Naval Academy at Annapolis, recognized the possibil-
ities offered by the spherical screen process in the training of gunners
to hit fast-moving targets. That was the real start of work on the
gunnery trainer.

LOWER GUN
POSITIONS "

FIG. 1 . Bird's-eye view of Waller Flexible Gunnery Trainer: — Mark 2, showing
unified assembly of elements.

The fundamental theory of the spherical screen process is that for
the average individual the perception of distance, beyond about 20
ft, is not so much the result of binocular stereopsis as it is of peripheral

July, 1946

THE WALLER FLEXIBLE GUNNERY TRAINER

vision, relative movement, size of object, and atmospheric perspec-
tive. By peripheral vision I mean what the eye sees outside of its
central area of sharp focus. This screen process simulates what the
eye normally perceives by filling a screen, shown in Fig. 2, which is a
portion of the inside of a sphere, with a motion picture. The angular
dimensions of the screen, 150 deg in the horizontal and 75 deg in the

FIG. 2. Perspective view of spherical screen upon which
scene is projected.

vertical, are nearly those encompassed by the normal human eye,
and the angular relationships of any object, fixed or moving, on the
screen are the same as those seen by the eye in actuality. Thus,
the requirements of peripheral vision and movement perspective are
satisfied.

In the photography, the size and atmospheric perspectives are re-
produced. Therefore, the observer finds himself surrounded by a
normal visual effect. The success of the curved screen process in ac-
complishing this is evident to anyone* who has ever seen it.

F. WALLER

Vol 47, No. 1

For gunnery training purposes, a picture of the desired target, say
an airplane, is produced. To the observer this target does not re-
main more or less fixed upon a single square screen covering only a

FIG. 3. Front view of trainer showing method of mounting and grouping of
guns and turrets around the projection unit.

small angle but moves within his field of vision in an entirely normal
manner thus enabling him to exercise his judgment of distance and
motion as though he were in the field.

The observer is placed behind a dummy gun, located near the
optical center of the screen, with which he attempts to hit the target.

July, 1946 THE WALLER FLEXIBLE GUNNERY TRAINER 77

By means of suitable apparatus described later, when the trigger of
the gun is pulled, and the gun is aimed so that a hit would be made,
this fact is instantly announced audibly in the gunner's earphones.
This enables the person being trained to make an immediate mental
note of the judgment and actions which led to success. In this way
the Waller flexible gunnery trainer not only reproduces for the ob-
server any desired environment and target, but also correctly sim-
ulates conditions of firing in a way that otherwise could only be found
in actual combat. Since anything can be produced on the screen
that can be photographed, and since operation of the trainer is inde-
pendent of weather, time, and the availability of actual equipment,
it offers a valuable means of training in preparation for and supple-
menting actual firing.

In order to cover a screen of 150 deg in width by 75 deg in height
with motion picture projection, it was found necessary to have 5
projectors to obtain sufficient light on the screen. This dictated the
number of cameras needed to take the pictures.

The camera consists of five 35-mm motion picture cameras syn-
chronously driven and operating as a single unit. This unit has been
kept sufficiently small and light so that it can be mounted in the gun
or turret positions available on bombers, or be used on a tripod ashore
or afloat. The cameras are arranged to cover, to the best advantage,
a spherical angle of 150 deg by 75 deg, and each one covers approxi-
mately a fifth of this total image. By operating the camera unit in a
gunner's position,. it photographs what he would see from this same
position. In the trainer, these pictures are projected on a spherical
screen of the same total angles by means of 5 projectors which are
arranged in the same relative positions as the cameras. They repro-
duce the picture as photographed, that is, as if the gunner had been
in the same position which the camera occupied.

The Screen. — The screen is shown as a section of a hollow sphere
of 20-ft radius. The supporting framework is made of plywood,
/-beams and intercostals. The frame is covered with preformed
plywood panels that are screwed in place. The projection surface of
the screen is given a special semispecular finish which reflects light
principally to the center, where the gunners are placed. By doing
this, it minimizes the degradation of the projected images by cross re-
flection from one part of the screen to another.

Arrangement of Dummy Guns. — Placed at even distances around
the center line of the projectors and the center of the screen are 4

78 F. WALLER Vol 47, No. 1

dummy guns shown in Fig. 3. Each gun is mounted on a heavy
tubular mount and is free to train and elevate so as to cover the
screen. On the outside of the mount is a bearing for a seat which
slides on 2 tubes so that it may be adjusted for men of different
heights. The seat swings on a horizontal axis and is supported by
heavy spiral springs which are also adjustable for varying weights of
men. The seating arrangement gives full flexibility so that a gunner
can keep his eye in line with the sight.

Each gun is provided with a pair of handles, the right one contain-
ing a trigger. When the trigger is pulled, the handles are vibrated
by a pair of motors in the dummy gun, simulating the recoil of a 50-

:m ."'"•' ."^ • •...'-. ..••• - x' ' .-.. ,..

FIG. 4. Sighting the target through the Mark 9 gun sight.

caliber gun. The instructor can disconnect the vibrator circuit if he
so desires. On each gun in the original model is mounted a Mark 9
collimator sight. Fig. 4 shows what the gunner sees when he looks
through his sight at the target in the field.

Subsequently, the trainer has been adapted to train men for sight-
ing with Sperry and Martin waist turrets, Sperry ball turrets, G. E.
fire control stations, as well as several different mountings for 50-
caliber and 20-mm guns, with and without lead-computing sights,
and the Navy Mark 51 Director. Various combinations of these devices
were installed on individual trainers as required. For training crews
for the B-29's, 3 pedestal-type G. E. Directors and one ring-type di-
rector are used.

July, 1946 THE WALLER FLEXIBLE GUNNERY TRAINER

Firing at Target. — As the gunner looks through the sight and fol-
lows the target the resulting movement of the gun, in train and in
elevation, is transmitted, as shown in Fig. 5, by a pair of light

HORIZONTAL
SHEAVE

FIG. 5. Schematic drawing showing flexible steel cable system from gun to

flexible steel cables running over ball-bearing rollers to train and
elevate lever arms in the unit called the "register." Each pair of
cables is kept under constant balanced spring tension so that stretch

F. WALLER

Vol 47, No. 1

or expansion and contraction from temperature changes have no ef-
fect on its accuracy.

The lever for train and the lever lor elevation each connect with
opaque masks having a transparent pattern of 2 fine lines. These
masks slide horizontally across the face of an aim scoring film in the
register unit associated with each gun. Figs. 6, 7, and 8 show de-
tails of linkage and scanner bars and masks.

i>I

ftl

FIG. 6. Close-up of sector arm to scanner bar assembly.

Scoring Mechanism in Register. — The function of the register
unit is to determine whether the gun is aimed at any instant to hit
the target plane. It is this unit which receives the existing train
and elevation from the gun, and if the aim is correct, it provides the
means of sending electrical impulses to the instructor's console where
the hit recording counter for each gun is located.

The register is similar to one of the screen projectors and the film
used in the register operates in step with the screen picture films —
at a speed of 24 pictures per sec. The film used in the register is not a
picture film, as may be seen from Fig. 8, but is a hit scoring film spe-
cially prepared as described under the scoring machine. There is a
frame of register film to correspond to each frame of picture film.

July, 1946

THE WALLER FLEXIBLE GUNNERY TRAINER

The register film is opaque and on each frame of the film are small
transparent areas. The areas are so spaced that they represent the
position at which the gun should be aimed to hit the target in the
position shown by the corresponding picture frame. When the gun
is pointed at the correct point of aim, transparent areas of the masks
which the levers move will then register with the transparent areas
in the film. This allows the light in the register projector to be trans-
mitted to a photocell which, through an amplifier and relay, actuates
the hit counter mounted in the instructor's console.

FIG. 7. Close-up view of scanner bar and gate assembly.

Each time the gunner pulls the trigger one burst for his gun is re-
corded on the corresponding burst counter on the instructor's con-
sole. At the same time, the bullet counter for his gun will record the
number of bullets that would be fired during the length of time in
which he holds the trigger down. If the gunner has his gun pointed
at the correct point of aim when he pulls the trigger, he will hear a
high -pitched tone in his earphones instantaneously and he will score
as many hits as the number of bullets fired while he maintained the
correct point of aim. If the gunner is not on the correct point of aim

F. WALLER

Vol 47, No. 1

when he pulls the trigger, he will still score the burst and the bullets
fired but no hits.

On trainers adapted for devices where range and aim are fed in
separately, 5 counters are used. The fourth records the number of

fLOWEB) TRAIN SCANNER
LEFT SIDE OF SCRETN^

RIGHT SIDE OF SCREE^
BAND FOR UPPER
FILM PATTERNS

FILM GUIDE EDGE

ELEVATION-TOP OF SCREEN
TRAIN-CENTER OF SCREE

i-ELEVATION-i

BAND FOR [-1

FILM PATTERNS

FIG. 8. vSchematic drawing showing register film in relation to scanner bars
and method of registering film for scoring.

bullets fired while the gunner is putting in the correct range, and the
fifth the number of bullets fired while he is aiming correctly. On
these trainers the hit counter scores only when both range and aim
are scored simultaneously.

July, 194G THE WALLER FLEXIBLE GUNNERY TRAINER 83

In order that the scoring on this machine may represent actual com-
bat conditions, the scoring film in the register unit may have differ-
ent-size transparent areas to allow for different-size vulnerable areas
and targets, or to allow for the area of the cone of gun dispersion. These
areas may be placed on the scoring film at different distances apart.
If they are on every other frame, they will record hits at the rate of
720 per min; if on every fourth frame, at 3GO per min, etc. By using
this method, the gunner will not only score hits in proportion to his
accuracy of aim but he will also score hits in the proportion which the
vulnerable area of the target is to the area of the cone of fire at the

FIG. 9. Close-up view of the control panel on the instructor's console.

distance of the target. That is, if the target at 600 yards has a
vulnerable area of 30 sq ft, and the area of the cone of fire is 300 sq ft,
only one- tenth of the number of bullets fired would hit the target.
Both of these factors are taken into consideration and the scoring
gives a real indication of the man's ability as a gunner. Some
branches of the Armed Services desired to omit these features and
for them a constant angular tolerance of aim was used and all hits re-
corded. By using film for the production of the register bands,
great flexibility is obtained.

Instructor's Console. — The instructor's console is mounted to the
upper forward portion of the structural framework, above the 5
projectors. From his position at the console, shown in Fig. 9, the

84 F. WALLER Vol 47, No. i

instructor may view the entire screen and the 2 upper gun positions.
The console has the following general controls and indicators:

(a) A running time meter to indicate total hours the trainer has been run.

(b) Switch to illuminate counters when house lights are off.

(c) A "still" button which allows instructor to hold pictures on screen for 5
min so that he may point out errors students are making in their aim.

(d) A "start" button.

(e) A "stop" button.

(/) Switch to permit talking to all 4 students at once.

(g) Projection room signal light to indicate "ready."

(h) Switch to communicate with office or projection room.

(i) Microphone head to contact students or projection room.

(j) Intercommunicating telephone with projection room.

In addition, 3 counters and the following controls are provided for
each of the 4 gun positions :

(a) Total number of rounds fired.

(&) Number of bursts fired.

(d) Pilot light which flashes as each hit is scored.

(e) Microphone switch so instructor can talk to individual gunner.

(/) Vibrator switch by which vibrators on individual guns may be dis-
connected.

(g) Aim projector switch which shows a one-inch diameter ring of light
at place on screen at which the gun is aimed. The student does not see this light,
since the central spot in his own sight covers it up but it allows the instructor to
see where that student is aiming. The ring is projected on the screen by small
projector mounted on the gun.

(h) Control knob, to return counters to zero.

(i) Pilot lights to indicate switches "on" or "off."

The Scoring Machine. — For the preparation of the aim scoring
films which are used in the registers, we have developed a special
machine called the scoring machine. This machine consists of the
following parts:

(a) Five projectors for projecting the films one frame at a time.

(b) A 10-ft radius screen laid off in degrees of train and elevation on which
these pictures are projected and analyzed.

(d) Four cameras connected to these pointers for making the original nega-
tives from which the scoring films are printed.

On this scoring machine, the pictures are analyzed and a plot is
made showing the range for each frame of film, that is, each J/24 sec.

July, 1940

THE WALLER FLEXIBLE GUNNERY TRAINER

Simultaneously, a record is made of the angle of the target, in both
train and elevation, in relation to the gun-carrying plane. A record
is kept by the camerman, who makes the original negative, of the
air speed and altitude of the gun-carrying plane. With all this in-
formation, the Aberdeen Tables give us the time of flight of the bullet.
Knowing how many twenty-fourths of a second it will take the bullet

-TRAIN

^AZIMUTH)

LEFT SIDE OF SCREEN!

(LOWER) TRAIN SCANNER BAR
AND FILM PATTERN

ICHT SIDE OF 5CREEN-

(UPPER)TRAIN SCANNER 6

AND FILM PATTERN | p— T

ric.

3t>,

SPHERICAL SCREEN

— BOTTOM ROW OK
PROJECTORS

FIG. 10.

Schematic plan showing method of projecting mosaics oil
the screen.

to reach the target, we then aim the 4 pointers, which represent the
4 gun positions, at the picture which is that many twenty- fourths of
a second, or frames of film, later. This gives the correct angular
lead. By then displacing each pointer the number of gunnery mils
in both train and elevation which the tables give as the ballistic cor-
rections, we have the correct point of aim.

By a system of cables and levers similar to those connecting the
gun and its register, each pointer is connected to a scoring camera.

86 F. WALLER Vol 47, No. l

This camera makes a master negative scoring band for its gun. When
the scoring band is run. in synchronism with the picture print from
which it has been made, it will record, to within a few gunnery mils,
the correct point of aim for the moving target shown on the picture
film.

Projector Unit. — The projector unit is a group of 5 Century pro-
jectors operating in synchronism as a single mechanism. All are of
the same mechanical design and each projects a portion of the whole
picture upon the screen.

The optical axis of each projector passes through a common point,
the focal point of the screen, and radiates to 5 different areas on the
screen, as illustrated in Fig. 10.

The projectors are designed for use with 35-mm motion picture
film, operating at 24 frames per sec. The running time of a 3000-
ft reel is approximately 33 min.

A heat shield, which is a circular heat-absorbing Aklo glass filter
and associated mechanisms, is located close to the rear wall of the pro-
jector. It is mounted on a counterbalanced pivoted arm and oper-
ated by a solenoid in conjunction with a limit switch which is related
to a sequential circuit. The purpose of the heat shield is to absorb
and reduce the heat at the aperture in the film trap and protect the
film when still pictures are being projected. Incidentally, during
this period the radiant heat from the lamp is also automatically re-
duced by dimming to a degree but still providing sufficient light for
still picture projection.

When the solenoid operating the heat shield is energized, it pulls
the heat shield into its place just back of the condensing lens assembly,
where it absorbs a portion of the radiant heat from the lamp before it
reaches the film.

Lamp House. — A cylindrical lamp house is mounted on the rear
of the shutter guard. A large hinged door extending half way
around the housing permits access to the interior for lamp replace-
ments and optical adjustments. It accommodates a 2100-w incan-
descent filament lamp as the light source for the projector.

An inner tube, which surrounds the lamp, is part of the cooling
system. It serves to direct an air stream all around the lamp house
to carry away the heat generated by the lamp.

Cooling System. — The system of forced air circulation in the lamp
house has been devised to remove normal generated heat that
would endanger the film in the projector and to provide cool operat-

July, 1946 THE WALLER FLEXIBLE GUNNERY TRAINER &7

ing conditions. It includes the assembly of distribution ducts and
air tubes associated with the projector. They in turn are con-
nected to an air supply and exhaust system provided in the building.

The air is forced in at the top of the lamp house, streahis past the
full length of the lamp all around the bulb as directed by the air tube
within the house, and the heat is carried off through the exhaust at
the bottom of the house to be dissipated at a distant point.

An additional cooling system is provided for the film. It is devised
to force a high- velocity sheet of air downward on both sides and over
the entire surface of the film in the film trap. The air is distributed
through a forked inlet pipe connection on the driving side of the pro-
jector and passes through the center wall to the film or operating
side. One tube leads to the nozzle on the film trap, the other to noz-
zle on the film gate. After the cooling curtains of air flow past the
film surfaces, they circulate about in the immediate vicinity of the
mechanisms in the projector.

Main Drive. — Each projector is driven by its own electric motor,
but all motors are mechanically coupled together and held in syn-
chronism by an arrangement of beveled gears on a common syn-
chronizing shaft that keeps all projectors running at a speed of 24
frames of film per sec.

Framing Motor. — An additional motor called the framing motor,
mechanically connected with the synchronizing shaft, serves to
bring the projector mechanisms to a stop with the film in frame and
shutters open so that the still picture projected is properly composed
on the screen.

Photoelectric Controls. — The photoelectric controls of the trainer,
designed by W. Robert Dresser, are most elaborate and although
not covered here could easily be the subject of an entire paper.

60th SEMIANNUAL

TECHNICAL CONFERENCE

HOLLYWOOD-ROOSEVELT HOTEL
Hollywood, California

OCTOBER 21-25, 1946

Officers in Charge

D. E. HYNDMAN President

HERBERT GRIFFIN Past-President

L. L. RYDER Executive Vice-President

M. R. BOYER Financial Vice-President

J. A. MAURER Engineering Vice-President

A. C. DOWNES A Editorial Vice-President

W. C. KUNZMANN Convention Vice-President

C. R. KEITH Secretary

E. I. SPONABLE Treasurer

General Office, New York

BOYCE NEMEC Engineering Secretary

HARRY SMITH, JR Executive Secretary

Directory of Committee Chairmen

Pacific Coast Section and Local Ar-
rangements H. W. MOYSE, Chairman

Papers Committee C. R. DAILY, Chairman

BARTON KREUZER, Vice-

Chairman

Publicity Committee HAROLD DESFOR, Chair-
man

Registration and Information W. C. KUNZMANN, Chair-
man, assisted by C. W.
HANDLEY

Luncheon and Dinner-Dance Commit-
tee L. L. RYDER, Chairman

Hotel and Transportation Committee S. P. SOLOW, Chairman

TECHNICAL CONFERENCE S9

Membership and Subscription Commit-
tee H. W. REMERSCHEID, Chairman

Ladies Reception Committee Hostess MRS. H. W. MOYSB

Projection Program— 35-mm W. V.WOLFE, Chairman, assisted

by Members Los Angeles
Locals 160 and 165
16-mm H. W. REMERSCHEID

HOTEL RESERVATIONS AND RATES

The Hollywood-Roosevelt Hotel, Hollywood, Calif., will be the Conference
Headquarters, and the hotel management extends the following per diem room
rates, European plan, to SMPE members and guests:

Room with bath, one person $4.40-5.60

Room with bath, two persons, double bed $5.50-6.60

Room with bath, two persons, twin beds $6.60-7.70

Desired accommodations should be booked direct with Stewart H. Hathaway,
Manager of the hotel, who advises that no parlor suites will be available unless
confirmed by him. All reservations are subject to cancellation prior to October
14, and no reservations will be held after 6:00 p.m. on the anticipated date of arrival
unless the hotel management has been advised otherwise.

HOUSING COMMITTEE

An acute housing condition exists in Hollywood and it is expected that most
of the available reservations at the Hollywood-Roosevelt Hotel will have been
taken by the time this issue of the JOURNAL reaches the membership. In order
to be of assistance to members, desiring room accommodations, the Pacific Coast
Section has set up a Housing Committee under the Chairmanship of Past- Presi-
dent Herbert Griffin.

The Housing Committee expects to mail a return post card to all members out-
side of the Hollywood area on which the member may state whether he desires
room accommodations and for what length of time. The returned cards will be
checked against available reservations and an effort will be made to place Eastern
and Midwestern members who plan to attend the Conference. However, the
demand is very apt to exceed the supply and reservations will be made on the
basis of "first come, first served." It will be of assistance to all concerned to have
the cards returned as quickly as possible.

RAIL, PULLMAN, AND AIR ACCOMMODATIONS

SMPE members and guests who have received confirmed room reservations,
should then consult local transportation agents as early as possible, and book their
desired transportation accommodations immediately.

REGISTRATION

The Conference Registration Headquarters will be located in Room 201 on the
mezzanine floor of the hotel, where Luncheon and Dinner-Dance tickets can be
procured prior to the scheduled dates of these functions. Members and

90 TECHNICAL CONFERENCE Vol 47, No. l

guests are expected to register. The fee is used to help defray Conference
expenses. e

BUSINESS AND TECHNICAL SESSIONS

Day sessions will be held in the hotel, and evening sessions at locations away
from the hotel, which will be listed in the preliminary, and final printed Confer-
ence programs. ^

Authors who are planning to present papers at the 60th Semiannual Technical
Conference should mail the title of their paper to the West or East Coast Chair-
man of the Papers Committee, or to the Society's New York Office, as soon as
possible. As a prerequisite to inclusion on the program, authors' abstracts must
be received by the Papers Committee by Sept. 1. Complete manuscripts must
be submitted by Oct. 1, 1946. Only through your cooperation can a preliminary
program be drafted early enough for publication in the industry trade pape'rs
and mailing to the membership at least a month prior to the Conference.

GET-TOGETHER LUNCHEON AND DINNER-DANCE

The Society will again hold its regular pre-war social functions and accordingly
a Get-Together Luncheon is scheduled in the California Room of the hotel on
Monday, October 21, at 12:30 P.M. The luncheon program will be announced
later. Members in Hollywood and vicinity will be solicited by a letter from S. P.
Solow, Secretary of the Pacific Coast Section, to send remittances to him for the
Conference registration fee and luncheon tickets. Ladies are welcome to attend
the luncheon.

The 60th Semiannual Dinner -Dance will be held in the California Room of the
hotel on Wednesday evening, October 23, at 8:30 P.M. Dancing and entertain-
ment. (Dress optional.) A social hour for holders, of Dinner-Dance tickets will
precede the Dinner-Dance between 7: 15 P.M. and 8: 15 P.M. in the Hotel Terrace
Room (Refreshments).

LADIES' PROGRAM

A reception parlor for the ladies' daily get-together and open house with Mrs.
H. W. Moyse as hostess will be announced on the hotel bulletin board and in the
final printed program.

Ladies are welcome to attend technical sessions of interest, also the Luncheon
on October 21, and the Dinner-Dance on October 23. The Conference badge and
identification card will be available to the ladies by applying at Registration
Headquarters.

The ladies' entertainment program will be announced later.

MOTION PICTURES AND RECREATION

The Conference recreational program will be announced later when arrange-
ments have been completed by the local committee. Identification cards issued
only to registered members and guests will be honored at the deluxe motion picture
theaters on Hollywood Boulevard. Those desiring other recreation during the
Conference should consult the hotel bulletin board or inquire at Registration
Headquarters.

July, 1946 TECHNICAL CONFERENCE (.U

Monday, October 21, 1946

Open Morning.
10:00 a.m. Room 201, Hotel Mezzanine Floor: Registration. Advance sale of

Luncheon and Dinner-Dance tickets.
12:30 p.m. California Room: SMPE Get-Together Luncheon.

Program announced in later bulletins.
2: 00 p.m. Aviation Room, Hotel Mezzanine Floor: Opening business and

Technical Session.
8:00 p.m. Evening Session: Location to be announced later.

Tuesday, October 22, 1946

Open Morning.
10:00 a.m. Room 201, Hotel Mezzanine Floor: Registration. Advance sale of

Dinner-Dance tickets.

2: 00 p.m. California Room: Afternoon Session.
8: 00 p.m. Evening Session: Location to be announced later.

Wednesday, October 23, 1946

9: 30 a.m. Room 201, Hotel Mezzanine Floor: Registration. Advance sale of

Dinner-Dance tickets.
10: 00 a.m. California Room: Morning Session.

Open Afternoon.
7: 15 p.m. Hotel Terrace Room: A social hour for holders of Dinner-Dance

tickets preceding the Dinner-Dance (Refreshments).

8:30 p.m. California Room: 60th Semiannual Technical Conference Dinner-
Dance. Dancing and entertainment. Program will be an-
nounced later.

Thursday, October 24, 1946

Open Morning.

1:00 p.m. Room 201, Hotel Mezzanine Floor: Registration.
2:00 p.m. California Room: Afternoon Session.
8: 00 p.m. Evening Session. Location to be announced later.

Friday, October 25, 1946

Open Morning.

2: 00 p.m. California Room: Afternoon Session.

8:00 p.m. Evening Session. Adjournment of the 60th Semiannual Technical
Conference. Location to be announced later.'

Note: All sessions during the 5-day Conference will open with an interesting
motion picture short.

92 SOCIETY ANNOUNCEMENTS Vol 47, No. 1

Important

Because of the existing food problem, your Luncheon and Dinner-Dance
Committee must know in advance the number of persons attending these func-
tions in order to provide adequate accommodations.

Your cooperation in this regard is earnestly solicited. Luncheon and Dinner-
Dance tickets can be procured from W. C. Kunzmann, Convention Vice-President,
during the week of October 13 at the Hollywood-Roosevelt Hotel.

All checks or money orders for Conference registration fee, Luncheon and
Dinner-Dance tickets should be made payable to W. C. Kunzmann, Convention
Vice-President, and not to the Society.

W. C. KUNZMANN
Convention Vice-President

SOCIETY ANNOUNCEMENTS

MIDWEST SECTION MEETING

A large audience of Midwest Section members and guests was addressed by
Frank E. Carlson of General Electric Company, Nela Park, Cleveland, who dis-
cussed "Tungsten Filament Light Sources" at a meeting held in the Paramount
Preview Theater, Chicago, June 20.

Future improvements in tungsten sources for projection will not supply such
gains and output as achieved previously through higher efficiency. Mr. Carlson
suggested that engineers should choose the correct source size for a particular
system of projection optics by making practical use of the reversibility of the
optical systems. Diffuse illumination of the objective will allow the image size
to be measured at the filament position.

A summary of the influence of filament coiling on sound reproduction systems
was given.

The meeting concluded with the showing of selected portions of the German
Agfa negative-positive process color film, The Golden City. Discussion of the film
brought out the fact that the cyan color was apparently suppressed either for
esthetic reasons or because of technical difficulties. The increased resolution of
the print was favorably received.

The first meeting of the fall series is scheduled for September 12, at 8:00 P.M.
in Western Society of Engineers' Hall, 205 West Wacker Drive, Chicago. All
Society members in the area who are not now receiving Midwest Section meeting
notices should communicate with Robert E. Lewis, Secretary-Treasurer, Armour
Research Foundation, Chicago 16, 111.

PACIFIC COAST SECTION MEETING

Captain E. M. Senn, U. S. Navy, and Captain William C. "Bill" Eddy, U. S.
^Slavy, retired, addressed a joint meeting of the Pacific Coast Section and the
Institute of Radio Engineers held on June 10 in the Walt Disney Studio Theater,

July, 1946 SOCIETY ANNOUNCEMENTS 93

Burbank, Calif. The meeting was opened with a screening of an interesting Navy
film showing the use of radar in naval engagements.

Captain Senn described the Navy's extensive electronics training course, which
is accredited at Purdue University for 2 years toward an electrical engineering
degree. He also pointed out the Navy's serious need for qualified young men to
maintain Navy electronic equipment.

Captain Eddy discussed some of the wartime electronic devices which were so
vital in bringing the war to its successful conclusion. He conducted an open
forum during his address answering many interesting questions.

Captain Eddy has returned to his civilian activity as director of the Balaban
and Katz television station in Chicago, and gave a description of the new 700-ft
antenna structure now under construction.

A large number of Navy personnel swelled attendance of the combined meeting
to over 350.

EMPLOYMENT SERVICE

POSITIONS OPEN

Position available for Optical Designer, capable of handling the calcula-
tion and correction of aberrations in photographic and projection lens
systems. Junior designers or engineers will be considered. Write
fully giving education, experience, and other qualifications to Director
of Personnel, Bell and Ho well Company, 7100 McCormick Road, Chi-
cago 45, 111.

Motion picture studio in Bombay, India, has positions open for profes-
sional motion picture camerman with studio and location experience;
sound recording engineer experienced in installation, maintenance and
operation of recording equipment; motion picture processing labora-
tory supervisor; and professional make-up artist. Five-year contracts
at favorable terms are offered to those qualified. Write or cable direct
to Personnel Manager, Dawlat Corporation Ltd., Patel Chambers, French
Bridge, Bombay 7, India, giving experience, etc., in detail.

New film production unit to be located at Athens, Georgia, needs film
editor-writer and film director. Experience in 16-mm as well as 35-mm
production desirable. Southern background or interest in South pre-
ferred but not essential. Write giving full details of experience, etc., to
Nicholas Read, The National Film Board, Ottawa, Canada.

POSITIONS WANTED

Projectionist-newsreel editor with 15 years' experience just released
from service. Willing to locate anywhere. Write P. O. Box 152, Hamp-
den Station, Baltimore 11, Maryland.

Honorably discharged veteran with 10 years' experience in projection
and installation of projection and sound equipment, both for booth and
back-stage. Prefer to locate in California, Oregon or Nevada. For ad-
ditional details write F.A.N., Box 113, Holley, Oregon.

Cameraman, honorably discharged Army veteran, 'desires re-enter indus-
trial, educational production with independent producer or studio. Ex-
perienced in 35- and 16-mm color and black-and-white. References
and complete record of experience available. Write, wire or telephone
T. J. Maloney, 406 Oak St., Ishpeming, Mich. Telephone 930.

94 SOCIETY ANNOUNCEMENTS

Sound Recordist. Former Signal Corps sound instructor and Army
Pictorial Service newsreel recordist-mixer, 35-mm equipment. Honor-
ably discharged veteran, free to travel. Write Marvin B. Altman, 1185
Morris Ave., New York, N. Y. Telephone Jerome 6-1883.

16-mm Specialist. Honorably discharged veteran with many year's
ezperience, specializing in 16-mm. Linguist. Available for special
assignments. Write J. P. J. Chapman, ARPS, FRSA, The Huon,
Branksome Hill Road, Bournemouth, England.

Cameraman. Veteran honorably discharged from Air Force Motion
Picture Unit desires to re-enter industrial, documentary, or educational
film production. Experienced in 35- and 16-mm, sound, black-and-
white and color cinematography. Single, willing to travel. Write S.
Jeffery, 2940 Brighton Sixth St., Brooklyn 24, N. Y. Telephone Dewey
2-1918.

JOURNAL OF THE SOCIETY OF
MOTION PICTURE ENGINEERS

Vol 47 AUGUST, 1946 No. 2

CONTENTS

PAGE
Reports of SMPE Committees :

Report of the Committee on Motion Picture Instruc-
tion 95

Report of the Committee on 16-Mm and 8-Mm Mo-
tion Pictures 107

Report of the Committee on Standards 110

Report of the Committee on Studio Lighting 113

Report of the Committee on Television Projection

Practice 118

Motion Pictures Tomorrow W. F. RODGERS 120

Citations to Thomas Armat and Warner Brothers 124

A Simplified Recording Transmission System

F. L. HOPPER AND R. C. MOODY 132

The Photometric Calibration of Lens Apertures

A. E. MURRAY 142

A New Film for Photographing the Television Monitor

Tube C. F. WHITE AND M. R. BOYER 152

Television Reproduction from Negative Films

E. MESCHTER 165

Current Literature 182

60th Semiannual Convention 184

Indexes to the semiannual volumes of the JOURNAL are published in the June and December
issues. The contents are also indexed in the Industrial Arts Index available in public libraries.

JOURNAL

OF THE

SOCItTY of MOTION PICTURE CNGINtEPiS

MOT«L PfHNSYLVANIA • NCW YOP.K I. N-Y- • TCL. PtNN. 6 O62O

HARRY SMITH, JR., EDITOR

Board of Editors
ARTHUR C. DOWNES, Chairman

I. CRABTREE ALFRED N. GOLDSMITH EDWARD W. KELLOGG
YDE R. KEITH ALAN M. GUNDELFINGER CHARLES W. HANDLEY
ARTHUR C. HARDY

Officers of the Society
^President: DONALD E. HYNDMAN,

342 Madison Ave., New York 17.
*Pasl-P resident: HERBERT GRIFFIN,

133 E. Santa Anita Ave., Burbank, Calif.
^Executive Vice-President: LOREN L. RYDER,

5451 Marathon St., Hollywood 38.
** Engineering Vice-President: JOHN A. MAURER,

37-01 31st St., Long Island City 1, N. Y.
*Editorial Vice-President: ARTHUR C. DOWNES,

Box 6087, Cleveland 1, Ohio.
**Financial Vice-President: M. R. BOYER,

350 Fifth Ave., New York 1.
* Convention Vice-President: WILLIAM C. KUNZMANN,

Box 6087, Cleveland 1, Ohio.
^Secretary: CLYDE R. KEITH,

233 Broadway, New York 7.
^Treasurer: EARL I. SPONABLE,

460 West 54th St., New York 19.

Governors

*fFRANK E. CAHILL, JR., 321 West 44th St., New York 18.
**FRANK E. CARLSON, Nela Park, Cleveland 12, Ohio.
**ALAN W. COOK, Binghamton, N. Y.

*JOHN I. CRABTREE, Kodak Park, Rochester 4, N. Y.

*CHARLES R. DAILY, 5451 Marathon St., Hollywood 38.
**JOHN G. FRAYNE, 6601 Romaine St., Hollywood 38.
**PAUL J. LARSEN, 1401 Sheridan St., Washington 11, D. C.
**WESLEY C. MILLER, Culver City, Calif.

*PETER MOLE, 941 N. Sycamore Ave., Hollywood.
"JHoLLis W. MOYSE, 6656 Santa Monica Blvd., Hollywood.

*WILLIAM A. MUELLER, 4000 W. Olive Ave., Burbank, Calif.
*°A. SHAPIRO, 2835 N. Western Ave., Chicago 18, 111.

*REEVE O. STROCK, 111 Eighth Ave., New York 11.

Term expires December 31, 1946. tChairman, Atlantic Coast Section.
**Term expires December 31, 1947. jchairman, Pacific Coast Section.
*° Chairman, Midwest Section.

Publication Office, 20th & Northampton Sts., Easton, Pa.

General and Editorial Office, Hotel Pennsylvania, New York 1, N. Y.

Entered as second-class matter January 15, 1930, at the Post Office at Easton,

Pa., under the Act of March 3, 1879.

JOURNAL OF THE SOCIETY OF
MOTION PICTURE ENGINEERS

Vol 47 AUGUST, 1946 No. 2

REPORTS OF SMPE COMMITTEES

REPORT OF THE COMMITTEE ON MOTION PICTURE
INSTRUCTION*

JOHN G. FRAYNE**

The Committee on Motion Picture Instruction was formed in
October 1945, with the aim and purpose of providing the Society with
a list of the various institutions, colleges, and universities that teach
courses on motion pictures. The Society would then list these sources
for education in motion pictures, enumerating courses that are taught
at each institution, thereby permitting the Society to furnish con-
crete information in answer to inquiries from individuals seeking in-
formation on where such courses and instruction might be obtained.

After some preliminary correspondence between the members, a
questionnaire was drawn up and approved for submission to educa-
tional institutions of technical school or college grade. A copy of the
questionnaire is attached as an appendix to this report. It will be
noted that it requested information from each institution on the sub-
jects listed below :

(1) Cinematography (including color)

(2) Photography (including color)
. (3) Sound Recording

(4) Motion Picture Film Editing

(5) Motion Picture Projection

(6) Motion Picture Distribution

(7) Economic Problems in Motion Picture Production and Exhibition

(8) Film Processing— still

(9) Film Processing — motion picture
(10) Miscellaneous

The questionnaire was sent to 155 institutions of higher learning
scattered throughout the United States. To date, replies have been

* Presented May 8, 1946, at the Technical Conference in New York.
** Chairman.

REPORTS OF SMPE COMMITTEES

Vol 47, No. 2

received from 102, or approximately 66 per cent of those sent out. Of
this number, 60 were universities, 32 were colleges, 8 were technical
schools, one was a military academy, and one was an Armed Forces
institute. The information has been broken down into courses which
are listed below in the order of their appearance on the questionnaire :

CINEMATOGRAPHY (INCLUDING COLOR)

Name of College

Courses

Semester Hours

Credits

New York University
Dept. of Motion Pict.
Washington Square
New York 3, N. Y.

Motion Pictures 1-2

3 per wk

Oregon State College
Corvallis, Ore.

Educational Cinema-
tography (Summer
Session only)

3 term hr
(2sem.hr)

(2)

University of Denver
Denver, Colo.

Motion Picture Mak-
ing

2 quarter hr

Ohio State University
Columbus, Ohio

Cinematography

University of Southern
California
Dept. of Cinema
3551 University Av.
Los Angeles, Calif.

Cinema 115ab, Cinema
16 Sab, and others
(see bulletin)

1st yr, 8 hr
per wk
2nd yr, 8 hr
perwk

8 units
8 units

PHOTOGRAPHY (INCLUDING COLOR)

Iowa State College
Ames, Iowa

1. Physics 316
2. Physics 650

1. 4'/2
2. 3 or more

These 2 courses
are to de-
velop pho-
tography in
scientific
fields

University of Oregon
Eugene, Ore.

Rudiments of Photo-
graphic Journalism

i*/i

2 term hr

Oberlin College
Oberlin, Ohio

In Dept. of Chemistry
(Photography)

Lab.— 1 or 2
3 -hr periods
per wk

2 or 3 hr

Baylor University
Waco, Tex.

Photography

3V.

University of Detroit
Detroit, Mich.

P41b

3 hr or more
per wk for
1 sem.

University of Minne-
sota
Minneapolis, Minn.

News Photography
Photography

3 quarter hr
perwk
5 quarter hr
perwk

3
3

Gustavus Adolphus
College
St. Peter, Minn.

211 Photography
212 Advanced Photog-

%
2

raphy

Aug. 1946

MOTION PICTURE INSTRUCTION

Name of College

Drake University
Des Moines, Iowa

College of Emporia
Emporia, Kan.

University of Colorado
Boulder, Colo.

Brigham Young Uni-
versity
Provo, Utah

State College of Wash-
ington
Pullman, Wash.

Northwestern Univer-
sity

Medill School of
Journalism

Evanston, 111.

New York University
Dept. of Motion Pict,
Washington Square
NewYork3,N.Y.

Colgate University
Hamilton, N. Y.

Miami University
Oxford, Ohio

Oregon State College
Corvallis, Ore.

Courses
News Photography

Semester Hours
2

University of Idaho
Moscow, Idaho

Photography in Phys- Sem. I — 2 hr
ics Dept.

Photography 2

Photography
Elementary
Advanced

Photography — in planning stage at present

Elements of Photog- 4 quarter hr

raphy
Press Photography 4 quarter hr

St. Olaf College
Northfield, Minn.

Motion Pict. 3-4

Photography

Elementary Photog-
raphy (Still)

Advanced Photogra-
phy (Still)

Ph 161— Rudiments of
Photography

Ph361 (Hand Camera)
Ph 362 (Commercial)

Ph 363 (Composition
Enlarging)

Ph 461, 462, 463— Ad-
vanced Photography
(Color, Photomicrog-
raphy, Microscopic
Motion Pictures)

"Photographic Tech-
nique" (Zool. 151-
152}. Does not in-
clude motion picture
but includes color-
correct photography
and some color pho-
tography

Photography & Art

2 per wk

Sem. I -- 3

hr
Sem. II — 2

4hr

Credit*
2

2hr
2

2 term hr

(1.33 sem.

hr)

(1.33)

3 term hr

(2 sem. hr)

(2)

3 term hr

(2 sem. hr)

(2)

3 term hr

(2sem.hr)

(2)

3 term hr

(2sem.hr)

(2)

2 sem.

REPORTS OF SMPE COMMITTEES

Vol'47, No. 2

Name of College

Ohio State University
Columbus, Ohio

University of Southern

California
Dept. of Cinema
3551 University Ave.
Los Angeles, Calif.

Courses
3 courses

Cinema 90, 91, 92;
Cinema 1 21ab

Cinema 50ab will re-
place 90, 91 (Color-
begins 1946)

Semester Hours Credits

3 (each course)

3 (each
course)

Each course 4
hr per wk
7 hr per wk

2 units
course
3 units

per

It will be noted that 21 schools report courses in photography.
These appear to range all the way from elementary photography
through news photography to the more scientific aspects of the
subject, such as photomicrography. It will be noted, however, that
in none of the replies is there any indication that photography is con-
sidered as a major subject leading to a degree. On the other hand, it
appears to be a minor subject associated with a wide variety of major
courses.

SOUND RECORDING

Courses Semester Hours

Sound Recording Part of course

in public
speaking

None, except in courses
offered in radio tech-
niques

Motion Pictures 9-10 2 per wk

Name of College Courses Semester Hours Credits

Georgia School of

Technology
Atlanta, Ga.

State College of Wash-
ington
Pullman, Wash.

New York University
Dept. of Motion Pict.
Washington Square
New York 3, N. Y.

Oregon State College Ed. 533— Correlation 3 term hr (2 3

Corvallis, Ore. of radio recordings sem. hr.) (2)

with visual aids

(Occasionally given in summer session. Occasionally
a few students are trained on an apprenticeship basis in
connection with radio station KOA C — very fragmentary.)

»
University of Southern Cinema 140 3 hr per wk 2 units

California
Dept. of Cinema
3551 University Ave.
Los Angeles, Calif.

Only five schools list any courses in sound recording, and two of
these can only be considered as dealing with operation of equipment
rather than study of the fundamental engineering problems involved
in sound recording. In fact, it is very doubtful that any strictly
technical courses in this most important field are offered in any of the
higher institutions of learning in the United States.

Aug. 1946

MOTION PICTURE INSTRUCTION

MOTION PICTURE FILM EDITING

Name of College

New York University

Dept. of Motion Pict.

Washington Square

New York, 3 N. Y.

University of Southern

California
Dept. of Cinema
3551 University A ve.
Los Angeles, Calif.

Antioch College

Yellow Springs, Ohio

Courses
Motion Pictures 31

Cinema 135

Motion Picture Film
Editing

Semester Hours
2 per wk

3 hr per wk

20 wk

Credits
2

University of Kansas
Lawrence, Kan.

Pennsylvania State
College

State College, Pa.
University of Ken-
tucky

Lexington, Ky.
New York University

Dept. of Motion Pict.

Washington Square

New York 3, N. Y.
Antioch College

Yellow Springs, Ohio

MOTION PICTURE PROJECTION
Study of the history, 2 2

technique, art, and

the social and edu-
cational significance

of the motion picture
Teach projection but None

not for credit. 16-

mm only.

Given both informally upon request of individuals and
included in educational audio-visual instructional aids
courses.
Included in Motion

Pictures 3-4

Classes conducted by students under extra-curricular
committee.

It will be noted that five institutions indicate some kind of courses
in motion picture projection. However, the courses offered either
give no university credit or are considered a part of other courses in
motion pictures or audio-visual instruction.

Name of College
Pennsylvania State

College

State College, Pa.
University of Ken-
tucky

Lexington, Ky.
New York University
Dept. of Motion Pict.
Washington Square
New York 3, N. Y.
University of Southern

California
Dept. of Cinema
3551 University A ve.
Los Angeles, Calif.

MOTION PICTURE DISTRIBUTION

Courses Semester Hours

Motion Picture Dist.

Credits

Included in graduate courses on audio-visual aids in
instruction. Commercial'distribution for entertainment
is not included.

Motion Pictures 2 per wk 4

19-20

Cinema 150

2 hr per wk

2 units

100

REPORTS OF SMPE COMMITTEES

Vol 47, No. 2

ECONOMIC PROBLEMS IN MOTION PICTURE PRODUCTION AND EXHIBITION

Name of College Courses Semester Hours Credits

New York University Motion Pictures 19-20 2 per wk 4

Dept. of Motion Pict.
Washington Square
New York 3, N.Y.

University of Denver
Denver, Colo.

University of Southern

California
Dept. of Cinema
3551 University Ave.
Los Angeles, Calif.

University of Detroit
Detroit, Mich.

State College of Wash-

ington
Pullman, Wash.

University of Southern

California
Dept. of Cinema
3551 University Ave.
Los Angeles, Calif.

Courses planned for 1946-1947 in School of Commerce

Cinema 250ab

2 hr per wk
for 2 sem.

2 units per
sem.

FILM PROCESSING — STILL

Film Processing — Still
In Planning Stage

Covered in Cinema 90,
91, and 92

As a part of the Lab. work,
etc., of General Photography

FILM PROCESSING— MOTION PICTURE

Oregon State College
Corvallis, Ore.

Included (16 mm) in
courses in Photogra-
phy

Baylor University
Waco, Tex.

Pennsylvania State

College
State College, Pa.

University of Michigan
Ann Arbor, Mich.

Kansas State College
of Agriculture &
Applied Science
Manhattan, Kan.

MISCELLANEOUS
Course in Visual Ed.

Visual Education — Ed.
423 Lab. in visual
and other aids

Ed. 424 Visual & other
sensory aids in edu-
cation

B-133 Visual-Sensory aids

in education
B-300 Research Semi-

2
1 or 2

1
1 or 2

2hr

Credit ar-
ranged

"We make a few films for public relations and instruc-
tional purposes but offer no instruction in any technical
phase of the motion picture industry, although we do
offer a course in visual instruction and we have three
courses in still photography. We would be glad to have
syllabi and other information which might be useful in
formulating plans for technical course work."

Aug. 1946

MOTION PICTURE INSTRUCTION

101

Name of College

Louisiana Polytech.

Institute
Ruston, La.

Courses

Semester Hours

Credits

"I am teaching a class in 'Audio- Visual Aids to In-
struction.' It has to do with making effective utiliza-
tion of films in the classroom. Although motion picture
production, distribution, etc., should concern itself with
the 16-mm documentary and educational film as well
as the entertainment film. I find a tendency to center all
courses around the 35.-mm 'Hollywood' entertainment
industry. This was my observation during the past
three years spent with the Army's film center in New
York and observation of courses in the colleges of that
city." — Robt. H. Mount, Dir. of Visual Instruction.

Oregon State College
Corvallis, Ore.

University of Okla-
homa
Norman, Okla.

University of Kansas
Lawrence, Kan.

Ed. 431 — Construction

and use of Visual

Aids
Ed. 531 — Organization

and supervision of

visual aids (summer

only)
I Ed. 474— Written

and visual teaching

aids.
Wild Life Photography

(a new course to be

given by Dept. of

Zoology)

Audio-visual Educa-
tion: a course de-
signed principally for
teachers

Methods and Adminis-
tration of Visual
Instr.

3 term hr (2
sem. hr)

(2)

A. & M. College of "Some interest is being developed toward use of films
Texas in connection with teaching. Committee actively work-

College Station, Tex. ing on these details — I am Chairman of that Commit-
tee."— G. B. Wilcox, Head, Dept. of Educ. & Psychology

University of Colorado
Boulder, Colo.

Otis Art Institute
2401 Wilshire Blvd.
Los Angeles, Calif.

Pasadena Junior Col-
lege

East Colorado St.
Pasadena, Calif.

Photochemical Physics

Theater & Set Design
Course

Course re- First year basic
quires 3 yr earns 31
term pt.
Total term
pt. for
course, 114

Stage Technology 2-yr course

"Students completing this course are qualified to ac-
cept positions in radio, motion pictures, television, re-
cording, and legitimate stage. In the past we have
found it most helpful to use as teaching aids advertising
materials such as graphs, illustrations, charts, data
sheets, etc., put out by the various companies supplying
these fields. This enables the students to become ac-
quainted with available equipment, its care, use, and
applications, and also to receive up-to-the-minute in-

102

REPORTS OF THE COMMITTEES

Vol 47, No. 2

Name of College

Georgia School of

Technology
Atlanta, Ga.

University of South

Dakota

Extension Division
Vermillion, S. D.

Harvard University
Cambridge, Mass.

University of Wiscon-
sin

Univ. Extension Div.
Madison 6, Wis.

Courses

Semester Hours

Credits

Cornell University
Dept. of Extension
Teaching and Info.
Ithaca, N. Y.

New York University
Dept. of Motion Pict.
Washington Square
New York 3, N. Y.

Boston University
84 Exeter St.

formation on new developments. We should appreciate
being included on your mailing list to receive such ma-
terials as you may have. If possible, we should like to
have this material in lots of 200 so that each student may
have a copy for reference."

Parts of courses in Phys-
ics and Elec. Eng.

Visual Instruction
(During summer ses-
sion)

Audio-Visual aids to 2 2

teaching

Visual Instruction: 2

"Designed to meet the needs of administrators, class-
room teachers, supervisors, and directors of visual educa-
tion. The history of visual education, relevant prin-
ciples of learning, scientific studies in the field, and
methods of evaluation are investigated. Individual
assistance is given in co-ordinating audio- visual materials
to course of study or general curriculum needs, and dem-
onstration of tested methods of classroom use of audio-
visual materials at elementary and secondary school
levels hi requested subject areas."

Local Production of Audio- Visual Materials:

"An opportunity for teachers, supervisors, and ad-
ministrators actually to produce those audio-visual
teaching materials which adapt themselves for produc-
tion in the local school situation. Following a brief
treatment of audio-visual teaching psychology and phi-
losophy, the persons enrolled will be taken through ex-
periences in developing poster files, bulletin-board dis-
plays, the construction of terrain models and filmstrip,
slide and motion picture production. Laboratory facili-
ties will be available to a maximum enrollment of 15.
People coming into the course should have their own
cameras, to be approved by the instructors. Other ar-
rangements may be made for a limited number of inter-
ested people who own no equipment but who intend to
do so."

"At present we give only one course in photography
which is more an introduction to all phases of photo-
graphic work than anything else. It is intended to give
prospective teachers an idea of the scope and methods
applicable to their teaching."

Motion Pictures 91

Visual Education —
Mgt.

2 per wk

2x/4 per wk

Aug. 1946

MOTION PICTURE INSTRUCTION

103

Name of College

Boston, Mass.

Antioch College

Yellow Springs, Ohio

Massachusetts Insti-
tute of Technol-
ogy
Cambridge, Mass.

University of Southern

California
Dept. of Cinema
3551 University Ave.
Los Angeles, Calif.

Indiana University-
Extension Division
Bloomington, Ind.

Course Semester Hours Credits

Problems in Visual 2V4perwk 3

Ed.— Mgt.

"None of our courses limit themselves to the use of the
educational motion picture although a considerable por-
tion of the courses discuss the possibility of the motion
picture in education."

Aspects of the Film

12 wk

"MIT offers no courses that exactly fit into these
categories. For example, we offer a course on color, of
which color reproduction is one important aspect. In
normal times we offer courses on photography and
cinematography but they were temporarily discon-
tinued because of the war."

A.B. Degree with
major in Cinema

M.A. Degree with
major in Cinema

See Bulletins. Related courses in physics and engi-
neering— Electronics, Recording, etc.

"At the present time, we have only one course in
audio- visual aids in production, but will break this course
down into 3 to 5 courses as our production program gets
under way at Indiana University."

Of the courses listed above as offered by the various institutions,
few, if any, can be construed as offering technical information on a
level corresponding to that of other well-established curricula. On
the other hand, they appear to come under the broad heading of
"Visual Education." Some of these courses do include the actual
making of films, which involves the artistic as well as the technological
aspects of motion picture making. Undoubtedly, students taking
these courses do obtain a certain familiarity with motion picture
equipment, especially of the 16-mm variety, but it is doubtful that
any basic training in cinematography or sound recording is included
in these visual education courses.

The results of the survey of the higher institutions of learning in
the United States by the Committee indicate very clearly that the
technology of motion picture making has not been given serious con-
sideration in institutions of this caliber. While it must be admitted
that courses in physics, chemistry, electrical engineering, electronics,
etc., offer the basic groundwork for proficiency in the various fields
of picture making, it would appear that the industry has advanced to
a point where training for it should be recognized as calling for special
instruction in our schools and colleges. It would certainly appear

104 REPORTS OF SMPE COMMITTEES Vol 47, No. 2

that it is high time for cinematography to be emerging from the
strictly "craft" classification into that of a full-fledged profession.
The modern cinematographer, it seems, should have a basic education
in physics, chemistry, illumination, photography, camera design, and
all the other elements that go into the work of a cinematographer.
The results of the questionnaire, however, show that this condition
does not exist in our American schools.

In the newer fields of motion picture activity, such as sound re-
cording, it is hardly to be expected that formal courses in this sub-
ject should be listed at this time. It may be argued that sound re-
cording calls for training in the basic physical and electronic sciences,
and as such should be covered in the regular courses given in this
field. To those of us who are familiar with the ramifications of mod-
ern sound recording for motion pictures, public address, and radio, it
would seem that it is about time that specific courses leading to a
degree in this most important field should be offered by some of our
more progressive institutions.

It is of particular interest to the Committee that of the well-known
institutions of learning in Southern California — the home of the
motion picture — only the University of Southern California lists any
courses in the field of motion picture making. However, these courses
are pointed more toward training people for teaching similar courses
in other institutions and for providing a background for those inter-
ested in the cinema rather than providing solid engineering training
pointed toward the professional aspects of the industry. One would
expect that here at least some attempt should be made to train young
men and women for positions in what is probably the largest local
industry. The failure of these institutions to do so reflects the general
opinion that motion picture making is in the classification of a craft
rather than a profession, and as such does not warrant the serious
attention of institutions of higher learning. Another important
factor is the reported difficulty of graduates from local schools finding
employment in the studios, where a rigid closed shop bars employ-
ment of anybody except members of the various labor unions. It
will be necessary, therefore, to secure the co-operation of the various
labor unions with the local colleges in setting up high grade courses
of technical instruction. Until some avenue of employment is opened
to prospective graduates of these courses, there seems to be little hope
of improvement in the present situation.

Aug. 1946 MOTION PICTURE INSTRUCTION 105

Appendix A

QUESTIONNAIRE

COMMITTEE ON MOTION PICTURE INSTRUCTION
SOCIETY OF MOTION PICTURE ENGINEERS

The Society of Motion Picture Engineers has organized a Committee on Motion
Picture Instruction. This Committee was formed in response to numerous re-
quests received from members of the Society, individuals outside of the Society,
and members of the Armed Forces asking for information as to where educational
courses might be taken which would help them gain a better understanding of the
motion picture film industry, in its three major phases: namely, production,
distribution and exhibition.

The following questionnaire has been prepared in an effort to procure this
information, and your cooperation in completing it will be appreciated:

Name of Institution.

Address-

Check Type of Institution: University College Technical School—

Trade School Other ,

Courses Pre-Requisites Semester Hours Credits

Cinematography
(including color)

Photography
(including color)

Sound Recording

Motion Picture Film
Editing

106

Courses

REPORTS OF SMPE COMMITTEES

Pre-Requisites Semester Hours

Credits

Motion Picture

Projection

Motion Picture
Distribution

Economic Problems in

Motion Picture

Production and Ex-

hibition

Film Processing — Still

Film Processing —

Motion Picture

Add here any courses
in topics not listed
above which may
have a bearing on

•

the motion picture

industry and which

deal with the techni-

cal rather than the

artistic phases of

motion picture pro-

duction and exhibi-

tion

JOHN G. FRAYNE, Chairman

Committee on Motion Picture Instruction

REPORT OF THE COMMITTEE ON 16-MM AND 8-MM
MOTION PICTURES*

D. F. LYMAN**

Early in 1945 this Committee, formerly called the Committee on
Nontheatrical Equipment, was reorganized, principally because the
chairman became Engineering Vice-President of the Society.

From the very start, there was some dissatisfaction with the former
name, "Committee on Nontheatrical Equipment." It seems to us
that 16-mm film, at least, is destined to become more and more
theatrical, in both senses of the word. Moreover, we did not wish to
have our activities confined to "equipment," nor to the 16-mm size,
which was the only one mentioned in the wording of the scope of the
Committee. J. A. Maurer discussed the question with the proper
authorities of the Society. They gave us their permission to have the
name "Committee on 16-Mm and 8-Mm Motion Pictures" proposed
to the Committee in the form of a letter ballot. Affirmative votes
were received from 27 of the 31 members who voted. The other four
suggested alternatives that were, for the most part, similar. Hence
the new name.

We needed also a new expression of the scope of the Committee
because the old statement limited our activities to problems related
to the projection of film, and to the 16-mm size, as stated above. In
the April 1946 issue of the JOURNAL, the new wording is as follows:

"16-MM AND 8-MM MOTION PICTURES (formerly Nontheatrical Equip-
ment).— To make recommendations and prepare specifications for 16-mm and
8-mm cameras, 16-mm sound recorders and sound recording practices, 16-mm
and 8-mm printers and other film laboratory equipment and practices, 16-mm
and 8-mm projectors, splicing machines, screen dimensions and placement, loud-
speaker output and placement, preview or theater arrangements, test films, and
the like, which will improve the quality cf 16-mm and 8-mm motion pictures."

Our Committee now has 36 members, which means that it has
been enlarged considerably. It has a definite job to do; one that

* Presented May 10, 1946, at the Technical Conference in New York.
** Chairman.

107

108 REPORTS OF SMPE COMMITTEES Vol 47, No. 2

should be completed as soon as possible. In 1941, the previous Com-
mittee issued a report entitled "Recommended Procedure and Equip-
ment Specifications for Educational 16-Mm Projection." It was
prepared for the Committee on Scientific Aids to Learning, of the
National Research Council. There has been so much interest in the
report that the supply of several thousand reprints has been de-
pleted, and it was necessary to print a new issue. But there is an-
other reason for revising the recommendations. There has been con-
siderable progress since then, perhaps not so much in the equipment
as in our appreciation of the more important aspects of the problem,
accentuated by the requirements of the Armed Forces. In 1944 and
1945 a great deal of thought and effort was expended on American
War Standards, through the Z52 Committee of the American Stand-
ards Association. The various subcommittees of Z52 were able to
make use of the 1941 report of this Committee in preparing drafts
for a number of the War Standards. With all the mental power that
was available to Z52, it was inevitable that there should be advances
in the understanding of the essential specifications and tests for 16-mm
equipment. Those advances, in turn, made our 1941 report somewhat
out of date. It is now our task to review the recommendations with
the idea of using as much of the material as applies to classroom
projection. Furthermore, we should make as much further progress
as we can. This work is now under way.

Here is an illustration of the need for widening the scope of the
report. Recently there has been a movement among those who are
concerned with the future of the 16-mm educational program to ac-
quaint the architects who are designing schools with the requirements
for projection. There has been enough experience now, in the schools,
to indicate that all the audio-visual aids should be co-ordinated and
that the planning should start with the architect. There are so many
factors, such as the shape of the room, provision for darkening the
room, ventilation of darkened rooms, treatment of surfaces for good
acoustics, and provision of electrical outlets, that it is essential for
us to gather them together for the guidance of the architect. This
is especially important in view of the large amount of school building
that is now in the planning stage or in prospect.

Then there are other ways in which we can help the school authori-
ties more than we have in the past. One of their problems is the
training of volunteer student projectionists. It should be possible for
us to help a great deal in this respect. We are fortunate in having, as

Aug. 1946 16-MM AND 8-MM MOTION PICTURES 109

members of the Committee, men who have had experience in both
administrative and field work in school projection, and we are planning
to make the most of that experience. Moreover, we expect that they
can tell us how to write and present our material in such a way that it
will appeal to the educational authorities and be intelligible to them.

A tentative outline for the revised instructions has been prepared.
Chapter headings include:

(7) Origin and Growth of Audio-Visual Aids to Education
(77) Architectural Design of Schools, Auditoriums, and Classrooms
(777) The Function of the Projector

(7V) Rearrangement of the Existing Classroom and Selection of the Screen
( V} Specifications for the Projector and the Sound Apparatus
( VI) Specifications for Arc Lamps
( VII) Duties of the Projectionist
(V777) Care of the Film
(IX) Care of the Apparatus

A meeting of the Committee was held on November 2, 1945. At
that time, the order of the outline was changed, and the material was
classified into several groups.

Since then, previous specifications that bear on each subject listed
in the outline have been combined in preparation for writing the
revised copy. We hope that all the material on each subject can thus
be kept together in the new issue.

When the time came to assign this material to the members for
revision, it was apparent that we needed smaller working groups of
four or five members each. For greater efficiency and speedier ac-
tion, nine subcommittees are being appointed: Cinematography,
Sound Recording, Test Films, Projection Practice, Projector Speci-
fications (dealing with the projection of the picture), Projector Spe-
cifications (dealing with the reproduction of the sound), Laboratory
Practice, Liaison and Advisory, and Editorial. These are temporary
subcommittees, and there is no intention of making them obligatory
for future chairmen of the parent Committee.

There has been some discussion about how the Committee can ex-
pand its activities so that its influence will be felt in other fields.
There are organizations that can use our help if they know that they
can turn to us. The Liaison and Advisory group will be useful in this
work. For the immediate future, however, the school projection
program is enough to require our undivided attention. The Editorial
subcommittee will specify the form and style of the various sections
of the new issue.

110 REPORTS OF SMPE COMMITTEES Vol 47, No. 2

We wish to appeal to the members of the Society to give us all
possible help for our preparation of better recommendations. Any
technical information about the use of 16-mm sound projectors for the
imparting of knowledge will be welcome. It will reach us safely if it
is sent to Boyce Nemec, Engineering Secretary of the Society. We
are especially interested in reaction to American War Standard
Z52. 1-1944 issued by the American Standards Association and to the
allied specifications included in that standard as appendices.

REPORT OF THE COMMITTEE ON STANDARDS*

F. T. BOWDITCH**

During recent months the Committee on Standards has become
increasingly active. While the war was in progress, most standardi-
zation was done in the interests of the Armed Forces, under the
auspices of special war committees of the American Standards Asso-
ciation on which many members .of the Committee on Standards
served. A great quantity of war standardization resulted which now
requires consideration from a peacetime point of view. In addition,
these new standards called attention to the need for modernization
of older ones. First consideration of this matter was given by ASA
Sectional Committee on Motion Pictures Z22, under the chairman-
ship of Clyde R. Keith.

In line with this procedure, Z22, in a meeting last October, reviewed
all Motion Picture Standards then in existence, both the prewar Z22
Standards and the War Standards developed by Z52. Of the 53
standards so reviewed, 20 were reaffirmed in their existing form and
six others have since been approved with minor editorial changes.
These 26 have now been referred to the United Nations Standards
Co-ordinating Committee for inclusion in their agenda. Of the re-
maining 27, three are under consideration by subcommittees of Z22,
two have been referred to the Research Council of the Academy of

* Presented May 10, 1946, at the Technical Conference in New York.
** Chairman.

Aug. 1946 STANDARDS 111

Motion Picture Arts and Sciences, and 22 have been referred to the
Committee on Standards of the SMPE.

This large number of standards was referred to our Committee pri-
marily because each one of them represents a special technical
problem. After several years of relative inactivity, we were thus
suddenly faced with a task sufficient to occupy us for several years
at the normal prewar pace of such work. Moreover, the ASA spe-
cifically requested prompt action on all these standards in order that
as many as possible might be made ready for submission to the United
Nations Committee before its first technical meeting.

In order to implement this work six subcommittees of the Com-
mittee on Standards have been formed, and each assigned a group of
related standards for early consideration. These include subcom-
mittees on

(a) Projection Reels, with three projects, under the chairmanship of D. F-
Lyman ;

(b) Photographic Density and Sensitometry, with two projects, under the chair-
manship of D. R. White;

(d) IQ-Mm and S-Mm Camera and Projector Apertures, with six projects, under
the chairmanship of John A. Maurer;

(e) 16- Mm and 8- Mm Projector Sprockets, with four projects, under the chair-
manship of Otto Sandvik; and

(/) Film Splices, with two projects, under the chairmanship of Wm. H. Offen-
hauser, Jr.

All of these subcommittees are actively at work and the revision of six
of the 22 standards has now been agreed to in subcommittee and sub-
mitted to letter ballot of the parent Committee.

Another project of perennial interest to the Committee is that of
the Glossary of Motion Picture Terms. After working on this task
intermittently during the war years, the job, instead of diminishing
toward completion, has actually grown larger as the result of the intro-
duction of many new terms, while the manifestation of increased
trade interest has further justified the early preparation of a glossary..
This project has, therefore, been broken down into twelve parts ac-
cording to field of interest. Nine of these are to be handled by present
engineering committees, and three by special subcommittees of the
Committee on Standards. It is hoped that in this way the task can be
advanced to earlier completion with the publication of each section
as it is finished rather than waiting for the entire job to be completed.

112 REPORTS OF SMPE COMMITTEES Vol 47, No. 2

In 1941 an "SMPE Recommended Practice" with respect to the
edge-numbering of 16-mm film was published in the JOURNAL for
comment before consideration as an American Standard. A sub-
committee of the Committee on Standards under the chairmanship
of Lloyd Thompson was appointed about a year ago to review this
situation. Mr. Thompson's subcommittee has recently recom-
mended that this "Recommended Practice" be advanced to an
American Standard, without change, that is with the specification of
a 40-frame interval between numbers. The parent Committee at
its meeting on May 8, 1946, approved this recommendation and
authorized the first step toward American Standardization, that
is, a letter ballot of the entire Committee on this Proposed Standard.

For many years a discussion has been active relative to the pos-
sible advantage of a 16- tooth intermittent projector sprocket larger
in diameter than the 0.935-in. value now in use with 35-mm film.
Laboratory tests have consistently indicated a much longer film life
with a larger sprocket, but the practical application of such a sprocket
had never been successfully accomplished. As long ago as 1930,
American Standard Z22.35 called for a diameter of 0.945 in. How-
ever, when this larger size was supplied to the trade in 1934, many
complaints of noisy operation arose, attributed to sprocket wear by
the film. Consequently, an expensive reversion to the 0.935-in.
diameter followed with the result that Z22.35 has never been a truly
observed Standard.

In the recent war emergency the necessity for film conservation was
responsible for a renewed consideration of this subject. A special
Subcommittee on Intermittent Projector Sprockets for 35-Mm Film
was formed under the chairmanship of Dr. E. K. Carver for this
purpose. The committee had laboratory data indicating that at least
double the film life could be obtained through the use of a larger
sprocket. The problem was to determine how such a sprocket would
stand up in service and if the initially apparent increase in film life
persisted throughout sprocket life. After extensive tests with sprock-
.ets of several diameters in a number of theaters, the subcommittee
found that the initial film saving does persist, and that sprocket wear
is in no case faster and in many cases much slower than with the
present 0.935-in. diameter sprocket. Increased projector noise, a
possibility, originally the subject of much controversy, was simply
nonexistent. The parent Committee has since approved by letter
ballot the subcommittee's recommendation of an 0.943-in. diameter

Aug. 1946 STUDIO LIGHTING 113

sprocket and this recommendation had been in turn passed on to ASA
Sectional Committee Z22.

Particularly in this last instance does the value of standardization
become apparent. Potential saving in decreased film wear through
adoption of this standard has more than justified the work of this
committee, and we hope is typical of the engineering service which
can now be supplied to the industry as a result of the recent central
office expansion. A most important step in this direction is the
acquisition of our full-time Engineering Secretary, Boyce Nemec,
who can apply the needle when required to keep our projects on the
move. Certainly we could not ask for better technical representation
of the industry than that now provided on our present engineering
committees. Under John Maurer's able direction, as Engineering
Vice-President, and the application of Boyce's needle, we can expect
real progress in the months to come.

REPORT OF THE COMMITTEE ON STUDIO LIGHTING*

C. W. HANDLEY**

Previous papers and reports have catalogued and described motion
picture studio lighting equipment. The purpose of this report is to
show the light output at various beam divergences of some of the
popular types of equipment and to give an indication of the light
levels used by some directors of photography. This information
should give the reader a basis for general conclusions on the question
of how much light is used.

Practically all of the lighting equipment used around the tops of
sets on parallels and much of the floor lighting is accomplished by
means of spotlamp units equipped with Fresnel-type lenses. These
units are controlled as to spot diameter by moving the light source
toward or away from the lens and are reduced in intensity at a given
spot diameter by the use of frosted gelatin diffusers placed in front of

* Presented May 10, 1946, at the Technical Conference in New York.
** Chairman.

114

REPORTS OF SMPE COMMITTEES

Vol 47, No. 2

30 25 20 15 IO 5 05 10 15 20 25 30
DEGREES OF DIVERGENCE

FIG. 1. Candlepower distribution from a
Mole-Richardson "midget" incandescent spot
type 404, with a 200-w, T-10 bulb d-c bayonet
base lamp.

120,000

100,000

80,000

60,000

40,000

20,000

30 25 20 15 10 5 0 5 10 15 20 25 30
DEGREES OF DIVERGENCE

FIG. 2. Candlepower distribution from a
Mole-Richardson "baby" solar spot, type 406
with a 750-w, T-24 bulb medium bipost base
M.P. type lamp.

Aug. 1946

STUDIO LIGHTING

115

30 25 20 15 10 5 05 10 15 20 25 30
DEGREES OF DIVERGENCE

FIG. 3. Candlepower distribution from a
Mole-Richardson "junior" solar spot type 410
with a 2000-w, G-48 bulb mogul bipost base,
M.P. type lamp.

B - 20'BEAM

I9.0OO LUMENS

C- JO-BEAM

26,500 LUMENS

D - 60'BEAM

47.000 LUMENS

1,000,000

800,000

600,000

400.0OO

200.0OC

30 25 20 15 10 5 0 5 10 15 20 25 30
DEGREES OF DIVERGENCE

FIG. 4. Candlepower distribution from a
Mole-Richardson "senior" solar spot type 414,
with a 5000-w, G-64 bulb mogul bipost base
lamp.

116

REPORTS OF SMPE COMMITTEES

Vol 47, No. 2

80 70 60 50 40 30 20 10 0 10 20 30 40 50 60 70 80

DEGREES OF DIVERGENCE

FlG. 5. Candlepower distribution from a Mole-Richardson "duarc"
type 40 operating from a 120- v, d-c line, arc current 41 amp.

1/400,000

1,200,000

i 1.000,000

800.000

600,000

400.000

200.00C

V
2,000,0001

1,800,000

1,600,000

1400000*;

18* BEAM
20,400 LUMENS
C -44'BEAM (MoxSprtOd)
- 62,500 LUMENS

30 25 20 15 10 5 0 5 10 15 20 25 30
DEGREES OF DIVERGENCE

FIG. 6. Candlepower distribution from a
Mole-Richardson high intensity arc spot, type 90
with 115-v, d-c arc operating at 110 amp, 60

Aug. 1946

STUDIO LIGHTING

117

the lens. Figs. 1 to 6 show the average apparent candlepower and
lumens output of a number of the spotlamp type units.

Fig. 7 shows the average apparent candlepower and lumens out-
put at various angles of a carbon arc type broadside lamp which does
not have adjustable beam spread and is used for general floodlighting.

For black-and-white cinematography, tungsten filament lamps are
usually the main light sources, particularly on small sets. Carbon

30 25 20 15 10 5 05 10 15 20 25 30
DEGREES OF DIVERGENCE

FIG. 7. Candlepower distribution from a
Mole- Richardson high intensity arc spot, type 170
with 115-v, d-c arc operating at 140 to 145 amp,
60 to 70 arc v.

arc lamps are used for "streak lighting," shadow effects, and on larger
sets where it is necessary to project light for considerable distances.

"Key-light" levels on black-and-white sets vary from 50 to as high
as 400 ft-c.

Professional color cinematography is balanced to sunlight, there-
fore, carbon arc lamps are usually the main light sources. The flood-
type carbon arc lamps are used without niters and the high-intensity
rotating positive-carbon type spotlamps are equipped with light
straw-colored gelatin niters known as "F-7". Tungsten filament

118 REPORTS OF SMPE COMMITTEES Vol 47, No. 2

lamps, fitted with blue filters for sunlight balance, are used on color
where fill light is indicated on small sets and for softening the front
illumination in closeups. The key-light levels in color cinematog-
raphy-vary from 250 to as high as 900 ft-c.

REPORT OF THE COMMITTEE ON TELEVISION PROJEC-
TION PRACTICE*

P. J. LARSEN**

About a year ago this Committee was organized as a subcommittee
of the Society's Theater Engineering Committee. At that time,
its scope as outlined included specification, design, construction, in-
stallation, maintenance, and method of use of equipment for projec-
tion of television pictures in the theater. This entails recommenda-
tions for arrangement of television equipment in the theater or pro-
jection room, including definite plans and layouts necessary for
such equipment including its location and electrical and mechanical
association with the normal film projection equipment. This scope,
therefore, also includes the dimensions of the projected picture, color
spectrum of light source, and the characteristics of the reflective or
translucent screen that may be used for viewing the theater television
performances.

As noted in the above, the scope includes the specifications, de-
sign, construction, installation, maintenance, and method of use of
equipment. This means that all matters dealing with theater tele-
vision transmitters, relays from studio to transmitters and from city
to city, receivers, projectors, and all the associated gear fall within
the scope of the Committee.

The Committee is made up of members of the Society representing
manufacturers of television equipment, theater circuits, motion pic-
ture producers and distributors, including newsreel companies, tele-
vision broadcasting companies, architects, theater equipment deal-
ers, and other interested members of the Society.

At its first meeting held on June 1, 1945, it was decided that four

* Presented May 10, 1946, at the Technical Conference in New York.
** Chairman.

Aug. 1946 TELEVISION PROJECTION PRACTICE 119

Task Groups would be formed, each to make a study and furnish the
Committee with information along the following lines :

(1) Task Group A — To furnish engineering information regarding existing and
proposed theater television equipment with respect to size, weight, and shape of
equipment, minimum and maximum size of projected image brightness attain-
able, type of system, line definition, type of screen and maximum viewing angle
recommended, recommended location of equipment, etc.

(2) Task Group B — To furnish information regarding existing conditions in
different types of motion picture theaters in this country with respect to physical
condition and sizes of projection rooms, balconies, auditoriums, stages, viewing
angles, etc., for purposes of determining location of proposed theater television
equipment.

(5) Task Group C — To furnish information in connection with picture quality,
including resolution, color, gamma, contrast range, and screen characteristics,
etc., now available from monochrome and color film in order to determine the
optimum equipment designed for theater television.

(4) Task Group D — To visualize what theater television is to be ; namely, how
theater television can be presented in theaters, its commercial aspects, the types of
distribution systems required for inter- and intra-city, the problems of program
pickup, storage of program material and scheduling distribution, the question as
to the privacy of its programs and where reception and projection of broadcasting
programs should be provided.

At the meeting held on September 26, 1945, preliminary reports
were offered on Task Group B and C. At this same meeting there
was some discussion led by D. E. Hyndman, President of the Society,
with respect to the formation of Task Group D.

Formal reports were offered to the Committee at its meeting on
April 4, 1946, by Task Groups A, B, and C. As a result of the dis-
cussion at this meeting, certain additional information is to be ob-
tained by these Task Groups. However, it was agreed that further
progress in the functions of the Committee now depend on the forma-
tion of Task Group D so that it might consider the information pre-
sented by the other Task Groups and make certain specific recom-
mendations which would permit the Committee to proceed with its
work. It appears that before much further progress can be made
certain problems included in the scope of Task Group D would have
to be agreed upon by the industry.

It is hoped that through co-operation with the Motion Picture As-
sociation, which represents the major motion picture film producers
and distributors, and with the leading motion picture theater owner
organizations, Task Group D can be formed in the near future and,
without too much delay, contribute its assistance to the Committee
and the Society.

MOTION PICTURES TOMORROW*

W. F. RODGERS**

As one who has given a business life of 35 years to the motion picture
business, I feel grateful indeed for the opportunity to speak to the
Society of Motion Picture Engineers. Individually and collectively
you, as an organization, have made the most important contribution
to the growth and stability of this great industry.

The technical or, as I would rather put it, the scientific and engineer-
ing progress you have made in sight and sound, in presentation, and
projection, are the factors in a highly mechanized business that have
made possible these wonders, for it remained for someone with the
vision and the patience of a scientist and the foresight of an engineer
to plan and nurse through experiments so important to the steady de-
velopment of this business. Therefore, I salute you, gentlemen, for
your ability and tenacity of foreseeing the coming of the scientific
approach to this business.

That the days of the "hit-and-miss" approach to our problem are
over must be evident. These are the days when no national merchant
can afford to put a product on the market without first thorough, care-
ful analysis to ascertain the market acceptance of that product, its
probable source of sales, and which market is most likely to succeed.

As distributor, I, too, have seen a new approach — a business ap-
proach that first thinks out a problem and then analyzes the effects;
and its success is, therefore, by no means accidental. Just as blue-
prints are necessary to an engineer, so does the new businessman's
equipment require the facts, figures, charts and graphs.

Today motion pictures are receiving the greatest acceptance in the
history of the business. That knowledge of making motion pictures on
which you and others have burned so much midnight oil is paying its
dividend. Mobilized of necessity during the war, the science of dis-

* Presented May 6, 1946, at the opening luncheon, 59th Semiannual Tech-
nical Conference in New York.

** Vice- President in charge of Distribution, Loew's Incorporated, New York.

120

MOTION PICTURES TOMORROW 121

tribution, coupled with the science of presentation and the science of
production, teamed well to do the greatest morale job in the history
of the world.

In the tomorrow of the business that same scientific approach to our
problems will continue to pay a dividend, for ours is a great responsi-
bility if this war-torn world is to ever settle on the basis of a permanent
peace. No medium, as so often it has been said before, can carry the
message of the brotherhood of man as well as can motion pictures.
And so, we see through our international departments the appearance
of 16-mm projectors in the most remote localities throughout the world
that all peoples may know that the world does not need bloodshed and
starvation, but that the world they fought for is and can be a world of
clean living, of peace and of plenty. That is our job and we here who
make these pictures possible must feel the responsibility of doing the
best job possible, that the by-product of our efforts will be the wider
distribution of American-made motion pictures to the world.

To accomplish this we must face the future of our business ready to
assist producers everywhere in their efforts to re-establish motion pic-
tures of their own, and just as we will send them our product so must
we be ready to accept whatever of their products are suitable to our
audiences and throw into the discard once and for all that too-old
bias against so-called "foreign" pictures. We could not maintain the
high standard of production here if we could not expect to achieve
world distribution over there. Any barriers that are erected because
of our reluctance to give proper presentation to good pictures, re-
gardless of their source, will only react on the quality of our own
production. There is no need for a scientific approach to that one,
it is just plain arithmetic.

In the tomorrow of the business, just as with the coming of sound
we learned such new words as "acoustic," "photoelectric cells,"
"faders," etc., so have we commenced to acknowledge the wisdom of
research in the pursuit of scientific analysis of proposed titles, scien-
tific analysis of market possibilities and proper approach, polling of
communities to find out what mediums they react to the best — a con-
stant search that old methods of "trial and error" be abandoned in
favor of the facts on which an intelligent plan can be made. Adver-
tising agencies are quietly investigating the proper type faces to
which the public react best, how can this picture best be merchan-
dised through radio, newspapers, word of mouth, magazines, and
where are they the strongest.

122 W. F. RODGERS Vol 47, No. 2

We in the Sales Department are pouring over charts to see whether
the public demand for a picture' has reached the necessary point that
will enable it to reach the greatest number of patrons, analyzing
these reports that we may price our merchandise for the individual
situation, all aimed to permit all types of theaters in all types of
localities better to withstand the various types of competition which
is bound to come once the world returns to its peacetime activities.

Ours, too, is a great responsibility that we encourage our theater
owners to try to influence more people to appreciate the marvel of mo-
tion pictures as the greatest amusement value in the world. It is also
our responsibility that this great product conceived in Hollywood
through the use of all the latest authentic and mechanical methods for
which you are responsible be given every aid that more and more of
our public become motion picture conscious.

And so with you unceasingly continuing the work of the Society of
Motion Picture Engineers, with Hollywood alert to pick the best in
literature and transfer it into great productions through the science of
their creative genius, with the science of advertising and market
analysis available to us and the science of sales well past the formula of
"I want and you'll give", and with the science of exhibition, one that
is today occupying the attention of every thinking theater owner who
knows full well that the physical appointments of his theater and
science of public relations are vital to his success, it seems to me we
lack just one science; at least there is one that has not been given the
attention it must receive if, in the new scientific world of tomorrow,
motion pictures are to be ready to march side by side with the other
great businesses of the world.

If s the science of industry team play. I am sure during this last
bitter struggle there must have been many disagreements over policy,
many disappointments, yet all interested faced the enemy as a team,
one part making up for the weakness of the other toward their mutual
self-preservation and eventual victory.

We, too, will always have our internal differences, regrettable as
they may be, but the importance of a unified pride in our business, the
vital necessity of an appreciation that the other fellow, too, has
problems, and an appreciation of his efforts toward the success of the
business as a whole, that science of team play must be redeveloped
if we are to be able to make the progress ahead that we can make.
As I said in the beginning, I have been identified with this business
for more than 35 proud years, and anytime I hear any element of

Aug. 1946 MOTION PICTURES TOMORROW 123

the business carelessly referred to, or anytime it is unfairly attacked
and the industry does not rise in righteous indignation to defend an
industry that has done so much, not only for all of us, but for the
world at large, I shudder at our thoughtlessness. The science of
team play — of appreciating the other fellow's contribution to the
business, of an all-out one-for-all love of the business and a solemn
resolve to make our individual contribution to a better public appre-
ciation of our business — that, gentlemen, is a must for tomorrow;
that, gentlemen, is a science that does need developing and will spell,
in the final analysis, achievement.

We who are in the business are looked on as experts on motion pic-
tures, our opinions are sought, people like to talk to us about movies.
Let us then, therefore, in the tomorrow of peace highly resolve that
our individual part in its success over and beyond our business contri-
butions shall be to say and say again, "It's a great, fine business, with
a majority of great, fine people in it doing, on the whole, a great, fine
job of making motion pictures the world's greatest entertainment
value."

PRESENTATION OF SCROLL TO THOMAS ARMAT*
DONALD E. HYNDMAN**

The date of April 23, 1946, marked 50 years since Thomas Armat
gave the first exhibition in a theater of motion pictures as we know
them today. The exhibition was given in Koster & Bial's Music
Hall in New York City where Mr. Armat personally operated the
projector on this first historic night marking the beginning of what
we know today as one of the seven major industries of the United
States.

The projector he used was designed by him and embodied a new
feature of relatively long periods of rest and illumination of each suc-
cessive picture on the film. This projector was then known as the
Vitascope.

Approximately ten years ago, Mr. Armat was made an Honorary
Member of our Society and, tonight, we are honoring him with a cita-
tion scroll of his pioneering work in our motion picture industry. It is
appropriate to report for history on this most historic occasion that,
when he brought the motion picture to the screen, he made it strictly
a silk-hat occasion.

The fact is that Mr. Armat, then a blithe figure of a young business
man of thirty and launched on a successful career in Washington busi-
ness and with all the tradition of a F.F.V. (First Families of Virginia)
behind him, was tremendously impressed with an invitation to bring
his Vitascope to New Jersey and demonstrate it before the great
Thomas A. Edison. In the 1890's in Washington all big events
called for formal dress.

Mr. Armat arrived at West Orange, New Jersey, one afternoon
in a silk hat and frock coat and with the Vitascope in a trunk. He
was a bit overwhelmed when he was invited into a smoky, dusty barn
of a foundry with a group of shirt-sleeved laboratory assistants and
over ailed mechanics there to help Mr. Edison look at the new projector.

* Presented May 8, 1946, at the dinner-dance, 59th Semiannual Technical
Conference in New York.

** President, SMPE.
124

CITATIONS

125

Doubtless Mr. Edison and the hired hands were impressed with
Mr. Ararat's silk hat, too. Anyway, we can be assured that projec-
tion came to the movies in style. Mr. Armat rates special honors, then,
not only for his contribution to the theater screen but also for being
the first genuine silk-hat engineer of our industry.

ja>-a c?-vd

Society of QJotion picture Qnainecrs

homos jHrmat

""Mnwntor of r)u-
lecture

•'
Unction fjhcture g>rojector
^pn the occ43*lon of rhc

^fiftieth Anniversary

of the first exhibition of" motion picture* m a rhcafiv.
Roster and Biale aXisnc Hall in Dew UorU. A|.vil
23.1896. DKat owning ,T3hotna« Am^itop.

a j.->tO)«rctoi of" In* ou»ti dcstqri which vsvi^ i*-.
ptOK-'ctii-ug machiiiv- ctJij.^yuitT ti Uvp-fot -muui mcoi>»
and cju-irtct rhc film ei longer j.vxiocl ^f ivdC ami
illumitiiation than the time featured for niov..-.^-,!:
fl-om fratnc to fvciinc . Ohc»c fcotnivd so uni.
the vjualitv of'mot ion picture projection flu it Th.-v
were incorporcitect •ub*cquently irx TTU>«=.t
mcrciciUy succc»s>f"ul ptv»)vVtoi-&

ii» scroll ift presented b>' Hie Society of" 0"Jv>t ion
'IP Picture Cn9uwcr» mjycocjmt-ion of Ch

d inventions of Ghomcw Ar«r»cir which
hocl a cot\tinumvT inHueiivrc on riic d.

FIG. 1. Scroll presented to Thomas Armat.

On behalf of our Society, I now take great pleasure in presenting to
Brooke Armat, son of Thomas Armat, this anniversary scroll which
reads as follows :

"To Thomas Armat, pioneer inventor of the motion picture pro-
jector on the occasion of the Fiftieth Anniversary of the first exhibi-
tion of motion pictures in a theater, Koster and Bial's Music Hall in
New York, April 23, 1896. That evening, Thomas Armat operated a

126

CITATIONS

Vol 47, No. 2

projector of his own design which was the first projecting machine
employing a loop-forming means and giving the film a longer period of
rest and illumination than the time required for movement from
frame to frame. These features so improved the quality of motion pic-
ture projection that they were incorporated subsequently in most
commercially successful projectors.

FIG. 2. THOMAS ARM AT.

"This scroll is presented by the Society of Motion Picture Engi-
neers in recognition of the distinguished inventions of Thomas Armat
which have had a continuing influence on the development of motion
picture projection for half a century."

[Mr. Armat accepted the illuminated scroll on behalf of his father, who was
unable to be present, with deep appreciation and gratitude. ]

Aug. 1946 CITATIONS 127

PRESENTATION OF SCROLL TO WARNER BROTHERS

[President Hyndman continued:]

We have already been privileged tonight to honor a distinguished
motion picture pioneer, Thomas Armat — and now it gives me equal
pleasure to announce another citation by our Society.

Society ofjfljotioufiirturc Qmiinws

mill of j-JchicucnuMit

J^rc»*titccl

»amcr

FIG. 3. Scroll presented to Warner Brothers.

After motion pictures, there came sound pictures. But they did not
just come. They had to be visualized, engineered, developed, and
made into a practical, commercial article.

Thomas A. Edison and Dr. Lee de Forest were among the first to
visualize this dream and to demonstrate its possibilities.

128

CITATIONS

Vol 47, No. 2

Many other inventors, technicians, and engineers also worked on
the idea — but except for Edison, de Forest, Fox-Case, the Bell Labo-
ratories, Western Electric, Eastman, RCA, General Electric, du Pont,
and a few others, most of them eventually gave it up as impractical.

Nearly everybody in the amusement business, too, after attending
dozens of experiments and demonstrations, became fed up with the
idea of talking pictures and declared that it could never be turned into
a commercial success.

FIG. 4. Left to right, MAJOR ALBERT WARNER, t PRESIDENT
HYNDMAN. AND BROOKE ARMAT.

Then along came four young men who had more faith, more de-
termination and more perseverance. Those four boys were Harry,
Jack, Albert, and the late Sam Warner.

They not only defied the skeptics who scoffed at the idea of talking
motion pictures, but they gambled everything they had on the new
invention — and it's no secret to many of you that in the year just
before they presented their first synchronized picture, Don Juan, on
August 6, 1926, and in the year that followed, when they brought out
The Jazz Singer with the first line of dialogue ever spoken from the

Aug. 1946 CITATIONS 129

screen, the Warner finances were so low that employees often had to
be asked to hold off cashing their weekly pay checks for a few days
after receiving them.

But the Warners saw it through, and you know the rest.

We are honored to have with us here tonight one of those four coura-
geous brothers, Major Albert Warner — and on behalf of our Society,
I now take great pleasure in presenting to Major Warner this Scroll of
Achievement, which reads as follows :

"In recognition of their pioneering courage and efforts in the de-
velopment of sound recording and sound reproduction for motion
pictures; their faith in the technical inventions that gave a new enter-
tainment and educational medium to the world; their leadership in
the adoption of new technical creations and improvements in the field
of motion pictures ; and the encouragement they have thereby given
to engineers to bring forth new ideas and create better standards in the
motion picture art; this Scroll of Achievement is presented to Warner
Brothers by the Society of Motion Picture Engineers in this Twentieth
Year of the successful introduction of sound motion pictures.

"By order of the Board of Governors, Society of Motion Picture
Engineers."

[Accepting the scroll on behalf of his brothers and himself, Major Warner ac-
knowledged the honor with the following words : ]

Mr. President, Honored Guests , Ladies and Gentlemen : 1 am very
deeply honored to accept this scroll on behalf of my brothers and
myself.

It is true that the Warner Brothers brought sound and talking
pictures out of the laboratory and gave them to the world as a new
form of entertainment. But a large measure of the credit for that
development goes to you engineers.

The success of talking pictures was a victory for engineering genius
as well as an achievement of the motion picture studio. You men
perfected the machine. We supplied the showmanship. The im-
portance of the engineer in the field of motion picture development
has never been fully appreciated. Few persons know how much we
owe you men, not only for what you did in the field of sound, but
also in color, in working out equipment standards for theaters and
studios, and in the new work you are constantly doing to improve

130 CITATIONS Vol 47, No. 2

our sound equipment so that pictures can be produced more effi-
ciently and shown to the best advantage.

This year not only marks the twentieth anniversary of talking
pictures, but it is also the fiftieth anniversary of motion pictures as
commercial entertainment, and I should like to take this opportunity
to pay my respects to Thomas Armat, the "father of the projector,"
whom you have just honored here tonight. If Armat had not followed
through with his engineering skill and perfected the device that made
it possible to project pictures on a big screen, there would have been
no motion pictures — and no talking pictures — at least, not until some
other engineer had the perseverance to solve the technical problems
involved.

Giving talking pictures to the world was not a one-man or one-
company proposition. Many talented men, many members of the
Society of Motion Picture Engineers who are here tonight, and
several big companies like Western Electric, Bell Telephone, Eastman
Kodak, Edison, RCA Victor, and others, played a big part. So our
twentieth sound anniversary, and this Scroll of Achievement, is a trib-
ute to all of them as well as to Warner Brothers. I only regret that
my brother Sam, who worked hardest to overcome the early difficul-
ties in the making of talking pictures, is not here tonight to share in
this honor.

I do not know what will be the next great contribution of the engi-
neers to the entertainment industry — whether it will be third dimen-
sion, or television, or new refinements in the equipment we now have,
or whatever it may turn out to be. But I know that your work is
not finished. You engineers are just as much a part of our industry
as the actors, writers, directors, producers, film salesmen and exhibi-
tors.

New horizons are in sight for motion pictures — in the fields of
education, culture, human relations, business — and above all, in the
great work of promoting American ideals of democracy, world unity
and peace. Your work can help to rid civilization of demagogues
who preach the doctrine of "divide and conquer" that destroyed a
large part of the world. We do not want agitators of that kind to
gain a foothold here and destroy this great land of liberty.

Right now, if we are not careful, we will lose the peace we fought so
hard to win. So it is up to all of us to keep working together to assure
a peaceful world. We must not leave the way open for another war
by settling for a soft peace.

Aug. 1946 CITATIONS 131

To attain these ends for humanity, we need your talents and
your efforts more than ever. The screen cannot march forward to
new and higher achievements unless it marches hand-in-hand with the
men who work out the technical problems — the motion picture engi-
neers.

A SIMPLIFIED RECORDING TRANSMISSION SYSTEM*
F. L. HOPPER AND.R. C. MOODY**

Summary. — This paper describes a recording transmision system in a single,
compact, lightweight unit, capable of excellent performance together with reliable and
simple operation.

There are many applications for a simplified recording transmission
system in both the domestic and foreign fields. Such a system would
supply all the necessary equipment to operate from a microphone and
into a sound recording modulator. A minimum number of operating
controls would be provided, in order that the operation of the system
be as simple as possible. Reliability, with a high standard of per-
formance, is another of the principal requirements. The system
should operate from a variety of power sources and should be capable
of operating several types of light valve modulators. Applications
include: Use of the system with newsreel cameras, with a small re-
corder for superportable field use where a double film system is re-
quired, or for use with double film systems where a relatively simple
and inexpensive equipment is required. It is the purpose of this
paper to discuss such a transmission system.

A survey of the fields to which this unit might be applied indicated
the following requirements :

(Jf) The equipment should be light, compact, sturdy, and all parts should be
easily accessible.

(2) Maximum reliability of operation is essential, hence the circuits should be
simple, a minimum of controls should be employed, and components and wiring
should be moisture resistant. Each component should justify its inclusion in order
that the system have minimum components.

(5) Power consumption should be a minimum so that the unit may be operated
from batteries, or if alternating current is generally available, from a-c operated
rectifiers.

(4) Performance should be adequate for the types of recording service for which
the equipment is proposed. In this respect, the system provides considerable
improvement compared over those developed in the past.1' 2

* Presented Oct. 15, 1945, at the Technical Conference in New York.
** Electrical Research Products Division, Western Electric Company, Los
Angeles, Calif.
132

SIMPLIFIED RECORDING TRANSMISSION SYSTEM

133

(5) Ease of service of the equipment in the field when away from permanent
test facilities.

The transmission equipment is housed in a rectangular duralumi-
num case 15 in. long, 11 in. high, and 7 in. wide, and weighs 25 Ib.
The chassis, which includes the top panel, carrying operating con-
trols, is easily removed from the case, even with all connecting cables
attached. The vertical portion of the chassis is employed to support
the transformers, retards, and vacuum tubes, single-side mounted.
Terminal cards, carried across the back of the vertical chassis, are

FIG. 1. RA-1253 amplifier.

employed to support resistor and condenser components. All com-
ponents are designated, and operating voltages which appear across
such items as plate and cathode resistors are indicated in order to
facilitate rapid checking or test. These cards are easily removed,
giving access to equipment terminals underneath. This method of
assembly permits of simple direct wiring and minimum use of cable
forms, resulting in reduced wiring costs. Fig. 1 shows the equipment
in its case, and Fig. 2 the chassis only.

Reference to Fig. 3, which is a schematic, will indicate the various
electrical components comprising the system. The circuits nomi-

134

F. L. HOPPER AND R. C. MOODY

Vol 47, No. 2

nally divide into two parts, the main transmission system which sup
plies program material to the modulator, and the noise reduction.
Facilities associated with the amplifier are :

(1) Two position mixer,

(2} Variable dialogue equalization,

(5) Interstage gain control,

(4) Peak limiting,

(5) Volume indicator,

(6) Headset monitoring,

(7) Modulator equalizer,

(5) Modulator "off-on" switch,

(5) Plate and heater "off-on" switches.

FIG. 2. RA-1253 amplifier chassis.

Facilities associated with the noise reduction are:

(1) Margin control (internal),
(2} Bias control (internal),
(5) Bias "off-on" switch.

In addition to these controls, certain others are provided which
may or may not be used depending upon whether the film recorder is
of the single or double film type. These facilities generally are con-
cerned with the recorder lamp and its associated controls.

The amplifier consists of 3 pentode connected stages with feedback
around the last 2 stages. Adequate feedback is used to insure a wide

Aug. 1946 SIMPLIFIED RECORDING TRANSMISSION SYSTEM

135

margin of stability and a high degree of damping. With all feedback
removed the amplifier is reasonably stable and has 125-db gain. The
application of feedback, which reduces the gain to 95 db, contributes
its total effect to increased stability and damping.

Plate current drain was reduced as much as was consistent with the
desired over-all performance from the viewpoint of harmonic distor-
tion. The plate consumption of the 3-stage amplifier is 1 1 milliam-
peres at 180 v. Examination of the distortion-frequency character-
istic, curve B of Fig. 4, shows one per cent distortion or less at all fre-

NOtSE REDUCTION

FIG. 3. RA-1253 amplifier schematic.

quencies from 50 to 7500 cycles at an output of +12 dbm, which is a
nominal value for the modulator overload.

The maximum output amplitude is limited at +18 dbm to protect
the light valve from accidental overload which might cause damage.
The limiting point occurs 6 db above nominal modulator overload so
that the distortion generated by the peak chopping action is not ob-
served. This limiting action is accomplished by a cold cathode type
of voltage regulator tube placed in shunt with the feedback resistor.
When the peak signal voltage exceeds the ignition voltage of this
tube the amplitude is effectively limited by a corresponding increase
in the amount of feedback. So that the blocking condenser associ-

136

F. L. HOPPER AND R. C. MOODY

Vol 47, No. 2

ated with the cold cathode tube will not accumulate a charge, a high
resistance is placed across the tube. Limiting is equally effective on
both halves of the wave. The cold cathode regulator tube may be
visually observed by the operator through a bull's eye on the front
panel. Action of the tube, i. e., peak chopping, is indicated by the
visual glow discharge of the tube when the output of the amplifier is
sufficient to cause it to operate. Curve A of Fig. 4 shows the ampli-
fier distortion at the threshold of amplitude limiting.

FIG. 4.

Two microphone input positions are provided. By means of an
adaptor either of the microphone circuits may be connected to a trans-
mission line or bridging bus. Low level mixing is employed in the
interests of simplified design and reduction in size, weight, and num-
ber of components. The mixer inputs are designed to operate from
a nominal 30-ohm impedance, thus accommodating most types of
microphones. A switch-type interstage gain control for large gain
adjustments is provided on the front panel. Maximum gains of 95,
80, and 65 db are available.

Dialogue equalization is obtained by screen grid degeneration. A
4-position switch placed on the front panel gives zero-, 4-, 7-, and 10-
db equalization at 100 cycles. Degenerative equalization does not

Aug. 1946 SIMPLIFIED RECORDING TRANSMISSION SYSTEM

137

impair the circuit stability and improves it at low frequencies. The
gain-frequency characteristics of the amplifier and the 3 steps of dia-
logue equalization are shown in Fig. 5.

Fig. 5 also shows a modulator equalizer characteristic which com-
pensates for a particular modulator resonance characteristic. The
equalizer is the constant resistance type so that the amplifier works
into a constant load at all frequencies. By the same token, the mod-
ulator is supplied from a constant resistance generator, hence no iso-
lating pads are required, and the volume indicator may be bridged
across either end of the circuit with the assurance that correct read-

FIG. 5.

ings will be obtained at all frequencies. The equalizer has a fre-
quency response characteristic which is complementary to that of the
light valve. The combination of equalizer and light valve is there-
fore uniform with frequency, and material improvement in transient
performance of the light valve is realized.

Headset monitoring is provided by means of a high impedance
bridge across the output of the amplifier. Either the 705-type
moving coil headset, or the 71 3- type molded earpiece receiver may be
accommodated.

The noise-reduction circuit is of the carrier type and has been
greatly improved from a standpoint of simplicity of adjustment and
operation.

138

F. L. HOPPER AND R. C. MOODY

Vol 47, No. 2

The modulating voltage is obtained from a triode-diode amplifier-
rectifier and a timing filter. The grid of this tube is supplied from a
margin potentiometer which is bridged across the amplifier output.
The plate circuit of the tube supplies a transformer which in turn pro-
vides full wave connections for the diode. Bias threshold is obtained
by adjustment of the cathode resistor. Fig. 6 shows three possible

10
9

7
6
5

4
3

-2
-3
-4

-5
-6

-7
-8
-9
-IP
-II
-12

;

-T-

I • .

X -5f -

— -

-. j

X^x

-r-

r*S

^X*^

-•::.

• :i • 1

~iO!

s? ' f

if-

^E^QX

--p ;

-£

***

:4_

•a

a
a

— r-

•4

4- P

_| :.

v-^-4-^

/• t

— -

HS1

,:«

EDI

JCLI

I0f

J,;.F

>ER

tpt

Efc=-:_

-V

/]J E

j-;

)WII

«3"

^EE

biR

-^P

"iinii

*£»

£--

-P

-jA -;-{ ~|— -

;

ric

•

»F

•>

« „

Bit

'••

B (

1 P

R
4AR

DIN

•N

i--

.„:....

TTv

-f-

..L..

-84

-82

-80 -78 -76 -74
FIG. 6.

-72 -70 -68 -66

threshold adjustments. Fig. 7 shows three possible margin adjust-
ments. While it is apparent that a wide variety of threshold and
margin conditions are available, it is expected that proper adjust-
ments will be selected for any one type of modulator and film. Such
adjustments do not appear upon the control panel, and must be made
internally. The operation of the system is sufficiently stable so that
these adjustments need be made only infrequently.

Aug. 1946 SIMPLIFIED RECORDING TRANSMISSION SYSTEM

139

The timing filter shown in the schematic diagram, Fig. 3, is used
for standard track recording. Push-pull filters, or filters having dif-
ferent attack and release times, may of course be substituted.

A single suppressor gird pentode tube is used as the oscillator and
modulator. The total current consumption of this tube is less than
one milliampere. The oscillator is of the conventional Hartley type

-12

with a center tapped coil. This makes the feedback factor /3 positive
and unity. The consequent mismatch between tube and tank cir-
cuit improves the frequency stability. There is no need for oscillator
kc tuning as the tank coil and condenser are selected to 30 =*= 2 kc tol-
erance in manufacture. The oscillator output is sufficiently stable
so that there is no necessity for adjustment over long periods of time.
Modulation is applied to the third grid of the suppressor grid pen-
tode. No critical bias conditions are required for operation as a mod-
ulator. Only 2 v peak are required for complete carrier cancellation.

140

F. L. HOPPER AND R. C. MOODY Vol 47, No. 2

The output of the oscillator-modulator is of the order of 12 v max-
imum. This is applied to a potentiometer. The potentiometer
slider is connected tp a pentode carrier amplifier. The carrier output
transformer is uniform =*= 10 per cent from 20 to 40 kc so that any oscil-
lator drift from a 30-kc mean frequency does not affect the output.
Feedback is applied from a third winding on the transformer in series
with the output pentode cathode. The effect of this feedback is two-
fold, the linearity of the copper-oxide rectifier is greatly improved and
the tube constants have little effect on the output current. The out-
put of the copper-oxide rectifier is filtered with a 500 mfd condenser.
This noise-reduction circuit is capable of supplying 300 milliamperes

FIG. 8. RA-1254 power supply schematic.

to a 1.5-ohm load. Total current drain of the noise reduction is 11
milliamperes at 180 v.

The total power consumption of the amplifier-noise reduction is
one ampere at 6 v and 22 milliamperes at 180 v.

Power for the system may be obtained from a-c operated rectifiers
when it is associated with a film recorder which utilizes such power.
For use with newsreel systems where the primary power source is a
12-v storage battery, a dynamo tor is provided to supply plate poten-
tial. Thus, the battery supplies power for the camera motor, re-
corder lamp, amplifier tube heaters, and the dynamotor. The dyna-
mo tor is small, weighs less than 3 Ib and requires only 18 w for its
operation. Fig. 8 shows power supply schematic and indicates that
180-v regulation is secured through the use of two regulator tubes.
These draw only 8 milliamperes on the average. In addition, a bal-

Aug. 1946 SIMPLIFIED RECORDING TRANSMISSION SYSTEM 141

last lamp is used in series with the vacuum tube heaters, insuring
proper operating voltage to the tubes. With these regulating pro-
visions, system performance is consistent over the useful range of the
storage battery voltage from full charge to discharge. Fig. 9 is a
photograph of this particular power supply unit.

Two 10-ft, 6-conductor cables are used between the amplifier-
noise reduction unit and the battery power supply. An additional
6-conductor cable is connected from the power supply to the modu-
lator on the camera.

•••••••••••••••••••••1

FIG. 9. RA-1254 power supply.

The described system affords in a single compact, lightweight unit,
a transmission system capable of excellent performance together with
reliable and simple operation.

REFERENCES

1 HOPPER, F. L., MANDERFELD, E. C., AND SCOVILLE, R. R.: "A New High-
Quality Portable Film Recording System," /. Soc. Mot. Pict. Eng., XXVIII, 2
(Feb. 1937), p. 191.

2 HOPPER, F. L., MANDERFELD, E. C., AND SCOVILLE, R. R.: "A Lightweight
Sound Recording System," /. Soc. Mot. Pict. £ng.,XXXHI,4 (Oct. 1939), p. 449.

THE PHOTOMETRIC CALIBRATION OF LENS APERTURES*

ALLEN E. MURRAY**

Summary. — An absolute and physically sound method developed at Bausch &
Lomb is described for the photometric calibration of lens apertures.

Essentially the method consists in comparing the total flux from a depolished opal
glass aperture with the flux through a given lens at a definite stop opening when focused
on the opal glass aperture. An integrating sphere is used to collect the flux in the
two cases and readings are made proportional to the flux with two matched barrier
layer photocells. The theoretical development and some numerical results are given.

It is well known that the square of the ratio of the diameter of the
entrance pupil to the equivalent focal length of a lens is an inadequate
characterization of its transmission. For one thing, the individual
surfaces reflect a certain amount of light, which is thus lost to the
image, though some may be passed eventually to the image plane
as "flare." In addition, the glass elements absorb a certain amount
and the natural vignetting takes its toll so that the total transmission
over a finite image area is relatively less than over a small area about
the lens axis. Further, the unavoidable manufacturing tolerances
conspire to make the marked //values only approximately representa-
tive of the relative amount of light transmitted to the image. In
some instances, the latter cause alone may lead to about 10 per cent
error in //number1 even in closely controlled lens manufacture.

These problems have been appreciated for some time by the cine-
matographers, who must exercise close control over exposure to in-
sure the desired photographic quality. Numerous schemes2 ~4>1° have
been proposed in the past all looking to establishing a method of
measuring the transmission of an objective over its range of //values,
thus to the setting up of a series of effective transmission values more
closely correlated with the exposures required to produce a given
density than are the geometrical //numbers. These schemes are all
arbitrary in that the proposal is made either to measure the trans-

* Presented May 10, 1946, at the Technical Conference in New York.
** Scientific Bureau, Bausch & Lomb Optical Co., Rochester, N. Y.

142

PHOTOMETRIC CALIBRATION OF LENS APERTURES 143

mission of the test lens against the transmission of a geometrically
defined aperture, or against a lens chosen as standard.

Both of these proposals have disadvantages, though the former, by
techniques herein indicated, could be made to provide a satisfactory
standard method of photometric calibration. The objections to the
latter method are so obvious they will not be noted beyond indicating
that the method is valid essentially for only one laboratory.

In the search for a method which would be universally valid, and
which could be used in any laboratory desiring to build the equipment
and which would yield significant results without the necessity of
standardization, the present physically sound, absolute method of
photometric lens calibration was developed at Bausch & Lomb. This
method does not depend on arbitrary apertures or lenses, and yields
immediately either the effective //number or, assuming the //number
is marked, the more elegant transmittance5 which is known8 to be a
function of the geometrical //number.

Essentially, the method consists in comparing the total flux trans-
mitted by a lens at a given stop with the total emitted by a Lamber-
tian aperture. The ratio of these two gives a quantity proportional to
the transmittance of the lens and inversely proportional to the square
of twice the //number.

In practice, a depolished opal glass aperture, which is masked to the
standard 35-mm sound film aperture dimensions, is placed flush with
the wall of the integrating sphere. The total flux from the film aper-
ture is thereby measured. The second step consists in introducing
between the aperture and the sphere opening the lens under test,
which is focused on the aperture by autocollimation. The flux
through the objective at the various stops is collected by the inte-
grating sphere, thus yielding in conjunction with the previous meas-
urement, the desired primary data.

Theoretical. — The total flux emitted by a perfectly diffusing
source of area A and brightness B is7 F = irBA . Now, if the area is
small, and we introduce before it at a distance r a parallel circular
area of radius h, the solid angle subtended by the second area at the
position of the first is simply (h/r)z and the total flux passing through
the second area is given by the amount of the flux emitted and this

solid angle

F = *BA(hW.

(This development is tantamount to assuming that all distances are
large enough that the inverse square law is sufficiently accurate?

144 A. E. MURRAY Vol 47, No. 2

However, the corrections are not difficult to derive8 and can be shown
to amount at most to a fraction of one per cent with magnitudes of
present concern.)

If a lens is now focused on the film aperture of brightness B and
area A, h in the previous equation becomes the radius of the exit
pupil and r the equivalent focal length of the lens, assuming that the
pupils are very close to the principal planes. Since the exit and en-
trance pupils are conjugate to one another, the radius of the latter can
be substituted for the former, and the flux through the lens becomes

or, upon introducing the //number defined as the ratio of the focal
length to the diameter of the entrance pupil, and introducing also
the transmittance K, the flux through the lens becomes

Or, in terms of the total flux emitted by the film aperture

Fo = irBA

the flux transmitted by a lens at a given stop is

It is clear that the ratio F/FQ is a measure of the photometric effi-
cacy of the lens, and may be used to define the effective stops ("photo-
metric //numbers," or, as suggested by Berlant,4 "//numbers").
In accordance with the traditional relationships defining the geo-
metrical //numbers, the photometric //numbers are defined in the
following manner :

(photometric //number) = /„ * =

and thus the relationship between the "photometric //number" and
the traditional geometrical //number becomes

f*- A.

Jn -7=-

To avoid all questions of possible circuit nonlinearity and photocell
response drift, a balanced 2-cell circuit was chosen whereby to meas-
ure these fluxes. The lamp house is provided at the rear with a

Aug. 1946 PHOTOMETRIC CALIBRATION OF LENS APERTURES 145

window of depolished opal glass similar to that used at the film aper-
ture on the latter of which the lens is focused, the flux in the inte-
grating sphere being measured in terms of the brightness and area of
the comparison window and the distance of the comparison photo-
cell from that window.

Using the same notation as before, the flux through the face of the
comparison photocell of radius hc, at distance r from the window of
brightness BQ and area A 0 is

F =

Now in the null condition, when the film aperture is part of the sphere
wall and a balance is attained, the flux through the comparison cell
face is

And when a lens is being measured, at balance

where it is assumed that both the brightness and the area of the com-
parison window may be changed to accommodate the scale used on the
instrument. Forming the ratio of these two fluxes as on the other side
of the lamp house, and since the corresponding fluxes measured on the
two sides are assuredly proportional to each other, the working equa-
tion becomes in terms of primary parameters

It was discovered in practice that the areas involved at the rear of
the lamp house were so small that the highly desirable use of circular
diaphragms whose diameters could be measured accurately was in-
feasible, so it becomes necessary to use neutral filters to modulate the
light on passage from the null condition to the condition of measure-
ment. Under these circumstances, if T represents the transmittance
of the filter

BI «• TB0 AI — AQ

and the working equation becomes in terms of observable parameters

>o\»

146

A. E. MURRAY

Vol 47, No. 2

from which either the transmittance K of the lens can be derived im-
mediately, using the engraved aperture stops, or the effective photo-
metric //number can be found. The relationships involved are ob-
viously

and /„* =

Practical Realization. — The apparatus built in our laboratory is
composed essentially of an integrating sphere to collect the flux,
a lamp house containing a 500-w lamp, and a movable bracket for

FIG. 1. The lens calibration equipment.

the comparison photocell, whose position can be measured with
respect to the rear of the lamp house.

The integrating sphere is 213/4 in. in diameter, and from a previous
user we inherited a 3-in. port with a cylindrical projection which
serves very well to support the sphere on the optical bed. The sphere
is baffled inside so that no direct light from either the aperture or the
intercepted beam can strike the photocell. It was found that this
still was insufficient, and that the addition of an auxiliary collecting
sphere of 8-in. diameter sufficed to smooth out the distribution of
light so that the same conditions of illumination on the photocell pre-
vail whether the lamp house aperture is in place, or the flux through
a lens is being measured.

Aug. 1946 PHOTOMETRIC CALIBRATION OF LENS APERTURES 147

The lenses under test are supported in a hinged rack and pinion
mount with a set of adapters for different lenses. The lens support
can be swung out of the way of the lamp house for establishing ro.
In focusing a lens, a plane mirror is fastened over the front of the lens
and the reflected image focused to sharpness on the matte white face
of the lamp house muzzle. This insures that the film aperture is in
the focal plane of the lens.

The lamp house is built with a projection fitting into the inte-
grating sphere. This projection is long enough to place the depolished
opal glass film aperture flush with the sphere wall, the condition re-
quired for accurate measurement of the total flux issuing from the
film aperture. The face of the projection is lacquered with the same
matte white lacquer used inside the sphere and lamp house. Forced
draft ventilation cools the lamp. The rear of the lamp house bears
another depolished opal glass aperture, whose exposed area can be
varied by the insertion of masks. This area faces the comparison
photocell, which is carried on a movable bracket carrying an index,
so that the distance between the comparison aperture and the com-
parison cell can be measured. These distances are read on the cali-
brated optical bed.

In the preliminary stages of this development it was decided to use
the logically unobjectionable photometric balance system of meas-
urement. With this system and two sufficiently well-matched photo-
cells, virtually all sources of primary photometric and electric inac-
curacies can be caused to drop out of consideration, leaving a sound
physical measurement of flux.

This whole system of measurement is more suitable in the labora-
tory instrument than in factory equipment, because of (a) the several
separate measurements entering into the evaluation, and (b) the
fatigue of the photocells. The cells fatigue inevitably at different
rates, and the balance point shows a drift in time whose rate is variable
with illumination level, which requires a delay in reading increasing
with the desired accuracy.

Two Weston Model 594YRO type 3 photoelectric cells are used in
the balanced current bridge type of circuit, which is superior to the
traditional potentiometric bridge because of the nonlinear response
curve of the barrier layer cells. The circuit is shown in Fig. 2.

It was quickly discovered that the edges of the originally preferred
apertures which were used to modulate the flux on the comparison cell
side of the lamp house could not be made sufficiently nonreflecting to

148

A. E. MURRAY

Vol 47, No. 2

serve for a geometrically defined modulator in the required diameters
of the order of one mm. Furthermore, these small apertures empha-
size the local inhomogeneities in brightness of the second opal source.
For these reasons it was felt that the light could be modulated
more accurately and reproducibly by means of neutral filters altering
the brightness of the fixed aperture. Evaporated metal neutral
filters of very flat spectrophotometric characteristics were used to ac-
complish this end. These filters are inserted in a holder between the
comparison aperture and the photocell, and have been found to serve
very well. The filters are so oriented that the reflected light from the
comparison aperture falls clear thereof.

•AAAAA/

FIG. 2. The electrical circuit.

To gain a measure of the suitability of the independently calibrated
filters for this application, a series of measurements was made with
a set of filters whose transmittances were accurately measured in
visible light. A set of blank apertures was placed in the lens holder
at a definite distance from the film aperture, and the flux measured in
the customary way. The mean transmittance of the blank apertures
under these circumstances was 0.992, demonstrating the feasibility of
using the independently measured transmission values with the
barrier layer cells.

In practice, densities of the order of 2.5 are needed, which are diffi-
cult to measure with accuracy on standard equipment. The present
apparatus as designed can be adapted to the measurement of the trans-
mittances of filters of this density fairly easily by suitable modulation
of the flux through the front of the lamp house, but it proved expedient
to measure the factors of the denser filters by using blank apertures.

Aug. 1946 PHOTOMETRIC CALIBRATION OF LENS APERTURES

149

The consistency attained in a series of measurements is easily
8 per cent in transmittance, or I/IQ stop,9 and in careful work there is
no trouble in attaining 1/zo stop. The methods required to improve
this consistency, and through it the accuracy of the equipment, are
straightforward.

The flux radiated from the film aperture when it forms part of the
sphere wall is mathematically equivalent to the flux through a perfect
lens of aperture ratio //0. 5, the theoretical maximum. Our technique
is a generalization of the method using an arbitrary standard blank
aperture at a given distance from the film aperture, since our standard
stop (0.5) is optically defined.

The method, then, is absolute, depending however on the meas-
urement of the transmittance of the neutral filters used to modulate

LAMPHOUSE

COMPARISON
PHOTOCELL

FIG. 3. Disposition for measurement of r0.

the brightness of the comparison aperture. This can be done in the
same equipment as used for the lenses without begging the question,
or more expediently by calibrating against blank apertures. (The
filters could be eliminated entirely with a fixed comparison aperture
by using the inverse-square law, if very bulky equipment were ac-
ceptable. A range of at least 100 : 1 in the comparison aperture-photo-
cell distance would have to be provided.)

A series of B altar lenses of different focal lengths was measured with
the following results :

2.3 2.7 2.8

5.6 8

152-mm//2.7 /„* ... 2.97

BS 972

Filmed

Nominal 0.93 K 0.83

4.04 5.70 8.30 11.38

0.98 0.96 0.96 0.93

150

152-mm//2.7 /»*
BF2564
Unfilmed
Nominal 0.75 K

100-mm//2.3 /»*

VA 5884

Filmed

Nominal 0.90 K

100-mm//2.3 /„*
VA 5596
Unfilmed
Nominal 0.63 K

40-mm //2.3 /»*

BF4256

Filmed

Nominal 0.90 K

40-mm //2.3 /»*
BF4211

Unfilmed
Nominal 0.66 K

A. E. MURRAY

3.32

4.46

Vol 47, No. 2
6.30 9.13 12.6

0.66

...

0.78

0.79

0.77

0.76

2.57

2.98

4.15

5.66

8.19

12.37

0.80

0.94

0.93

0.98

0.95

0.79

2.98

3.35

4.82

6.64

9.63

14.01

0.59

0.70

0.69

0.70

0.69

0.62

2.70

2.96

4.13

5.86

8.03

11.07

0.74

0.92

0.97

0.91

1.00

0.99

3.10

3.38

4.65

6.31

9.12

11.27

0.56

0.70 0.76 0.78 0.77

0.89

In those cases where the transmittance was measured at certain
stops by aid of two neutral filters, the mean of the measurements
is reported. The "nominal" value of the transmittance is the value
measured on a transmissometer of conventional design. The values

COMPARISON

APERTURE
LAMPHOUSE

NEUTRAL

FILTER

COMPARISON
PHOTOCELL

FIG. 4. Disposition when the flux through a lens is being measured.

in the table above include all sources of inaccuracy of setting of the
diaphragm except that of backlash, since the diaphragms were all set
from the same side (from the larger apertures) . Notable in these re-
sults is the fact that without exception the transmittance over the
film aperture is markedly lower at the maximum opening than at any
other stop.

Aug. 1946 PHOTOMETRIC CALIBRATION OF LENS APERTURES 151

It is clear that the method here described provides information re-
garding only the "photometric efficacy." It is per se insufficient to
standardize the measure of transmission (photographic //number or
"//number"). This can be achieved through an arbitrary choice of
transmittance, such that the measure of transmission is sufficiently
close to present practice. From our observed transmittances and
from the calculated values for lenses with six to eight air-glass sur-
faces a value in the range 0.60 to 0.75 would be close to current prac-
tice.

The proposal made by Berlant4 is essentially the one made here.

ACKNOWLEDGMENT

The author wishes to acknowledge his indebtedness to Dr. W. B.
Ray ton, Director of the Scientific Bureau, at whose request the prob-
lem was undertaken, and whose ever-resourceful interest lias been a
continual source of encouragement. He wishes also to acknowledge
the help given in many discussions with Dr. K. Pestrecov, and the
assistance of Mrs. M. Tarplee in making tedious readings, and that
of the Misses L. Frey and B. Marble in reducing the observations
and in making numerous auxiliary computations.

REFERENCES

1 PESTRECOV, K. : Private communication (1945) .

2 CLARK, D. B., AND LAUBE, A.: "Twentieth Century Camera and Acces-
sories," /. Soc. Mot. Pict. Eng. 36, 1 (Jan. 1941), p. 50.

3 SILVERTOOTH, E. W.: "Stop Calibration of Photographic Objectives," /. Soc.
Mot. Pict. Eng. 39, 2 (Aug. 1942), p. 119.

4 BERLANT, E.: "A System of Lens Stop Calibration by Transmission,"
J. Soc. Mot. Pict. Eng., 46, 1 (Jan. 1946), p. 17.

6 The term "transmittance" is used for the ratio of the light transmitted to the
light incident in accordance with the Optical Society of America Committee on
Colorimetry Report, J. Opt. Soc. Amer., 34, 4 (Apr. 1944), p. 184.

6 MARTIN, L. C.: "An Introduction to Applied Optics," Pitman & Sons,
(London), 1932, p. 206.

7 HARDY, A. C., AND PERRIN, F. H.: "The Principles of Optics," McGraw-
Hill Book Co., (New York), 1932, p. 272.

8 WALSH, S. W. T.: "Photometry," Constable (London), 1926, p. 102.

9 As here used, a fractional stop is understood to mean the ratio given by the
fractional power of \/2, the ratio between two geometrical full stops; e. g., Va-
stop difference corresponds to the ratio 21/4 between the fractional stops, Vio stop
21/*, etc.

10 DAILY, C. R.: "A Lens Calibrating System," /. Soc. Mot. Pic. Eng., 46, 5
(May 1946), p. 343.

A NEW FILM FOR PHOTOGRAPHING THE TELEVISION
MONITOR TUBE*

C. F. WHITE** AND M. R. BOYERf

Summary. — A film which is specially adapted for photographing images on the
P-4 monitor tube surface has been prepared. Optical sensitization is adjusted to
yield peaks of sensitivity within the blue to yellow spectral region corresponding to the
emission of the P-4 screen. Resolving power of the film has been found of controlling
importance when used in 16-mm size and this factor has affected the choice of emulsion
for this purpose. The film may be employed either as a negative or reversed.-

As soon as the cathode-ray tube came into use in the laboratory, it
became obvious that records of the traces would be most valuable.
In the past, numerous articles have been written on the photography
of cathode-ray tube traces; Morse,1 Feldt,2 and recently Goldstein
and Bales,3 have reported on the various films suitable for this type
of photography.

A specific application arises in the motion picture photography of
the P-4 phosphor screen at an exposure time of V«o of a second as re-
quired by the present 525-line television transmission. This should
be accomplished with currently available lenses, at reasonably small
apertures, and with tube voltages which allow adequate tube life.

Resolving Power. — In numerous 16-mm records taken of pictures
on the P-4 tube, it was noted that films now commonly sold as ' 'high
speed" did not give satisfactory pictures. Apparently this resulted
from lack of resolution of the picture currently transmitted on 525-line
television. This is surprising since a study of the published resolv-
ing power figures on currently available 16-mm films indicated that
films of lowest resolution were theoretically capable of resolving
all the transmitted lines. This is shown by the following calculations.

On a 525-line transmission 10 per cent is lost owing to blanking
time, leaving the net received lines as 473. Taking these on a 16-mm

* Presented May 10, 1946, at the Technical Conference in New York.
** Research Division, tSales Research Division, Photo Products Department,
E. I. du Pont de Nemours & Co., Inc., Parlin, N. J.
152

PHOTOGRAPHING TELEVISION MONITOR TUBE 153

154

C. F. WHITE AND M. R. BOYER

Vol 47, No. 2

film, and using the American Standards Association4 standard pro-
jector aperture height as 7.2 mm the required resolution of the film
is 66 undivided lines per mm. Since all film data on resolution are
given in double lines, or a black line plus an equivalent white
space, the "photographic resolution" required of the 16-mm film is 33
lines per mm. This is considerably less than the published resolu-
tion of the highest speed film.

One explanation for the apparent discrepancy between the observed
facts and the estimate formed on the basis of published figures

CURVE I. ZNS, <NO ACTIVATOR)
2 " 0.002 % Ac
3. " 0.032%A6

4000 4500 5000 5500 6000 6500 7OOOA
WAVE LENGTH
FIG. 2.

seemed to lie in the method of determining resolving power in photo-
graphic emulsions.

When an emulsion is evaluated for resolution, the object to be
photographed is normally one of high contrast, of the order of 200:1.
It is usually composed of black lines on a transparent background or
white lines on an opaque background or some combination of the two.
A conventional resolving power chart is shown in Fig. 1 .

However, in considering the contrast of pictures on the television
tube, Beers5 states that for a 441 -line picture the contrast of large
areas can be considered as 50: 1 and for small areas as 10:1.

Aug. 1946 FILM PHOTOGRAPHING TELEVISION MONITOR TUBE 155

CURVE I.-4ZNO. 2BEO. 3Si02,(!yMN. l200°C-6p'
3.L (3)MN.' "

4500 5000 5500 6000 6500 7000 7500A

WAVE LENGTH
FIG. 3.

CURVE I.-9ZNO. BeO. 6Si02. 8MN, IIOO°C.-60'

2- I200°C.- "

3- " I250°C-"

4500 5000 5500 6000 6500 7000 7500-A
WAVE LENGTH

FIG. 4.

156

C. F. WHITE AND M. R. BOYER

Vol 47, No. 2

This figure 10.1 for small areas was probably not based upon areas
so small as to approach closely the limit of resolution of the systems
and a decrease below the 10:1 figure would be expected in any system,

100

3000

7000 A

SPECTRAL DISTRIBUTION OF THE EMISSION RADIATION OF A P- 4 SCREEN

FIG. 5.

UNSENSITIZED

ORTHOCHROMATIC

PANCHROMATIC

ORTHOCHROMATIC

PANCHROMATIC

V V 1

4000

5000

6000

7000 A

WAVE LENGTH

OPTICAL SENSITIZATION

FIG. 6.

as the areas considered approach the resolution limit. In addition,
the mathematical studies of Cawein6 on the relation of contrast to
television bandwidth suggest still further reduction in brightness
ratio when increasing the number of lines transmitted from 441, for

Aug. 1940 PHOTOGRAPHING TELEVISION MONITOR TUBE

157

which the figures were given, to the present 525 lines for equal total
transmission band widths.

The requirement for the resolving power of an emulsion to photOr
graph the 525-line television tube evolves as 33 lines per mm at con-
trasts well under 10: 1. This is considered as of primary importance.

No mention has been made of the effect of the lens on resolution,
because it has been assumed in this discussion that the lens will be
good enough to take care of a resolution in excess of 33 lines per mm.

Exposure Time. — In most of the data previously published, the
exposure time could either be long or extremely short, as compared
with the exposure time required for recording the image on the tele-

030 0.60 0.90 1.20 1.50 1.80 2.10 2.40 DENSITY Do
2 4 8 16 32 64 128 256 CONTRAST C

VARIABLE OBJECT CONTRAST-NEGATIVE PROCESSING

FIG. 7.

vision tube. The recording of the tube image must be done in
Yso of a second. Longer times may be used so long as they are mul-
tiples of this figure, but the longer the exposure time, the more blurred
will be any rapid picture action on the tube.

Speed. — Assuming that the exposure time is fixed at Vao of a sec-
ond, and the resolution at 33 lines per mm for a low-contrast ob-
ject, the final requirement of a satisfactory emulsion is speed to the
P-4 screen.

The P-4 screen is chosen, not because it happens to be the type
most commonly used in receiving tubes, but because at the present
time the monitor tube must be set by eye for the best quality pic-
ture. Since judgment of quality is based on experience gathered from

158 C. F. WHITE AND M. R. BOYER Vol 47, No. 2

o o

§ I I

I I I I

co S to ^

a s
s

1 <?
I .« I

fe Co

N N

> «-> 5 fci bo

1 > 9 > |
cH § o3 S

« '% ™

8 3 g 6

« « § s

I -S I I ^

1 ^ o3 3 $

_J_ 42 CO PQ

S 8 S

•?4 '-^ •«»

N N N

.y -*&

Q ^

Aug. 1946 PHOTOGRAPHING TELEVISION MONITOR TUBE

159

viewing black-and-white pictures, it is most natural to work with a
screen as nearly white as it is possible to obtain. It is conceivable
that after a period of time it would be possible for a particular person
to adjust a green or blue tube for picture quality just as readily as a
white tube.

It is important to recognize that the Radio Manufacturers Associ-
ation designation of a P-4 screen has no meaning at the present time
so far as spectral emission is concerned. As many as 20 P-4 screens
have been compared visually at one time and no two matched for
color.

80
70

5 .

0.30 0.60 0.90
248

',1°

'ff

I 80 2 10 2.40 DENSITY D.
64 128 256 CONTRAST C

VARIABLE OBJECT CONTRAST-REVERSAL PROCESSING

FIG. 8.

Because emulsion speed is dependent on the spectral quality of the
exposing radiation, the emission of the P-4 screen must be considered.
The difficulties involved will be recognized by a study of the following
data taken from Krushel7 (Table 1, Figs. 2, 3, and 4).

The RMA designation of a P-4 screen is shown in Table 1.

The curves in Fig. 2 show the change in emission spectra as silver
activator is added to a ZnS phosphor.

Also, since the P-4 screen is a mixture of Zn sulfide and Zn Be
silicate, the change in emission of the Zn Be silicate as changes are
made in the manganese activator concentration is shown in Fig.
3. Fig. 4 shows the relation of crystallization temperature to emis-
sion.

As can readily be seen from the above figures and from recognition
of the fact that the proportions of Zn sulfide and Zn Be silicate are not

160

C. F. WHITE AND M. R. BOYER

Vol 47, No. 2

specified, the exact spectral emission of the P-4 cannot be given.
However, the ratio of ZnS to Zn Be'SiOs is usually high and emission
in the blue region will accordingly be high. Also, as higher tube

FIG. 9.

voltages are employed in the future to obtain greater brilliance the
emission may be further shifted toward the blue. Finally, it can be
said that the amount of emission in the red is very low. A typical

„ 10000
E 1000

« 100

1 10

* 0
35C

^ ^

> '

0 4000 4500 5000 5500 6000A
WAVE LENGTH

FILM TYPE 323-DAYLiOHT EXPOSURE
FIG. 10.

emission curve of a P-4 screen is given in Fig. 5. Thus we find that
the film for photographing images on such a surface should possess
the highest possible blue sensitivity while taking full advantage of the
green and yellow emission.

Aug. 1946 PHOTOGRAPHING TELEVISION MONITOR TUBE

161

An additional factor influencing the choice of film for photographing
the P-4 tube is found in phenomena associated with optical sensiti-
zation of emulsions. As sensitization is carried farther and farther
toward the red end of the spectrum, the blue sensitivity is often de-
creased. This is illustrated in a diagrammatic manner in Fig. G. This
factor, coupled with the foregoing information concerning tube
emission, suggests the probability that an orthochromatic emulsion
will be most suitable for photographing the P-4 tube.

The above data establish to a certain degree the necessary resolu-
tion and spectral sensitivity of a motion picture film for photographing
the P-4 tube receiving 525-line television.

100

3000

SCREEN

EMISSION

10000
1000
100
10

4000

5000
WAVE LENGTH

6000

7000A

P-4 SCREEN EMISSION AND FILM SENSITIVITY
FIG. 11.

The following data indicate the way du Pont Emulsion type 323
meets these requirements.

Resolution with Type 323. — Published data (Sandvik)8 show that
the resolution of a film is markedly affected by the contrast of the
test object and as we have pointed out, the object contrast in 525-
line television is very low in regions of fine detail. Tests were made
to obtain data on the specific emulsions used here, and processing was
extended to include reversal development. Results are shown in
Figs. 7 and 8. It will be noted that the resolution falls with the con-
trast, and that a higher white light speed film (type 301) is worse than
either types 314 or 323, the ortho film.

162

C. F. WHITE AND M. R. BOYER

Vol 47, No. 2

Further tests were made by projecting into a television system the
slide shown in Fig. 1. Great attention was given to maintain con-
stancy in the system and from emulsion to emulsion. Photographs of
the screen were taken and processed by reversal. A frame of the re-
sulting picture on type 323 is shown in Fig. 9.

Labelling the four center blocks 1, 2, 3, 4 clockwise and beginning
with the upper left-hand block, the resolving power for type 323 in
lines per mm at//2.0 is as follows :

Block 1 — 15 lines per mm (white vertical lines, black background)
Block 2 — 15 lines per mm (black vertical lines, white background)
Block 3 — 20 lines per mm (white horizontal lines, black background)
Block 4 — 20 lines per mm (black horizontal lines, white background)

FIG. 12.

The difference in the resolution of the vertical lines (Blocks 1 and
2) and the horizontal lines (3 and 4) should be noted. This ap-
parently is a confirmation of the calculations of Cawein and Hartley.

Considering the lowering in contrast owing to small areas and 525-
line television, the recording of 20 lines per mm seems quite adequate
and within 3 lines per mm of the observed image on the tube.

Speed of Type 323. — For the reasons given previously orthochro-
matization was used in type 323 and the spectral sensitivity is given
in Fig. 10.

A combination of the response of the film and the emission of a
P-4 tube is shown in Fig. 11.

Aug. 1946 PHOTOGRAPHING TELEVISION MONITOR TUBE

163

The speed of this emulsion was checked against du Pont Emulsion
type 314, which has an equivalent resolving power, by exposing to
two different P-4 screens, each at different voltages.

The first test was run on a television system by projecting into the
iconoscope, by means of a slide projector, the slide shown in Fig. 12.

The system, with 15,000 v on the tube, was set to reproduce all
shades in the 6-step wedge and measurements of the light output were
taken at this setting by means of a photocell and microammeter.
Keeping the same setting on the tube, and checking the intensity
regularly, the two films were exposed in a 16-mm camera at three
apertures. The films were reversed by machine.

2.00

150

t 1.00

.50

.00

RELATIVE LOG E
RELATIVE SPEED TO P-4 TUBE

FIG. 13.

Table 2 shows net reversed density values for five of the steps re-
corded with an//2 setting.

Film

323
314

0.24
0.39

TABLE 2
Net Densities-

0.36 0.98

0.54 1.21

1.60
1.80

1.71
2.00

A second speed check on a different P-4 tube was run as follows :

A raster was put on the tube and a neutral density wedge placed in
front of the exposing plate carrying the two films, types 314 and 323

164 C. F. WHITE AND M. R. BOYER

The films were exposed and developed as a negative. Fig. 13 is a plot
of these results.

The above two tests indicate that the new ortho type 323 is at least
one-half stop faster than the panchromatic type 314 to a P-4 tube
surface. These tests, coupled with actual motion picture recordings
in a synchronized camera further indicate that the new film has ample
speed to be useful under entirely practical television operating con-
ditions.

Conclusion. — The above tests and practical picture tests indicate
that du Pont type 323 is suitable for photographing the P-4 monitor
tube screen as it has a useful combination of resolving power and
speed.

REFERENCES

1 MORSE, R. $.: "Materials Available for the Photography of Cathode Tube
Traces," Electronics, XI (Apr. 1938), p. 37.

2 FELDT, R.: "Photographing Patterns on Cathode- Ray Tubes," Electronics,
XVII (Feb. 1944), p. 130.

3 GOLDSTEIN, H., AND BALES, P. D. : "High Speed Photography of the Cathode-
Ray Tube," Rev. Sci. Instr., XVII (Mar. 1946), p. 89.

4 A. S. A.— Z22. 14-1941.

6 BEERS, G. L., ENGSTROM, E. W., AND MALOFF, I. G.: "Some Television
Problems from the Motion Picture Standpoint," /. Soc. Mot. Pict. Eng., XXXII,
2 (Feb. 1939), p. 121.

6 CAWEIN, M. : "Relation of Contrast to Width of Television Band," FM and
Television, IV (Nov. 1944), p. 28.

7 KRUSHEL, I.: "Phosphors and Their Behavior in Television," Elec. Ind.t
IV (Dec. 1945), p. 100.

8 SANDVIK, O.: "The Dependence of the Resolving Power of a Photographic
Material Upon the Contrast of the Object," /. Opt. Soc. Amer., XVI (Apr.
1928), p. 244.

TELEVISION REPRODUCTION FROM NEGATIVE FILMS*

E. MESCHTER**

Summary. — The expected reproduction characteristics are examined for the
cases where film is included as one step of the television process. Features of per-
formance to be expected from both negatives and prints as image sources are predicted
from average characteristics of elements of the television system. A dynamic test
procedure for the investigation of the over-all reproduction curve involving film and
television is described. Actual tests confirm the theoretical prediction that a negative
film with a rising shoulder characteristic may provide superior television images.

General Background. — The end objective of the television process
is the production, on the picture tube, of an image which will be
pleasing to the observer. Into this term "pleasing" enter both sub-
jective factors, such as subject matter, state of observer's visual adap-
tation and contrast with nearby objects, and factors capable of exact
objective specification of which scene brightness values, electrical and
optical characteristics of the reproduction equipment are typical. It
is the purpose of this discussion to examine some of the objective fac-
tors entering into the production of television images derived from
photographic film as an intermediate, and particularly to study the
conditions peculiar to the use of negatives for television broadcasting.

Objectively, the television process attempts to achieve "straight-
line reproduction," in which, for every picture element, the logarithm
of the brightness of the picture tube face divided by the logarithm of
the brightness of the original scene is a constant. This ratio will be
referred to as the over-all contrast of the system. For perfect repro-
duction the value of the over-all contrast is one and the brightnesses
of the picture are proportional to the first power of the bright-
nesses of the original scene. However, this ideal often cannot be
achieved in practice, particularly for outdoor scenes, where the aver-
age brightness range is too great to be reproduced accurately by the
picture tube. Some compression of the picture brightness scale is

* Presented May 10, 1946, at the Technical Conference in New York.
** Research Division, Photo Products Department, E. I. du Pont de Nemours,
& Co., Parlin, N. J.

165

166

E. MESCHTER

Vol 47, No. 2

necessary in such cases and the over-all contrast will be less than one.
The discussion which follows is not limited to any particular value of
the over-all contrast, but may be applied to the study of any degree
of brightness scale compression or expansion.

Progress of a television image through the various electrical and
optical stages of the reproducing system usually involves at least two
nonlinear steps.

(1) The electrical output of the iconoscope (pickup tube) is not proportional to
the amount of light falling upon it.

(2) The light output of the kinescope (picture tube) is not proportional to the
electrical signal applied to it.

1.2

NEGATIVE FILM
CHARACTERISTIC

LOG SCENE BRIGHTNESS (EEL-)

FIG. 1. Characteristic curve of a normal negative
film.

However, when an original scene is imaged on the iconoscope and
reproduced on the kinescope these nonlinearities are of a nature such
as substantially to cancel each other out, and good quality reproduc-
tion with an over-all contrast of about one is obtained for scenes of
moderate brightness range.

If the original scene is first recorded on a negative, printed on a pos-
itive, and this in turn imaged on the iconoscope, little change is intro-
duced into the system. We know that the photographic process,
properly carried out, gives a result in which the logarithms of the
brightnesses of the projected image are closely proportional to those
of the original scene, and nothing is done in this case to disturb can-

Aug. 1946 TELEVISION REPRODUCTION FROM NEGATIVE FILMS

167

cellation of the equipment nonlinear! ties mentioned above. The final
image obtained from scanning positive film should therefore be of a
quality approximating that of direct pickup; it may perhaps exhibit
a slightly different over-all contrast, depending on the exact film
processing.

It is possible to record an original scene on a negative, image this
negative on the iconoscope and, by modification of electrical connec-
tions (essentially reversal of amplifier polarity) cause a positive image
to appear on the kinescope. This procedure is very attractive in

2.0

1.5

1.0

OVECALL PHOTOGRAPHIC

CHARACTERISTIC:

LOG SCENE BRIGHTNESS

(BEL)

1.0

1.5

FIG. 2. Over-all reproduction characteristic of a
negative and print system, showing density of print as a
function of original scene brightness.

some respects, notably those of film processing speed and simplicity.
However, the over-all characteristics of the electro-optical system
are considerably disturbed by the polarity change ; the curvatures of
the kinescope and iconoscope characteristics that formerly tended to
cancel now add, with the result that tonal values of the final image
are appreciably distorted.

The exact manner in which this comes about may be demonstrated
graphically, employing published average characteristic curves for
film, iconoscope and kinescope. Fig.l shows the characteristic curve
of a negative film, in which density of negative has been plotted
against log brightness of original scene. The distance 0 to 1.3 on the

168

E. MESCHTER

Vol 47, No. 2

abscissa corresponds to a range in brightness of 20 to 1, typical of a
rather flatly lighted scene, but which also represents about the max-
imum brightness range which can be handled by present television
systems.

If a print is made from this negative following normal cine pro-
cedures^the resulting film will have a characteristic similar to that of
Fig. 2. Print density varies almost linearly with the logarithm of the
brightness of the original scene ; reproduction is reasonably faithful,
with an over-all photographic gamma of 1.3.

ICONOSCOPE
CHARACTERISTIC

ILLUMINATION

FIG. 3. Average iconoscope characteristic (after
Zworykin and Morton).

Average characteristics of the elements of a television system are
shown in Figs. 3,* 4, and 5.* Fig. 3 shows that the electrical signal
produced by an iconoscope is proportional to the intensity of illum-
ination only for low values of the latter; saturation effects cause dis-
tinct departures from linearity at higher illumination levels.

The amplifier (in which term we will include all electrical elements
except the iconoscope and kinescope) is essentially linear, departures
in a good system being rather small. The change from curve A to
curve B of Fig. 4 represents the change in polarity required when
switching from positive to negative film in the pickup. Change of

* Data from Zworykin and Morton, "Television."

Aug. 1946 TELEVISION REPRODUCTION FROM NEGATIVE FILMS

169

slope of the amplifier curve represents a change in gain, while a trans-
lational shift corresponds to a bias adjustment.

The kinescope characteristic possesses even poorer linearity than
that of the iconoscope; this is shown by Fig. 5, in which the curve
exhibits no real straight line portion.

Another curve which will be helpful during the following discussion
is that of Fig. 6. This is merely a graphical representation of a table
of logarithms, giving directly the relation between the brightnesses
usually used in describing iconoscope and kinescope performances

FIG. 4. "Amplifier" characteristics, showing method
of indicating reversal of polarity.

and the "log brightness" values that are rather more convenient in
discussing the original scene and its photographic aspects.

These several characteristic curves can be combined to give an esti-
mate of the linearity of over-all reproduction from the original scene
through the successive steps of film, projector, iconoscope, amplifier
and kinescope.

Reproduction Through a Positive Transparency. — Consider first
an original scene, the brightness values of which are represented as
abscissas on the left quadrant of Fig. 7. The over-all photographic
characteristic (introduced originally as Fig. 2) is represented by
curve A in this quadrant, and print density values corresponding to
original brightnesses may be read off directly. When this film is

170

E. MESCHTER

Vol 47, No. 2

placed in a projector the (logarithm of the) illumination from each
element is directly obtainable from the density values, as indicated
by the arrows from axis Yl to axis Y2. Variation in brightness of
projector light is represented by merely sliding Y2 along Yl.

Axis Y2> in turn, may be considered as one of Fig. 6. This curve
already has been introduced to form the right quadrant of Fig. 7a.
The combination of Figs. 2 and 6 may be represented more conven-
iently as in 7b, where the double ordinate axis has been eliminated but
both density and log illumination scales have been retained.

KINESCOPE
CHARACTERISTIC

VOLTS INPUT

-30

-HO

FIG. 5. Average kinescope characteristic (after
Zworykin and Morton).

Continuing through the television system, the projector output be-
comes the iconoscope input. The iconoscope step can be added by
placing the abscissas of Fig. 3 along Xi of Fig. 7b, inverting Fig. 3 in
the process. The result, shown in Fig. 8, represents progress from the
original scene through the iconoscope.

Other elements of the complete system may be added successively.
The iconoscope output becomes amplifier input, as in Fig. 9. Am-
plifier output is fed to the kinescope and becomes kinescope input as
in Fig. 10. It is convenient to have final brightness on a logarithmic
scale; the reintroduction of the "logging curve" of Fig. 6 on the kine-
scope output accomplishes this (Fig. 11).

The logarithms of the brightnesses of the original scene and of the
reproduction on the kinescope screen now appear in the same figure,

Aug. 1946 TELEVISION REPRODUCTION FROM NEGATIVE FILMS 171

along axes A- A and C-C of Fig. 11. These should form the two axes
of a separate quadrant; this can be achieved by transferring values
of A- A straight down to B-B as in Fig. 12, and by carrying values of
C-C along the path shown to D-D.

The complete diagram is now ready for use. In order to represent
a normal situation the projector light (which determines the level of
iconoscope input) has been adjusted so that the entire useful range
of the iconoscope is utilized, and amplifier gain has been set so that
the whole inconoscope output swing corresponds to the useful kine-
scope input range (Fig. 13).

l-O

BEIGHTNESS (o« ILLUMINATION)

FIG. 6. Curve for conversion between brightness (or
illumination) and log brightness (or log illumination).

To determine the over-all reproduction curve select any original
scene brightness," represented by point M . Draw a vertical line to
the film characteristic curve, then a horizontal line to the "delogging"
curve, and so on to each curve successively as indicated by the arrows.
Point N on the kinescope output scale corresponds to M; these two
serve to locate P, which is one point of the over-all reproduction char-
acteristic. The process is repeated, starting with other values (Q,
R, S) on the scale of original scene brightness, to obtain other points
of the reproduction curve (T, U, V). When carefully carried out on
a large scale diagram the reproduction curve of Fig. 14 results.

For the positive transparency under consideration the values so de-
termined closely approach the ideal straight line. The tonal scale is

172

E. MESCHTER

Vol 47, No. 2

somewhat expanded in the center of the range and compressed at the
ends, but reproduction is good for the brightness range considered.
The deficiency actually takes the form of some loss of shadow and
highlight contrast and therefore detail, even for the limited bright-
ness range considered. Attempts to reproduce scenes of greater tonal
range will of course introduce more serious losses at the ends of the
scale.

Reproduction Through a Negative Transparency. — The graphical
procedure carried out for the case of a positive transparency may

LOG BRIGHTNESS

o ^ ! Ts i"

LOG CUEVE

LLUM- FBOM PKOJ-

LOG BRIGHTNESS

LOG CURVE

I LLUM. FROM PCOJ.

20 to o Xi

FIG. 7. Combination of Figs. 2 and 6 into one dia-
gram, from which can be found illumination from the
projector as a function of log brightness of original
scene.

be applied to the study of transmission from a negative with only
very minor change in the details of the diagram. The film charac-
teristic in section A of Fig. 15 is now that of a negative instead of the
over-all negative and print combination, and the amplifier polarity
of section B has been reversed from (a) to (b) .

The over-all reproduction characteristic C obtained when this is
carried out for a normal negative of gamma 0.7 presents a number of
interesting features. In the first place, the slope of the straight line
portion is almost exactly correct, a somewhat surprising and certainly
very fortunate result. From this one may conclude that negatives

Aug. 1946 TELEVISION REPRODUCTION FROM NEGATIVE FILMS 173

of normal contrast characteristics should be about right for television
transmission.

Second, the contrast decrease in the shadows which was observed
for the positive case is no longer present; shadow detail should there-
fore be improved when the transmitted image is derived from a nega-
tive.

Third, there is a rather more serious loss of highlight contrast (and
therefore detail) than when positives are employed. This highlight
loss appears to be the most serious defect associated with the use of

LOG BE

ILLUMINATION

ICONOSCOPE

FIG. 8. Further combination of individual charac-
teristic curves, representing progress through the
iconoscope.

ordinary negatives. It appears distinctly worthwhile to seek means
of correction, so that a high quality image may be achieved simulta-
neously with the other negative conveniences of processing speed and
simplicity.

By far the simplest solution, from the motion picture viewpoint, is
to have the television engineers design an amplifier of special charac-
teristics, the insertion of which during the transmission of negatives
will eliminate the distortion in question. The television engineers
assure us that this can be done, on paper at least, but there appears
to be some sentiment to the effect that they already have enough
trouble without inventing special amplifiers to please the motion pic-
ture engineer.

174

E. MESCHTER

Vol 47, No. 2

Another "paper" solution, based solely on the graphical procedure
under discussion, is shown in Fig. 16. If a negative slightly softer
than normal is used, and if the projector light is adjusted to use only
the high illumination end of the iconoscope characteristic, and if the
amplifier gain and bias are adjusted to still fill the kinescope input
scale with the resulting signal, then the over-all reproduction curve
turns out almost exactly perfect. However, increased noise from the
increased gain is only one of the drawbacks which make this solution
a rather impractical one.

LOG

O -IP -20 -3O

ICONOSCOPE

VOLTS OUTPU1

Z /"AMPLIFIER"

FIG. 9. The amplifier characteristic has been added to
the film-projector-iconoscope combination.

An important element of the system which can be varied at pleas-
ure within certain limits is that of the film characteristic curve. It
is possible to discover the shape of the film characteristic required for
accurate reproduction by inverting the original graphical procedure,
inserting the desired straight line in the over-all reproduction quad-
rant and treating the film as the unknown. The mechanical pro-
cedure for accomplishing this, following from point to point through
the various elements of the system, is exactly the same as before.

The result, the negative film characteristic required for accurate ob-
jective reproduction, is shown in Fig. 17. At low brightness levels
the straight-line characteristic is retained, but higher contrast is in-
troduced in the highlights to compensate for the loss experienced with

Aug. 1946 TELEVISION REPRODUCTION FROM NEGATIVE FILMS 175

film exhibiting the standard straight-line curve. Formulation of a
negative stock with such a rising shoulder is entirely possible and has
much to recommend it as the best solution to the problem of obtain-
ing improved television images from negatives.

Experimental Verification of Theoretical Predictions. — The con-
clusions concerning the nature of the reproduction characteristics to
be expected from the transmission of positive and negative trans-
parencies have been checked experimentally through the kind co-
operation of F. J. Bingley, Chief Television Engineer, Philco Radio

.FILM

KINESCOPE

O VOLTS

ICON-

AMPLIFIER

FIG. 10. Addition of the kinescope characteristic
permits determination of image brightness as a func-
tion of log brightness of original scene.

and Television Corporation. It should be pointed out that these
tests are purely of an exploratory nature, having been carried out on
the monitor tubes of one station, for a limited number of scenes and
for only one group of control settings. However, the general method
is a dynamic one and is directly applicable to more thorough studies.
The "original scenes" were high-quality transparencies; relative
brightnesses of various picture elements were easily determined by
simple density measurements. These transparencies were placed on
an illuminator and photographed on du Pont Superior 2 in a standard
35-mm cine camera. The negatives received a standard type of de-
velopment and normal prints were made on du Pont Fine Grain Pos-
itive type 225. A long-range transparent gray scale composed of

176

E. MESCHTER

Vol 47, No. 2

LOG BR. A

AMP.

FIG. 11. Introduction of a second logarithmic
curve puts scene brightness and image brightness on
similar logarithmic scales.

FIG. 12. Log scene brightness and log image
brightness are now on the two axes of one quadrant
for convenience in determining the over-all reproduc-
tion characteristic of the entire system.

Aug. 1946 TELEVISION REPRODUCTION FROM NEGATIVE FILMS 177

simple blocks was also photographed on the same illuminator and
prints prepared. The use of such blocks insured areas in the cine
frame of a size sufficient to permit convenient density measurement.
Two rolls were then prepared for television scanning, one negative
and one positive. Each scene was about 100 ft in length, which gave
ample time to adjust the electrical controls to obtain the best
possible quality on that particular subject, and each was followed by
50 ft of gray scale.

SCENE

FIG. 13. The geometrical method of determining
a point on the over-all reproduction characteristic is
indicated: following the arrows through the char-
acteristics of the successive stages of the system lo-
cates N corresponding to the starting point M.

Previously calibrated illuminated wedges were placed on either
side of the monitor tube and a 5 X 7 camera was focused to include
both the monitor and these standard illuminated gray scales.

The test procedure was then to run the cine material, adjusting
controls to obtain the most pleasing quality on each scene. At the
end of the scene the controls were left fixed when the cine gray scale
appeared, and the gray scale image together with the standard scales
beside the kinescope were photographed on a single piece of du Pont
X.F. Pan. A record was made in this manner for each scene of each
roll, comprising both positive and negative inputs to the system
High agitation development of the 5X7 films permitted determina-
tion of absolute brightness levels of the kinescope face by simple den-

178

E. MESCHTER

Vol 47, No. 2

/EEL. LOG BRIGHTNESS OF ORIGINAL SCENE

-6 -8 1-0 t-Z ••«*

FIG. 14. Graphically predicted over-all reproduc-
tion curve of a television system employing positive
film, compared with ideal reproduction.

FIG. 15. Graphical arrangement for the prediction
of over-all reproduction characteristic when a negative
film is used. The film curve at A and the amplifier
polarity at B have been changed, yielding the new
reproduction curve at C.

Aug. 1946 TELEVISION REPRODUCTION FROM NEGATIVE FILMS 179

sity comparison with the image of the standard wedge. These data,
taken in combination with the relative brightnesses of the original
scene, give an indication of the over-all reproduction characteristics
of the photographic-electro-optical system. The general method is
of special interest since it is a true dynamic test, carried out under
actual operating conditions. It does not involve the insertion of
glass slides, auxiliary projectors or alternate light sources.

The results are shown in the graph of Fig. 18, in which logarithms
of kinescope brightnesses have been plotted against the logarithms of

SCENE

FIG. 16. A theoretically possible method of obtain-
ing straight-line reproduction via a negative. Film
gamma is low, only the high illumination portion of
the iconoscope characteristic is used and amplifier gain
has been increased.

the relative brightnesses in the original scene. Reproduction by way
of cine positive is rather better than expected, as shown by the good
straight-line -characteristic. The decrease in contrast for the high-
lights when standard negative is used is very evident, confirming the
predicted performance. Visual estimates of picture quality agreed
with these calculated results; images from the negative film were
superior in shadow detail but rather poorer in the highlights. Qual-
itatively, this failing is of a nature to be remedied by a rising shoulder
negative as predicted in the earlier section of this discussion.

It should be re-emphasized that this represents a limited series of
tests, and that such elements as the exact placing of these curves can
be affected by projector light brightness, negative exposure, kine-

180

E. MESCHTER

Vol 47, No. 2

scope beam current and many other factors. However, it appears
that the curve shapes may be regarded as truly representative of the
general performance to be expected.

Practical picture tests of experimental negative materials embody-
ing the rising shoulder characteristic have given very satisfactory re-
sults in a number of locations, indicating that the basic reasoning and
first series of tests were sound. A complete quantitative evaluation
based on the procedure described above has not been possible up to
the present time; the recent reallocation of frequencies has made

SPECIAL

NEGATIVE FILM

CHARACTERISTIC

LOG SCENE BRIGHTNESS (REL.)

•9

FIG. 17. Special negative film characteristic re-
quired for straight-line reproduction if gain and pro-
jector output are to remain normal.

such tests not merely difficult to arrange, but actually impossible in
many cases. However, it is hoped that the return of stations to the
air will soon allow the performance of this experimental stock to be
studied quantitatively under a variety of conditions.

Conclusion. — Theoretical considerations based on the published
average characteristics of the elements of the television system and
actual dynamic experimental tests indicate that superior results
may be expected from the use of a rising shoulder negative as a
source of images for television broadcast. Fortunately manufactur-
ing methods of achieving such a film characteristic are known; the
good quality images obtainable in this was to provide additional incen-

Aug. 1946 TELEVISION REPRODUCTION FROM NEGATIVE FILMS 181

tive for the general study of the use of negatives for television trans-
mission. These improved results can be obtained without throwing
any added burden of equipment modification on the television engi-

LOG BRIGHTNESS OF ORIGINAL SCENE

FIG. 18. Experimentally determined over-all re-
production characteristics of a television system using
normal prints and normal negatives, compared with
ideal straight-line reproduction curve.

neer. It seems likely that such negatives, offering high-quality images,
will find a definite place in the television world, since they also offer
the advantages of speed and simplicity of processing.

CURRENT LITERATURE OF INTEREST TO THE MOTION PICTURE

ENGINEER

The editors present for convenient reference a list of articles dealing with subjects
cognate to motion picture engineering published in a number of selected journals.
Photostatic or microfilm copies of articles in magazines that are available may be
obtained from The Library of Congress, Washington, D. C., or from the New York
Public Library, New York, N. Y., at prevailing rates.

American Cinematographer

27, 5 (May 1946)

Psychology and the Screen (p. 160)
World-Wide Celebration Planned on 20th Anniversary of

Sound Films (p. 162)
Survey of Current Processes of Color Kinematography in

England (p. 164)
. Soviet's War Documentary (p. 170)

27, 6 (June 1946)

Specialized Photography Applied to Engineering in the
Armed Forces (p. 195)

Miniature Camera Models (p. 202)

Russia Grabs German AGFA Plant, Process, Equipment
(p. 206)

British Kinematograph Society, Journal

9, 1 (Jan.-Mar., 1946)
Presidential Address (p. 2)
Technicians' Ideas for Improving Equipment:
Cameras and Studio Equipment (p. 5)
News Reel Equipment (p. 7)
Sound Recording (p. 8)

Processing Equipment and Procedure (p. 9)
Projection and the Projection Room (p. 10)
Sub-Standard Equipment and Processes (p. 11)
Television in the Kinema (p. 13)
Newsreels in War-Time:

The North African Campaign (p. 17)
War Filming .in the Far East (p. 19)
The Liberation of Europe (p. 21)
The Norwegian Campaign (p. 26)
Make-Up of Newsreels (p. 27)
182

H. A. LlGHTMAN

J. H. COOTE
A. K ALTS AT Y

P. M. THOMAS AND

C. H. COLES
I. BROWNING

A. G. D. WEST

C. VlNTEN

D. FORRESTER

E. WILLIAMS AND
N. DAINES

E. THORNE
R. PULMAN
G. H. SEWELL
A. G. D. WEST

T. ASHWOOD
A. TOZER

S. BONNETT

L. MURRAY
J. C. STAGG

CURRENT LITERATURE

183

Electronics

19, 2 (Feb. 1946)
Design of Compact Two-Horn Loudspeaker (p. 156) P. W. KLIPSCH

International Projectionist

21, 5 (May 1946)

Victor Animatophone 16-Mm Projector (p. 7) L. CHADBOURNE

Elements of Projection Optics, Pt. II (p. 10) A. MONTANI

21,6 (June 1946)

Switzer Electronic Arc Control (p. 5) G. W. SWITZER

Illusion of Depth in Motion Pictures (p. 8) H. T. SOUTHER

Television in the Movie Theatre? (p. 14) L. B. ISAAC

Basic Radio and Television Course, Pt. 24 — Receiving

Systems (p. 18) M. BERINSKY

The Photographic Journal

86A, (Apr. 1946)
Technical Progress in Kinematography (p. 96) R. H. CRICKS

60th SEMIANNUAL CONVENTION

HOLLYWOOD-ROOSEVELT HOTEL
Hollywood, California

OCTOBER 21-25, 1946

Officers in Charge

D. E. HYNDMAN President

HERBERT GRIFFIN Past-President

L. L. RYDER Executive Vice-President

M. R. BOYER Financial Vice-P resident

J. A. MAURER Engineering Vice-President

A. C. DOWNES Editorial Vice-President

W. C. KUNZMANN Convention Vice-President

C. R. KEITH Secretary

E. I. SPONABLE Treasurer

General Office, New York

BOYCE NEMEC Engineering Secretary

HARRY SMITH, JR Executive Secretary

Directory of Committee Chairmen

Pacific Coast Section and Local Ar-
rangements H. W. MOYSE, Chairman

Papers Committee C. R. DAILY, Chairman

BARTON KREUZER, Vice-
Chairman

Publicity Committee HAROLD DESFOR, Chair-
man

Registration and Information W. C. KUNZMANN, Chair-
man, assisted by C. W.
HANDLEY

Luncheon and Dinner-Dance Commit-
tee L. L. RYDER, Chairman

Hotel and Transportation Committee S. P. SOLOW, Chairman

184

SMPE CONVENTION 185

Membership and Subscription Commit-
tee H. W. REMERSCHEID, Chairman

Ladies Reception Committee Hostess MRS. H. W. MOYSE

Projection Program — 35-mm W. V. WOLFE, Chairman .assisted

by Members Los Angeles
Locals 150 and 165
16-mm H. W. REMERSCHEID

HOTEL RESERVATIONS AND RATES

The Hollywood-Roosevelt Hotel, Hollywood, Calif., will be the Convention
Headquarters, and the hotel management extends the following per diem room
rates, European plan, to SMPE members and guests:

Room with bath, one person $4.40-5.50

Room with bath, two persons, double bed $5.50-6.60

Room with bath, two persons, twin beds $6.60-7.70

HOUSING COMMITTEE

An acute housing condition exists in Hollywood and it is expected that most
of the available reservations at the Hollywood-Roosevelt Hotel will have been
taken by the time this issue of the JOURNAL reaches the membership. In order
to be of assistance to members desiring room accommodations, the Pacific Coast
Section has set up a Housing Committee under the Chairmanship of Past-Presi-
dent Herbert Griffin.

The Housing Committee expects to mail a return post card to all members out-
side of the Hollywood area on which the member may state whether he desires
room accommodations and for what length of time. The returned cards will be
checked against available reservations and an effort will be made to place Eastern
and Midwestern members who plan to attend the Convention. However, the
demand is very apt to exceed the supply and reservations will be made on the
basis of "first come, first served." It will be of assistance to all concerned to have
the cards returned as quickly as possible.

RAIL, PULLMAN, AND AIR ACCOMMODATIONS

REGISTRATION

The Convention Registration Headquarters will be located in Room 201 on the
mezzanine floor of the hotel, where Luncheon and Dinner-Dance tickets can be
procured prior to the scheduled dates of these functions. Members and

186 SMPE CONVENTION Vol 47, No. 2

guests are expected to register. The fee is used to help defray Convention

expenses.

BUSINESS AND TECHNICAL SESSIONS

Day sessions will be held in the hotel, and evening sessions at locations away
from the hotel, which will be listed in the preliminary, and final printed Conven-
tion programs.

Authors who are planning to present papers at the 60th Semiannual Con-
vention should mail the title of their paper to the West or East Coast Chair-
man of the Papers Committee, or to the Society's New York Office, as soon as
possible. As a prerequisite to inclusion on the program, authors' abstracts must
be received by the Papers Committee by Sept. 1. Complete manuscripts must
be submitted by Oct. 1, 1946. Only through your cooperation can a preliminary
program be drafted early enough for publication in the industry trade papers
and mailing to the membership at least a month prior to the Convention.

GET-TOGETHER LUNCHEON AND DINNER-DANCE

The Society will again hold its regular pre-war social functions and accordingly
a Get-Together Luncheon is scheduled in the California Room of the hotel on
Monday, October 21, at 12:30 P.M. The luncheon program will be announced
later. Members in Hollywood and vicinity will be solicited by a letter from S. P.
Solow, Secretary of the Pacific Coast Section, to send remittances to him for the
Convention registration fee and luncheon tickets. Ladies are welcome to attend
the luncheon.

The 60th Semiannual Dinner- Dance will be held in the California Room of the
hotel on Wednesday evening, October 23, at 8:30 P.M. Dancing and entertain-
ment. (Dress optional.) A social hour for holders of Dinner- Dance tickets will
precede the Dinner-Dance between 7: 15 P.M. and 8: 15 P.M. in the Hotel Terrace

Room (Refreshments).

LADIES' PROGRAM

A reception parlor for the ladies' daily get-together and open house with Mrs.
H. W. Moyse as hostess will be announced on the hotel bulletin board and in the
final printed program.

Ladies are welcome to attend technical sessions of interest, also the Luncheon
on October 21, and the Dinner-Dance on October 23. The Convention badge and
identification card will be available to the ladies by applying at Registration
Headquarters.

The ladies' entertainment program will be announced later.

MOTION PICTURES AND RECREATION

The Convention recreational program will be announced later when arrange-
ments have been completed by the local committee. Identification cards issued
only to registered members and guests will be honored at the following deluxe
motion picture theaters on Hollywood Boulevard :

Egyptian Theatre

Grauman's Chinese Theatre

Hollywood Pantages Theatre

Hollywood Paramount Theatre

Warner's Hollywood Theatre

Aug. 1946

SMPE CONVENTION

187

Monday, October 21, 1946

Open Morning.
10:00 a.m. Room 201, Hotel Mezzanine Floor: Registration. Advance sale of

Luncheon and Dinner-Dance tickets.
12: 30 p.m. California Room: SMPE Get-Together Luncheon.

Program announced in later bulletins.
2: 00 p.m. Aviation Room, Hotel Mezzanine Floor: Opening business and

Technical Session.
8: 00 p.m. Evening Session: Location to be announced later.

Tuesday, October 22, 1946

Open Morning.
10:00 a.m. Room 201, Hotel Mezzanine Floor: Registration. Advance sale of

Dinner-Dance tickets.

2: 00 p.m. California Room: Afternoon Session.
8:00 p.m. Evening Session: Location to be announced later.

Wednesday, October 23, 1946

9: 30 a.m. Room 201, Hotel Mezzanine Floor: Registration. Advance sale of

Dinner-Dance tickets.
10:00 a.m. California Room: Morning Session.

Open Afternoon.
7:15 p.m. Hotel Terrace Room: A social hour for holders of Dinner-Dance

tickets preceding the Dinner-Dance (Refreshments) .

8: 30 p.m. California Room: 60th Semiannual Convention Dinner-Dance.
Dancing and entertainment. Program will be announced
later.

Thursday, October 24, 1946

Open Morning.

1:00 p.m. Room 201, Hotel Mezzanine Floor: Registration.
2: 00 p.m. California Room: Afternoon Session.
8: 00 p.m. Evening Session. Location to be announced later.

Friday, October 25, 1946

2: 00 p.m.

Open Morning.

California Room:

Afternoon Session.

8:00 p.m. Evening Session. Adjournment of the 60th Semiannual Conven-
tion. Location to be announced later.

Note: All sessions during the 5-day Convention will open with an interesting
motion picture short.

188 SOCIETY ANNOUNCEMENTS

Important

Because of the existing food problem, your Luncheon and Dinner-Dance
Committee must know in advance the number of persons attending these func-
tions in order to provide adequate accommodations.

Your cooperation in this regard is earnestly solicited. Luncheon and Dinner-
Dance tickets can be procured from W. C. Kunzmann, Convention Vice- President,
during the week of October 13 at the Hollywood-Roosevelt Hotel.

All checks or money orders for Convention registration fee, Luncheon and
Dinner- Dance tickets should be made payable to W. C. Kunzmann, Convention
Vice- President, and not to the Society.

W. C. KUNZMANN
Convention V ice-President

SOCIETY ANNOUNCEMENTS

EMPLOYMENT SERVICE

POSITIONS OPEN

Position available for Optical Designer, capable of handling the calcula-
tion and correction of aberrations in photographic and projection lens
systems. Junior designers or engineers will be considered. Write
fully giving education, experience, and other qualifications to Director
of Personnel, Bell and Howell Company, 7100 McCormick Road, Chi-
cago 45, 111.

Photographer. Large manufacturer with well-organized photographic
department requires young man under 35 for industrial motion picture
and still work. Must be experienced. Excellent opportunity. Replies
held in confidence. Write stating age, education, experience and
salary to The Procter and Gamble Co., Employment Dept., Industrial
Relations Division, Ivory dale 17, Ohio.

SOCIETY ANNOUNCEMENTS 189

POSITIONS WANTED

Projectionist-newsreel editor with 15 years' experience just released
from service. Willing to locate anywhere. Write P. O. Box 152, Hamp-
den Station, Baltimore 11, Maryland.

Cameraman, honorably discharged Army veteran/desires re-enter indus-
trial, educational production with independent producer or studio. Ex-
perienced in 35- and 16-mm color and black-and-white. References
and complete record of experience available. Write, wire or telephone
T. J. Maloney, 406 Oak St., Ishpeming, Mich. Telephone 930.

16-mm Specialist. Honorably discharged veteran with many year's
experience, specializing in 16-mm. Linguist. Available for special
assignments. Write J. P. J. Chapman, ARPS, FRSA, The Huon,
Branksome Hill Road, Bournemouth, England.

Experienced and licensed projectionist and commercial radio technician
desires employment with 16-mm producer as sound recordist. Thor-
oughly familiar with principles and practices of sound-on-film recording.
Write F. E. Sherry, 705l/z West San Antonio St., Victoria, Texas.

We are grieved to announce the death of Leon Gaumont, Honorary
member of the Society, on August 11, 1946, in Paris, France.

SOCIETY of MOTION PICTURE ENGINEERS

WOTtL PENNSYLVANIA • MCW YOftKl, N-Y- » TCL. PCNN. 6 O62O

APPLICATION FOR MEMBERSHIP

(This page should be completely filled out by applicant in conformity with Qualifications and
Requirements given on the opposite page for grade desired. References given should be members
or nonmembers who will supply information on applicant's experience and serve as sponsors.)

Name Age

Address .

City

Employer . .

Occupation.

Grade Desired: Associate Q Active
Education *

Record of Employment* (list companies, years, and positions held).

Other Activities1

REFERENCES**

(Name) (Address) (City)

The undersigned certifies that the statements contained in this application
are correct, and agrees, if elected to membership, that he will be governed by the
Society's Constitution and By-Laws so long as his connection with the Society
continues.

Date 19 (Sgd)

* If necessary, use additional sheet to give complete record.

** References should be members of Society. If not, supply two letters of reference from individuals
acquainted with applicant's work.

JOURNAL OF THE SOCIETY OF
MOTION PICTURE ENGINEERS

Vol 47 SEPTEMBER 1946 No. 3

CONTENTS

PAGE
Synchronization Technique W. A. POZNER 191

The Past and Future Activities of the Society of
Motion Picture Engineers

D. E. HYNDMAN AND J. A. MAURER 212

Modernization Desires of a Major Studio

L. L. RYDER 225

Dubbing and Post-Synchronization Studios

W. A. MUELLER 230

The Relation of Television to Motion Pictures

A. B. Du MONT 238

Nonintermittent Motion Picture Projector with Vari-
able Magnification F. G. BACK 248

A Film-Splicing and Repair Machine

A. WALLINGSFORD 254

American Standards on Motion Pictures 258

60th Semiannual Convention 265

Society Announcements 268

Indexes to the semiannual volumes of the JOURNAL are published in the June and December
issues. The contents are also indexed in the Industrial Arts Index available in public libraries.

JOURNAL

OF THE

SOCItTY of MOTION PICTURE ENGINEERS

MOTCL PENNSYLVANIA • NCW VOP.K I, N-V- • TCI. PSNN. 6 O62O

HARRY SMITH, JR., EDITOR

Board of Editors
ARTHUR C. DOWNES, Chairman

JOHN I. CRABTREE ALFRED N. GOLDSMITH EDWARD W. KELLOGG

CLYDE R. KEITH ALAN M. GUNDELFINGER CHARLES W. HANDLEY

ARTHUR C. HARDY

Officers of the Society
^President: DONALD E. HYNDMAN,

342 Madison Ave.t New York 17.
*Past-President: HERBERT GRIFFIN,

133 E. Santa Anita Ave., Burbank, Calif.
^Executive Vice-President: LOREN L. RYDER,

5451 Marathon St., Hollywood 38.
** Engineering Vice-President: JOHN A. MAURER,

37-01 31st St., Long Island City 1, N. Y.
*Editorial Vice -President: ARTHUR C. DOWNES,

Box 6087, Cleveland 1, Ohio.
** Financial Vice-President: M. R. BOYER,

350 Fifth Ave., New York 1.
* Convention Vice-President: WILLIAM C. KUNZMANN,

Box 6087, Cleveland 1 , Ohio.
^Secretary: CLYDE R. KEITH,

233 Broadway, New York 7.
*Treasurer: EARL I. SPONABLE,

460 West 54th St., New York 19.

Governors

*fFRANK E. CAHILL, JR., 321 West 44th St., New York 18.
**FRANK E. CARLSON, Nela Park, Cleveland 12, Ohio.
**ALAN W. COOK, Binghamton, N. Y.

*JOHN I. CRABTREE, Kodak Park, Rochester 4, N. Y.

*CHARLES R. DAILY, 5451 Marathon St., Hollywood 38.
**JOHN G. FRAYNE, 6601 Romaine St., Hollywood 38.
**PAUL J. LARSEN, 1401 Sheridan St., Washington 11, D. C;
** WESLEY C. MILLER,. Culver City, Calif.

*PETER MOLE, 941 N. Sycamore Ave., Hollywood.
*tHoLLis W. MOYSE, 6656 Santa Monica Blvd., Hollywood.

*WILLIAM A. MUELLER, 4000 W. Olive Ave., Burbank, Calif.
*°A. SHAPIRO, 2835 N. Western Ave., Chicago 18, 111.

*REEVE O. STROCK, 111 Eighth Ave., New York 11.

*Term expires December 31, 1946. tChairman, Atlantic Coast Section.
**Term expires December 31, 1947. tChairman, Pacific Coast Section.
*° Chairman, Midwest Section.

Publication Office, 20th & Northampton Sts., Easton, Pa.

General and Editorial Office, Hotel Pennsylvania, New York 1, N. Y.

Entered as second-class matter January 15, 1930, at the Post Office at Easton.

Pa., under the Act of March 3, 1879.

JOURNAL OF THE SOCIETY OF
MOTION PICTURE ENGINEERS

Vol 47 SEPTEMBER 1946 No. 3

SYNCHRONIZATION TECHNIQUE*
W. A. POZNER**

Summary. — This paper consists of an introduction describing the history of
dubbing technique, a section describing the importance of sound perspective, and a
detailed description of the dubbing method used by M-G-M International Films
Corporation.

M-G-M International Films Corporation greatly appreciates the
opportunity created by the Society of Motion Picture Engineers to
present this paper on the dubbing technique in general and, specifi-
cally, the technique employed in its studios. It must be made clear
at the very beginning that there is no such thing as isolated dubbing
technique. There is, however, a well-known motion picture tech-
nique, within which there exists a process that some call "dubbing"
and others "synchronization." When the well-established and well-
known technical methods of the motion picture industry are properly
applied for the purpose of synchronizing or dubbing, a process is
created which may be called dubbing technique. I would like to
outline briefly how, in the last fifteen years, motion picture technique
has progressed to a point permitting us to substitute voices, to trans-
pose stories from one language to another, creating the illusion of
reality, and thus enabling us to entertain greater and greater au-
diences.

If we bear in mind that one of the basic goals of the motion pic-
ture industry is to make the screen look alive in the eyes of the audi-
ence, we will more easily follow the problems that the dubbing proc-
ess had to overcome.

History. — With the advent of the "talkies," the motion picture

* Presented Feb. 13, 1946, at a meeting of the Atlantic Coast Section of
the Society in New York.

"* M-G-M International Films Corporation, New York.

191

192 W. A. POZNER Vol 47, No. 3

industry had to face several problems, one of the most important be-
ing the problem of the language. The technique of producing mo-
tion pictures, the acting, everything had to be changed. There was
the necessity for a different tempo. The subtitles and cut-in titles
of the silent movies were replaced by speech. That speech could not
be under stoojl by audiences speaking a different language. The only
means of overcoming that difficulty was to use the superimposed
title method, which still had the defect of not being a means of con-
veying all of the dialogue nor all of the story, and distracted the public
from watching the action. It was only a partial answer to the prob-
lem.

In 1931 some major American companies decided to experiment
with another method. Similar experiments, by the way, were
carried out simultaneously by German firms, and consisted of try-
ing to substitute the original language of the picture with a foreign
language, synchronizing the foreign language with lip movements of
the actors on the screen. It is interesting to note that the approach
to the problem in the USA was entirely different. from that of the
Germans. The Germans decided that this was a purely technical
problem, which, therefore, had to be solved by highly technical
means. In this country it was just a problem that was to be solved
by whatever means proved to be the best.

Mechanical Guide Method. — In trying to transpose a motion
picture by only technical means, the German method fell into ex-
aggerated mechanical details. No picture can be technically per-
fect if it is not the result of collective work of a group of people with
creative minds. The same holds true for any and all motion pic-
ture processes and, therefore, for the process of dubbing.

The German process consisted basically of:

(a) A method which permitted detection of the speech components of the
original version by electromechanical means which gave a design similar to a
cardiogram.

(6) Transcription of this speech and its graphic representation on a pape*r
or film in a manner similar to the one used for transcribing lyrics and music on
a music sheet, the only difference between such transcripts and music sheets
being that instead of having the notes, there was a graphic representation of the
syllable, and — underneath — instead of having the lyrics there were the syllables
themselves.

After establishing this type of strip or guide, the text was trans-
lated. Great care was taken that each syllable, word, and sentence

Sept. 1946 SYNCHRONIZATION TECHNIQUE 193

of the new language coincided exactly, syllable for syllable, with the
original text. The new language was then transcribed on clear,
transparent film or on a large disk, and was spaced exactly in the
same manner as was the original dialogue on the "guide." The
guide bearing the dialogue in the new language was then projected
on a screen simultaneously and in synchronism with the picture print
from which its "cardiogram" was detected.

The text appeared from right to left (horizontal projection) or
counterclockwise (on a rotating disk) and the actors were to read
their lines as soon as a syllable hit a determined point on the screen.
The result was a more or less perfect synchronization of the new
dialogue with the lip movements of the original version, but that
synchronization was not a true one. The length was there, but the
phonetic emphasis and the emphasis of thought and of interpretation
were totally absent.

It could not be otherwise, since no actor can interpret his part if
he reads it at a speed dictated to him by something like an electric
signboard. There is a doubt in our mind that anybody could inter-
pret aloud and with dramatic effect the news bulletins that appear on
the Times Building, if he had to read each syllable exactly at the time
that it reaches the 42d Street corner.

Of course, this method presented the advantage of giving a syn-
chronized dialogue at a very low cost. The preparation required the
work of only a few people — no studio time was involved; therefore,
there is no studio overhead — all of which accounts for a very low
cost. But it also had a big disadvantage, since there are very few
actors who can read their lines from a mechanical guide with any
real effectiveness. Those few who are able to do so, became, as some
people say, "expert dubbers."

As an example, one actor alone has dubbed the voices of George
Raft, Edward Robinson, Gary Grant, Dick Powell, and Paul Muni.
We all have admiration for these stars and we know that each one of
them has a personal approach to the part he plays. No one would
expect William Powell to play Zola, or Paul Muni to play Nick
Charles, since the temperament and interpretation of every actor is
different and personal. It is, therefore, inconceivable that an actor,
good as he might be, could impersonate several outstanding perform-
ers, although the excuse is often heard that "he does not fit the
part, but he is a good synchronizer," which means that he can read
in synchronism.

194 W. A. POZNER Vol 47, No. 3

Of course, these are only the basic elements of the mechanical
guide method born in Germany and later developed in France, Spain,
and the USA. This method has improved since, mainly because the
theater-going public was dissatisfied and demanded a better product.

On the other hand, the method of detecting the original dialogue
became more perfect and gave not only the graphic representation of
syllables, but also the phonetic emphasis of the word within a sen-
tence. The technicians learned how to differentiate a labial sound
from an explosive; the difference between the vowels and the con-
sonants, and the equivalent sounds such as d and /, a and i, s and /,
giving the writers a somewhat greater flexibility for translation.
But the basic mistake of that mechanical guide method still re-
mains; i.e., the actor is still merely reading his lines, and not often
giving them much dramatic effectiveness.

Visual Synchronization Method.— ^The other rnethod we should
like to call the visual synchronization method.

The problem of substituting the original dialogue by a new one in a
different language was. approached from the viewpoint that such
substitution should be done on the basis of emotional equivalence.
The new dialogue, of course, should be in synchronism with the lip
movements of the actor appearing on the screen, but, what is more
important, the emphasis of the new dialogue and its meaning should
coincide with the emphasis of the facial expressions and gestures that
the audience follows.

Since there were no mechanical reference points to guide the work
of people who tried to make such language conversions, most of this
work was done in the experimental, empiric way.

The new dialogue had to be created simply while projecting the
picture on a screen, on a cutting moviola, or on a projection
moviola, and trying to read the lines of the new dialogue simultane-
ously with the dialogue of the original version. Soon enough it was
found out that it was impossible to write a dialogue in such a fashion,
since the speed at which words are delivered on the screen is much too
great, and the reels in their regular form are too long. The standard
projection machine and the cutting moviola were discarded; the
former because of the time loss involved in rethreading it, and the
latter because of its small picture size. The so-called projection
moviola with reverse drive was used and reels were run back and
forth, projected on the screen, while the writer tried to adapt the-
new dialogue line by line.

Sept. 1946 SYNCHRONIZATION TECHNIQUE 195

Such a process involved a tremendous amount of time loss and
could not give the necessary assurance for synchronization. It also
became apparent that no actor could memorize his part for scenes
that lasted several minutes. It was at this point that the idea of an
endless loop was born. This meant that every picture to be syn-
chronized had to be cut into short scenes, leader had to be added at
the beginning and end of scenes, forming an endless band of film,
and such film band projected for as many times as it was necessary
for the writer to write the dialogue and later for the actor to enact the
scene.

Special equipment that could safely allow a projection of such type
of loops had to be devised. A technique of breaking a complete
motion picture into scenes had to be developed.

It might seem that such breakdown can be done arbitrarily, and,
let us say, all scenes can be 50 ft long, 100 ft long, or X ft long. It is
not really as simple as that. It must be remembered that an attempt
is being made to re-create an illusion and therefore the emotional as-
pect of each scene composing the picture must be taken into considera-
tion. The work of the writer, the actor, and the sound technician
must be facilitated to the greatest extent to enable them to achieve
their goal. Therefore, when in practice a motion picture is broken
down into loops, their lengths vary from as little as 20 ft to as much
as 150 ft.

There is no average length. It is dictated by the amount of dia-
logue that the writer has to adapt and the actor to memorize, as well
as by the dramatic emphasis of the scene. It is impossible to break
up a love scene in the middle, or a hysterical scene at an arbitrary
spot, just for the sake of the loop's length.

It would also involve needless difficulties to combine in the same
scene several camera angles, since each camera angle may or does call
for a different acoustical interpretation.

Another very important factor which must not be overlooked in
comparing the two methods is one of talent. Since the emphasis of
the visual synchronization method, from the start, was on the dra-
matic value of the new version, it was extremely important to find
actors who could really "get under the skin" of their counterparts
of the original version. The voices in the new language version had
to match the voices of the actors of the original one, but even more,
the manner of speech, the manner of articulation, to a certain extent
the mouth formation, had to be very similar.

196 W. A. POZNER Vol 47, No. 3

If an identical voice could not be found, it was necessary to find a
voice that would have the same general characteristics — perhaps in
slightly higher or lower register, but still being dry if the voice of the
original actor was, or warm in the event that the original actor's voice
was such.

To summarize in brief: the visual synchronization method first
requires a lot of imagination on the part of the writer, then makes
it imperative for the actor to know his lines by heart and to be able
to interpret them at a tempo which is dictated to him by the action
that he is observing on the screen.

To this description of the two processes, I should like to offer you
my conclusion. The mechanical guide method was far superior to
the visual synchronization at the early stage of the game, but when
this last process started adopting the generally known motion pic-
ture technique, it outdistanced its competitor by a very great mar-
gin.

Sound Recording. — Before going into a detailed description of the
visual synchronization method as employed in the studios of M-G-M
International, it is necessary to point out the extreme importance of
the sound recording technique.

At this time I would like to quote a paragraph by Wesley Miller
from "Motion Picture Sound Engineering," published by the Re-
search Council of the Academy of Motion Picture Arts and Sciences :

"In the natural world, certain combinations of objective elements
of sight and sound are familiar to us. It is only when our expecta-
tions are disturbed that we commence to wonder and to investigate.
The character of a sound informs us of its proper source and time of
origin. We know what to expect when we start out to identify
that source and we have but to trace it back to find it.

"The recording medium introduces a new element — time. The
reproduced sound may no longer be traced directly to its source in
point of time. Any period may elapse between the original incep-
tion and the final reproduction. The identification of source be-
comes a voluntary effort, and a multitude of questions arise to per-
plex us in the technique of the reproduction system. The motion
picture craftsman desires to create in his product an illusion which
plays upon the imagination of his audience to make them forget
these artificial factors. By the many artifices at his command he
may often transport them from their own sphere to the entirely new
surroundings which he provides for them at the screen."

Sept. 1946 SYNCHRONIZATION TECHNIQUE 197

This quotation clearly indicates how much crc-ativi ability is
generally expected from the sound engineer and this is even more
true in the case of a sound engineer working on the production of a
synchronized version.

There is no difference between the recording apparatus employed
in direct production and in synchronization. The mixer panels,
volume indicators, loudspeakers, amplifiers, recording machines are
identical.

However, the problems that the sound engineer must solve are
multiple. In a recording auditorium with a given acoustical char-
acteristic, he has to create an illusion that would match an infinity
of acoustical interpretations. The scene that is recorded takes place
in the hall of a railroad station. The next, in a boudoir. Then come
scenes in an airplane, in the woods, in the mountains, in a business
office. Each and every one of them calls for different acoustics, but
the sound engineer is still in the same auditorium. He cannot move
his recording equipment by means of a magic wand from one sur-
rounding to another. Nevertheless, he must give the movie-going
audience the illusion of doing just that.

Another problem: a big close-up appears on the screen. You see
only the actor's head. You expect to hear his breath; you expect
almost to hear his mouth move. In the next shot you can barely
see people; they are at the end of a long hall, but the auditorium
remains the same size. It is up to the ability of the sound engineer
to create the illusion of space.

Of course, a question may arise in the reader's mind: "Well,
after all, how about the man who gives you a variety of illusions dur-
ing radio broadcasts?" Far be it from me to call his job an easy one,
but how much easier it is compared to the one of the sound engineer
who works on a synchronized version ! Do not forget that he is not
supposed to record sound of the quality that you would hear in any
big hall, but it must be of a quality that in your mind will correspond
to the big hall that has already been photographed, and that you
actually see on the screen.

The distance between you and the actor on the screen must be
faithfully reproduced and the acoustical illusion must be exactly of
the same order as the visual one. The engineer must relate the ac-
tor's voice to what is seen on the screen. If an actor on the screen,
while speaking, has his back toward the public and then turns
around, the quality of the sound must change, to underline that

198 W. A. POZNER Vol 47, No. 3

movement, but it must change exactly at the time that the public
feels such change as a natural necessity. It cannot be too early and
it cannot be too late. That is why the imaginative capacity of the
recording engineer who is entrusted with creating a new sound track
for a synchronized version is of basic importance to the success of
the whole enterprise.

M-G-M International Films Corporation. — Having established
the basic differences and the basic requirements of both synchroniza-
tion methods, I would now like to give you an idea of how this work
is performed in the studios of M-G-M International.

It was found a good policy to organize production units, the num-
ber of such teams being in direct proportion with the program of the
studio. Our experience has been that a high-standard production
requires an average of three months' continuous work from time of
inception until a finished product can be thrown on the screen.
Therefore, one team, allowing for overlaps in production, can com-
plete five to six pictures in a year. As we go along, we shall de-
scribe in detail all of the basic steps and all duties and responsibil-
ities of the members of each production team.

To begin with, such a team is composed of four permanent mem-
bers and one semipermanent one, the permanent members being the
director, the assistant director, the film editor, and the sound engi-
neer. The semipermanent one is the dialogue writer.

Preparation. — The first step is preparation for production. When
a motion picture in its original version reaches the synchronization
studios of M-G-M International, and a production of the new lan-
guage version is decided upon, the motion picture is screened for all
members of various production teams, supervisors and studio man-
ager personnel.

Such screening has two purposes : first, to give the writers and di-
rectors an opportunity to express their choice; second, to give the
technicians of our studios the opportunity to become acquainted
with the type of product coming from Hollywood (or from abroad)
and to familiarize themselves with the pictures they will have to work
upon.

A meeting will be held after the screening and the production as-
signed to a team. At this time, cutting, action, and dialogue con-
tinuities have reached our studios.

With a continuity in hand, the director, the assistant director,'
the writer, and the supervisor will screen the feature at least twice

Sept. 1946 SYNCHRONIZATION TECHNIQUE 199

more. They will mark on the continuity the proposed scene break-
down, they will discuss such scene breakdown after each screening
until a decision is reached. In addition to that, the writer will in-
dicate on his original dialogue continuity the phonetic emphasis of
sentences, the basic pauses or peculiarities in pronunciation of each
actor. The film editor will take note of any special information
which might be required in the preparation of inserts, titles, opticals,
etc. When the continuity is broken -down we say the picture has
been "cued" into loops and we number such loops consecutively
from one to whatever the last number might be. Several copies of
cued continuities are prepared, one going to the production super-
visor, the others to all members of production teams.

Following a loop breakdown, the film editor will break down the
film into individual scenes, splicing the beginning and end of each
scene into a loop, which is numbered in agreement with the scene
numbers appearing on the continuity.

The production supervisor will prepare a character breakdown
chart. Such chart indicates in how many scenes and in which ones
each character will appear. A chart of that sort is therefore ex-
tremely useful in preparing a shooting schedule, a daily program for
recording.

But before any actors can be called, a decision must be reached as
to who those actors are going to be, which is the work of the director,
his assistant and the casting department.

Casting. — When synchronizing American pictures into a foreign
language, it has been our aim to find exact voice counterparts for
each important American actor. In order to do so, we have con-
ducted an extensive series of tests. For our Spanish version pro-
gram, tests are being made currently in New York, Hollywood, and
throughout Central and South America.

Such actors' tests are transcribed on phonograph records and for-
warded to our Casting Department. The records are played back and
the voices of the Spanish-speaking actors are typed.

To begin with, the types are very broad, such as young or old,
high-pitched or low-pitched, pleasant or unpleasant. Then each
type is reviewed again. For example, deep, middle-aged, male
voices will have pleasant and unpjeasant voices among them, will
have character and lead types, and eventually a voice is found that
sounds like Edward Arnold's. In order to make sure, we will com-
pare the recording of this voice with a film or disk recording of Ed-

200 W. A. POZNER Vol 47, No. 3

ward Arnold himself. If this comparison is satisfactory, this Spanish
actor will become the permanent counterpart of Edward Arnold in all
our Spanish versions.

It goes without saying that the acting ability of all prospective
Spanish performers is thoroughly examined even prior to a detailed
analysis of their voice quality.

After having applied such a process of talent classification for
about two years, our studios -have today successful counterparts of
most of the M-G-M stars for their Spanish versions. Nevertheless,
the scouting for talent is a continuous task and is being carried out
daily by our casting department. Cross-reference files, index card
files are established. Each Spanish actor who ever played a part in
our studios has such a card. Each identifiable actor who played a
part in the original version of any picture synchronized in our studios
has a card. On the Spanish actor's card the name of the original
actor for whom he substituted in the Spanish version appears, as well
as the title of the picture and the name of the character. On the
original actor's card similar information is recorded, the only dif-
ference being that it will bear the name of the Spanish actor who sub-
stituted his voice for the original one.

Once such a system is established, the casting problem in its major
part becomes a problem of classification. A detailed original cast
for each picture to be synchronized is received by our casting de-
partment. Referring that cast to our cross-index files, we can easily
find the counterparts of Spencer Tracy, Ingrid Bergman, Agnes
Moorhead, Donald Crisp, etc.

Should a new actor appear in a forthcoming production, advance
information will reach us and a thorough search will be conducted
until a proper Spanish voice is found for such an actor.

The Shooting Schedule. — Once the casting is finished, it becomes
necessary to prepare the shooting or recording schedule.

The preparation of such a schedule is the responsibility of the pro-
duction supervisor, in co-operation with the director.

The first step will be to establish the basic sequences that compose
the picture as a whole. Then, to place such sequences in their chrono-
logical order, and finally, to analyze whether each sequence, short
or long, can be recorded in one day's work.

One must be very careful not to overestimate the working capacity
of actors and directors. It has been found that an average of 30
scenes or loops can be satisfactorily recorded during a normal day's

Sept. 1946 SYNCHRONIZATION TECHNIQUE 201

work, provided that no actor has more than 20 scenes in the day.

The director will see to it that the dramatic continuity is not in
terrupted by breaking down one sequence into two or three, and that
the continuity of the development of each character is kept in its
chronological sequence, which is not necessarily the continuity of the
motion picture in its final form. Such nonchronological develop-
ment is found in the flash-back technique as used in such pictures
as The White Cliffs of Dover, Mrs. Parkington, and Waterloo Bridge,
where the chronological development of the characters does not coin-
cide with the continuity as it appears on the screen.

Writing the New Dialogue. — As important as all the other problems
might be, the one of writing dialogue for the dubbed version is really
the most important. We stated previously that our goal was to
create a new language version which will retain all of the character-
istics of the original one, refracted in the specific medium of the new
language with all its peculiarities, traditions, and idiosyncracies.

When two characters appear on the screen, one having a Bronx
accent and the other a Middle Western one, everybody knows where
these people come from and what their background is. They are
definite types. But how do people in Colombia and Peru know what
the Bronx is, and why should they understand that there is a dif-
ference between Bostonian and Texan English? The how and why
are the questions to be answered by the writer who prepares the new
dialogue. The answer cannot be given by the interpretation of the
actors, since not only the accents vary in English, but basically the
structure and the texture of the language are different. The actor
can only handle the language material that is given to him. He can
express it more or less successfully, but if the style of his dialogue does
not correspond to the character, the discrepancy creates an artificial
type.

The writer, therefore, must possess a deep knowledge of the lan-
guage in which he writes. He is not a mere translator. He really
does re-create the types which once appeared in the original version.
But he also has another problem to solve. The dialogue must be
spoken in synchronism with the lip movement of the original char-
acter. The basic thought expressed in the original dialogue must be
retained. It is a very tedious and difficult problem to solve. A
study in phonetics, a deep knowledge of phonetic equivalents must
accompany an idiomatic knowledge of the language.

A mind that can take advantage of any situation provided by the

202 W. A. POZNER Vol 47, No.3

action on the screen, by every off-screen dialogue, by every shadow,
every movement on the screen, is one of the basic requirements for
successful dialogue writing. I should like to give a few examples of
how writers overcome the difficulties of such problems.

The lines below are part of a prayer spoken by a young girl in a
French convent. The writer, though not making a literal translation,
has preserved the spirit of the original, as well as its rhythm. In
both languages the basic phonetic values are the same— meaning,
particularly that in both versions the labials fall in the same places —
and that the new lines are in naturally expressed English. Here are
the examples :

Delivrez-moi/de I'angoisse/d'ou je s/uis plongee.
Deliver me/from the anguish/that is/upon me.
Protegez-nous d/e votre main.
Protect us now and /ever more.

Technically, the process of writing the new dialogue can be boiled
down to the following elements :

As you remember, the writer screens the picture with the original
continuity in hand, on which he makes notes indicating basic pauses,
phonetic emphasis of sentences and the off-screen dialogue, etc.
Usually the writer requires three screenings before he can assimilate
the picture and gather all the pertinent information. Having the
picture clearly in mind, the writer then translates the dialogue, keep-
ing the basic conformation of his text as closely as possible to the
conformation of the original. In other words, the tempo of the dia-
logue, the length of sentences are kept as closely as possible to those
of the original.

The writer definitely does not translate word for word, or sentence
for sentence, since this would produce artificial speech. We all know
that nouns, adjectives, and verbs do not present themselves in the same
sequence in, let us say, a language of the Anglo-Saxon group and in a
language of the Latin origin. Trying to place the words in the same
sequence as they appear in the original language would by necessity
mean artificiality. Of -course, this first draft of the new dialogue is
by no means synchronized with the original. It is merely the basis
which later on will be adjusted and transformed so as to become syn-
chronized with the lip movements on the screen.

Some writers require an additional step in their work. The orig-
inal dialogue is rerecorded from film onto phonograph records of

Sept. 194G SYNCHRONIZATION TECHNIQUE 203

commercial si/.e and speed, and the writer can play those records back
nil a pickup whenever he wants to check the length of a sentence, the
manner and rhythm in which the sentence is pronounced, the em-
phasis, the speed, and the pauses that occur within the sentence.

After having prepared a final draft, the writer will discuss it with
the director and the assistant director. It will be made sure that the
characters have not been modified, and that all thoughts contained
in the original have been clearly transposed into the draft.

It will be necessary at this point to check the synchronism of the
new dialogue with the original action. For that purpose, the writer
and the assistant director will screen the picture scene by scene, each
scene being in loop form, and while the assistant director will enact
every line of dialogue in synchronism with the action appe'aring on
the screen, the writer will eventually rewrite, correct, and check the
synchronization. The director will very often be present at this
time, and all three men will discuss any change that might be neces-
sary for synchronization. It has been our experience, however, that
the synchronism is not so important as the naturalness, the fluidity
of the dialogue.

However, by this we do not mean that any degree of synchronism
is acceptable for a synchronized version. Should we face a situation
where, in order to achieve a perfect synchronization, we would de-
stroy the naturalness of our dialogue, we would discard such a solu-
tion, since the fluidity of the dialogue is a greater psychological fac-
tor in giving perfect illusion than the mechanical synchronization.

Such checking of dialogue necessitates two or three weeks of constant
and tedious work for each feature. Once completed, the dialogue is
discussed again by the members of the production team and the
editorial staff of the script department. Should the subject call
for strictly medical, naval, or military expressions, a technical ex-
pert's help is used. Once all these questions are clarified, the script is
prepared with as many copies as necessary, and distributed to the
cast and all members of the production team.

Recording the New Dialogue. — Once the script is ready, the cast is
set, the recording schedule is established, and the technicians know
the subject they are going to work on almost by heart, we can safely go
into a studio and start recording.

Fig. 1 is a vertical section of a recording studio.

A recording studio should be a rectangular room approximately
40 ft long and 25 ft wide, and 16 to 18 ft high. At one end of such

204

W. A. POZNER

Vol 47, No. 3

an auditorium there is a screen. At the other extreme, in a room ad-
jacent and separated by a large glass window, is located the sound
recording control room. A projection booth, equipped with one or
two standard projection machines with special loop magazines, is
located above or alongside the control room. The control room
should be as large as possible, to accommodate the necessary sound
recording equipment, the sound engineer and one or two assistants.
It is very important that the people inside the control room have an
unobstructed view into the auditorium, and can readily observe the
projection screen.

The auditorium should be acoustically treated for both sound-
proofing, in order to eliminate outside noises, and a minimum amount
of reverberation. In practice, the deader the acoustics of such room,
the better the results. However, it is desirable that the acoustic treat-

FIG. 1. Recording studio.

ment of the room be of the variable type, the walls being treated by
hinged panels, one s?'de of which is sound-absorbent and the other
sound-reflectant. This arrangement permits a variety of acoustical
interpretation simply by changing the position of such panels.

The actors are usually placed facing the screen and about 15 to 20 ft
away from it. Three microphones are generally employed during re-
cording. They are placed at approximately 2, 6, and 12 ft away from
the actors, toward the screen, and represent the average distances at
which close-ups, medium shots, and long shot sound can be re-created.
These distances are by no means a "must" and will vary with each
scene. The microphones will be used simultaneously or alternately,
depending upon the requirements of the scene. The sound engineer
may change from one microphone to another at a split second, since he
has special switch-over keys at his disposal.

The actors and director work in the studio, the assistant director
and the sound engineer in the control room.

Sept. 1940

SYNCHRONIZATION TECHNIQUE

205

We have already indicated that speeial equipment allowing a safe
projection of film loops had to be devised. M-G-M International
Films Corporation studios have developed a special loop magazine
which can he adapted to any standard projection machine. Once

FIG. 2.

Projector with special double film and loop at-
tachment.

equipped in such manner, the standard projection machine assumes a
threefold purpose. It permits the regular projection of movietone
prints and also permits simultaneous projection of two films, picture
and track, since it incorporates the well-known feature of the preview
attachment. In addition, it permits a continuous projection of a loop.

206

W. A. POZNER

Vol 47, No. 3

Fig. 2 is a general view of the special loop and double film magazine
in operating condition. All film is enclosed throughout its entire travel.

Fig. 3 shows the arrangement for projection of separate sound and
picture. The upper left-hand magazine is sound track feed. The

FIG. 3.

Projector threaded for separate sound and
picture.

lower left-hand magazine is sound track take-up. A removable
spindle in the rectangular loop magazine provides picture take-up.

Fig. 4 shows the film path for the projection of separate sound and
picture. Both picture and sound engage together over the projector
hold-back sprocket, whereupon they follow separate paths through
the sound head. The picture is kept from interfering with the track

Sept. 1940

SYNCHRONIZATION TECHNIQUE

207

during its passage through tin- sound head by the addition of four
supplementary rollers. The two films engage again in the sound
head hold-back sprocket, after which |x>int they separate to their
appropriate pickup magazines.

Fig. 5 shows the operating condition for loop projection. The fire
trap seen at the bottom of Fig. 4 is replaced by a shoe which fits over

FIG. 4.

Close-up of sound head threaded for sep-
arate sound and picture.

the sound head hold-back sprocket. This is needed because there is no
tension on the film as it follows into the loop magazine. The spindle,
required in the loop magazine when a reel is used, is removed by Ipos-
ening a wing knot and a liner is inserted which reduces the width of
the magazine to approximately 38 mm. The loop magazine casting
must be considerably deeper than this to accommodate a standard
reel. However, in order to minimize the possibility of film twisting as
it lies in the loop magazine, this excess width must be eliminated and a

208

W. A. POZNER

Vol 47, No. 3

liner box of this sort has been found to be the quickest and easiest way
to do this. The film pulls up from the bottom of the pile as seen on the
right, passes over a series of rollers and enters the extension collar on
top of the projector mechanism through a fire trap, from which point
it follows the normal film path.

FIG. 5. Projector threaded with picture loop.

A loop is projected upon the screen, the actors watch it, and follow-
ing the director's indications and guided by the sound of the voices
coming from the screen, place their lines. When the general outline of
the scene has been achieved, the original sound is cut off and the loop
is run silent. The actors know their cues. They will now rehearse the
scene for dramatic interpretation . Meanwhile, the sound engineer has
placed his microphones in positions that will give him a sound record
which matches exactly the action on the screen. He obtains the effect
of distance at any particular spot of the scene by using different micro-

Sept. 1946 SYNCHRONIZATION TECHNIQUE 209

phones. If the scene starts with a medium shot and then goes over to
a long shot, he will use two microphones ; the first one being, let us
say, 6 ft away from the actors, the second 10. At the point where
the camera angle changes, the sound engineer switches "off" the first
microphone and "on" the second, giving the audience the illusion
of depth.

Each loop is provided with a standard leader. That leader bears a
"bloop" or scratch in the sound track area and a diagonal line in the
picture area that appears from the left top corner down to the right
bottom corner on the screen. The purpose of the bloop is to give a
synchronization reference point. The diagonal line is to warn the
actors that the action will start at the moment the line reaches the
bottom right-hand corner of the screen. The synchronization mark or
bloop is used as follows : once the original sound of the scene has been
cut off on the studio speakers, it is automatically fed to the mixing
table in the sound engineer's control room. By throwing a key, the
recording engineer can record that sound, but instead of recording the
sound of the whole scene, he will only open his key at the time when
the bloop is about to appear on the screen. The click produced by the
bloop will go through the recording system and will register on the
film in the recording machine, giving a reference point for lining up the
new sound track with the original picture.

Obviously, there will be more than one take for each scene — on an
average there are three or four such takes. It is important that each
take number be properly announced and that the bloop be recorded
every time.

The sound track recorded in this manner is then developed and
printed in the laboratory and returned to the film editor, who in the
meantime has received a copy of the director's report with all neces-
sary instructions. Guided by the slate at the beginning of each take,
the editor breaks down the sound track ; he also breaks down the loops
and restores them to their original form of film rolls. He will line up
the synchronization mark located in the track area of the loop leader
with its recorded counterpart appearing on the new sound track. He
then cuts off the leaders an'd assembles the picture and the new sound
track in its original continuity. This work is done the day after the
recording session, so that at the end of the next day the production
team screens the material recorded during the preceding day.

At the screening of such "dailies," the director is able to check
whether the results obtained are satisfactory from the point of view

210 W. A. POZNER Vol 47, No. 3

of action and interpretation. The sound engineer checks the quality
of the sound and the perspective of his recording. The film editor can
judge how much work there will be to synchronize properly the new
dialogue with the picture.

Work of the Film Editor. — As the recording progresses, the film edi-
tor assembles more and more scenes and sound tracks, restoring the
picture to its original reel form. To obtain perfect synchronism, he
runs the picture and track on a moviola, advances or retards a word
or a sentence, lengthening or shortening pauses between them, creat-
ing new ones, and eliminating others.

Contrary to some opinions, this is by no means a mechanical proc-
ess. One must have the feeling of the word. One must have imagina-
tion in order to respace words whenever this is necessary. One must
also be able to judge where the emphasis in certain sentences is sup-
posed to fall, so that the emphasis of the sound coincides with that of
the action. After the completion of the recording, the film editor will
have his first cut ready. The new dialogue will be more or less in syn-
chronism with the picture. Several screenings will take place. The
director and the assistant director will indicate corrections until the
production team is satisfied that the best results are achieved. At that
time a final screening takes place, at which the production supervisor
and two or three other directors, the editor of the script depart-
ment, and eventually the writer will assist. Final corrections will be
indicated, and all scenes that do not satisfy this audience will be re-
taken. Such retakes go through the same series of operations as the
original recording, and once completed, are incorporated in the final
version of the new dialogue track.

Music and Effects Tracks. — It will be necessary now to rerecord the
new dialogue tracks, together with the musical background and sound
effects that the original version calls for. A special music and effects
track is produced. This track will contain all of the musical back-
ground and noises required by the action. The music tracks, of
course, are prepared in the same fashion as they are for the original
version. The sound effects are either reclaimed from those portions of
the original sound track that are not covered by dialogue, or re-cre-
ated. All these elements are rerecorded on the new sound track at
proper levels. They are prepared in reel form to match the picture
reel by reel. The dialogue track of the new version, once edited, will
appear also in the same form.

Rerecording. — The new dialogue, the music, and the effects are then

Sept. 1946 SYNCHRONIZATION TECHNIQUE 211

rerecorded at proper relative levels to create the best possible illusion
of reality. Sometimes it is necessary to have more than one dialogue
track. As a rule, the singing voices are assembled on a different track
to permit a greater flexibility for correction, compensation, and bal-
ance during rerecording. Sound for all special effect scenes will also
be placed on a different track. At times the music or some of the
sound effects are, for practical reasons, prepared in a similar manner.
The total number of tracks per reel will therefore average four and
sometimes as many as six or eight.

The rerecording of a synchronized version calls for exactly the same
operations as those used for rerecording an original film. Such re-
recording technique has been thoroughly discussed in the past and
does not require any additional discussion within the framework of
of our subject. However, it is extremely important to follow the basic
rerecording procedure while working on a synchronized version.

The rerecorded negative is then developed in the usual manner,
lined up with the picture negative, which already has the new lan-
guage main and end titles and inserts, and then printed in the same
manner as any other motion picture film is printed in the laboratory.

The first sample print is then screened for the production depart-
ment, discussed, and if the consensus of opinion is that portions or
sequences of that print are found unsatisfactory, these portions will be
retaken until the final result is unanimously approved. The print is
then turned over to the distributing organization, which may accept,
reject, or request changes. Once the print is accepted, the picture is
ready to go into the theaters and face its final test, the scrutiny of that
very important person, the paying customer.

THE PAST AND FUTURE ACTIVITIES OF THE
SOCIETY OF MOTION PICTURE ENGINEERS *

DONALD E. HYNDMAN** AND JOHN A. MAURERf

The Society of Motion Picture Engineers is an international engi-
neering organization composed of a group of individuals associated
in a general partnership to conduct a business, paying no salaries to
officers or members, but operating on a nonmonetary principle to
recommend engineering procedures; to guide, to some extent, re-
search and development; to encourage improvement; and to lead
standardization within the motion picture industry. It enjoys all the
normal legal privileges of a nonprofit organization.

The Society was organized in 1916 by a group of engineers under
the leadership of C. Francis Jenkins, of Washington, D. C., who be-
came its first president. The founders of the Society had three ob-
jectives in view: (1) the advancement of motion picture engineering
and the allied arts and sciences; (2) the standardization of mecha-
nisms and practices employed in the motion picture industry; and (3)
the dissemination of scientific knowledge through publication.

Since its organization the Society has held semiannual conventions
at which engineering papers were presented and general discussion
invited. These papers and discussions have been published in the
TRANSACTIONS of the Society, which were issued first semiannually
and later quarterly from 1916 through 1929, and in the JOURNAL,
which has been published monthly since the beginning of 1930. No
other source of information about the scientific and technical side of
the motion picture industry is comparable in scope, in completeness,
or in continuity to the accumulated TRANSACTIONS and JOURNAL of
the Society of Motion Picture Engineers. The knowledge contained
in these publications and in the separate reprints and reports issued

* Presented before a joint meeting of The Royal Photographic Society of
Great Britain and the British Kinematograph Society, London, Apr. 10, 1946.

** President, f Engineering Vice-President, Society of Motion Picture Engi-
neers.
212

PAST AND FUTURE SOCIETY ACTIVITIES 213

by .the Society is of incalculable value to the industry, and represents
an actual cost, for research work, of many millions of dollars.

The present membership of the Society comprises approximately
2300 engineers and technicians who are employed either directly or
indirectly within the international motion picture industry and allied
industries.

Because of the mutual understanding and close co-operation of
these men, who know the problems in the related fields of production,
distribution, and exhibition, it has been possible for the Society to
bring about engineering advances that might otherwise have re-
mained dormant for many years. These engineering advances origi-
nate in studios, research and engineering laboratories, and companies
manufacturing film, equipment, and accessories. Discussion of these
new techniques and products at the conventions of the Society, and
publication in the JOURNAL of papers describing them, leads to their
prompt acceptance by the industry, and has often resulted in major
improvements in the efficiency of its operations in all departments.

When a program of continuous activity has been carried on over a
period of many years, it would be an unfair misrepresentation to
select certain steps in that program and say that they are outstanding
accomplishments, thereby implying that the rest were routine and
unimportant. A dynamo in a power plant is not less important be-
cause it emits only a steady hum instead of showers of sparks. The
power and light that it furnishes are made visible in other places.
Similarly, the value of the Society of Motion Picture Engineers to the
motion picture industry has been in its continuous program of collect-
ing and disseminating information, evaluating practices, recommend-
ing improved methods, and promoting standardization rather than
in any isolated spectacular accomplishments, though the latter have
not been lacking. The high technical quality of the motion pictures
shown in theaters and the efficient operation of the equipment in
studios, film exchanges, and motion picture processing laboratories all
over the world give evidence that the Society has performed its func-
tions well.

At its first meeting the Society organized four engineering com-
mittees. The names of these committees are sufficient to demon-
strate the seriousness with which the Society attacked its stated ob-
jectives of advancement of the theory and practice of motion pic-
ture engineering and the allied arts and sciences, the standardization
of the mechanisms and practices employed therein, and the main-

214 D. E. HYNBMAN AND J. A. MAURER Vol 47, No. 3

tenance of a high professional standing among its members. These
four committees were (1) Committee on Cameras and Perforations,

(2) Committee on Motion Picture Electrical Devices, (3) Committee
on Motion Picture Theater Equipment, and (4) Committee on
Optics.

From this beginning the committee work of the Society has been
continuously carried on and expanded until today there are sixteen
regular engineering committees on (1) Cinematography, (2) Color,

(3) Exchange Practice, (4} Film Projection Practice, (5) Laboratory
Practice, (6) Preservation of Film, (7) Process Photography, (8)
Screen Brightness, (9) 16-Mm and 8-Mm Motion Pictures, (10)
Sound, (11) Standards, (12) Studio Lighting, (13) Television, (14)
Television Projection Practice, (15) Test Film Quality, and (16)
Theater Engineering, Construction, and Operation. These com-
mittees have truly studied all phases of "motion picture engineering
and the allied arts and sciences."

At its first meeting, in October 1916, the Society began its work of
standardization by considering the dimensions of film perforations.
The record shows that up to that time many difficulties in the produc-
tion and editing of motion pictures had resulted from the lack of one
generally recognized standard for film perforations. At the same
time many prints were being damaged in projection because the di-
mensions of the film perforations and of the projector sprockets were
not in the proper relationship. As the author of one paper on stand-
ardization expressed it, "Fellow members, upon your decisions at
this and coming meetings rest the savings of untold amounts of un-
necessary waste in time, money and material." Standardization of
film dimensions by the SMPE caused these difficulties rapidly to
disappear, and today the industry has almost forgotten that they
ever existed — a most happy state of affairs.

Other fundamental problems which the Society considered at its
early meetings were the principles of operation of the lens systems used
in the projector and the choice of proper equipment for operating the
projection arc lamp with direct instead of alternating current. A
correct understanding of the lens system led to the selection of more
efficient condenser lens combinations, which made possible larger
and brighter pictures, suitable for larger theaters. The arc lamp was
the subject of the first committee report to be published in the TRANS-
ACTIONS, by the Committee on Electrical Devices. This report
contains an admirably clear analysis of the causes of the unsteady

Sept. 1946 PAST AND FUTURE SOCIETY ACTIVITIES 215

and flickering screen illumination often obtained with the then gen-
erally used alternating-current arcs, together with a convincing ex-
position of the inherent superiority of the direct-current arc lamp.
This authoritative statement undoubtedly did much to speed the
general adoption of the superior direct -current equipment, which gave
brighter pictures free from flicker.

Another important activity which began in the first year of the
Society's existence was the creation of an accurately defined motion
picture nomenclature. This work has been carried forward con-
tinuously by the Committee on Standards and Nomenclature (known
today simply as the Committee on Standards) and is being promoted
actively at the present time. Commonly accepted terms and defini-
tions avoid confusion, dispute, and waste.

Any reasonably adequate review of the work done by the SMPE
during the 30 years of its existence would require a book rather than a
brief account such as is being given here. The indexes alone, cover-
ing the engineering papers and committee reports published in the
TRANSACTIONS and JOURNAL, fill 369 pages of small type. Reports
by engineering committees, not counting the reports of the Historical
Committee and the Progress Committee, totaled 244. These figures
give impressive evidence of the continuity of the Society's work and
of its success in stimulating research and the exchange of engineering
information. They fail to show the thoroughness with which all
phases of motion picture technique have been studied by the authors
and committees represented. The work of the Society has been a
perpetual backlog of valuable information on which industry success
has been built.

Many of these committee publications concerned standards.
Collected editions of standards were published in 1920, 1928, 1930,
1934, 1938, 1941, 1944, and will be published again this year, 1946.
The standards adopted by the Society have been recognized and
followed by manufacturers all over the world. They eventually
became official American Standards, and today they provide a
secure basis for international standardization of all dimensions per-
taining to motion picture film and the machinery used with it.

An outstanding accomplishment of the Society was the prepara-
tion, by the Committee on Projection Practice, of comprehensive
plans and safety specifications for projection rooms in theaters.
These plans have been followed in the great majority of theaters
constructed since they were first published in 1*031, and they have

216 D. E. HYNDMAN AND J. A. MAURER Vol 47, No. 3

been officially recognized by being incorporated in the building codes
of several states, including the State of New York. The advantage
of having reasonable and practical safety standards thus formulated
by the motion picture industry itself, instead of having them im-
posed by less well-informed outside authorities, hardly needs to be
emphasized.

Since the formation of the Research Council of the Academy of
Motion Picture Arts and Sciences in 1934, the SMPE has at all times
co-operated with the Research Council to the fullest extent, as it did,
for example, in adopting and publishing the Academy's Standard
Electrical Characteristics for Theater Sound Reproducing Systems.
It has kept the Research Council informed of the engineering work
being carried on by its committees, and has sought the advice of the
Council on all projects of motion picture standardization. The
Society and the Research Council look forward to increasingly close
co-operation in the immediate future, when both will be studying the
engineering 'and industrial problems of television and its relation to
the motion picture industry.

As good a criterion as any of the strength of an organization is its
ability to meet emergencies and deal with them successfully. Several
times in recent years the Society of Motion Picture Engineers has
met and passed this test. In 1935 a serious problem arose when the
German standardizing body adopted a set of standards of 16-mm
sound film which would have made their projectors and film noninter-
changeable with those made in America. During that year and 1936,
the SMPE steadily sought to achieve world standardization in this
then comparatively new industry. These efforts were supported by
Great Britain which, early in 1936, adopted the SMPE standards for
16-mm sound film. Representatives of the SMPE were sent to
Europe, and by the end of 1936 full international standardization was
achieved, involving only one minor change from the earlier SMPE
standards.

A more comprehensive test was presented in the latter part of 1943
when representatives of the Armed Forces requested the SMPE to
undertake an extensive program of war standardization. First the
Standards Committee and then the Board of Governors of the So-
ciety gave prompt consideration to this request, and the Engineering
Vice- President, who has general charge of all engineering committees,
was authorized to proceed with the necessary work of organization.
It was decided that the work should be carried on directly under the

Sept. 1946 PAST AND FUTURE SOCIETY ACTIVITIES 217

auspices of the American Standards Association, but that the personnel
of the committees on motion pictures should be supplied by the SMPE
and by the Research Council of the Academy. The subcommittees
thus organized to work with the War Committee on Photography and
Cinematography Z52, of the ASA, began their work promptly and
within six months had completed the major tasks assigned to them.
These included the development of specifications for 16 -mm Sound
Motion Picture Projectors especially suited to the needs of the Armed
Services, a specification for 16-mm Motion Picture Release Prints,
Methods of Determining Resolving Power of Lenses, and specifica-
tions for eight test films for checking and measuring the performance
of 16-mm projectors.

Later work by these committees included specifications and di-
mensions for screens, for 16-mm projector reels, specifications for
tests required for quality control, standardization of sound records
and scanning area for 35-mm sound motion picture release prints,
and standardization of synchronization marks for release print nega-
tives. Many of the Society's members have also served on other
committees which have arrived at war standards for the field of still
photography.

It is important to note that the rapid rate of progress in this war
standardization work resulted from the fact that most of the problems
presented were closely related to ones that had been studied by the
committees of the SMPE. For example, Subcommittee D of the
War Standards Committee, which prepared specifications for the
Service Model 16-Mm Sound Projector, was able to begin its work
with a draft based on specifications for projectors for educational use
contained in a report of the Committee on Nontheatrical Equipment
published only two years earlier, and this material was found to be so
complete that only three meetings of the war standardization sub-
committee were necessary to arrive at a satisfactory specification.
Similarly the specifications for lens testing and for test films were
based on previous studies by the Society. No organization which
participated in any of the several war standardization programs in-
itiated by the Army and Navy has more right to be proud of the
promptness with which it was able to meet the needs of the Armed
Services.

An important aspect of this wartime work is that it forms a basis
for new peacetime standards and specifications of great value to the
industry.

218 D. E. HYNDMAN AND J. A. MAURER Vol 47, No. 3

A more recent example of the ability of the Society to act in an
emergency is the part it has played in securing frequency allocations
by the Federal Communications Commission of the United States
Government for the needs of theater television. When the Tele-
vision Committee of the Society of Motion Picture Engineers, at its
meeting on September 18, 1944, studied the recommendations of the
Radio Technical Planning Board on frequency allocations for experi-
mental television, it was considered that these recommendations did
not explicitly incorporate the needs of the motion picture industry.
It was decided that the Committee should take steps to insure ade-
quate protection of the future requirements of theater television by
making specific requests for the necessary channels at the Federal
Communications Commission hearings in October 1944.

Accordingly, a delegate was appointed by the Committee to pre-
sent the frequency allocation needs of theater television before the
Commission.

Later in the hearings before the Federal Communications Com-
mission, the Columbia Broadcasting System, Inc. filed a brief in
part of which it opposed the granting of frequency allocations for
theater television. The Society, through its representative, Paul J.
Larsen, promptly presented a rebuttal which must be judged to
have been effective, inasmuch as the Commission on May 25, 1945,
issued a report in which it granted substantially the allocations re-
quested by the Society. Thus the alertness and energetic action of
the SMPE have safeguarded an opportunity for what is likely to
prove to be a major development in the motion picture theater field.
Not stopping with this, however, the Society has begun to work for
the realization of this important development by setting up a new
committee, the Committee on Television Projection Practice, to
study the special problems of installing and operating television
equipment in the theater. As usual, the Society remains in the fore-
front of progress.

Earlier in this paper an example was given of how the standardizing
activities of the SMPE eliminated waste in the production and ex-
hibition of motion pictures. Lack of space made it necessary to
omit mention of many other instances of this kind, among which a
a series of committee reports on wartime conservation is noteworthy.
One such activity that is going on at the present time will be described
because it illustrates the importance of services that the Society is
continually rendering to the motion picture industry.

Sept. 1946 PAST AND FUTURE SOCIETY ACTIVITIES 219

It has been known to a number of engineers for several years that
improved characteristics of the film stock being manufactured today
make possible a more nearly ideal choice of the diameter of the inter-
mittent sprocket of a theater projector than the diameter which has
been in general use.. Projector manufacturers and theaters have been
reluctant to make such a change, however, until the facts were proved
by a sufficiently long series of practical tests.

Accordingly, in 1943 the Standards Committee of the Society set
up a Subcommittee on 35-Mm Projector Intermittent Sprockets, and
this Subcommittee arranged to carry out a comprehensive series of
practical tests with sprockets of different diameters in theaters in
New York City and in Rochester, N. Y. A report on these tests was
presented at the 57th Semiannual Technical Conference of the SMPE
in Hollywood, California, on May 16, 1945. This report showed that
no trouble was experienced with the larger sprockets, while it was
proved that a general change from the currently used diameter of
0.935 in. to the recommended diameter of 0.943 in. would double the
number of projection runs obtainable with any given release print.
The Society has proposed to the American Standards Association
that this new diameter (0.943 in. ) be adopted as an American Stand-
ard, and is taking steps to give proper publicity to the results of its
tests so that the entire industry may be made aware of this oppor-
tunity to conserve film.

Nonengineering Activities of SMPE Committees. — Much work of
importance to the motion picture industry is accomplished by non-
engineering committees and groups of the Society. There are some
14 such committees, among which the Papers Committee is outstand-
ing. This committee is responsible for obtaining material on engi-
neering developments in the industry for presentation at the con-
ventions of the Society. The Technical News Committee gathers
items of current interest to the industry for publication in the JOUR-
NAL. The Historical and Museum Committee collects and assembles
data on early motion picture equipment.

The technical achievements of industry pioneers are reviewed and
considered by the Progress Medal Award Committee, and those
deemed worthy of such recognition are awarded the Society's gold
medal.

These and many other nontechnical groups within the Society
contribute in large measure to the technical literature made avail-
able to the industry and to the public.

220 D. E. HYNDMAN AND J. A. MAURER Vol 47, No. 3

Journal. — Perhaps of greatest importance to the industry is the
JOURNAL of the Society which is published monthly. Over 2500
papers, exclusive of committee reports, have been published in the
JOURNAL on standardization and other industry engineering subjects.

The contents of the JOURNAL have been referred to and praised by
leading engineers and technicians throughout the world, who regard
it as the only complete source for motion picture knowledge.

The Society also publishes booklets on standardization, committee
reports, and recommended procedures.

Engineering Conventions.— Since 1916 the Society has held 59
conventions attended by engineers, technicians, executives, and other
representatives of the motion picture and allied industries. These
semiannual meetings give opportunity for members to exchange hew
developments and to discuss processes and equipment used by the
industry. Papers are presented which are later published in the
JOURNAL. By attendance at these group discussions industry repre-
sentatives are kept informed on subjects of mutual benefit.

Regional Sections. — Between general engineering conventions,
monthly meetings are held by the Atlantic Coast Section in New
York, the Midwest Section in Chicago, and the Pacific Coast Section
in Hollywood, at which similar discussions of industry engineering
subjects are conducted. Thus, the motion picture engineer has fre-
quent opportunities to exchange views and obtain information on
problems confronting him.

Engineering Information Service. — The Society is continually
called upon to supply engineering and technical information to all
branches of the motion picture industry. Letters, telephone calls,
and telegrams are received from studios, exchange branches, and
theater circuits requesting data on a wide variety of subjects. Al-
though the Society cannot at present meet all demands for such in-
formation because of insufficient facilities, it has contributed (as far
as possible) in the general distribution of engineering knowledge for
the mutual benefit of the entire industry.

Proposed Projects. — For several years it has been apparent to the
Board of Governors and Officers of the SMPE that certain activities
ought to be undertaken and others carried on at an accelerated rate
in order to meet properly the present and future needs of the motion
picture industry. Much of this additional work could not be under-
taken in the past because sufficient finances and adequate personnel
were not available.

Sept. 1946 PAST AND FUTURE SOCIETY ACTIVITIES 221

With additional financing now available and with an increased
secretarial staff, the SMPE is now carrying on group engineering at
a much accelerated rate on problems and projects related to produc-
tion, distribution, exhibition, films, equipment, and accessories.

It also has under way the most ambitious standardization program
in its history. Virtually all the motion picture standards in existence
before the war as well as the numerous emergency standards adopted
during the war have been or are being reviewed by a number of sub-
committees of the Standards Committee, and many useful facts
have already become apparent as a result of this study. Many
standards, of course,. have been found entirely satisfactory. Others,
while not changed in substance, have been improved in accuracy and
clearness of presentation. In some cases it has been found that
changed conditions in the industry make definite changes in standards
desirable. In still other cases the need for better techniques than
those known at present has been revealed by this searching study. A
number of important technical papers, discussing problems which
have come to light in the course of this review of motion picture
standards, are to be presented at coming meetings of the Society and
will be published in the JOURNAL.

All standardization work of the Society is done in close co-opera-
tion with the American Standards Association. The Sectional Com-
mittee on Motion Pictures Z22, of the ASA, is sponsored by the So-
ciety, and many of its members are also members of the Committee
on Standards of the SMPE.

Detailed studies of the inter-relations of the television art and the
entertainment field of motion pictures have been under way for the
past year. This work involves such specific projects as studies of
frequency allocation and bandwidth requirements of television in re-
lation to screen definition, private addressee systems, study of the
problems of installing and operating television equipment in theaters,
and correlation of the technical terms used in television with those
used in photographic technology.

Past issues of the JOURNAL and TRANSACTIONS of the Society con-
tain many papers of fundamental importance relating to such sub-
jects as cinematography, sound recording and reproduction, motion
picture laboratory practice, the optics of projection systems, etc.
The usefulness of much of this information, however, is impaired be-
cause of the lack of correlation of the work of the various authors
and because some of the material is out of date. One of the most im-

222

D. E. HYNDMAN AND J. A. MAURER Vol 47, No. 3

COMMITTEES-

CONVENTION
ARRANGEMENTS
PUBLICITY (CONV.)
APPARATUS EXHIBI
PROJECTION (CON

! !

! i

Sept. 1946

PAST AND FUTURE SOCIETY ACTIVITIES

223

portant projects being undertaken by the Society is the correlating,
assembling, editing, and preparation of original material where
needed for engineering reference books and reports on the above men-
tioned subjects and on film exchange practice, motion picture process
photography, motion picture theater engineering, preservation of

PRESENT AND PROPOSED ORGANIZATION

OF
SMPE EXECUTIVE OFFICE

LEGEND:

PRESENT PERSONNEL
'! PROPOSED ADDITIONAL PERSONNEL

FIG. 2.

film, and for a motion picture projectionist's handbook. These
books are urgently needed not only in the industry but also as text
books for the teaching of courses on motion pictures in colleges and
universities. Such courses are now proposed in answer to numerous
requests from members of the Armed Forces as well as from civilians
who, in past years, have often asked the Society to recommend in-

224 D. E. HYNDMAN AND J. A. MAURER

stitutions giving courses in motion picture production, distribution,
and exhibition.

Performing these tasks with an adequately staffed executive office
the Society, in co-operation with the Research Council of the Acad-
emy of Motion Picture Arts and Sciences, will be able to bring about
improvement in engineering practices that will increase economy,
advance public relations, increase the entertainment value of motion
picture productions, and in general raise the stature of the motion
picture industry.

The organizational increases required to carry out this work are
shown in Figs. 1 and 2.

MODERNIZATION DESIRES OF A MAJOR STUDIO*

LOREN L. RYDER**

Summary. — This paper is a discussion of technical things to be accomplished
if motion pictures are to remain the best form of entertainment presentation at com-
petitive cost. It includes ways and means of better utilizing the developments of
World War II, also suggested usages of some of these developments.

Most of the articles in the JOURNAL of the Society of Motion Pic-
ture Engineers are technical discussions of things accomplished.
This is a discussion of things still to be accomplished. It is a state-
ment of the problems facing Hollywood and the industry.

In the future as in the past the success of motion pictures is de-
pendent upon retaining the best form of entertainment presentation
at competitive cost. The motion picture industry has enjoyed lavish
prosperity hinged largely on a technique — a mechanical means of
presentation. The same writers, directors, producers, and actors have
found no equivalent means of entertainment expression.

We have arrived at the present state of the art through two
phases — first, silent and, then, sound pictures. Sound in pictures
along with radio was a derivative of World War I. We are just be-
ginning to feel the impact of the vast developments of World War II.
The objective of this paper is to stimulate thinking and aid in bring-
ing into this industry those developments and devices which have ap-
plication to 35-mm motion picture work. The same thinking and ex-
change of ideas will aid television and 16-mm production.

It must be kept in mind that the studios are manufacturers of enter-
tainment and not designers or manufacturers of equipment and mate-
rials. Most of the development work in the studios is done within
individual departments to meet specific problems in production or
showmanship. The effort in this regard is properly directed and will
no doubt continue in the industry. On the other hand the extension
of this noncentralized engineering practice has and is resulting in an
increasing divergence in the industry equipment and methods. It is

* Presented May 9, 1946, at the Technical Conference in New York.
'* Director of Recording, Paramount Pictures, Inc., Hollywood.

225

226 L. L. RYDER Vol 47, No. 3

cumbersome and costly. It is of necessity short-range thinking; it is
a deterrent to legitimate manufacturers.

In the past most of the worth-while basic research and much of the
applied engineering for this industry has been done by the manu-
facturers. It has been profitable to manufacturer and consumer alike.
The lack of a new horizon and a war have slowed down progress to
incremental improvements. If this industry is to advance with the
trend of the times, both supplier and consumer must discuss the prob-
lems and requirements until there is general understanding.

Development activity in the motion picture industry may be con-
sidered under three main headings: to accomplish new effects in
showmanship; to obtain improved technical quality; and to achieve
economy. Most projects are directed at one of these objectives with
little regard for the other two.

As indicated earlier most development activity on the part of the
studios is directed toward showmanship. The studios are endeavor-
ing to make more real and more spectacular a fiction story. War
experience with stereoscopic gun trainers in a hemisphere or plane-
tarium-type dome was certainly more real and effective to the trainee
than a flat screen. Perhaps the theater of tomorrow should be a
planetarium in which the audience sees everything and hears every-
thing in normal surroundings as in real life. Possibly a sector of a
sphere of grandeur size will accomplish the desired effect. War de-
velopments in optics have exceeded our fondest expectations. Will
it be possible to produce a picture equally satisfactory from the
front, middle, and rear seats of a theater? How can we obtain
stereoscopy? Twentieth Century-Fox has made a test demonstra-
tion of 50-mm color film on a large screen backed by stereophonic
sound. Some people feel that theater sound reproduction would be
enhanced by either a two-speaker system or stereophonic sound.
Others feel that sound to match stereoscopy will have to be truly
three-dimensional rather than subjective from, three horns. We all
know that we need more color.

Progress during World War II was made by men who refused to be
restricted by the limitations of the present. The same approach
should be used in the present thinking, after which the decisions can
be tempered by economics and good business.

Quality improvement has received more industry-wide attention
than either showmanship or economics. It is a more obvious need to
the technician. It is usually more direct to accomplish and is imme-

Sept. 1946 MODERNIZATION DESIRES OF MAJOR STUDIO 227

diately satisfying when accomplished. Unfortunately quality im-
provements do not pay in the box office in a manner comparable to
improvements in showmanship, nor do they pay on the balance sheet
comparable with economy.

Economy is most evasive. The studios are very cost conscious but
the answer is not quite clear. Most industries compete on a cost
basis. A few compete on a quality basis. In picture work the compe-
tition is in showmanship and entertainment. Cost is something that
seems to affect the final balance sheet, making it red or black. In an
effort to gain showmanship the industry has gone through a period of
"gadgeteering" almost without reference to operating costs and with-
out unification. As compared to cost-competitive manufacturers,
motion picture making is obsolete, old-fashioned, and inefficient.
The individualism of showmanship is carried into the mechanics of
production. Everything is still done by force of manpower. This
is not a plea for fewer men but for better tools and devices for the men
to use. Production time is the major cost item in 35-mm production,
16-mm production, and television. With old and obsolete equip-
ment in the field, now is the time to plan the unification and moderni-
zation of the equipment and materials required for these three great
industries. Equipment effecting real economy may prove to be first
in demand.

The following is a discussion of certain activities which may have
application to motion picture making .

Lighting. — In the field of lighting it has been stated that the pres-
ently used 120-v potential was arrived at as being the highest
voltage which could conveniently be taken in shock by a man and
therefore may not be the most effective and efficient voltage for
gaining illumination. Likewise, the present 60-cycle frequency is
merely a marginal improvement over the old 25-, 40-, and 50-cycle
frequencies which are even yet being abandoned. A 400-cycle
supply has become standard for aircraft and much of the Army and
Navy equipment both on shore and the sea. At this frequency trans-
formation is simple and the desired voltage can always be available at
the point of usage. It may be that the industry should make a long-
range review of this situation. Some of the lights now being de-
veloped may operate more effectively and without flicker at this
higher frequency. This includes such lamps as the discharge lamp,
the high-intensity fluorescents, the incandescents of either the stand-
ard type or low voltage high-amperage type such as were developed

228 L. L. RYDER Vol 47, No. 3

for signaling and aircraft landing. Other lamps which show promise
are the zirconium lamp and mercury arc. It is also hoped that some-
thing can be done to our "inkies" and arcs to make them lighter in
weight, more convenient and more effective. Work is in progress in
an effort to increase arc illumination for background projection.

Back-lot shooting areas should be enclosed because of shadow
trouble and sound interference from airplanes. This cannot be done
because there is not enough lighting equipment in Hollywood, ex-
cluding the sun, to light an entire street for Technicolor shooting.
There is a need for some type of general lighting for large fixed sets
and backings.

Photography. — In an effort to gain greater utility with the camera
some of the studios are already developing gyro-stabilization with
servo-control in a manner similar to that used for gun pointing.
This stabilization will be most effective on dolly, camera boom, and
camera car shots. It may also eliminate the costly present practice
of building tracks and special roadways for such shots. The same
type of servo-control mechanisms will be used to gain repetition of
mechanical movement during the picture shooting and during special
effects work. It can be used to time foreground action to background
projection.

Thirty-five millimeter cameras are too large, too heavy, too noisy,
and too covered with gadgets. One Paramount camera was noted to
have 19 gadgets associated with it.

The increase in film speed which has been accomplished is sincerely
appreciated but the industry could use to advantage still faster films
and lenses. The improved image orthicon now available to television
may point the way to a higher speed, lighter, and noiseless camera
with a picture recorder operating off-stage in a manner similar to
sound recording. The light amplifier demonstrated by Dr. Zworykin
may have application to existing types of cameras and lenses.

Sound. — The writer's desire in regard to sound equipment in-
cludes : a microphone, directional at low frequencies and relatively
nondirectional at high frequencies, preferably weighing not more *
than one or two pounds; a microphone boom capable of changing
the angle of the microphone as well as rotation, plus all of the normal
movements — this microphone boom and the cables associated with it
should cast the minimum shadow and, if possible, soft shadow lines ;
a mixer unit about the size of a large book; an amplifier and recorder
mounted in a suitcase or on a lightweight dolly; and a motor system

Sept. 1946 MODERNIZATION DESIRES OF MAJOR STUDIO 229

with motors the size of aircraft motors, also free from all the com-
plexities and inefficiencies of our present motor systems.

Set Construction. — The system of set construction, like the
fabrication of houses, has not been modernized in years. The
theatrical business needs a completely new expendable material for
set construction, for example, a material which may be put together
in a manner similar to carton construction and abandoned after
picture shooting.

The industry needs fast-drying paints, especially paints with high
gloss that can be sprayed on floors between takes to retain the mirror-
like flawless effect which is so spectacular in reviews and dance num-
bers.

The industry needs new carry-alls for set handling and cranes or
elevators for overhead rigging.

Plastics. — New plastics, plywood, adhesives, and glass products
should ajl find their way into motion picture making.

It is the writer's hope that in the immediate future it will be pos-
sible for the industry to give the manufacturers a more complete and
satisfactory statement of the requirements which should make manu-
facturing more certain, more profitable, and more modern.

DUBBING AND POST-SYNCHRONIZATION STUDIOS*

WILLIAM A. MUELLER**

Summary. — The paper covers design and operating considerations for two foreign-
dubbing and post-synchronization studios recently built on top of the Music Building
of the Warner Brothers lot in Hollywood, California. After describing constructional
details to provide satisfactory acoustical conditions in the rooms, the paper concerns
itself with the technique of adding foreign dialogue to completed pictures and dubbing
in replacement lines to photographed sequences which, originally, were too noisy to
permit the recording of intelligible speech.

Before the war the dubbing of foreign dialogue into American pic-
tures was done in the countries in which the picture was released. As
a result of the war, this work necessarily had to be transferred to the
United States and at Warner Brothers we were required to dub a
number of Spanish, French, and Italian versions of domestic releases ;
work which had been previously done in Madrid, Paris, and Rome.

When this program was first started, a review room was equipped
for this purpose, but shortly thereafter the project assumed such size
that it was necessary to build special recording stages to handle the
work.

Inasmuch as foreign dubbing was a temporary wartime adjunct of
our normal operation, the rooms were to be designed to serve equally
well as studio review rooms or narration recording studios. Fig. 1
shows a plan view of two recording studios built specifically for for-
eign dubbing with these specifications in mind. As may be noted, the
projection room is located between the two recording studios, so that
the projection and sound equipment is centralized, and maintenance
and operation problems are simplified. This also reduces film han-
dling, as the rooms were built ont he second floor of another building,
and an elevator was provided to carry filmd irectly from the street
into the projection room.

The rooms are 61 ft long, 35 ft 5 in. wide, and 18 ft high, giving a

* Presented May 9, 1946, at the Technical Conference in New York.
*•* Warner Bros. Pictures, Inc., Burbank, Calif.
230

DUBBING AND POST-SYNCHRONIZATION STUDIOS

231

dimension ratio of approximately 3:2:1, which is in the optimum
range for rooms of this size. This ratio assumes that the length of the
room is measured to the front of the screen, since the space behind the
screen is separated from the room by a heavy drape and cannot be
considered as part of the main enclosure.

In considering the acoustic design, it was desired to use nonparallel
walls, reducing room width toward the front or screen end, which not
only results in good acoustics but also good motion picture presenta-
tion and design, as the audience interest is focused toward the screen.
However, wartime material restrictions prevented this, as splayed
and nonparallel surfaces called for double walls and doubled the

.FILM ELEVATOR

•)
n

[

STRIP Ml
z PROJ. co

KIN

•ga

G
)LJ

z PRO J] ROOM'S

j6MM.]l6MM.

€=S3 Jl T»a^=>
=SD |Rro=
— IBl —

Ml
CC

Kir
N;

G STRIP
OLE PROd.

U S

(/)

N
1 . BEATER .2 ^

THEATER 14

HALLWAY

SIDEWALK

STREET

FIG. 1. Plan view of recording studios.

quantity of strategic materials required but not available. An older
method of acoustic design was therefore resorted to, using alternate
reflecting and absorbing surfaces, with a reflecting surface facing an
absorbing surface on the opposite side of the room. The rear wall of
the room was entirely covered with a sound-absorbing material, hav-
ing a uniform characteristic with respect to frequency. The entire
ceiling was surfaced with hard plaster and the floor covered with car-
peting. The floor was terraced toward the rear wall, which is very de-
sirable from an audience viewing standpoint and extremely important
acoustically as it tends to increase sound diffusion in the room.

The rooms were adjacent to a public street adjoining the studio, so
that considerable attention was given to the matter of noise insula-
tion. Noise measurements on this street showed a maximum noise
level of +74 db referred to 10 ~16 w when using a 70-db weighting net-

232

W. A. MUELLER

Vol 47, No. 3

work. Our previous experience indicated that a maximum room noise
level of about +30 db, using a 40-db weighting network, was neces-

^%^^^^^^^l1j&@r— ^4'B"TT0N LA™ AND PLASTER

-2X6 STUDS 16 O.C.

I/2CELOTEX

IX 3" STRIPS I6"O.C.

&&^&$$&$l£&&&&j? *~3/ 4 BUTTON LATH AND PLASTER

THEATER

FIG. 2. Detail of walls between theaters and hallways and exterior walls.

sary for satisfactory recording conditions, so that 45 db of insulation
was indicated.

This was obtained, first, by placing the corridor of the building next
to the street, which helped to isolate the rooms, and, next, by using a

~3/4BUTTON LATH AND PLASTER
(^-IXS'STRIPS I60.C.
^I/2"CELOTEX

2'X4"STUDS I6"O.C.

ROCKWOOL BLANKET

2X4 STUDS 1 6 O.C.

"*"1' SOLID DIAGONAL SHEATHING

— 3/8'X! 1/2*LATH STRIPS
—3/8 BUTTON BOARD

CEMENT PLASTER 8 METALLATH

PROJ. ROOM SIDE

FIG. 3. Detail of wall between projection rooms and theaters.

laminated wall construction with interior dead-air spaces, as shown on
Fig. 2.

Sept. 1946 DUBBING AND POST-SYNCHRONIZATION STUDIOS

233

The insulation of the projection room wall was also important, as
the projectors have a high normal noise level and must be adequately
insulated from the recording studios. This partition is of the double
wall construction, as shown in Fig. 3, and as a further precaution, the
noise level at the source was reduced by treating the upper parts of the
walls and ceiling of the projection room with a fireproof absorbent.
The air-conditioning system was provided with acoustic baffles at the
inlets and outlets to eliminate noises coming through these openings,

THEATER

4 ROCKWOOL

2*4" CEILING JOISTS I6"OC.
4* ZONOLITE

!" WOOD LATH AND PLASTER
I" ABSORBENT TILE

REHEARSAL ROOM

FIG. 4. Detail of theater floor construction.

and a low air velocity was employed to reduce noise caused by air
flow.

These rooms were constructed by adding a second floor to the pres-
ent single story music building in which were housed several vocal
rehearsal rooms. It was held essential that the different working
units should be able to operate simultaneously without noise interfer-
ence, and the ceiling and floor construction shown in Fig. 4 provides
sufficient insulation between the rooms to achieve this end.

All of these precautions in design and construction have resulted in a
room (Fig. 5) which is excellent for recording from the standpoints of
noise level and sound quality. The actual noise level of the room is

234

W. A. MUELLER

Vol 47, No. 3

+32 db, using a 40-db weighting network while the projection
machine and strip projector are in normal operation.

Since the rooms were designed to serve as review rooms when not
needed as foreign dubbing rooms, a special motor system, was pro-

FIG. 5. Screen end of recording studio.

vided, a schematic drawing of which is shown on Fig. 6. As will be
noted from the illustration, one of the projectors is driven either by a
synchronous or an interlock motor mounted on a common shaft.
When the room is used as a review room, the synchronous motor
drives the projector in the usual fashion. When it is used as a foreign

•

PROJECTOR

INTERLOCK MOTOR
OR
DISTRIBUTOR

SYNCHRONOUS

MOTOR

FIG. 6. Projector drive.

dubbing stage, the interlock motor drives the projector in synchro-
nism with a dialogue strip projector and recording machine during
takes. For rehearsals, the synchronous motor drives its associated
interlock motor as a distributor, and this distributor, in turn, drives
the interlock motor on the strip projector. This is a greatly simpli-

Sept. 1946 DUBBING AND POST-SYNCHRONIZATION STUDIOS 235

fied form of the motor systems previously used and has proved very
efficient and time saving for operations of this nature.

It was found, in dubbing foreign languages into American pictures,
that the best means of cueing the actors was visual. Not only was it
necessary for them to see the picture of the scene to be dubbed, but it
was also found best to supply them with a visual projection of the dia-
logue, written in the foreign language. A special strip projector wa?
used to project this dialogue, as well as a synchronizing mark, on the
screen underneath the picture.

In practice, a scene in foreign dubbing is made as follows :

The picture is divided into short sequences, and the sequences are projected
with the American dialogue once or twice for the actors to get the feeling of the
scene. The loudspeakers are then turned off and rehearsals made, projecting the
picture, and the foreign dialogue beneath it, on the screen. When rehearsals
indicate satisfactory performance, a take is made. Usually, several takes are nec-
essary in order to get proper synchronization throughout the entire scene, and, at
times, parts of several takes are used to complete a more perfect one.

With the end of the war, foreign dubbing in the United States
stopped and was returned to the countries in which the films were to
be released. These rooms are now being used for recording of narra-
tion and post-synchronization of dialogue for our American releases.

The shooting of exterior scenes in all the Hollywood studios has
become more and more troublesome, owing to the tremendous num-
ber of takes that are spoiled by airplanes. On Saturdays, when the
traffic is heaviest, it is very difficult to shoot outdoors. The Sound
Department of this studio has surmounted this difficulty by the tech-
nique of post-synchronization; that is, the sound track which has
been recorded on the set, and was spoiled by extraneous noises, is
used as a cue track, and the actors are taken into our post-synchroni-
zation stages and new dialogue is recorded to correspond with this cue
track. It has been found that some of the techniques developed for
foreign version synchronization are useful in post-synchronizing do-
mestic releases. Many actors respond better to visual dialogue cues
than to the audible dialogue played back for them through a head-
phone. Others work much better with a combination of visual and
audible cues.

In post-synchronizing sound pictures, it is very important that the
mood, feeling, and dramatics of the scene be preserved as they were re-
corded originally. Every effort is made to preserve these characteris-
tics, and it is especially important that the acoustics and pickup

236

W. A. MUELLER

Vol 47, No. 3

quality of the post-synchronized scene match exactly the scenes
which precede or follow it. This is essential, since any change in
acoustics, or acoustic perspective, betrays to the listener that the
scene is a ' 'phoney." For this reason, these rooms are equipped with
every facility for changing the acoustics to match those of the sets
being projected on the screen.

Fig. 7 shows the room in operation and particularly the adjustable
panels which are used to control the acoustics of the space surround-

FIG. 7 A typical post-synchronizing setup.

ing the actor. They are surfaced on one side with a soft absorbent,
and on the other side with a hard reflecting material, and the side
which best simulates the set conditions is used. In addition, the
nature of the floor can be changed from a rug-covered, or carpeted
type, to a hard-surfaced floor, such as linoleum or wood. With this
equipment, it is possible to match the acoustics of practically any
type of scene encountered in normal picture production.

In order to supply proper realism to a scene being post-synchro-
nized, it has been found that the actors must make the movements
which they are shown making on the screen. They absolutely must
not stand still in front of the microphone and read their lines. For

Sept. 1946 DUBBING AND POST-SYNCHRONIZATION STUDIOS 237

instance, if the scene shows an actor seated for a portion of the scene
and then rising to deliver the remainder of his lines, this action must
be repeated in post-synchronization, as there is sufficient difference
in the voice of a person when seated and standing to cause an unnat-
uralness in the recording. Also, the jarring effect, caused by walking,
creates a modulation of the voice that must be duplicated by the ac-
tor by walking as he did when he was photographed. The actor must
walk into a scene, out of it, or toward the microphone in post-syn-
chronization, just as he did in the actual scene, if true realism is to be
achieved. If a person is shown in bed, it is impossible to match the
sound track except by having the actor lie down, and the correct
handling of countless details such as these is necessary to achieve a
natural duplication of the original scene.

The technique of post-synchronization has achieved considerable
savings in motion picture production in Hollywood. By its use, pic-
tures may be "shot" on locations that were previously too noisy to
secure intelligible dialogue, and photographic effects, such as wind,
rain, or lightning, can be used at an intensity which would previously
have ruined the dialogue.

In one of our recent pictures, A Stolen Life, much of the action took
place in a small lighthouse, located on a rocky peninsula jutting out
into the ocean. This rock became an island at high tide, and the
noise of the waves and surf was so loud as to make the dialogue un-
intelligible. Photography at the location was ideal, and by post-
synchronizing the dialogue, a much more satisfactory result was se-
cured, at lower cost, than would previously have been possible.

In another production, there was a scene showing the principals ice
skating on a small woodland pond. The cost of providing a refriger-
ated surface, so that ice skates could be used, was too great, so roller
skates were substituted, which, naurally, were extremely noisy on the
hard floor. The scene was photographed so that it was not apparent
that the principals were not on ice skates, and the noise was eliminated,
for this sequence alone.

In conclusion, it may be stated that actors, directors, and produc-
ers have shown an enthusiasm, totally unexpected, for the facilities
thus provided for extending the application of sound recording to
their art.

The author wishes to thank Michael Rettinger, of RCA, for his as-
sistance in the acoustic design of these rooms and Thomas Sharpe of
the same company for his suggestions on the described motor sys-
tem.

THE RELATION OF TELEVISION TO MOTION PICTURES*

ALLEN B. DU MONT**

Summary. — This paper describes how these two picture-reproducing techniques
can work together with growing advantages to both, rather than engaging in bitter
rivalry as erroneously anticipated.

It is altogether natural that a new art should be viewed with some
suspicion by the older art. The actual scope of the newcomer is a
matter for speculation. There is no* telling at first how great or how
small that scope may be. But in the vast majority of cases the new
art soon fits into its own well-defined groove, serving a heretofore
unserved need. As often as not the new art eventually supplements,
rather than replaces, the older art, thereby rounding out the over-all
services to the public. So what were erroneously regarded as bitter
rivals are ultimately drawn into a workable partnership as mutual
advantages become increasingly apparent and real.

Your motion picture industry is a case in point. When movies
progressed from the level of scientific curiosity, or side show, to that
of the crude entertainment of the nickelodeon, or poor man's theater,
there were some misgivings among theatrical folks. Such misgivings
gave way to genuine apprehension and even strong opposition when
producers of the first full-length photoplays put in bids for topflight
stage stars. And when talkies gave voice to the screen, with natural
color thrown in for good measure, the legitimate stage really had
something to worry about.

However, the movies that seemed such a serious threat at first to
the general welfare of the legitimate stage have in time fitted into
their own particular groove. The recorded or canned show now
parallels that of the time-honored stage. There is little direct con-
flict. Rather, there is close co-operation today. Actors perform for
the stage, and then for the movies, and back again to the stage; the

* Presented Mar. 13, 1946, at a meeting of the Atlantic Coast Section
of the Society in New York.

** Allen B. Du Mont Laboratories Inc., Passaic, N. J.
238

RELATION OF TELEVISION TO MOTION PICTURES 239

better for their broadened experience. Movies have their place.
The legitimate stage has its place. Folks go to see the movie version
and then insist on seeing the stage version. Both can be delightfully
different. Actually, the movies are feeders for the theaters. Theater
receipts were never greater. And so the legitimate stage has found it
profitable to supply the movies with performers, writers, and best
plays. There is no longer the slightest fear of one putting the other
out of business.

History is about to repeat itself with the advent of commercialized
television. However apprehensive the motion picture industry may
have been with regard to television, such doubts have given way to
growing interest and a closer collaboration. It is the purpose of this
paper to deal with some mutual interests that must bring movies and
television still closer together as time goes on.

It must be immediately obvious that the movie-television partner-
ship is already well under way. Movies play a large part in today's
television programming, because film brings the same advantages to
television that it has for theater presentation, plus certain other ad-
vantages. Film images are of excellent pictorial quality, especially
when specifically selected for television reproduction. Film provides
a permanent record for use at any time and in any place. Film pro-
grams can be handled with a minimum of technical personnel in the
television studies, let alone the elimination of actual studio performers
or again the mobile pickup unit out in the field.

Just as film permits simultaneous presentation of a program in any
number of theaters supplied with prints, so it provides a simple and
economical means of syndicating production among any number of
scattered television stations.

Also, film overcomes the problems of timing. The film can be pro-
duced when and as it is most convenient, yet the film can be shown at
any time thereafter. At least half of the news and sporting events
happen during the day. Yet the television audience expects to see
the televised versions at night. Happily, the film recording spans
that awkward gap in timing.

Still another angle: the film recording permits a program to be
shown again and again. Until now, with only three television sta-
tions sharing time in the New York metropolitan area, the audience
has been viewing just the one program available on most evenings.
Yet even at this early stage of commercialized television, there have
been evenings when two and even three of the original television sta-

240 A. B. Du MONT Vol 47, No. 3

tions were on the air simultaneously with outstanding programs
actually competing for audience attention. In the future there will
be seven stations on the air each evening. The audience will ob-
viously be missing interesting programs, just as much as is missed
at a three-ring circus, if we continue the practice of a different show
by each station every night. In order that a given show may be en-
joyed by the greatest audience, it may be that telecasters will borrow
a leaf from the movie industry and repeat their best shows, by means
of film recordings. The same show might be run for three evenings
in a row, after the manner of neighborhood movie houses. Or the
recorded show might be shown again later in the evening or the
following afternoon, in order to reach the maximum audience. At
any rate, film recordings are to television what the transcribed pro-
gram is to broadcasting.

In the case of the sponsored program, particularly the so-called
"commercial" or advertising plug, film is the ideal means of insuring a
uniform identity of product or company. Already many such films
have been made and used to good effect in telecast advertising.
Such films are carefully produced. The whole world of scenery is
available for such shooting. Larger items, such as automobiles,
trains, steamers, airplanes, and so on, can best be filmed in their natu-
ral settings. Film recordings can then be shown over and over
again with that positive uniformity of presentation so vital to good
advertising. Film is to television advertising what the stereotype
mat is to newspaper advertising — the foolproof reproduction.

Even in studio production, film has its vital place. Time and again
a studio production has troublesome gaps or pauses that must be
bridged over by some suitable pictorial action, since the television
public will not tolerate a blank Teleset screen even for a few seconds.
Film "shorts" help fill in such pauses.

Too, television plays have been considerably enhanced by the in-
clusion of movie scenes, frequently made with the same performers
amid the desired outdoor scenes. For instance, if the television play
calls for a bit of action, say on Fifth Avenue, or for a train, steamer,
bus, or airplane, it is evident that such a scene can best be made out-
doors with the same actors, rather than attempt a synthesized ver-
sion in the studio. Filmed scenes also gain time for shifting from one
studio scene to another.

Economically, film production presents certain advantages. Tele-
vision studio problems can be frequently solved by film shooting.

Sept. 1946 RELATION OF TELEVISION TO MOTION PICTURES 241

For one thing, studio space limitations can be overcome by having
the production made in a movie studio and recorded on film. Like-
wise, if the studio schedule is overflowing, the production may be
filmed even in the television studio itself, thereby dispensing with
the lengthy rehearsals of the live-talent show. If performers are
available only at certain times that do not conform with telecasting
hours, the film recording again solves the problem. At any rate, the
entire production can be filmed and used at will, without tying up
limited studio facilities.

Of course the simple performance can be handled at lower cost with
studio live talent and direct television pickup. This will always be
the logical choice especially when a single television station must bear
the entire cost. But for more elaborate productions and where
several stations are participating in a syndicated program, then the
movie method of production becomes increasingly more attractive.
In all cases the cost comparisons should largely settle the choice of
direct television pickup of film recording.

Many of the programs handled by Du Mont Television Station
WABD in New York City are recorded on film as standard produc-
tion routine. We have a threefold purpose in recording many of our
programs :

(1) Such films provide a handy record that may be studied by our studio
personnel, technicians, and again the performers, directors and writers, to improve
their respective talents as time goes on.

(2) Such films are supplied to advertisers, as a permanent record of their pro-
grams.

(5) Such films serve to build up a growing library of recorded programs that
can be used again either over our own stations, or syndicated to other stations.

Du Mont technicians have worked for several years on the many
problems of recording television images on movie film. We have
evolved a satisfactory technique, whereby television images of a
repetitive rate of 30 pictures per second on the cathode-ray screen
can be recorded on movie film at 24 frames per second or any other
rate required. Our own recordings are made on standard 16-mm
film, with sound track included, for a completely recorded television
program.

The greatest problem in film recording of television programs di-
rectly off the cathode-ray tube of the television monitor is the diffi-
culty of synchronizing the 30 frames per second speed of television to

242 A. B. Du MONT Vol 47, No. 3

either the 16 frames per second of silent motion pictures or the 24
frames per second of standard sound film.

There are two practical solutions to these problems :

(2) To record silent pictures at 15 frames per second using a synchronous
motor drive on a standard camera and projecting this film at the standard speed
of 16 frames; or

(2) Recording at standard sound speed of 24 frames per second using a
specially constructed shutter and pull-down in a camera also driven by a syn-
chronous motor. This will allow the film to be projected at sound speed from
a standard projector.

15 Frames per Second Silent. — As stated before, television
operates at 30 frames per second. If a standard motion picture
camera with a shutter of approximately 204 deg is driven by a syn-
chronous motor at 15 frames per second, half of the alternate 30
television frames will be recorded, the other half will be lost during
the pull-down time of the camera with the result that 15 frames per
second will be recorded. In projecting a film taken by this method
at the standard 16 frames per second, no particular speeding up of the
subject action is noticeable.

24 Frames per Second. — In recording television 30 frames per
second at standard 24-frame sound speed the difficulties are not so
easily overcome; however, these problems are almost entirely of a
mechanical nature.

Again a synchronous motor is used to drive a standard camera at 24
frames per second, but both the shutter and pull-down mechanism
must be altered so that 6 television frames out of every 30 are lost
during the pull-down time of the camera, resulting in a 24 frame per
second recording of the 30-frame television picture.

As the film travels through the camera at sound speed, sound can
be recorded in the usual ways, either on the same film using a single
system or by a separate sound camera using a double system.*

The motion picture business is based on sequence of runs, which is
based on pricing. So far not enough money has been offered by
telecasters for film to warrant any deviation from their normal ar-
rangements. Because of this, it is evident that motion picture pro-
ducers have been unwilling to supply first-run feature pictures or,
for that matter, even news reels or short subjects to telecasters.

Consequently, telecasters have had to depend on entertainment

* The film shown at the end of this paper was recorded at 15 frames per second,
but projected at sound speed with background music and voice dubbed in.

Sept. 1940 RELATION OF TELEVISION TO MOTION PICTURES 243

films of more or less ancient vintage — films from which the movie
industry has already extracted just about the last dollar of box-office
revenue. Speaking for my own organization which has pioneered
in telecasting through our New York Station WABD (originally Sta-
tion W2XWV), and more recently through our Washington Station
W3XWT, I cannot point with particular pride to much of our film
programming during the past several years. Certainly we would not
pick many of the old-time films we have shown as a matter of choice.
Along with other pioneer telecasters, we have had to show films of
five, ten, and even fifteen years ago. Time and again our audience
has witnessed the all-too-obvious turning back of the hands of time
as we have flashed on their Teleset screens a less polished Bing Crosby
of the early '30s, a precocious Shirley Temple when she was only knee
high to a grasshopper, and a youthful Charlie Chaplin at the peak of
his career, and so on. The pictorial quality of such ancient times is
positively an imposition before the eyes of today's critical audience.
Yet such ancient films can prove interesting and worth while at times.
Indeed, Station WABD has even capitalized the antique touch, and
its audience had positively enjoyed such backward glimpses into the
"good old days" with the showing of the Charlie Chaplin classics.
No one will deny that the libraries of ancient films have been of
inestimable help to pioneer telecasters. Such films have served to
fill in a third to a half of our evening programs until such time as we
have been able to build up our studio live- talent features.

In addition to the ancient films, we have been fortunate in having
an ever-increasing supply of documentary films, notably in connection
with the war. British and other films have been available to tele-
casters; also good documentary films produced in this country by
governmental bureaus, industrial concerns, universities, and others.
Such documentary films can provide fair entertainment at times, al-
though they definitely lean toward enlightenment and even sheer
propaganda. But since such films are usually not shown in theaters
and may generally go unseen by the general public, television now
provides a logical means of making such documentary films available
to a general audience.

For years past the producers and sponsors of documentary films
have sought an audience for their wares. They knew such an audi-
ence existed, but had not the means of reaching it. It would seem
to me that television, especially as it reaches out into schools, pro-
vides a logical means of bringing worthy documentary films to the

244 A. B. Dti MONT Vol 47, No. 3

attention of the vast nontheatrical audience. A single print can be
shown simultaneously to a huge audience limited only by the number
of Telesets that may be installed.

Sooner or later, and it should be soon, telecasters must seek new
sources of film. True, there is a rising proportion of studio live-talent
material now in the making, but as telecast programs are lengthened
to cover afternoons as well as evenings, we may be obliged to use an
increasing amount of film entertainment.

The time is coming when television should have suitable film pro-
ductions of its own. Such films should be geared to television re-
quirements— technically as well as in subject matter. Already some
telecasters have covered news events with their own cameramen.
Such presentations have been well received by the television audience.
Too much cannot be said for the splendid news reporting of such
events as the return of General Eisenhower to Washington last
summer; the signing of the Japanese surrender aboard the Missouri
in Tokyo Harbor, telecast a week later to the television audience in
the New York area; and the San Francisco meeting of the U.N.O.
Much of this news reporting has been handled with 16-mm camera
equipment, and it is important to point out that this smaller film
televises about on a par with the 35-mm size. Thus the telecaster
can work with the convenient and economical 16-mm equipment
and film, which is especially significant for the smaller television
station covering local news and sports.

Motion picture producers, knowing of the need for better films by
the telecasters now on a truly commercial basis, may soon be pro-
ducing films strictly for television use. That is logically their job.
They are now being invited to supply the need. As more and more
telecasters take to the air, the economies of special television film
productions will be worked out so that this market can prove worth
while for film producers.

So far this paper has dealt with the contributions past, present,
and future of movies to the up-and-coming art of television. But
this movie-television collaboration is definitely a two-way proposi-
tion. In time television will contribute much to movie technique, as
well as movie economics.

Among the earliest television contributions to the motion picture
art is the Du Mont film-recording method of presenting televised
events in theaters. Our engineers have long worked on the problems
of projecting bright television images of adequate detail on full-sized

Sept. 1946 RELATION OF TELEVISION TO MOTION PICTURES 245

theater screens. Today we have high-brilliancy tubes and large aper-
ture lenses for approximating these results. Nevertheless, we have
developed another method which offers more satisfactory results by
way of recording the television images on film and then projecting
the film in the conventional manner.

The Du Mont equipment for this purpose— covered by the broad
basic U. S. Patent No. 2,373,114— includes a high-brilliancy cathode-
ray tube carrying the television image. The image is photographed
on movie film, along with the sound track. The film is then auto-
matically developed, fixed, rinsed, and dried, ready for conventional
projection in a matter of minutes. The film can be cut and spliced,
titled and edited, as necessary. The televised news event, taken off
the air or coaxial cable, is on the screen almost as soon as it happens,
so that it still qualifies as seeing- while-happening reporting. Yet the
film makes possible the showing of the event as often as may be de-
sired, which is a prime requisite of the theater or movie house.

The bulk of the revenue for big boxing bouts of the near future
will no doubt come from theater television. Likewise, with other
sporting events. For the box office is still the logical place to collect
for such features, and this television-filmed technique provides the
practical means of multiplying the paying audience to untold pro-
portions. This service is not to be confused with usual news reels,
since it presents the event in the matter of minutes as against hours
for the usual movie version. And a televised event — seen as it
happens — must always have fresher and greater box office appeal.
Meanwhile, the televised event will also be available to news reels
for usual distribution.

The possibilities of television-filming are simply unpredictable.
Even at this early date the television studio and its control room may
well be the envy of the movie producers. In television we have a
plurality of cameras on the studio floor, each transferring its pickup
instantly to a respective monitor screen in the control room. The
production director has before him the respective pickups of all
cameras. By means of the intercommunicating system, with ear-
phones worn by the cameramen, he can instruct any cameraman as to
desired shots. Any single pickup can be selected and transferred to
the transmitter for placement on the screens of the telesets of the
audience. More than that, any combination of scenes can be used
by corresponding switching. Also, there are electronic faders, lap
dissolves, and other effects largely duplicating movie camera tech-

246 A. B. Du MONT Vol 47, No. 3

nique. The control-room operators can obtain simple or intricate
montage effects by electronic manipulations of the pickups of two or
more cameras, while each component of such a montage is under
complete and immediate control.

As an interim step in television-filming, the remote electronic view
finder idea may interest movie producers. The usual film cameras
are still used in the conventional manner, but attached to such
cameras is a miniature television camera which transmits the view
finder image to a screen before the director. Thus the director has
before him the exact scene for which any camera is set at that given
moment. The director can phone the cameraman and give instruc-
tions, while viewing the new setup of the camera as such instructions
are followed. When the scene is properly set in the electronic view
finder, the order to "roll" the camera follows. What such co-ordina-
tion could mean to the director of gigantic spectacles, covered by
many cameras in scattered locations, is left to your imagination.

As time goes on the pictorial quality of televised images will
steadily improve until it is on a par with motion picture film. Tele-
vision-film recording will then be fully feasible, with television
cameras transferring their images to a central control room where the
director and his technicians will select the choicest scenes and actions
for recording.

Another fascinating television-filming possibility is found in the
growing sensitivity of the television cameras. The new image -
orthicon tube, with a sensitivity 100 times greater than that of pre-
vious television tubes, now picks up scenes in moonlight, by candle-
light, and in any kind of weather. Already we have reached a point
in television camera technique whereby poorly lighted scenes that
cannot be filmed directly on film emulsions can now be recorded
through the intermediary of television. An entirely new world of
movie possibilities is opened up by this supersensitive television
pickup.

Television is certain to be a powerful influence in future educational
methods. Telesets placed in various classrooms can bring an edu-
cator or lecturer or educational features before vast numbers of
students at one time. This facility may well have an important
bearing on the economics of visual education films, since a single
print can now be shown simultaneously to many different classes
and in many different schools at one time, through telecasting over
cable and over the air.

Sept. 1946 RELATION OF TELEVISION TO MOTION PICTURES 247

Television likewise finds its place in the merchandising field. In
addition to its use in theater, home, and school, television is entering
the department store for the purpose of reaching more people in
more departments with the offerings of other departments. Telesets
are located at strategic points throughout the large department
store. A center studio is set up, with the necessary cameras and as-
sociated equipment. Before the cameras may be placed certain
goods to be displayed, or models, to fashion the latest garments, or a
demonstrator with something to demonstrate. Instantly the image
and the voice are made available at strategic points throughout the
store, thereby overcoming to a large degree the penalty that has been
paid heretofore for magnitude.

In place of the live pickup, suitable merchandising films may be
used. Or the given live- talent pickups may be recorded on films for
repetition. Again television and film work hand in hand.

There are many heretofore inaccessible places to be filmed. Con-
ventional equipment is cumbersome, and lighting conditions may be
hopeless with usual film emulsions. But here again, television can
step in. Television cameras, devoid of moving parts and motors,
can be reduced to extreme compactness and light weight, if these are
the prime considerations. Such cameras can be carried to places
heretofore considered inaccessible, and the pickups flashed over
coaxial cable or ultra-short-wave link to a central recording point
where the images are copied on film. The production director at
the central point can follow the camera work and ask for precisely
what he wants.

Movies and television are natural partners. One supplements
the other. Movies are the permanent record. Television is the
more advanced way of getting the picture. Television owes much to
movies up to this time. But from here on movies will be receiving
increasing benefits from the rapidly refining television technique.

NONINTERMITTENT MOTION PICTURE PROJECTOR
WITH VARIABLE MAGNIFICATION*

F. G. BACK**

Summary. — jn the course of the Navy aviation training program, a projector
had to be designed to project the image of a target vessel on a curved cyclorama screen,
and to make this projected image perform all the real and apparent motions of an
actual battleship, as seen from the cockpit of a maneuvering aircraft.

The projected image had to wander all around the horizon. It had to become larger
and smaller under due consideration of the angle of depression corresponding to that
particular range. Also, the projected target had to be able to make all kinds of turns
and maneuvers to simulate actual combat conditions. To achieve all this a special
nonintermittent 16-mm film projector for variable speed and variable magnification
was built, and is described in the following paper.

The problems confronting the motion picture projector designer in
the past have been manv and varied, but never has it been his task
to cast upon the screen an image which executes various movements
at the operator's command through remote electronic control.

To produce a satisfactory Naval Aviation Training Projector the
image of a military object has to move around the spectator; it has
to vary in distance from the spectator, and it must make movements
around its own vertical axis. A combination of these movements
must give the same impression as viewed by an aerial gunner under
actual battle conditions.

All these movements must be performed at different speeds in all
possible directions strictly related to electrically controlled devices at
the operator's command, and in relation to target movements.

After intensive and thorough research it was found that the best
performance could be obtained only with a specially developed non-
intermittent motion picture projector fixed in the center of a cyclo-
ramic screen.

To solve the problem, the required independent motions were re-

* Presented Oct. 15, 1945, at the Technical Conference in New York.
** Research and Development Laboratory, 381 Fourth Ave., New York.
248

PROJECTOR WITH VARIABLE MAGNIFICATION

249

garded as basic movements, the combination of which would result
in the required performance.

The basic movements are as follows :

(1) The movement of the image around the observer,

(2} The movement of the image
toward and away from the ob-
server,

(5) Altering the angle of depres-
sion under which the image on the
screen is seen,

(4) The movement of the image
around its own axis.

The first component move-
ment is comparatively simple ;
the image has to move around
a horizon in the cyclorama.
It is effected by rotating the
whole projector around its own
vertical axis, Fig. 1A.

The second component
movement (the movement to-
ward and away from the ob-
server) is achieved by a special
varifocal projection optic.
This special optic allows the
change of magnification with-
out changing the location of
the object or the image;
namely, the film and the
screen, and without impairing
the optical quality of the
image.

The angle of depression,
under which the image on the
screen of the cyclorama is
seen, is changed by the movement of a mirror B which deflects the
projection beam (Figs. 1 and 2).

The fourth component movement (the movement of the image
around its own axis) is obtained through a special nonintermittent
film transport mechanism with a 16-face prism as optical compensa-
tor, which allows an endless film loop to project, with variable speed,

FIG. 1. Projector on demonstration stand.

250

F. G. BACK

Vol 47, No. 3

backward and forward, without slippage or play. The endless film
loop, which is 16 ft long, shows a complete revolution of the target
around it own vertical axis.

By the use of remote control it is possible to operate the projector
by combining either all basic movements, or by using only requested
movements; e. g.y a combination of the movement toward the ob-

FIG. 2. Optical system.

server with the movement of the image around its own axis would
result in the following illusion : The observer sees the image under
a certain angle of depression coming toward him, becoming larger
and larger by means of the varifocal projection optic. At the same
time, the operator turns the image by starting the nonintermittent
film transport mechanism.

Sept. 1946 PROJECTOR WITH VARIABLE MAGNIFICATION

251

If the operator lets the image make only a partial turn and reverses
this movement by letting the projector run backward, and repeats
this action, the observer then gets the impression that the image
meanders. The above-described movement combination can be per-
formed at any required speed and as often as desired according to the
revolution of the projector mechanism.

It is possible, if required, to repeat this performance under another
angle of depression, thus giving the observer the impression that he
is viewing the whole performance from another altitude. This image

FIG. 3. Projector mechanism.

performance can be varied according to the given requirements and
can be combined with the movement around its own axis. The image
then travels around the cycloramic screen.

The speed of all the above-described movements, either separately
or combined, is controlled by the use of electronic devices and regis-
tered on a central control point at the operator's desk. On this con-
trol point is also registered the reaction of the observer relative to
the movements of the target image. This makes it possible for the
operator to perform all target movements at the speed required by
the simulated conditions set up in the various phases of the training
program.

252 F. G. BACK Vol 47, No. 3

Consequently, the design of this instrument, built with the utmost
precision, had to follow in the outline the mechanical requirements
which would produce a perfect illusion.

(1) The vertical drive (Fig. 1G), which produces the movement around the
observer, including the slip-ring arrangement (Fig. ID),

(2) The nonintermittent film transport mechanism (Fig. 3),

(5) The lens tube arrangement (Fig. 2), which produces the movement of the
image toward and away from the observer, and altering the angle of depression
under which the image on the screen is seen by the observer.

The vertical drive, together with the nonintermittent film trans-
port mechanism is the upper unit of the projector. The lens tube
with the mirror (Fig. 2) is the lower unit.

The vertical drive (Fig. 1) comprises the slip-ring arrangement
(D) ; drive motor (G), the upper and lower arm, (H, I), which supports
the whole unit with the aid of the column (/) on the base plate (K).

The motor (L) is a Selsyn motor, part of the electronic equipment
which signals the required actions. The slip-ring arrangement makes
it possible to locate the various necessary connections for the electric
remote control system in central position. It is fixed to the projector
body and forms the axis around which the projector turns.

The projector body (Fig. 3) carries on its base plate the film drive
motor, together with a Selsyn motor.

Attached to two fixed arms (M] and one pivoted idler arm (AT) are
16 film idler rollers (P) which guide the endless film loop.

Two guide rollers lead the film in the sprocket which transports the
film.

Two special gate rollers (Q) are mounted above the 16-face prism.

One compensating arm (R) and one framing arm (S) enable the
operator to obtain the correct frame.

The light source is a 1000-w projection lamp. A blower motor
(T) with two fans serves for cooling purposes. The lamp house can
be tilted backward, as shown, to facilitate the changing of the endless
film loop. The light beam is transmitted through a special aplanatic
condenser, which gives a maximum of illumination without applying
undue heat to the film. A 90-deg deflecting prism (U) an integral
part of the condensing system, changes the direction of the light beam
from horizontal to vertical.

The lens tube arrangement contains the optical system, the mirror,
and in addition, all the necessary electrical and mechanical equipment
to co-ordinate the various movements which are essential to achieve

Sept. 1946 PROJECTOR WITH VARIABLE MAGNIFICATION 253

the desired requirements, namely, changing the size of image, altering
the angle of depression, and retaining the brightness of the projected
image.

In Fig. 2, the cams ( V, W, X) are turned by a small electric motor
which gets its signals through remote control. Microswitches pre-
vent over-travel in the entire system.

The lens tube is flanged to the projector body and rotates with it.

The optic of the lens tube consists of two objectives, one short-
focus and one long-focus. The short-focus lens produces a real image
of the film in a field lens, and this image in turn is cast upon the screen
by the long-focus objective with the aid of the deflecting mirror (£).
The change of magnification is obtained by moving the short-focal
lens relatively to the film and by compensating the resulting change
in focus with an appropriate movement of the long-focus lens.

It was found after plotting the necessary movements of the two
lenses against the range, to which the required image size is co-ordi-
nated, that one lens movement is linear, while the other movement
follows an irregular curve of the fourth order. The linear movement
was achieved by using a steel tape rolled on a disk ( Y) . The irregular
movement was effected by the use of the cams (X), roller, rack, (X2)
combined with the disk and steel tape (Y). The cams (V, W) are
connected with a mirror bracket (B). They are used selectively,
depending on the required altitude. The changing from one cam to
the other is accomplished by using the clutch (Z), and the lever move-
ment with the aid of the shift lever (Zi) .

Fig. 1 shows the assembled projector as it was installed in the vari-
ous Navy establishments (with the exception of the three chromium-
plated legs attached to the base plate, which have been used for
demonstration purposes only. They are replaced by special fixing
bolts.)

A number of these projectors have been in use for some time in
various Navy Training Centers.

A FILM-SPLICING AND REPAIR MACHINE*
ARMOUR WALLINGSFORD**

Summary. — Editing film with its constant splicing and movements through
moviola, synchronization, and projection machines subjects the prints to severe wear
and damage. Splices and repairs must be made speedily and accurately. Time also
is an important factor in meeting the release date.

This paper describes a new type of splicing machine which does not use cement,
heat, or require scraping of film.

This machine is a compact portable metal device capable of mak-
ing straight or diagonal splices, restoring lost and torn perforations,
and repairing either picture or sound tract without replacing sections.
This is accomplished by the application of a good quality of pressure-
sensitive tape which .is applied to the film and perforated in exact
registration with the film perforations. By using a specially prepared
tape, any splice may be "blooped" as it is made without extra opera-
tions.

The splicer is composed of a cutting anvil which removes one com-
plete frame or any desired section of film, and a splicing anvil which
completes the straight or diagonal splice, makes the perforations,
and trims the edges. The perforating is controlled by four re-
tractable registration pins which may be operated in unison, or in
any desired combination. An upper arm containing eight punches,
a pressure plate and trimming shears, when lowered onto the die
anvil completes the film splice on one side. When this operation
is repeated on the reverse the splice is ready for use.

The tape is contained in a readily accessible built-in housing which
facilitates easy replacement of the roll.

The film splicer reduces, by approximately one-half, the manual
movements required by the present method. It can be placed on the
editor's table, and splices may be made without removing the reels,
thus eliminating the use of paper clips.

* Presented Oct. 15, 1945, at the Technical Conference in New York.
** Editorial Sound Department, Republic Studios, North Hollywood, Calif.

254

A FlLM-SPLICING AND REPAIR MACHINE 255

At the present time- the three models of this machine for S-, !(>-,
and 35-mni film are in the process of development. The description
herein is made from the author's model which has been in constant
use for the past six months in the Editorial and Sound Department of
Republic Studios, North Hollywood, California. The machine was in
actual use splicing picture and sound track area without effecting
sound track for dubbing purposes. This was accomplished without
a single failure.

The cutter is a stationary split anvil. It has four registration pins,
two on each side. The machine improvements will include four other
retractable pins for the preparation of diagonal splices. Two holding
bars are hinged to fit over the register pins and anvil to hold the

FIG. 1. Cello-vision splicing machine showing film
cutter, stationary anvil, tape container, and overhead
male punch.

film in cutting. A cutting bar is also hinged on the opposite side of
the anvil and cuts out one frame of the film. The cut is designed to
produce sufficient overlap for the next operation of splicing.

The splicing anvil is also stationary and consists of a male and fe-
male punch die. The female die has eight- perforations which are
the same dimensions as film perforations. It is also provided with a
set of four registration pins which may be retracted independently of
each other by the use of four separate cams, which are controlled by
two knobs.

The male punch, is mounted on the frame, is a hinged U-arm
which is held in the open position by a spring. The punch has eight
punches to fit the female die. It is also provided with a pressure
plate to maintain the film in correct alignment during the splicing

256

A. WALLINGSFORD

Vol 47, No. 3

operation. The punch also contains two trimming or shear cutters
which coincide with the exact width of the film.

C ^JR

FIG. 2. Sub view of cam-operated knobs controlling
movement of registration pins in order to restore per-
forations and make repairs.

The film is placed on the cutting anvil and the desired cut is made.
The adhesive, which is a pressure-sensitive tape, is now obtained from
a container which is a part of the machine itself. Sufficient tape is
pulled out and the end is fixed on the near side of the splicing anvil

FIG. 3.

Tape from container in position and cut
end of film ready for splicing.

in readiness to make the splice. The ends of the film are placed on
the registration pins of the anvil in overlap relation under the adhe-
sive tape. The adhesive tape is now brought down on the film and
the anvil by mere finger pressure.

Sept. 1946 A FlLM-SPLICING AND REPAIR MACHINE 257

The overhead punch is brought down and perforates and trims the
adhesive tape from the edges in a single operation. The splice is now
cemented on one side. The film is turned over and the operation is
repeated on the other side of the film to complete the splice.

The machine is capable of making straight and diagonal splices,
and by the use of a special prepared tape, it will bloop automatically
every splice that is desired. Repairs may be made on any type of
break or tear without replacing the damaged sections, in sound
track or picture. Repairs have been made on push-pull modulation
where the tear occurred in the perforations and sound track area
without noise for dubbing purposes.' This is accomplished with the
application of the tape to the damaged section and in successive opera-
tions of the punch.

AMERICAN STANDARDS ON MOTION PICTURES

FOREWORD

THe six newly revised American Standards on Motion Pictures pub-
lished here were recently approved by the American Standards As-
sociation and represent another forward step in the present program
of motion picture standardization. All such existing standards within
the scope of Sectional Committee Z22 of the ASA have been reviewed
within the past year, and the first 20 revisions appeared in their new
distinctive format in the April 1946 JOURNAL. The six following
comprise the second group in this series, published first in the
JOURNAL and then made available to the industry, on 8l/z X 11-in.
sheets, punched to fit the new SMPE Standards Binder.

Revision of the first three of these Standards, Z22.28, Z22.29, and
Z 22.31, consists of title changes required by current American Stand-
ards Association editorial policy. Revision of the other three Stand-
ards, Z22.37, Z22.38, and Z22.39, which were originally published
in the August 1944 JOURNAL, had been at first thought unnecessary,
but title inconsistencies between two printed versions of the three
standards, together with a desire to have all Z22 Standards fit the
new binder, seemed to justify having them set in the new format.

Copies of these six Standards, and the twenty published in the
April 1946 JOURNAL, may be secured from the General Office of the
Society.

AMERICAN STANDARDS

Z22.28-1946 Projection Rooms and Lenses for Motion Picture

Theaters

Z 22. 29-1 946 Theater Projection Screens

Z22.31 -1 946 Motion Picture Safety Film

Z22.37-1944 Raw Stock Cores for 35-Mm Motion Picture Film

Z22.38-1944 Raw Stock Cores for 16-Mm Motion Picture Film

Z22.39-1944 Screen Brightness for 35-Mm Motion Pictures
258

AMKRKAN vSTANDARDS ON MOTION PICTURES 2.V.)

American Standard Dimensions for

Projection Rooms and Lenses for
Motion Picture Theaters

Z22.28-1946

1. Projection Lens Height

1.1 The standard height from the floor to the center of the projection lens
of a motion picture projector should be 48 inches.

2. Projection Angle

2.1 The projection angle should not exceed*! 2 degrees.

3. Observation Port

3.1 The observation port should be 12 inches wide and 14 inches high, and
the distance from the floor to the bottom of the openings shall be 48 inches.
The bottom of the opening should be splayed 15 degrees downward. If the
thickness of the projection room wall should exceed 12 inches, each side
should be splayed 15 degrees.

4. Projection Lens Mounting

4.1 The projection lens should be so mounted that the light from all parts
of the aperture shall traverse an uninterrupted part of the entire surface
of the lens.

5. Projection Lens Focal Length

5.1 The focal length of motion picture projection lenses should increase in
'/it-inch steps up to 8 inches, and in Vz-inch steps from 8 to 9 inches.

6. Projection Objectives, Focal Markings

6.1 Projection objectives should have the equivalent focal length marked
thereon in inches, quarters, and halves of an inch, or in decimals, with a plus
(-(-) or minus ( — ) tolerance not to exceed 1 percent of the designated focal
length also marked by proper sign following the figure.

NOTE: Complete plans for projection rooms are contained in the Journal of the Society of
Motion Picture Engineers, p 484, November, 1938.

260

AMERICAN STANDARDS ON MOTION PICTURES Vol 47, No. 3

American Standard Dimensions for

Theater Projection Screens

Reg. V. S. Pat. Off.

Z22.29-1946
First Edition
Z22.29-1941

1. Screen Size

1.1 Sizes of screens shall be in accordance with the table below.

2. Grommet Spacing

2.1 The spacing of grommets shall be 6 inches. In rare instances, however,
12 inches will be permitted. The ratio of width to height of screens shall be
4 to 3.

3. Screen Placement

3.1 The width of the screen should be equal to approximately 1/6 the dis-
tance from the screen to the rear seats of the auditorium. The distance
between the front row of seats and the screen should be not less than 0.87
foot for each foot of screen width.

Screen Sizes

Size No.
of Screen

Picture
Width
(Feet)

Picture Height,

Size No.
of Screen

Picture
Width
(Feet)

Picture Height,

Feet

Inches

Feet

Inches

1,2

3 '

Sept. 1946

AMERICAN STANDARDS ON MOTION PICTURES

261

American Standard Definition for

Motion Picture Safety Film

Ktt- V. S. Pat. Og.

Z22.31-1946

First Edition

Z22.31-1941

1. Safety Film

1.1 The term "Safety Film" as applied to motion picture materials shall
comply with American Standard Definition of Safety Photographic Film
Z38.3.1-1943. All 32-mm, 16-mm, and 8-mm film must be of the safety type.

262

AMERICAN STANDARDS ON MOTION PICTURES

Vol 47, No.

American Standard

Raw Stock Cores
For 35-Millimeter Motion Picture Film

Rr». V. S. Pat. Off.

Z22.37-1944

Millimeters

Inches

A
B
C

25.90 ± 0.20
50.00 ± 0.25
34.50 ± 0.50

1 .020 ± 0.008
1.968 ±0.0 10
1 .358 ± 0.020

Recommended Practice

R
S

16.70 ±0.30
4.00 ± 0.20

0.657 ±0.0 12
O.I 57 ±0,008

Bore A to fit freely to hub 25.40 ± 0. 1 mm or
1 .000 ± 0.004-inch diameter.

NOTE: Reprinted August 1 946, without change.

. 1 (.UC'

A.MKRICAN STANDARDvS o\ MoTlON PICTURES

203

American Standard

Raw Stock Cores
For 16-Millimeter Motion Picture Film

/•«/. ng.
Z22.38-1944

Millimeters

Inches

A
B
C

25.90 ± 0.20
50.00 ± 0.25
15.50 ±0.50

1 .020 ± 0.008
1.968 ±0.0 10
0.61 0± 0.020

Recommended Practice

R
S

16.70 ±0.30
4.00 ± 0.20

0.657 ±0.0 12
O.I 57 ±0.008

Bore A to fit freely to hub 25.40 ± 0. 1 mm or
1 .000 ± 0.004-inch diameter.

NOTE: Reprinted August 1 946, without change.

264 AMERICAN STANDARDS ON MOTION PICTURES

American Standard

Screen Brightness
For 35-Millimeter Motion Pictures

Ret. U. S. 1'at. Of.

Z22.39-1944

I. Screen Brightness

I . I The brightness a I" the center of a screen for viewing 35-mm
motion pictures shall be 101? foot-lamberts when the pro-
jector is running with no film in the gate.

NOTE: Reprinted August 1946, without chdnge.

60th SEMIANNUAL CONVENTION

HOLLYWOOD-ROOSEVELT HOTEL
Hollywood, California

OCTOBER 21-25, 1946

Officers in Charge

D. E. HYNDMAN President

HERBERT GRIFFIN Past-President

L. L. RYDER Executive Vice-P resident

M. R. BOYER Financial Vice-P resident

J. A. MAURER Engineering Vice-President

A. C. DOWNES Editorial Vice-President

W. C. KUNZMANN Convention Vice-President

C. R. KEITH Secretary

E. I. SPONABLE Treasurer

General Office, New York

BOYCE NEMEC Engineering Secretary

HARRY SMITH, JR Executive Secretary

Directory of Committee Chairmen

Pacific Coast Section and Local Ar-
rangements H. W. MOYSE, Chairman

Papers Committee C. R. DAILY, Chairman

BARTON KREUZER, Vice-
Chairman

Publicity Committee HAROLD DESFOR, Chair-
man

Registration and Information W. C. KUNZMANN, Chair-
man, assisted by C. W.
HANDLEY

Luncheon and Dinner-Dance Commit-
tee L. L. RYDER, Chairman

Hotel and Transportation Committee S. P. SOLOW, Chairman

265

266 SMPE CONVENTION Vol 47, No. 3

Membership and Subscription Commit-
tee H. W. REMERSCHEID, Chairman

Ladies Reception Committee Hostess MRS. H. W. MOYSE

Projection Program— 35-mm W. V.WOLFE, Chairman, assisted

by Members Los Angeles
Locals 150 and 165
16-mm H. W. REMERSCHEID

HOTEL RESERVATIONS AND RATES

The Hollywood-Roosevelt Hotel, Hollywood, Calif., will be the Convention
Headquarters, and the hotel management extends the following per diem room
rates, European plan, to SMPE members and guests:

Room with bath, one person $4.40-5.50

Room with bath, two persons, double bed $5.50-6.60

Room with bath, two persons, twin beds $6. 60-7. 70 (

Your Convention Vice-President has arranged with the management of the
Hotel Sir Francis Drake, San Francisco, Calif., to provide accommodations for
members who will visit this city while on the West Coast. Accordingly, reserva-
tions should be made direct with George T. Thompson, Managing Director, at
least two weeks in advance of expected arrival in San Francisco. When making
reservations, advise Mr. Thompson that you are a member of the SMPE.

REGISTRATION

expenses.

BUSINESS AND TECHNICAL SESSIONS

Day sessions will be held in the hotel, and evening sessions at locations away
from the hotel, as given below.

GET-TOGETHER LUNCHEON AND DINNER-DANCE

The Society will again hold its regular pre-war social functions and accordingly
a Get-Together Luncheon is scheduled in the California Room of the hotel on
Monday, October 21, at 12:30 P.M. The guest speaker will be Byron Price.
Members in Hollywood and vicinity will be solicited by a letter from S. P. Solow,
Secretary of the Pacific Coast Section, to send remittances to him for the Con-
vention registration fee and luncheon tickets. Ladies are welcome to attend the
luncheon.

The 60th Semiannual Dinner-Dance will be held in the California Room of the
hotel on Wednesday evening, October 23, at 8:30 P.M. Dancing and entertain-
ment. (Dress optional.) A social hour for holders of Dinner-Dance tickets will
precede the Dinner-Dance between 7: 15 P.M. and 8:15 P.M. in the Hotel Terrace
Room (Refreshments).

Sept. 1946 SMPE CONVENTION 267

LADIES' PROGRAM

A reception parlor for the ladies' daily get-together and open house with Mrs.
H. W. Moyse as hostess will be announced on the hotel bulletin board and in the
final printed program.

The ladies' entertainment program will be announced later.

MOTION PICTURES AND RECREATION

Egyptian Theatre

Grauman's Chinese Theatre

Hollywood Pantages Theatre

Hollywood Paramount Theatre

Warner's Hollywood Theatre

A Tentative Program and abstracts of papers were recently mailed to the gen-
eral membership of the Society in the United States. The complete program as
followed during the Convention will be published in the November JOURNAL.
However, for those who failed to receive the Tentative Program, the technical
sessions scheduled, location, and time are given here to facilitate making plans.

Monday, October 21, 1946

Open Morning.
10:00 a.m. Room 201, Hotel Mezzanine Floor: Registration. Advance sale of

Luncheon and Dinner-Dance tickets.

12:30 p.m. California Room: SMPE Get-Together Luncheon.
2: 00 p.m. Aviation Room, Hotel Mezzanine Fhor- Opening business and

Technical Session
8:00 p.m. Evening Session: Republic Studios Scoring Stage, Hollywood.

Tuesday, October 22, 1946

Open Morning.
10: 00 a.m. Room 201, Hotel Mezzanine Floor: Registration. Advance sale of

Dinner-Dance tickets.

2: 00 p.m. California Room: Afternoon Session.
8:00 p.m. Evening Session: Paramount Studios, Hollywood.

268 SMPE CONVENTION Vol 47, No. 3

Wednesday, October 23, 1946

9: 30 a.m. Room 201, Hotel Mezzanine Floor: Registration. Advance sale of

Dinner-Dance tickets.
10:00 a.m. California Room: Morning Session.

Open Afternoon.
7: 15 p.m. Hotel Terrace Room: A social hour for holders of Dinner- Dance

tickets preceding the Dinner-Dance (Refreshments) .

8: 30 p.m. California Room: 60th Semiannual Convention Dinner-Dance.
Dancing and entertainment.

Thursday, October 24, 1946

Open Morning.

1:00 p.m. Room 201, Hotel Mezzanine Floor: Registration.
2: 00 p.m. California Room: Afternoon Session.
8:00 p.m. Evening Session: Walt Disney Theater, Disney Studios, Burbank.

Friday, October 25, 1946

Open Morning.

2: 00 p.m. California Room: Afternoon Session.
8:00 p.m. Evening Session: Marquis Theater, Hollywood.

Note: All sessions during the 5-day Convention will open with an interesting
motion picture short.

Important

Because of the existing food problem, your Luncheon and Dinner-Dance
Committee must know in advance the number of persons attending these func-
tions in order to provide adequate accommodations.

Your cooperation in this regard is earnestly solicited. Luncheon and Dinner-
Dance tickets can be procured from W. C. Kunzmann, Convention Vice- President,
during the week of October 13 at the Hollywood-Roosevelt Hotel.

All checks or money orders for Convention registration fee, Luncheon and
Dinner-Dance tickets should be made payable to W. C. Kunzmann, Convention
Vice-President, and not to the Society.

W. C. KUNZMANN
Convention Vice-President

SOCIETY ANNOUNCEMENTS

AMENDMENT TO BY-LAWS

During the past year the Board of Governors has received several requests to
establish Student Chapters of the Society in educational institutions of recognized
standing. Since no provision exists in the Constitution and By-Laws which
would permit establishment of Student Chapters, the Board of Governors, at a
meeting held July 10, 1946, considered and submits the following new proposed

Sept. 1946 SOCIETY ANNOUNCEMENTS 269

By-Law for consideration by the membership. The proposal is published here-
with in accordance with By-Law XII, Section 1, and will be voted on by qualified
members at the 60th Semiannual Convention, Hollywood-Roosevelt Hotel,
Hollywood, Calif., October 21-25.

Proposed By-Law XIII

Student Chapters

"Sec. 1. — Student Chapters of the Society may be authorized in any college,
university, or technical institute of collegiate standing.

"Upon written petition, signed by 12 or more Society members, or applicants
for Society membership, and the Faculty Adviser, for the authorization of a
Student Chapter, the Board of Governors may grant such authorization.

Chapter Membership

"Sec. 2. — All members of the Society of Motion Picture Engineers in good
standing who are attending the designated educational institution shall be eligible
for membership in the Student Chapter, and when so enrolled they shall be
entitled to all privileges that such Student Chapter may, under the General
Society's Constitution and By-Laws, provide.

"Sec. 3. — Should the membership of the Student Chapter fall below ten, or
should the technical quality of the presented papers fall below an acceptable
level, or the average attendance at meetings not warrant the expense of maintain-
ing the organization, the Board of Governors may cancel its authorization.

Chapter Officers

"Sec. 4. — The officers of each Student Chapter shall be a Chairman and a
Secretary-Treasurer. Each Chapter officer shall hold office for one year, or until
his successor is chosen. Officers shall be chosen in May to take office at the be-
ginning of the following school year. The procedure for holding elections shall be
prescribed in Administrative Practices.

Faculty A dviser

"Sec. 5. — A member of the faculty of the same educational institution shall be
designated by the Board of Governors as Faculty Adviser. It shall be his duty to
advise the officers on the conduct of the Chapter and to approve all reports to the
Secretary and the Treasurer of the Society.

Chapter Expenses

"Sec. 6. — The Treasurer of the General Society may deposit with each Chapter
Secretary-Treasurer a sum of money, the amount to be fixed by the Board of
Governors. The Secretary-Treasurer shall send to the Treasurer of the General
Society at the end of each school year, an itemized account of all expenditures
incurred during that period.

Chapter Meetings

"Sec. 7. — The Chapter shall hold at least four meetings per year. The Secre-
tary-Treasurer shall forward to the Secretary of the General Society at the end of
each school year a report of the meetings for that year, giving the subject, speaker,
and approximate attendance for each meeting."

270 SOCIETY ANNOUNCEMENTS Vol 47, No. 3

EMPLOYMENT SERVICE

POSITIONS OPEN

The Communicable Disease Center of the U. S. Public Health Service is en-
gaged in the production and distribution of motion pictures, lantern slides, film
strips, and other audio- visual aids to be used as media of information and instruc-
tion for colleges, universities, public health agencies, and other groups. The
above agency is currently recruiting for a number of positions open to experi-
enced technicians at salaries ranging from $3397.20 to $8179.50 per annum. Be-
cause of lack of space, details of only six of the 16 positions available are given in
this issue of the JOURNAL; others will be listed next month. Applicants should
address inquiries to Personnel Officer, U. S. Public Health Service, 605 Volunteer
Building, Atlanta 3, Georgia.

(1) PROJECT SUPERVISOR, $4902 per annum, requiring a thor-
ough knowledge of production of technical training films. Applicants
must supervise and advise employees in production analysis, choice of
script, plotting sequence, and other related duties in film production.

(2) CHIEF, DEVELOPMENT BRANCH, $5905.20 per annum, re-
quiring responsibility for the development of training films and col-
lateral aids in field of communicable disease. Must incorporate fac-
tual data, outline background and objectives to subordinate employees,
review completed plans, and make final evaluation of production for
policy, context, and effectiveness.

(3) CHIEF, UTILIZATION SECTION, $5905.20 per annum. Re-
quires a thorough knowledge of the utilization and evaluation of audio-
visual aids as applied to the dissemination of information. Must be
able to recommend changes in substance of films, maintain liaison
with various agencies, both domestic and foreign, determine appro-
priate film for given group, maintain continuous analysis of the pro-
duction program, and formulate distribution program.

(4) CHIEF, PRODUCTION BRANCH, $4902 per annum, re-
quiring a thorough knowledge of the production of training films, in-
cluding motion picture production management, laboratory techniques,
sound recording, film editing, animation, slide series, and other closely
related operations. Must be able to give executive direction to Chief
of Motion Picture Section, Chief of Film Strip Section, Chief of Gra-
phics, and other necessary personnel.

(5) CHIEF, FILM STRIP SECTION, $4149.60 per annum. Appli-
cants must have a thorough knowledge of photography, both still and
motion picture, color photography, photomicrography, sound recording
and music effect, recorded narration and dialogue, and other related
fields. Work is reviewed for propriety, effectiveness, and conform-
ance with project objective and general training film policy.

(6) CHIEF, PRODUCTION DIVISION, $7102.20 per annum. Re-
sponsible for the execution of the entire production and distribution
program in the field of audio-visual training as applied to communi-
cable disease control. Will be responsible for the improvement of the
present program, liaison with organizations or institutions concerned
with communicable disease control, evaluation and utilization analysis
of training methods, direct a staff of 50 to 60 people.

Position available for Optical Designer, capable of handling the calcula-
tion and correction of aberrations in photographic and projection lens
systems. Junior designers or engineers will be considered. Write
fully giving education, experience, and other qualifications to Director
of Personnel, Bell and Howell Company, 7100 McCormick Road, Chi-
cago 45, m.

line, SOCIETY ANNOUNCEMENTS 271

operation of recording equipment; motion picture processing labora-
tory supervisor; and professional make-up artist. Five-year contracts
at favorable terms are offered to those qualified. Write or cable direct
to Personnel Manager, Dawlat Corporation Ltd., Patel Chambers, French
Bridge, Bombay 7, India, giving experience, etc., in detail.

POSITIONS WANTED

Projectionist-newsreel editor with 15 years' experience just released
from service. Willing to locate anywhere. Write P. O. Box 152, Hamp-
den Station, Baltimore 11, Maryland.

Cameraman, honorably discharged Army veteran, desires re-enter indus-
trial, educational production with independent producer or studio. Ex-
perienced in 35- and 16-mm color and black-and-white. References
and complete record of experience available. Write, wire or telephone
T. J. Maloney, 406 Oak St., Ishpeming, Mich. Telephone 930.

We are grieved to announce the deaths of John E. McAuley, Fellow
of the Society, on August 22, 1946, in Chicago, III., and F. C. Coates,
Active member of the Society, on September 7, 1946, in Los Angeles,
Calif.

SOCIETY of MOTION PICTURE ENGINEERS

MOTCL PENNSYLVANIA • N«W YORKt, N*Y» • TtL. PCNN. 6 O62O

APPLICATION FOR MEMBERSHIP

Name Age

Address '.

City

Employer.

Occupation.

Grade Desired: Associate Q Active
Education *

Record of Employment* (list companies, years, and positions held).

Other Activities*.

REFERENCES**

(Name) (Address) (City)

Date 19_ (Sgd)_

* If necessary, use additional sheet to give complete record.

** References should be members of Society. If not, supply two letters of reference from individuals
acquainted with applicant' e work.

JOURNAL OF THE SOCIETY OF
MOTION PICTURE ENGINEERS

Vol 47 OCTOBER 1946 No. 4

CONTENTS

PAGE

A Unified Approach to the Performance of Photographic
Film, Television Pickup Tubes, and the Human Eye

A. ROSE 273
The High Cost of Poor Projection C. E. LEWIS 295

Factors Governing the Frequency Response of a Vari-
able-Area Film Recording Channel

M. RETTINGER AND K. SINGER 299

Wide-Range Loudspeaker Developments

H. F. OLSON AND J. PRESTON 327

Current Literature 353

Indexes to the semiannual volumes of the JOURNAL are published in the June and December
issues. The contents are also indexed in the Industrial Arts Index available in public libraries.

JOURNAL

OF THE

SOCItTY of MOTION PICTUPiE ENGINEERS

MOT«L PENNSYLVANIA • NEW YORKl, N-Y- • T«L. PCNN. 6 O62O

HARRY SMITH, JR., EDITOR

Board of Editors
ARTHUR C. DOWNES, Chairman

JOHN I. CRABTREE ALFRED N. GOLDSMITH EDWARD W. KELLOGG

CLYDE R. KEITH ALAN M. GUNDELPINGER CHARLES W. HANDLEY

ARTHUR C. HARDY

Officers of the Society
^President: DONALD E. HYNDMAN,

342 Madison Ave., New York 17.
*Past-P resident: HERBERT GRIFFIN,

133 E. Santa Anita Ave., Burbank, Calif.
^Executive Vice-President: LOREN L. RYDER,

5451 Marathon St., Hollywood 38.
** Engineering Vice-President: JOHN A. MAURER,

37-01 31st St., Long Island City 1, N. Y.
* Editorial Vice-President: ARTHUR C. DOWNES,

Box 6087, Cleveland 1, Ohio.
** Financial Vice-President: M. R. BOYER,

350 Fifth Ave., New York 1.
* Convention Vice-President: WILLIAM C. KUNZMANN,

Box 6087, Cleveland 1, Ohio.
^Secretary: CLYDE R. KEITH,

233 Broadway, New York 7.
^Treasurer: EARL I. SPONABLE,

460 West 54th St., New York 19.

Governors

"tFRANK E. CAHILL, JR., 321 West 44th St., New York 18.
**FRANK E. CARLSON, Nela Park, Cleveland 12, Ohio.
**ALAN W. COOK, Binghamton, N. Y.

*JOHN I. CRABTREE, Kodak Park, Rochester 4, N. Y.

*CHARLES R. DAILY, 5451 Marathon St., Hollywood 38.
**JOHN G. FRAYNE, 6601 Romaine St., Hollywood 38.
**PAUL J. LARSEN, 1401 Sheridan St., Washington 11, D. C.
**WESLEY C. MILLER, Culver City, Calif.

*PETER MOLE, 941 N. Sycamore Ave., Hollywood.
*JHoLLis W. MOYSE, 6656 Santa Monica Blvd., Hollywood.

*WILLIAM A. MUELLER, 4000 W. Olive Ave., Burbank, Calif.
*°A. SHAPIRO, 2835 N. Western Ave., Chicago 18, 111.

*REEVE O. STROCK, 111 Eighth Ave., New York 11.

*Term expires December 31, 1946. tChairman, Atlantic Coast Section.
**Term expires December 31, 1947. jChairman, Pacific Coast Section.
*° Chairman, Midwest Section.

Publication Office, 20th & Northampton Sts., Easton, Pa.

General and Editorial Office, Hotel Pennsylvania, New York 1, N. Y.

Entered as second-class matter January 15, 1930, at the Post Office at Easton.

Pa., under the Act of March 3. 1879.

JOURNAL OF THE SOCIETY OF
MOTION PICTURE ENGINEERS

Vol 47 OCTOBER 1946 No. 4

A UNIFIED APPROACH TO THE PERFORMANCE OF PHOTO-
GRAPHIC FILM, TELEVISION PICKUP TUBES,
AND THE HUMAN EYE*

ALBERT ROSE**

Summary. — The picture pickup devices — film, television pickup tube, and eye —
are subject ultimately to the same limitations in performance imposed by the discrete
nature of light flux. The literature built up around each of these devices does not re-
flect a similar unity of terminology. The present paper is exploratory and attempts
a unified treatment of the three devices in terms of an ideal device. The performance
of the ideal device is governed by the relation

(signal-to-noise ratio)2

scene brightness = constant —

picture element area X quantum efficiency

The three devices are shown to approximate this type of performance sufficiently
well to use it as a guide in treating their common problems. Simple criteria are de-
rived for characterizing the performance of any one device as well as for comparing
the performance of different devices. • For example, quantum efficiency is used to meas-
ure sensitivity; the signal-to-noise ratio, associated with a standard element area, is
used to measure both resolution and half-tone discrimination. The half-tone discrimi-
nation of the eye governs the visibility of "noise" in the reproduced picture and, in
particular, requires that pictures be photographed or picked up at increased scene
brightness when the brightness of the reproduction is increased. The observation and
interpretation of visual "noise" are discussed.

Introduction. — There are three picture pickup devices that have
separately been the subject of considerable investigation. These
are the human eye, motion picture film, and television pickup tubes.
For each of these, a large technical literature has been built up rela-
tively independently of the others. The language, the units, the con-
cepts, and the conclusions of the separate arts are not in a form that
allows them to be readily compared. This situation is understand-

* Presented May 10, 1946, at the Technical Conference in New York.
** RCA Laboratories, Princeton, N. J.

273

274 A. ROSE Vol 47, No. 4

able in the early stages of the arts because the primary emphasis is
then to get something — anything — that will transmit a usable picture.
As the art progresses, however, interest shifts naturally to an exami-
nation of the theoretical limits of expected improvements. Such an
examination is especially significant because all three devices are sub-
ject ultimately to the same simple statistical limitations arising from
the discrete nature of light flux. The time is opportune for the three
devices to profit from a consideration of their problems in common
terms.

Some illustrations will make the present situation clear. In films,
graininess is a familiar concept. Its origin, control, and visual effects
have been treated extensively and for a long time. In pickup tubes,
signal-to-noise ratio is an ever-present consideration for getting pic-
tures of good quality. For human vision, interest has frequently been
centered on the minimum discernible contrast. There is good reason
now to say that graininess, signal-to-noise ratio, and minimum dis-
cernible contrast are only three different names for the same property
of a picture pickup device. Again : the limiting resolution of film is a
standard and advertised characteristic ; the frequency response curve
of a television pickup tube is an important specification of the tube's
performance; the minimum resolvable angle of the eye is a well-
known figure and one which, perhaps, has received more than its just
share of attention. It is obvious that in all three instances, an at-
tempt has been made to count the number of separate picture ele-
ments.

A third illustration concerns sensitivity. v There is little need to
remind one of the variety and confusion of sensitivity scales that have
been proposed for film. On the other hand, the sensitivity of a tele-
vision pickup tube can, with reasonable adequacy, be defined by its
microampere signal output per lumen input. The sensitivity of the
eye has variously, and often with deliberate dramatic emphasis, been
described in terms of the farthest distance at which one can still see a
lighted candle; "the order of magnitude of the faintest visible star;
the number of lumens falling on the retina necessary for a visual
sensation; and so on. Only recently have there been more funda-
mental attempts to measure the sensitivity of the eye in terms of its
quantum efficiency.

These illustrations serve to show, first, that the basic properties of a
picture pickup device — resolution, sensitivity, and contrast discrimi-
nation— are indeed of common concern to the eye, film, and pickup

Oct. 1946

FILM, TELEVISION AND THE EYE

275

tube; and, second, that the specification of these properties has not
enjoyed an appropriately common treatment.

The purpose of the present discussion is to explore the extent to
which such a common or unified treatment is both possible and profit-
able.

The order of the discussion will be :

(1) The development of the properties of an ideal picture pickup device;

(2) The examination of eye, film, and pickup tube for the purpose of finding
out how well they approximate ideal performance;

(3) A re-examination of a number of current problems in the light of (1) and

(2).

It will become clear that the performance of an ideal device is com-
pletely specified by a single number, the quantum efficiency of its
photo process, taken together with some simple optical relations;

SCENE

LENS

PHOTOSENSITIVE
TARGET

FIG. 1. Essential parts of a picture pickup system.

REPRODUCED
PICTURE

that the performance of eye, film, and some pickup tubes approach
sufficiently close to ideal performance to suggest a unified approach to
many of their current problems and that such an approach leads to
simplifying concepts.

Ideal Picture Pick-Up Device. — Fig. 1 shows the essential parts
of a system for picking up and reproducing a picture. Attention
will be centered on the target of the pickup device, and, in particu-
lar, on one picture element of that target. A picture element is
here taken to be an element of area of arbitrary size, not necessarily
the smallest resolvable area. Let that element have a length of side
h, and absorb an average number N, of quanta in the exposure time
allowed. The absorption of each quantum will give rise to a separate
event such as the release of an external photoelectron, or an internal
photoelectron or the dissociation of a molecule. These are uncorre-
lated chance events. For this reason, the average number N has
associated with it fluctuations whose root mean square magnitude is

276 A. ROSE Vol 47, No. 4

the square root of the average number. Thus, if N is taken to be the
measure of the signal, TV1/2 is a measure of the smallest discernible dif-
ference in signal. In particular, the ratio

is the signal-to-noise ratio. We may write, therefore,

Signal-to-noise ratio = R = N1/2 (1)

and the geometric relation :

N
Scene brightness = B = constant p. (2)

Combination of Eqs (7) and (2) yields:

5 = Constant^. (3)

Eq (5) is the characteristic equation for the performance of the ideal
picture pickup device. It must be emphasized that Eq (3) is not
concerned with the particular mechanism used to generate a picture
so long as full use is made of all the absorbed quanta. For this rea-
son, it is meaningful to inquire whether the performance of such diverse
mechanisms as the eye, film, and pickup tubes can all be described by
the same characteristic equation.

Eq (3) defines the scene brightness B required to transmit a picture
having a signal-to-noise ratio R associated with picture elements of lin-
ear size h. It says that the scene brightness must be increased as the
square of the signal-to-noise ratio demanded, and as the square of the
number of lines in the picture, the number of lines being proportional
to l/h.

The constant term on the right-hand side of Eq (3) contains,
among other parameters, the quantum efficiency of the photo process.
It is this quantum efficiency* alone which sets the performance range
of the ideal pickup device. The complete constant term will be given
later. For the moment, it will be useful to examine a plot of Eq (3) .

* If the term "ideal pickup device" were to receive its full emphasis, the quan-
tum efficiency of the photo process should, of course, be taken to be 100 per cent.
The emphasis here, however, is on the complete utilization of all absorbed quanta
rather than on the absorption of all incident quanta.

Oct. 1946

FILM, TELEVISION AND THE EYE

277

Fig. 2 is a plot of Eq (3) for several values of scene brightness.
Fig. 2 shows that the signal-to-noise ratio increases linearly with the
size of picture element considered. In particular, there is a smallest
element which is determined by the smallest signal-to-noise ratio
that can be observed. The smallest element would be called the
limiting resolution. The smallest observable signal-to-noise ratio
has often been taken to be unity. Actually, by virtue of its statistical
origin, the smallest observable R is a function of how often one pre-
fers to have his observations correct. This much is verifiable both
from analysis and from the use of physical instruments as observers.

LINEAR ELEMENT SIZE-

FIG. 2. Performance curves for an ideal picture pickup
device.

For a human observer, tests1 have been made which suggest a
threshold value of R in the neighborhood of five. Whatever this
threshold is, one may draw on Fig. 2 a horizontal line whose inter-
sections with Bi, B2, and Bs mark the limiting resolutions for the
several scene brightnesses.

The complete form of Eq (3) may be readily obtained2 from well-
known optical relations and is

Here / = the// value (numerical aperture) of the lens
t = exposure time (seconds)

278

A. ROSE

Vol 47, No. 4

6 = quantum yield of the photo process (6 = 1 means 100 per cent quan-
tum efficiency)

h = length of side of element (cm) ,
1 lumen = 1.3 X 1016 quanta per sec (average for white light) .

If one takes the hyperfocal distance as a measure of depth of field, the
performance of the pickup device is completely specified by Eq (4)
together with the relation.3

2h (5)

hyperfocal distance = -^-r-

where F = focal length of lens
D = diameter of lens.

Complete specification means that one selects the desired values for
the hyperfocal distance, exposure time, signal-to-noise ratio, angle of
view, and size and number of picture elements, and from them com-
putes the scene brightness required.

CATHODE (ZERO)

DECELERATING RING

(ZERO)

/ — SECONDARY
ELECTRONS

ELECTRON IMAGE

SECONDARY |
ELECTRONS, ,

1LECTRODE \
(tISOOV.) \

ALIGNMENT COIL

PHOTO-CATHODE
-600V.)

TARGET SCREEN

(ZERO)
TWO-SIDED TARGET

IMAGE ORTHICON
FIG. 3. Image orthicon (schematic).

The scale factors for the curves of Fig. 2 are based on Eq (4) with
the choice of/ = 2, t = 1/30, 6=1. These curves show what may be
expected from an ideal device with 100 per cent quantum efficiency.

Television Pickup Tubes. — No operable pickup tube has yet
been reported which completely fulfills the properties of the ideal
pickup device. The effective exposure time of the image dissector,4
or other nonstorage devices, is limited to a picture element time and
such devices are correspondingly insensitive. The performance of the
iconoscope5 and orthicon6 is limited by noise currents in the television
amplifier rather than by the smaller noise currents inherent in the

Oct. 1946 FILM, TELEVISION AND THE EYE 279

primary photo process. The image orthicon7 (Fig. 3) goes a long way
toward removing this limitation in so far as the high light signal-to-
noise ratio of its output is, within limits, determined by the signal-to-
noise ratio in the primary photo process. It is handicapped, as are
the other storage-type tubes, mainly by having as much noise in the
low light portions of a picture as in the high lights. Eq (4) may, how-
ever, be used to describe the performance of the image orthicon if
signal-to-noise ratio is interpreted to mean the signal-to-noise ratio
in the high light portions of the picture.* The quantum yield of the
primary photo process is about 0.01 and the noiseless amplifier to be
compared with Fig. 1 is its electron multiplier.

Photographic Film. — One does rot readily think of film as having
a signal-to-noise ratio. Yet, the separate grains randomly situated
in film are immediately comparable with the separate and randomly
spaced electrons in the scanning beam of a television pickup tube.
And, in fact, a number of recent objective measurements as well as
analyses of graininess have led to the expression8

AD == ^ = constant X a~1/2 (6)

where AD and AT" are the rms deviations in density and transmission,
respectively, of an area a of film.** With the notation of Eqs (1)

A T"1

and (2), - - = R~l and a~l /2 = h~l and one may write for film

R = constant X h. (7)

The value of this constant is proportional to the reciprocal grain di-
ameter. There is good evidence that, for the same type of photo-
graphic grain, the film speed is proportional to the grain area. The
last two statements combined with Eq (7) give

R2
B = constant

* The beam current used to scan the target must be sufficient to discharge the
high light portions of the picture. Under these conditions, the signal-to-noise
ratio inherent in the beam is approximately that of the high lights. The same
beam current, however, scans the low lights and adds appreciable noise over and
above the noise inherent in the low lights.

** Eq (6) obviously cannot hold for values of a in the neighborhood of and less
than the grain size. Krevald and Scheffer9 and Raudenbusch10 have observed
such departures and more recently Jones and Higgins14 have reported them. The
problem is further involved by a range of grain sizes in any one film.

280 A. ROSE Vol 47, No. 4

just as for the ideal device (Eq 3) . One can accordingly use Eq (4) to
describe the performance of film with the understanding that the ratio
R2/h2 is characteristic of film with a given average grain diameter
and changes in R2/h2 are obtained by use of other films with different
average grain diameters. The quantum yield is the reciprocal of the
number of incident quanta required to make a grain developable*
and from published statements11 is in the neighborhood of 0.001.
The noiseless amplifier to be compared with Fig. 1 is the complete
development of a silver grain after only a few silver atoms have been
formed by the action of the light.

Human Eye. — Eq (4) is not immediately applicable to the
human eye because there is no way of directly measuring the signal-
to-noise ratio that the brain perceives. It is necessary, therefore,
to replace signal-to-noise ratio by its equivalent in terms of minimum
discernible contrast in the test object viewed.** The signal-to-noise
ratio R has already been referred to as a measure of the minimum dis-
cernible difference in signal. This allows one to write with reasonable
assurance

minimum discernible contrast = C = — - X 100 per cent. (£)

J\.

To get a value for the constant, let C take on its maximum value,
viz., 100 per cent. This defines the constant to be equal to the mini-
mum perceivable value of R. As mentioned earlier, the measure-
ments of Romer and Selwyn may be interpreted to give a value of
about five. Unpublished measurements by O. Schade on television
pictures yield a value of three. The determination of this constant is
of considerable importance in estimating the quantum efficiency of
the human eye and deserves more experimental work.j For the

* Strictly, this use of the term "quantum yield" is in accord with its normal
definition only if a grain is made developable by the absorption of a single quan-
tum. If more than one quantum needs to be absorbed for this purpose, the proc-
ess still may be looked upon for noise computations as rlie equivalent of the ab-
sorption of one quantum because the noise arises mostly from the random dis-
tribution of grains rather than from the fluctuations in rate of absorption of light
quanta.

T? _ 7?

** Contrast is denned as — X 100 per cent, where BD is the bright-

BL
ness of a gray test object immersed in a white surrounding of brightness BL.

f An interpretation of the experimental results of Jones and Higgins14 in which
the blending distances and signal-to-noise ratios were measured for the same films
also leads to a value of about five.

Oct. 1946

FILM, TELEVISION AND THE EYE

281

present it will be included as an undetermined constant. Substitution
of Eq (8) in Eq (3) gives

constant

(9)

for the characteristic equation which the eye would satisfy if its per-
formance were "ideal." Eq (9) may be rewritten with the minimum

/SCENE MCT4To 10*
PERFORMANCE DflTA /LUMINANCE) Foor LAMBERTS

<% CONTRAST Zro loo
FOE EYE IN RANGES MINIMUM \

2* TO 40*

100

h
•/)
ct

j* 30

O
O

5CENE LuMltMANCC-FboT LAMBERTS

-4

IO~

ID'

i.o 10 100

CONNOR e GHNOUNG
COBB £ Moss

o.oi

O.O3

RECIPROCAL OF MINIMUM RESOLVABLE ANGLE

FIG. 4. Comparison of experimentally observed performance of
the eye with ideal performance.

resolvable angle a in place of distance h to make it more readily com-
parable with published data. Thus,

B = constant

(10)

How well the performance of the eye matches Eq (10) may be seen
from Figs. 4 and 5. Fig. 4 shows a plot12 of C versus a~l for a large
range of scene brightnesses and, as expected from Eq (10), the data fall
closely on 45-deg lines. Data in the immediate neighborhood of a =
1 minute and C = 2 per cent are omitted because these represent

282

. ROSE

Vol 47, No. 4

limits to the performance of the eye set by other than statistical
considerations. The smallest angle that the eye can resolve at high
lights, for example, is set by the physical size of the retinal elements
or cone structure. A more precise treatment would include, and be
slightly modified by, the shape of the eye curve near its "cutoff" limits.

a 30

z
o
o

^ 10

/SCENE: IO-TO 1O

PETRFOCMflNCE D/9TA / LUMINANCC J FOOT
FOR EYE IN RANGES | % CONTRAST 2 TO 100

/ MINIMUM "*j

• •• CONNOR

OOO COBB £ Moss

0.03 O.I

3.0

FIG. 5. Replot of data in Fig. 4.

The complete characteristic equation for the eye, from Eqs (10) and
f),is

B = 1.4 J?SL X 10-2 ft L (11)

where a is the angle in minutes of arc subtended by a picture element
at the eye and k is the undetermined constant relating C and R.
Fig. 5 is a replot of the data in Fig. 4. It is a more complete test of
the characteristic Eq (11) and shows the small range* of the factor

* If the full range of this factor is ascribed to the variation of quantum efficiency
from low to high lights, one is presented in this approach with at most a ten-to-one
variation in sensitivity of the eye from low to high lights as opposed to the usual
statement that the dark adapted eye is 103 to 104 times as sensitive as the light
adapted eye.

Oct. 1946 FILM, TELEVISION AND THE EYE 283

&2//0 from very low to very high lights as well as its actual value.
At low lights the value of kz/te is 2800. If pne takes the exposure
time / to be 0.2 sec, k-/B = 560. It is known13 that at threshold
about 150 quanta (near 5300 A) must be incident on the eye to
generate a sensation. This corresponds to about 500 quanta if white
light is used. Various measurements and computations13 of the
number of quanta actually used in generating the sensation vary
from one to 50, giving 6 the range from 0.002 to 0.1 and k the range
from one to 7. This range of k is to be compared with the independ-
ently obtained values of five from Romer and three from Schade.

All of the above discussion has been for the purpose of showing that
the performance of the eye satisfies the same type of equation as that
obtained for the ideal pickup device. The quantum efficiency, as-
suming k = 5, is about 5 per cent at low lights and about 0.5 per cent
at high lights. The noiseless amplifier to be compared with Fig. 1
may be some catalytic or triggering action induced by the absorption
of quanta in the retina.

General Discussion. — The classes of picture pickup problems
that have received frequent attention are :

(1 ) Specification of the performance of any one pickup device ;

(2) Comparisons of the performance of two pickup devices of the same kind,
or of different kinds ;'

(3) The setting of standards of performance for pickup devices that would
"satisfy" the human eye.

The particular problems to be discussed here are intended only to be
representative, rather than exhaustive.

Sensitivity. — The simplest test for the relative "sensitivities" of
two devices is accomplished by observing the lowest scene bright-
nesses at which they can still record a picture. This type of test is
immediately subject to the questions: Was the exposure time the
same for the two devices? What were the relative lens speeds used?
What were the relative picture sizes? While these are obvious ques-
tions, there is no essential reason to pause here. The further ques-
tions of relative angles of view, numbers of picture elements and
signal-to-noise ratios are of equal importance. In brief, the com-
parison of the sensitivities of two devices is not meaningful until the
devices and their transmitted pictures are completely specified. But
complete specification, as pointed out earlier, means that the quan-
tum efficiency of the primary photo process is the only parameter
that can vary the range of performance of an ideal device. And

284 A. ROSE Vol 47, No. 4

accordingly, the quantum efficiency is the measure of sensitivity.
Not all devices, however, are ideal. For this reason, a more general
figure of merit, based on Eq (4) is here proposed. The figure of merit
is proportional to the reciprocal of the total light flux required to pro-
duce a picture of specified signal-to-noise ratio and resolution in a
specified exposure time. The figure of merit is

where / is the numerical aperture of the lens, B the scene brightness,
and A the area of target. If the performance of the device is ideal,
the figure of merit becomes also a measure of its quantum efficiency.

It is recognized that the signal-to-noise ratio of a given picture is
not a readily accessible parameter and that there is no general agree-
ment on a measure of resolution. The evaluation of sensitivity, how-
ever, can be no more accurate than the knowledge of these parameters.
It is of interest to apply the figure of merit to the interpretation of
several familiar problems.

Film Speeds. — Consider the range of film speeds advertised.
For the most part, these are films of the same quantum efficiency
but different grain size* and, for the most part, the essential sensitivity
performance of these films is the same. A simple example will make
this clear. Two films, A and B, are rated at the relative speeds of
one and four, respectively. Their quantum efficiencies are equal
and the average grain area of B is four times that of A . Normally,
one might say that B can pick up a scene with one-fourth the light
required by A . While this statement is true, it is misleading. Sup-
pose one wants the same resolution and depth of focus in both pic-
tures. This would mean a film area of B four times as large as A to
match resolutions and consequently a lens for B stopped to twice the
numerical aperture (//number) of the lens for A to match depth of
focus. The result is that both films require the same scene brightness
to transmit the same picture — a result which could have been antici-
pated from their equal quantum efficiencies or from their figures of
merit evaluated for the same transmitted picture quality .

Comparison of Eye and Film.- — An interesting application of the
figure of merit is to the taking and viewing of motion pictures.

* The relative speeds of Super XX and Eastman High Resolution plates are
in the ra*tio of about 104 to 1. The relative grain areas are in the ratio of about
103to 1.

Oct. 1946

FILM, TELEVISION AND THE EYE

285

For obvious reasons, the quality of the motion picture (signal-to-
noise ratio and resolution) is aimed at equaling or exceeding the
quality of picture which the eye can transmit at the brightness of the
motion picture screen. For equal quality one can anticipate that the
figure of merit for the eye would be at least a factor of twenty better
than for film based on relative quantum efficiencies. But, in so far
as film aims at better quality and attempts to compensate for some
of its limitations by projecting pictures at a higher than unity gamma,
an additional factor can be expected in favor of the eye.

Table 1 gives approximate values for /, B, and A to be associated
with the camera that takes the pictures and the eye that views them.

KINESCOPE

PICTURE
TRANSMITTED
BY IMAGE
ORTHICON

ORIGINAL
SUBJECT

35MM CAMERA

WITH / /

SUPER XX FILM / / LIGHT

AND f/z LENS / / SOURCE

IMAGE ORTHICON

CAMERA
WITH % LENS

FIG. 6.

Setup for comparison of low light performance of
Super XX film and an image orthicon.

The area of target used for the eye is that area of retina occupied by
the motion picture at a 4 : 1 viewing distance. The figure of merit for
the eye is seen to be 250 times that for film.

Eye
Film

2.5

TABLE 1
B, Ft-L

10
100

A, In.'
0.03

0.5

20
0.08

Comparison of Film and Television Pickup Tubes. — Fig. 6
shows the setup for comparing the low light performance of Super
XX film and an image orthicon. The original subject was illuminated
with an ordinary 40- w bulb attenuated with neutral filters. The
television camera was focused on the subject alone and its picture was

286

A. ROSE

Vol 47, No. 4

reproduced on a receiver located alongside the subject. The 35-mm
camera photographed simultaneously the original and reproduced
pictures. Both cameras used f/2 lenses and an exposure time of I/M
sec. Fig. 7 shows the results. Only in the first exposure, at 2 ft-L
brightness of the subject, do both original and reproduced pictures

INCANDESCENT LiGHT SOURCE

0.2

FOOT
LAMBERTS

0.07

0,02

FOOT
LAMBERTS

FIG. 7. Comparison of low light performance of Super XX film and an image
orthicon. (Image orthicon picture is on the left of each frame.)

appear. At 0.2 ft-L only the picture reproduced by the television
camera is present. And, in fact, the television camera continues to
transmit a picture even at 0.02 ft-L which is the brightness of a white
surface in full moonlight.

One interpretation of this test is that the image orthicon is 50 times
as "sensitive" as Super XX film because it can transmit a picture with
Ysoth of the light required by the film. The present paper argues

Oct. 1946 FILM, TELEVISION AND THE EYE 287

against this interpretation and sets the factor at ten. This is based
on the fact that the area of target (photo-cathode) used by the image
orthicon was five times the area of the- .'J.Vmm film frame. If the
cameras were to be set up to transmit the same picture with the same
angle of view and depth of focus, the lens on the image orthicon
would have to be stopped to 51/2 times the numerical aperture of the
lens for the 35-mm camera. The threshold scene brightnesses would
then be in the ratio of 10 : 1.

Graininess and Signal-to-Noise Ratio.— An excellent survey
of the extensive history of the problem of specifying the graininess
of film has recently appeared by Jones and Higgins.8 In this paper
and in a second one14 they undertake to compare two general methods
of measuring graininess. Method I, which they describe as a psycho-
physical measurement, records the distance from the observer at
which the grainy film appears to blend into a uniform surface. (After
introducing an observer for his special virtues as a measuring instru-
ment, he is ushered part way out again by the device of normalizing
his results with a standard test chart.) Method II is an objective
measurement of the transmission or density fluctuations of the film
using scanning apertures of various sizes. Broadly, Jones and Hig-
gins argue (1) that the objective measurements should match the
"blending distance" measurements in order to be considered valid;
(2) that the two types of measurements do not match ; and (3) that a
major discrepancy is that the blending distance measurements tend
to decrease at large densities while the objective measurements tend
to increase.

In contrast to the above, the present paper would argue that the
two types of measurement, I (by the eye) and II (by a scanning aper-
ture), should, so far as the eye and film satisfy the same physical
equations derived for an ideal device, show good* agreement. A
large part of the discrepancy noted above under (3) is removable
when reference is made to Fig. 4. Here it is seen that in the range of
0.1 to 10 ft-L the discrimination of the eye for small contrast differ-
ences varies by about five to one. This would correspond to a five-to-

* A precise correlation between eye and instrument observations must, of
course, take into account the detailed performance of eye and film near their
limiting resolution — performance which both for eye and film is determined more
by the finite size of its mosaic elements than by statistical fluctuations. The sig-
nificance attached to precise visual observations, however, should be tempered by
the known large spread of eye characteristics from individual to individual.

288 A. ROSE Vol 47, No. 4

one ratio of blending distances for the same film viewed at a bright-
ness of 10 and at a brightness of 0.1 ft-L. Because the visual observa-
tions of blending distance are made with a fixed source brightness
attenuated by films of varying density, the resulting blending dis-
tance measurements are a product of the graininess properties of the
film and the contrast discrimination properties of the eye as a func-
tion of scene brightness. When the latter term is separated out, the
graininess versus density measurements by the two methods (ob-
server and instrument) show relatively good agreement.

A further rough confirmation may be obtained by reference to some
"blending distance" measurements of Lowry15 in which a constant
viewing brightness was preserved. These showed about a factor of
two increase in graininess for a variety of films in the range of densities
from 0.2 to 1.0. This increase is in good agreement with the objective
(large aperture) measurements of Jones and Higgins14 shown in their
Fig. 16.

It is worth commenting briefly on another item emphasized in the
second paper by Jones and Higgins.14 The concept of the "effective
scanning area" used by the eye in evaluating graininess is introduced.
This is thought to be a useful concept particularly because the results
of objective measurements, using different scanning aperture sizes,
suggest the possibility of matching visual observations with small
apertures rather than large apertures.

Arguments, similar to the above, were at one time current in evalu-
ating the "noise" in a television picture. It was often remarked that
it was only the high-frequency noise that was objectionable. This
would correspond, for example, to selectively emphasizing the ob-
servations of graininess of film obtained with small scanning apertures
(either retinal or instrumental) . It is a relatively simple experiment
in a television system to increase the. effective scanning aperture
several fold, either by reduction of pass band or by defocusing the
kinescope spot. Such aperture changes are accompanied by large
changes in total noise power as viewed on the kinescope. Yet the
effect on visibility of noise of cutting out the high-frequency noise
components is small compared with the same changes in noise power,
distributed uniformly over the noise spectrum. This latter state-
ment is borne out by the curves in Fig. 7. In brief, the visibility, or
annoyance, of noise must be assessed over the full range of picture
element sizes from elements at the limiting resolution of the eye to the
largest element, which is the picture itself considered as a unit.

Oct. 1946

FILM, TELEVISION AND THE EYE

289

Resolution. — The most frequently used, because it is the most
easily observed, specification of resolution is the finest detail that a
system can resolve. This is true for film, pickup tubes, the eye, and
optical lenses. In general, this, specification is satisfactory if it is
appreciated that the limiting resolution itself has only narrow utility
and that the limiting resolution is more an indirect measure of what
detail is well resolved by the system. The "well-resolved" detail may
be two to four times coarser than the finest detail. And in the judg-
ment of picture quality, the eye attaches little weight to the picture
elements in the neighborhood of limiting resolution.

B, Bj

LINEAR ELEMENT SIZE - h

FIG. 8. Dependence of scene brightness on reproduction
brightness.

One illustration of the confusion caused by the use of limiting reso-
lution is the comparison frequently made between the resolution of
motion picture film and of a television system. The limiting resolution
of film is compared with the "cutoff" resolution of a television picture.
The picture detail at the "cutoff" resolution of a television system,
however, as limited by the amplifier pass band, has at least the pos-
sibility of being clearly resolved. It is misleading to attach the same
weight to this type of resolution as is attached to the limiting resolu-
tion of film. It would be nearer a true evaluation if the resolution of
film were specified at that number of lines at which film matched the
signal-to-noise ratio of a television system at its "cutoff." Such a

290

A. ROSE

Vol 47, No. 4

comparison would place the resolution of 35-mm motion picture film,
normally quoted at a limiting resolution of 1000 to 2000 lines, nearer
to the resolution of a 500-line than a 1000-line television picture.

In general, the specification of the signal-to-noise ratio that a picture
pickup device can transmit 'at an intermediate resolution is a more
accurate and significant specification, not only of resolution, but also

(c)

LINEAR ELEMENT SIZE-h

FIG. 9. (a) Noise reduction by lowered repro-
duction brightness; (&) Noise reduction by in-
creased observer distance; (c} Noise reduction
by bandwidth reduction.

at the same time of the half-tone discrimination of the device, than
is the specification of limiting resolution.

Satisfying the Human Eye. — Only one problem, that of presenta-
tion brightness, will be discussed here. Fig. 8 shows the signal-to-
noise ratio curves of a picture taken at scene brightness BI, and viewed
by the eye at presentation brightness B\. The viewed picture is as-
sumed to be "noise free"and accordingly the BI curve lies above

Oct. 1946 FILM, TELEVISION AND THE EYE 291

the B\ curve. If, now, the presentation brightness is increased to
B»f, the original scene brightness must also be increased, other
things being constant, by the same factor to B2' in order to match
the increased discrimination of the eye. These considerations are
significant because both motion pictures and television pictures aim
at higher presentation or screen brightness.

The converse of the above operations makes an interesting test.
Given a grainy motion picture or a "noisy" television picture, the
most effective way of eliminating the fluctuations with the least cost
to picture detail is to interpose a neutral filter between the eye and
the picture. The discrimination of the eye is thereby readily reduced
below the fluctuation limits of the picture. At the same time, the
picture is shifted to a portion of the eye characteristic which shows
higher apparent contrast and thus partially compensates for the loss
of brightness. Fig. 9 shows schematically the effect on picture de-
tail of three ways of trying to eliminate "noise" : reduction of picture
brightness, increase in viewing distance and reduction of bandwidth
of the picture. The last-named operation is peculiar to a television
system and, while it reduces the total noise in the system, has little
effect on the visibility of noise until an extremely coarse picture is ob-
tained.

The curves BI are the signal-to-noise ratio characteristic of the pic-
ture. The curves BI ', B\ , B2' are the signal-to-noise ratio character-
istics of the eye at the brightnesses BI and B2f. In order that the
fluctuations in the picture not be observed by the eye, the signal-to-
noise ratio of the picture should be above B\ , B\" , or Bz'. The limits
of performance of the eye are shown by the three dotted lines. They
mark out the minimum area that the eye can resolve by virtue of its
cone structure, the minimum signal-to-noise ratio that it can perceive,
and the maximum signal-to-noise ratio it can generate corresponding
to the Weber-Fechner limit of 2 per cent brightness discrimination.
The "cutoff" characteristics of the eye are shown as idealized sharp
breaks to simplify the argument.

Starting with a noisy picture, that is, B\ lying above BI as in Fig.
9a, there are several formal operations that can be performed to get
rid of the noise, that is, to insure that all parts of B\ lie above the eye
curve. Each of these operations corresponds to a physical operation
and each affects the observed picture detail differently. In Fig. 9a
the eye curve B\ is transformed into B2' by a change of ordinate scale
factor. This corresponds to interposing a neutral filter at the eye.

292 A. ROSE Vol 47, No. 4

The finest detail observable is still at the "cutoff" limit of the eye.
In Fig. 96, the eye curve B\ is transformed into B\ by a change of
abscissa scale factor. This corresponds to backing away from the
picture. Although the finest observable detail remains at the "cut-
off" of the eye, this "cutoff" now corresponds to coarser detail in the
picture. In Fig. 9c, the pass band of the amplifier through which the
original picture is transmitted is reduced to the point where the pic-
ture fluctuations are below the Weber-Fechner limit. This is an
expensive way to remove noise — expensive in picture detail.

A final aspect of the significance of presentation brightness arises
in the comparison of the low light performance of a man-made device
with that of the human eye. Assume, for example, that the man-
made device is as sensitive as the eye. If one picks up a scene whose
brightness is 0.1 ft-L and views the reproduction at a presentation
brightness of 10 ft-L, noise should be visible in the reproduction while
it was not visible in the original scene. The higher presentation
brightness gives the eye an unfair advantage. A more valid procedure
would match the presentation brightness of the reproduction with
the brightness of the original scene.

Visual Noise. — The phrase "signal-to-noise ratio of the eye"
has been used frequently in the preceding discussion. One might
expect to be able to "see" these fluctuations just as one sees the graini-
ness of film or the noise in a television picture. The writer is con-
vinced that such fluctuations are observable* particularly at low
lights around 10~4 ft-L. A white surface then takes on a grainy ap-
pearance not unlike that of motion picture film. The observations
in more detail are: in complete darkness little or no fluctuations are
detectable, a fact which attests the substantial absence of local noise
sources in the eye. Near threshold brightnesses, large area, low am-
plitude fluctuations appear. At higher brightnesses these fluctuations
increase in amplitude and decrease in size. In the neighborhood of
10~2 ft-L the fluctuations tend to disappear and a white surface takes
on a "smooth" appearance and remains so at normal brightness
levels. A secondary observation is that low-level blue light appears
distinctly more grainy than low-level red light.

The last observation, together with known data on dark adapta-
tion, fits in well with the assumption of a gain control mechanism in
the eye. This gain control, just as the gain control in a. television

* See also DeVries.

Oct. 1946 FILM, TELEVISION AND THE EYE 293

receiver or the lamp brightness used for film projection, does not alter
the signal-to-noise ratio but does alter the visibility of noise by pre-
senting the picture at a higher or lower brightness level. Thus, at
high scene brightnesses, the gain control in the eye may be turned
down to the point where the fluctuations are just not visible. (The
sensitivity of the eye is apparently high enough to afford this luxury.)
If one suddenly reduces the scene brightness, the gain control is still
momentarily set at a low value and the picture is dim or not visible.
As the gain control resets itself at a higher level, the picture appears
to get brighter. This corresponds with the experience of dark adap-
tation. At these low light levels (10~4 ft-L) one has only to assume
that the gain control is set high enough to make the fluctuations
visible.

To account for the observation that low-level blue light appears
to have more fluctuations than low-level red light, the gain control
mechanism can be assumed to be set higher for blue than for red.
This is not as ' 'ad hoc' ' as it may appear. The reason is that, although
at low-light levels* blue appears brighter and grainier than red, they
both present the same resolution to the eye.16 And since the resolu-
tion is determined by signal-to-noise ratio, this is in agreement with
the assumption of a gain control that varies presentation brightness
but not signal-to-noise ratio.

Acknowledgments. — The writer would like to acknowledge,
without committing the acknowledged to the conclusions presented
above, his indebtedness to Dr. D. O. North and Dr. G. A. Morton
of these laboratories, and O. H. Schade of the Victor Division of
RCA for many profitable discussions of the subject of this paper.

REFERENCES

1 ROMER, W., AND SELWYN, E. N: "An Instrument for the Measurement of
Graininess," Phot. Jour., 83, (1943), p. 17.

2 ROSE, A.: "The Relative Sensitivities of Television Pickup Tubes, Photo-
graphic Film and the Human Eye," Proc. I.R.E., 30, 6 (June 1942), p. 295.

3 DEVORE, H. B., AND IAMS, H. A.: "Some Factors Affecting the Choice of
Lenses for Television Cameras," Proc. I.R.E., 28, 8 (Aug. 1940), p. 369.

4 FARNSWORTH, P. T.: "Television .by Electron Image Scanning," Jour.
Frank. Inst., 218, 4 (Oct. 1934), p. 411.

5 ZWORYKIN, V. K., MORTON, G. A., AND FLORY, L. E.: "Theory and Per-
formance of the Iconoscope," Proc. I.R.E., 25, 8 (Aug. 1937), p., 1071.

* The test is performed by starting with red and blue at the same brightness at
high-light levels and attenuating both by the same neutral filter.

294 A. ROSE

6 ROSE, A., AND IAMS, H. A.: "The Orthicon, a Television Pickup Tube,"
RCA Rev. ,4,2 (Oct. 1939), p. 186.

7 ROSE, A., WEIMER, P. K., AND LAW, H. B. : "The Image Orthicon, A Sensi-
tive Television Pickup Tube," Proc. I.R.E., 34, 7 (July 1946), p. 424.

8 For summary of literature, see: JONES, L. A., AND HIGGINS, G. C.r "The
Relationship Between the Granularity and Graininess of Developed Photographic
Materials," /. Opt. Soc. Am., 35, 7 (July 1945), p. 435.

9 VAN KREVALD, A., AND SCHEFFER, J. C.: "Graininess of Photographic Ma-
terial in Objective and Absolute Measure," J. Opt. Soc. Am., 27, 3 (Mar. 1937),
p. 100.

10 RAUDENBUSCH, H.: "Measurements of Graininess and Resolution of Photo-
graphic Film," Phys. Zeits., 42, 15/16 (Aug. 1941), p. 208.

11 SILBERSTEIN, L., AND TRiVELLi, A. : "Quantum Theory of Exposure Tested
Extensively on Photographic Emulsions," /. Opt. Soc. Am., 35, 2 (Feb, 1945),
p. 93. (The writers of this paper avoid emphasizing the physical implications of
their analysis. At the same time they do interpret their results to show that the
intrinsic sensitivity of film is increased by longer development times. So far as
other measurements have shown that the increase in speed resulting from long
development time is paralleled by an increase in graininess, the present paper
would argue that the intrinsic sensitivity is unchanged but that the developed
grains are made larger by longer development time.)

12 CONNOR, J. P., AND GANOUNG, R. E.: "An Experimental Determination of
Visual Thresholds at Low Values of Illumination," J. Opt. Soc. Am., 25, 9
(Sept. 1935), p. 287; COBB, P. W., AND Moss, F. K., "The Four Variables of
Visual Threshold," /. Frank. Inst., 205, 6 (June 1928), p. 831.

13 References to number of quanta used by the eye for a threshold sensation:

1 quantum — DEVRIES, H.: "The Quantum Character of Light and Its Bear-
ing on Threshold of Vision, Differential Sensitivity and Visual Acuity of the
Eye," Physica, 10, 7 (July 1943), p. 553.

2 quanta — VAN DER VELDON, H. A. : "The Number of Quanta Necessary for
a Light Sensation for the Human Eye," Physica, 11,3 (Mar. 1944), p. 179.

4 quanta — HECHT, S. : "Quantum Relations of Vision," /. Opt. Soc. Am., 32,
. 1 (Jan. 1942), p. 42.

25 to 50 quanta — BRUMBERG, E. M., VAVILOV, S. I., AND SVERDLOV, Z. M.:
"Visual Measurements of Quantum Fluctuations," /. Phys. (Russian), 7, 1
(1943), p. 1.

14 JONES, L. A., AND HIGGINS, G. C.: "Photographic Granularity and Graini-
ness," J. Opt. Soc. Am., 36, 4 (Apr. 1946), p. 203.

15 LOWRY, E. M.: "An Instrument for the Measurement of Graininess of
Photographic Materials," J. Opt. Soc. Am., 26, 1 (Jan. 1936), p. 65.

16 LUCKIESH, M., AND TAYLOR, A. H. : "A Summary of Researches in Seeing at
Low Brightness Levels," Ilium. Eng., 38, 4 (Apr. 1943), p. 189.

THE HIGH COST OF POOR PROJECTION*

CHARLES E. LEWIS**

Summary. — Only five per cent of the capital invested in the motion picture in-
dustry is in studios and distribution facilities. Ninety-five per cent, aggregating a
total of $1,900,000,000, is invested in motion picture theaters and their equipment.
This staggering investment is in the hands of the exhibitor. He needs technical ad-
vice.

The Society of Motion Picture Engineers has built up the necessury body of tech-
nical knowledge; but the exhibitor cannot follow the Society's work in detail because
he is equally harassed with many theater problems. The Society should arrange to
segregate and collect those items of information appearing throughout its JOURNAL
which would be of direct help to him. These could then be published in handy refer-
ence form for dealing with theater projection problems. There is no other single aid
to good projection that could be more helpful.

It has very often been said that the process of putting a motion
picture before the public involves a long chain of operations. Each
link is equally indispensable to the final result. A broken chain is a
broken chain no matter where the break occurs. This Conference has
heard some highly valuable and important papers that deal with
chemical analysis of developing solutions. That is one link. The
decision of a theater owner or manager about signing a check for im-
provements in projection equipment is another — and equally im-
portant.

Weakness anywhere in the chain is expensive to the industry. As
for poor projection, the industry pays for that in at least three dif-
ferent ways.

On the question of what relationship exists between good projec-
tion and box-office receipts, Showmen's Trade Review recently conduc-
ted a survey. Selected groups of representative theater managers
were queried. These are the men who are on the spot and ought to
know. In the opinion of 80 per cent of those who answered, an un-
questionable increase in box-office revenue follows improvements in
projection or sound equipment.

* Presented May 9, 1946, at the Technical Conference in New York.
** Publisher and Editor, Showmen's Trade Review, New York.

295

296 C. E. LEWIS Vol 47, No. 4

The industry loses this increase in revenue wherever projection is
neglected.

There is an additional loss, however, which is not confined merely
to those theaters that are neglectful. The best equipment and the
best care cannot get good projection out of mutilated prints. The
industry suffers a further box-office loss through the carelessness of
those theaters which neglect their equipment to the point where it
damages prints.

A third loss to the industry results from shortening the life of the
prints.

Good projection, therefore, is a matter of practical financial con-
cern to everyone who draws his livelihood from this industry. The
cost of poor projection is a direct or indirect tax on all of us, and a
drain that helps to limit appropriations for research and development.

The practical questions would seem to be: (1) What is good pro-
j ection ? (2) Ho w can we get it ?

Good Projection — It is not the place of an exhibitor to tell
the Society of Motion Picture Engineers what good projection is.
So far as the exhibitor is concerned, good projection is that which
conforms to the standards this Society has been so patiently elabo-
rating for many years.

Good projection is that in which the brightness at the center of the
screen is between 9 and 14 ft-L when the projector is running with no
film in the gate. I do not say this. Your Committee on Standards
says so. And in this connection I hope it will not be out of place to
add a word, from the exhibitor's point of view, with respect to the
very valuable work done by your Film Projection Practice Committee.
It would, in my opinion, be a great service to the industry if, now that
the war is over, the activities of the Film Projection Practice Commit-
tee could be resumed.

It is you — the Society — who tell this industry what good projec-
tion is. But when the Society of Motion Picture Engineers estab-
lishes standards for the motion picture industry there still remains a
vital practical problem. It is the problem of keeping that informa-
tion effectively at the elbow of a critically important class of men who
have urgent need of it — and I propose to confine this discussion to
that one link only — the exhibitor — the man who has to make up his
mind about signing the check.

That man needs help. It is easy to point out that the SMPE puts
invaluable technical guidance at his disposal. But is that poor,

Oct. 1946 HIGH COST OF POOR PROJECTION 297

harassed man also to become a member of the American Society of
Heating and Ventilating Engineers, the American Institute of Archi-
tects, the Illuminating Engineering Society, and the American In-
stitute of Accountants? All these organizations no doubt could give
him information that would be of great value to him. He is also ex-
pected to be an expert on advertising, exploitation, and employee rela-
tions, and on the supervision of plumbing.

The Exhibitor. — The exhibitor is the forgotten man of this indus-
try. Yet his stake in it is vastly greater than that of all other
branches combined. Recent figures of the Motion Picture Produc-
ers and Distributors Association show a total sum of 125 million
dollars invested in American studios, an additional 25 million dol-
lars investigated in American distribution facilities, and the tre-
mendous total of one billion, nine hundred million dollars invested
in American motion picture theaters.

That is to say, approximately 95 per cent of all the capital in this
industry is in theaters and their equipment. This stupendous ag-
gregate of capital value is in the hands of the exhibitor — the forgotten
man.

With respect to projection, the exhibitor has advice at his disposal.
The projectionist is another very important link in the chain. The
exhibitor can secure invaluable technical advice from his projection-
ists. Service inspectors and supply dealers also can and do advise him.
But the trouble is he is the man who must make the decision. The
advice he gets from these sources is often conflicting. And he must
keep in mind that while the advice is almost invariably honest, it is
not always disinterested. He needs some little information of his
own — some standards of reference — to help him decide.

It is true that in considering what to do about conflicting advice,
the exhibitor can turn to Vol 43 of the JOURNAL of the SMPE and
find there what his screen brightness ought to be according to the
Committee on Standards. If he reads carefully through Vol 39 of
the JOURNAL he will find out that he ought to install a voltage reg-
ulator if line variations exceed ± 3 per cent. Vol 36 will tell him
that his screen width should equal one-sixth the distance from the
screen to the rear seats, and so on.

But probabilities are that he does not have a file of the JOURNAL,
and that if he did he could not find what he wanted when he needed
to find it.

Yet, he is the man whose decisions are final. He signs the checks.

298 C. E. LEWIS

He and no one else, in the great majority of theaters, determines what
shall be done and what left undone.

Getting Better Projection. — This brings me to the practical sug-
gestion I would like to place respectfully before you for your con-
sideration. Can the Society arrange to segregate and collect those
items of information appearing throughout the JOURNAL which
would be of direct help to the exhibitor, and to publish them in a form
which the average exhibitor could use as a small reference encyclope-
dia to his projection problems? Or if the Society is not willing to
undertake this work, will you permit others to do it?

You have compiled technical data about projection that is of the
utmost practical value. The highest kind of skill and competence
and an immense amount of labor have gone into building up this body
of technical knowledge.

I can think of no one single aid to good projection that would be
more helpful than to put this mass of expert information into the
hands of the exhibitor in a form in which he can make effective use of
it.

FACTORS GOVERNING THE FREQUENCY RESPONSE OF A
VARIABLE-AREA FILM RECORDING CHANNEL*

M. RETTINGER** AND K. SINGER**

Summary. — This paper is an analysis of the essential factors governing the most
desirable dynamic frequency response of an RCA variable-area sound recording chan-
nel, and includes a study of the most suitable location of the various equalizers em-
ployed in the system. It consists of five parts:

(1 ) Derivation of equalization characteristic for the RCA sound-on-film recording
channel;

(2) Review of effort equalization and of relative spectral energy distortion in
electronic compressors;

(3} Recommended recording channel equalization;

(4) Experimental -work;

(5) Conclusions.

By dividing the paper into these interrelated parts, the different phases of the sub-
ject can be studied more conveniently than if no such subdivisions were made. Also,
the fifth part contains the results and recommendations reached in the first four parts.

PART 1

Derivation of Equalization Characteristic for the RCA Sound-on-
Film Recording Channel. — In the field of sound recording, as in
every other branch of modern industry, effort is constantly being
made to improve the product. Operating conditions are frequently
checked and the results critically examined to learn whether any
change in equipment or technique would provide more natural or
more effective sound recording.

The frequency response of sound recording channels has been deter-
mined largely by empirical methods. These methods will probably
continue in use for some time, at least to the extent of providing the
sound director with a means of controlling the finer adjustments to
suit his judgment.

A study of the essential fixed factors determining the equaliza-
tion characteristics of an RCA variable-area sound recording chan-

* Presented May 9, 1946, at the Technical Conference in New York.
** RCA Victor Division, Radio Corporation of America, Hollywood.

299

300

M. RETTINGER AND K. SINGER

Vol 47, No. 4

nel has been made. The purpose of this study was to observe how
closely the calculated characteristics came to agreeing with those
determined by empirical methods and to determine whether the re-
sults could be improved by replacing the empirically determined char-
acteristics by those obtained by calculation.

The electrical characteristics of a high-fidelity sound recording
system are as important as are those of the physiology and psychology
of speech and hearing. There are also a number of acoustic and
electroacoustic factors which must be critically examined.

0
DB
-5

0
DB-5
-10

0
DB
-5

LOSS DUE TO FILM

PROCESSING (NEGATIVE t PRINT)

0
5

— U

CURVE*
1

^
3

—

• .

— ^

— .

•^

LOSS DUE TO FILM
AND RECORDER SLIT

PROCESSING (NEGATIVE £ PRINT)
ATTENUATION WITH 1/4- MIL SLIT

— M

CURVE*

—

— -.

— " •»

"*•

^ ^

CALCULATED RECORDER
WITH t/4- MIL SLIT

SLIT LOSS FOR RECORDER

CURVE*

IpOO
FREQUENCY IN CYCLES PER SECOND

FIG. 1A.

10,000 20,000

The following discussion deals with each of the important phases of
sound recording and reproduction, with which we are concerned, in
order to ensure satisfactory results in a typical theater.

We will first discuss the factors which determine the optimal fre-
quency response of an RCA variable-area sound-on-film recording
system.

Curve 1 of Fig. 1A shows the high frequency loss owing to a Y^mil
recorder slit which is currently used in all of our studio- type record-

ers/

Curve 2 of Fig. 1A illustrates the loss of high frequencies caused by
film processing and recorder slit attenuation. The curve is not based
on theoretical calculation, but represents experimental results. The

Oct. 1946

FACTORS GOVERNING FREQUENCY RESPONSE

301

recorder slit was J/4 mil m width. The film processing losses include
not only the so-called negative processing losses, but also the printer
and print losses. Ultraviolet light was used in the recording, and
the film processing was carried out in a commercial Hollywood labora-
tory. These negative processing, printer, and print losses may be
assumed as correct for standard, push-pull, and class-5 recording.
They may not be correct for all types of film, and slight deviations
may occur in different laboratories.

Subtracting Curve 1 from Curve 2 of Fig. 1 A, we obtain Curve 3 of
Fig. 1A which represents the processing losses proper.

Fig. IB shows the frequency response characteristic of the RCA
type MI -3 121 low-pass filter for different cut-off frequencies. Ordi-

MHO/00

Ml-3121- *

6000

10.000-

-10

20 100 1000 IQOOO 20,000

FREQUENCY IN CKCtfS P£Q SECOND

FIG. IB. Frequency characteristics of MI-10100 high-pass filter and MI-
3121 low-pass filter.

narily the attenuation of the filter at 6000 cycles is not considered in
the derivation of the recording characteristic. The reason for this
is that the attenuation is small at this frequency.

Fig. IB also shows the frequency response characteristic of the RCA
type MI-10100 high-pass filter. This characteristic also is ordi-
narily not considered in the derivation of the recording characteris-
tic. The reason for this is that in most instances, the filter char-
acteristic has little effect in the region of dialogue equalization.

Curve 1 of Fig. 1C represents the amount of low frequency ac-
centuation caused by the reproducing level being (on the average) 5 db
above normal speech level which is the level employed by a person in
somewhat noisy surroundings. This increase in level is occasioned
by noises from the audience and other sources in the theater; the
resulting low frequency accentuation is caused by the hearing char-

302

M. RETTINGER AND K. SINGER

Vol 47, No. 4

acteristic of the human ear. The ear sensitivity at different frequen-
cies varies as a function of the intensity of the signal. The higher
the intensity of the signal, the more sensitive the ear becomes to the
low frequencies. To illustrate: Consider a 100-cycle tone and a
1000-cycle tone, each emitted from a speaker at the same intensity
level, 60 db above the threshold of audibility. If the signal level for
each is increased by a certain amount, 10 db, the 100-cycle tone will
now sound louder to the ear, in reference to the 1000-cycle tone, than

-25

•2.5

CURVE. I

LOW FREQUENCY ACCENTUATION DURING REPRODUCTION

CURVE E

RELATIVE SPEECH CHARACTERISTIC CHANGE WE TO LQW£R£D i/O/Cf

-2.5

CURVE m SPEECH LtVCL CHANGE DUE TO REVERBERATION CHARACTERISTIC OF STUDIO

MI -3043 UNI -DIRECTIONAL MICROPHONE CHARACTERISTIC (6 • 45')

+10

CURVE. Y

ALGEBRAIC SUM OF THE ABOVE CURVES

ipoo

FREQUENCY IN CYCLES PEP SECOND

FIG. 1C.

JQOOO ZQPOO

the 100-cycle tone sounded in reference to the 1000-cycle tone before
the signal level was increased.

Curve 2 of Fig. 1C illustrates the relative speech characteristic
change, plotted relative to 1000 cycles.2 Dialogue spoken on a sound
stage is usually uttered at a level 5 db below the level at which it
would be uttered if the surroundings were not so quiet. When a
person speaks in a low tone, the low and high frequencies become rela-
tively more distinctive as compared to a voice characteristic of nor-
mal intensity. If this low voice were reproduced at normal speech
level — the level usually employed by a person situated in the more or

Oct. 1946 FACTORS GOVERNING FREQUENCY RESPONSE 303

less noisy surroundings depicted on the screen — it would tend to
sound unnatural ; that is, it would sound both somewhat heavy and
sibilant.

Curve 3 of Fig. 1C indicates the average accentuation of the low
frequencies and the loss of the high frequencies due to the reverbera-
tion characteristic of the sound stage. Practically all acoustic mate-
rials are less absorbent for the low frequencies and become increas-
ingly absorbent for the high frequencies. For this reason, the rever-
beration time at the accentuation or loss may be Expressed by

db = lOlogio-^1

L o

where T\ is the reverberation time at the frequency under considera-
tion, and To is the reverberation time at 1000 cycles.

It should be noted that Curve 3 of Fig. 1C depicts the sound-level
change caused only by the variation in reverberation time in the
sound stage. It does not cover any losses which may be occasioned
by peculiar set conditions; that is, pronounced panel resonance of a
set, or room resonance of a set, or room resonance in a small enclosure.
This last condition of room resonance deserves some careful con-
.sideration, particularly if the trend of providing ceilings in sets is
continued. Experience has indicated that room resonance effects are
the most difficult to be compensated for in rerecording. This is
because usually, in the low-frequency spectrum, only a narrow fre-
quency band is intensified, with the result that the ordinary type of
low-frequency attenuation available in the rerecording channel can-
not eliminate this selective emphasis without producing at the same
time an unnatural effect on the character of the voice itself.

Curve 3 of Fig. 1C does not show the variation in high-frequency
sound level owing to the varying reverberation time. This variation
is occasioned by changes in the relative humidity of the air in the
stage. The absorption of sound in the air becomes very great at the
high frequencies for low relative humidities. It doubles at 6000
cycles when the normal relative humidity of 40 per cent is reduced to
20 per cent. This effect would point to the use of a variable high-
frequency equalizer to compensate for varying conditions of humid-
ity. At present no compensations for this effect are made for dialogue
recording. One studio employs a variable high-frequency compensa-
tor for their music recording, to equalize the effects of varying humid-
ity during a scoring session. The studio does not use it however,

304

M. RETTINGER AND K. SINGER

Vol 47, No. 4

for dialogue recording when changes of location produce as much,
if not more, of a change in the high frequency transmission of sound
in air. Their reason for this procedure lies in the blanket statement
that "the extreme high frequencies are more important in the case of
music than in speech."

Curve 4 of Fig. 1C shows the RCA type MI-3043 unidirectional
microphone response characteristic for 45-deg incidence of sound.
This angle was chosen because it represents the normal pick-up angle

10
5
OB 0
-5
-10
-15

,*-— -^^

Vfc^

^^--

CURVE*!

ALGEBRAIC SUM OF CURVE*? Of

FIG. IA t, CURVE *5 OF FIG. 1C

100

1000

10.000 20.000

15
10

5
DB 0

15
10
5
0
5
10
15

AVERSE OF CURVE*!. THIS fS T
QUALIZATION CHARACTERISTIC
RECORDING CHANNEL EMPLOYING
JNI-DIRECTIONAL MICROPHONE,
/4 RECORDING SLIT, $ STANDAF
ROCESSING

HE
FC
IV

1-
V,
Fl

ALCL
(\Ct

LATED

E
' F
— I.
1
F

304J
LM

^ '

CURVED?

-10
-15

— *

100 (POO

FREQUENCY IN CYCLES PER SECOND

FIG. ID.

10,000 20,000

employed in the studios. For the purpose of this discussion it is as-
sumed that the high-frequency equalization available in the micro-
phone is not employed. If any equalization is introduced in the
microphone, a corresponding compensation must be made in the re-
cording characteristic. Present studio practice seldom avails itself
of the high-frequency equalizer in the microphone. The coil in this
equalizer together with a suitable resistor, is most frequently em-
ployed to provide microphones having "matched," that is, practically
identical, low-frequency response.

Oct. 1946 FACTORS GOVERNING FREQUENCY RESPONSE

305

Curve 5 of Fig. 1C shows the algebraic sum of Curves 1, 2, 3, and 4.
For the sake of convenience in the following discussion, the curve rep-
resenting this algebraic sum will be referred to as the curve of "acous-
tic losses," in comparison with the curve representing film processing
losses and slit attenuation, which will be labeled "optical losses."

Curve 1 of Fig. ID shows the algebraic sum of the acoustical and
optical losses. Curve 2 of Fig. ID represents the theoretically de-
rived recording characteristics and is the inverse of the summation
Curve 1. This curve has been replotted on Fig. IE (top curves)

-5

-10

-IS

CALCULATED LOW FREQUENCY
CHANNEL CHARACTERISTIC

•^

CALCULATED HIGH FREQUENCY
CHANNEL CHARACTERISTIC

- i-

• '

„ — •"

0
-5

-10
-15

FREQUENCY RE
MI-31

SPONSE OF THE
6 A COMPENSATOR

, —

= .•

FREQUENCY RESPONSE OF THE
MI-I02O9 PRE-AMPUFIER

J^1

^-*

, — -!

-— ""•"'""

10.000 20.000

50 100 500 IPOO 1500 500 1,000

FREQUENCY IN CYCLES PER SECOND

FIG. IE.

where it has been "smoothed out," a process justified on the basis
of the many variables included in the analysis. Fig. IE (bottom
right) shows the frequency response of the RCA type MI-10209 pre-
amplifier ordinarily used. This might indicate that the high-fre-
quency equalization usually employed by the studios in connection
with the RCA type MI-3043 unidirectional microphone would give
rise to a slight prominence of "highs" (frequencies between 1500 and
6000 cycles).

An objection to the prominence of "highs" has actually been made
by one studio. On the other hand, a number of other studios claim
that a slight prominence of this frequency range gives the product

306

M. RETTINGER AND K. SINGER

Vol 47, No. 4

"presence." It should be noted that for frequencies above 6000
cycles, the low-pass filter becomes effective.

Fig. IE (bottom left) shows the frequency response of the RCA
type MI -3 116 A low-frequency compensator, set on step 2.

PART 2

Review of Effort Equalization and Relative Spectral Energy
Distortion in Electronic Compressors. — In the recording of sound on
film, various types of frequency discriminations occur, some of
which are compensated by equalizers in the recording channel.
It is usual practice in the art thus to equalize for the frequency dis-
crimination of the microphone, film processing, reproducer slit, etc.

H» 70
1

LOUD --

NORMAL

SOFT

20 100 1.000 10,000 20,000

FREQUENCY IN CYCLES PER SECOND

FIG. 2 A. Average voice characteristics of men and women.

Another type of frequency discrimination is a variable function of the
sound-level difference between normal speech levels and the voice
level actually employed by the actor on the set. The correction for
this latter effect has been termed "effort equalization."2

Another type of frequency discrimination results from the action
of the compressor usually employed in the variable-area sound record-
ing channel.3 It is the purpose of this section to review these two
effects.

Oct. 1946 FACTORS GOVERNING FREQUENCY RESPONSE

307

In the case of "effort equalization," cognizance is taken of the fact
that the voice level of the actor during recording is rather low, be-
cause a person in a very quiet surrounding, such as a sound stage,
involuntarily lowers his voice. At low levels the voice characteris-
tic shows a preponderance of both low- and high-frequency compo-
nents, as compared to a voice characteristic of normal intensity. If
this voice were reproduced at normal speech level, the level usually
employed by a person situated in the more or less noisy surroundings
depicted on the screen, it would tend to sound unnatural; that is,
both somewhat heavy and sibilant.

-10

-20

-30

-40

NORMAL SPEECH
SOFT SPEECH
LOUD SPEECH

100 1.000

FREQUENCY IN CYCLES PER SECOND

10.000 ^O.OOO

FIG. 2B. Average voice characteristics, men and women, "matched" at 500

cycles.

The only corrective measures considered in this connection are
those that will compensate for the change in the voice characteristic,
a change caused by the actor speaking in a lower tone than is normal.
In the case where the low voice level of the actor corresponds to his
normal tone of voice called for by the surroundings in which he ap-
pears, no correction is required.

Fig. 2A shows the average voice characteristics of men and women.
The data for this figure were taken from the paper by Loye and Mor-
gan. These curves were redrawn on Fig. 2B and matched at 500
cycles to show more readily that soft or low speech contains a pre-

308

M. RETTINGER AND K. SINGER

Vol 47, No. 4

ponderance of both low- and high-frequency components, and that
loud speech is considerably lacking in low-frequency components, as
compared to a voice of normal intensity.

Let us now consider the effects described by B. F. Miller. Accord-
ing to his paper, voice signals, after having traversed the compressor,
show a preponderance of high-frequency components. This effect is
explained as follows: Fig. 2C shows the average relation between
rms speech pressure per cycle and speech component frequency.
Miller obtained the data for this curve from a paper by Dunn and
White.4 It is observed that the presssures corresponding to the
lower-frequency (vowel) sounds of speech are very much greater than

-40

20 100 1.000 IOPOO 20,000

FREQUENCY IN CYCLES PER SECOND

FIG. 2C. Average relation between rms speech-pressure per cycle and speech
component frequency.

those corresponding to the higher-frequency (sibilant) sounds.
Now, if the amplification of the compressor is inversely proportional
to the instantaneous signal voltage, the amplification of the sibilants
will be higher than that of the vowels, since the instantaneous volt-
age corresponding to the sibilants is less than the instantaneous volt-
age corresponding to the vowels. To correct this condition, Miller
recommended an equalizer, inserted between the compressor output
terminals and the control rectifier, the frequency response of which
was designed to vary according to the inverse of the pressure-fre-
quency distribution of Fig. 2C.

Miller's fundamental and highly commendable paper deserves de-
tailed examination and review. In the first place, it assumes that
the frequency characteristic of the speech signal entering the com-

Oct. 1946 FACTORS GOVERNING FREQUENCY RESPONSE

309

pressor has approximately the characteristics of Fig. 2C. However,
in the recording of sound various equalizers are usually inserted
in the circuit ahead of the compressor. Some of the factors on
which the design of these equalizers are based tend to support this
curve, but some do not. Factors which support this curve are the
corrections applied for the frequency response of the microphone,
the reverberation characteristic of the sound stage, and the changed
voice characteristic of the actor speaking in quiet surroundings
(effort equalization). The factors which produce a change in the
curve of Fig. 2C are the compensations employed to correct for film

10
5
DB 0
-5
-10

.-^^*

— —

• — — ^^B^.

«=:

— ..

^ >

FILM PROCESSING £ RECORDER SLIT ATTENUATION

EFFECT PRODUCED BY REPRODUCING LEVEL BEIN<
5DB ABOVE RECORDING LEVEL

100 1.000

FREQUENCY IN CYCLES PER SECOND

FIG. 2D.

10,000 20JOOO

losses, for the recorder slit attenuation, and for the increase in low
frequencies occasioned by the reproducing level being (on the aver-
age) 5 db above normal speech level. These three factors are repre-
sented graphically on Fig. 2D. When they are taken into consid-
eration, the electrical signal characteristic of the human voice at
normal level just before entering the compressor assumes the solid
curve of Fig. 2E. Therefore, this curve, and not Fig. 2C, should be
used in the design of the compressor equalizer.

In the second place, the curve of Fig. 2C will not be the same for
extremely low or extremely loud speech and, therefore, if best results
are to be expected, it cannot be used indiscriminately for the deter-
mination of the compressor equalization characteristic.

310

M. RETTINGER AND K. SINGER

Vol 47, No. 4

In the third place, the question arises as to whether there are sig-
nificant differences between male and female voices, and also be-
tween individual male and individual female voices, which would
justify the use of the inverse of the curve of Fig. 2E for the frequency
characteristic of the compressor equalizer for normal speech levels.

It is the purpose in the following paragraphs to consider these fac-
tors in some detail.

Fig. 2F shows the extreme voice characteristics of men, and Fig.
2G the extreme voice characteristics of women. The data for these

-10

^ ^-^^*"

'~--|

'** ^

X ^

^s^^.^^

^>-^

"**•-

""• ^

**>•

'. AVERAGE RE
PER CYCLE

1LATION BETV
AND COMPO

</EEN RM S SPEECH - PRES
IENT FREQUENCY

URE

CHARACTERISTIC OF

SIGNAL ENTERING COMPRESSOR

1 1

100 1000

FREQUENCY IN CYCLES PER SECOND

FIG. 2E.

10,000 20,000

curves were obtained from the aforementioned paper.4 Fig. 2H
shows the envelope of the extreme voice characteristics of both men
and women. This limit curve was obtained by plotting the extreme
voice characteristics for both men and women, and then tracing the
envelope along the curves. The limits are most pronounced between
4000 and 5000 cycles, the difference amounting to 15 db.

None of the above objections, however, should make Miller's in-
trinsic argument invalid; namely, that, for faithful recording, ir-
respective of the (speech) characteristic of the signal entering the
compressor (not the microphone), the inverse of this characteristic
should be employed for the frequency response characteristic of the
compressor equalizer. If the speech to be recorded is extremely low

Oct. 1946 FACTORS GOVERNING FREQUENCY RESPONSE

311

or loud, or if its characteristic departs noticeably from the normal
characteristic of the speech (individual voice differences), the char-
acteristic of the compressor equalizer should then vary accordingly.
Extremely low and extremely loud speech has a characteristic the
inverse of which differs from the normal frequency response char-
acteristic of the compressor equalizer to such an extent that this
equalizer cannot completely correct for "selective distortion" at all
times. Fig. 2J indicates the changes in this equalizer characteristic
necessary to accommodate the extremes in voice levels, by showing

03
OHO

tf>

> -20
in

5
a

-30

-40
2*

EXTREME \/OICE CHARACTERISTICS OF MEN
AVERAGE VOICE CHARACTERISTIC OF MEN

~rf

/ /

' *^\-*

^*~^\!

•5^

•^ V,

100

000 10,000 20,0

FREOJUENCV IN CYCLES PER SECOND

FIG. 2F.

the average voice characteristic of soft, normal, and loud speech
"matched" at 1000 cycles.

For best results a variable equalizer should be used in the compres-
sor to accommodate all possible conditions: A compromise curve
such as that in Fig. 2E must be employed in the compressor. It will,
at least, cover a large number of cases, and will do much to eliminate
the hiss of many voices recorded on a recording channel that contains
a compressor.

It is sometimes claimed that, since the compressor is an energy-
actuated device, no "selective distortion" would occur if the timing
in the compressor were kept short enough. Within practical limits,
acceptable results can be obtained in this manner. However, the

312

M. RETTINGER AND K. SINGER

Vol 47, No. 4

argument does not appear to be completely true; for, even. if no
condenser had to be charged in the compressor, and the compressor
could be made to react instantaneously with the voltage changes,

DB 0

-10
w>

-30

-40

-">/

J •

">^\

^Sj

's*~*

EXTREME VOICE
OF WOMEN ~I_~

TH/

mACTERISTIC
RACTERISTIC

CHfi

AVERAGE VOICE

OF WOMEN

0 IOO 1,000 10,000 2C

FREQUENCY IN CYCLES PER SECOND

FIG. 2G.

"selective distortion" might still occur because the signal voltage of
the sibilant at the beginning of a word can very well be below the
"threshold of the compressor" (the volume level below which the
compressor will not act).

DB 0

OO IPOO 10,000 20,000

FREQUENCY IN CYCLES PER SECOND

FIG. 2H. Envelope of extreme voice characteristics of men and women.

The colloquialisms, "de-esser" or "de-icer," are sometimes applied
to this compressor equalizer.

The conclusions reached as a result of the foregoing observations
are:

Oct. 1946 FACTORS GOVERNING FREQUENCY RESPONSE

313

(1) An equalizer of the type described by B. F. Miller, to be inserted between
compressor output terminals and the control rectifier for controlling "selective
distortion," appears to have merit in recording sound on film.

(2) While the frequency characteristic of the equalizer should essentially con-
form to the inverse of the speech characteristics (Fig. 2C)f in practice the amount
of equalization in the channel ahead of the compressor, must be taken into con-
sideration in designing the equalizer.

(3) The consideration of the region in the speech characteristic below 1000
cycles appears important for very loud speech. It may therefore be advisable
to employ two insertion-loss characteristics in the compressor equalizer — one of
normal and one for declamatory speech.

•10

-10

/— s

OFT

• —

^--

'^1-

^<'

K^ N

-LO

NORMAL-

- NORMAL SPEECH _,

- SOFT SPEECH
• LOUD SPEECH

I0OO
FREQUENCY IN CYCLES PER SECOND

IO.OOO 20OOO

FIG. 2J. Average relation between rms speech-pressure per cycle and speech
component frequency; curves matched at 1000 cycles.

PART 3

Recommended Recording Channel Equalization.— In the early
days of sound recording, the placement of equalization in various
parts of the channel, mainly in the preamplifier and in the mixing
console, was determined chiefly by considerations of convenience and
immediate economy. For instance, by locating all the required high-
frequency compensation in the preamplifier (a method which could be
carried out with the aid of only a carbon resistor and a small con-
denser), there was provided not only a relatively inexpensive means of
equalization, but also one which was readily accessible and could be
changed quickly. However, as this method became an established
practice and the technique of sound recording increased in complex-
ity, the early economic advantages assumed a less obvious character.
For instance, when it was decided to change from a nondirectional

314 M. RETTINGERA ND K. SINGER Vol 47, No. 4

to a directional microphone, as during the recording of dialogue within
a small "boomy" set, it became necessary (in the absence of a handy
soldering iron) to interchange also the preamplifier. Many sound
trucks, therefore, carried two, and some trucks even three preamplifi-
ers to accommodate the various microphones.

Soon other complications arose. The recording of a comparatively
high-level signal, such as that obtained from a telephone line, called
for a separate high-frequency equalizer. The signal could not con-
veniently be applied to the input of the preamplifier because of hum
and general noise trouble resulting when the telephone signal was
sufficiently attenuated to meet the input-level requirements of the
preamplifier. The signal, therefore, had to be introduced after the
preamplifier. Because such occasions were not numerous, the sound
recording departments looked upon them as necessary evils which
could not easily be avoided.

With the introduction of high-fidelity monitoring and wax or
acetate disk recording equipments, further complications arose. Sat-
isfactory operation of these units could be obtained only if the equali-
zation for film processing losses, recorder slit attenuations, and other
factors was omitted from the signal. This meant, of course, that
suitable "decompensators" had to be employed.

If an abnormal amount of high-frequency compensation is re-
quired, it is often difficult to provide it in the preamplifier. For in-
stance, if a velocity microphone is desired for dialogue recording,
which may call for an equalization in excess of 10 db at 6000 cycles,
it is not easy to locate all this compensation in the preamplifier. The
difficulty lies not so much in increasing the high-frequency response
of the amplifier by 10 db at 6000 cycles, but in securing the most
desirable response characteristic between 1000 and 6000 cycles by any
simple, compact means of equalization.

Other disadvantages could be enumerated in connection with the
described method of placing the high-frequency equalization in the
preamplifier. These, no doubt, would vary with the modes and proc-
esses of the different studios, and hence might be more serious in
one studio than in another. However, it is generally agreed that the
assignment of the high-frequency compensation to the preamplifier is
no longer the economic nor the convenient solution it once was.

Location of the entire low-frequency (so-called "dialogue") equali-
zation in the mixing console is likewise an established outgrowth of
an early procedure of expediency. It was at one time considered

Oct. 1946 FACTORS GOVERNING FREQUENCY RESPONSE 315

satisfactory to allow the mixer on the set to control the general char-
acter, and to some extent also, the intelligibility, of the voices to be
recorded. With the advent of the compressor, however, some of the
functions of the mixer along this line became curtailed. Moreover,
by locating all dialogue equalization in the mixing console, which is
always ahead of the usual compressor in the circuit, a certain ac-
centuation of the low frequencies occurs with increasing voice level.
This accentuation is caused by the compressor action, which tends to
"flatten out" the signal characteristic by acting only on those signal
components which are above its threshold. If the low-frequency
components are severely attenuated ahead of the compressor, so that
they lie below the compressor threshold and hence must pass through
the instrument uncompressed, the resulting output from the compres-
sor will show an increased amount of low-frequency response. This
change in voice quality is particularly noticeable when an actor de-
livers a declamatory speech. Acoustical studies have proved that
declamatory speech contains fewer low-frequency components than
normal speech, just as intimate speech contains a preponderance of
low notes. Compressor action, however, tends to "wipe out" or to
eliminate this lack of lower registers during declamation, making the
voice sound less high-pitched. Whether this characteristic is bene-
ficial is at present a moot question. It has been said by some that it
results in greater carrying power of declamatory speech.

Having stated the problem of high-fidelity recording on film, we will
now discuss steps that may be taken in the direction of a satisfactory
solution.

Low-Frequency Equalization. — A consideration of the various
conditions which occur in the normal course of dialogue recording
favors the "splitting" of the low-frequency equalization and dis-
tributing it both before and after the compressor. It appears
logical that a mixer should be able to control to some extent the in-
telligibility of the recorded sound, as by attenuating some low-fre-
quency components in an unusually heavy voice or in speech uttered
in a small, boomy set. For this purpose the mixer should have at
his control in the mixing console a small amount of variable low-fre-
quency compensation, for example, 6 db at 100 cycles. If micro-
phones are changed during a production, a change in the required
equalization can just as easily be effected by a variable equalizer in-
serted after the compressor as by locating all the compensation in the
console. In the case of a sound truck, all that the mixer has to do is

316

M. RETTINGER AND K. SINGER

Vol 47, No. 4

to telephone to the recordist in the truck to adjust the dialogue equal-
izer in the rack to the required amount, a procedure which can also be
employed very easily in the case of a centralized recording system.
If this system (of "splitting" the low-frequency equalization) is
adopted, the following benefits will result :

(1) The low-frequency components will not be emphasized so much by the
compressor. To make this point quite clear, consider Fig. 3A, which represents a
somewhat simplified schematic of the prevalent conditions, and consists of a top
and a bottom figure. Curve A of the top figure shows the present type of dialogue
equalization, which attenuates 100 cycles by approximately 12.5 db. In the
normal voice-characteristic the 100-cycle component is about 2.5 db below the

-5

-ro

-15
-20

DB 0
-5
-10
-1 5
-20

s/s

^>B

,. — •

/ „

^•••^

^- *"*c

100 1,000

10,000

~^**

^"^^

— -

— • — "

\ 10. 5 D

~ "D

100 1.000

FREQUENCY IN CYCLES PER SECOND

FIG. 3A. Compressor characteristics.

10,000 20,000

500-cycle component, consequently the signal entering the compressor has more
nearly the characteristic indicated by Curve B. Curve C shows the frequency
characteristic of the compressor output signal when the device is set for a com-
pression ratio of 20 into 10 db. As shown on the figure, the 100-cycle component
is now only 7.5 db below the 500-cycle component, indicating that the 100-cycle
component was accentuated 12.5 — 7.5 = 5 db. In the bottom part of Fig. 3A,
Curve A represents half of the total dialogue equalization which is introduced
ahead of the compressor. Because in the normal voice characteristic the 100-
cycle component is about 2.5 db below the 500-cycle component, the signal
entering the compressor has more nearly the characteristic indicated by
Curve B. Curve C shows the frequency characteristic of the compressor
output signal when the compressor is set for a compression ratio of 20 into 10
db. Since this signal will be subjected to the effect of the second part
of the dialogue equalization placed after the compressor, the resulting signal
will have more nearly the characteristic of Curve D. As shown on the figure,

Oct. 1946 FACTORS GOVERNING FREQUENCY RESPONSE

317

318

M. RETTINGER AND K. SINGER

Vol 47, No. 4

the 100-cycle component is now 10.5 db below the 500-cycle component, indi-
cating that the 100-cycle component was accentuated only 12.5 — 10.5 = 2 db, as
against 5 db when all of the dialogue equalization was placed ahead of the com-
pressor.

(2) Incidental low-frequency set noises, stage rumble, etc., will not be ac-
centuated so much when part of the dialogue equalization is placed ahead of
and part after the compressor, for the same reason as described above.

High-Frequency Equalization. — The disadvantages of locating
all the high-frequency equalization in the preamplifier have been
discussed in some detail. From a consideration of these factors,

FIG. 4A. Compressor schematic.

it becomes evident that a much more desirable high-frequency
equalization system consists in locating the so-called fixed com-
pensation (film losses, recording slit attenuation) after the compres-
sor and to introduce the so-called microphone compensation ahead
of the compressor.

Limitation. — When changing from one kind of microphone to an-
other kind in the course of recording, it is still necessary to change the
preamplifier. This means that a sound truck may have to carry
more than one preamplifier if more than one kind of microphone is to
be employed.

Oct. 1946 FACTORS GOVERNING FREQUENCY RESPONSE 319

Fig. 3B shows the recommended arrangement of equipment items
and equalizer characteristics for a film recording channel which ful-
fills the requirements previously discussed.

PART 4

Experimental Work. — A recording channel was set up in accord-
ance with the block schematic illustrated in Fig. 3B. This block
schematic shows the recommended arrangement of equipment items
and their frequency characteristic. The combination results in a

COMPfffSSIO. J RAT/0

COMPRESSOR RESPO* SE

RECTIFIER

EQUAL ZER

20: 10

EQ1ALIZLR

WOO

FP£QU£NCY IN CYCLES PERSKOM)

FIG. 4B. Compressor frequency characteristic.

variable-area film recording channel which conforms to the theo-
retical requirements derived in the preceding parts of this paper.
All tests were made in the RCA Hollywood Film Recording Studio,
as the equipment available there provided great flexibility of fre-
quency characteristic, level relationships between various amplifiers,
change of compression ratio, and other equally valuable considera-
tions. Instead of the mixer, which is normally employed in studio
channels, the rerecording console with its array of equalizers was
used.

It should be understood that the tests to be described later were
made for the purpose of proving or disproving the validity of the
theoretical conclusions arrived at in the preceding text, without re-

320

M. RETTINGER AND K. SINGER

Vol 47, No. 4

gard for the practical limitations which might be encountered in
the studios because of presently available equipment or operational
technique. Live talent, male and female, was used as a source of
sound, and a standard recording channel was used as a basis of
comparison.

0
-2
-4

\-6

' -14
-16
-18
-20
-22
-24
-26
-2A

UPIH

ISSH

RAT,

20:

10 ,

•>B

UNC

OMP

<?£S.

"ED

SS£1

y£

;

-6

0-5

8 -S

6 -5

4 -5

2 ~5

0 -4

3-4

6 -<

4 -<

2 -4

0 -3

3-3

6 -3

i -3

2 -3

1 -2

8-2

INPUT LEVEL- DB
FIG. 4C. Compression characteristic of compressor.

Let us consider each equipment item used in this experimental
channel and discuss the differences between it and corresponding
items used in recording studios.

A unidirectional microphone connected for an output impedance
of 500 ohms was fed to a microphone amplifier of 250 ohm nominal
input impedance. This microphone amplifier had been modified
for a frequency response as shown on Fig. 3B. Feeding the 500-ohm
microphone output into a nominal 250-ohm input results in 3 db more
output from the microphone with a negligible loss of high-frequency
response, and a 3-db improvement of signal-to-noise ratio.

Oct. 1946 FACTORS GOVERNING FREQUENCY RESPONSE

321

A variable rcrecording compensator was set to duplicate the vari-
able equalizer characteristic shown on Fig. 3B to take care of boomy
sets, heavy voices, etc.

A compressor was modified to contain the circuit shown on Fig.
4 A. This was done in order to duplicate the compressor fre-
quency characteristic illustrated on Fig. 4B to eliminate the "rela-
tive spectral energy distortion."

COb

'PRL

SSH

RAJ

INCC

IMP

JES.

*ED

x^"

MPf

CSS

-6

0 -5

8 -5

6 -S

4-5

2 -5

0 -<

8-4

6 -4

4 -4

2 4

0 -3

8 -3

6 -3

4 -3

2 -3

o -;

8 -:

INPUT IB/EL- DB

FIG. 4D. Compression characteristic of compressor.

-16

-id

-20
-22
-24
-26
-28

Compression characteristics for compression ratios of 20 into 10
db and 30 into 20 db are shown on Fig. 4C and 4D.

A film loss and dialogue equalizer, whose schematic and frequency
characteristic are in accordance with the schematic shown on Fig.
4E, was used.

A 6000-cycle low-pass filter was employed, together with an 80-
cycle high-pass filter. Both filters are standard equipment items and
were not changed.

322

M. RETTINGER AND K. SINGER

Vol 47, No. 4

In order not to conceal compressor action by the effects of noise
reduction (clipping of the beginning of sounds because of the normal
noise-reduction circuit timing), it was decided to make all tests with-
out noise reduction.

An RCA type MI-3233B bridging amplifier was employed to drive
the recording galvanometer.

The regular rerecording monitoring system, consisting of a suitable
monitoring decompensator, 50-w power amplifier, and a two-way
speaker system, was used for monitoring.

-12

FIG. 4E.

/poo
CLtS KR SECOND

Film loss and dialogue equalizer characteristic.

10.000

Results obtained from this experimental channel were compared
against a standard recording channel whose block schematic is shown
in Fig. 4F. The block schematic is self-explanatory.

A series of tests was made during which a male and a female voice
(live talent) repeated the same dialogue under varying conditions.
These variations in conditions cover changes of compressor acting
time from 0.7 to 10 milliseconds and compressor release time from
0.1 sec to 0.5 sec, change of compression ratio between 20 into 10
db and 30 into 20 db, and use or omission of the de-esser. These
tests were divided into two groups, namely, listening tests and
recording tests. Listening tests were made for all conditions. Dur-
ing these tests the performance of the experimental channel was com-
pared against the standard channel. An instantaneous comparison

Oct. 1946 FACTORS GOVERNING FREQUENCY RESPONSE

323

was possible by means of a suitable switching arrangement. During
the listening tests the fact was definitely established that an unmis-
takable improvement in quality was obtained when the de-esser
circuit was used in the compressor. Consequently, no recording
tests without de-esser have been made. Actual recordings under all
conditions with de-esser and through the standard channel as shown
on Fig. 4F were made. Particular attention was paid during these
tests to maintaining equal modulation level on the film for all varying
conditions.

FIG. 4F. Standard recording channel.

Prints of these recording tests were run through the RCA re-
recording channel before a group of listeners, who voted for what they
thought sounded best. These listening tests were spaced over
several days in order to avoid fatigue of the observers. Six of the
original eleven tests were finally eliminated. Further listening tests
brought the unanimous opinion that test two sounded best. The
conditions under which these test recordings were made were as fol-
lows:

Compression ratio

Compressor acting time.
Compressor release time.
De-esser. .

20 into 10
0.0007 sec
0.5 sec
in circuit

In order to illustrate compressor acting time, Figs. 5 and 6 are
shown here. Fig. 5 shows compressor action when the level of a
5000-cycle tone is suddenly increased by 20 db. The compressor
used in the standard recording channel had an acting time of about
0.002 sec and a release time of about 0.5 sec which corresponds to the
timing trace shown on the top.

324

M. RETTINGER AND K. SINGER

Vol 47, No. 4

An acting time of 0.0007 sec and a release time of about 0.5 sec
gave the best results in the compressor in the experimental channel.
The second trace from the top corresponds to this timing. The bot-
tom trace shown on Fig. 5 corresponds to an acting time of 0.0002
sec and a release time of 0.5 sec.

Recording tests under these conditions were not included in this
investigation, as earlier experiments covering the use of such ultra-

START

C\R.CL)\T

START

T\M\NG, C\R.CU\T

0. 05 MFD., R.- Z.5

C\RCO\T

FIG. 5.

Operating time characteristics of MI-10206 electronic mixer; com-
pression ratio 20 db into 10 db (/ = 5000 cycles).

fast compressor timing have shown that it is impractical for produc-
tion conditions. In order to utilize such fast acting time, it would be
necessary to maintain very accurate tube balance in the compressor
which, owing to limitation of tube and power supply stability and
the pressure under which motion pictures are made, is not possible.
Fig. 6 shows the appearance of a speech recording made with the
different timing circuit constants shown on Fig. 5. The overshoot-
ing with the slow acting time of 2 milliseconds is quite evident.

PART 5

Conclusions. — From the results of the foregoing tests, the fol-
lowing conclusions have been formed :

Oct. 1946 FACTORS GOVERNING FREQUENCY RESPONSE

325

(1) When comparing the experimental channel, set up in conformity with the
block diagram on Fig. 3B with the standard channel, an improvement in sound
quality is obtained, even \\lun the compressor in the experimental channel is
operated without de-esser, having an acting time of 2 millisec and release time
of 0.5 sec. This improvement results from placing part of the dialogue equaliza-
tion as well as part of the high frequency equalization after the compressor, which
arrangement reduces the "wiping out" action of equalization by the compressor.
This, observation was made when comparing conditions 1C and ID against the
standard channel.

HO COMPRESSVOU

CVROJIT COKVTAUTS :

MFD.. R=.5W\LG.

QR.CO\T

O.05MFD.. R

C\RCU\T

R-

FIG. 6. Operating time characteristics of MI-10206 electronic mixer; com-
pression ratio 20 db into 10 db (speech).

(2} Adding the de-esser circuit to the compressor in the experimental channel
eliminates to a large extent the hisses and accentuated sibilants which can be
noticed without it.

(5) Speeding up the acting time of the compressor in the experimental channel
to 0.0007 sec reduces sibilant accentuation still further. The sound quality
thus obtained is smooth and natural and compares to what a listener would hear
on a recording stage if he were located in the position of the microphone.

(4) It has definitely been established that the addition of the de-esser circuit
and speeding up in acting time together result in the best recording quality.

(5) When the compressor acting time is speeded up to 0.0007 sec, the compres-
sor release time should be slowed down to 0.5 sec to maintain adequate filtering of
the gain change control voltages. If adequate filtering is not maintained, repro-
duction becomes rough.

326 M. RETTINGER AND K. SINGER

(6) A close qualitative agreement between theoretical analysis and practical
test results has been established. Quantitatively, the tests differed from the
theoretical optimum only in this 'respect that because of practical limitations it
was not possible to obtain exactly the calculated sibilant de-accentuation in the
compressor. (See Fig. 4B.) This deviation, however, was rather small, and
cannot influence the qualitative trend.

^(7) An analysis has been presented which is applicable to all types of record-
ing channels, which may use any of the available recording media as, for instance,
variable-area or variable-density on film, lateral or vertical cut on acetate or wax,
steel- tape recording, or any other recording media.

REFERENCES

1 COOK, E. D.r "The Aperture Effect," /. Soc. Mot. Pict. Eng., XIV, 6 (June
1930), p. 650.

2 LOYE, D. P., AND MORGAN, K. F.: "Sound Pictuie Recording and Repro-
ducing Characteristics," /. Soc. Mot. Pict. Eng., XXXII, 6 (June 1939), p. 631.

3 MILLER, B. F.: "Elimination of Relative Spectral Energy Distortion in
Electronic Compressors," /. Soc. Mot. Pict. Eng., XXXIX, 5 (Nov. 1942), p. 317.

4 DUNN AND WHITE: "Statistical Measurements of Conversational Speech,"
/. Acous. Soc. Am., (Jan. 1940), p. 278.

WIDE-RANGE LOUDSPEAKER DEVELOPMENTS*
H. F. OLSON** AND J. PRESTON**

Summary. — Two unit direct-radiator loudspeakers may be constructed in
many different ways. In order to determine some of the characteristics, a number
of experimental designs were built and tested. As a result of these experiments, it
appeared that the duo-cone loudspeaker, consisting of two coaxial, congruent, sepa-
rately driven cones, possessed many constructional, theoretical, and experimental
advantages. Consequently, a detailed theoretical and experimental investigation of
the duo-cone loud-speaker was carried out to determine the optimum values for the con-
stants of the system from the standpoint of the following characteristics: pressure
response, directional pattern, distortion, and transient response. The results of
these investigations are included.

Introduction. — The almost universal use of the direct-radiator
loudspeaker is a result of its simplicity of construction, small space
requirements and the relatively uniform response frequency char-
acteristic. Uniform response over a moderate frequency band may
be obtained with any simple direct-radiator loudspeaker. How-
ever, reproduction over a wide frequency range is restricted by
practical limitations. The portion of the speech range required for
intelligibility falls in the midaudio band. The range of the funda-
mental frequencies of most horn, reed, and string instruments also
falls within this band. This is rather fortunate because it is a very
simple task to build mechanical and acoustical vibrating systems to
cover only this midfrequency band. The two extreme ends of the
audio-frequency band are the most difficult to reproduce with effi-
ciency comparable to the midfrequency range. Inefficiency at the
low frequencies is primarily caused by a small radiation resistance.
Inefficiency at the high frequencies is primarily caused by large mass
reactance.

The volume range is another factor involved in sound reproduc-
tion. In the middle frequency band the ear has a volume range of a
million to one in pressure, or a trillion to one in energy. To build

* Reprinted from RCA Review, VII, 2 (June 1946), p. 155.
** Research Department, RCA Laboratories Division, Princeton, N. J.

327

328

H. F. OLSON AND J. PRESTON

Vol 47, No. 4

linear reproducing apparatus for this tremendous range is practically
impossible today. As a matter of fact, it is not practical to reproduce
the volume range of all musical instruments.

An increase in the volume and frequency ranges of the loudspeaker
multiplies the problems connected with obtaining the proper direc-
tional pattern, low nonlinear distortion and suitable transient re-
sponse. The directional characteristics of the conventional direct-

INPUT

FRONT VIEW SECTION A-A"

FIG. 1. A coplaner combination of low- and high-fre-
quency direct-radiator loudspeaker units.

radiator loudspeaker are quite adequate for the frequency range of the
present-day broadcast receivers. However, when the high-frequency
range is increased by one to two octaves, the directional pattern be-
comes quite narrow and some consideration must be given to this
problem. The problem of nonlinear distortion is multiplied several'
times by the addition of one or two octaves. The additional volume
range, of course, complicates the problem of nonlinear distortion. It
has been found that poor transient response is not objectionable in the
case of a loudspeaker with a limited frequency range — in some cases
it actually enhances the reproduction. However, a wide-range high-

Oct. 1946

WIDE-RANGE LOUDSPEAKER DEVELOPMENTS

329

fidelity loudspeaker should exhibit good transient response. From
the above discussion it will be seen that additional volume and fre-
quency ranges increase the complexity of the technical problems in
loudspeaker design and manufacture.

Wide frequency range, low distortion loudspeakers are required foi
monitoring in radio and television broadcasting, phonograph, and
sound motion picture recording and high quality sound systems. The
direct-radiator loudspeaker is particularly suited for these applications
because the acoustic power required is relatively low and the space re-
quirements rather limited.

It is the purpose of this paper to describe the following: the de-
velopment work on a wide-range direct-radiator loudspeaker; the
performance of an experimental duo-cone direct-radiator loudspeaker.

FIG. 2. The directional characteristics in the overlap
region of the coplaner combination of low- and high-fre-
quency direct-radiator loudspeaker units shown in Fig. 1.

Two-Unit Loudspeakers. — Two-unit loudspeakers may be con-
structed in many different ways. In order to determine some of the
characteristics, a number of experimental designs were built and
tested. Some of the theoretical and practical advantages and dis-
advantages will be described.

The simplest two-unit, direct-radiator loudspeaker consists of a
small cone unit and a large cone mounted on the front face of a flat
baffle as shown in Fig. 1. If the response covers the frequency range
from 40 to 15,000 cycles the natural overlap region will be somewhere
between 1000 and 2000 cycles. A system of the type depicted in
Fig. 1 consists of a cone 15 in. in diameter in the low-frequency unit
and 2 in. in diameter in the high-frequency unit. Owing to the mount-
ing arrangements of the two units the spacing between the two units

330

H. F. OLSON AND J. PRESTON

Vol 47, No. 4

in the baffle was 15 in. The middle of the overlap region was placed
at 1500 cycles. The directional pattern at 1500 cycles is shown in
Fig. 2. Complete destructive interference occurs when the distance
between the two units is one-half wavelength and odd multiples of
one-half wavelength. The type of directional characteristic shown
in Fig. 2 introduces frequency discrimination for points removed from
the axis in a very important frequency band.

In the next experiment, the high-frequency loudspeaker unit was
placed coaxially inside the low-frequency unit as shown in Fig. 3.

INPUT

FRONT VIEW SECTION A-AV

FIG. 3. A coaxial combination low- and high-frequency
direct-radiator loudspeaker units.

This construction improved the directional pattern in the overlap
region. However, the sound which was diffracted around the high-
frequency unit and reflected from the low-frequency cone interfered
with the direct radiation. The result of this process is a nonuniform
response frequency characteristic as shown in Fig. 4.

In the next experiment a small cellular horn loudspeaker was used
as the high-frequency loudspeaker. The horn loudspeaker was placed
in a baffle above the low-frequency unit as shown in Fig. 5. This
system exhibited the same type of directional pattern in the overlap
frequency region as the system of Fig. 1.

Oct. 1946

WIDE-RANGE LOUDSPEAKER DEVELOPMENTS

331

Following the above experiment the cellular horn loudspeaker was
arranged coaxially with respect to the low-frequency loudspeaker as
shown in Fig. 6. This system exhibited diffraction characteristics
similar to those of Fig. 3. There was an additional factor; namely,
the source of the high-frequency sound was several inches behind the
source of the low frequency sound. This path amounts to almost a
wavelength in the overlap frequency region. This is an undesirable
feature, particularly, in the case of the reproduction of transient
sounds.

400

rooo

10000 20000

FREQUENCY IN CYCLES PER SECOND

FIG. 4. The response frequency characteristic of the
high-frequency unit of the coaxial combination of low- and
high-frequency direct-radiator loudspeaker units shown
in Fig. 3.

From the above experiments it appeared undesirable to place the
high-frequency unit in front of the low-frequency unit. This feature
can be obviated by making the pole for the low-frequency unit a
portion of the high-frequency horn as shown in Fig. 7. The response
frequency characteristic obtained on this system was smooth. In
addition, the directional pattern was acceptable particularly when a
wide angle low-frequency cone was used. The difference in path
length between the source of the low- and high-frequency units was
still an undesirable feature.

There is another problem when a horn and direct-radiator loud-
speaker are combined; namely, the difference in efficiency. The
efficiency of a horn loudspeaker is from 10 to 20 db greater than the
direct-radiator loudspeaker. This means that an attenuation net-
work must be used with the horn unit to obtain comparable efficiencies
and uniform response from the combination of the two units.

332

H. F. OLSON AND J. PRESTON

Vol 47, No. 4

In the next experiment two cone loudspeaker units were com-
bined so that the large cone was a continuation of the small cone as
shown in Fig. 8. This system has been termed a duo-cone loud-
speaker. The combination system shown in Fig. 8 eliminates the
path difference factor because in the overlap region the two cones vi-
brate together as a single cone.

H.F.UNIT

. *A"
FRONT VIEW

INPUT

SECTION A-A

FIG. 5. A coplaner combination of a direct-radiator
low-frequency loudspeaker unit with a cellular horn high-
frequency loudspeaker unit.

As a result of the above experiments it appeared that the duo-cone
loudspeaker possessed many constructional, theoretical, and experi-
mental advantages. In view of this, it was decided to make a detailed
investigation of the duo-cone loudspeaker. It is the purpose of the
sections which foilow to describe in detail some of the characteristics
of the duo-cone loudspeaker.

Oct. 1946

WIDE-RANGE LOUDSPEAKER DEVELOPMENTS

333

Theoretical Considerations. — The performance of a direct-radi-
ator loudspeaker may be obtained from theoretical considerations.1
Theoretical investigations are useful in determining the dimen-
sions of the units, the masses of the voice coils and cones, the air gap
flux, the fundamental resonant frequencies, and other relevant fac-
tors. Proper evaluation of these factors is important in obtaining
a scientifically co-ordinated loudspeaker system. It is the purpose of
this section to outline theoretically the action of the duo-cone loud-
speaker consisting of two congruent, coaxial direct-radiator loud-
speaker systems.

L.F.UNIT

/ H.F.UNIT

'A"
FRONT VIEW

INPUT

SECTION 'A-A'

FIG. 6. A coaxial combination of a direct-radiator low-
frequency loudspeaker unit with a cellular horn high-fre-
quency loudspeaker.

A cross-sectional view, voice coil circuit, and the mechanical circuit
of the low frequency unit of the duo-cone loudspeaker are shown in
Fig. 9. The total mechanical impedance of the vibrating system at
the voice coil is

ZMT =

+ jo>mc + J

(1}

where

rms = mechanical resistance of the suspension system, in mechanical

ohms,

?M A = mechanical resistance of the air load, in mechanical ohms,
me = mass of the cone and the voice coil, in grams,

334

H. F. OLSON AND J. PRESTON

Vol47, No. 4

mA = mass of the air load, in grains, and

CMS = compliance of the suspension system, in centimeters per dyne.

Eq (1) may be written as follows :

ZM T = rMS + fMA + JXMC '+ JXMA — JXMS

(2)

where rMs = mechanical resistance of the suspension system, in mechanical

ohms,

TM A = mechanical resistance of the air load, in mechanical ohms,
XMC = u>mc = mechanical reactance of the voice coil and cone,

L.F.UNIT

H.F.UNIT

FRONT VIEW

INPUT

SECTION *A-A

FIG. 7. A coaxial combination of a direct-radiator low-
frequency loudspeaker unit with a horn high-frequency
loudspeaker unit.

= urn A = mechanical reactance of the air load, in mechanical ohms,
and

XMS = ~^ — = mechanical reactance of the suspension system, in me-
chanical ohms.

The mechanical resistance and mechanical reactance of the air load
may be obtained from Fig. 10.

The motional impedance,2 in abohms, of the mechanical system is

ZEM

ZMT

(3)

Oct. 1946 WIDE-RANGE LOUDSPEAKER DEVELOPMENTS

335

INPUT

FRONT VIEW

SECTION A-A'

FIG. 8. A congruent coaxial combination of low-frequency
and high-frequency direct-radiator loudspeaker units.

INPUT

FIELD
STRUCTUREU^CMSrMS

[•vw^nnflp-vw*!

BAFFLE

CROSS SECTIONAL
VIEW

VOICE COIL
ELECTRICAL CIRCUIT

MECHANICAL CIRCUIT

OF THE
MECHANICAL SYSTEM

FIG. 9. Cross-sectional view, electrical circuit, and mechanical circuit
of the low-frequency portion of a duo-cone loudspeaker. In the electrical
circuit: TEG, the internal electrical resistance of the generator; TEC and L,
the electrical resistance and inductance of the voice coil ; ZEM, the electrical
motional impedance, e, the voltage of the electrical generator. In the me-
chanical circuit: me, the mass of the cone and voice coil; CMS, the compli-
ance of the suspension system ; TMS, the mechanical resistance of the suspen-
sion system ; WA and TMA, the mass and mechanical resistance of the air load ;
/AT, the mechanomotive force in the voice coil.

336

io6.

H. F. OLSON AND J. PRESTON

, ,, ,,, ,, . io6

Vol 47, No. 4

io5

I'0*
o

to2

•o2 2

lO3'

IO4

FREQUENCY

roo

O.I

V 2 V 2

FREQUENCY

EFFICIENCY
2 - 5 1

-v

11 ^ •

8 * 8

o22 4 V2 4 V2

FREQUENCY

FIG. 10. Mechanical impedance and efficiency frequency
characteristics of the low- and high-frequency units of the duo-
cone loudspeaker. In the low-frequency unit: XMC, the me-
chanical reactance of the cone and coil; XMA and rMA, the me-
chanical reactance and mechanical resistance of the air load;
XMS, the mechanical reactance of the suspension system. In
the high-frequency unit : XMCI and XMCZ, the mechanical re-
actances of the voice coil and cone: XMA and TMA, the me-
chanical reactance and mechanical resistance of the air load;
XMS, the mechanical reactance of the suspension system: XMC,
the mechanical reactance of the air cavity behind the cone.

where B = flux density in air gap, in gausses,

I = length of the conductor in the voice coil, in centimeters, and
ZMT = mechanical impedance of the mechanical system, in mechanical
ohms.

The efficiency of the loudspeaker is the ratio of the sound power
output to the electrical input. The efficiency may be obtained from
the voice coil circuit of Fig. 9 and expressed as follows :

Oct. 1946 WIDE-RANGE LOUDSPEAKER DEVELOPMENTS 337

TER X 100% (4)

TEC +

where TER = component of the motional resistance caused by the radiation

of sound, in abohms,

TEM = total motional resistance, in abohms, and
TEC = damped resistance of the voice coil, in abohms.

The components rER and rEM may be obtained from Eqs (1), (2),
and (3).

From Eqs (2), (3), and (4) the efficiency, in per cent, of the loud-
speaker is

(£/)2 (rus -

X 100 (5)

Above the fundamental resonant frequency the mechanical react-
ance caused by the suspension system is small compared to the me-
chanical reactance of the cone and coil. Since rMA is small compared
to XMA and XM c, Eq (5) becomes

(Biy rMA x loa

TEC (XMA. + XMC)Z 109

In terms of the resistivity and density of the voice coil, Eq (6) be-
comes,

R2 <r,r ,m.

X 100 (7)

pKr(xMA -

where mi = mass of the voice coil, in grams,

p = density of the voice coil conductor, in grams per cc, and
Kr = resistivity of the voice coil conductor, in ohms per cc.

The relation between the efficiency and the ratio of the mass of the
coil to the mass of the cone and air load may be obtained from Eq.
(7). The maximum efficiency occurs when the mass of the cone is
equal to the mass of the coil.

The cone diameter of the low-frequency unit used in the duo-cone
loudspeaker is 15 in. The mechanical resistance and reactance char-
acteristics of the elements of the vibrating systems are shown in Fig.
10. For the air load on the large cone it is assumed that it is mounted
in an infinite baffle.

The efficiency in which all the elements of the vibrating system are
included may be obtained from Eq (5) . The resistance rMC caused by

338 H. F. OLSON AND J. PRESTON Vol 47, No. 4

suspension system is also a factor in the efficiency in the region of
resonance. The mechanical resistance, rMs> of the suspension system
of the large cone is 2400 mechanical ohms.

The efficiency characteristic is shown in Fig. 10. It will be noted
that the efficiency is higher at the resonant frequency. However, when
coupled to a vacuum tube driving system the motional impedance is
also increased which produces the power input to the voice coil. For
this reason, the response is not accentuated to the degree depicted by
the peak in the efficiency characteristic. It will be seen that the effi-

PEC I" PEC ^Cl mC2 CMS°MCPM» m* r

_^vwTnnr^wv^ p-ryirinYnnrrHMH^

e [ZE^ fM ZTCu.

i V > I .

VOICE COIL MECHANICAL CIRCUIT

OF THE
ELECTRICAL CIRCUIT MECHANICAL SYSTEM

CROSS SECTIONAL
VIEW

FIG. 11. Cross-sectional view, electrical circuit and mechanical circuit
of the high-frequency portion of a duo-cone loudspeaker. In the electrical
circuit: TEG, the internal electrical resistance of the generator; TEC and L,
the electrical resistance and inductance of the voice coil; ZEM, the electrical
motional impedance, e, the voltage of the electrical generator. In the me-
chanical circuit: ma, the mass of the voice coil; met, the mass of the cone;
CMS, the compliance of the suspension system ; r MS, the mechanical resistance
of the suspension system; mA and TMA, the mass and mechanical resistance
of the air load; CMC, the compliance of the air cavity behind the cone; CM\,
the compliance between the voice coil and cone; fM, the mechanomotive
force in the voice coil.

ciency decreases very rapidly below the resonant frequency. There-
fore, in a direct-radiator loudspeaker the response limit at the low-
frequency end of the frequency range is determined by the resonant
frequency of the system.

The motional impedance of a dynamic loudspeaker is given by
Eq (3). The normal impedance, in abohms, of voice coil is given by

ZEN = ZEM + ZED (#)

where ZEM = motional electrical impedance, in abohms, and

ZED = electrical impedance of the voice coil in the absence of motion, that
is blocked, in abohms.

Oct. 1946 WIDE-RANGE LOUDSPEAKER DEVELOPMENTS 339

A cross-sectional view, voice coil circuit, and mechanical circuit of
the high-frequency unit of the duo-cone loudspeaker is shown in
Fig. 11. In the case of the high-frequency unit there are two addi-
tional compliances as contrasted to the low-frequency unit, namely,
the compliance of the chamber behind the cone and the compliance
between the coil and cone. The mechanical impedance at the voice
coil, assuming the latter compliance, to be zero is given by

where rMS = mechanical resistance of the suspension system, in mechanical

ohms

TMA = mechanical resistance of the air load, in mechanical ohms,
mci = mass of the voice coil, in grams
rwc2 = mass of the cone, in grams
MA = mass of the air load, in grams

CMS — compliance of the suspension system, in centimeters per dyne, and
CMC = compliance of the air chamber behind the cone, in centimeters per
dyne.

The efficiency, from Eqs (3), (4), and (9), is

(rMc + TMA) + rEo [(TMC + rMA)2 + XMA + XMCI +

X 100 (10}

where TMS = mechanical resistance of the suspension system, in mechanical

ohms

TMA = mechanical resistance of the air load, in mechanical ohms
XMA = com A = mechanical reactance of the air load, in mechanical

ohms
XMCI = wwci = mechanical reactance of the voice coil, in mechanical

ohms
XMCZ = cornea = mechanical reactance of the cone, in mechanical ohms

1
x.\fs — — — = mechanical reactance of the suspension system, in me-

chanical ohms, and

XMC = — — — = mechanical reactance of the air chamber behind the
cone, in mechanical ohms.

The cone diameter of the high-frequency unit used in the duo-cone
loudspeaker , is 2 in. The mechanical resistance and reactance char-
acteristics of the elements of the vibrating system are shown in Fig.
10. For the air load upon the cone it is assumed that the large cone

340

H. F. OLSON AND J. PRESTON

Vol 47, No. 4

forms a conical horn. The mechanical resistance of the suspension
system is 3600 mechanical ohms. It will be seen that mechanical
reactance caused by the air chamber behind the cone is three times the

jdMm

FIG. 12. A photograph of a duo-cone direct-radiator loudspeaker.

mechanical reactance resulting from the suspension system. There-
fore, in the range where the compliances are the controlling mechani-
cal reactances the compliance caused by the air chamber is the con-
trolling compliance. This expedient reduces the distortion caused by

100 1000 10000 20000

FREQUENCY IN CYCLES PER SECOND

FIG. 13. Response frequency characteristics of the low- and high-
frequency units of the duo-cone loudspeaker mounted in a large
baffle.

a nonlinearity of the suspension system. The efficiency characteristic
is shown in Fig. 10. It will be seen that the efficiency falls off about
10,000 cycles. This results from the fact that the system is mass con-
trolled and the radiation resistance does not increase as the square of

Oct. 1946

WIDE-RANGE LOUDSPEAKER DEVELOPMENTS

341

the frequency above 10,000 cycles. By introducing a compliance,
Cji/i, between the voice coil and cone the effective mass of the system
is reduced and uniform efficiency is maintained to 15,000 cycles as
shown by the dotted efficiency characteristic of Fig. 10.

The combination of the low- and high-frequency units as outlined
should yield uniform output from 30 to 15,000 cycles. A photograph
of an experimental duo-cone loudspeaker having the constants given
in this section is shown in Fig. 12.

Response Frequency Characteristics. — The measured response
frequency characteristics of the low- and high-frequency units of the
duo-cone loudspeaker mounted in a large flat baffle are shown in

100

1000

10000 20000

FREQUENCY IN CYCLES PER SECOND

FIG. 14. The electrical impedance frequency characteristics of
the low- and high-frequency units of the duo-cone direct-radiator
loudspeaker.

Fig. 13. These characteristics are in substantial agreement with
the efficiency characteristics of Fig. 10. The response frequency
characteristics in a phase inverter cabinet will be considered in a
later section.

Cross-Over Network. — The cross-over network is an important
consideration in a direct-radiator loudspeaker. In the design of any
two-unit loudspeaker, when there is considerable path length be-
tween the two units, a relatively sharp cross-over network is re-
quired in order to prevent destructive interference between the two
units in the cross-over region. In the duo-cone loudspeaker, since
the large cone is a continuation of the small cone, the cross-over fre-
quency range need not be confined to a narrow band because the two
cones vibrate as a single cone in this frequency region. This fact
makes it possible to use a very simple cross-over network. The elec-

342

H. F. OLSON AND J. PRESTON

Vol 47, No. 4

trical impedance characteristics of the low- and high-frequency units
of the duo-cone loudspeaker are shown in Fig. 14. The inductance
of the large low-frequency voice coil is large. As a consequence, it
is not necessary to use an inductance in series with the low-frequency

CAI

FRONT VIEW

Mi CAI PAI

SECTION A-A

ACOUSTIC CIRCUIT

FIG. 15. Front and sectional views and the acoustic cir-
cuit of the acoustic phase inverter used with the duo-cone loud-
speaker. In the acoustic circuit: MI, CAI, and TAI, the inert-
ance, acoustic capacitance, and acoustic resistance of air load
and cone and coil of the loudspeaker unit ; M% and TAZ, the in-
ertance and acoustic resistance of the port; CAZ, the acoustic
capacitance of the cabinet volume.

unit to reduce the current at the high frequencies. The only external
element required for the cross-over network is a condenser in series
with the high-frequency unit to limit the current through the high
frequency unit at the low frequencies. The cross-over frequency in
this system extends over about two octaves. However, as pointed

Oct. 1946 WIDE-RANGE LOUDSPEAKER DEVELOPMENTS 343

out before, this is not objectionable because in the overlap region the
two cones vibrate as a single cone.

Phase Inverter with a Variable Port. — The preceding considera-
tions have been concerned with the performance of the duo-cone

FIG. 16. Photograph of the duo-cone direct-radiator
loudspeaker mounted in a phase inverter cabinet with the
variable port with the grill removed.

loudspeaker operating in a large flat baffle. The large flat baffle
is not a practical mounting arrangement for general applications.
A cabinet is the conventional housing for direct-radiator loudspeaker
systems. It is the purpose of this section to consider a phase inverter-
type cabinet suitable for the duo-cone loudspeaker.

344 H. F. OLSON AND J. PRESTON Vol 47, No. 4

The term "phase inverter loudspeaker" is used to designate a system
consisting of a loudspeaker mechanism mounted in a closed cabinet
with an opening or port which augments the low-frequency response
by the addition of the sound radiated from the port. The reason that

FIG. 17. Photograph of the complete duo-cone loud-
speaker.

the addition of the port augments the low-frequency response is be-
cause the particle velocity of the air in the port is in phase with the
velocity of the cone.

The amount of low-frequency accentuation required for a particu-
lar condition of reproduction depends upon the program material, the

Oct. 1946

WIDE-RANGE LOUDSPEAKER DEVELOPMENTS

345

room in which the sound is reproduced, etc. Therefore, it is desirable
to provide a. variable means for adjusting the low-frequency response
to a loudspeaker. It is the purpose of this section to describe a phase
inverter- type cabinet with a variable port.

100 1000 10000 20000

FREQUENCY IN CYCLES PER SECOND

FIG. 18. Response frequency characteristics of the duo-cone
direct-radiator loudspeaker unit operating in the phase inverter
cabinet of Fig. 15 for various openings of the port.

The acoustic circuit of the system shown in Fig. 15 shows the action
of the acoustic phase inverter. When the port is closed, the inert-
ance M2 = °°, the action is the same as that of a completely en-
closed cabinet. If the inertance of the port is approximately equal to

FIG. 19. Directional characteristic of the duo-cone direct-
radiator loudspeaker at 1000 cycles.

the inertance of the cone the low-frequency response will be accentu-
ated. The performance can be deduced from the acoustic circuit as
follows :

The volume current in ZA\ is given by

346 H. F. OLSON AND J. PRESTON Vol 47, No. 4

where SAI = fi + JcoM H

ZAlZAS ~T

r AI = acoustic radiation resistance on the cone,
MI = inertanceof the cone and the air load, and
CAI = acoustic capacitance of the cone,
1

= acoustic capacitance of the cabinet volume,

TAT*. = acoustic radiation resistance of the cone,

M% = inertance of the port,

p = ABU = sound pressure which drives the acoustic system,

B = flux density in the air gap,

/ = length of conductor in the air gap,

i = current in the voice coil, and

A = area of the cone.

The volume current in zA3 is

-T

The total power radiated is given by real part of

P = rAi Xi* + rA2 Xz2.

Eq (13) shows the effect of the port in altering the response in the
low-frequency range.

A photograph of the duo-cone loudspeaker mounted in a phase in-
verter cabinet with the grill removed is shown in Fig. 16. The same
cabinet with the grill in place is shown in Fig. 17.

The measured response frequency characteristics of the duo-cone
loudspeaker operating in a phase inverter cabinet are shown in Fig.
18. These characteristics show the effect of the port opening upon
the response and also show that the response is uniform in the overlap
region.

Directional Characteristics. — The directional characteristics of a
loudspeaker used for monitoring and high-quality sound reproduc-
tion should be substantially independent of the frequency over at
least a total of 90 deg. The directional characteristics of a cone
loudspeaker are a function of the frequency. At the low frequencies

Oct. 1946

WIDE-RANGE LOUDSPEAKER DEVELOPMENTS

347

where the dimensions are small compared to the wavelength the sys-
tem is nondirectional. When the dimension of the cone becomes com-
parable to a wavelength the system becomes directional. Above this

FIG. 20. Directional characteristic of the duo-cone direct-
radiator loudspeaker at 3000 cycles.

frequency the directional pattern becomes progressively sharper with
increase in the frequency. The directional pattern of a cone is also a
function of the cone angle. This is because of the finite transmission
of sound in the cone. By increasing the angle of the cone the direc-
tional pattern becomes broader at the higher frequencies. Relatively

FIG. 21. Directional characteristic of the duo-cone direct-
radiator loudspeaker at 6000 cycles.

wide angle cones were used in both the low- and high-frequency units
of the duo-cone loudspeaker in order to obtain uniform response over
a total angle of 90 deg up to 15,000 cycles. The directional patterns
for 1000, 3000, 6000, 10,000, 13,000, and 15,000 cycles are shown in
Figs. 19, 20, 21, 22, 23, and 24. The directional pattern is practically

348

H. F. OLSON AND J. PRESTON

Vol 47, No. 4

nondirectional over the 90 deg angle below 1000 cycles. Referring
to the directional characteristics it will be seen that the directional
patterns show very little variation over an angle of 90 deg over the
frequency range to 15,000 cycles.

FIG. 22. Directional characteristic of the duo-cone direct-
radiator loudspeaker at 10,000 cycles.

Nonlinear Distortion. — Nonlinear distortion occurs when a non-
linear element is present in a vibrating system. The outside sus-
pension system is one nonlinear element in a direct-radiator loud-
speaker. The stiffness is not a constant but is a function of the

»0* 0 10 '

FIG. 23. Directional characteristic of the duo-cone direct-
radiator loudspeaker at 13,000 cycles.

amplitude and, in general, increases with larger amplitudes. The
theoretical and experimental considerations of nonlinearity in a direct-
radiator loudspeaker have been considered elsewhere3 and will not
be repeated here. The conclusion of this investigation was that the
nonlinear distortion caused by the suspension system may be elimi-

Oct. 1946

WIDE-RANGE LOUDSPEAKER DEVELOPMENTS

nated by placing the fundamental resonant frequency of the loud-
speaker at the lower limit of the reproduction frequency range.
Above the fundamental resonant frequency, the velocity of the cone is
not appreciably affected by the suspension system because the me-
chanical reactance resulting from the compliance of the suspension
system is small compared to the mechanical impedance of the re-
mainder of the system. In this loudspeaker the fundamental resonant
frequency of the low-frequency unit of the duo-cone loudspeaker was
placed at 30 cycles. Under these conditions, the nonlinear distortion
caused by the suspension system was minimized.

Another nonlinear element is the cone. In the range from 100
cycles to 1000 cycles nonlinearity of the cone produces both har-

FIG. 24. Directional characteristic of the duo-cone direct-
radiator loudspeaker at 15,000 cycles.

monic and subharmonic distortion. Since the range from 100 to 800
cycles contains the maximum power in both speech and music it is
very important that the distortion be reduced to a minimum in this
range. This can be done by employing a very rigid cone. In order
to obtain sufficient rigidity to insure low distortion it was necessary
to make the thickness of the cone about two and one-half times that
of the conventional cone. This increased the rigidity by a factor of
about 15 times.

Inhomogeneity of the flux density through which the voice coil
moves is another source of distortion. This type of distortion can be
eliminated by making the summation of the product of each turn and
the flux density associated with that turn independent of the ampli-
tude. This requirement was satisfied by making the voice coil large
and slightly longer than the air gap. In order to obtain reasonable

350

H. F. OLSON AND J. PRESTON

Vol 47, No. 4

efficiency with the heavy cone it is necessary to employ a heavy voice
coil. A voice coil of 25 grams was used in this loudspeaker which is
about 25 times the mass of the voice coil used in console-type radio
loudspeakers.

FIG.

too 1000 10000

FREQUENCY IN CYCLES PER SECOND

25. Second harmonic distortion frequency characteristics
for 1-, 2-, and 10-w input.

In the case of the high-frequency unit of duo-cone loudspeaker, the
nonlinear distortion caused by the suspension system was minimized
by making the stiffness of the space behind the cone the controlling

Z 5

tr 4

LJ
CL

Z 2
O

if) 0
O

10 WAFTS

100 1000 10000

FREQUENCY IN CYCLES PER SECOND

FIG. 26.

Third harmonic distortion frequency characteristics
for 1-, 2-, and 10 w input.

mechanical impedance. (See section on Theoretical Considerations.)
For example, the resonance of the high-frequency unit without the
back enclosure occurs at 750 cycles. With the back 'enclosure as used
in the duo-cone loudspeaker the resonant frequency is 1500 cycles.

Oct. 1946 WIDE-RANGE LOUDSPEAKER DEVELOPMENTS 351

With the above expedients the nonlinear distortion in the duo-cone
loudspeaker is quite low as shown in Figs. 25 and 20. The average
input for normal monitoring and listening use is about 100 to 200 milli-
watts which means that under these conditions the distortion is very
small.

Transient Response. — The sounds of speech and music are of a
transient rather than a steady-state character. Therefore, practically
all the sounds which are reproduced by a loudspeaker may be con-
sidered to be of a transient nature. In view of this, the transient re-
sponse of a loudspeaker is an important factor in sound reproduction.
One way of testing the transient response of a loudspeaker is to apply
a square wave current to the voice coil and record the output by

LTLTLTLT

FIG. 27. Acoustic output of the duo-cone direct-radiator loudspeaker with an
electrical square wave input of "900 cycles."

means of a microphone and cathode-ray oscillograph. For a test of
this type it is very important that the microphone be capable of re-
producing square waves. The velocity microphone is a mass con-
trolled system in the frequency range above 15 cycles. Since the
driving force is proportional to the frequency, the system can be re-
placed by a constant driving force and a resistance instead of mass
element. The transient response of this system is perfect. A special
velocity microphone was built in which the free field response as
determined by reciprocity calibrations was uniform to within one
decibel from 25 cycles to 16,000 cycles. The important frequency
region from the standpoint of transient response in double unit loud-
speakers is near or below the overlap frequency band. The response
of the duo-cone loudspeaker to a square wave having a fundamental
component of 900 cycles is shown in Fig. 27. It is not a perfect

352 H. F. OLSON AND J. PRESTON

reproduction of a square wave but is quite comparable to other audio
elements covering this frequency range. It may be mentioned in pass-
ing that to obtain a semblance of square waves from a loudspeaker re-
quires a very good acoustical system.

REFERENCES

1 Olson, H. F.: "Elements of Acoustical Engineering," D. Van Nostrand
Company, Inc., New York, N. Y., 1940.

2 Olson, H. F.: "Dynamical Analogies," D. Van Nostrand Company, Inc., New
York, N.Y., 1943.

3 Olson, H. F.: "The Action of a Direct Radiator Loudspeaker with a Non-
Linear Cone Suspension System," /. Acous. Soc. Am., 16, 1 (July 1944), p. 1.

CURRENT LITERATURE OF INTEREST TO THE MOTION PICTURE

ENGINEER

American Cinematographer

27, 7 (July 1946)
The New Norwood Exposure Meter (p. 254)

27, 8 (Aug. 1946)

Evolution of the Camera in Sound-Film Pro-
duction, 1926-1946 (p. 276)
Soviet Film Scenarios (p. 282)
Sound and the Visual Image (p. 284)

British Kinematograph Society, Journal

9, 2 (Apr.-June, 1946)
Some Applications of Photography and Kine-

matography in War-Time (p. 39)
The Past and Future Activities of the Society
of Motion Picture Engineers (p. 43)

Two New Sound Recording Films (p. 51)

R. A. WOOLSEY AND C. H.

COLES

D. EREMIN

H. A. LIGHTMAN

G. H. SEWELL

D. E. HYNDMAN AND J. A.

MAURER
I. D. WRATTEN

British Kinematograph Society, Proceedings of the Film Production Division

(Session 1945-16)
The Evolution of Motion Picture Technique

(p. 3) W. M. HARCOURT

Practical Sound Problems in Film Production.

I. Production 'Requirements (p. 34) J. J. CROYDON

Location and Planning of Studios (p. 45) H. JUNGE

International Photographer

18, 6 (July 1946)
New Concentrated-Arc Lamp (p. 12) G. S. OSLIN

18, 8 (Sept. 1946)
New Kodachrome 16-Mm Commercial Film

(P- 9)

Evolution of Motion Picture Technique (p. 11) W. M. HARCOURT
Kodak Etachrome Film (p. 18)

353

354

EMPLOYMENT SERVICE

Vol 47, No. 4

International Projectionist

21, 7 (July 1946, Section 1)
The Retiscope Fiberglas Screen (p. 5)
Video and the Somnolent Cinema (p. 12)
21, 7 (July 1946, Section 2)
Studio Projection Engineering (p. 20)
Evolution of the Carbon Arc for Projection

(P- 24)

Case History of the Simplex Projector (p. 33)
If It Isn't on the Film (p. 36)
Projection Arc Lamps — Then and Now (p. 45)
Present and Future 16-Mm Projection (p. 48)
Twenty Years of Horn Progress (p. 52)
Projection Room Design Advances (p. 54)
Motiograph: The Early Years (1896-1924)

(p. 60)

21, 8 (Aug. 1946)
Basic Radio and Television Course, Pt. 25 —

Receiving Systems (p. 17)
Projectionist's Role in Sound Reproducer

Development, 1926-46 (p. 22)

The Photographic Journal

86B, 2 (Mar.-Apr., 1946)
The Scophony High-Speed Camera (p. 42)

RCA Review

7, 2 (June 1946)

An Experimental Color Television System
(p. 141)

P. BETHEL

F. WALDROP AND J. BORKIN

A. ROST AND W. McCORMICK

W. C. KALB
M. STEPHAN
P. MOLE
H. H. STRONG
E. W. D'ARCY
W. W. SIMONS

B. SCHLANGER

A. C. ROEBUCK

M. BERINSKY
F. LOVETT

R. D. KELL,

G. L. FREDENDALL,

A. C. SCHROEDER

AND R. C. WEBB

EMPLOYMENT SERVICE

POSITIONS OPEN

We are listing below additional positions available with the U. S. Public Health
Service, Communicable Disease Center, as described on page 270 of the September
JOURNAL. Applicants should address inquiries to Personnel Officer, U. S- Public
Health Service, 605Volunteer Building, Atlanta 3, Georgia.

(7) CHIEF, EVALUATION SECTION, $4902 per annum. Requires a
thorough knowledge of the evaluation of audio-visual aids as applied to
the dissemination of information. Applicants must be able to devise
programs of testing audio-visual aids. Must maintain liaison with
personnel using same; must be able to advise on audience interpreta-
tion, attitude, motivation, dramatic presentation, for cinematic tech-
nology, upon the teaching impact; must supervise one or more part-time
educators, and perform other related duties as assigned.

Oct. 1946 EMPLOYMENT SERVICE 355

(8) PHOTOGRAPHER, MOTION PICTURE, $3397.20 per annum.
Applicants must have a thorough knowledge of general motion picture
and still photography and the ability to operate animation equipment.
His experience should be of such productive nature as to indicate con-
cisely the ability to perform the duties involved.

(9) CHIEF, WRITERS SECTION, $4149.60 per annum. Applicants
must have had progressively responsible experience in the preparation of
written material of a scientific or general nature for motion pictures, both
general or training films, and other media of dissemination. Must be
able to develop and prepare film continuity scripts and of collateral
training material to accompany film production.

(10) CHIEF, ANIMATION SECTION, $3397.20 per annum, involving
the ability to depict ideas of a scientific or technical nature for production
through motion pictures and other audio-visual media. Supervises
several animation artists.

(11) CHIEF, TRAINING AND PRODUCTION SERVICE, $8179.50
per annum. Applicants must be able to accept the responsibility for the
development of the training and audio-visual production services of the
Communicable Disease Center. Directs the program of production and
distribution of all audio-visual training aids and the orientation and
specialization training in public health. Advises with the Officer in
Charge on program and policy formulation.

(12). FILM WRITER, $3397.20 per annum. Requires a knowledge
of thie development and preparation of film script in training film produc-
tion. Must be able to prepare a shooting script adequate for the effec-
tive presentation of materials by audio-visual means.

New film production unit to be located at Athens, Georgia, needs film
editor- writer and film director. Experience in 16-mm as well as 35-mm
production desirable. Southern background or interest in South pre-
ferred but not essential. Write giving full details of experience, etc., to
Nicholas Read, The National Film Board, Ottawa, Canada.

POSITIONS WANTED

356 EMPLOYMENT SERVICE Vol 47, No. 4

16-mm Specialist. Honorably discharged veteran with many years'
experience, specializing in 16-mm. Linguist. Available for special
assignments. Write J. P. J. Chapman, ARPS, FRSA, The Huon,
Branksome Hill Road, Bournemouth, England.

JOURNAL OF THE SOCIETY OF
MOTION PICTURE ENGINEERS

Vol 47 NOVEMBER 1946 No. 5

CONTENTS

PAGE

The Newsreel— Its Production and Significance :

Editing the Newsreel D. DOHERTY 357

Foreign Editions H. LAWRENSON 361

Women's Fashions V. DONNER 364

The Film Library B. HOLST 365

The Field Unit ' J. GORDON 367

The Newsreel Cameraman W. MclNNis 80S

Newsreel Sound W. M. McGRATH 371

Characteristics and Applications of Concentrated-Arc
Lamps W. D. BUCKINGHAM AND C. R. DEIBERT 376

Optical Problems of the Image Formation in High-
Speed Motion Picture Cameras J. KUDAR 400

An Improved Method for the Determination of Hydro-
.quinone and Metol in Photographic Developers

H. L. BAUMBACH 403

Application of Methyl Ethyl Ketone to the Analysis of
Developers for Elon and Hydroquinone

V. C. SHANER AND M. R. SPARKS 409

Naval Training-Type Epidiascope for Universal Projec-
tion of Solid Objects J. BOLSEY 418

A New Method of Counteracting Noise in Sound

Film Reproduction W. K. WESTMIJZE 426

Society Announcements 441

Indexes to the semiannual volumes of the JOURNAL are published in the June and December
issues. The contents are also indexed in the Industrial Arts Index available in public libraries.

JOURNAL

OF THE

SOCIETY of MOTION PICTURE ENGINEERS

MOT«L PENNSYLVANIA • NCW YOP.KI. N-Y- • TCL. PCNN. 6 O62O

HARRY SMITH, JR., EDITOR

Board of Editors
ARTHUR C. DOWNES, Chairman

JOHN I. CRABTREE ALFRED N. GOLDSMITH EDWARD W. KELLOGG

CLYDE R. KEITH ALAN M. GUNDELFINGER CHARLES W. HANDLEY

ARTHUR C. HARDY

Officers of the Society
'President: DONALD E. HYNDMAN,

342 Madison Ave., New York 17.
*Past-President: HERBERT GRIFFIN,

133 E. Santa Anita Ave., Burbank, Calif.
'Executive Vice-President: LOREN L. RYDER,

5451 Marathon St., Hollywood 38.
** Engineering Vice-President: JOHN A. MAURER,

37-01 31st St., Long Island City 1, N. Y.
^Editorial Vice-President: ARTHUR C. DOWNES,

Box 6087, Cleveland 1, Ohio.
"Financial Vice-President: M. R. BOYER,
350 Fifth Ave., New York 1.

* Convention Vice-President: WILLIAM C. KUNZMANN,

Box 6087, Cleveland 1, Ohio.

* Secretary: CLYDE R. KEITH,

233 Broadway, New York 7.
^Treasurer: EARL I. SPONABLE,
460 West 54th St., New York 19.

Governors

*fFRANK E. CAHILL, JR., 321 West 44th St., New York 18.
**FRANK E. CARLSON, Nela Park, Cleveland 12, Ohio.
**ALAN W. COOK, Binghamton, N. Y.

*JOHN I. CRABTREE, Kodak Park, Rochester 4, N. Y.

*CHARLES R. DAILY, 5451 Marathon St., Hollywood 38.
**JOHN G. FRAYNE, 6601 Romaine St., Hollywood 38.
**PAUL J. LARSEN, 1401 Sheridan St., Washington 11, D. C.
**WESLEY C. MILLER, Culver City, Calif.

*PETER MOLE, 941 N. Sycamore Ave., Hollywood.
*JHoLLis W. MOYSE, 6656 Santa Monica Blvd., Hollywood.

* WILLIAM A. MUELLER, 4000 W. Olive Ave., Burbank, Calif.
*°A. SHAPIRO, 2835 N. Western Ave., Chicago 18, 111.

*REEVE O. STROCK, 111 Eighth Ave., New York 11.

Term expires December 31, 1946. tChairman, Atlantic Coast Section.
**Term expires December 31, 1947. jChairman, Pacific Coast Section.
*° Chairman, Midwest Section.

Publication Office, 20th & Northampton Sts., Easton, Pa.
General and Editorial Office, Hotel Pennsylvania, New York 1, N. Y.
Entered as second-class matter January 15, 1930, at the Post Office at Easton.
f Pa., under the Act of March 3. 1879.

JOURNAL OF THE SOCIETY OF
MOTION PICTURE ENGINEERS

Vol 47 NOVEMBER 1946 No. 5

THE NEWSREEL— ITS PRODUCTION AND SIGNIFICANCE*

Summary. — The following symposium on the production and significance of the
present-day newsreel was one of the discussions in the series of popular subjects ar-
ranged by the Atlantic Coast Section of the Society to promote a wider knowledge of
motion picture industry techniques and practices. Newsreel operation in peace and
war and a resume of the details of production are described by staff members of Movie-
tone News.

EDITING THE NEWSREEL
DAN DOHERTY**

On behalf of our producer, Edmund Reek, it gives me sincere pleas-
ure to welcome you here tonight. We have full knowledge and appre-
ciation of the splendid contributions to the motion picture industry
•made by members of your learned Society and it is with a certain
degree of temerity that we stand before you to tell you something
about our little segment of motion picture making.

Screen journalists believe that the newsreel is one of the most vital
•units in the industry, but for all that, we are often neglected and sadly
misunderstood. There is a tendency to too casual an acceptance of
our noblest efforts and to too bitter criticism of our slightest foibles.
One of the most rankling criticisms leveled at us is made by way of
odious comparison, or I might put it "invidious comparison," in that
it is based on a false premise.

Why, we are asked over and over again, cannot the newsreels
present the news like newspapers, or like news magazines, or like
picture magazines? To ask such questions, in our corporate opinion,
shows a complete lack of understanding of a newsreel' s function.
W^hile the newsreel at times can be the most complete reporter — wit-

* Presented Apr. 17, 1946, at a meeting of the Atlantic Coast Section of the
Society held in the studios of Movietone News, Inc., New York.
** Assignment Editor, Movietone News, New York.

357

358 PRODUCTION AND SIGNIFICANCE OF NEWSREELS Vol 47, No. 5

ness recent UNO coverage, the battles of World War II, the assasi-
nation of King Alexander, Foreign Minister Barthou of France at
Marseille, the Hindenburg disaster, the Jap attack on the Panay, the
Pearl Harbor disaster — it cannot cover every item presented in the
New York Times, or Time magazine, or Life, or the Reader's Digest.

The newsreel editor cannot, for many reasons, attempt to cover all
the news. In the first place, the cost of keeping a camera staff capa-
ble of the noble effort would be prohibitive. In the second place,
what would we do with it when we had it? Newsreel presentation
time in theaters is limited. The newsreel editor has at most 1800
ft a week in which to present the news of the world!

By this confining fact alone his task is, therefore, not one of attain-
ing total coverage but of selection, and selection based on an intimate
knowledge and understanding of the medium.

But now I hear our critic saying, "Well, if that is so, why select all
the trivia that newsreels are notorious for presenting, and why all
those hardy annuals that year after year crash their way to the
screens of the world via the newsreel releases?" Well, gentlemen,
there you step on our pet corn, and we plead guilty. But there are
extenuating circumstances, as I shall describe.

In the first place, a lot of the hardiest annuals are of national in-
terest— the horse races, football games, world baseball series, Rose
Bowl, Mardi Gras, etc. On occasion we have taken rein and passed
up covering one or another of these stories. The howl which arose
from exhibitors over our neglect would put a pack of wolves to shame.
A given section of the public likes to see these well-advertised spec-
tacles and all we can do is give it to them. You should see our mail
with requests from exhibitors for stories of sectional interest only.
For instance, exhibitors in the Michigan area want that hardy annual
from Holland, Michigan, which shows the goodly descendants of
those highly sanitary Dutch burghers turning out en masse to clean
their city streets. This is picturesque only the first time you see it,
but Michiganders expect it every year. So, we make a special of it
for that territory. Almost every section of the country has a fete or
celebration of this kind that the populace thereof thinks as much of
as do the Michiganders about their exhibition of cleanliness.

Other trivia is partly our editorial responsibility. The policy of
Movietone when there is a slack in the international picture is to
present a balanced and entertaining reel. Therefore, we have our
animal corner, our daffy Lew Lehr, our Donner Fashions, sport

Nov. 1946 PRODUCTION AND SIGNIFICANCE OF NEWSREELS 359

features, "cheesecake" charm, and baby shows. You cannot always
have Papal Consistories, with their medieval pomp and UNO meet-
ings with their dramatic fireworks. You have to include a launching
now and then, cover Washington doings, and so forth.

In fact, some of our greatest pictures have come from routine
assignments. We had seriously considered passing up the arrival of
the Ilindenburg on the day it exploded because we had taken it
arriving at Lakehurst many times without incident. Can you imag-
ine what would have happened to our whole staff if we had not sent
our man out there that day? We would all have been fired and jus-
tifiably so! An editor must have a sixth sense and anticipate things
like that. Another incident happened during the very routine assign-
ment of covering Mrs. Truman christening a plane. We got the
laugh picture of last year. A subject we released on three ducks
playing with a kitten was kept on many theater screens as long as
eight weeks.

We are asked, "Why not controversial subjects?" I will be ready
after this meeting to take suggestions from anyone who will tell me
about a controversial subject that will fit into newsreel footage. Our
newsreel policy is to be strictly objective, to let the camera tell the
story. During the recent steel strike we had a prominent steel com-
pany thinking we were against them. Because of its far-reaching
implication, when the strike broke we decided to cover it and sent
men to the mills, etc. Then we went to the union involved and the
company and asked for statements from the heads of each. We got
one from the union but failed to get one from the company. Our
release followed and then a call came from the company: the presi-
dent would make a statement for us now. They were told politely
that we did not want it then. We had three other talks in our reel
coming up and the steel company was just backed off the front
page as far as we were concerned. They swallowed hard, but took
it. What else could they do — and for that matter, what else could
we do?

That the n\swsreels do as good a job as they do with their limited
staffs is a major wonder to me. However, we are by no means satisfied
with ourselves and we are continually looking to improvement.
About the future and television, I am sure you technically minded
men are better informed than we, but we believe television to be our
"oyster," that is, editorially. W7e have developed the techniques of
covering news for motion and sound, and whether the picture goes on

360 PRODUCTION AND SIGNIFICANCE OF NEWSREELS Vol 47, No. 5

a theater screen or a television receiving set, we know how to get it
there with the greatest efficiency. You men have made the machines
to do it — we think we have the editorial acumen to use them. Mer-
genthaler did not run Horace Greeley or Charles Anderson Dana out
of business.

We have many violent critics, but in high places we have a few
friends. We have the sympathetic understanding of such executives
as our own president, W. C. Michel, and our parent company's presi-
dent, Spyros Skouras, its vice-president in charge of distribution,
Tom Connors, and our chief production executive, Darryl Zanuck.
All of these men take a lively interest in our affairs, are first to praise
us when we are good, and are not bashful with criticism when they
believe it is justified. Qne of the most sincere tributes ever paid a
newsreel was made by our late president, Sidney Kent. I would
like, in conclusion, to read it to you:

"The newsreel is a Gulf Stream flowing through the motion picture industry,
warming it with its vigorous, ever-young spirit of enterprise.

"With a long record of accomplishments the newsreel has earned an enduring
place on every theater program. It is the standard short reel — never failing to
make its screen time memorable, instructive, and entertaining. Its multitudinous
activities, always carried out with breathtaking speed, amaze and inspire those
who know the difficulties of production.

"We have only to review the record of Movietone News in any given year to
appreciate the perfection of newsreel organization.

"Editorially and technically Movietone News has been a thing of clarity and
precision. Balanced, poised, daring, and courageous it has steered an honest
course through the labyrinth that is the history it records.

"In every country in the world it commands respect for the strict impartiality
of its policies, wherein no whit of its independence is sacrificed. We are pleased
to congratulate its executives and associates on the newsreel for the unsparing
zeal with which they carry on through the endlessness and unexpectedness of
their task of reporting the news of the world."

And that, gentlemen, is the significance of newsreels. It makes me
sick to see some producers and theater men cringe before cheap
politicians when their industry commands such an influential journal-
istic arm that can call on the Bill of Rights to maintain its journal-
istic prerogatives of free speech.

Nov. 1946 PRODUCTION AND SIGNIFICANCE OF NEWSREELS 361
FOREIGN EDITIONS

HARRY LAWRENSON*

As a newsreel editor, which I have been for about 25 years, both
in the "silent" days and since "sound," my chief interest in what goes
on behind all the knobs and dials, on both cameras and amplifiers, is
simply what comes out, on and behind the screen, when I press the
buzzer to the projectionist. As long as the picture is in focus and the
"noises" sound good, I am happy. When the pictures are fuzzy or
shaky, or when the noises come out "sour," I can always cuss the
technical department! And that, now and again, can be a blessing
in disguise, for when, on rare occasions, the picture or the sound
track is not so good, people will notice that, but they do not think to
criticize occasional poor editing of the picture they are looking at!

But perhaps you would be interested in what happens to our news-
reel pictures after the cameraman, the soundman, the recording de-
partment, and the editors have finished with them. Movietone
News, as you see it in the Roxy or any American theater, is only the
first product. That same newsreel, or most of the items in it, is sub-
sequently shown in no less than 47 foreign countries and in more
than a dozen different languages! From New York, and from very
well-equipped production centers in London, Paris, Sydney, Aus-
tralia, Brussels, Belgium, Rome, and now in South Africa, regular
weekly or twice weekly editions of the newsreel are prepared and re-
leased. Movie audiences in Canada, England, France, Italy, Sweden,
Australia, India, and every one of the Latin American countries are
very familiar with the words Movietone News, even if, on occasion,
they do appear on the screen as "Actualities Parian te," "Actualidades
Movietone," or "Foxuv Zvukovy Tydenk." We have estimated the
weekly world audience of Movietone at over 200 million people. The
50 leading newspapers of the world do not have anything like that
circulation, all together!

If the newsreel has developed in its world appeal, I am happy to
say that camera development and sound recording have definitely
kept pace. One of my fonder memories is of 1927 when the first
"portable" recording outfit went overseas. I was in Paris at the
time, editing our then silent Fox Newsreel, when two gentlemen —
Ben Miggins and Eddie Kaw — descended upon me in a huge moving

* Foreign Editor, Movietone News, Xew York.

362 PRODUCTION AND SIGNIFICANCE OF NEWSREELS Vol 47, No. 5

van with the words Movietone News on its sides! I said a "portable"
recording outfit advisedly, for when Miggins and Kaw explained this
newfangled sound business to me, they kindly offered to take me
along on some of their jobs. I soon found that they needed me to
help lift and carry the half-ton cameras and amplifiers and stuff with
which the truck was loaded! Yes, it was portable, if you had a
strong back and a weak mind. That outfit, incidentally, made the
first sound films of such people as Clemenceau, Foch, Mussolini, and
the Pope.

Speaking of portability, I went along with the sound crew on one
of those early sound recording jobs in Europe, and a young priest put
his finger on what was then our number one problem. We arrived
at a small parish church in a provincial village to film a colorful cere-
mony and festival. We arrived barely in time, and the young priest
who was to officiate was in a dither because of the delay. As we
started to unpack all the heavy boxes and gear and dump the stuff
outside his little church, he became visibly more agitated. When he
saw that still more boxes and batteries and things were yet to be
unloaded, he approached me with a look of concern. "M'sieur," he
said, "if you have to carry all that equipment into our little church,
we'll never get started. Don't you think it might be better if I
brought the congregation and the ceremony out here in. the yard and
you made your picture from your truck?"

Today, thanks to you gentlemen, all our cameramen carry Eyemos,
and I understand the soundmen carry all their gear in a very small
cigar box! That, incidentally, might account for the occasional
aroma that seems to emanate from the sound track!

When we look back to those very early days of the motion picture,
when an Akeley was the latest thing in cameras, and sound was some-
thing Al Jolson was dreaming about, we have really come a long way.
Who would have thought then that a United Nations Conference
held in New York would be faithfully reported in sight and sound,
and that the pictures would be shown, for example, in London, by
Movietone News, only two days after they were made. Who, in-
deed, in these days when presidents, kings, prime ministers, and
politicians go out of their way to get in front of a newsreel camera and
microphone, would not wonder at the time, not too many years ago,
when a news sound outfit was looked upon as a toy that was not to be
taken seriously, or perhaps even something to be carefully avoided
lest the speaker's words of wisdom be only too faithfully recorded and

Nov. 1946 PRODUCTION AND SIGNIFICANCE OF NEWSREELS 363

presented to the public as he really said them, if not exactly as he
meant them!

When the French statesman Clemenceau was alive, fear of the
recorded word was quite real. We had tried and tried to get the old
man to give us a sound interview, but the answer was always the
same. In French, or English, it was "Nothing doing." But, one
day, the old Tiger finally consented to a camera interview in his
garden. No sound, however!! That was the condition. So, news-
reel men, being always the soul of honor, carefully planted a micro-
phone and cables behind a row of cabbages and set up our camera.
Mr. Clemenceau came out and in a fairly genial mood began to walk
around the garden while we took his picture. After a little of this,
one of our boys asked him to move in nearer to the camera for a
close-up, carefully leading him to the spot where our "mike" was
hidden. The sound recorder was ensconced in the truck some dis-
tance away and out of sight.

All was well until our guide started to ask Mr. Clemenceau, in an
off-hand fashion, what he thought of the world situation. Mr.
Clemenceau appeared to think for a moment and his face darkened.
Then suddenly he raised his thick walking stick into the air. "This
is what I think!" he shouted, and brought his stick crashing down
into the cabbage patch, right where our microphone was hidden! Or
at least where we thought it was hidden! From then on we always
kept our gear in full sight — microphones were expensive.

On another occasion in Sydney, Australia, where I was assigned to
start our Australian Edition of the newsreel, we had an illustration of
the sort of thing that helps remove hair from editors' skulls. Our
task was to get out an all- Australian newsreel every week, and in a
country where important news just does not happen every day that
was quite a job. We managed to get out the reel for the first few
weeks by running subjects somewhat longer than we might here,
padding them a little, so to speak, and then came the inevitable. It
was make-up day. We had to go to press that night and I needed
200 ft more.

I called the boys together — we had two sound cameras and one
silent — and dispatched all three of them to go out and film something,
anything; I had to have enough footage to fill up a newsreel. One
went off to track down a talking bird about which he had heard.
Another went to a union leader to try and get a statement on a
threatened strike, and the silent cameraman went down to the

364 PRODUCTION AND SIGNIFICANCE OF NEWSREELS Vol 47, No. 5

beach to look for a possible bathing girl picture. I was desperate!
I went to the corner pub to get a drink. Some hours went by and
eventually all three cameramen returned, each with the same report —
nothing to photograph! By that time, however, / did not mind.
The studio had caught fire and burned down while they were gone!
And we did not have even a shot of our own plant going up in flames!
Dog bites man!

I really have not a great deal either technically or scientifically
to contribute to this session. But I will add this, in closing: Every
week, when I receive my pay check for editing Movietone News, I say,
"Thank heaven for the Motion Picture Engineers!"

WOMEN'S FASHIONS
VYVYAN CONNER*

Movietone News fashion subjects are miniature productions. In
proportion, as much time and effort go into their make-up as goes into
the preparation of a feature picture. We often spend weeks in prepa-
ration for a one-minute subject, and we spend a complete eight-hour
day of shooting to get a one-minute subject on the screen. The
whole world is tapped for ideas ; and sets are especially built on which
to present them. Girls are interviewed by the hundreds, and finally
picked for their beauty and charm, poise and personality, figure and
smartness.

For their movie appearance their make-up is as carefully put on as
a Hollywood star's. Our cameramen light them to their best advan-
tage, so that both girl and gown are exquisitely set forth in all their
beauty. Every effort is used to make the newsreel fashion clip a
small bit of perfection, from news and subject matter, suitability of
set and location, and on up to the height of feminine American
Beauty.

I work with the finest, most brilliant American designers, and
bring their designs before the eyes of the world, showing American
taste, American settings, American girls, and the "American look"
to some hundred-million moviegoers.

In normal times, we make about one fashion picture every ten
days. The commentary is translated at once into Portuguese and

* Women's Editor, Movietone News, New York.

Nov. 1946 PRODUCTION AND SIGNIFICANCE OF NEWSREELS 365

Spanish, and shipped to all the sister Republics of Latin America.
The fashions are then sent abroad for translations into French, Ger-
man, Swedish, and all the other languages.

THE FILM LIBRARY

BERT HOLST*

Situated on the top floor of the Movietone News building is the
Film Morgue or, to be more official, the Movietone News Film Li-
brary over which I preside. Few people in the news business have
any idea of the true importance of this unpublicized department with
its history in cans. As of this day there are stored in our vaults
approximately 42 million ft of negative scenes shot all over the world,
from the North to the South Poles, from every conceivable angle —
ground, air, and underseas.

This film is carefully catalogued with a simplified but all-informa-
tive cross index, which allows us to put our hand on any given foot of
film in the shortest possible time.

Thirty- two fireproof vaults store this film. Because of space
limitations in addition to the millions of feet already mentioned,
there are other millions in our out-of-town vaults at Ogdensburg,
New Jersey. This latter footage we consider as the most unlikely
to be called for by our own company, or by the other clients who use
our library for stock shots.

We also have a sound library containing every conceivable sound
or a good facsimile of any sound. This is also catalogued in our
simplified system and before you could say "Jack Robinson" we
could give you anything from an artillery barrage to a Bronx cheer.

One of the greatest worries of a film librarian is that old bogey,
fire hazard. A constant vigilance must be maintained where so
much inflammable film is kept. Extreme temperatures, either hot
or cold, are always a problem with us and at such times we must be
careful of gas accumulations and see to it that the entire vault spaces
are properly aired. We must also guard against dampness that
rots film and dry ness that shrinks it. Keeping this invaluable col-
lection of canned history in a good state of preservation means main-
taining as near as possible a constant temperature.

* Librarian, Movietone News, New York.

366 PRODUCTION AND SIGNIFICANCE OF NEWSREELS Vol 47, No. 5

From our library we yearly supply the industry with millions of
feet of stock shots. Practically every West Coast production com-
pany uses our facilities, that is, when they do not conflict with our
service to our parent company, Twentieth Century-Fox Film Cor-
poration and of course our own newsreel and short subject depart-
ment.

I do not have time to tell you of the demands made on us which we
satisfy in the majority of cases. Being film men you could easily
imagine what these are. Some are really tough to fulfill but we
manage to do them most of the time. However, there are limits to
our ingenuity and we are stumped occasionally. For instance, here
are some actual requests we have received from some of our clients.

"Have you a shot," we were once asked, "of an English railroad
train running without lights during a blackout?"

Another time a lady editor seriously wired for — and this is the
exact description she sent us — two love doves cooing, one gets vexed
and draws away, the other seems to pine.

A West Coast producer once asked us for an alligator waving an
American flag!

A religious organization in the process of making a biblical picture
wired us, "What have you got in your library on Abraham and
Moses?"

A big powder manufacturing company making a commercial,
wired us to send them a 2-f t shot showing a close-up of the impact of
an explosive shell hitting an iceberg.

A famed explorer wishing to illustrate the fiction he dishes out in
his lectures wrote us that we would greatly help his film presentation
if we could supply him with a pelican diving for a fish; camera to
follow him under the water; pelican to catch school of fish; fish
swimming in pelican's bill.

Then, as innocently as a newborn babe, another West Coast pro-
ducer wired us for "A Wednesday afternoon scene — -Landscape."
This is a complete description of his request!

Of course there are many requests for sound effects, legitimate and
otherwise, but I actually got this one: "Can you send us the sound
of a moccasin on soft snow?"

That, gentlemen, is a little about the Movietone News Library.

Nov. 194G PRODUCTION AND SIGNIFICANCE OF NEWSRKELS 367

THE FIELD UNIT

JACK GORDON*

An assignment to cover a national political campaign offers an op-
portunity to explain the many problems and obstacles encountered
by a newsreel unit in the field. As I was "Mr. Newsreel" with one of
the candidates on the last campaign I can tell you this is so.

Shortly after the conventions, I was appointed, at the suggestion of
our Producer Edmund Reek (after he had been asked about it by the
Republican National Committee), to take charge of the newsreel
coverage of the campaign and to be the liaison between all the news-
reels and the Republican Committee. On paper it looked easy and a
nice chance to see the country at the expense of the Republican Na-
tional Committee. Well, you live and learn. The only things I saw
were stars when our special train collided with the rear end of another
train.

To start, my first duty was to make arrangements on the campaign
train for a representative from each of the reels. When we were
ready to depart we had only a rough itinerary. First, Philadelphia,
next Louisville, then across the country to Washington, then Oregon,
California, and then back east again. At every major stop plans had
to be made for camera positions in each of the many auditoriums
where Dewey was to speak. Lights had to be rented or bought and
set up, and last but not least unions in different localities had to be
contacted so that there would be no difficulties from that end.

Each auditorium and stadium was different, which meant many
problems. Some had balconies and some did not. To offset this,
platforms were built for each place large enough for all the newsreel-
men to work comfortably. And lighting these huge places was always
a problem. Some of these cities, where addresses were to be made,
did not have sufficient equipment to supply our lighting needs.

Wre had to call New York, Chicago, and California to furnish this
equipment; big city coverage was difficult, but easier to handle than
stops enroute. When the special train stopped at some small town,
the local populace would be out at the station en masse. Of course,
our candidate would be prevailed upon to make a short address.
Plans for these platform addresses would be announced but a short
time before the train would be in the station. Cameramen would

* Unit Director, Movietone News, New York.

368 PRODUCTION AND SIGNIFICANCE OF NEWSREELS Vol 47, No. 5

have to be ready to jump off the train and rush back to the observa-
tion car, fight their way through the crowds to get a good location,
and be ready when the candidate appeared. At no time could the
cameramen afford to let the candidate appear without safeguarding
themselves by being on the spot with their cameras. None could
foresee what might occur; witness the shooting of Mayor Cermak
when an attempt was made in Florida to assassinate President
Roosevelt.

In all of the major cities where the candidate campaigned the local
authorities would arrange for quite a celebration and parade from the
station to his hotel headquarters. This route had to be covered and
arrangements had to be made for a special automobile to carry the
cameramen at the head of the parade. Although this arrangement
was agreed upon with the candidate and his staff it very often happened
that the camera car would have to battle against being pushed out
of the parade. Passes, police cards, and other identifications meant
nothing to some of them. There is a perpetual feud between news-
reel men and police. The constabulary always have their own ideas
as to where the newsreel fits in. It was very important that the can-
didate be covered completely but it meant constant fighting to do it.

There is never a dull moment for the hard-working cameraman.
He never knows what the morrow will bring for him. Some local
assignment, city fathers pinning a medal on visiting heroes, a political
speech, amateur boxing or basketball, Atlantic City beauty pageant,
or a trip around the world. Perhaps just a fire that will keep him on
the job all night and day, or a strike. Whichever the case, it will be
on film, and the next day he will be back for more.

THE NEWSREEL CAMERAMAN
WALTER McINNIS*

Fifty years ago this October, Hammerstein's Olympia Music Hall
rang to the cheers of an enthusiastic audience as President McKin-
ley's Inaugural Parade was re-enacted in all its pomp on a motion
picture screen.

In 1927, the Fox-Case Corporation launched its famous Fox Movie-
tone News. It was instantly popular and the public who had become

* Cameraman, Movietone News. New York.

Nov. 10-Hi PRODUCTION AND SIGNIFICANT!-: OF NEWSKKKI.S

sound conscious overnight, received it with greal acclaim. HclWe
long the silent-type newsreel became antedated and just as quickly
so did the silent type of coverage suffer a momentary lapse. All
newsreel stones were covered with the prime thought in mind, "How
is it for sound?" This was purely a transition period. The news-
reels had found their voice, but had not yet learned to talk.

Public acclaim for the sound newsreels was not to be denied. Be-
fore long all five major newsreel producing companies were operating
sound trucks throughout the world, although none quite so exten-
sively as Fox Movietone. It surely seemed as if the day of the silent
camera had waned, but already the pendulum had reached the limit
of its swing and was enduring that split-second battle with inertia
before returning. In other words, motion pictures with sound were
no longer newsworthy just because they had sound. Now, the sound
had to be justified, and thus the newsreel commentator was born. It
soon became apparent that many newsreel shots could be covered
"MOS" — or in newsreel parlance, "mitout sound" — and joyfully,
cameramen rushed to their respective attics and reverently dusted
off the old silent cameras.

Aided and abetted by the newsreel editors who once more had be-
come "coverage conscious" the pendulum raced back across its arc
with increased momentum while the amount of field-recorded sound
that was heard in the newsreel became reduced to nearly the vanish-
ing point.

During this period many improvements in sound camera equip-
ment were made by the Wall Camera Company of Syracuse. A new
compact self-contained camera, comparable in weight to a silent
Mitchell camera, and requiring a light 12-v storage battery for its
operation, was delivered to Movietone cameramen. It had the first
of the popular D-type intermittent, a rack-over arrangement for
critical focusing through the objective lens, and a right-side-up finder.
This camera could be used with a 400-ft magazine as well as the 1000-
ft type, an important weight decreasing factor. All of the restrictions
imposed upon cameramen by the use of the old-type sound camera
equipment were eliminated with this new camera. The sound equip-
ment, too, had become very portable and movement became almost
as unrestricted as with the silent camera. Another important im-
provement was the Akeley gyrotripod permitting smooth "pam" and
tilts coupled with ruggedness of construction. For nearly all types
of stories this tripod is still unsurpassed.

370 PRODUCTION AND SIGNIFICANCE OF NEWSREELS Vol 47, No. 5

Newsreels today show the result of 50 years of progress. No small
tribute to the newsreel institution is the record of combat coverage in
the last war. The pool of war correspondents of the five major pro-
ducing companies was responsible for much of the documentary film
which will provide the motion picture history of the war. Combat
cameramen attached to the Armed Forces rapidly acquired the news-
reel technique under the tutorship of many old timers in the business.
In fact, at this office and at the March of Time Newsreel Cameramen
School operated during the first year of the war, hundreds of fighting
cameramen were put through these two institutions. The graduates
in turn trained the great body of fighting cameramen who have given
us the complete history of World War II in motion pictures.

In October 1929, in the pioneer days of sound newsreels, I was given
an assignment to go to India on a tiger-hunting expedition. Before
leaving, a rush call came for a lightweight outfit to make pictures
with sound aboard the dirigible Los Angeles on a test run over Phila-
delphia and New York. We took off at 5:00 P.M. and made pictures
of the crew's quarters and other interiors. Approaching New York
approximately at 8:30 P.M., with the light not too good, we took a
chance and made some night shots of the City. The Commander of
the Los Angeles was good enough to describe the time and places we
were passing over; his voice recorded exceptionally well, considering
the drone from the motors. We were later complimented on the
quality of the negative and sound, and were informed that these were
the best night shots made over New York up to that time.

Our first jungle assignment was in Hydrabad, where our camera
was placed on a platform built in a tree at a 15-ft elevation. The
microphone was set where we hoped a tiger would make an appearance
to kill a water buffalo or cow. This continued for nearly three months
with only partial success — the tiger refused to co-operate.

We then moved to the native state, Cooch Behar, north of Calcutta.
The Mahareni of Cooch Behar was very co-operative and we made
several tiger hunts from elephants. In all we had 32 elephants. We
were more successful on this enterprise. On our first trip into the
jungles the two elephants mentioned before had quite a battle over a
maiden elephant, and that nearly ended our hunt before we even got
to our jungle location, but the mahouts appeased the elephants.

About 20 elephants would form a large circle and drive in any
tigers or other animals that were encircled in the ring toward the fire
line where we, the Mahareni, and her guests were set up. Our first

Nov. 1946 PRODUCTION AND SIGNIFICANCE OF NEWSREELS 37 1

drive netted a huge black bear with two cubs clinging tightly to their
mother's back. The huge bear came out of the jungle so close to my
camera elephant that he became startled and reared back on his hind
legs, and in turn gave me quite a scare not knowing whether he would
crash back on me and the equipment. The elephant finally settled
down and surprisingly enough we had some very good pictures with
sound of the elephants trumpeting wildly and loudly. The Mahareni's
guests made no attempt to shoot the bear because it is not considered
sportsmanlike to shoot a mother bear with cubs.

In our next beat-in, we rounded up a leopard, and the following day
we rounded up a large tiger weighing nearly 400 Ib. This is really a
sport of kings as one has to be an invited guest to take part in a hunt
of this nature.

In April 1930 we returned to Calcutta. The city was in an uproar.
The Ghandi riots had started; so instead of returning to New York
we shot a lot of material in Calcutta, and then proceeded to Bombay
where the fighting was more intense. We made thousands of feet of
riot pictures.

NEWSREEL SOUND
WARREN M. McGRATH*

The wedding of sight and sound in motion pictures was pioneered
in no small extent by newsreel soundmen and engineers. Since the
remarkable sound picture record of Lindbergh's takeoff on the .first
Trans-Atlantic flight early in 1927, a small group of intrepid field
soundmen have brought back a library of sounds and sound effects
that should remain a lasting tribute to their skill, ingenuity, and dar-
ing. The few field soundmen who remain continue to record sound
under acoustic conditions that would be the despair of the average
studio mixer.

It was inevitable that newsreel sound should pass through an era of
growing pains before settling down to a specific treatment acceptable
to all major newsreel producing companies. Since early 1932, the
commentary type of newsreel story has increased in popularity until
today it is Accepted as the most lucid manner in which to present

* Sound Engineer, Movietone News, New York.

372 PRODUCTION AND SIGNIFICANCE OF NEWSREELS Vol 47, No. 5

current events. This, of course, has resulted in a steady decrease in
the amount of natural sound recorded in the field and thus the work
of the newsreel synchronizer has become increasingly important. It
is through his efforts that commentary mixed with music and sound
effects, and an occasional interpolation of natural sound, results in a
pleasing composite sound track at a level constant throughout the
reel and unvarying from week to week.

The newsreel synchronizer, or recording engineer, handles the final
stage through which newsreel make-up proceeds. His work com-
mences when the film has been edited, music carefully selected from
a vast library of prerecorded tracks, and sound effects tracks and
script all prepared in final form. The tools of his trade are :

(1) An acoustically treated narration stage, equipped with a pickup micro
phone and a motion picture screen visible to the commentator and the mixer ;

(2) Several film rerecorders, or film phonographs, used for the rerecording of
music and sound effects;

(3) Several loop machines. These machines are rerecorders so arranged that
a continuous loop of sound track can be run through them during the scoring of a
picture and thus furnish a constant source of a particular sound, available to the
mixer whenever required ;

(4) Disk recorders and playbacks for the premixing of complicated sound
tracks when required;

(5) A recording console with its associated amplifiers, mixers, and equalizers ;
(6} A high-quality monitoring system;

(7) A film recorder;

(8) An interlock drive system which will furnish the motive power for all re-
recorders, loop machines, projection machines, disk machines, and film recorders,
and which will keep all of the machines being driven by the system in perfect
synchronism.

All of this equipment is maintained at a consistent high efficiency.
Routine measurements, gain runs, and film tests .are compared with
standard equipment data to insure a minimum of breakdowns and a
maximum of quality. Spare equipment units and a jack panel pro-
vide a flexible means by which the mixer may substitute apparatus,
cascade amplifiers, or introduce equalizers for a desired effect.

Newsreel subjects are infinite in their variety. A routine proce-
dure for mixing sound can have no application here. Each subject
must be handled with tact and discernment befitting its especial

Nov. 1946 PRODUCTION AND SIGNIFICANCE OF NEWSREELS 373

nature. The editorial department furnishes the mixer with a "spot"
sheet on which each scene of the newsreel subject is carefully listed in
its proper sequence. The spot sheet also indicates the desired sound
that is to be synchronized with the particular scene and the footage.
A comparison between the narrator's script and the spot sheet will
give a fairly close idea of the treatment the newsreel subject should
receive.

Two music tracks are provided for most subjects. The tracks are
prints of the same music negative but have "start" marks so placed
that one is synchronized to start with the beginning of the picture
and the other to finish with the end of the picture. The mixer must
use a suitable spot during the recording to change over from the first
music track to the second. This is usually done during sound effects,
natural sound, or long periods of narration in order to mask the opera-
tion. Of course, careful note must be made of the key in which the
particular part of the music track is played, as changing from one key
to another is instantly apparent. This system of using two identical
music tracks eliminates the necessity of having music passages re-
corded to a precise length.

The newsreel sound crew consisting of two soundmen, a projection-
ist, and the mixer, work as an efficient unit. Each man has his duties
and co-operates with his fellow department members to insure a
swift and efficient handling of the newsreel scoring. The mixer de-
pends upon the efforts of the machine room soundmen to thread
correctly the music and sound effects tracks in the rerecorders and
loop machines, and properly to thread and "sync-mark" the record-
ing film. The recordist, i. e., the soundman in charge of the film re-
corder, must also keep a careful check on the over-all recording level
and the recording lamp current; each man must carefully check the
machines assigned to him to insure their smooth operation. The pro-
jectionist's duties are too well known to enumerate here.

One and sometimes two rehearsals are required before the timing
and co-ordinating of all sound is mastered. During these rehearsals,
the mixer must find time to check the tonal quality of the narrator's
voice, his volume level, and the general level and synchronization of
the sound effects that are to be used. Important, too, is the spotting
of the story wherein the commentator and mixer carefully check the
script for timing. Each line of copy must be spoken at precisely the
spot for which it is intended and there must be a smooth transition
from narration to field recorded dialogue when required.

374 PRODUCTION AND SIGNIFICANCE OF NEWSREELS Vol 47, No. 5

Rehearsals completed, we are now ready for a take. , A swift re-
sume of the sound to be used on the subject might indicate that two
channels are required for music tracks, one for synchronized sound
effects, one for the continuous running loop machine, one for the pick-
up of field recorded sound from the picture film, and a narration
channel. Six channels which the mixer must manipulate in an aver-
age time of less than two minutes and with only two hands.

All recording for our national newsreel is done by the double-system
method wherein the sound is recorded on a separate piece of film than
that used for the picture. For this type of recording, and as we use a
variable-density type of recording, Eastman Type 1373 Fine-Grain
Recording Positive is used. Excellent quality is obtained when de-
veloped to a density of 0.55 and a gamma of 0.55. After much ex-
perimentation we have determined that little is to be gained by the
use of noise reduction when using this type of film. Although the film
is used solely for the sound track, it has been found advantageous
to print a picture image on it before developing in order to facilitate
the work of the editorial department and to aid the final check on
the sound recording work that has been done. The lavender picture
used by the mixer and narrator in scoring is used for this purpose,
and lining up the start mark on this with a corresponding one on the
sound track enables the laboratory to print a backward, negative,
"in-sync" image beside the sound track. At least once during the
scoring session the mixer must check with the laboratory to assure
himself that the sound track is properly exposed for density and
gamma heretofore mentioned. For this purpose a small strip of un-
modulated sound track is sent to the laboratory well in advance of
the start of the evening's recordings. The report returned by the
laboratory enables the mixer to correct the recording lamp current
accordingly.

From the foregoing it can be seen that the recording engineer's re-
sponsibilities are many and varied. Equally important, however, is
that sixth sense which for want of a better definition, can be called a
sense of timing. Newsreel subjects being essentially fast-moving and
of short duration, it is often necessary to bring in sound effects pre-
cisely on a frame. Then again, the time in which all recording must
be completed is very limited. Although some lengthy subjects have
taken as long as one hour to score, the average time taken by the re-
cording room, from the start of rehearsals to the completed take, is
less than 15 min.

Nov. 1946 PRODUCTION AND SIGNIFICANCE OF NEWSREELS 375

It is not possible to discuss the many electronic and mechanical de-
vices by which the recording room accomplishes the varied sound re-
cordings which it is called upon to produce in the course of a single
newsreel. Like any sound department, improvements are always
under way at Movietone News. To Earl Sponable and William
Jordan go unstinted praise in their constant pioneering in the elec-
tronic field. The installation of sound recording equipment at this
studio remains as a tribute to their skill and forethought.

CHARACTERISTICS AND APPLICATIONS OF CONCEN-
TRATED-ARC LAMPS*

W. D. BUCKINGHAM AND C. R. DEIBERT**

Summary. — The concentrated-arc lamp is an arc lamp provided with permanent
electrodes which are sealed into an argon-filled glass envelope. The light source is a
sharply defined luminous disk on the end of a specially prepared zirconium oxide
cathode. The radiation has a gray body distribution with the superimposed atomic
spectra of zirconium and argon. In the various sizes of lamps now made the light-
emitting spot ranges from 40 to 100 candles per sq mm in brightness and from 0.003
to 0.06 in. in diameter.

Small-sized lamps furnish a close approach to a point source and have application
in optical testing and demonstrating. Medium-sized lamps make increased detail
rendition and depth of focus possible in microscopy and the photographic enlarger.
Large-sized lamps are a pplicable in the field of projection.

The concentrated-arc lamp is a new type of light source that was
invented just prior to the war and developed during the war under a
contract issued through the Optics Division of the National Defense
Research Committee. The new lamp is an arc lamp, but differs
from the usual carbon arc in that it has permanent, fixed electrodes
which are sealed into a glass bulb filled with an inert gas. The name
"concentrated-arc" comes from a characteristic of the lamp which
makes it possible to concentrate the arc activity upon a small portion
of the electrode so as to produce a very high-intensity light source'in
the form of a luminous circular spot, which is fixed in position, sharply
defined and uniformly brilliant.

A line of standard size lamps has been developed in 2-, 10-, 25-,
and 100-w sizes. Pictures of these are shown in Fig. 1. Lamps have
been made in sizes as large as 1500 w, but they are considered ex-
perimental as yet, and their designs have not been standardized. The
actual physical sizes of the lamps shown in the. picture ra'nge from the
2-w lamp, which is 5/» m- m diameter and 2 in. high, to the 100-w lamp
which is 23/8 in. in diameter and 6 in. high.

* Presented May 9, 1946, at the Technical Conference in New York.
** The Western Union Telegraph Company, Electronics Division, Water Mill,
N. Y.

CONCENTRATED-ARC LAMPS

377

The actual source of the light is a flat circular luminous disk that
forms on the end of the specially prepared central wire, which is the
negative electrode or cathode. The diameter of this disk in the' 2-w
lamp is only 0.003 in. As the current is increased, the spot grows
larger so that a 100-w lamp has a spot O.OGO in. in diameter, while the
spot of a 1500-w lamp is 0.375 in. in diameter.

With a 2-w concentrated-arc lamp in operation, it is difficult to
believe that the source is but 0.003 in. in diameter because it is so
bright that the eye sees it as a disk of light which is apparently l/s in.
or more in diameter. If a dense welding filter is put in front of the
lamp, it is seen as a very tiny source.

FIG. 1.

Some of the characteristics of the lamps are tabulated in Fig. 2.
The brightness of the 2-w lamp is about 100 candles per sq mm. Or-
dinary tungsten filament lamps operate at about 10 candles per sq mm
and have a life of 1000 hr. By increasing the current through tung-
sten lamps, they can be -operated at brightnesses ranging up to 20 or
25 candles per sq mm but under these conditions, their life drops to
10 hr or less. Concentrated-arcs are thus several times brighter than
tungsten lamps and have a longer life.

The 2-w lamps have an average life of 175 hr. Larger lamps last
longer, 100-w lamps averaging 1000 hr. By average life it is meant
that if a number of 100-w concentrated-arc lamps were started to-

378

W. D. BUCKINGHAM AND C. R. DEIBERT Vol 47, No. 5

gether on life test, half of them would be good at the end of 1000 hr.
Since this is almost a year of continuous operation, and since life
tests cannot be accelerated, such data are collected very slowly.

The tabulation of Fig. 2 shows the brightness of the positive crater
of the ordinary carbon arc to be from 175 to 800 candles per sq mm,
and the sun to have a brightness of 1600 candles per sq mm. The
brightness of concentrated-arcs thus falls between that of tungsten
filament lamps and that of the carbon arc.

The internal construction of a typical, concentrated-arc lamp is
shown in the drawing of Fig. 3. The negative electrode or cathode is
the unique element of the new lamp. It is made by packing zirconium

LAMP RATING

0.003
0.060
0.375

2 WATT
100 WATT
1500 WATT

TUNGSTEN

CONG. -ARC

CARBON - ARC
LOW HIGH
INTENSITY

SUN

BRIGHTNESS
CANDLES /MM2

100

175

800

1600

LIFE -HOURS

1000

175 - 1000

FIG. 2.

oxide into the open end of a tube which is made of tungsten, molyb-
denum, or tantalum, these metals being selected because of their high
melting temperatures.

The positive electrode or anode, also made of a metal with a high
melting point, consists of a simple sheet or plate which has sufficient
radiating surface so that during operation, it will reach no more than
a dull red heat.

These two electrodes are mounted in the bulb so that the exposed
oxide surface of the cathode is but a few hundredths of an inch from
and directly behind a hole in the center of the anode. This hole is
slightly larger in diameter than the cathode tube and serves as a
window for the emergence of light from the cathode.

Nov. 1946

CONCENTRATED-ARC LAMPS

379

After the bulb has been evacuated, it is filled with an inert gas,
usually argon, to almost atmospheric pressure. The cathode is then

ANODE i PARTIALLY CUT AWAY )
CATHODE

CUTAWAY SECTION

ZIRCONIUM OXIDE CORE
TANTALUM TUBE

CRATER OR LIGHT SOURCE

FIG. 3. Internal elements of the concentrated-arc lamp.

put through a "forming" process. To do this, a high potential direct-
current source, with suitable current limiting resistors in series is
connected to the electrodes so that an arc strikes between the anode
and the metallic tube of the cathode. After a few seconds, the

LIGHT SOURCE

^$5$^^5^^^ / E$$$$$$$$$$$^^^

,/ j

TANTALUM TUBE

s<^r

' •.'•*-

ZIRCONIUM OXIDE
-^-— "~ CORE

FIG. 4. Cross section of an old concentrated-arc lamp.

cathode tube becomes red hot and heats the zirconium oxide packed in
it to a temperature where the oxide becomes electrically conductive.
The arc then strikes between the anode and the oxide and the heat of
the arc raises the temperature of the surface of the oxide to or above

380

W. D. BUCKINGHAM AND C. R. DEIBERT Vol 47, No. 5

its melting point of 3000 K. The molten oxide flows and bonds it-
self to the sides of the metal tube forming a smooth glassy surface.
In the molten state and under the intense ionic bombardment of
the arc, some of the zirconium oxide is reduced or decomposed to
metallic zirconium forming a very thin layer of this metal over the
surface of the cathode. Zirconium metal is a better electron emitter
at high temperatures than is the oxide, and it also has a lower melting
temperature; thus, as soon as the metallic zirconium surface layer is
formed, the temperature of the cathode drops slightly, and the under-
lying oxide solidifies and supports the film of molten metal on its
surface. It is this film of molten metal which is the chief source of the
visible radiation from the lamps. The film, once formed during man-

CONCENTRATED-ARC

270*

LUMENS-TTXC.P LUMENS-4TT xC.P

FIG. 5. Spatial distributions of concentrated-arc and tungtsen filament lamps

ufacture, remains to be heated and become incandescent whenever
the lamp is relighted. It is so thin that surface tension holds it to
the oxide backing so the lamps may be burned in any position.

This light source, consisting as it does of a thin metallic film sup-
ported by a refractory backing, has several unique advantages. Or-
dinary tungsten filament lamps can be made to give more light if
burned hotter, the radiation increasing as the fourth power of the ab-
solute temperature. A small increase in temperature thus produces
a comparatively large increase in radiation. This process is limited
in the tungsten filament lamp by the melting point of tungsten, for if
it is reached or even closely approched the lamp quickly burns out.
Concentrated-arc lamps are not so limited. In these lamps, the in-
candescent metallic light source can be operated and is operated at a
temperature which is above the melting point of the metal, thereby

Nov. 1946

CONCENTRATED- ARC LAMPS

381

producing light of a color quality similar to that which would be emit-
ted by tungsten if it could be operated at a temperature at or slightly
above its melting point.

A second' advantage of the concentrated-arc is its life characteristics
when operating at these high brilliancies. As the temperature of the
filament of a tungsten lamp is increased, lamp life decreases because
of evaporation of the filament material. Since the metal source of
the concentrated-arc lamp operates in a molten condition, it might be
expected that it, too, would evaporate.

Spectrograms taken of the portion of the arc stream very near the
cathode show the presence of very strong zirconium lines. This in-

ANODE GUpW

CATHODE GL0*
-

FIG. 6. Spectrogram of a 100-w concentrated-arc lamp. The diagram indicates
in what part of the arc stream the various spectra originate.

dicates that some evaporation of free zirconium occurs and under the
excitation of the arc, the characteristic spectrum is emitted. It is
found, however, that in addition to the normal zirconium spectrum,
the singly and doubly ionized zirconium spectra are also present.
Furthermore, there is practically no zirconium found in any portion
of the arc stream except that portion which is within a few thou-
sandths of an inch from the cathode surface.

These phenomena are explained as follows : An atom of zirconium
gains sufficient energy to leave the cathode surface and enters the
cathode glow region of the arc which extends for a few thousandths
of an inch from the cathode surface. Here, under the intense argon
ion bombardment, the zirconium atom has one or more electrons
knocked off of it, or in other words, it is ionized. In the normal atom,

382

W. D. BUCKINGHAM AND C. R. DEIBERT Vol 47, No. 5

the positive nuclear charge is just balanced by the negative charges
of the surrounding electrons so the atom as a whole is neutral.
When electrons are removed, as in the ionized atom, the atom is left
with a positive surplus and thus has a positive charge and is attracted
and drawn back to the negative cathode it just left. If any zirconium
atoms do escape permanently from the cathode, they are replaced by
reduction of the underlying oxide. As a result of these processes, the
lamps have lives which are measured in hundreds of hours.

During the normal life of a concentrated-arc lamp, the amount of
material evaporated from the cathode is so small and that little is so
well distributed by the strong convection currents in the argon gas
within the bulb that bulb blackening is not serious.

FIG. 7. Spectral distribution of radiation from a 100-w
concentrated -arc lamp.

Fig. 4 shows a sectional view of a cathode which has reached the
end of its useful life because of loss of material. The difficulty lies in
starting the lamp. Because of the shielding effect of the protruding
side walls, it is difficult to get the arc to strike to the zirconium sur-
face, it being more apt to strike to the side wall itself. If the arc does
strike to the zirconium surface, the lamp will operate satisfactorily
until it is turned off, then the same difficulty will be experienced when
it is turned on again. Warning that the end of the life of the lamp is
approaching is given by difficult starting.

An important characteristic of the lamps is shown by Fig. 5. Here
it is seen that the spatial distribution of light from concentrated-arcs
follows Lambert's law and has a cosine distribution. That is, the

Nov. 1<)4<»

CONCENTRATED-ARC LAMPS

383

light emitted in a given direction may be calculated by multiplying the
maximum candlepower by the cosine of the angle between the axis of
the cathode and the direction considered.

If a 100-w concentrated-arc lamp and a 100-w tungsten filament
lamp are measured with a foot-candle meter, it will be found that the
readings are nearly equal. Thus on a candlepower output per watt
input basis, they are similar. It must be pointed out, however, that
the light from the tungsten filament lamp is emitted in all directions,
as is shown in the drawing, while that from the concentrated-arc is
but in one general direction. When these volumes are integrated to
obtain the total light or lumen output of the lamps, it is found that
the concentrated-arc lamp has but one-fourth as many lumens as a

— J

. i

10 r-

3 f^* -S

' •

OOWATT

— 21 «

/^"\

)W
I A

aT
Mr

•

O.I J -

*H" •

NOTE

AR
OP

TOWS INO
IRATING C

GATE NORM
URRENTS

•^^

"2 WA

- LAM

:> .

1 1 1

AMPERES

1 1 1

| 3

1.0

FIG. 8. Change of light source diameter with current of
concentrated-arc lamp.

tungsten filament lamp of equal candlepower. It is for this reason
that concentrated-arc lamps are not recommended for general illu-
mination applications such as room lighting or flood lighting, but
rather for those uses where its small size or high brightness are of
major importance.

The cathode-current densities in the concentrated-arc lamp vary
from about 250 amp per sq cm for the 100-w lamp to about 900 amp
per sq cm for the 2-w lamp. Assuming the electron emitter to be
the thin zirconium layer at a temperature slightly below 3000 K, the
melting point of the oxide and using the constants commonly given
for a zirconium filament in a vacuum, values of electron emission are
obtained which are of the order of 500 amp per sq cm. This value is
entirely in line with the actual current densities found in the lamps

384

W. D. BUCKINGHAM AND C. R. DEIBERT Vol 47, No. 5

and seems to confirm the present belief that the arc is maintained
largely by thermionic emission, and that the active surface of the

FIG. 9. Change of maximum brightness with current of
concentrated-arc lamp.

cathode consists of a thin layer of zirconium atoms at a temperature
considerably above the normal melting point of the bulk metal.

The radiation from the concentrated-arc lamp appears to be divided
into three parts as follows :

FIG. 10. Average cathode brightness distribution of
concentrated -arc lamps.

(1) Continuous radiation from the molten cathode surface,

(2) Line radiation from the excited gas and vapor,

(3) Continuous radiation in the spectral region from at least 3500 A to 5000 A
originating in the excited gas and vapor.

Nov. 1 it-Mi

CONCENTRATED-ARC LAMPS

385

The existence of the three types of radiation is shown by the spec-
trograms in Fig. (). The three exposures were made from the cathode

FIG. 11. Change of candlepower with current of concen-
trated-arc lamps.

spot, cathode glow", and anode glow portions of the arc of a specially
constructed 100-w lamp. The cathode spot is by far the brightest
portion. Even though the exposure time for the three traces varied in
the ratio of 1:10:100, it is found that the trace of the cathode spot

<0
L
1 |

XMP

:: i

-/•

06O ' -

060 „

<r
0 48 fcl -

'25 WATT
LAMP

0 »

Vfl
M

-'//

= /

'^*

2 WAT1
LAMP,

^11

NOTE . ARROWS INDICATE NORMAL '

nil

PERAT

MG CURRENTS.

<^\

AMPERES _

• 1

0.1

i 1 1 1 MO

FIG. 12. Change of efficiency with current of concentrated-
arc lamps.

area is more than 10 times as intense as that of the cathode glow area.
Thus concentrated-arc lamps emit radiation from two main sources,
the white-hot zirconium cathode surface and the cloud of excited zir-
conium vapor and argon gas in the cathode glow region which extends

386

W. D. BUCKINGHAM AND C. R. DEIBERT Vol 47, No, 5

for a few thousandths of an inch from the cathode. The portion
which originates from the cathode surface has a continuous spectral
distribution. It extends in measurable amounts from 2500 A in the
ultraviolet, through the visible, reaching a maximum near 10,000 A
and on into the infrared. That portion of the radiation which comes
from the cloud of excited vapor and gas shows three principal spectra,
a continuum extending from the ultraviolet to about 5000 A, the
normal, singly, and doubly ionized zirconium spectrum and the normal
and singly ionized argon spectrum.

-CANOJ.E POWER
-SPOT DlAMFTtR

FIG. 13. Change of characteristics with age of 10-w
concentrated-arc lamps.

The spectral distribution characteristic shown in Fig. 7, thus rep-
resents the combination or sum of these several individual spectra.
Radiation shorter than 3000 A or longer than 5 microns is not trans-
mitted by the type of glass vused for the bulbs of standard-type lamps.

The diameter of the cathode spot of a given lamp depends upon the
current. If the current is increased, the spot slowly grows larger,
taking several seconds to adjust itself to the new condition. Fig. 8
shows how the diameter of the light sources vary from 0.05 mm to
3.5 mm as the lamp currents are changed in the several standard sizes
of lamps. While the lamps are designed to operate at a definite cur-
rent value, it is possible to adjust the spot size by changing the current.

One of the advantages of the concentrated-arc is its high brightness.
As shown by Fig. 9, the maximum brightness of standard lamps at their

Nov. 1946

CONCENTRATED-ARC LAMPS

387

normal opera ting current varies between 40 and 100 candles per sq mm.

To the eye, the cathode spot appears to have a uniform and con-
stant brilliance. Measurements show that the brightest part is near
the center. The average brightness variation across the spot, for the
several sizes of lamps, is shown in Fig. 10.

The candlepower increases with the current, maintaining an almost
linear relationship over a very wide range as shown by the curves of
Fig. 11.

The efficiency of concentrated-arc lamps, as measured in candle-
power per watt input to the lamp, varies between 0.15 for the 2-w
lamp to 0.8 for the 100-w lamp. This characteristic is shown in Fig.

•>

CO =

=0 u

HOURS

•

100

ipoo\

jogoo

FIG. 14. Mortality curves for 2- and 10-w concentrated-
arc lamps.

12. Comparable figures for tungsten filament lamps range from
0.54 for a 6-w lamp to 1.29 for 100-w lamps.

The average changes in the major characteristics of 10-w lamps
during aging are shown by the curves of Fig. 13. These show that
during the first few hours of running, the candlepower and light spot
diameter will decrease, while the maximum brilliance increases.
After about 100 hr of operation, these characteristics become rea-
sonably stable.

The mortality curve of Fig. 14 shows the average life of 2-w lamps
to be 175 hr. Similar data on larger lamps give 700, 800, and 1000
hr on 10-, 25-, and 100-w lamps although individual lamps have shown
lives up to 5000 hr. Failure is usually caused by loss or shrinkage of
the cathode filling material.

388

W. D. BUCKINGHAM AND C. R. DEIBERT Vol 47, No. 5

Concentrated-arc lamps have a negative volt-ampere characteristic
as is shown by the curves of Fig. 15. Consideration must be given
this fact in the design of their power supplies which will be considered
a little later in this paper.

While it is impossible to predict more than a few of the many spe-
cific uses which will be found for these new lamps, it is thought that
many of them can be put into three general classifications. The
first is the use of concentrated-arcs as point sources. Of course, there
is no such thing as a true point source, for if it has no area it must of
necessity be infinitely bright. The smaller sizes of concentrated-arcs
are a close approach, however, to point sources and many interesting
and useful things can be done with them.

BB\

-50 —

,VLAMP

2V
L

KA-
AN

; \

25V
Lt

/ATT
MP

\^^

00 V

WTT

'••^^

^-.

. JJ

."du

NOT

: ' ARROWS INDICATE NORMAL
OPERATING CURRENTS.
1 1 i 1 i i

AMPERES

1 1 II 10

FIG. 15. Volt-ampere characteristics of concentrated-arc
lamps.

Since the light rays radiate from what is almost a single point, the
lamps can be used to throw very sharp shadows and used as lensless
enlargers as shown by the shadow of the child's mitten in Fig. 16.
The lamp and mitten can be seen in the lower corner of the picture.
The shadow is projected on the wall with considerable enlargment,
but even so, the shadows of the fuzz of the yarn stand out clearly. It
will be noted that each tiny strand is outlined with a diffraction pat-
tern.

A small concentrated-arc makes an excellent source with which to
test lenses, adjust optical devices, and demonstrate lens aberrations
and other optical phenomena. Fig. 17 shows the caustics produced
when point source lamps are placed slightly off the axis and inside the

Nov. 1946

CONCENTRATED- ARC LAMPS

389

principal focus of two short focus plano-convex condensing lenses. A
lens designer would need a little time to calculate this figure. Many
interesting and instructive demonstrations and tests can be made with
these brilliant point sources.

A second general field of application of concentrated-arc lamps is
their use in conjunction with lenses. Fig. 18 shows a point source at
the principal focus of a condensing lens. Since the source is so small,

FIG. 16.

the rays leaving the lens are almost exactly parallel. Such an ar-
rangement makes an unusual contact printer for photography in which
close contact between the negative and print are not necessary.

A use of point source lamps which has even wider application is
shown in Fig. 19. This is the case where the point source is used as
the source of illumination in optical systems. The particular system
pictured is supposed to represent a photographic enlarger. Since the
source is a point, the rays leaving the condenser are substantially
parallel. Thus, the ray passing through point A on the negative or
film goes on to strike the enlarging lens at only point A on the lens

390 W. D. BUCKINGHAM AND C. R. DEIBERT Vol 47, No. 5

and passes through the lens to form the image on the screen. The
same sequence is followed by the rays passing through each point of
the film, the important fact being that there is but little scattering of
the rays at the film so the light from each microscopically small ele-
ment of the film passes through only one equally small area of the lens.
Thus for each elementary area, the lens acts as if it were stopped
down to an extremely small aperture, //200 or less, and forms an
image having the extreme sharpness and depth of focus which corre-
sponds to such an aperture, but there is not the corresponding loss of
light, for the whole lens is working. This explains why it is that,

FIG. 17.

when point source lamps are used with optical devices, images are
formed which have unusual definition and depth of focus.

In actual applications, the system might appear as shown in Fig.
20. Here the point source is so positioned in respect to the condens-
ing lens that the rays converge in leaving the condenser. This ar-
rangement results in exactly the same stopping effect as before and
has the advantage that the enlarging lens need not be as large as the
negative.

Fig. 21 shows the extreme depth of focus of the image projected
with a photographic enlarger equipped with a point source lamp as
compared to that obtained when the same enlarger uses a large source

Nov. 1946

CONCENTRATED- ARC LAMPS

391

PRINTING SURFACE

tungsten filament lamp. This group of pictures shows that when the
enlarger is focused for a 36-in. projection distance, a fairly well focused
image is found at a 27-in. projection distance when using the point
source. When a large source lamp is used in the same enlarger, the
image is very poor at this 27-in. position.

The design of a practical lens usually involves a series of compro-
mises. One error cannot be fully corrected without causing some
other to increase to serious magnitudes. Since many of the common
lens aberrations decrease as
the lens is stopped down,
and since using a point
source lamp with lenses pro-
duces a stopping effect, it
may be that a lens could
be designed particularly for
use with point sources and
produce results far superior
to those produced by any
lens now available.

The increased sharpness
and depth of focus which
result from the use* of point
source lamps in photographic
enlargers are also secured in
many other optical devices
when point source lamps are
substituted for the large
lamps normally employed.
In microscopy the results are
quite marked as shown by the
photomicrographs of Fig. 22.

The third general field of application of concentrated-arc lamps is
in projection. Fig. 23 diagrams a simple projector system. The
problem in many such systems is to get the maximum amount of light
from the source through a small opening such as the film gate and on
through the projection lens to the screen. Optically, the way to
get maximum light through a small opening is to image the source at
the opening or film gate. The upper view of Fig. 23 shows a projec-
tor using a concentrated-arc lamp source adjusted to this condition.
Since the concentrated-arc has a uniformly brilliant disk of light, its

POINT SOURCE

CONTACT PRINTER

FIG. 18. Contact printer.

392

W. D. BUCKINGHAM AND C. R. DEIBERT Vol 47, No. 5

image when placed at the film gate results in a uniformly illuminated
screen.

FIG. 19. Diagrammatic representation of a hypothetical photo-
graphic enlarger with point source illumination.

This adjustment cannot be used when a tungsten filament projec-
tion lamp is employed. In this case, the projector must be set up, as
shown in the lower part of Fig. 23, so the coils of hot tungsten are

p--

FIG. 20. Diagrammatic representation of a conventional photo-
graphic enlarger with point source illumination.

imaged not at the film gate but in front of the projection lens. If
they were imaged at the film gate, the image of the hot coils would
appear on the screen. Since so much light is lost at the gate under
the adjustments necessary with tungsten filament lamps, the effi-

Nov. 1946

CONCENTRATED-ARC LAMPS

393

ciency of transferring light to the screen is low. Concentrated-arc
lamps can thus be used much more efficiently in such projector sys-
tems.

A second factor in favor of the concentrated-arc in projection sys-
tems of this type is that the screen brightness is a direct function of
the source brightness. Since the new lamps are brighter than tung-
sten filament lamps, they also show an advantage from this stand-

mi ill
= in"
= n

25-WATT CONCENTRATED-ARC LAMP

He-ail

33" 30"

LENS TO EASEL DISTANCE

'212 PHOTO-ENLARGING LAMP

nilln

= n

Illili!

= 11^

iiiiliT

&i*s

36"

33" 30".

LENS TO EASEL DISTANCE

27"

FIG. 21. Enlargements illustrating depth of focus with //4.5, 2-in. lens
focused for 15 X magnification.

point. As a practical result of these advantages, it was found in a re-
cent test that a 100-w concentrated-arc lamp would put more lumens
on the screen through an 8-mm film projector than could be obtained
when a 500- w tungsten filament type projection lamp was used.

The largest size of concentrated-arc now in production is the 100-w
lamp. This lamp has a source 0.060 in. in diameter. Using a good
condenser system, it is possible to magnify this small spot to cover
the film gate of an 8-mm film projector.

An experimental concentrated-arc lamp which has a source spot
.diameter of 3/ie in. and operates on 450 w has been made for 16-mm

394

W. D. BUCKINGHAM AND C. R. DEIBERT Vol 47, No. 5

projectors. Fig. 24 shows an experimental 1500-w concentrated-arc
lamp which, with its 4000 cp coming from a spot 3/8 in. in diameter,
seems to be adapted for use in 35-mm film projectors.

There are probably many uses which will be found for the new con-
centrated-arc lamps; the few which have been discussed are intended
only to show the peculiar advantages of the lamps in several general
types of applications.

The usual type of concentrated-arc lamp requires a high volt-
age pulse to break down the gap between the anode and the cathode
and to establish the arc and a supply of direct current to maintain the

FIG. 22. Photomicrographs illustrating the increased detail rendition
and depth of field made possible by the concentrated-arc in A as compared
with conventional illumination in B.

arc. In the laboratory, this is easily accomplished with the simple
circuit shown in Fig. 25. Here the lamp is connected through a re-
sistance and radio frequency choke coil to the 110-v, d-c supply line
taking care to connect the lamp with correct polarity. When a hand
held spark coil, sometimes known as a Tesla coil, vacuum leak tester
or violet ray coil, is touched to the lamp lead, the high-frequency
spark jumps the gap in the lamp and the arc is established on the
direct current from the line. This circuit is suitable for 10-, 25-, and
100-w lamps. Two- watt lamps can be operated in a like manner if
the supply main has a potential of 200 v or over. In every case, the
series resistance must be adjusted to limit the current through the
lamps to the value recommended for the particular size of lamp used.
A convenient power supply for a 2-w lamp is shown in Fig. 26.
This operates from 110 v, a-c and with a simple voltage doubler rec-

Nov. 1946

CONCENTRATED- ARC LAMPS

395

tifier delivers 220 v, d-c to the lamp. The high starting voltage is
secured with the aid of the series choke coil and a momentary contact
switch which is connected across the lamp terminals. When this
switch is operated, the inductive pulse generated by the choke coil is
sufficient to start the lamp.

Fig. 27 pictures a group of power supplies which operate from 110 v,
a-c and supply the necessary high voltage starting and direct-current

PL»NE_OFF

CONCtNTBtUO A»C

FIG. 23. Projection systems with concentrated-arc lamp (top) and
with tungsten filament lamp (bottom).

running power for the various size lamps. The general type of cir-
cuit used is shown in Fig. 28. This circuit combines two rectifiers, a
high voltage transformer and vacuum tube rectifier for the starting
potential and a low voltage transformer and selenium rectifier for
the running current. A relay automatically switches the lamp from
one to the other as soon as the arc strikes. This type of power supply
is very satisfactory from an operating standpoint, but is heavy, bulky
and comparatively expensive.

An alternating-current type of concentrated-arc lamp is being de-
veloped to avoid these difficulties. The direct-current type lamp

396

W. D. BUCKINGHAM AND C. R. DEIBERT Vol 47, No. 5

consists of the cathode and a single anode. The alternating-current
type has two anodes which, when properly connected to a center
tapped transformer, allow the lamp to rectify within itself the current
flowing through the cathode. Incidentally, the lamps make such
good rectifiers that they may eventually find their major use in this-
field, rather than that of illumination. The necessity of a high volt-
age for starting is avoided in the alternating-current type lamp by
the use of an auxiliary tungsten filament which is built into the lamp

FIG. 24. An experimental 1500-w concentrated-arc lamp.

This filament is heated for an instant during the starting sequence
and provides the ionization necessary to establish the arc, the whole
process requiring but a fraction of a second.

The appearance of the power supply for the new 100-w, a-c type
concentrated-arc lamp is shown in Fig. 29. As compared to the
power supply for the direct-current type lamp, the new supply has
the advantage of less than one -half the weight, one-third the bulk,
and one-quarter the cost. The circuit diagram of the supply for the
alternating-current lamp is shown in Fig. 30. The comparative
simplicity of this new power supply is apparent.

Nov. 1946

CONCENTRATED-ARC LAMPS

397

115

CONCENTRATED-ARC LAMP

FIG. 25. A circuit for running lamps from direct-current mains.
The tesla coil and RF choke are included for starting purposes.

i i

— 8 MFO. 450V. i/ mm ^
1 t

<
I • MFD 450V. <

i o^^^fi^^^^ • — i

FIG. 26. Schematic diagram of a power supply for 2-w concen-
trated-arc lamps.

FIG. 27 Concentrated-arc lamp power supplies.

398

W. D. BUCKINGHAM AND C. R. DEIBERT Vol 47, No. 5

In starting this lamp, first, the filament is lighted, then the arc
strikes between the filament and the cathode, heating the cathode.

FIG. 28. Schematic diagram of a 25-w concentrated-arc lamp
power supply.

Finally, the filament is turned off and the thin zirconium surface of
the cathode is maintained in a molten condition by the arcs from the
two anodes working alternately. If the lamp is turned off for only
an instant, it may be restarted without the use of the hot filament.

FIG. 29. A power supply for a 100-w, a-c type concentrated-arc

lamp.

If the interruption is long enough to allow the cathode to cool too

much, the lamp will not restart without the aid of the heated filament.

As the arc is turned off and on again, the boundaries of the molten

pool spread out. It should also be noted that the bright zirconium

Nov. 1946

CONCENTRATED- ARC LAMPS

399

pool always remains in the same position, is sharply outlined, and
appears to be uniformally bright.

The lamp can be burned in any position and moved around in any
desired manner. In this respect, it is quite different from the usual
tungsten filament projection lamp which must be burned in only the
specified position and is very sensitive to motion and shocks.

FIG. 30. Schematic diagram of a power supply for 100-w
a-c type concentrated-arc lamps.

In this discussion, an attempt has been made to show what the
concentrated-arc lamp is , how it works, what its characteristics are,
and some of the things it can be used for. Its characteristics and
properties are so different from those of any other type of lamp now
available, and the results which can be obtained with it are so unique
that it is hoped that concentrated-arc lamps will prove to be useful
in the solution of many of the problems of science and industry.

OPTICAL PROBLEMS OF THE IMAGE FORMATION
IN HIGH-SPEED MOTION PICTURE CAMERAS*

JOHN KUDAR**

Summary. — The optical design of high-speed cameras can be improved by con-
sidering the results of a systematic analysis of the various optical aberrations which
arise with the rotation of the polygonal refracting prism.

High-speed motion picture cameras with rotating polygonal prisms
have recently been described.1-2 The nonlinearity of the parallel
displacement of the image formed by the rotating prism (plane-
parallel plate) results in optical aberrations which must be kept within
reasonable limits.3 This is achieved in high-speed cameras by using
a rotating shutter represented by dark parts between the consecutive
prism faces. However, the rotating plane-parallel plate produces
several other aberrations which, although less obvious but much more
intricate than the nonlinearity of the parallel displacement of a ray,
must nevertheless be considered in the optical design of high-speed
cameras.

The optical aberrations produced by the rotating plane-parallel
plate can be classified as analogies to the well-known lens aberrations
as follows.

Lens calculations based upon the approximation that sin x = x —
xz/6 give account of spherical aberration, coma, curvature of field,
astigmatism, and distortion. The camera lens and the polygonal prism
behind it, in the position in which the optical axis of the lens is per-
pendicular to two parallel faces of the polygonal prism, can be de-
signed of course as a well-corrected lens system. Then the rotation
of the prism produces periodic aberrations, which have no axial sym-
metry and are to be related to the tangential and sagittal sections of
the rotating prism. These prismatic aberrations are analogous to the
five lens aberrations. For instance, the nonlinearity of the parallel

* Submitted Aug. 3, 1946.
** London, England.

400

OPTICAL PROBLEMS 401

displacement of a ray is an effect corresponding to distortion in lens
optics.

There is a prismatic astigmatism, too; the field curvature of lens
optics corresponds here to the varying positions of the astigmatic
image planes. Finally, the periodically varying prismatic coma is
also a very interesting aberration of great practical importance.
The analogy between these periodic aberrations caused by the rotat-
ing prism and the axially symmetrical aberrations of lens optics be-
comes immediately obvious by considering the variation of the former
during the rotation of the prism. Thus the axially symmetrical aber-
rations caused by the prism, when its two faces are just perpendicular
to the optical axis, will change continuously to the tangential and
sagittal prismatic aberrations. ' However, the spherical aberration of
the prism is independent of the rotation. It is sufficient, therefore,
as far as -spherical aberration is concerned, to achieve the optical
correction for the camera lens and the prism together.

It is of practical importance to compare the numerical values of the
varying prismatic aberrations. This task involves detailed optical
calculations4 the results of which regarding high-speed cameras can be
summarized as follows.

Let D represent the distance between parallel faces of the rotating
prism, n the refractive index of the prism, x the angle of incidence of
the principal ray belonging to any image point, r and a the half
angular aperture of the rays participating in the formation of any
image point in tangential and sagittal prismatic sections.

The prismatic coma is an aberration perpendicular to the direction
of the principal ray and to the prism axis. The numerical value of
the tangential coma is

-" to

and the sagittal coma : •

The tangential and the sagittal image planes vary continuously
their positions along the principal ray during the rotation of the prism
and the distance between them is the numerical value of the prismatic
astigmatism :

H + l-X* (3}

•V • l*^/

402 J. KUDAR

The nonlinear term in the parallel displacement of the principal ray
is

:'. (4)

One can easily obtain approximate information about the average
values of these aberrations in considering principal rays parallel to
the axis of the lens system. Thus the angle of incidence x equals the
angle of rotation of the prism starting from the position in which its
faces are perpendicular to the axis of the lens system.

Comparing Eqs (1), (2), (3) with (4) in practical cases in which the
maximum angle of rotation of the prism2 may be x = 1/5 ( = 1 1 l/* deg.)
and an aperture //2 might be used,1 prismatic coma and astigmatism
result in much greater aberrations than the nonlinearity (4) could ever
produce. Thus the correction of the latter by suitable dimensioning
of D does not mean necessarily that the image formation will be of
perfect quality regarding the other aberrations.

Careful consideration of these results in the optical design of high-
speed cameras could open the possibility for improvements which may
be of practical importance.

REFERENCES

1 WADDELL, J. H.: "A Wide Angle 35-Mm High-Speed Motion Picture Cam-
era," /. Soc. Mot. Pict. Eng., 46, 2 (Feb. 1946), p. 87.

2 SMITH, H. J.: "8000 Pictures per Second," J. Soc. Mot. Pict. Eng., 45, 3
(Sept. 1945), p. 171.

* TAYLOR, H. D. : Proc. Phys. Soc., 49, (1937), p. 663.

4 KUDAR, J.: "Optical Problems of the Rotating Prism Cinematograph Pro-
jector," Proc. Phys. Soc. 58, (Sept. 1946),p. 598.

AN IMPROVED METHOD FOR THE DETERMINATION OF

HYDROQUINONE AND METOL IN PHOTOGRAPHIC

DEVELOPERS*

H. L. BAUMBACH**

Summary. — A new method for the quantitative chemical analysis of hydroqui-
none and metol is suggested that is more rapid and, in some cases, more accurate than
previous methods.

Molecular hydroquinone and methyl p-amino phenol are extracted from the de-
veloper at pH 8.0 to 8.5 with methyl acetate. The extract is dissolved directly in
water and titrated with hydrochloric acid to determine metol, and then with iodine at
pH 6.5 to 7.0 to determine the sum of the metol and the hydroquinone.

Introduction. — Accurate chemical analyses of photographic
developing solutions are a very necessary part of the chemical
control of continuously replenished developers. The analytical de-
terminations of bromide and sulfite and the measurement of £H are
practically as rapid and accurate as could be desired. The deter-
mination of hydroquinone and metol, however, has been time con-
suming and, in some cases, not sufficiently accurate. For example,
many negative solutions utilize a low concentration of metol in the
presence of a high concentration of hydroquinone. At the pH values
at which such developers operate, metol is practically the only effec-
tive developing agent, and its low concentration needs to be deter-
mined accurately, often to =*=0.01 gram per liter.

Present methods of analysis for hydroquinone and metol are usually
based upon an extraction of the hydroquinone alone from the acidified
developer, using an immiscible organic solvent, such as ethyl ether,
and a second extraction of hydroquinone and metol base from another
sample of developer, which, in this case, has been adjusted to a pH
of about 8.5. The metol concentration is thus determined by differ-
ence. Extraction of the hydroquinone from the developer sample at
a pH of about three and subsequent extraction of the same sample for

* Presented May 6, 1946, at the Technical Conference in New York.
** West Coast Laboratory, Paramount Pictures, Inc., Hollywood.

403

404 H. L. BAUMBACH Vol 47, No. 5

metol at £H 8.5 is not always practical, since metol base is easily
oxidized by the air, in the absence of hydroquinone.

The methods of Lehmann and Tausch,1 Evans and Hanson,2
Baumbach,3 Atkinson and Shaner,4 Evans, Hanson and Glasoe,5 and
Stott6 have all been based upon this idea of two separate extracts at
two pH. values.

In an effort to shorten the analytical methods and to obtain more
accuracy in the determination of metol, research performed in this
laboratory has resulted in a new procedure which permits hydro-
quinone and metol base to be determined in a single extract. A new
solvent has been selected which has the property of being only slightly
soluble in a salted developer and yet very soluble in pure water,
thereby eliminating the need for evaporating the extracting solvent
in order to obtain the hydroquinone and the metol base in a form
suitable for analysis. This solvent also possesses a high extraction
coefficient for both hydroquinone and metol base, being superior to
ethyl ether in this respect.

Selection of a Suitable Solvent. — In the course of the testing of
a number of solvents that might be used to extract hydroquinone
and metol base from developing solutions, one solvent was found
that possessed the desired properties to a high degree. Methyl ace-
tate has the ability to extract 70 per cent of the hydroquinone and 80
per cent of the metol from an equal initial volume of typical develop-
ing solution at a pH. of §.5 and at a temperature of 70 F. When the
developer is saturated with potassium bromide, over 90 per cent of
both agents is extracted by an equal initial volume of the solvent.
Two such extractions result in less than one per cent of each agent
remaining in the developer. Methyl acetate does not extract de-
veloping agent oxidation products (sulfonates) from a developer con-
taining sulfite nor at this pH is there any sulfur dioxide extracted.

There are several advantages to the use of methyl acetate in place
of ethyl ether. Methyl acetate is a somewhat safer solvent to use,
since its flash point is 20 F and its lower explosive limit is 4.1 per cent
by volume in air, while the flash point of ethyl ether is — 20 F with a
lower explosive limit of 1.7 per cent. The lower vapor pressure of
methyl acetate likewise makes this solvent superior to the use of
ethyl ether. The solubility of methyl acetate in water is consider-
able, being 32 grams per 100 ml of water at room temperature.
This solubility enables the extract to be dissolved directly in water for
analysis and obviates the need for solvent evaporation.

Nov. 1946 IMPROVED METHOD FOR PHOTOGRAPHIC DEVELOPERS 405

The Determination of Hydroquinone and Metol in a Common
Solution. — Methyl />-amino phenol is amphoteric ; it can act
either as an acid or a base. On the other hand, hydroquinone
possesses only hydroxyl groups and can function only as an acid.
Therefore, metol base can be determined in the presence of a large
excess of hydroquinone by titrating the mixture with strong acid.
Fig. 1 shows the relationship between pH and the volume of standard
hydrochloric acid during the course of the titration of a dilute solu-
tion of metol base.

Since the basic properties of metol are so mild, the use of a color
indicator in such a titration will not yield results of sufficient accuracy
for most purposes. The end-
point can be determined ac-
curately, however, by the usual
potentiometric procedure, the
continuous reading pH meters
being very satisfactory for this
purpose. Unfortunately, a pure
water solution of metol base per-
mits the latter quickly to be oxi-
dized by the dissolved oxygen,
unless precautions are taken to
provide an inert atmosphere.
This is especially true at pH
values above six. The rate of
oxidation can be greatly reduced,
however, by reducing the polarity
of the solvent. Thus metol base is relatively stable in a solvent
mixture of 75 per cent water and 25 per cent isopropyl alco-
hol, or in the same percentages of water and methyl acetate.
Decreasing the polarity of the solvent also decreases the sharp-
ness of the inflection for the acid titration, as is shown in Fig. 2, but
the titration is still a practical one. The pH at which the inflection
occurs is a function of the concentration of the metol base ; therefore,
it is necessary to plot a titration curve for an unknown developing
solution. For control purposes, where the concentration of metol
in a developer will not vary greatly, it is practical to titrate the ex-
tract with acid to a given pH. Fig. 2 shows the pH values of the in-
flections for developer concentrations of metol, if the procedure that
is described below is followed.

I 2

MLS O.IN HCI

FIG. 1. Titration with hydro-
chloric acid of metol base in water
solution.

406

H. L. BAUMBACH

Vol 47, No. 5

After the titration of the metol base with acid, the same solution
may be titrated with standard iodine solution. Iodine oxidizes hy-
droquinone and metol base quantitatively to quinone and methyl
quinone imide, respectively, if the resulting hydriodic acid is neu-
tralized with alkali as the reaction progresses. At £H values much
below seven, the reaction does not go to completion and at pH values
in excess of eight, the reactions between oxygen and the reducing
agents are rapid enough to cause errors.

I 2

MLS O.I N HCI

FIG. 2. Titrations with hydrochloric acid of metol
base in a solution of 25 per cent methyl acetate and
75 per cent water. Numerals refer to concentrations of
metol in the original developer samples.

Procedure. — Pipette 25.0 ml of the developer sample into a
250-ml separatory funnel. Add 0.5 ml of thymol blue indicator
solution and neutralize the alkali of the developer with concentrated
hydrochloric acid until the blue color just changes to yellow, or ad-
just the pH of the developer sample between 8.0 and 8.5 by any other
means.

Add 15 grams of fine, granular potassium bromide and 25 ml of
methyl acetate. Shake the funnel vigorously for 3 min and allow
the layers to separate for 3 min. Drain the water into a clean beaker
and pour the organic layer into a clean, dry 100-ml beaker. Return
the water layer to the separatory funnel and rinse the beaker with
25 ml more methyl acetate, which should be added to the funnel.

Nov. 1946 IMPROVED METHOD FOR PHOTOGRAPHIC DEVELOPERS 407

Repeat the shaking and the separation, using the second methyl
acetate extract to rinse the surfaces that held the first.

Pour the extract of the first separation successively through two
additional dry 100-ml beakers, in order to make the removal of water
complete, and finally place the extract into a 400-ml beaker. After
the second extract has followed the first, add 150 ml of distilled water
and mount the beaker into a titration unit that provides the calomel
and glass electrodes, burettes for the hydrochloric acid and the iodine,
and a motor stirrer.

Titrate the solution containing methyl acetate, hydroquinone and
metol base with 0.0500 N hydrochloric acid, plotting a titration curve
of pH versus volume of reagent or, in routine work, titrate to the
specific pH corresponding to the point of inflection.

Add an additional 150 ml of water with 10 ml of starch indicator
and titrate with 0.100 N iodine solution. During the titration, add
sufficient 10 per cent disodium phosphate solution to maintain a £H
between 6.5 and 7.0. The end-point is taken when a stable bluish
color is produced.

CALCULATIONS

ml of 0.0500 N HC1 X 0.344 = grams per liter of metol in developer,
(ml of 0.100 N KI3-ml of 0.0500 N HC1) X 0.220 = grams per liter of hydro-
quinone in developer.

NOTES

(1) Some samples of methyl acetate contain sufficient methyl alcohol to make
the mixture completely miscible with water. Such material is not suitable for
this analysis. If approximately two-thirds of the methyl acetate remains un-
dissolved after shaking with an equal volume of water, the material is satisfactory.

(2) Potassium bromide is used to "salt out" the developing agents because this
salt raises the density of the water layer to a degree that permits rapid separation
of the two phases after they have been intimately mixed.

REFERENCES

1 LEHMANN, E., AND TAUSCH, E.: "Zum Chemismus der Metol-Hydrochinon
entwicklung," Photographische Kor respondent, 71 (Feb. 1935), p. 17.

2 EVANS, R. M., AND HANSON, W. T., JR.: "Chemical Analysis of an MQ De-
veloper," /. Soc. Mot. Pict. Eng., XXXII, 3 (Mar. 1939), p. 307.

3 BAUMBACH, H. L.: "The Chemical Analysis of Metol, Hydroquinone, and
Bromide in a Photographic Developer," /. Soc. Mot. Pict. Eng., XXXIII, 5
(Nov. 1939), p. 517.

4 ATKINSON, R. B., AND SHANER, V. C.: "Chemical Analysis of Photographic

408 H. L. BAUMBACH

Developers and Fixing Baths," J. Soc. Mot. Pict. Eng., XXXIV, 5 (May 1940), p.
485.

5 EVANS, R. M., HANSON, W. T., JR., AND GLASOE, P. K.: "Synthetic Aged
Developers by Analysis," /. Soc. Mot. Pict. Eng., XXXVIII, 2 (Feb. 1942), p. 188.

6 STOTT, J. G.: "The Application of Potentiometric Methods to Developer
Analysis," /. Soc. Mot. Pict. Eng., XXXIX, 1 (July 1942), p. 37.

APPLICATION OF METHYL ETHYL KETONE TO THE

ANALYSIS OF DEVELOPERS FOR ELON

AND HYDROQUINONE*

VAUGHN C. SHANER** and MARY R. SPARKS**

Summary. — A method of analysis for Elon and hydroquinone in developers is
described involving the use of methyl ethyl ketone as the extracting solvent. Tests
showed it to be a better extracting solvent for Elon and hydroquinone than methyl ace-
tate or ethyl acetate. Analyses made employing methyl ethyl ketone with the U-tube
extraction method showed it to have the necessary accuracy and reproducibility for
use in production control.

During the past few years, photographic processing control has be-
come increasingly important. Many processing laboratories employ
chemists for the express purpose of analyzing their developers daily
in order that the exact chemical concentrations of the constituents
may be known at all times. For this reason there has been a constant
search for ways to improve the existing methods of developer analysis
and many articles have been published describing various ways of de-
termining the concentrations of developing agents. Evans and
Hanson1 published a colorimetric method, and Baumbach2 revealed
a volumetric method of determination for Elon and hydroquinone.
Atkinson and Shaner3 used a volumetric method involving extraction
of Elon and hydroquinone in a £7- tube ; Stott4 described a potentio-
metric method ; Evans, Hanson, and Glasoe5 employed a polarogra-
phic method for determination of Elon and hydroquinone. Recently,
Baumbach has described a method of developing agent analysis in-
volving the extraction of Elon and hydroquinone with methyl ace-
tate. In his method, concentrations of Elon and hydroquinone are
determined in the same solution by acid titration of Elon, using a
pH meter with a glass electrode as the end-point indicator and then
an iodine titration of both Elon and hydroquinone. This method is
satisfactory from the standpoint of convenience and accuracy but

* Presented May 6, 1946, at the Technical Conference in New York.
** Motion Picture Film Dept., Eastman Kodak Company, Hollywood.

409

410

V. C. SHANER AND M. R. SPARKS

Vol 47, No. 5

difficulty was encountered in obtaining methyl acetate of sufficient
purity for the analysis. For this reason, the authors set out to find a
new extracting solvent for Elon and hydroquinone.

Procedure. — In the analytical procedure described by Baum-
bach, a 25-ml sample of developer is pipetted into a 250-ml separa-
tory funnel. Add 0.5 ml of thymol blue indicator solution and
sulfuric acid until the color just turns yellow. Add 15 grams of
potassium bromide. This is necessary because methyl acetate is

^ 5

1212
ML O.I N HCL

FIG. 1. Elon titration curves, model G
Beckman />H meter, model 015 glass electrode
—(A) 50 ml methyl ethyl ketone extract 200
ml water; (B) 50 ml methyl ethyl ketone ex-
tract 10 ml methanol 10 ml water.

quite soluble in water which is not saturated with salt. Add 25 ml
of methyl acetate and shake 3 min. Let stand 3 min, and separate
into three dry 100-ml beakers in series. Repeat the extraction step
with an added 25-ml portion of methyl acetate, and combine the two
extracts in a 400-ml beaker. Add 150 ml of distilled water. Titrate
the Elon base with 0.1 N hydrochloric acid, using a glass electrode
as the end-point indicator. Plot the titration curve of £H versus
milliliters of hydrochloric acid. Add 150 ml of distilled water, and
fresh starch solution. Titrate slowly with 0.1 N iodine, after the pH
is raised to a value of 7.0 by the addition of a buffer, such as disodium
phosphate. Compute the hydroquinone concentration by the differ-

Nov. 1946 APPLICATION OF METHYL ETHYL KETONE 411

ence between the Elon titration and the total as determined by the
iodine titration.

In this laboratory it was found more convenient to employ a £/-tube
extractor of the type used by Atkinson and Shaner rather than a
separatory funnel. It was deemed desirable, therefore, to adapt
Baumbach's procedure to this £/-tube extractor. A developer sample
volume of 10 ml has been found about the largest practical to use
with a £7-tube extractor. However, it was discovered that the large
water dilution of the methyl acetate extract prior to the Elon titra-
tion led to a curve, the end-point of which was hard to determine.
Preliminary tests showed that this large dilution could be avoided
by the use of 10 ml of methyl alcohol and 10 ml of water to make the
solution sufficiently polar for the titration. In Fig. 1 are reproduced
titration curves which show the comparison of titrations carried out
with and without methyl alcohol. Curve A was obtained from the
titration of solvent extract with a large dilution of water, while curve
B was obtained from the titration of solvent extract with only 10 ml
of water and 10 ml of methyl alcohol. By reference to the figure it
may be seen that curve B has a precise inflection point, making it
possible to read the end-point of the titration more easily. Accord-
ingly, an experimental procedure was set up to include these modi-
fications.

Pipet a 10-ml sample of developer into the funnel attached to a £7-tube extrac-
tor. Add 5 drops of 0.04 per cent thymol blue indicator dye and one to one sul-
furic acid until the color just turns yellow. Saturate the developer sample with
potassium bromide. Run 50 ml of solvent through the extractor and collect at
the delivery end of the £7-tube in a dry graduate cylinder. Pour the solvent ex-
tract into a mixture of 10 ml of distilled water and enough methanol to make the
solution miscible. Titrate with 0.1 N hydrochloric acid, using a pH meter with a
glass-calomel electrode system as an end-point indicator, and plot the titration
curve. Compute the Elon concentration as follows:

(ml Hcl X N Hcl) X 172

Elon in grams per liter = — — .

ml of sample

Add 200 ml of distilled water, starch solution and disodium phosphate crystals to
bring the pH to 8.0 and titrate the combined Elon and hydroquinone with 0.1 N
iodine. Compute the hydroquinone concentration as follows:

2 X NI2)
Hydroquinone in grams per liter =

ml of sample

-P

(ml Hcl XN Hen

— & — - — Xoo.

ml of sample

412

V. C. SHANER AND M. R. SPARKS

Vol 47, No. 5

Tests of Solvents. — The solvents selected for the tests in this
work were methyl acetate, ethyl acetate, and methyl ethyl ketone.
A developer with high concentrations of Elon and hydroquinone
was mixed according to the following formula:

Elon

Hydroquinone
Sodium sulfite
Sodium carbonate
Potassium bromide
Water to make

3.3 grams

9.5 grams
40.0 grams
20.0 grams

2.0 grams

1 liter

Samples of this developer were extracted according to the pro-
cedure just given. The 50-ml portions of solvents were collected and
analyzed in 10-ml aliquots to determine the rate of extraction of Elon
and hydroquinone by each of the different solvents. In order to bring
the solution to workable volume it was therefore necessary to use 50
ml of distilled water. It was necessary to use 25 ml of methanol with
ethyl acetate but 10 ml of methanol was sufficient for methyl acetate
and methyl ethyl ketone. The data thus collected are shown in Tables
land 2.

TABLE 1

Elon Found in Grams per Liter

Extraction
Number

Total Volume
of Solvent
(in ml)

Methyl Ethyl
Ketone

Ethyl
Acetate

Methyl
Acetate

1.84

1.29

1.70

2.96

2.03

2.52

3.15

2.58

3.07

3.28

3.05

3.25

3.28

3.19

3.25

Extraction
Number

2
3
4
5

TABLE 2

Hydroquinone Found in Grams per Liter

Total Volume

of Solvent

(in ml)

10
20

30
40
50

Methyl Ethyl
Ketone

5.35

8.70
9.14
9.28
9.28

Ethyl
Acetate

3.31
5.52

7.04
8.15
8.67

Methyl
Acetate

5.04
7.17
8.49
8.97
8.97

Nov. 1946 APPLICATION OF METHYL ETHYL KETONE 413

By referring to Table 1 it may be seen that methyl ethyl ketone
extracts 1.84 grams per liter of Elon, ethyl acetate extracts 1 .29 grams
per liter, and methyl acetate extracts 1.70 grams per liter of Elon,
in 10 ml of solvent. Similarly, in 20 ml of solvent, methyl ethyl ke-
tone extracts 2.96 grams per liter of Elon, ethyl acetate extracts 2.03
grams per liter of Elon, and methyl acetate extracts 2.52 grams per
liter of Elon. From these data, it was decided that methyl ethyl ke-
tone is a better extracting solvent for Elon than methyl acetate or
ethyl acetate. Since 50 ml of methyl ethyl ketone extracted no more
Elon than 40 ml of methyl ethyl ketone, it was concluded that 50 ml
of methyl ethyl ketone is a sufficiently large volume for extracting all
the Elon present in a 10-ml developer sample.

From Table 2, it may be seen that, in a 10-ml volume, methyl ethyl
ketone extracted 5.35 grams per liter of hydroquinone, ethyl acetate
extracted 3.31 grams per liter of hydroquinone, and methyl acetate
extracted 5.04 grams per liter of hydroquinone. Similarly, in a 20-
ml volume, methyl ethyl ketone extracted 8.70 grams per liter of
hydroquinone, ethyl acetate extracted 5.52 grams per liter of hydro-
quinone, and methyl acetate extracted 7.17 grams per liter of hydro-
quinone. From these data it was decided that methyl ethyl ketone
is a better extracting solvent for hydroquinone than methyl acetate
or ethyl acetate. Since 50 ml of methyl ethyl ketone extracted no
more hydroquinone than 40 ml of methyl ethyl ketone, it was con-
cluded that 50 ml of methyl ethyl ketone is a sufficiently large volume
to extract all the hydroquinone present in a 10-ml sample of a prac-
tical developer with a safety factor included.

TABLE 3

Methyl Ethyl Methyl Ethyl

Solven Ketone Acetate Acetate

Weight of Flask 78.360 75.260 71.350

and Residue

Weight of Flask 63 . 970 65 . 4 10 63 . 892

Grams of Hydro- 14.390 9.850 7.458

quinone

In order to compare the amounts of hydroquinone which could be
dissolved in methyl acetate, ethyl acetate, and methyl ethyl ketone,
50-ml portions of each solvent were saturated with hydroquinone.
The solvent was evaporated to dryness and the residual hydroquinone
weighed. The data thus collected are shown in Table 3.

414

V. C. vSHANER AND M. R. SPARKS

Vol 47, No. 5

By inspection of Table 3 it may be seen that 50 ml of methyl ethyl
ketone will dissolve 14.39 grams of hydroquinone while 50 ml of
methyl acetate will dissolve 9.85 grams of hydroquinone, and 50 ml
of ethyl acetate will dissolve 7.46 grams of hydroquinone. From these
data it is concluded that methyl ethyl ketone is a better solvent for
hydroquinone than is methyl acetate or ethyl acetate.

Of the two solvents, methyl acetate and methyl ethyl ketone,
methyl ethyl ketone has the advantage that in California it is more
easily obtainable in a purer form than methyl acetate. Also, methyl
ethyl ketone in a pure form now costs only about one-fourth as much
as methyl acetate of 95 per cent purity.

Reproducibility.— To test the reproducibility of the methyl ethyl
ketone method of developer analysis, six analyses each were made
of a production negative developer, a production positive developer,
a high Elon-high hydroquinone type developer, a low Elon-low
hydroquinone type developer, and the SD-21 developer. The re-
sults of these analyses are shown in Tables 4 through 8.

TABLE 4

Production Negative Developer

Analysis
Number

Elon (grains per liter)
Hydroquinone (grams per
liter)

TABLE 5

Production Positive Developer

Maximum
Deviation
from Mean

(Per Cent)

1.63

.63

.69

3.6

2.12

.98

.06

.10

2.4

Analysis
Number

Elon (grams per liter)
Hydroquinone (grams per
liter)

Maximum
Deviation
from Mean
(Per Cent)

2.22

2.34

2.24

2.35

2.39

4.3

7.31

7.71

7.55

7.33

7.71

7.61

3.2

TABLE 6

High Eton-High Hydroquinone Type Developer

Analysis
Number

Elon (grams per liter)
Hydroquinone (grams per
liter)

Maximum
Deviation
from Mean
(Per Cent)

3.22

3.31

3.13

3.31

3.4

£.46

8.96

8.80

9.04

8.91

9.14

4.5

Nov. 1946 APPLICATION OF METHYL ETHYL RRTONR

415

TABLE 7
Low Elon-Low Hydroquinone Type Developer

Analysis
Number

Elon (grams per liter) 0 . 20

Hydroquinone (grams per 0.33
liter)

Maximum
Deviation
from Mean
(Per Cent)

0.21

0.20

0.21

0.20

2.4

0.31

0.32

0.33

4.6

TABLE 8

SD-21 Developer

Maximum
Deviation
from Mean

(Per

Cent)

1.93

.87

.93

1.87

4.81

.85

4.69

Analysis
Number

Elon (grams per liter)
Hydroquinone (grams per
liter)

From Table 4 it is evident that the maximum deviation from the
mean in six analyses of a production negative developer is 3.6 per
cent for Elon and 2.4 per cent for hydroquinone. According to

TABLE 9

Elon

Sodium Sulfite

Hydroquinone

Boric Acid

Borax

Sodium Carbonate

Citric Acid

Potassium Metabi-

sulfite

Potassium Bromide
Water to make

Developer 1
in grams

Developer 2
in grams

Developer 3
in grams

Developer 4
in grams

2.0

3.0

0.22

0.31

100.0

40.0

64.9

39.6

5.0

10.0

0.28

6.0

8.0

. . .

8.0 '

2.0

20.0

18.7

0.68

...

1.5

0.25

0.86

1 liter

Table 5, the maximum deviation from the mean in six analyses of a
production positive developer is 4.3 per cent for Elon and 3.2 per cent
for hydroquinone. In Table 6, it may be seen that the maximum de-
viation from the mean in six analyses of a high Elon-high hydro-
quinone type of developer is 3.4 per cent for Elon and 4.5 per cent for
hydroquinone. According to Table 7, in six analyses of a low Elon-
low hydroquinone type developer, the maximum deviation from the
mean is 2.4 per cent for Elon and 4.6 per cent for hydroquinone. In

416

V. C. SHANER AND M. R. SPARKS

Vol 47, No. 5

Table 8 it may be seen that in six analyses of SD-21 developer the
maximum deviation from the mean is 2.0 per cent for Elon and 2.2
per cent for hydroquinone. These reproducibility values indicate
that the methyl ethyl ketone method of developer analysis is quite
satisfactory for production control of developers.

Accuracy. — To test the accuracy of a method of developer analy-
sis using methyl ethyl ketone, analyses according to the procedure
just outlined were made of each of the four developers, mixed ac-
cording to the formulas given in Table 9.

By inspection of Table 10, it may be seen that the methyl ethyl
ketone method of developer analysis has an accuracy quite adequate
for production control analyses of fresh developers.

TABLE 10

Developer
Number

Elon

Hydroquinone

Elon

Hydroquinone

Elon

Hydroquinone

Elon

Hydroquinone

Grams per
Liter
Mixed

2.0
5.0
3.0
10.0
0.22
0.28
0.31
6.0

Grams per
Liter
Found

1.9

4.9

2.9

9.6

0.21

0.29

0.30

5.90

Per Cent
Error

4.0

1.8
4.2
4.0
4.5
3.5
3.2
1.6

However, since these developers were all freshly mixed it was neces-
sary to determine what effect the monosulfonates of Elon and hydro-
quinone might have on the accuracy of this method. Therefore, one
gram per liter of Elon monosulfonic acid and one gram per liter of
sodium hydroquinone monosulfonate, respectively, were added to
developer 1 of Table 9. The resulting developers were analyzed and
the data thus collected are shown in Table 1 1 .

Elon
Hydroquinone

TABLE 11

Developer 1 plus one gram per
Developer 1 liter Elon Monosulfonic Acid

1.92 1.93

4.91 4.87

Developer / plus

one gram per liter

Hydroquinone

Monosulfonate

1.94
4.94

Reference to this table will show that the presence of monosul-
fonates in the developer apparently has little effect on the result of
the analysis. Hence, it is concluded that the methyl ethyl ketone

Nov. ItUti APPLICATION OF METHYL ETHYL KETONE 417

method of developer analysis has satisfactory accuracy for production
control.

Conclusions. — (1) It was concluded that methyl ethyl ketone is
more satisfactory than methyl acetate or ethyl acetate as a sol-
vent in the analysis of photographic developers for Elon and hy-
droquinone in the following respects :

(a) Methyl ethyl ketone when used with the f/-tube extractor showed a better
rate of extraction for Elon and hydroquinone than methyl acetate or ethyl ace-
tate.

(6) Hydroquinone is more soluble in methyl ethyl ketone than in equal vol-
umes of methyl acetate or ethyl acetate.

(d) Methyl ethyl ketone is only one-fourth as expensive as methyl acetate.

(2) Analyses made using methyl ethyl ketone as the extracting
solvent showed this method to have the necessary accuracy and re-
producibility for production control use.

REFERENCES

1 EVANS, R. M., AND HANSON, W. T., JR.: "Chemical Analysis of an M. Q.
Developer," /. Soc. Mot. Pict. Eng., XXXII, 3 (Mar. 1939), p. 307.

2 BAUMBACH, H. L.: "The Chemical Analysis of Hydroquinone, Metal and
Bromide in a Photographic Developer," /. Soc. Mot. Pict. Eng., XXXIII, 5 (Xov.
1939), p. 517.

3 ATKINSON, R. B., AND SHANER, V. C.: "Chemical Analysis of Photographic
Developers and Fixing Baths," /. Soc. Mot. Pict. Eng., XXXIV, 5 (May 1940), p.
485.

4 STOTT, J. G.: "The Application of Potentiometric Methods to Developer
Analysis," /. Soc. Mot. Pict. Eng., XXXIX, 1 (July 1942), p. 37.

6 EVANS, R. M., HANSON, W. T., JR., AND GLASOE, P. K.: "Synthetic Aged
Developers by Analysis," /. Soc. Mot. Pict. Eng., XXXVIII, 2 (Feb. 1942), p. 188.

NAVAL TRAINING-TYPE EPIDIASCOPE FOR UNIVERSAL
PROJECTION OF SOLID OBJECTS*

JACQUES BOLSEY**

Summary. — The Special Devices Division of the Navy's Office of Research and
Inventions conceived the idea of projecting the image of a solid model on a screen. For
this particular trainer it was desired to project an airplane model on a spherical
screen. Moreover, the image itself had to be movable so as to be positioned at any
point on the spherical screen. Among the many problems which had to be overcome
were depth of field, as in microphotography, the level of illumination, as in all epidia-
scopes, and finally the wide range of magnification.

Many optical combinations were computed and tested, as well as a variety of light
sources and condensing designs. The most suitable combination was adopted and
built into the first model.

In its development of ever more realistic and flexible training de-
vices, the Special Devices Division of the Navy's Office of Research
and Invention commissioned us to build a new type of projector to
project the image of a solid model airplane. The attitude of the
airplane should be variable at will, allowing full freedom in pitch,
roll, and turn, so that the airplane might appear to climb, dive, and
roll. The image should be continuously variable in size in order to
simulate continuous changes in range.

The Navy's requirement was to set up such a projector at the center
of a spherical screen (similar to a planetarium screen, 26 ft in diam-
eter). Means were also required to project the image anywhere on
the spherical screen, and all of these variables were to be remote con-
trolled.

The prescribed range of magnification was to be such that the air-
plane could appear at any distance from 300 to 4000 ft. This called
for a variable focus optical system with a range of more than one to
thirteen.

* Presented May 8, 1946, at the Technical Conference in New York.
** Bol, Ltd., New York.

418

NAVAL TRAINING-TYPE EPIDIASCOPE

419

It was obvious from the start that the greatest difficulty would be
to get a sufficiently high light output, because, since the airplane was
to be seen in any attitude (that is, head-on, from above, from below,
etc.) a transparency or film strip was out of the question. The Navy
wished to project a solid model. Moreover, a dark silhouette against

FIG. 1.

Complete projector (front view, with-
out covers).

a light background would not have been suitable ; a light image was
required in order to stand out on the screen when background effects
such as clouds or sea were thrown on the screen by auxiliary projectors.
The size of the model airplane was not specified but a maximum of
one-inch wing span was chosen in order to avoid excessive length of
the projection optics.

The level of illumination on the model would have to be extremely

420

J. BOLSEY

Vol 47, No. 5

high, since requirements showed that the relative aperture of the sys-
tem could not be higher than about //20.

Illumination. — A large number of known light sources were con-
sidered. The high-pressure mercury vapor tube is of course excel-
lent in so far as total output is concerned, but the very elongated

FIG. 2.

Complete projector (rear view, with-
out covers).

shape of this source was not suitable because the light had to be
concentrated into a small circular area. Moreover, it required a
fair bulk of auxiliary equipment which we wished to avoid.

At the other end of the scale, in lamps then available, the photo-
micrographic lamp is excellent for size and shape of the source, but
its total output is such that dozens would have been required.

Nov. 1946

NAVAL TRAINING-TYPE EPIDIASCOPE

421

Several attempts were made to work out a means of bringing the
model itself to incandescence. A thorium oxide model was made to
glow in the flame of a welding torch. It was found that a tungsten
model could be held at incandescence by a high-frequency coil. These
methods were ruled out because of the element of danger in the open
flame and of the high disintegration rate of the model and the diffi-
culty of mounting the model under these conditions.

FIG. 3. Model suspension mechanism.

We investigated also a vacuum tube developed by the Flight
Training Research Association in which a tungsten foil shaped like an
airplane was brought to incandescence by electronic bombardment.
The foil airplane lacked realism and the development on this tube
had not at that time progressed to the use of a solid model.

Eventually, the light source that was adopted was the tungsten
projection lamp. Its filament area is relatively small, its efficiency
good, and it is reliable and available. It was found that the neces-

422

J. BOLSEY

Vol 47, No. 5

sary illumination could be obtained by using a suitable number of
these lamps in high-efficiency condenser systems.

Arranged in a conical pattern around and in front of the model are
six standard projection bulbs of 1000 w each. They all illuminate the
same one-inch area and the concentration of light — and of heat — is
extremely high. The projection lamps are, of course, designed to
burn in a vertical or near vertical position. Our tests have shown,
however, that if the lamps are well cooled, their life is still good when
burned in a changing position as in this projector.

FIG. 4. Experimental optical system.

Model Suspension. — The problem of mounting the model so
that its attitude could be varied at will led to the construction of
several experimental setups.

In one the airplane was supported by a tiny ball joint on the end
of a thin rod; other rods were similarly attached to the tail and one
wing. The airplane could thus be controlled from below, much as a
marionette is controlled from above.

Another system considered consisted in mounting the model in the
center of a transparent sphere. This had the advantage that no
supporting members were visible at any time. However, the control
for pitch, roll, and turn of the model would have been so complicated
that it was abandoned. Also the heat radiated by the lamps and
concentrated in the center of the sphere required a complicated and
bulky heat evacuation system.

Nov. itMC) NAVAL TRAINING-TYPE EPIDIASCOPE 423

All in all, a gimbal mount was found simplest and most efficient.
Inside the projector the model, which is easily interchangeable, is sup-
ported on a slender shaft in a gimbal mount. Differential gear trains
are eliminated by placing the two smaller Selsyn motors on the gim-
bals themselves. This mechanism being relatively light, no servo
system is required, and the three Selsyns control pitch, roll, and turn.

Optical System. — In developing the optical system the prime
considerations were light transmission and over-all length. The
magnification had to be varied in the ratio of one to thirteen, which
called for a new type of very high ratio lens. The image on the
screen is 40 in. in wing span at the highest magnification, and 3 in. at
the lowest; the equivalent focal lengths are consequently 3.7 in. and
29.25 in., yet this maximum equivalent focal length of almost 30 in.
must be obtained in a system as short as possible. The design which
was incorporated in the projector has a mount 27.5 in. long and com-
prises only two optical groups. It will operate at apertures up to
//9, but as stated previously it is diaphragmed to //20 at maximum
magnification in order to reproduce the full depth of the model with
sufficient sharpness.

The lens groups are assembled in telescoping tubes in the pointed
end of the egg-shaped housing. The simplest means to co-ordinate
the movement of these two lens groups is a differential cam which
varies the space between them as the rear group is driven forward or
back. This cam is mounted above the lens tube and a bell crank
transmits the change in spacing by means of a gear sector and rack.
The assembly is finally spring-loaded to avoid all backlash.

As with the optical system, the condenser units were thoroughly
studied in order to reach an efficiency as high as possible for this
particular device. Each unit consists of only two elements aside
from the spherical reflector; it accepts a 65 deg cone and condenses
the light to an area of one-inch in diameter. All lenses and conden-
sers are coated for maximum light transmission.

Zenith and Azimuth Control. — In order to move the image in
zenith and azimuth on the screen, it was decided to swing the whole
projector. This avoided the loss of light which would occur in a
stationary projector equipped with mirror or prism systems for
zenith and azimuth movements of the image. The projector proper,
which is somewhat egg-shaped and measures 45 in. in length and 24 in.
in diameter, is mounted in a yoke so that it may be pointed 20 deg
below the horizon, raised to a vertical position, and swung over on its

424 J. BOLSEY Vol 47, No. 5

back, down again to 20 deg below the horizon. This gives the zenith
of the image.

The yoke rotates around a vertical axis for azimuth. Both move-
ments are worm gear driven. Since the image must be positioned
at any point on the screen by remote control, a servo system is used
to rotate the projector around these axes. An all-aluminum con-
struction was adopted to lighten the structure and facilitate these
movements.

FIG. 5. Servo-mechanism for azimuth and elevation.

The servo-system consists of a variable speed hydraulic unit of the
type used in standard aircraft turrets, the booster valves of which are
controlled by Selsyns.

The response is practically instantaneous and the maximum speed
of rotation of the projector is 26 rpm. If the airplane represented is at
300 ft, a crossing speed of 550 mph can be represented. The zenith
drive is equipped with a limit mechanism which acts directly on the
booster valve.

Range Control. — To meet the Navy's specifications, the range
control must be such that speeds up to 50,000 ft per min may be
simulated with low response delay and high accuracy. The

Nov. 1946 NAVAL TRAINING-TYPE EPIDIASCOPE 425

change in magnification was obtained by the vario-lens system
previously described. A hydraulic system was adopted to drive the
vario-lens because it delivers a great range of speeds with immediate
response. It is mounted on the bulk head behind the model support
assembly and is powered by constant speed electric motor of l/3 hp.
The lens assembly can be driven its full 17 in. of travel in 6 sec in
either direction; the closing speed represented can also be brought
down to zero and reversed immediately without any lag. The
booster valve on the hydraulic unit is controlled by a Selsyn motor;
when the lens assembly reaches the end of its travel this Selsyn is over-
powered by a limit mechanism to prevent damage to the apparatus.

These hydraulic units are normally equipped with a breather which
allows the oil to expand as it warms up in the course of running, and
any bubbles to escape. In this application the breather is replaced
by expansion chambers so that no oil spillage will occur and no bub-
bles will be trapped as the unit rotates to varying positions with the
projector.

Power. — The entire machine is operated on 110 v, a-c single
phase, and requires about 7 kw power.

Remote Controls. — Besides the two motors which power the hy-
draulic servo-systems, in all, six Selsyns are built into the machine;
one for zenith and one for azimuth movements, one for the range
control drive of the vario-lens, and one each for pitch, roll, and turn
of the model. Each of the Selsyn motors is connected to its Selsyn
generator on the control panel. This panel controls the movement
of the airplane.

Acknowledgment. — We wish to express our appreciation to the
Special Devices Division of the U. S. Navy Office of Research and
Inventions for their assistance, encouragement, and advice which
helped us greatly in the development of this device.

A NEW METHOD OF COUNTERACTING NOISE IN SOUND
FILM REPRODUCTION*

W. K. WESTMIJZE**

Summary.— Reproduction of the sound recorded on sound film is usually accom-
plished by means of a narrow beam of light thrown upon the film in a direction per-
pendicular to that in which the sound track is moving. The fluctuations in the light
flux passed through are converted into sound. With this method a noise results
which is caused by the fact that part of the light passed through is intercepted by
specks of dust, scratches, etc., on the sound track, especially when the film has already
been used several times. This article describes a method of counteracting this noise in
cases where the sound is recorded as so-called amplitude writing. The beam of light
is replaced by a series of equidistant light spots moving with great velocity perpendicu-
lar to the sound track. In addition to the theoretical fundamentals of the method,
a practical form of application is also discussed.

The Ordinary Method of Reproduction. — The reproduction of
sound film is usually reproduced in the following manner. A nar-
row beam of light is thrown on the film perpendicular to its direc-
tion of motion. Confining ourselves to the case where the sound is
recorded as so-called amplitude-writing, such as, for example, with
the Philips-Miller film,1 the quantity of light passing through the
film depends upon the width of the sound track (and of course of the
beam). The light passes through to a photocell and is converted
into an electric current which may be considered as a direct current
upon which an alternating current is superposed. The magnitude
of this direct current depends upon the width of the so-called zero
track, i. e., the track which is made when no sound vibrations are
being recorded. The zero track is unavoidable, since otherwise
modulation would be impossible. It is easy to understand that its
width must be equal at least to once or twice the maximum modula-
tion amplitude, according as the modulation takes place on one
side or on both sides of the track.

|!* Reprinted from Philips Technical Review, 8, 4 (Apr. 1946), p. 97.

** Research Laboratory, N. V. Philips' Gloeilampenfabrieken, Eindhoven,
Holland.
426

COUNTERACTING NOISE IN FILM REPRODUCTION

427

The alternating current depends upon the modulation of the track
and thus on the sound vibrations recorded, and if the light beam were
infmitesimally narrow the trend of this current would be an exact
copy of the sound vibrations. Actually the beam has a finite width
A, but even so the relation between the sound vibrations recorded
and the corresponding vibrations of the light flux can easily be
determined. Let us assume that the sound track is modulated by one
harmonic vibration. Such a vibration is represented in Fig. 1.

163 93

FIG. 1. Diagram of the usual method of scanning.
The film with the modulated sound track travels past this
beam S. The variations in the light flux passed through
are registered by a photocell. In the diagram the track is
modulated on both sides by a purely sinusoidal vibra-
tion ; d = width of the unmodulated track, q = amplitude
of the vibration with which the track is modulated, y =
depth of modulation at the point with the abscis x, A =
width of the slit.

When this vibration corresponds to a tone of v oscillations per sec,
and when the velocity at which the film is traveling is v cm per sec,
there are v/v vibrations per cm of film and the vibration can be repre-
sented by the equation y = q cos 2ir v/v.x where y is the depth of
modulation and x the length of film passed, measured from an arbi-
trary zero point. The amount of light passed through is then pro-
portional to

f i

v
sin IT- A

- X

d representing the width of the zero track. From the result it is
immediately clear that a d-c and an a-c component are present, while
it is also clear that the amplitude of the a-c component is multiplied

428 W. K. WESTMIJZE Vol 47, No. 5

by a factor which depends upon the frequency v. This factor

v A
sin TT - A

is equal to unity when v = 0, and then decreases. In order that the
highest frequencies to be reproduced should not be attenuated by
more than about 3.5 db compared with the lowest (such an attenua-
tion is still permissible) it is necessary that

7r.^.A< 1.5.

witn "max = 8000 and v = 32 cm per sec this results in A < 0.002
cm. . The light-beam may therefore not be wider2 than 20 ju-

When there are specks of dust or dirt on the sound track or when it
has been scratched, as is particularly the case with much used sound
films, these tiny specks and scratches, irregularly distributed over
the surface of the film, cause a noise. They cannot, however, be ob-
served individually, as is the case with larger particles ( < 80 M), which
cause an annoying ticking or bubbling sound. It would mean a con-
siderable improvement in reproduction if this noise could be avoided.

For some time already a system has been in use which diminishes
this noise. It is based on the following principle. The noise is most
annoying during the soft passages, i. e., when the depth of modula-
tion is slight. In sound recording it is now arranged, by means of
suitable connections, that during these passages the zero track be-
comes narrower, thus reducing the area upon which the troublesome
specks or scratches may occur and thereby also the noise. During
the louder passages the zero track again becomes wider, and thus
also the noise becomes louder, but this is less troublesome here be-
cause for the greater part it is drowned out by the music or speech.

This method, therefore, does not eliminate the noise, but only re-
duces it during the soft passages.

Principle of High-Frequency Scanning. — We have seen that in the
scanning method described above the noise is caused by contamina-
tions on the transparent part of the film between the two edges of
the sound track. This phenomenon therefore also occurs when the
edges of the track, which actually represent the sound, are ideal.
With the method of high-frequency scanning, about to be discussed,
only the edges of the track are scanned ; the influence of the part

Nov. 1940 COUNTERACTING NOISE IN FILM REPRODUCTION

429

between the edges is eliminated and thus also the noise, so far as
it is caused by specks on the transparent part of the film. Of course
the noise resulting from imperfections in the edges of the sound
track, to which we shall return later, still remains, just as with the
method of zero track adaptation discussed above.

With this method of scanning, instead of a narrow slit of light, we
have a series of light spots moving at a very high velocity and at
regular intervals perpendicularly across the film. Since the sound
track is also moving, the light spots actually move in an oblique

•

BCD

FIG. 2. Diagram of the high-frequency method of scan-
ning. A series of equidistant spots of light travel at a
high velocity across the film. Owing to the fact that the
film is also traveling at the same time, the light spots
describe paths which are oblique with respect to the
film and which are given in Fig. a. The slope of these
paths is very much exaggerated for the sake of clearness.
As long as a light spot is inside the edges of the track, a cur-
rent flows in the photocell. The form of the signal leaving
the photocell is shown in Fig. b. The block AB corre-
sponds to the path ab, etc.

direction across the film. Here, too, the light passing through falls
on a photoelectric cell, which gives a current impulse during the
time that the light spot is moving between the edges of the track.
The image of -this impulse is approximately rectangular. The in-
tensity of the impulse is determined by the intensity of the beam of
light employed. The duration of the impulse depends upon the
width of the track at the point where the light spot crosses it. Thus
in Fig. 2 AB in the lower half corresponds to ab in the upper half,

430

W. K. WESTMIJZE

Vol 47, No. 5

the same being true of CD and cd, etc. It is essential to note that the
beginning and end points of the blocks are fixed by the edges of the
sound track. (For the sake of clearness the obliqueness of the paths
of the light spot across the film is exaggerated.) Contaminations on
the film are manifested by variations in the beam of light passed
through and consequently the image of the current impulses is not
actually as shown in Fig. 2b, but as in Fig. 3 ; between A and D the
current is not constant, variations occurring of an accidental nature.
The great advantage achieved lies, however, in the fact that the dis-
turbances are separated from the phenomenon to be reproduced, the
former affecting the height of the blocks, while the latter only affects
the beginning and end points of the blocks. Therefore the disturb-

46395

FIG. 3. Image of a current impulse from the photocell.
The intensity variations are due to contaminations on
the film in the path of the scanning light spot. The es-
sence of the method lies in the fact that the influence of
the contaminations can be eliminated by passing the
signal through a limiter. Limitation to the level EF
would in this case be sufficient.

ances can easily be eliminated by sending the whole signal through a
limiter which only passes signals up to a certain amplitude. In this
way the disturbances are, as it were, cut off. For the- current varia-
tion shown in Fig. 3 a limitation to the level EF would be sufficient
to bring about this elimination. If the signal is afterwards so ampli-
fied that the amplitude is increased in the ratio BA/EA, a signal is
obtained which is absolutely identical with what would have been
obtained if the sound track had been everywhere uniformly trans-
parent.

We must now consider the question as to how we can derive the

Nov. 10 a; COUNTERACTING NOISE IN FILM REPRODUCTION 431

original sound frequencies from the block-signal. The frequency
spectrum of this signal must first be investigated. This involves
complicated calculations which will be further dealt with on another
occasion. Suffice it here to go into a few qualitative considerations.
Let us first examine the unmodulated signal. It consists of con-
gruent blocks having a frequency M (the scanning frequency). If a
Fourier analysis is made of this signal, vibrations with the frequen-
cies //, 2ju, 3/x, etc., are obtained. If we now modulate- the block signal
with a frequency v, secondary frequencies then appear in the spec-
trum: M =*= v\ /z ± 2*>; . . . 2/jL =*= v\ 2fjL ± 2i>; . . . 3/z ± v\ etc. It is,
however, quite obvious that also the frequency v itself will occur. Let
us again consider Fig. 2. The blocks corresponding to the wide
parts of the track are wide and consequently the wide parts give rise
to relatively long current impulses with short interruptions. In the
case of the narrow parts of the track it is just the reverse. If we now
pass this signal through a suitable filter, i. e.t a low-pass filter, with
limiting frequency coinciding with the highest frequency that has
. to be passed through, the result is that the signal, roughly speaking,
is replaced by a progressive average over a certain time interval ap-
proximately of the order of one-quarter of the time of vibration of the
limiting frequency. Thus in each case a number of successive blocks
is averaged and the result is a signal which is strong when tfye blocks
are wide and weak when they are narrow, thus an alternating current
with a frequency v corresponding to the frequency of the vibration
originally registered.

For reproduction it is essential that the frequency v should occur
but that 2v, 3i>, etc., should be absent. That this is indeed the case is
proved by calculation, though it is not easy to imagine. It is ob-
vious, however, that this is of importance, for, as a rule, with v also 2i>
and possibly 3v, etc., lie in the audible range.

We can now also make it clear that the scanning frequency ju must
be much greater than the highest frequency v to be reproduced, be-
cause in addition to M owing to the modulation also the tones n — v,
ju — 2v, etc., occur. These tones become weaker as we get farther
away from the frequency //.

Calculation shows that the frequency p. — 5v is already 60 db
weaker than the frequency v. The frequency ju — 4v would still be
strong enough to be disturbing. If we are to eliminate this by
means of a filter, then it must fall outside the audible range, and this
means that:

432 W. K. WESTMIJZE Vol 47, No. 5

M — 4j/max > Vmax, Or jU > 5j>max

where *>max represents the highest frequency of the audible region
which is to be reproduced. Taking 7max = 8000 c per sec, it follows
that fi = 40,000 c per sec.

Limitations of the Effect of the Method. — It must be pointed out
that not all disturbances can be eliminated in the manner described.
Two cases must be examined separately.

In the first place a contamination may be so large and conse-
quently intercept so much light as to cause the photocurrent to fall
below the value that passes through the limiter. The result is a
"dent" in the corresponding block which again causes a disturbance.

This is especially the case when the light is entirely cut off by the
contamination, in which event one light spot produces two current
impulses (blocks) . However, by giving the light spot an oblong shape
it is possible to ensure that this case seldom occurs. Already in the
beginning of this article it was observed that the width of the light
beam in the ordinary method of scanning may not be more than 20 n,
because otherwise the high tones would be weakened. This applies
also for the width of the light spots, but not for their height. (By
width we mean here the dimension perpendicular to the motion of the
light spots and by height the dimension parallel to that motion.)
An increase in the height, for instance to 100 M> has by first approxi-
mation the same effect on the fluctuations of the transmitted light
flux as if the zero track had been taken 100 — 20 = 80 /x wider and
the height left the same. This can easily be explained: Owing to
the finite height of the light spots the photocurrent impulses do not
have the form of rectangles (apart from the disturbances due to con-
taminations), but of equilateral trapezia.

During the time that the light spot is moving over the edge of the
track, the intensity increases from zero to the maximum value and
decreases again from the maximum to zero. In Fig. 4 two cases are
depicted for different heights of the light spots. It is assumed that
they begin to pass over the track at the same moment. The photo-
current impulses then begin at the same moment for both, at the
point A . We further assume, of course, that the two light spots move
at the same velocity, so that the light intensity increases in the same
way and the trend of the photocurrent will be the same in both cases,
for instance along AB. A difference occurs only when the lowest
light spot is completely over the track, let us say at B. From that
moment the corresponding photocurrent (except for disturbances)

Nov. 1946 COUNTERACTING NOISE IN FILM REPRODUCTION 433

remains constant. For a short while, however, the current corre-
sponding to the highest spot continues to increase at the same rate,
until this spot is also entirely over the track, let us say at Bf, from
which moment the second current, too, is (practically) constant.

As soon as the upper edge of one of the spots has reached the other
side of the track, the corresponding current begins to decrease again.

A DO

46116

FIG. 4. Influence of the height of the light spots on the
form of the photocurrent impulses excited by the light spot.
Owing to the fact that the spot has,a finite height, some time
elapses before the whole spot is over the track. During that
time the current increases continually. The impulse A BCD
is due to a low spot, the impulse AB'C'D' to a higher one,
the top side of both spots having reached the edge of the
track at the same moment. In the second case the average
current of the photosignal is larger. Limitation of the sig-
nal to the level EF is therefore sufficient to eliminate all dis-
turbances in the second case but not in the first case.

Under our assumptions this will take place at the same moment for
both currents and the points at which this takes place, namely C and
Cf, lie vertically above each other. The decrease is at the same rate
as the increase and thus equal for both spots (the current curves are
equilateral trapzeia). The currents thus decrease according to two
parallel straight lines, CD //CD'. Therefore they do not end at the
same moment. The difference DD', however, is entirely determined

434 W. K. WESTMIJZE Vol 47, No. 5

by the difference in intensity CC' (and the velocity of the spots,
which is however, the same for both), and this in turn depends ex-
clusively on the difference in height of the spots.

If we pass the two signals through the same limiter then from our
reasoning it follows that the signals finally obtained differ only in
length, but that this difference is the same for all blocks and therefore
has no effect on the sound to be ultimately reproduced. It only alters
the d-c component of the photocurrent signal, just as a change in the
width of the zero track would do, and this is suppressed by a filter.
If the height of the spot is greater than the width of the track, the
situation is somewhat different, but a closer investigation shows
that in this case too the length of the blocks of the limited signal,
except for a constant, is proportional to the width of the track at the
place where the light spot passed.

From the foregoing it will be clear that it is possible to choose such
a height of the spot that, practically speaking, the transmitted light
cannot be cut off by contaminations to such a degree that after limita-
tion such disturbances still have any effect. This is in fact demon-
strated in Fig. 4. The absolute changes in intensity of the transmit-
ted light beams resulting from contaminations are the same for both
spots. Therefore the noise assumed to be present in this case is
without influence on the limited signal with the higher spot, but with
the lower spot it does leave a disturbance in the limited signal.

The second possibility of disturbances occurs when a contamina-
tion lies exactly tangent to or across the edge of the track. This
alters the form of the limitation. The disturbance caused by such
an imperfection in the edge of the track is not eliminated by the
method discussed here. The chance of such a disturbance occurring,
however, is slight compared with that caused by a speck elsewhere
on the track. The modulated track is at least 1 mm wide, so that the
chance of contaminations, even of the size of 100 ju> coming to lie at
the edge is only 20 per cent; most contaminations, however, are
much smaller and there is therefore still less chance of their lying at
the edge of the track.

One Possible Construction of the Apparatus. — The above-de-
scribed high-frequency scanning can be realized in different ways.
In the first place, the sound track can be scanned by a moving light
spot, as has been assumed in the foregoing. In principle the same
results can be attained by projecting the image of the sound track
and causing this image to vibrate with respect to a diaphram. In

Nov. 1946 COUNTERACTING NOISE IN FILM REPRODUCTION

435

both cases we may consider the vibration as being brought about
with a moving light source and a stationary optical system, but also
with a stationary light source and a moving optical system. Finally
the vibrations may be construed as being brought about by elec-
trical means as well as by mechanical means. We shall here con-
fine ourselves to the description of a method worked out by us in
which the scanning is accomplished with a moving light spot ob-
tained from a mechanically moved optical system.

In Fig. 5 a diagram is given of the arrangement employed. The
light from a linear source is projected by a lens several mm from the
edge of a disk which can be rapidly rotated. In this disk radial slits
have been sawed beginning at the edge. When the disk is rotating
rapidly, therefore, each slit allows a fraction of the light from the

FIG. 5. Diagram of the setup for high-frequency scanning. G = source
of light with linear filament. L = lenses. 6" = rotating disk with slits. F
= film with modulated sound track. V = light spot. C — photocell.

image to pass through. The image of the illuminated opening is
focused on the sound film by means of a second lens. The light
passed through the sound track falls on a photocell and gives rise to
the photocurrents already mentioned.

The practical realization of such a setup involves a number of
technical difficulties which we shall now discuss.

The Choice of Light Source. — We have already remarked that the
width of the light spot on the film may not amount to more than
20 M. Furthermore, it must be very sharp (the transition from
light to dark must take place within a distance of not more than a
few M) and not only when the projection is along the axis of the sys-
tem, but also when the image is about 1 mm above or below it.
Finally the light must be of sufficient intensity to excite a reasonably
amplifiable photocurrent. These conditions make certain demands
on the optical system and the source of light.

436 W. K. WESTMIJZE Vol 47, No. 5

Linear light sources whose incandescent body is narrower than 80 ju
are difficult to produce. This implies that the optical system must
be a reducing one. The same conclusion is reached from the re-
quirement of sharpness of projection. A fivefold reduction suffices
for both requirements. This reduction is mainly effected by the
second lens. The first lens gives practically an image of 1:1. The
requirements for sharpness of the image make it necessary to work
with small opening angles.

Finally from the minimum required light intensity of the beam
that falls upon the photocell and from the dimensions of the optical
system it is to be deduced that the brightness of the light source em-
ployed must be at least 1000 candle power per cm.2 In order to
satisfy these requirements a special lamp was constructed.

Construction of the Rotating Disk. — The greatest difficulty lay in
the construction of the disk. As already mentioned, the required
frequency of the light spots is 40,000. The width of the track for
Philips-Miller film can be set at a maximum of 1.6 mm, hence a veloc-
ity of the light spots of 6400 cm per sec. Since, as mentioned above,
the second lens reduces by a factor 5, this leads to a peripheral veloc-
ity of the disk of 32,000 cm per sec. Now the peripheral velocity
determines the stresses occurring in the disk. Similar disks of differ-
ent diameters but with equal peripheral velocities exhibit exactly
the same stresses at corresponding points. At a velocity of 32,000
cm per sec, these stresses are enormous and approach the yield point.
It is clear that this sets an upper limit for the velocity. In fact, if this
limit is reached the disk flies to pieces.

Since for different materials under otherwise similar conditions
the stresses are proportional to the specific weights, a material had
to be found with the most favourable ratio of yield point to specific
weight. Moreover, having regard to the motive power for the disk,
the material had to be electrically conductive, so that practically only
duralumin and electron could be considered. Furthermore,
since the highest stresses occur where the hole is drilled for the spindle,
the disk was given a very slightly conical profile.

It can then be calculated that both for dualumin and for elec-
tron the maximum stresses occurring, even at a velocity of 40,000
cm per sec, still remain below half the yield point value. This was
in fact confirmed experimentally by investigating at what peripheral
velocity a test disk flew to pieces. This was found to be at 60,000
cm per sec (the stresses are proportional to the square of the velocity) .

Nov. 1946 COUNTERACTING NOISE IN FILM REPRODUCTION 437

Furthermore, as it was desirable not to make the apparatus too
cumbersome, the disk could not be made too large. Its radius was
therefore fixed at 5 cm. This means that a speed of rotation of
32,000/107r = 1000 rev per sec is required. Since the slits have to be
about 5 X 1.6 = 8 mm apart, 107T/0.8 = approximately 40 slits can be
made on such a disk. They are 0.6 mm. wide and 3.5 mm long (from
this it follows that the length of the light spots on the film is 120 M).
The cutting of the slits requires much care.

In the first place they have to be spaced at exactly equal distances
and must be exactly alike, as otherwise the frequency of revolution
of the disk appears in the frequency spectrum, and since this lies in
the audible region there will be a whistling tone in the sound re-
produced. The scanning frequency, which is 40 times as high, lies,
as we know, outside this region.

In the second place very careful finishing is essential because other-
wise at the high speeds of rotation the disk might crack at the slits.
For that reason before the slits are cut small holes are drilled at the
spots where the slits end.

Bearings and Motive Power of the Disk. — With the above-mentioned
very high number of revolutions of 1000 per sec special demands are
of course made of the bearings. Even a slight eccentricity of the
center of gravity of the disk with respect to the center of the bearings
gives rise to enormous centrifugal forces as the speed increases, re-
sulting in high pressures on the bearings, vibration of the motor,
high friction and heavy wear. In order to avoid this the principle
of the de Laval shaft was employed, with a thin flexible spindle
instead of the usual rigid shaft. Owing to the centrifugal force
the spindle will sag already at a low number of revolutions, and
this sag becomes greater as the speed of rotation increases.
When a certain speed is reached, the so-called critical speed, the sag
will theoretically even be infinite. Above that speed the sag de-
creases rapidly and at the limit for infinitely high speed the disk
will rotate about its center of gravity. When this state is reached
the sage of the spindle and consequently the pressure on the bearings
is very small. The bearing pressure is then mainly determined by the
disk's own weight.

A difficulty in working with a de Laval spindle lies in the passing
of the region of the critical frequency when starting up. It is pos-
sible to do so without breaking the spindle if that region is passed so
quickly as to leave no time for the disk to assume large deflections.

438 W. K. WESTMIJZE Vol 47, No. 5

In our case, however, the driving couple was not large enough for this
and we therefore decided to suppress the dangerously large deviations
by applying a suitable damping arrangement to the spindle. For that
purpose the spindle is passed through eyelets at a short distance from
the disk on either side. These eyelets are connected by rods to small
pistons moving up and down with a little play in small cylinders
containing oil. By this means the lateral movements of the disk are
damped, and by choosing suitable dimensions for this device the
vibrations in the critical region can be kept sufficiently low. Once the
critical region is passed, the disk runs very quietly and speeds of 1000
and 2000 rev per sec are easily attainable.

As already remarked in passing, the disk is driven electrically.
It is placed in the field of two mutually perpendicular magnetic cir-
cuits activated by alternating currents with a frequency of 1500 c per
sec and shifted 90 deg in phase with respect to each other. Each
circuit consists of two pole shoes, between which air gaps of about l/%
cm have been cut. The disk is placed in these air gaps. The com-
bination of the two alternating magnetic fields produces a rotating
field which turns the disk — made of a conducting material especially
for this purpose — and is able to give it sufficient velocity. In order
to minimize friction the disk with the complete driving mechanism is
placed in an air-tight housing, so that it can function in a vacuum.

The Limitation of the Signal and Its Conversion into Sound. — The
current impulses from the photocell, which are of the order of 10~~7 A,
are first very strongly amplified. For this purpose a wide-band am-
plifier is used which gives amplication constant within 6 db in a region
from 30 to 500,000 c per sec. These voltage impulses are modulated,
in the first place by fluctuations resulting from contaminations on the
sound film, but in addition a noise connected with the powerful ampli-
fication is superimposed on the whole signal.

As has already been mentioned in discussing the principle of the
method, these disturbances are eliminated by limiting the signal.
For this purpose a pentode with high anode resistance is used. As is
known, by introducing a sufficiently high resistance in the anode cir-
cuit of such a valve the Ia — V0 characteristic can be made to as-
sume the shape3 of the curve k in Fig. 6. If, then, we apply to the
valve a negative grid voltage so high that Ia = 0 even for the most
powerful disturbances occurring, in the absence of a signal, and make
provision for the signal, on the other hand, to be so powerful as always
to generate the maximum anode current, likewise for the most power-

Nov. 1940 COUNTERACTING NOISE IN FILM REPRODUCTION

439

ful disturbances, then the object has been attained. (See Fig. 6).
Finally the signal prepared in this manner needs only to be sent
through a filter that allows the frequencies of the audible region to
pass through and eliminates all the others. It may then be fed to the
loudspeaker via a power pentode.

46778

FIG. 6. Diagram of the double limitation of the photo-
current signal by means of a pentode. Si = incoming signal
showing disturbances caused by contaminations on the film
and disturbances caused by the powerful amplification.
Su = outgoing signal, k ia-Vg = characteristic of the pen-
tode. Starting from an arbitrary point PI of Su\ the corre-
sponding point P-i of Su can be constructed with the aid of
the auxiliary points Q and R. Since the time units on the t\
and /2 axes are similar, OR cuts the angle between the / axes
through the center.

Conclusion. — By means of the method of counteracting noise
described here it is possible to obtain a perceptible improvement
in quality of the sound reproduced. At the present stage of de-
velopment the improvement in the case of new films, which are

440 W. K. WESTMIJZE

therefore practically free of contamination, is of no significance.
In the case of films which have been used several times, however, the
improvement is considerable. The method described thus makes it
possible to use a film much longer than was previously possible, with
retention of the original quality.

REFERENCES

1 For a description of the Philips Miller system, see Philips Tech. Rev., 1
(1936), pp. 107, 135,211.

2 SCHOUTEN, J. F.: "Synthetic Sound," Philips Tech. Rev., 4, (1939), p. 167.

3 Cf. also Philips Tech. Rev., 5 (1940), p. 61.

SOCIETY ANNOUNCEMENTS

ATLANTIC COAST SECTION MEETING

The first fall meeting of the Atlantic Coast Section of the Society was held at the
Hotel Pennsylvania on October 16. Frank S. Cillier, Associate Director of Re-
search, Encyclopaedia Britannica Films, Inc., presented a most interesting paper
entitled "Blueprinting the Classroom Film."

Following the showing of a 16-mm motion picture, Using the Classroom Film,
Dr. Cillier described the unique characteristics developed by the classroom film in
the 20 years since the advent of sound. He drew conclusions regarding classroom
film research, writing, production, and distribution from the practical experiences
of classroom film producers, here and abroad, and in the light of present-day
trends. Dr. Cillier showed two typical classroom films.

Interesting questions and answers developed in the discussion period which fol-
lowed. The paper will be published in a subsequent issue of the JOURNAL.

MIDWEST SECTION MEETING

Three speakers addressed the October 10 meeting of the Midwest Section of the
Society held in Chicago. Jack Kielty described the special problems of making a
traffic safety documentary 16-mm film, X Marks the Spot. A showing of the film,
which was produced by the Jam Handy Organization for the State of New Jersey,
opened the meeting.

Clyde R. Keith, Chairman of the American Standards Association Sectional
Committee on Motion Pictures Z22, spoke on the work of the American Standards
Association in processing motion picture standards. The democratic structure of
the committee was emphasized and the procedure for establishment of a standard
was described.

John Boyers, assisted by R. J. Tinkham of Magnecord, Inc., Chicago, pre-
sented a paper on "High-Fidelity Magnetic Recording for Motion Picture Produc-
tion." The decision to drive the 0.004-in. OD wire by a flywheel capstan was ex-
plained, as well as details of magnetic head construction. An enjoyable demon-
stration included musical passages, direct playback, echo and other effects.

The last two papers were also presented at the Hollywood Convention and will
appear in the JOURNAL.

INCREASE IN MEMBERSHIP DUES

Personal letters were recently mailed to all Associate and Student members of
the Society by M. R. Boyer, Financial Vice-President, announcing an increase in
annual membership dues. At the meeting of the Board of Governors held during
the 60th Semiannual Convention in Hollywood, it was brought to the attention of
the Board that our present Associate and Student membership dues were insuffi-
cient to cover the increased cost of JOURNAL publication and administration.

441

442 SOCIETY ANNOUNCEMENTS

The Board, therefore, took the only action possible and voted to raise the dues
of Associate members from $7.50 to $10, and of Student members from $3 to $5,
annually. Bills for 1947 dues for these two grades, therefore, will show this in-
crease.

At this time, Mr. Boyer would like to urge the many Associate members who are
eligible for Active membership to consider applying for this higher grade member-
ship in the Society. Many members find that active participation in Society
affairs materially increases the value of the Society to them and their companies.
Since only members in the higher grades are eligible to vote and hold office, oppor-
tunities for participating in Society affairs are obviously better for members in the
Active grade.

INCREASE IN JOURNAL SUBSCRIPTION RATE

Owing to increased costs of JOURNAL publication and administration, the Board
of Governors of the Society has voted to raise the nonmember subscription rate
to the JOURNAL from $8 to $10 annually, effective January 1, 1947. Single copies
will be increased to $1.25 each. The Board also voted to discontinue discounts
for subscriptions and single copies received through accredited agencies, effective
January 1, 1947.

JOURNAL OF THE SOCIETY OF
MOTION PICTURE ENGINEERS

Vol 47 DECEMBER 1946 No. 6

CONTENTS

PAGE
Sensitometric Control of the Duping Process

J. P. WEISS 443

Rapid Test for Ferricyanide Bleach Exhaustion

L. E. VARDEN AND E. G. SEARY 450

Tone Control for Rerecording C. O. SLYFIELD 453

Postwar Test Equipment for Theater Servicing

E. STANKO AND P. V. SMITH 457

Zoom Lens for Motion Picture Cameras with Single-
Barrel Linear Movement F. G. BACK 464

A New Selsyn Interlock Selection System

D. J. BLOOMBERG AND W. O. WATSON 469

A Processing Control Sensitometer G. A. JOHNSON 474
Television and the Motion Picture Theater L. B . ISAAC 482

Technical Problems of Film Production for the Navy's
Special Training Devices H. S. MONROE 487

An Improved 200-Mil Push-Pull Density Modulator
J . G. FRAYNE, T. B . CUNNINGHAM, AND V. PAGLIARULO 494

Current Literature 519

Society Announcements 520

Index to Journal, Vol 47 (July-December, 1946) :

Author Index 532

Classified Index 536

Copyrighted, 1947, by the Society of Motion Picture Engineers, Inc. Permission to republish
material from the JOURNAL must be obtained in writing from the General Office of the Society.
The Society is not responsible for statements of authors or contributors.

Indexes to the semiannual volumes of the JOURNAL are published in the June and December
issues. The contents are also indexed in the Industrial Arts Index available in public libraries.

JOURNAL

OF THE

SOCItTY of MOTION

MOTCL PENNSYLVANIA

NCW YORK I, N-Y

PICTURE ENGINttPvS

TCL. PCNN. 6 O62O

HARRY SMITH, JR., EDITOR

Board of Editors
ARTHUR C. DOWNES, Chairman

JOHN I. CRABTREE ALFRED N. GOLDSMITH EDWARD W. KELLOGG

CLYDE R. KEITH ALAN M. GUNDELPINGER CHARLES W. HANDLEY

ARTHUR C. HARDY

Officers of the Society
^President: DONALD E. HYNDMAN,

342 Madison Ave., New York 17.
*Past-P resident: HERBERT GRIFFIN,

133 E. Santa Anita Ave., Burbank, Calif.
^Executive Vice-President: LOREN L. RYDER,

6451 Marathon St., Hollywood 38.
** Engineering Vice-President: JOHN A. MAURER,

37-01 31st St., Long Island City 1, N. Y.
^Editorial Vice-President: ARTHUR C. DOWNES,

Box 6087, Cleveland 1, Ohio.
** Financial Vice-President: M. R. BOYER,

350 Fifth Ave., New York 1.
* Convention Vice-President: WILLIAM C. KUNZMANN,

Box 6087, Cleveland 1, Ohio.
^Secretary: CLYDE R. KEITH,

233 Broadway, New York 7.
^Treasurer: EARL I. SPONABLE,

460 West 54th St., New York 19.

Governors

"tFRANK E. CAHILL, JR., 321 West 44th St., New York 18.
**FRANK E. CARLSON, Nela Park, Cleveland 12, Ohio.
**ALAN W. COOK, Binghamton, N. Y.

*JOHN I. CRABTREE, Kodak Park, Rochester 4, N. Y.

*PETER MOLE, 941 N. Sycamore Ave., Hollywood.
*JHoLLis W. MOYSE, 6656 Santa Monica Blvd., Hollywood.

*WILLIAM A. MUELLER, 4000 W. Olive Ave., Burbank, Calif.
*°A. SHAPIRO, 2835 N. Western Ave., Chicago 18, 111.

*REEVE O. STROCK, 111 Eighth Ave., New York 11.

*Term expires December 31, 1946. {Chairman, Atlantic Coast Section.
**Term expires December 31, 1947. jChairman, Pacific Coast Section.
*° Chairman, Midwest Section.

Subscription to nonmembers, $8.00f per annum; to members, $5.00 per annum, included in
their annual membership dues; single copies, $1.00.f Order from the Society at addnss above.
Published monthly at Easton, Pa., by the Society of Motion Picture Engineers, Inc.

Publication Office, 20th & Northampton Sts., Easton, Pa.

General and Editorial Office, Hotel Pennsylvania, New York 1, N. Y.

Entered as second-class matter January 15, 1930, at the Post Office at Easton,

Pa., under the Act of March 3. 1879.

t See special notice on page 528.

JOURNAL OF THE SOCIETY OF
MOTION PICTURE ENGINEERS

Vol 47 DECEMBER 1946 No. 6

SENSITOMETRIC CONTROL OF THE DUPING PROCESS*

J. P. WEISS**

Summary. — Because of head-and-tail development effects, sensitometric exposures
of the standard type cannot be wholly relied upon to give a true indication of picture
contrast. A method is described whereby sensitometric exposures of unusual form can
be used to establish processing techniques for making dupes. Good correlation with
visual judgment of quality of a standard picture is obtained.

In testing films designed for use in the duplicating process, we in
the du Pont Laboratory have considered it desirable to make a sen-
sitometric evaluation as well as a practical picture duping test. The
latter tells us if the over-all result is good, but only the former can
give quantitative information on how close the match is at all density
levels throughout the picture. Unfortunately, ordinary sensitom-
etry falls short of providing a true measure of film performance and
of the conditions for obtaining most accurate reproduction. This
undoubtedly has been the experience of everyone who has tried to use
ordinary sensitometric methods in the selection of processing condi-
tions for making dupes.

Confining our discussion to exposures made with intensity-scale
sensitometers exclusively, the chief source of error in ordinary sensi-
tometry is the development phenomenon variously known as "Eber-
hard," "area," or "neighborhood" effect. In the case of unidirec-
tional development in a continuous developing machine it is called
the "head-and-tail" or "directional" effect. Briefly reviewed, it is
the fact that the density produced at any given point on a film de-
pends not only upon the exposure at that particular point and the
processing, but also upon the densities of the adjacent areas. Re-

* Presented May 6, 1946, at the Technical Conference in New York.
** Technical Division, Photo Products Department, E. I. du Pont de Nemours
& Co., Inc., Parlin, N. J.

443

•444

J. P. WEISS

Vol 47, No. 6

action products from the developing process tend to inhibit further
development. If an exposed area follows high densities through a
continuous developing machine, the resulting high concentration of
development products streaming over the area will cause it to have
lower density than if it had been preceded by unexposed emulsion.
Vigorous agitation minimizes this density depression by sweeping
away these reaction products, but in most cases not enough agitation
is provided in a motion picture developing machine to eliminate the
effect entirely.

The physical form of the usual sensitometric exposure causes it to
be particularly susceptible to modification by the directional effect.

DENS

2.0

1.5

MASTER POSITIVE

9.4 97 OO 0.3 0.6 0.9 1.2 1.5 1.8

FIG. 1. Varying sensitometric data from a single film.

Ordinarily it consists of relatively large exposure elements (each one
cm square) arranged in orderly progression from light to heavy. If
the strip travels through the developer low density end first, the re-
sulting characteristic curve will have somewhat higher toe, lower
gamma, and more pronounced shoulder than if the same strip had
been developed high end first.

Fig. 1 shows the effects obtained with four different area condi-
tions and unidirectional development. First is a standard sensi-
tometric exposure developed low end first; second is the same devel-
oped high end first. For curve C the exposure was confined to iso-
lated Y4-in. circles. The last curve represents 1/^-'m. circles sur-
rounded by very high densities. Although exposure and develop-

Dec. 1946 SENSITOMETRIC CONTROL OF DUPING PROCESS 445

ment were identical in all cases we have four different characteristic
curves for the same film. None of these truly depicts the H and D
curve actually followed by the picture elements, for obviously the area
conditions are not the same. In a typical picture the density ele-
ments are small and distributed in random fashion so that directional
effect is less evident.

Our approach to the problem of avoiding the pitfalls of directional
effect was to make sensitometric exposures of unusual form to approx-

FIG. 2. Gray-scale "mixed-density" test chart.

imate as closely as possible the neighborhood conditions prevailing in
a normal picture. To meet the latter requirement we had to:

(1) Keep the individual picture elements as small as could be read on a stand-
ard densitometer;

(2) Avoid the regular progression of densities from low to high as in ordinary
H and D exposures;

(5) Make the average density about the same as in an average picture;
(4) Avoid density extremes not ordinarily present in pictures.

We made a gray-scale chart of the desired form from pieces of print-
ing paper flashed and developed to various densities, then photo-
graphed the chart with a 35-mm motion picture camera to obtain a
negative suitable for print-through use. Fig. 2 shows a frame from
this negative.

Great care was taken to avoid a series of progressively increasing

446

J. P. WEISS

Vol 47, No. 6

densities in the pattern. As a further precaution against systematic
errors, in each frame there are two different elements with the same
value of density so that it will be adjacent to a greater variety of
other density areas. The picture area is divided into 16 elements.

We used a series of different exposures in making the original nega-
tives such as to approximate the density scales appearing in a wide
variety of picture negatives.

This "mixed-density" negative is used for print-through sensi-
tometry in exactly the same manner as if it were a standard sensito-
meter strip. A few frames are spliced to the negative being dupli-
cated, and the printed densities read at each stage of the operations.

DUPE NEG.
DENSITY

2.0l—

1.0

0.5

'MIXED DENSITY"
PRINT- THRU -/

'REGULAR" PRINT -THRU

1.0 1.5 2.O

ORIGINAL NEGATIVE DENSITY

FIG. 3. Comparison of over-all density reproduction curves.

Results from the mixed-density exposures have been quite gratify-
ing. We are able to obtain excellent correlation between sensito-
metric data and accuracy of reproduction as judged pictorially.

Fig. 3 illustrates how the mixed-density print-through results com-
pare with the print-through from a standard-type sensitometer strip.
The graph is a plot of dupe-negative density against density of the
original negative. Perfect duplication would of course be represented
by a straight line at a 45-deg angle. The average slope of the repro-
duction curve as indicated by the standard print-through strip is less
than unity, while it is a little greater than unity according to the
mixed-density curve. An actual picture carried through the duping
process at the same time was judged slightly too contrasty according

Dec. 1946 SENSITOMETRIC CONTROL OF DUPING PROCESS 447

to visual comparison of final prints. Note also that the considerable
degree of curvature to the "regular" plot emphasizes the cumulative
nature of errors caused by directional effects in the duping process.
Alteration of the densities of the master positive H and D strip by
directional effect introduces an exposure error when printing through
the strip onto the dupe negative stock, which error is in addition to the
directional effects in the dupe negative development itself. The final
"regular" curve would indicate that good reproduction could not be
obtained, a conclusion disproved by the fact that many laboratories

FIG. 4. Modified "mixed-density" test chart.

routinely produce duplicate prints almost indistinguishable from
originals.

The mixed-density print- through system is likewise superior to the
method of including original sensitometric exposures with the master
positive and dupe negative developments. If these are developed
low end first according to usual practice, the measured gammas of
both are lower than is actually the case in the picture image. If de-
velopment is adjusted to give a measured gamma product of unity, a
picture made under these conditions will appear too hard.

Fig. 4 illustrates a mixed-density type of test negative modified to
combine a picture with the series of small uniform densities. This
modified test object was used in some recent work using duping tech-

448 J. P. WEISS Vol 47, No. 6

niques to alter the contrast of the original rather than to obtain an
accurate reproduction. It provides a sensi tome trie and a pictorial
means of estimating contrast right in the same frame. As before,
we had gratifying success in being able to predict picture contrast
from sensitometric data.

We have found that the mixed-density technique spells the differ-
ence between success and failure in the application of sensitometry to
the evaluation of dupes, by avoiding the errors which render the
ordinary methods not wholly reliable for this use. To a processing
laboratory it offers a sound quantitative basis for estimating quality
of dupes and simplifies the diagnosis of unsatisfactory results. We
do not pretend, on the other hand, that it would automatically lead
to better quality than the artistic approach, for many laboratories
already produce dupes of superlative quality; nor does the mixed-
density offer any advantages over ordinary sensitometer exposures
for maintaining processing levels once they have been established ; it
is needed for process evaluation purposes rather than in subsequent
routine control.

Certain precautions must be observed in applying the mixed-
density print-through method.

First, it is mandatory to have even illumination and uniform ex-
posure in the printer. While this is a general requirement for first-
class printing, small irregularities which are tolerable in pictures will
lead to erratic results when a densitometer is used.

Second, if it is desired to set up printing and developing conditions
with the criterion for success that the gamma product shall be unity,
the densitometry must be beyond reproach. This means that the
densitometer must measure a value of image density appropriate in
both its spectral and geometric aspects to the actual conditions of
use. Thus the original and dupe negatives should be measured in
terms of printing density, while visual density should be used for
describing the master positive and final print. If a contact-type prin-
ter is used, a diffuse density is the proper characteristic to measure.
On the other hand, when duping is done on an optical printer the
effective density is something intermediate between diffuse and specu-
lar density, and a densitometer which measures diffuse density would
be inappropriate.

In conclusion, the mixed-density system of sensitometry is an illus-
tration of a photographic fact learned through long experience;

Dec. 1946 SENSITOMETRIC CONTROL OF DUPING PROCESS 449

namely, that to be reliable, test conditions must parallel closely the
conditions of use. The departure in the form of the standard sensi-
tometric exposure from the random density distribution in an actual
picture leads to errors from head-and-tails development effects which
are especially serious under duping conditions. Mixed-density ex-
posures sidestep these directional errors by closely imitating the neigh-
borhood density conditions in an average picture. By this method
we secure a more accurate evaluation of film performance in the dup-
ing process than by conventional sensitometry.

RAPID TEST FOR FERRICYANIDE BLEACH EXHAUSTION*
L. E. VARDEN** AND E. G. SEARY**

Summary. — Ferricyanide-type bleach solutions for removing silver in certain
monopack color processes become relatively inefficient upon the accumulation of but a
few per cent of ferrocyanide ion. A method is given for the rapid determination of the
state of exhaustion of such bleach solutions, based on colorimetric estimation of the
ferrocyanide concentration. The method is sufficiently reliable for processing control,
and equipment common to motion picture laboratories is adaptable.

Introduction. — In the processing of certain types of color film and
color printing materials, it is necessary to remove a silver image to re-
veal the desired color image. For most monopack color processes a
ferricyanide-type bleach is used in place of the stronger dichromate or
permanganate bleach solutions recommended for black-and-white
reversal processing. This is necessitated by the sensitivity of most
dyes to strong oxidizing agents.

A typical silver bleach for monopack color materials consists of a
soluble ferricyanide, a soluble halide, and suitable buffering com-
pounds for maintaining £H constancy of the solution. The silver
bleaching reaction in such a solution takes place in two steps. First,
the ferricyanide ion attacks the silver image, oxidizing it to the rela-
tively insoluble silver ferrocyanide, whereupon the lesser soluble silver
halide is formed by reaction of the silver ion with the halide ion pres-
ent in the solution. Thus,

4Ag + 4K3Fe.(CN)6 -> Ag4Fe(CN)6 + 3K4Fe(CN)6 (1)

Ag4Fe(CN)6 + 4KBr -* K4Fe(CN)6 + 4AgBr. (2)

It will be noted that the ferrocyanide ion increases in concentration
as the ferricyanide ion is depleted during exhaustion of the solution.
The determination of the ferrocyanide concentration forms a desir-
able criterion of exhaustion since a small increase in ferrocyanide pro-
duces a disproportionately large decrease in bleaching rate.

* Presented May 6, 1946, at the Technical Conference in New York.
** Pavelle Color Incorporated, New York.
450

TEST FOR FERRIC YANIDE BLEACH EXHAUSTION

451

A simple, rapid and reliable colorimetric method for ferrocyanide
determination has been found in the precipitation of the ferrocyanide
with ferric chloride. The resultant ferric ferrocyanide is the in-
tensely colored compound, Prussian blue.

3K4Fe(CN)6 + 4FeCl3 -* Fe4[Fe(CN6)]3 -h 12KC1 (5)

As the bleach is used and the ferrocyanide ion concentration increases,
the color produced by the addition of ferric chloride gradually changes
from yellow to green. The change in color is sufficient to allow visual

§14

o 12

£5

SQUARE FEET OF ANSCO PRINTON F
3 ro A o> OD c

PHOTOMETER READING IN MICROAMPERES

FIG. 1.

comparisons to be made against standard solutions or to permit direct
absorption measurements with a simple photometer. The latter
method is preferred since standard solutions are unstable.

Apparatus. — Since most processing laboratories have Eastman
Kodak Argentometers for estimating silver content of fixing solutions,
an attempt was made to adapt a Model B Argentometer, 1 equipped
with its customary photronic cell and microammeter, as a photom-
eter. It was found that the original rubber-ended Argentometer

452 L. E. VARDEN AND E. G. SEARY

cell is not suitable for use with ferricyanide solutions because the rub-
ber is attacked. A fused glass cell, having a light path of 10 mm,
was substituted with complete satisfaction.

Procedure. — In practice the bleach solution to be tested is diluted
100 : 1 with water to reduce the optical density to a measurable
range. The addition of about one per cent of 6N HC1 was found neces-
sary with the bleach used in this work (Ansco 705) in order to keep
buffering material from precipitating. Approximately 25 cc of the
diluted bleach are placed in the glass cell and the cell inserted in the
Argentometer. The instrument is then balanced at 150 /-ia (zero on
the silver scale). Two drops of 10 per cent FeCla are now added and
the solution stirred well. A reading of the microammeter is taken
immediately, since the density of the solution slowly changes as the
Prussian blue flocculates and eventually settles out. When first
mixed, however, the test solution is perfectly clear and reproducible
results are possible.

Experimental. — A series of freshly prepared bleaches of known
exhaustion was tested by the above procedure. The decrease in
light transmission of the test solutions was found to proceed in a regu-
lar mariner with extent of exhaustion, as is shown in Fig. 1. Here
the photocell current is plotted against the log of the area of material
processed.

Conclusion. — The writers do not feel that, a mere measure of the
area of material processed is a valid indication of bleach exhaustion.
Variation in exposure, and especially contamination, dilution, and
aeration effects can affect the exhaustion markedly. For this reason
a more positive measure of bleach exhaustion, such as the one out-
lined above, is pref err able for laboratory control. Although the data
given are for a monopack color print material, the method is appli-
cable for monopack taking and duplicating films where ferricyanide-
type bleach baths are employed.

REFERENCE

1 WEYERTS, W. J., AND HICKMAN, K. C. D. : "The Argentometer — An Appara-
tus for Testing for Silver in a Fixing Bath," /. Soc. Mot. Pict. Eng., XXV, 4
(Oct. 1935), p. 335.

TONE CONTROL FOR RERECORDING*

•

C. O. SLYF1ELD**

Summary. — This paper covers the use of a tone track for automatic control of
music and effects background in rerecording.

During World War II those of us who were left at the studios were
exceedingly busy working on government training films of a wide
variety. It was necessary that these films be finished as quickly as
possible as they were urgently needed for training the large number of
men being taken into the military and naval services. Some of these
films had no music or effects background behind the narration, but
many of them had. This meant that the mixers must use footage
counters, screen cues, or other means to indicate where the dialogue or
narration started and stopped in order to fade or bring the back-
ground volume up in the proper places. Watching a footage counter
or screen cues for long periods of time is very tedious and those of you
who have experienced this know that it is very fatiguing.

Our first attempt to avoid this process was to use the RCA so-
called "up-and-downer." This worked quite satisfactorily, but, when
the narrator stopped for breath or made a slight pause, the back-
ground would rise in volume which was not desirable during these
brief intervals. The timing constants of the rectifier and variable
gain amplifier were adjusted to avoid this to a degree, but in doing
so the fade and increase in volume time were not satisfactory.

In view of this difficulty, the idea was conceived of using a constant
amplitude tone to control the gain of the amplifier through which the
music and effects background was being transmitted. In other
words, a tone would be used to control the gain of the amplifier in-
stead of rectifying the dialogue signal itself as is done in the "up-and-
downer." In our particular case, a frequency of 1000 cycles was used
although other frequencies could be used just as well.

A great deal of experimental work was done on timing constants

* Presented May 9, 1946, at the Technical Conference in New York.
** Sound Department, Walt Disney Productions, Burbank, Calif.

453

454

C. O. SLYFIELD

Vol 47, No. 6

which would give optimum results. The one which was picked as
being most satisfactory for our service was that in which the fade took
place in 15 frames and the increase back to normal level occurred in
30 frames. With these timing constants, the listener does not feel
that the background is suddenly reduced for the dialogue or that the
rise in volume at the end of the dialogue is so drastic as to be too
noticeable. The music just seems to get out of the way without the
feeling that the background is suddenly changed or notched to make
way for the dialogue.

Some types of music and effects have to be dropped lower in volume
than others in order to clear the dialogue. For this purpose a tone
fader is provided on the rerecording console so that the amount of
tone to be rectified is under control of the mixer at all times.

To prepare the control track, the dialogue track is set up on a dual
counter with the second sprocket used for assembling the tone control

FIG. 1. Tone track arrangement.

track. A piece of 1000-cycle tone is spliced into this track so that it
starts and stops with the dialogue. This is done for the entire reel.
After the control track has been completed, it is advanced six frames
and a "sync" mark applied. This, it will be seen from Fig. 1, makes
the tone appear six frames ahead of the dialogue entrance. The tone
also ends six frames before the end of the dialogue or narration. This
six-frame advance allows */4 sec anticipation before the entrance of
the dialogue and causes the gain of the variable-gain amplifier through
which the background is passing to be reduced sufficiently so that the
first word of dialogue is not missed. The fade continues for 15
frames from the start of the tone and holds the background to a pre-
determined level until the tone ends — six frames before the finish of
the dialogue or narration. At this time the rise in volume of the back-
ground begins and continues for 30 frames.

The block diagram of Fig. 1 shows the arrangement of the tone

Dec. 1946

TONE CONTROL FOR RERECORDING

455

track in relation to the other tracks which are being rerecorded. The
block diagram, Fig. 2, shows how the rerecording system works with
the tone track controlling the gain of the amplifier through which the
background is being transmitted. It will be noted that the dialogue
on sound head No. 1 is amplified through preamplifier No. 1 and the
volume controlled by fader No. 1. The dialogue signal is then im-
pressed across one of the primary windings of transformer Tl .

Music and effects are run on sound heads Nos. 3 and 4 and con-
trolled by faders Nos. 3 and 4, respectively. The combined outputs
of these two tracks are sent through the variable-gain amplifier as
shown. The tone track is run on sound head No. 2, amplified by
preamplifier No. 2 and the volume of the tone controlled by fader
No. 2. The tone output is then rectified by the tone rectifier as

SOUND
HEAD 1

PRE-AMP.

FADER

T
O

NO. 1

NO 1

TONE

SOUND
HEAD 2

PRE-AMP

FADER

TONE

NO 2

RECTIFIER

MUSIC

SOUND
HEAD 3

PRE-AMP

FADER

NO 3

EFFECTS

VARIABLE
GAIN
AMPL FIER

SOUND
HEAD 4

PRE AMP

FADER

NO. 4

NO 4

FIG. 2. Tone track rerecording setup.

shown in the diagram. The resulting negative direct-current poten-
tial is applied to the grids of a pair of push-pull 6K7 tubes in the
variable-gain amplifier to reduce the gain of that device. The higher
the value of tone, the greater will be the negative voltage applied
to the grids of the 6K7 tubes thus controlling the volume of the
variable-gain amplifier. The output of this amplifier is then im-
pressed across the second primary coil of transformer Tl . The com-
bined outputs of the dialogue and the variable-gain amplifier in the
secondary winding of this transformer are transmitted to the re-
cording amplifier and then to the recorder itself. It will be noted
that the volume of the dialogue is controlled manually, but the vol-
ume of the music and effects is controlled both manually and by
means of the rectified tone which actuates the variable-gain amplifier
through which these elements are passing.

456 C. O. SLYFIELD

The 1000-cycle sections can be removed from the tone track after
the rerecording has been completed, spliced together and used over
and over again.

Other methods have been considered for accomplishing the same
results as are obtained with the tone control track, but these have
been discarded for various reasons. One method suggested was the
use of a notch on the edge of the dialogue track which would either
operate a tone or some other device for controlling the volume of the
amplifier through which the background was being transmitted. This
system could be used quite easily, but there is always the problem of
making changes after the rerecording tracks have been prepared. It
is somewhat difficult to change the position of a notch, whereas the
tone track can be changed quite readily by simply using more or less
tone or changing the position of the tone which has already been
spliced into the reel. The tone track, of course, requires no blooping
so changes can be made with speed and little difficulty.

The tone control track system works remarkably well and is par-
ticularly adapted for dialogue pictures which have more or less of a
continuous background of music and effects. It has been used for
the past two years and has proved itself to be highly satisfactory.

POSTWAR TEST EQUIPMENT FOR THEATER SERVICING*
EDWARD STANKO** AND PAUL V. SMITH**

Summary. — This article outlines the underlying reasons for the need of new and
modern test equipment for properly servicing theatrical sound reproducing equipment
with the minimum expenditure of time. A detailed description is given of a modern
test kit designed to Jill this need, with photographs of the kit and schematic diagrams of
its special instruments.

Early in 1946, several months after cessation of the war, new test
equipment for servicing theater installations was designed and is now
in production. With the advent of war in 1941, civilian test and
measuring equipment production was stopped. By the end of the
war the cumulative wear and tear on these instruments, plus emer-
gency repair expedients, made it necessary to replace the major por-
tion of the existing equipment.

Careful studies of the requirements of postwar test equipment in-
dicated the advisability of incorporating certain special test instru-
ments which would expedite field work, aid in obtaining accuracy of
measurement, speed the identification of defective components, and
permit the equipment to be of maximum usefulness under all possible
field conditions. These special instruments for field work will be ex-
plained in detail later in this paper.

The continual improvement of motion picture sound equipment,
with its increased fidelity, imposes more rigorous requirements on the
field test equipment necessary to maintain its circuits and its optical
and mechanical parts in their original condition and adjustment.
Also, this new field test equipment must be easily portable if it is to
be of maximum usefulness. Fortunately, there are now available
miniature tubes and other parts, developed during the war, which con-
tribute materially to this requirement.

During the war, instrument research has been greatly accelerated.
The advent of alnico magnets has made possible stronger magnetic

* Presented May 9, 1946, at the Technical Conference in New York.
** RCA Service Company, Inc., Camden, N. J.

457

458 E. STANKO AND P. V. SMITH Vol 47, No. 6

fields, more rugged moving systems, and higher torque springs, all
of which contribute to accuracy, ruggedness, and reliability, as well as
lighter weight. Miniature tubes have been developed which can
stand the shock of being fired from a gun in an antiaircraft shell.
Tubes incorporating these developments will withstand the shocks
incident to use in portable equipment much better than prewar tubes.
With these things in mind, a new 1946 model test kit has been de-
signed, consisting of a varnished, reinforced fiber case, with compart-
ments for the various instruments. When the lid is closed, these in-
struments are held firmly in place. The case is 19 in. long, 7l/z in.
deep, and 14 in. high when its carrying handle is folded. Including

FIG. 1. Theater servicing test equipment and carrying case.

all instruments, it weighs 35 Ib. The case contains an RCA Volt-
Ohmyst especially designed to operate on self-contained batteries
instead of on alternating current, which was required for previous
models. A neon pilot lamp is connected to the batteries through a
resister-capacitor network and the selector switch, and flashes re-
peatedly when the batteries are in use. This effectively attracts
attention, and consumes only a few microamperes of battery current.

The VoltOhmyst can be used to measure direct-current voltages
from 0.10 v to 1000 v over six ranges, with an error of less than 2 per
cent of full scale on each range.

Measuring is accomplished by a two-tube, balanced-bridge circuit
which permits a constant, high input resistance of 1 1 megohms on all
ranges, and protects the instrument moving system from burn-out

Dec. 1946 TEST EQUIPMENT FOR THEATER SERVICING

459

should excessive voltage be applied. Because the normally negative
lead is grounded to the instrument case, a reversing switch is provided
to allow measurement where the ungrounded side of a voltage source
is negative. The high-side input probe incorporates a one-megohm
isolating resister, thus permitting measurements to be made in sig-
nal-carrying circuits without adding a serious lead capacitance which
might introduce excessive circuit loading or distortion.

The resistance meter also utilizes the bridge circuit and can be used
to measure resistances from 0.1 ohm to 1000 megohms over six ranges.
A 3-v, heavy-duty battery furnishes voltage for all ranges. The
voltage applied to the part being measured does not exceed, and is

FIG. 2. Circuit diagram of special VoltOhmyst.

usually much less than, 3 v; this is of advantage when small lamps,
alloy-steel core transformers, and other items susceptible to damage
by excess current are to be measured.

Because the bridge circuit isolates the meter moving system from
the measuring circuit, accidental connection of the ohmmeter leads to
a live circuit or to a charged capacitor will not result in instrument
burn-out. The resistances themselves are rugged enough to with-
stand a large momentary surge or overload without damage.

The special VoltOhmyst can be used to measure alternating-current
voltages of commercial and low audio-frequencies from one volt to
1000 v, over five ranges. When so used, it becomes a conventional
copper-oxide-rectifier voltmeter with a resistance of 1000 ohms per
volt. No series resistor is used in the alternating-current leads.

460 E. STANKO AND P. V. SMITH Vol 47, No. 6

The measuring circuit is isolated from the case ground, when alter-
nating voltage is measured.

This special VoltOhmyst also measures direct current from 100
jua to 5 amp, over six ranges, with accuracy of ±2 per cent of full
scale, by the use of shunts on the meter moving system. The measur-
ing circuit is isolated from the case ground when measuring current.

A standard socket selector kit with a complete complement of
adaptors and sockets, including those for the new seven-prong mini-
ature tubes, is furnished for analysis of tube circuits. Two pin jacks
are installed in the top of the special VoltOhmyst described above;
these provide for mounting the socket selector block on the top surface
of the VoltOhmyst. Connections are readily made to the socket jacks
with the standard VoltOhmyst leads. When the sockets and adap-
tors are not in use, their prongs are pushed into rubber-lined dummy
sockets in three wooden carrier blocks which are then slid into corre-
sponding slots in the carrying case. The soft rubber dummy socket
hole linings grip the prongs firmly but easily. This facilitates acces-
sibility of the sockets and adapters, and helps prevent loss or damage.
The general knowledge of the use of these socket selector kits renders
any detailed explanation herein unnecessary.

An alternating-current voltmeter and decibel meter is furnished
with the test kit. This instrument is smaller and lighter than the
special VoltOhmyst and, since it can be used for alternating-current
voltage measurements from 0.25 to 150 v, may serve as a pocket in-
strument.

This decibel meter is necessary when testing personnel are making
frequency response runs, transmission tests, gain or loss tests, amplifier
overload checks, or similar system measurements. Instruction sheets
are furnished with this instrument to enable the user to correct the
readings should the instrument be used on lines of impedances differ-
ing from that for which the instrument was calibrated, or to convert
the readings to those corresponding to a different zero level, or both.
This voltmeter includes a series blocking capacitor which is to be
used when the instrument is connected to a circuit carrying both di-
rect and alternating current.

An alternating- or direct-current ammeter also is supplied. This
instrument can be used to measure currents of from 0.1 to 15 amp,
either alternating or direct current, over two ranges. In addition, a
pair of terminals is provided to connect to the leads from any standard
50-millivolt drop shunt, for measuring higher values of direct current.

Dec. 1946 TEST EQUIPMENT FOR THEATER SERVICING

461

The latter two instruments are held in place in the case by wooden
blocks, and when the cover is closed, cannot shift or rattle around.

A 150-amp, 50-millivolt drop shunt is also furnished, which can be
connected to the ammeter to enable measurements of direct current up
to 150 amp. This is necessary when adjusting arc lamp rectifiers,
checking arc lamp ammeters, etc. This shunt is permanently
mounted on a wood block which slides into a slot in the case adjoining
those on which the socket selector adapters are mounted.

The Triatic Signal Tracer, which is a standard unit in the kit, was
especially developed to fill a long-felt need for a universal tester. It
will measure capacitors of values between. 10 micromicrofarads and

FIG. 3. Circuit diagram of Triatic tester.

80 microfarads, over three ranges, with an accuracy of 5 per cent or
better. The signal tracer will also indicate whether the capacitor
being tested has low or high shunt resistance. Bridge balance is in-
dicated directly by a 6U5/6G5 electron-ray indicator tube. The use
of this tube eliminates delicate instruments or bothersome head-
phones, and the indication is not confused by any harmonics that may
be present in the supply voltage.

It incorporates a three-tube, high-gain audio-frequency amplifier
and a 3-in. PM speaker. The amplifier has approximately 100-db
gain and a power output of 200 milliwatts. Its high impedance input
(1.5 megohm) can be connected directly to a signal-carrying circuit
without appreciably disturbing the signal. The volume control is in

462 E. STANKO AND P. V. SMITH Vol 47, No. 6

the input circuit; therefore, a high-level signal at the amplifier input
can be reduced sufficiently to prevent overloading any part of the
amplifier. A series blocking capacitor in the input circuit prevents
any direct-current voltage from reaching the volume control or the
grid of the first tube ; the input connection can, therefore, be attached
directly to the amplifier plate circuit. An output jack, capacitively
coupled to the plate of the output tube, facilitates the use of head-
phones if desired. Power is supplied by a built-in power supply oper-
ating from the 120-v, 60-cycle alternating-current line.

A jack on the front panel is connected, through a resistor-capacitor
filter, to the integral power supply, and will supply anode voltage to
one or two photocells. This voltage may be connected, by the simple
throwing of a toggle switch, through a one-megohm coupling resistor
to the amplifier input jack. This connection permits the amplifier
to operate directly from a photocell without any other electrical con-
nections whatever.

These features permit the use of this device for many tests, such as:

(1) Focusing optical systems, adjusting lateral guide rollers, or balancing
push-pull photocells on one sound head while the other is in normal operation ;

(2) Checking one sound head for proper operation, normal photocell output
quality, film, exciter lamp, and photocell defects, hum or vibration pickup, elc.f
while the other sound head is in normal operation;

(5) Stage-by-stage tracing to locate sources of noise, distortion, intermittent
operation, or low gain;

(4} Testing auxiliary devices, such as microphones, phonograph pickups, etc.
for normal operation ;

(5) An emergency substitute for the normal photocell anode supply, and the
first stages of theater sound systems, in which failure has occurred ;

(6) Talk-back from stage to projection room when installing equipment (when
used with a microphone) ;

(7) In conjunction with a photocell, to find out whether room illumination
supply is direct current or alternating current. (With a photocell and test lamp,
to determine whether an unknown power source is direct or alternating current.)
If alternating current, hum will be heard from the loudspeaker; if direct cur-
rent, no hum will be heard, but clicks may be produced by interrupting the light
which falls on the photoelectric cell.

The Triatic Signal Tracer also includes a circuit probe tester incor-
porating a resistor-capacitor circuit and the electron-ray tube
6U5/6G5. This circuit provides a means of rapidly checking an
amplifier or similar circuit to determine the approximate voltages
present, their polarity, and whether they are direct or alternating.
Only one range is needed, which is not subject to burnout or damage

Dec. 1946 TEST EQUIPMENT FOR THEATER SERVICING 463

on voltages ordinarily found in an audio-frequency amplifier. This
range extends approximately from 2 v to 500 v. Direct-current vol- '
tages are indicated by the opening or closing of the electron-ray tube
shadow, depending on polarity; the voltage is estimated by noting
the time taken (after the probe has been removed from the circuit
being tested) for the resistor-capacitor circuit to discharge enough to
allow the electron-ray tube shadow to return to its normal position.
This is accurate enough for rapidly checking circuits in case of trouble,
when it is merely necessary to determine whether or not the voltage
present is somewhere near its normal magnitude.

The alternating-voltage indication is a blurring of the edges of the
electron-ray tube shadow, the result of the rapid opening and closing
produced by the applied alternating voltage. Since alternating volt-
age cannot charge up resistor-capacitor circuits to any permanent
value, voltages are indicated by the width of the blurred area.

All necessary cables, plugs, and cords are furnished with this equip-
ment. These are placed in space provided in the bottom of the cab-
inet. In addition, a bracket is provided on the back of the special
VoltOhmyst for carrying its own plug leads.

Additional space is available for other small instruments should
future developments require their use in addition to those now pro-
vided.

The large number of varied and comprehensive tests possible with
this equipment will allow the theater service engineer to keep equip-
ment running at its optimum, and to locate any trouble which may
develop. The service of a competent engineer equipped with modern
test equipment is the best possible guarantee of continuous high-
quality sound reproduction.

ZOOM LENS FOR MOTION PICTURE CAMERAS WITH
SINGLE-BARREL LINEAR MOVEMENT*

FRANK G. BACK**

Summary. — Previous varifocal lenses used two or three movable components which
had to be shifted against each other. This movement was necessarily nonlinear and
therefore had to be achieved by nonlinear cams. This presented many mechanical
difficulties and it was nearly impossible to obtain an accurately focused image over the
entire range. In addition, the shifting of the movable components against each other
caused numerous aberrations.

The new varifocal lens has only one movable barrel, and the compensation of the
image movement is achieved solely by optical, and not by mechanical means. There-
fore, the lens gives a well-focused image sufficiently free from aberrations at all posi-
tions.

In order to avoid monotony or accentuate details, it is desirable in
many instances to start a motion picture scene with a close-up and
end it with a long shot, or vice versa, in such a way that the transition
is not effected abruptly but that the object seems gradually to come
closer or go farther away on the screen. Such an effect is achieved in
three different ways.

First: The camera itself moves toward or recedes from the object.
This method is complicated and requires the teamwork of several
highly skilled camera operators in order to "follow focus," get a
smooth movement, and keep the target in proper range and frame.
At the same time, this method often necessitates complicated and
awkward installations and is mainly used in studio work.

Second: The shots are taken from a stationary camera with a wide-
angle lens, and the zoom effect is achieved later by means of an optical
printer. This method is rarely used because steady movements over
a large number of frames without change in exposure are difficult to
achieve. Moreover, if the detail to be shown is not in the center of
the frame, steady movements are almost impossible to accomplish.

* Presented May 8, 1946, at the Technical Conference in New York.
** M.E., Sc.D,, Research and Development Laboratory, 381 Fourth Ave.,
New York.

464

ZOOM LENS FOR MOTION PICTURE CAMERAS

466

Third: A stationary camera is used in conjunction with a varifocal
lens, a so-called "zoom lens." The requirements for a zoom lens are
as follows: Good definition over the entire range of focus; smooth
operation; no change in exposure during the zooming; and foolproof
performance.

The zoom lenses developed to date do not meet the above-described
requirements because of their rather complicated mechanical design,
and for this reason they have not been accepted in spite of their un-
deniable advantages. For each of the focal lengths required the

Front Vari-
Lens ator

Erector

Relays

Film

1,4.6 STATIONARY LENSES
2,3,5 MOVABLE LENSES

FIG. 1. Optical principle of Zoomar lens.

designer computed the exact position of each component, trying, of
course, to keep the number of movable elements as small as possible.
The shifting of the movable elements was then plotted against the
different focal lengths. The resulting graph was taken as the basis
for a cam movement needed to displace the movable lens elements
with relation to each other. The disadvantage of such a lens lay
mainly in the fact that it was nearly impossible to obtain an accurately
focused image over the whole focal range because of necessary toler-
ances in manufacturing. Only some points gave satisfactory resolu-

466

Vol 47, No. 6

Dec. 1946 ZOOM LENS FOR MOTION PICTURE CAMERAS 467

tion, while other points were definitely out of focus even if these lens
systems had been fully corrected for the seven aberrations. This in
itself was impossible to achieve with a system of the above-described
type.

An even graver disadvantage of these lenses lay in the fact that
although a skilled craftsman could produce a lens system of such
close tolerances, it is inevitable that wear, even in the most accurately
designed and manufactured mechanism, produces a certain play
sufficient to throw the system out of focus. Therefore, nearly all of
the varifocal lenses marketed so far became unusable after a relatively
short time.

FIG. 3. Zoomar mounted on Cine Kodak Special.

We have designed and developed a new zoom lens where the chief
goal was simplicity, as far as the mechanical movements were con-
cerned, with assurance that the picture quality would not suffer from
normal wear. We have been able to find certain equations and opti-
cal relations which permit full compensation with strictly linear dis-
placements of optical components. The result is a varifocal lens
with a single barrel movement. Any cams, gears, or nonlinear helices
are completely eliminated and the compensation is achieved by optical
means.

There are only two groups of elements in the zoom lens; namely,
stationary elements and coupled movable elements. The stationary
elements are mechanically connected by the housing; the coupled
movable elements are mounted in a common barrel. Movement of

468

F. G. BACK

FIG. 4. Compound
zoom shot of Empire
State Building.

the barrel to any position in the housing yields
a stationary image of varying size.

Fig. 1 shows a simplified schematic view of
our lens arrangement in three positions. Posi-
tion A is the wide-angle position; position B
the medium position ; and position C the tele-
photo position.

Elements 1, 4, and 6 are the stationary
group. Elements 2, 3, and 5 are the movable
group. The stationary elements do not
change their position within the lens-housing;
elements 2, 3, and 5 are moved simultane-
ously. The effect is a picture of variable size
but stationary as far as displacement along
the lens axis is concerned.

Although Fig. 1 shows the principle of the
new zoom lens, each lens element shown repre-
sents a group of lens components in itself,
because every one of these lenses has to be
corrected for chromatic and spherical aberra-
tion and for astigmatism. Coma and distor-
tion are eliminated by the concerted interplay
of all groups.

As already stated, this arrangement com-
pensates for any displacements of image in the
film plane, so that the image remains station-
ary in spite of changing its size. Fig. 2 shows
a cross section of the entire zoom lens.

Fig. 3 shows the zoom lens with a coupled
Zoom-Viewfinder mounted on a Cine Kodak
Special.

Fig. 4 shows a zoom shot taken with the
lens to illustrate its focal range.

In closing, I wish to thank A. L. Varges, of
News of the Day, and W. K. Jacobi, of the Long
Island Optical Company, for their valuable co-
operation. Also, I wish to express my thanks
to H. Lowen, our chief mathematician, for the
untiring and devoted services he rendered in
the course of this research work.

A NEW SELSYN INTERLOCK SELECTION SYSTEM*
DANIEL J. BLOOMBERG** AND W. O. WATSON**

Summary. — This paper describes a new type of Selsyn interlock selector switch
wherein six-pole, five-position rotary switches utilize a combination operating motion.
Position selection is obtained by rotary motion and contacting by plunger motion.

Selsyn motor and distributor systems are universally used in motion
picture production for the synchronizing of the sound, camera, and
projection motor operation. The involved multiple patching meth-
ods generally practiced, using a patch plug and cable connector sys-
tem for the selection and interlocking of motor and distributor busses,
is somewhat cumbersome.

The interlock selection system described in this paper eliminates
cable patching and provides a simple, convenient method of interlock
motor and distributor selection.

The design of a switching method, wherein only two distributor
busses are used, is comparatively easy. However, where three or more
distributors are used, then the problem arises of being able to select
any distributor without contacting a bus which is in operation. The
solution to this problem was the primary incentive to the Republic
Sound Department in designing the multiple interlock switch.

Fig. 1, a block diagram, and Fig. 2, a schematic diagram, depict a
typical Selsyn system. In this arrangement there are three distribu-
tors and five interlock motors. To connect each motor to a distribu-
tor bus it has been necessary in the past to connect a large 6-pin plug
into the de-energized bus at each machine or at a central patching
panel in or near the machine room. The new interlock selection
system provides a switch for each Selsyn motor, and in order to con-
nect a motor to any distributor at rest, it is necessary only to rotate
the switch to a designated position and depress the knob.

In Fig. 3 there are three switch units mounted rigidly by four
spacer rods. Construction of the base and contact arms in the closed

* Presented May 18, 1945, at the Technical Conference in Hollywood.
** Republic Productions, Inc., North Hollywood, Calif.

470

D. J. BLOOMBERG AND W. O. WATSON Vol 47, No. 6

or operating position may be seen in the cut-away cross section.
Each switch unit consists of ten "micro-action" type switches mounted
in a radial form on an insulated base. Two sets of contacts on each

FIG. 1. Typical Selsyn installation.

FIG. 2. Typical Selsyn switching circuit.

switch unit are actuated by means of two plungers spaced 180 deg
apart on the associated actuating disk. The actuating disks are me-
chanically connected together through their centers by extruded rod
as shown. These disks are not rigid, but float on the connecting rod

Dec. 1946 NEW SELSYN INTERLOCK SELECTION SYSTEM

471

and are held in position by the spring 174, Fig. 3. This spring acts
as a cushion and is designed to equalize the pressure required to op-
erate each set of switch contacts. A radial detent, spaced at 21 deg,
positions the actuating pin disks.

The switch is operated by depressing the trigger 144 and pulling out
the knob 141 retracting the actuating pins from the contact position.
The switch may then be rotated to the required position and pressed

1414'

FIG. 3. The new interlock selection switch.

in until the trigger latch engages the circular sear at 148. This motion
moves the actuating disks and plungers into contact with the micro-
type switches, snapping them into the closed circuit position. The
plungers also serve as indexing pins preventing accidental movement
of the knob. Thus the switch may be rotated through any operating
bus position without actuating the micro-type switch, and does not
make contact until the knob is pressed home at the selected bus.

This system has been installed in the Republic Studio Sound De-
partment and the main Selsyn distribution panel has 30 switches con-

472 D. J. BLOOMBERG AND W. O. WATSON Vol 47, No. 6

H *

IT ft

FIG. 4. Main Selsyn distribution panel installed
at Republic Studios.

FIG. 5. Projection room control panel.

Dec. 1946 NEW SELSYN INTERLOCK SELECTION SYSTEM 473

trolling 30 interlock motors and four distributors as shown in Fig. 4.
The projection room in stage 12 has a similar panel with six switches
and is shown in Fig. 5. This system has been in operation for over
a year and has proved highly satisfactory.

The advantages of the new interlock selection system are :

(1) Simplified operation results in reduction of time required to set up the
system,

(2) Ease of transferring any motor to another distributor, after setups have
been made,

(5) Distribution panel may be installed in a more convenient central location
than the patch-type panel because of greatly reduced space requirements (30
switches in a panel 30 in. X 30 in. X 9 in. deep),

(4) Improved appearance of panel with elimination of patch cables.

A PROCESSING CONTROL SENSITOMETER*

•

GERALD A. JOHNSON**

Summary. — A sensito meter which gives an intensity-scale exposure has been
developed as an aid to the control of photographic processing conditions. The illumi-
nation is modulated by a photographic step tablet of 21 steps in which the exposure
increases by increments of \/l?. Positive films are exposed to an incandescent lamp
operating at 2850 K, while with negative films, dyed gelatin filters are inserted in the
optical system to approximate sunlight quality. A pendulum mechanism furnishes a
l/i9-sec exposure time for positive and negative films. Where longer exposure times
are needed, as, for example, with photographic papers, an additional manual shutter
is provided. The instrument gives highly reproducible results for process control but
is not calibrated in absolute units.

As an aid in controlling photographic processing conditions, a small,
portable sensitometer was developed during the war for the Army,
Navy, and Marine Corps. It was called the Processing Control
Sensitometer and proved to be so useful that an improved model has
been built. A picture of it is shown in Fig. 1.

The instrument, built to utilize a voltage of 110 ac but capable of
operation on other voltages, with minor alterations, provides a rapid
and convenient means of making reproducible sensitometric expo-
sures. Film strips from control emulsions, exposed with the sensitom-
eter, can be compared with previously exposed and processed strips
from the same control emulsions. In this way, speed and contrast
trends can be followed closely, and accurate control of the uniformity
of developing conditions can be maintained. The sensitometer can
also be used to test relative speeds, contrasts, and other sensitometric
properties of various films.

In the last 50 years many sensitometers have been developed as
aids in investigating the characteristics of photographic films and
papers. Mees1 has summarized this progress. All sensitometers are
classified either as intensity-scale or time-scale instruments. An
example of the latter class is the Eastman I IB Sensitometer,2 which is

* Presented May 6, 1946, at the Technical Conference in New York.
** Eastman Kodak Company, Rochester, N. Y.
474

PROCESSING CONTROL SENSITOMETER

475

FIG. 1. Processing control sensitometer.

a popular precision instrument giving a constant-intensity, variable-
time exposure. The former group, intensity-scale sensitometers,
which are more common, furnishes constant-time, variable-intensity
exposure with the aid of devices like absorbing tablets. Accurate
step tablets and wedges sold by photographic manufacturers have
made it possible to construct simple and efficient instruments of this

FIG. 2. Optical system of processing control sensitometer.

476

G. A. JOHNSON

Vol 47, No. 6

FIG. 3. Step tablet, ammeter, and control knobs of proc-
essing control sensitometer.

FIG. 4. Electrical wiring diagram of processing control sensitometer.

Dec. 1010

PROCESSING CONTROL SENSITOMETER

477

type. Their precision is satisfactory for controlling photographic
processing and for general sensitometry, but it is not good enough for
primary instruments in a sensitometric laboratory. The Processing
Control Sensitometer is of this class.

FIG. 5. Typical exposed and processed film strip.

A schematic drawing of the optical system is shown in Fig. 2, where
1 is an incandescent light source, and 2At 2B, and 2C are light baffles.
A pendulum disk shutter 3 is used as a timing device. No. 4 indicates
the filter holder or manual shutter. No. 5 is a light modulator con-

1.0

DENSITY OF BASE

.2.0

10"

RELATIVE LOG E

•— 1232 SUPER XX 7-0.57
•°-^l23l PLUS X tf-0.64

— 1 2 30 BACKGROUND X U- 0.66

FIG. 6. Characteristic curves for motion picture nega-
tive films exposed with the processing control sensitometer.
Film developed in Kodak SD-21 at 65 F; processing con-
trol sensitometer negative setup.

sisting of a photographic step tablet, 6 is the film strip 'to be exposed,
and 7 the film cover door.

As shown in Fig. 1, the lamp house containing a 10-v, 7.5-amp
photocell exciter lamp is at the back of the sensitometer. Directly
in front of it is the pendulum timing mechanism, which must be level

478

G. A. JOHNSON

Vol 47, No. 6

to operate properly. Leveling of the timing mechanism is accom-
plished by rotating thumbscrews attached to the feet until the bubbles
of the two crossed levels are centered. By turning the crank clock-
wise the pendulum is prepositioned on the releasing latch, which when
tripped allows the pendulum to swing through about 260 deg. A 60-
deg opening in the pendulum disk permits the light to pass for Vio sec.
At the end of its swing the pendulum is caught, ready to be preposi-
tioned for the next exposure.

2.0

1.0

3.0

RELATIVE LOG E
• — • 1301 **I85 DEVELOPED IN KODAK D-16 AT 65°F
• — • 1302 *«2 30 " D-16 AT 65°F

«— * 1203 "8=0.58 DEVELOPED IN KODAK SD-21 AT 65°F

1365 TH.25

SD-21 AT 65°F
(EXPOSED 8X NORMAL)

FIG. 7. Characteristic curves for motion picture positive, master
positive, and duplicate negative films exposed with the processing
control sensitometer ; processing control sensitometer positive setup.

When the arrow on the filter control knob at the side is turned to the
P (positive) position, unfiltered light of 2850 K illuminates the film,
while with the knob in the TV (negative) position, a bluish Wratten
78 A A filter and a 0.60 neutral density filter are inserted in the light
beam to give approximate sunlight quality illumination. Film strips
to be exposed are placed inside the film cover door (Fig. 3) and against
the step tablet, which has 21 steps with a gradient of \/2. The rheo-
stat knob A turns on the exposing light and acts as a coarse adjust-

Dec. 1946

PROCESSING CONTROL SENSITOMETER

479

ment; rheostat knob B is for fine control. The ammeter is illumi-
nated by green light when the light switch button near the levels is
pressed. The lamp is operated at approximately 6.5 amp to give
2850 K, and the needle deflection may be watched through the scale
window. Fluctuations in line voltage are reduced by the use of a
voltage regulator. The electrical wiring diagram of the transformer,
rheostats, and lamps is shown in Fig. 4. A list of these electrical parts
for the instrument is given in the Appendix.

3.0

2.0

1.0

~ 1357

1373

RELATIVE LOG
3-2.40 DEVELOPED IN KODAK D-16 AT 65* F

8' 3.08
B-0.40

D-16 AT 65 F
SO-21 AT65*F

FIG. 8. Characteristic curves for motion picture sound
recording films exposed with the processing control sensi-
tometer; processing control sensitometer positive setup.

A reproduction of a typical exposed and processed strip is shown in
Fig. 5. The exposure area is 7/$ in. by 4x/4 in., each step being 5 mm
by 7/8 in. Strips exposed in this manner may be compared on an
illuminator to note differences or, if a densitometer is available, the
densities of the exposed samples may be plotted against the densities
of the step tablet. Fig. 6 presents typical characteristic curves for
three motion picture negative films exposed on the processing control
sensitometer to sunlight illumination for Vio sec.

480

G. A. JOHNSON

Vol 47, No. 6

Motion picture positive films, as well as master positive and dupli-
cate negative film, exposed on the same instrument to 2850 K radia-
tion for Vio sec, give characteristic curves shown in Fig. 7. Included
in this illustration are curves for Release Positive Film, type 1301;
Fine Grain Release Positive Film, type 1302; Fine Grain Duplicating
Positive Film, type 1365; and Fine Grain Panchromatic Duplicating
Film, type 1203. With the slower emulsions, it is sometimes neces-
sary to give multiple exposures; for example, with type 1365, an ex-
posure of eight times normal is required.

Characteristic curves for motion picture sound recording films,
types 1357, 1372, and 1373, are shown in Fig. 8. Samples of these

0 1.0 2.0 3.0 4.0

RELATIVE LOG E

• — • KODABROMIDE F-2 V\Q SEC EXP
• — • AZO F-2 10 SEC EXP

FIG. 9. Characteristic curves for photographic papers exposed
with the processing control sensitometer. Film developed in
Kodak D72 at 65 F; processing control sensitometer positive
setup.

films were exposed under the same conditions used for the positive
films.

Professional and commercial sheet films may also be exposed to
either sunlight or tungsten illumination by inserting an edge or a
corner of the sheet under the film cover door and setting the filter
control knob in the desired position.

Since photographic papers are much slower in speed, it is necessary
to use longer exposure times when exposing such materials. This is
accomplished by removing the filter from its holder and inserting a
sheet of metal or cardboard, which may be used as a manual shutter
for exposure times longer than 10 sec. Characteristic curves for

Dec. 1946 PROCESSING CONTROL SENSITOMETER 481

Kodabromide F-2 and Azo F-2 papers exposed with this sensitometer
appear in Fig. 9.

Since the intensity of the lamp changes with age and the lamp is
uncalibrated, an additional lamp is used as a reference standard or
monitor. After both lamps have been seasoned by burning for at
least 2 hr, periodic checks of the exposing lamp are made against
the monitor. In this way any drifts caused by the exposing light can
be detected. When the exposing light shows an appreciable drift,
the monitor should replace it, and a new monitor should be selected.

With reasonable care, the processing control sensitomer will give
highly reproducible results both for controlling processing conditions
and for making relative sensitometric film tests.

REFERENCES

1 MEES, C. E. K. : "The Theory of the Photographic Process," Macmillan Co.
(New York), 1942; Chap. XVI, p. 587.

2 JONES, L. A.: "A Motion Picture Laboratory Sensitometer," /. Soc. Mot.
Pict. Eng., 17, 4 (Oct. 1931), p. 536.

Appendix

Electrical Parts List

Index
No. Description

1 Lamp, Mazda Photocell Exciter, 10 v, 7.5 amp, single contact, pre-

focus flanged base (for exposing).

2 * Ammeter, 0-10 amp, with illumination feature, a-c Model, 744,

Weston Electrical Instrument Corporation.

3 Resistor, adjustable, 25 ohm, 25 w, Cat. No. 0365, Ohmite Manu-

facturing Company.

4 Lamp, 6.2 v, miniature screw base (for ammeter scale) ; and Socket,

Cat. No. CH109, Drake Manufacturing Company.

5 Switch, double-throw contacts, Cat. No. BZRO-1, Microswitch

Corporation.

6 Transformer, filament, 110 v primary, 10 v, 8 amp secondary, Cat.

No. T-19E96, Thordarson Electrical Manufacturing Company.

7 Rheostat, 15 ohm, 25 w, Model H, Cat. No. 0146, with Cat. No.

5129 Knob, Ohmite Manufacturing Company.
8 A Rheostat, 200 ohm, 100 w, Model K, Cat. No. 045-2, with Cat. No.

5129 Knob, Ohmite Manufacturing Company.
8B Switch, toggle, single pole, single throw, 3 amp, Cat. No. 354,

Ohmite Manufacturing Company (attached to and actuated by

200-ohm rheostat).

9 Cord, Type S, 2-conductor, No. 16 awg.

10 Plug cap, parallel blade, 250 v, 10 amp, Cat. No. 70124, Cutler

Hammer, Inc.

TELEVISION AND THE MOTION PICTURE THEATER*
LESTER B. ISAAC**

Summary. — The problems of television entertainment in motion picture theaters
are discussed from the exhibitors' standpoint, including present equipment available,
location in the theater, scheduling of programs, and the economic considerations. The
author wishes to emphasize that his comments in no way reflect the opinions of Loew's
Incorporated or any of its officers or executives. He offers the following paper as his
Personal thoughts on a widely discussed subject.

After much talk and probably some concern on the part of motion
picture exhibitors as to how television will affect their theaters, I
have come to the conclusion that it is about time that something was
said to alleviate their fears.

Merely as a novelty, television cannot bring patrons to the box
office. It must be entertainment comparable to that to which they
have been so long accustomed. For the past several years, I have
made a careful investigation and study of the television situation.
It is my opinion that motion pictures are here to stay, and will enjoy
their universal popularity for many, many years to come. Yet there
are many in the television field who will not agree witn me and, to
the contrary, even make claims that the new art will eventually re-
place the motion picture.

I think, however, it may be acknowledged that of all the claims that
have been made to date regarding the practical possibility of tele-
vision as a form of public entertainment, not one has developed as an
accomplished fact. This would seem to indicate that, while making
full allowance for the need for progress and vision, certain elements of
the television field are day-dreaming and are much confused as to the
future of television in the theater. The motion picture industry has
had long practical experience in the entertainment field and I am sure
those who have carefully studied this situation are not greatly
alarmed. I do hope, however, that this discussion may offer some

* Presented, May 10, 1946, at the Technical Conference in New York.
** Director of Projection and Sound, Loew's Incorporated, New York.

482

TELEVISION AND THEATERS 483

mental relief to those exhibitors who at the present time have fears
regarding the many complications and heavy expense which will be
entailed through the installation of television in their theaters.

Theory is necessary in engineering and medicine, but the exhibitor
cannot use the public as guinea pigs. When the product reaches the
theater, it must have very definite box-office possibilities. On an
average it must be profitable entertainment, for after all the real boss
of the motion picture industry is the man, woman, or child who lays
down the price of admission at the box office. Even with such a
phenomenon as television, we must be farsighted and practical. In
my opinion, what Barnum said about the credibility of the American
public is no longer true today. The public shops for the best enter-
tainment and for the most comfort.

First, let us assume that television projection room equipment is
now available for theater use. How could such programs be handled
in a practical manner? Particularly those on-the-spot pickups?
Certainly it would not be practical or desirable to break into our
regularly scheduled motion picture film programs. For example,
suppose the theater screening of a main feature is scheduled for 12
noon. At 12:50 P.M. you receive a phone call that a fire or flood is
taking place in one of our large cities, and this will be telecast at 12 :55
P.M. Now in order to bring this television program to your audience,
the feature picture must be stopped, after running for 55 min, with 35
mih more for cqmpletion. Certainly no one, by any stretch of the
imagination, can believe that theater patrons will cheerfully put up
with any such interruption of the show.

It is no doubt true, of course, that when televison programs of
national interest are prepared in advance, a schedule could be
arranged to meet the requirements of such telecast. But this telecast
must be arranged at a date which will allow the theaters to publicize
the fact to their patrons. If, however, the telecast program is of wide
national interest, why should the public go to the theater and pay
admission, when such a telecast might be seen on their own home re-
ceivers? There has been wide talk about the likelihood of using
television in the theaters at some comparatively immediate date.
Therefore, I think we must now seriously consider the fact that at the
present time, so far as I know, there is no television projector available
for practical use in motion picture theaters.

I want to emphasize the words "practical use," as this will be dis-
cussed later in this paper. I concede that there are at least two so-

484 L. B. ISAAC Vol 47, No. 6

called theater television projectors — there may be more — but I have
seen only two demonstrated so far. The first projector was demon-
strated at the old New Yorker Theater in New York City several
years ago. The second was demonstrated to me on April 5, 1946, in
Schenectady. I can merely give my impression of these two demon-
strations.

The demonstration at the New Yorker Theater several years ago
did not impress me favorably because of poor illumination and only
fair definition. The main portion of the program was a prize fight
being held at Madison Square Garden. This telecast was not
through the ether, but was brought to the projector by tuned tele-
phone wire circuits, not coaxial cable. The television projector was
located in the loge section of the balcony, at a distance of approxi-
mately 70 ft from the screen, and projected an image of about 15 X
20 ft. I am given to understand that this projector is limited to a
projection distance of from 70 to 80 ft.

In the Fall of 1945, I had an opportunity to examine this television
projector again. Although the television projector was not demon-
strated at that time, it was my conclusion that no obvious improve-
ments had been made upon it since it was demonstrated at the New
Yorker Theater.

Because of the design and bulk of this projector, it could not be in-
stalled in any theater projection room, even if it were possible to
project the image at greater distances than that for which this par-
ticular model was designed. This will be discussed later in the paper.

The demonstration of the second television projector was given in
the very small auditorium of the Schenectady Civic Playhouse. The
program consisted of motion picture film and live talent, which was
picked up by the television cameras and fed by coaxial cable through
a special low-power microwave frequency modulation transmitter
located on a tower near the studio. The transmitter output was
beamed by a directional transmitter antenna toward the theater,
where it was picked up by another directional antenna. It was then
energized by way of coaxial cable through a special frequency modu-
lation picture receiver, which fed the picture to the projector.
Sound was transmitted from the studio by means of radio link.

The size of the projected image was 11 X 15 ft, at a projection
distance of 30 ft. The light and definition from this projector are
superior to that of any others I have witnessed, although there is still
room for considerable improvement before it can be used for practical,

Dec. 1946 TELEVISION AND THEATERS 485

profitable entertainment in motion picture theaters. I have been in-
formed that 30 ft is the maximum projection distance for this projec-
tor. Projection distances in present-day theaters are from 60 to 210
ft, but no average is available.

The two television projectors as herein outlined are known as the
"instantaneous system." When a program is received from its
source, it is instantly projected on the screen. There is another type
known as the "storage system." The storage system takes the tele-
vision program from its source, either off the air or through wire, and
records it upon a supersensitive 35-mm film which is processed in from
2 to 5 min, and then projected through standard 35-mm motion pic-
ture projectors. This latter system seems practical, but I do not
know how it would look on the screen, as I have never had the pleasure
of witnessing a demonstration. Granted, however, that it seems
practical from an equipment point of view, why is it necessary, and
what obvious advantage has it? It seems to me to be just another
way of presenting a newsreel. We certainly could not present this
system to audiences as true television, and I should like to ask what is
wrong with our present-day newsreel. I think a splendid job is being
done in bringing visually and orally news subjects of great current
interest to the public. Certainly a bang-up job was done in World
War II.

Now let us consider the practicability of the two television projec-
tors which I have seen demonstrated. Both have very limited pro-
jection distances, and just where could they be installed in present-
day theaters? Engineers and manufacturers have made many
statements regarding the installation of television projectors in
theaters. Therefore, I will first give the claims as made, and then
give the negative side.

Claim: Because of limited projection distance, the television projector could
be installed on the rear of the stage and the image projected through a translucent
screen.

Negative: Most theaters throughout the United States have only sufficient
room for the present horn system. If there were sufficient room back of the
screen, what would happen to the motion picture screen and horn system? Most
theaters are not equipped to permit flying of the screen, let alone the horns.

In those theaters which may be equipped to fly the screen and move the regular
horns off stage to permit rear projection television, is there a definite idea as to the
added cost of labor required? The great majority of theaters cannot stand the
additional expense.

486 L. B. ISAAC

Claim: A television projector could be installed in the orchestra pit.
Negative: Most theaters do not have orchestra pits.

Claim: A television projector could be installed in the basement, just in front
of present picture screen, using either a regular mirror or periscope system to
project the image on the regular screen.

'Negative: Most theaters do not have a basement in the screen area. Base-
ments are usually under the lobby and rear of the auditorium for heating and
ventilating units and for use of stores which may be a part of the project.

There may be other theater television projectors on paper, but
whether on paper or not if television is to become a part of the motion
picture theater, the design and manufacture of television theater pro-
jectors must meet the requirements of practical, profitable theater
operation.

Theater television projectors should be designed and manufactured
in such a manner that they could be installed including the receiver
and all its controls, in the present theater projection room beside the
motion picture projectors. The present film amplifiers and loud-
speaker equipment should be utilized for the oral part of television
programs.

This type of design and installation is a must, otherwise any attempt
to install the equipment in places other than those mentioned here
will cause jurisdictional labor disputes. Such disputes would put
the exhibitor right in the middle and, further, make the cost of
operation so great that it would be much more economical to install
live stage shows.

In closing, I repeat what I said at the beginning of this paper : I
fail to see the box-office value of television in conjunction with motion
picture theater operation unless it may be the televising of coming
feature trailers to homes. Of course, nothing is impossible; but
things that are possible can often be highly impractical. The two
opinions I have indicated in this paper are that the show must go on,
and the show must pay. All that I have seen or heard about tele-
vision for motion picture entertainment makes me believe that it is
by no means ready for general use in motion picture theaters. And
it is unfair to halt progress in many other directions while waiting for ,
that extremely indefinite period when the overwhelming number, not
just a few theaters, can afford to install television.

TECHNICAL PROBLEMS OF FILM PRODUCTION FOR THE
NAVY'S SPECIAL TRAINING DEVICES*

H. S. MONROE**

Summary. — The production of films for the Navy's Special Training Devices
involved a large number of special problems peculiar to this work, in addition to all
the usual problems of film production. These special problems were caused by the
urgency of war, the conditions under which the films were used, the necessity for
maximum realism, and the need to provide for assessing the student's work.

Since many of the devices around which these problems revolve are
still in process of development, this will be in no sense a description of
the devices, or of the films which are used with them. I will attempt,
simply, to sketch out a few of the technical motion picture problems
which were encountered. Some of these were peculiar to this work
and all of them were encountered in what might be called a more
virulent form.

First were the problems caused directly by the fact that the devices
were for war training and were developed and produced under the
pressure of a war emergency. Any new device believed to have
merit for training was wanted immediately. If the device required
film, it was important that the film should be ready at the same time
as the device. With the very special requirements often involved for
the film, this meant starting production while the device was still in
the blueprint or even the early drawing-board stage.

As a result, the film was always a lap or two behind the changes
being made in the device, and it was frequently necessary to scrap all
we had done and start over with the deadline only a few weeks away.
Sometimes the film was completed only to discover, when it was tried
in the prototype device, that some unforeseen condition, or unreported
change, had made it entirely unsuitable. Everything, then, had to
be held up while a new film was made under the greatest conceivable
pressure.

* Presented May 10, 1946, at the Technical Conference in New York.
** Formerly, Atlas Educational Film Company, Oak Park, III.

487

488 H. S. MONROE Vol 47, No. 6

Another result of the war urgency was that little time could be
spared for experimentation to -determine the best method of obtaining
the desired results. It was usually necessary to proceed along a line
that looked promising, without trying others that might prove equally
so. If the adopted method worked out, so much time was saved
that it was considered worth the risk of having to scrap all that work
and start over. In some cases, as in equipment manufacture, more
than one method might be tried concurrently, and the most successful
be finally adopted.

Another group of problems arose from the conditions under which
the devices were to be used. A device such as the Waller trainer1
does a magnificent training job where there is sufficient space and
permanence for its installation; but the devices which produced our
headaches were for installation on shipboard and at advanced bases.
This meant that a lot of them were needed and, where we could use
standard projectors, they were 16 mm because we could get more of
them in a hurry.

To get sufficient movement of the gun or turret in tracking plane
images on the screen, a large screen was necessary, and the enlarge-
ment from 16-mm film was of the order of 380 diameters. ' Even using
fine-grain release positive for prints gave a noticeable loss of sharpness
when the screen was viewed from the gunner's position only ten feet
away.

When a 15-mm wide angle projection lens had to be used to save
space, the result left much to be desired. This gave us no leeway at
all for any loss of sharpness in any of the steps through which the pic-
tures passed during production. Also, using such a wide-angle lens
for projection on a flat screen introduced a distortion. This compli-
cated the problems of definition and of scoring, of which I shall say
more later.

Devices installed on shipboard, at advanced bases, and even at
some training stations within our own borders, encountered climatic
conditions which certainly did not prolong print life. This problem
was further complicated by the difficulty of giving projectors ade-
quate maintenance, and by the necessity of using many inexperienced
operators.

Many methods of print protection were tried, but even so, prints
were often returned with the emulsion scraped completely off of large
areas, and whole rows of sprocket holes pulled out. Because of
problems of synchronization, tp be discussed later, every effort was

Dec. 1946 FlLM PROBLEMS FOR NAVY DEVICES 489

made to keep them going as long as possible since such films had to be
returned to the distribution center for repair. However, no film pro-
jected continuously all day long, day after day, through a gate infre-
quently cleaned, at high temperature and humidity, will last indefi-
nitely.

The second major class of problems arose out of the nature of the
devices themselves. One of their main purposes was to reproduce, as
closely as possible, the conditions which the student would meet in
actual combat, and to keep him alive to learn from his mistakes.
This made important the greatest possible realism in all the films,
both as to sight and to sound.

Every effort was made to give the student the impression that he
was actually aboard a ship or plane. Where the conditions under
which the pictures were made precluded the showing of actual parts
in the foreground, these were introduced by traveling mattes. Since
the films were so highly magnified in use, the mattes had to be
aligned with great accuracy so as to avoid any suggestion of matte
line which would give the show away.

Special atmospheric conditions which the student might encounter
in the course of his duty also had to be reproduced. Often the right
conditions could not be obtained at the time the pictures were made
and, in at least one case, the required conditions were changed after
the pictures had been taken. Since it was impossible to arrange for
reshooting, it was necessary to produce the effects in the laboratory.
Any effects man will recognize the difficulties of producing an authen-
tic picture of a ship gradually appearing as you approach it through
the haze, from a negative made on a clear sunny day!

Realism was further enhanced by sound tracks carrying the sounds
which the student might expect to hear in actual combat. For the
most efficient training, the movements of the target had to be pre-
determined ; consequently actual battle photography could seldom be
used. Whether the scenes were staged for live photography or shot
from models, authentic sound could not be recorded at the same time.

The problem then, was either to cut recordings of actual battle
sound to synchronize with what was shown on the screen or to produce
a synthetic track which could be mixed with a general background of
battle noise to make the final track. There was also the problem of
reproducing the sound of gunfire at realistic volume levels with the
equipment available, but in this case we just did the best we could and
Jet it go at that.

490 H. S. MONROE Vol 47, No. 6

In order to be effective, a training device must also evaluate the
work of the student and inform him of his progress. This was the
most serious source of problems, and the cause of those most peculiar
to this special type of film. In the case of targets having rapid motion
relative to the gunner, always present when aircraft are involved, the
most important aspect was determining the correct point of aim.
Where the film was produced by live photography, it required careful
assessment to determine from it this relative movement, and ballistic
calculations to compute the correct lead and range correction.

In addition to the difficulties inherent in this process, we had to face
the following facts : No pilot, under actual flying conditions, can fly
a theoretically perfect attack. Such an attack, however, is far better
for teaching the principles of correct aim than one containing the
vagaries of actual flight. Consequently we made many of the films
by mechanical animation, using models.

In this method, the movement of the student's position, in either
plane or ship, and of the target, are calculated and then translated
into the movement of the target relative to the student's position as it
will appear on the screen. At the same time, the correct point of aim
for the target position of each frame is computed.

The point of aim may be indicated on another film which will be
operated in synchronism with the target film, or on the target film
itself. In the first case, the usual arrangement was two standard 16-
mm projectors with their film propelling mechanisms geared together
through a connecting shaft. The sound tracks of the two films were
used simply to increase the realism of the student's situation by repro-
ducing the sounds of the plane in which he was supposed to be flying
and of his own gun when he fired it, or other sound appropriate to the
particular type of trainer involved.

To make the scoring at all accurate, the images shown by the two
projectors had to be aligned exactly. Not only did individual projec-
tors vary as to their optical alignment, but it soon became apparent
that it was practically impossible to make 16-mm prints in quantity
which would show the center of the image at exactly the same spot.
We found it necessary, therefore, to place at the beginning of each
film a fairly long strip of alignment leader having a pattern by which
the two images could be exactly aligned for that particular installation
and that pair of films.

All steps leading up to the master negative from which the release
prints were made had to be performed with the greatest precision.

Dec. 1946 FlLM PROBLEMS FOR NAVY DEVICES 491

We even obtained special length rolls of IG-min stock so that the
whole film, including the alignment leader, could be printed on one
piece. These precautions were to ensure that alignment, once estab-
lished, would be preserved throughout.

Training activities using the films were instructed to keep pairs to-
gether and, in case one film required repair, to return both to the
distribution center rather than attempt to repair it in the field. The
loss of a frame or two in splicing would place the remainder of the film
out of synchronism, and field activities had neither the equipment nor
the experience to correct this.

Problems of synchronism were avoided when the scoring control
was placed on the same film as the target, but other problems were
introduced. In some of the single film trainers the sound track was
used as the scoring control. This might be done by having the track
carry a tone whose frequency was a function of a quantity involved
in the student's problem. If this track were to be made by the use of
a variable frequency oscillator, the problem arose of trying to adjust
the frequency with sufficient accuracy while the film was running
through the recorder at 90 ft a min.

Experiments were made with a method of producing the track
photographically, but that method had a lot of problems all its own,
and no entirely satisfactory method of producing this type of control
track had been developed up to the close of the war.

Another type was an unmodulated track having short "blips" of
fixed frequency which, when amplified, would operate a scoring
mechanism through a suitable relay. Because of the uniform length
of the blips, and the sharp attack and cutoff required, it was found
most convenient to produce such tracks photographically on a modi-
fied animation stand. The final prints in both the cases I have
described were made in the usual way from separate sound and picture
negatives, so the problems at that stage were simply the usual ones
encountered by any laboratory.

Another type of film, however, dispensed with the sound track
entirely and used the space it would ordinarily occupy on the film,
together with the space between frames, for signals to indicate the
correct point of aim. These had to be aligned with the picture within
tolerances of two thousandths of an inch, so they had to be on the
same negative.

Since the scoring mechanism employed photocells, accurate scoring
required the greatest possible difference in the light hitting the cell

492 H. S. MONROE Vol 47, No. 6

between the correct aim position and any other. This meant high
contrast in the scoring area of the film, and was analogous to the situa-
tion in a variable-area sound track. But in that case the track is
printed from a separate film, often having a different emulsion
specially designed for the purpose and almost universally given
different development. In these films, on the other hand, it was
necessary to produce the equivalent of a variable-area track on the
same film as the picture negative, without degrading contrast in the
scoring signals and without producing a "soot-and- white wash" pic-
ture.

The history of this work, throughout the war, was a seesaw of im-
proving control at the expense of picture, then picture at the expense
of control, the switch being made in each case when the neglected part
began to give trouble. Some research was begun on methods using
color filters to control the contrast of the two parts of the film.
Means of increasing the inherent contrast of the scoring signals which
were photographed were also sought, but the problem cannot be said
to have been entirely solved when hostilities ceased.

Even if it had been possible to make an ideal scoring signal with a
picture that still looked like something, machines would nevertheless
vary as to the score they gave a student because of variation in the
sensitivity of the photocell and its amplifier, in the width of the mask,
in the voltage of the power supply, and in the adjustment of the
trainer.

To a certain extent this was overcome by developing a leader which
projected a stationary target that the rawest novice could hit. Thus
eliminating the student's error, we were able to measure the variation
in machines and films. The scoring signals on the film, instead of re-
maining stationary at the point of correct aim, moved past it at a
uniform rate. The score thus obtained enabled a correction factor to
be applied to the scores subsequently made on that machine, in that
location, with that film, to make them comparable with scores ob-
tained in other situations.

In addition to all these problems, which were more or less inherent
in the training situation, others were introduced by the special design
of certain devices. These ranged from such comparatively simple
things as frames half the standard height, through extreme wide
angle projection from standard film onto a hemispherical screen, to
the use of polarizing emulsions to make an image, or part of one,
appear to disappear at will. Throughout, there was a gradual in-

Dec. 1946 FlLM PROBLEMS FOR NAVY DEVICES 493

crease in the use of color, just as there has been in the entertainment
and commercial fields, and that simply made the other problems the
more difficult of solution.

Although I have been able only to catalog the major problems we
encountered, it must be clear that the Special Devices Division's Film
Section in the Navy's Office of Research and Inventions was required
to be far more than a routine procurement agency.

REFERENCE

1 WALLER, F.: "The Waller Flexible Gunnery Trainer," /. Soc. Mot. Pict.
Eng., 47, 1 (July 1946), p. 73.

AN IMPROVED 200-MIL PUSH-PULL DENSITY
MODULATOR*

J. G. FRAYNE,** T. B. CUNNINGHAM,** AND V. PAGLIARULO**

Summary. — A completely new variable-density modulator utilizing a three
ribbon push-pull valve is described. The entire valve is sealed by the force of the Al-
nico V permanent magnet on the Permendur pole pieces. Signal is applied to the
center ribbon and noise-reduction currents are applied to the outer ribbons. True
class A push-pull operation is obtained from the two component single ribbon valves
by the use of an inverter prism which aligns the modulating and noise-reduction edges
of each aperture.

An anamorphote condenser lens is used to eliminate lamp filament striations at
the valve ribbon plane. An anamorphote objective lens gives a 4:1 reduction of the
valve aperture in the vertical plane at the film and a 2:1 reduction along the length of
the sound track. A meter is supplied to measure exposure as well as setting up
"bias." A photocell monitor is supplied and a "blooping" light for indicating
synchronous start marks.

Mathematical analysis of the exposure produced by the modulating ribbon is ap-
pended as well as a similar analysis of the four ribbon push-pull valve which the
new valve supersedes.

Introduction. — The Western Electric push-pull variable-density
system utilizing the R A -1061 type light valve has been in wide use in
sound picture recording for a number of years. The methods cur-
rently employed are essentially those described by Frayne and Silent.1
The original push-pull density modulator was built around the RA-
1061 light valve, which is a four ribbon structure, each pair of ribbons
denning a variable slit which is focused on the sound negative. The
sound currents actuate the ribbon pairs in opposite phase, one pair
closing while the other is opened. At the same time, superimposed
noise reduction currents are applied in phase to each pair of ribbons.
The cancellation' in reproduction of the in-phase noise reduction
frequencies has been one of the principal advantages of the push-pull
system.

* Presented May 9, 1945, at the Technical Conference in New York.
** Electrical Research Products Division, Western Electric Company, Holly-
wood.

494

PUSH-PULL DENSITY MODULATOR 495

The original push-pull valve was superimposed on a modulator
previously designed for recording a standard 100-mil density track.
Some improvements, however, were introduced in the optical system
in order to reduce the distortion attributed to the so-called "ribbon
velocity" effect.2 For example, a 4: 1 optical reduction of the one-mil
valve ribbon slit was attained by the use of a special 4 : 1 objective lens
which replaced the KS-7325 2:1 objective lens previously employed.
With this objective, a 100-mil push-pull density track was obtained.
Later, the demand grew for a high-quality 200-mil push-pull density
track, and this was obtained by mounting a small cylinder lens adja-
cent to the film working in conjunction with the standard 2:1 objec-
tive lens. With either arrangement, the width of the image was
effectively 0.25 mil.

MODULATING RIBBON-

FIG. 1. Illustrating push-pull modulation by single
ribbon.

The modulator described in this paper represents a completely in-
tegrated design retaining the well-proved principles of the light valve
and incorporating new and improved designs in every important
component of the complete modulator. Before finally deciding on
retaining the light valve as the basic light modulating device, much
time and effort were spent in searching for other devices and methods
which would give the same excellence of performance as had been
attained under 'operating conditions with the light valve. Since none
of the devices or methods studied gave comparable performance
when viewed from every possible standpoint, it was decided to retain
the light valve as the basic modulating device in the new modulator.

Design Principles of Three Ribbon Valve. — Since the light valve
is the most vital component in the modulator, the principles underly-
ing the performance of the new valve will first be described. The
guiding principles behind the design were simplification of manufac-
ture, ease of adjustment, stability of operation, and fidelity of re-

496 FRAYNE, CUNNINGHAM, AND PAGLIARULO Vol 47, No. 6

sponse. In line with simplification, a very thorough study was made
of the number of ribbons required to lay down a push-pull density
track. It will be recalled that the RA-1061 valve required the
accurate alignment and almost identical tuning of four separate
ribbons. A simple study showed that two push-pull tracks could be
made with a single ribbon, as shown in Fig. 1, the apertures A and B
being modulated in a push-pull manner by the movement of the
ribbon. In order to obtain noise reduction with a single modulating
ribbon, ribbons 1 and 3 of Fig. 2, which serve to define the fixed edges
of the apertures, may be made to move in accordance with the im-
pressed noise reduction currents. Thus, a total of three ribbons, one
for signal and two for noise reduction, is all that is necessary for push-
pull modulation of the light transmitted by the two apertures.

SPEECH

FIG. 2. Speech and noise-reduction connections for a three ribbon push-pull

valve.

The electrical connections for speech and noise reduction are shown
in Fig. 2. Ribbons 2 and 3 are effectively shunted by the 2-ohm
potentiometer which is used to correct for any variation in sensitivity
of either noise reduction ribbon, the bias currents in each ribbon being
adjusted in order to obtain equal biased spacing for each aperture.
It will be noted that the noise reduction and speech currents are
isolated electrically. This has several practical advantages. First,
it eliminates the heating effect which would otherwise be introduced
in the speech ribbon by the bias currents, thereby stabilizing its tuning
frequency. It permits the direct connection of the speech ribbon to
the secondary of the light valve transformer, instead of working
through a simplex circuit, as has been customary. This permits a
lower impedance light valve circuit, which is very effective in reducing
the height of the resonance peak. While the central ribbon is tuned

Dec. 1946 PUSH-PULL DENSITY MODULATOR 497

to a suitable high frequency the separate noise reduction ribbons may
be tuned considerably lower, thus reducing the amount of bias current
required and at the same time reducing the heating effect, with conse-
quent improved stabilization of their tuning and positioning.

It is obvious from Fig. 1 that the two apertures of the push-pull
valve are offset physically and that some means must be provided to
make their images collinear at the film plane. The offset apertures
in the RA-1061 valve are brought into line by means of refractor
plates, commonly known as "sawbucks." In the latter valve, the
alignment presents no great problem since the center line of the
images remains aligned irrespective of the amount of bias applied to
the ribbons. In the three ribbon valve, however, the problem is more
difficult in view of the fact that
the bias is applied to an opposite

I 7 1 MOD EDGE

edge of each aperture. Thus, if «>) L

MOD EDGE

the apertures are aligned for the N R EDGE

unmodulated unbiased condition,

they will not be aligned when any

bias current is applied to the "RE°GES "OP. EDGES

noise reduction ribbons. There M00. EDGES N. RADGES

are three possible alignments

which will be referred to as Cases

1, 2, and 3. N R EDGES

Case 1 is shown in Fig. 3a. It (c) [^ A — j — B ^
will be observed that the modu- MOD. EDGES

lated edges are in alignment for FIG. 3. Illustrating different align-
the unmodulated, unbiased con- ^ of aPertures in three ribbon
dition. The noise-reduction edges
are not aligned for any condition except that of 100 per cent closure.

Case 2 is shown in Fig. 3b, in which the noise reduction edge of one
aperture is aligned with the modulating edge of the second aperture.
The apertures will remain aligned only for the unbiased condition, be-
coming more and more out of alignment as the bias current is in-
creased.

Case 3 is shown in Fig. 3c and is obtained from Fig. 3b by the
optical inversion of Image B. This brings the noise reduction and
modulating edges, respectively, in line and the images of the apertures
will stay in alignment at the film plane irrespective of the amount of
bias applied to the ribbons.

The inversion of Image B is obtained by passing the light from the

498 FRAYNE, CUNNINGHAM, AND PAGLIARULO Vol 47, No. 6

corresponding aperture through an inverting prism of the type
shown in Fig. 4. The passage of the light from aperture B through
the glass prism is compensated by the addition of a rectangular prism,
generally known as the compensator, in the beam from aperture A.
The images of the two apertures are brought together at the film
plane simply by tilting the compensator. The method of mounting
and adjusting the prisms is discussed in detail later in the paper.

Distortion in Three Ribbon Valve. — The three ribbon push-pull
valve consists essentially of two single ribbon valves with noise reduc-
tion applied in each case to the outer ribbon defining the aperture.
The recording exposure characteristics of single and double ribbon
valves have previously been described.3 As a result, it is well known
that the distortion produced by the so-called ribbon velocity effect is
more pronounced in single than in double ribbon valves. For this
reason, the two ribbon valve has been generally favored in light valve
sound recording systems and the RA-1061 push-pull valve referred to

FIG. 4. Action of inverting prism on beam from one aperture.

above utilized a pair of such light valves to obtain a push-pull sound
track. The question arises, therefore, as to what distortion may be
introduced by using two single ribbon component valves in recording
a push-pull sound track. The three methods of aligning the apertures
shown in Fig. 3 have been analyzed mathematically. Only the con-
clusions will be referred to here, since the complete analysis appears
in the Appendix. Thus, for Fig. 3a the characteristic obtained from
ideal push-pull reproduction of the film recorded in this manner is as
follows :

Output - *[,, Ml sin „ (fc + 2) - I J, <*£> sin 2. (fc + ?

It will be noted that, in addition to the fundamental, both odd and
even harmonics are present. The presence of the even harmonics
indicates that the alignment of Fig. 3a does not permit true push-pull
reproduction from the track recorded in this manner.

Dec. 1946 PUSH-PULL DENSITY MODULATOR 499

The characteristic resulting from Fig. 3b is given by

4[-_ («6) oxr . / , a\ 1. /(2«6)\
Output = - \ Ji - — cos ^ sin co ( /0 -f 5- ) - 9 /2 ( -

<o|_ » 2t> \ 2v/ 2 \ v /

2w6 . / . a\ , 1 ,

cos -^ - sm 2a, (fa + 2J + 3 7, — cos —

Here again it will be observed that there is no cancellation of the even
harmonic components introduced by the ribbon velocity effect. This
condition results in both cases from the fact that the even harmonics
are recorded in the same phase as the fundamental, as will be noted
in the Appendix from Eqs (5) and (10) in Case 1 and Eq (14) in Case
2.

In the alignment of Fig. 3c, where the image of aperture B is in-
verted, the resulting push-pull characteristics is given by

Output -![/, M> sin. (h + 2) + 1,, Sf sin 3» (fc + |) + ...]•

The inversion of the image of aperture B results in complete cancella-
tion of the even harmonics since in this case these .components from
the two valve apertures are recorded in opposite phase relative to
their respective fundamentals, as indicated in Eqs (5) and (19). For
this reason, the alignment of Fig. 3c has been adopted in the present
three ribbon valve.

As a comparison with the performance of this valve, it is of interest
to note the characteristic of the present standard RA-1061 four ribbon
structure. It is given by the following equation :

r « . .

Output = E2 - E! = - \ Jj. -e - cos — sm w/o + o

3"' 2v

.
cos -z— sin

...]

It will be noted that even harmonics are absent in the track recorded
with this valve, but that the odd harmonics are present.

The theoretical exposure frequency characteristics obtained from
the mathematical expressions for Cases 3 and 4 are shown graphically
in Fig. 5. Because of the high second harmonic content of Cases 1
and 2, these valves have not been considered since a valve with such a
degree of second harmonic distortion would not be acceptable in
modern sound recording channels. The curves of Fig. 5 are calcu-
lated for an effective image width of 0.15 mil, which was determined

500

FRAYNE, CUNNINGHAM, AND PAGLIARULO

Vol 47, No. 6

from 60- to 7000-cycle inter modulation measurements.3 Although
the geometric width of the image is approximately 0.2 mil, the
effective width is considerably less, apparently because of uneven
illumination of the slit. Also, the theoretical harmonic content
calculated from Eq (23} of the Appendix on the basis of this image
width agrees very closely with the experimental values determined
from actual recordings. It should be pointed out that the harmonic
values shown in Fig. 4 are based on 100 per cent modulation of the
aperture. Since the light valve spacing only attains its nominal value
for the 100 per cent modulation condition, actual harmonic content
in a typical recording is considerably less since the action of the bias

I- FUNDAMENTAL 3 AND 4 RIBBON VALVE

2- 3£ST HARMONIC 3 RIBBON VALVE (CASE Bf)

3-3£4 HARMONIC A RIBBON VALVE (CASE ll)

FREQUENCY KILOCYCLES

FIG. 5. Frequency response and harmonic distortion of
three and four ribbon valve.

results in a smaller image width. The comparison of Cases 3 and 4
shows that the fundamental frequency response is the same for the
three and four ribbon valves, each being spaced to a nominal value of
one mil. The amplitude of the third harmonic is slightly greater for
the former. However, if we consider 9000 cycles as the upper limit of
reproduction, we are interested only in the third harmonics of funda-
mental frequencies lying below 3000 cycles. At these fundamental
frequencies, the curves of Fig. 4 show that the third harmonic in the
three ribbon valve is only one per cent of the value of the funda-
mental. With the second harmonic completely cancelled out, the
question of harmonic distortion in the exposure characteristic of this
light valve is academic. This is substantiated by experimental re-

Dec. 1946

PUSH-PULL DENSITY MODULATOR

501

cordings which agree very closely with the above theoretical data.
It should be pointed out, however, that either single component of the
push-pull track will show an appreciable second harmonic, since the
elimination of this harmonic is only attained in push-pull reproduc-
tion. The use of a single track in this modulator is, therefore, not
recommended for high-quality single track reproduction, but is a per-
fectly satisfactory medium for editing and other studio purposes.
Also, the current practice of using reverse bias on the ribbons is not
recommended with this valve, as any spacing wider than one mil leads
to increased distortion.

INDICIAL RESPONSE

FREQUENCY RESPONSE

(a)

+ 2
0
-2

-4
-6

]

—

___.

— -

—

\\\

-—

2 3456 789 2

X> FREQUENCY CPS 2000

(b)

tn-2
Q

-4

-6

1000

2 34567891 2

FREQUENCY C PS 20000

FIG. 6.

Dynamic Characteristics of the Valve. — In the design of the
RA-1238 valve, a marked reduction in the peak response as compared
to that of older type valves was considered imperative. It has been
well established that a source of some instability in existing light
valves is the relatively high amplitude oscillations induced at the
resonant frequency of the ribbons by the impact of sharp wave front
sounds. After careful consideration, the magnetic circuit discussed
elsewhere in the paper was selected. It was found that this design
resulted in a flux density in the air gap of around 32,000 gauss, and
this in turn produced a marked lowering of the resonance peak. The

502 FRAYNE, CUNNINGHAM, AND PAGLIARULO Vol 47, No. 6

use of a single ribbon of approximately 0.5-ohm resistance, instead of
the four ribbons of the RA-1061 valve, when associated with a corre-
spondingly low transformer secondary impedance, also was a contrib-
uting factor to lowering the height of the resonance peak. Further
damping was obtained by the use of an 0.5-ohm shunt directly across
the signal ribbon, the 3-db loss in power being more than compen-
sated for by increased efficiency of the valve.

The frequency response and the indicial or square wave response for
various conditions are shown in Fig. 6. Fig. 6a shows the valve
response with a 0.5-ohm shunt. An effective peak of about +3.5 db
is obtained while about two damped resonance oscillations result from

FIG. 7. Completely assembled three ribbon light valve modulator.

the application of a square wave to the light valve transformer.
When a constant impedance equalizer with an attenuation character-
istic conjugated to the resonance curve was inserted in the 600-ohm
side of the light valve transformer, the response shown in Fig. 6b was
obtained. A net rise of about 0.7 db was obtained under this condi-
tion and the oscillation under the impact of the square wave is reduced
to about one cycle. The use of this constant impedance equalizer has
therefore been standardized for use with the RA-1238 light valve.

An interesting feature of this valve which is not noticeable in older
designs is a pronounced shift of the resonance frequency by the high
damping. The valve used in making curves of Fig. 5 was tuned to

Dec. 1946 PUSH-PULL DENSITY MODULATOR 503

about 8300 cps in a weak magnetic field. When completely magne-
tized, the peak response shifts from this value to about 7400
cycles, or a shift of about 900 cycles. The true resonance point re-
mains at about 8500, and the valve will tend to oscillate weakly at this
frequency under a sharp impact. In order not to confuse the tech-
nician tuning the valve, the undamped peak frequency is referred to
as the tuning point rather than the frequency of 'maximum response
of the magnetized light valve

General Design of Modulator. — Fig. 7 is a photograph of the
completely assembled modulator. The lens assemblies, light valve,
lamp bracket, shutter,. deflector, exposure meter, and photocell optics,
all are solidly mounted on a vertically positioned base plate in the
modulator housing at right angles to the film plane. Each component
part is independently adjusted during manufacture and permanently
secured to the base plate. The entire optical assembly and valve can
be adjusted as a unit for final focus by sliding the base plate on engag-
ing studs located in the modulator housing. Provision is also made to
shift the plate parallel to the axis of the recording drum to locate the
sound track properly on the film.

The shutter operating solenoid, photocell monitor coupling unit,
and exposure meter photocell mesh are installed on the back of the
base plate, assuring positive alignment of all associated assemblies.
Connections are made to a remote photocell amplifier through a co-
axial cable. The electrical and mechanical lineup controls consisting
of lamp adjustments, variable lens stop, noise reduction balance
potentiometer, photocell balance potentiometer, and photocell
amplifier push-pull-standard switch, are conveniently located in the
modulator.

Optical System. — As indicated above in this paper, improvements
in the older 200-mil objective optical system had been obtained by
use of a small cylindrical lens mounted adjacent to the film and used
in conjunction with the 2 : 1 spherical objective lens, the combination
producing an effective image width of approximately 0.25 mil. This
system required the use of a weak positive cylindrical lens in conjunc-
tion with the spherical objective lens and oriented so that its axis is
"crossed" with the axis of the small cylindrical lens adjacent to the
film in order to bring the vertical septum and end masks of the light
valve into focus at the same plane as the ribbon image.

The new optical system schematic is shown in Fig. 8. The working
distance from lamp to film plane has been reduced from approximately

504

FRAYNE, CUNNINGHAM, AND PAGLIARULO

Vol 47, No. 6

Dec. 1946 PUSH-PULL DENSITY MODULATOR 505

13 in. in the old system to less than 11 in. in the new. This allows
for more compact and rigid design. Reading from right to left are
shown the RA-1236 cylinder lens and the new shorter focal length RA-
1235 spherical objective lens which is achromatized over the spectral
region to which the sound recording emulsion is most sensitive. The
two elements are combined in a single mount to form an anamorphotic
objective system. The septum and end masks are located behind the
ribbons and are focused in the same plane as the light valve ribbon
images. The inverting and compensating prisms are mounted in the
light valve in front of the ribbons. The next element appearing in
the system is the new RA-1237 aspheric cylindrical condenser lens
combination. This lens is designed to fill completely the objective
lens system and produce an image free of lamp coil striations at the
ribbon plane. This is accomplished by imaging the length of the
filament at the ribbon plane and imaging the vertical coil component
beyond the ribbons. An adjustable stop with a pair of blades
parallel to the ribbons is included in the condenser in order to provide
exposure control required by films of different speed and processing
variations. The last item in the system is the new recording lamp,
which is provided with a prefocused base. It is furnished in a smaller
envelope than previous lamps, and operates at a color temperature of
3150 K at a current of 7.5 amp.

The photocell monitoring deflector plate is placed between the light
valve and the objective lens. fms deflector transmits approximately
90 per cent of the total light to the objective system, the remainder
being directed to the monitor photocell. The optical system of the
photocell monitor follows previous practice in that the image of the
valve is focused at two separator lenses which divert each half of the
push-pull image to the appropriate photocell in the monitor amplifier.
A double reflecting shutter is also located in this portion of the objec-
tive path. In the closed position it blocks all light from the light
valve to the film but directs the image of a bloop lamp to the film for
the purpose of producing a synchronizing mark at the start of a re-
cording "take." The other side of the shutter in the closed position
deflects all of the light from the light valve into the exposure meter
photocell. The optical system produces an image of the light valve
at a collective lens which in turn transmits the light to the photocell.
An adjustable mask located at this image point provides means for
measuring the total light or each half of the image independently.

506

FRAYNE, CUNNINGHAM, AND PAGLIARULO Vol 47, No. 6

When the shutter is open for recording, neither the bloop lamp nor
the exposure meter is operative.

Light Valve. — The assembled light valve coded RA-1238 is
shown in Fig. 9. A disassembled view is shown in Fig. 10. The de-
sign follows the basic principles established by Wente4 and used in the
stereophonic recording system. Referring to Fig. 10, it will be seen
that the entire valve consists of three major parts: the permanent
magnet, the top pole piece, and the base pole piece assemblies.
These assemblies are located in opposite ends of the magnet by means
of concentric shoulders which fit into concentric recesses ground into
the ends of the magnet. Dowel pins in the end plates engage the
magnet and ensure proper orientation of the two pole pieces.

Fig. 11 shows a cross-sectional
^^^^m view through the magnet and pole

" pieces. A careful selection of

\ magnetic path, pole piece shape,

and materials was made in order
to ensure maximum flux density in
the air gap. Pole pieces were de-
signed to minimize leakage and to
conserve space without sacrificing
this requirement. These pole
pieces and end plates are con-
structed of Permendur, selected
largely because of its high flux
saturation. Alnico 5 was selected
for the magnet because of the
high magneto-motive force which

can be generated and retained in this type of material when used
in a permanently closed magnetic circuit. The valve is designed on
the basis of these factors and must be demagnetized and remagnetized
each time it is assembled, in order to retain maximum efficiency and
sensitivity. The design is such that in the assembled valve the pole
pieces are saturated well below the current used in the charging coil
during the magnetizing cycle. This results in an available magneto-
motive force considerably in excess of that required, which ensures uni-
form sensitivity in a large number of valves, assuming that all other
factors are held constant. The most important of these factors is the
air gap adjustment and this is held to a close tolerance during manu-
facture and assembly. Magnetic field strength tests reveal that the

FIG. 9. Assembled light valve.

Dec.. 1946 PUSH-PULL DENSITY MODULATOR 507

flux density reali/ed in the air gap averages about 32,000 gauss. This
is in marked contrast to the force of 10,000 to 1S,000 gauss obtained
in the RA-1061 valve. The benefits realized from these improve-
ments in design, are shown in Figs, Oa and (>b and were discussed earlier
in this paper.

Fig. 12 is a plan view of the light valve showing the arrangement of
the clamp carriages and ribbons. Each of the two groups of clamp
carriages is fastened solidly to the ends of the pole piece by means of
two relatively large insulated screws. Since the center of the vibrat-
ing span coincides very closely with those of the apertures, the caten-
ary effect or "bowing" of the ribbons is minimized. This, in turn,
results in a more uniform track density for the biased condition and

FIG. 10. Sections of disassembled light valve.

more uniform modulation across the track. The net result is a more
effective signal-to-noise ratio from a track recorded with this type of
valve.

Although the ribbon and pole pieces are constructed of different
materials, their temperature coefficients fall fairly close together and
the effects of temperature variations have been reduced to a very
small value. Variation in ribbon tuning is the most critical factor
affected by temperature changes in the light valve and it may be used
as a measure of the effect. Carefully controlled tests made over a
range of —40 to -4- 150 F produced a change of only 50 cycles in a tun-
ing frequency of 8500 cycles. The spacing and positioning of the
ribbons showed no measurable change over this range of temperature.

Benefits from the use of beryllium copper have been realized in the

508

FRAYNE, CUNNINGHAM, AND PAGLIARULO- Vol 47, No. 6

new clamp carriages shown in Fig. 13. The well-known physical
characteristics of this alloy5 have permitted the design of smaller and
lighter clamp carriages without any sacrifice of strength. This
material has additional advantages in this particular application be-
cause it is nonmagnetic, is highly resistant to corrosion, and provides a

FIG. 11. Cross-sectional view of valve.

relatively hard surface at the ribbon clamp. It is an excellent elec-
trical conductor, particularly after heat treatment. The mechanical
design of the carriage has been simplified over older designs to
facilitate manufacturing and assembly without loss in stability. The
slot and self-locked screw arrangement shown provides easy and
accurate ribbon height adjustment at assembly.

Dec. 1946

PUSH- PULL DENSITY MODULATOR

509

Silver contacts pressed into insulating bushings in the end plate and
connected to the carriages provide electrical connections to the ex-
ternal circuits. A 1/2-ohm shunt is permanently connected across
the speech ribbon to ensure maximum damping.

As shown in Fig. 11, the inverting and compensating prisms are in-
stalled in two small metal cylinders which are located immediately

FIG. 12. Illustrating clamp carriage and ribbon arrangement.

below the ribbon apertures. These cylinders are fitted into a hole
bored transversely through the pole piece so that the prism-supporting
cylinders lie parallel to the ribbons. Alignment of the prisms is
provided by the two opposing screws engaging the cylinders, causing
them to rotate and slightly tip the prisms. When adjustment is
completed, the screws are tightened to lock the cylinders in position.
Standard Valve. — For single track recording a new two ribbon

510

FRAYNE, CUNNINGHAM, AND PAGLIARULO Vol 47, No. 6

valve coded RA-270 will be available. It will mount in the modu-
lator in identically the same manner as the RA-1238, the correct
track position being obtained by moving the modulator in the manner
described above. All the mechanical advantages of the RA-1238 are
retained in this valve. The construction is, of course, simplified. by
virtue of use of one recording aperture.

ELEVATION ADJ

SPACING ADJ.

FIG. 13. Ribbon clamp carriage.

Exposure Meter. — The modulator includes a photoelectric ex-
posure meter which measures the total amount of light passed by the
valve, or by each component aperture. It thus provides a convenient
means of adjusting the light intensity to the proper value for the
particular film being used, and of balancing the light flux through each

wwv-

FIG. 14. Exposure meter schematic.

aperture. It may also be used for setting up noise reduction. The
essential parts of this device are a blue optical filter, a 929-type photo-
cell, direct-current amplifier, and current meter. The reflecting sur-
face of the shutter referred to above, when in the closed position,
diverts into the photocell all of the light which would normally fall on

Dec.

IVsii-PuLL DENSITY MODULATOR

511

tlu objective lens. The blue filter is interposed in the light beam
directly in front of the photocell. The photocell provides the input
voltage to a 6SJ7 pentode, which acts as a single stage direct-current
amplifier having considerable inverse feedback. The electrical
schematic is shown in Fig. 14. A microammeter connected in the
plate circuit of the pentode V2 indicates the change in plate current,
which is a measure of the quantity of light entering the photocell. A
balancing circuit cancels the zero signal plate current so that with no
light into the photocell the meter will read zero. A potentiometer
permits adjustment of this balancing current in order to compensate

-5

-10

FREQUENCY IN CYCLES PER SECOND

FIG. 15. Experimental film frequency characteristic of push-pull
modulator.

for slight drifts in plate current. A variable resistor shunting the
meter permits adjustment of sensitivity; however, the operator can
always return to a permanent calibration by throwing a switch which
disconnects this variable resistor and connects a calibrated shunt
across the meter. The meter has two scales, one a straight linear
scale, the other a decibel scale which is used primarily for setting noise
reduction. On this latter scale, full-scale reading of the meter is indi-
cated as zero db, half -scale as 6 db, quarter-scale as 12 db, etc.

The spectral sensitivity of the photocell in combination with the
blue filter results in an over-all spectral response of the exposure

512

FRAYNE, CUNNINGHAM, AND PAGLIARULO Vol 47, No. 6

meter which is very close to that of the film. This is indicated experi-
mentally by the fact that wide fluctuations in color temperature of the
recording lamp result in identical exposure of the film providing the
exposure meter reading is kept constant. There is sufficient inverse
feedback that wide variations in supply voltage produce negligible
changes in sensitivity.

Operating Characteristics. — The film frequency characteristics
of the modulator are shown in Fig. 15. These characteristics are
obtained on a print from a negative made with constant input modu-
lation of the single speech ribbon. The sensitivity of the speech
ribbon in the RA-1238 light valve for 100 per cent modulation at 1000

§.2

0.5

DENSITY

FIG. 16.

Push-pull inter modulation print density characteristic
of modulator.

cps requires a level of +4.5 db per 0.001 w including the power dissi-
pated in the built-in 1/2-ohm shunt.

The noise reduction ribbons which are tuned to approximately 5000
cycles require 0.080 amp for 0.001-in. excursion. Part of this current
is dissipated in the 2-ohm noise reduction balancing potentiometer.

Fig. 16 shows the intermodulation distortion in a print made from a
negative recorded by the RA-1238 light valve modulator. The inter-
modulation test procedure follows that originally proposed by Frayne
and Scoville,3 namely, the superposition of 1000 cycles on a 60-cycle
tone, the level of the 1000 cycles being 12 db below that of the lower
frequency. The combined wave forms were recorded at 80 per cent

Dec. 1946 PUSH-PULL DENSITY MODULATOR 513

peak modulation of the ribbon. The low value of 3 per cent indicates
that the modulator itself is essentially free of distortion, and the
broad curve indicates a wide choice of print density for this type of
push-pull recording.

The original model of the modulator and light valve were put in
production test at Sound Service Studios over a period from January
18, 1946 to March 25, 1946. An entire production was recorded on
the system during this period. No serious operating difficulties were
encountered during this time. A single valve was used and during
the entire period only one spacing adjustment was necessary. No
other variations were observed, although a daily check was made of
spacing and tuning. These checks were more of a precautionary
measure than a necessity, since only one valve was available for this
production test. The speech and music recordings made with this
modulator leave little to be desired in the way of quality.

Conclusion. — The 200-mil push-pull modulator described in this
paper is a completely integrated design, involving new or improved
design of all the component mechanical and optical elements. It has
proved to be very stable under operating conditions arid at the same
time gives a very high fidelity of response. The simplified structure
of the three ribbon valve makes the tuning and spacing operation more
simple, and the high damping results in greater stability under operat-
ing conditions. Its performance appears to be superior in every way
to the earlier RA-1061 valve that it supersedes.

REFERENCES

1 FRAYNE, J. G., AND SILENT, H. C.: "Push-Pull Recording with the Light
Valve," /. Soc. Mot. Pict. Eng., XXXI, 1 (July 1938), p. 46.

2 SHEA, T. E., HERRIOTT, W., AND GOEHNER, W. R.: "The Principles of the
Light Valve," /. Soc. Mot. Pict. Eng., XVIII, 6 (June 1932), p. 697.

3 FRAYNE, J. G., AND SCOVILLE, R. R. : "Analysis and Measurement of Distor-
tion in Variable-Density Recording," /. Soc. Mot. Pict. Eng., XXXII, 6 (June
1939), p. 648.

4 WENTE, E. C., AND BIDDULPH, R.: "A Light Valve for the Stereophonic
Sound-Film System," /. Soc. Mot. Pict. Eng., XXXVII, 4 (Oct. 1941), p. 397.

6 WILLIAMS, H. G.: "Predicting Spring Performance of Beryllium Copper
Wire and Strip," Iron Age, Reprint, July 8, 1943.

Mathematical Appendix

The following symbols are used throughout the analysis:
v = speed of film in recorder = 18,000 mils per sec
a = width of unmodulated image

514 FRAYNE, CUNNINGHAM, AND PAGLIARULO Vol 47, No. 6

b = amplitude of displacement of signal ribbon image under the influence of

signal subject to limiting value a = b
f = signal frequency
co = 2irf = circular frequency
t = time

y = =t b sin co/ = motion of the signal ribbon as a function of time
Jn(nub/v) = Bessel function of the first kind of order (») and argument nub/v
Jn( —nub/v) = ( — \}nJn (nub/v) Theorem in Bessel functions

51 = Instantaneous width of image in sound track No. 1

52 = Instantaneous width of image in sound track No. 2

Ci = Exposure received by a point on the film in crossing Si
c<i = Exposure received by a point on the film in crossing £2
S = symbol indicating summation of a number of terms in a series

CASE 1

Modulating edges of image are aligned as in Fig. 3a.

*c\ = /i — /o = Si/v = a/v — b/v sin co/i (/)

/! = to + a/v — b/v sin co/x . (2)

co/i = co(/o + a/v) — wb/v sin co/i (3)

= co(/0 + a/v) + 22 \/nJn(—rnab/v) sin »co(/0 + a/v) (4}

Cl = /! - /o = - + - S - Jn[ -^ sin mo U + - (5)

v co i n \ v ) \ vi

Cz = /2 - 4 = — = - + - sin co/! (6)

V V V

By similar development as above

a 2 " 1 / -wco&\ . / a\

Cz = /2 — /i = 2/ — /wl I sm wco I /j I (7;

ycoi«\yy \ v/

/2 - /o = 2a/z; (<?)

/2 = /o + 2a/y (5)

In push-pull reproduction the total output is Eq (10) minus Eq (5).

4*1 /-wcoA / a\

Output = c2 - ci = - - S - Jn{ ) sin wco ( /0 + - )

co i n \ • f , / \ f /

Dec. 1946 PUSH-PULL DENSITY MODULATOR 515

- 0-

CASE 2

Modulating edges of image are aligned as in Fig. 3b.

The theoretical development for the exposures in this case is identical to that of
Case 1, Eqs (5) and (10), except that for this case Eq (8) becomes

*/2 - /o = a/v (13)

and Eq (10) becomes

a 2 " l/-wco&\ .

c-2 = h — /i = --- 2i -I - I sin WOO/Q. (14)

v co i n\ v /

Push-pull output from the film is Eq (14) minus Eq (5).

Output = ct — ci = -- J5 -|./» I - — )

w 1«L \ v /J

sin wco/o + sin wco 1 fo + ~ 1 I (15)

L \ v/_\

4" 1 / — nub\ nua / a\

-2- 7n ( - ) cos - - sin wco I /o H -- )• (16)

o> in \ v / 2v \ 2v/

4f /co6\ coa / a\ 1 /2co6\

Output =» - 1 /il — I cos — sin w I /0 H -- I -- Jz I - 1 cos
w L \ * / 2y \ 2y/ 2 \ v /

CASE 3

Modulating edges of image are aligned as in Fig. 3c.

For this case the exposure in Si is identical to that for Case 1, Eq (5), but the
exposure in £2 becomes

*c-2 = h - to = (a/v) + (b/v) sin «fe. (18)

This is so because in this case the exposures in each sound track begin at the
same instant (to). Then, following the same development as in Case 1,

. . i — 'V* T I 1 1*1 / -t s\\

c% = fa — /o = — | — 2i - Jn[ ) sm wco [ /o H — )' (19)

516 FRAYNE, CUNNINGHAM, AND PAGLIARULO Vol 47, No. 6

Push-pull output from the film is Eq (19) minus Eq (5).

2 ™ 1 1~ /nub\ / — wco&\~l

Output = c2 - d = - 2 -\ Jn [- -} - Jni- -}\
w l w|_ \ v / \ * Vj

(20)

sin wco I /o H —

\ •

r /«co&\ r f —nub\ /nojb\

If w is an odd number /« I - I — /w I - I = 2/w I - 1

\ »•/ V * / V* /

T. . r /W«A r ( ~n

If w is an even number /« I — I — Jn I —

\ » / V "

I
Because of (21) and (22), the output in Eq (20) becomes

(21)
/

0. (22)

4f /co&\ / a\ 1 /3w6\

Output = - /if — ) sin to (/o+ - ) +-/3 - -

coLVz'/ \ v/ 3 \ v /

... .

sin 3co ^o + ~

CASE 4

Four ribbon light valve operating in push-pull.

In this case in order to refer the exposures in the two sound tracks to the same
instant of time (A>) , it is convenient to determine individually the following partial
exposures :

- = sin

v 2v 2v

^_?£17K(^

2v co in \2v /

sin null* (24)

Sz a b a 2

= tz — to = — = sin co/2 = \--

v 2v 2v 2v co

S - Jn [ — -

in \ 2v

53 a b a 2 * 1 / -n<ob\

cz = t0 — t3 = — = — -+ — sin co/3 = 2 - 7w I — — I

v 2v 2v 2v w i n . \ 2v /

( a

sin no)\ to

\ 2»

6*4 a 6 a 2 " 1 (nub\

d = /4 - tQ = — = — + — sin co/4 = — + - 2 - Jn { — - )
v 2v 2v 2v co i n \ 2v /

(25)

(26)

sin wco

Dec. 1946 PUSH-PULL DENSITY MODULATOR 517

The full exposure of sound track No. 1 will then be

and similarly for sound track No. 2

a 2 "

— J sin nu> ( to

/ \

2v I

'° - i)l

Push-pull output, Eq (29) minus Eq (2£), is

Output = E2 - El = /» - /»

w l n

sin wco Mo -h — I — \ Jn\ — — Jn I — — sin

f /wo; A /-nto&\"| r (nub\

for w odd \ Jn I - - I - /w ( - -II- 2/n I - )

L \2p/ \ 2v /J \2yy

for « even /n--

. /— na)b\ /no)b\

for w even \ Jn [ ] — Jn[ 1 = 0

L \ 2v ) \2v J

(50)

so -that in Eq (30} the odd harmonics exist and the even harmonics vanish.
Eq (30) can then be written

8|~" 1 (nub\ ( f a\

Output = E2 - E! = - S - Jn [ - - )< sin «u> ( ^o H -- )
co L l n \2v / ( \ 2v/

+ sinwcof/o ~fj\ W

s r^ i /«co6\ waja

= - S - Jn { — 1 cos — - sin nw/0 (5^)

w L 1 » \ 2y / 2v

518 FRAYNE, CUNNINGHAM, AND PAGLIARULO

and since in (32) the (n) are all odd numbers, (32) expands into

8f /<o&\ coa 1 /3coA

Output = E2 - Ei = - /! (— ) cos— - sin «/o+- /al-
to L \2z;/ 2y 3 \2p/

cos sin 3w*o + . . . |. (33)

* In the above analysis, the following symbols were used

Three Ribbon Push-Pull Light Valve
to = instant when the exposure in Si begins.
h = instant when the exposure in Si ends.
ti = instant when the exposure in S2 begins.
h = instant when the exposure in Sz ends.

Four Ribbon Push-Pull Light Valve
to = time of reference for Si, S2, S3, St.
ti = instant when the exposure in Si begins.
k = instant when the exposure in Sz begins.
tz = instant when the exposure in S$ begins.
/4 = instant when the exposure in £4 begins.

CURRENT LITERATURE OF INTEREST TO THE MOTION PICTURE

ENGINEER

American Cinematographer

27, 9 (Sept. 1946)

The Camera and Production Value (p. 312)

Photographing the Underwater Atomic Bomb
Test at Bikini (p. 315)

Carbon Arc Lighting for 16-Mm Color Pro-
duction (p. 318)

High Fidelity Sound Printing for 16-Mrfi Films
(p. 322)

duPont Perfects Film for Television (p. 325)
27, 10 (Oct. 1946)

Greatest Photographic Organization in History
Shot Bikini Blast (p. 352)

New Filter Technique for Color Cinematog-
raphy (p. 356)

Mitchell's New 16-Mm Professional Camera
(p. 376)

Bell Laboratories Record

24, 10 (Oct. 1946)
Fastax at Bikini (p. 358)

Electronic Engineering

18, 224 (Oct. 1946)
A Timer for Photo-Printing (p. 300)
Line Scanning Systems for Television (p. 302)
16-Mm Sound-on-Film Recorders (p. 309)

International Projectionist

21, 9 (Sept. 1946)

The Forest Electronic Arc Lamp (p. 5)
RCA's New Blue-Sensitive Phototube (p. 8)
The Laboratory Operator (p. 10)

H. A. LIGHTMAN

L. W. KNECHTEL

A. STENSVOLD

L. N. CHRISTIANSEN

G. WARRENTON

R. RENNAHAN

J. H. WADDELL

N. PHELP AND F. TAPPENDEN
A. M. SPOONER
J. NEIL

J. K. ELDERKIN

J. D. PHYFE

R. L. MCKNIGHT

519

520

SOCIETY ANNOUNCEMENTS

Vol 47, No. 6

Better Sound-Reproducing Equipment (p. 12)
Telefilm Race Track Control (p. 23)

21, 10 (Oct. 1946)

The Forest Electronic Arc Lamp (p. 12)
Some Historic Firsts : The Orthophonic Phono-
graph (p. 16)
The Technicolor Cameraman (p. 20)

Acoustical Society of America, Journal

18, 2 (Oct. 1946)
Measurement of Recording Characteristics by

Means of Light Patterns (p. 387)
The Effect of Non-Uniform Wall Distributions
of Absorbing Material on the Acoustics of
Rooms (p. 472)

Institution of Electrical Engineers, Journal

93, 69, Pt. 1 (Sept. 1946)

A Method of Transmitting^Sound on the Vision
Carrier of a Television System (p. 415)

93, 25, Pt. 3 (Sept. 1946)

Approximate Method of Calculating Reflec-
tions in Television Transmission (f>. 352)

C. VERITAS

J. K. ELDERKIN

W. HOCH

B. B. BAUER

H. FESHBACH
HARRIS

AND C. M.

D. I. LAWSON, A. V. LORD,
AND S. R. KHARBANDA

D. A. BELL

SOCIETY ANNOUNCEMENTS

JOURNAL AWARD

The SMPE Journal Award for 1946 was presented to Ralph H. Talbot for his
paper "The Projection Life of Film," published in the JOURNAL of August 1945.
The award, given annually for the most outstanding paper originally published in
the JOURNAL during the preceding year, was announced by President D. E.
Hyndman at the banquet held on October 23 during the 60th Semiannual Con-
vention of the Society in Hollywood. A suitably inscribed certificate was pre-
sented to Mr. Talbot.

The paper was first presented before the Society at the May 1945 Technical
Conference in Hollywood. A biographical sketch of the author, who is associated
with the Eastman Kodak Company, Rochester, will be published in an early
issue of the JOURNAL.

Honorable Mention was given to the paper by D. W. Epstein and I. G. Maloff,
of Radio Corporation of America, entitled "Projection Television," published in
June 1945; also to E. W. Kellogg, of the same organization, for his paper, "ABC
of Photographic Sound Recording," published in March 1945; and to M. H.
Sweet, of Ansco, for the paper entitled "The Densitometry of Modern Reversible
Color Film," published in the June 1945 JOURNAL.

Dec. 1946 SOCIETY ANNOUNCEMENTS 521

FELLOW AWARDS

In recognition of contributions made to the advancement of the motion picture
industry and for services to the Society, seven Active members were elected to
the grade of Fellow by action of the Board of Governors during 1946. At the
banquet on October 23 in Hollywood, appropriate certificates were presented by
President Hyndman to the following:

Ralph B. Austrian, RKO Television Corporation, New York.

Edmund A. Bertram, De Luxe Laboratories, New York.

John W. Boyle, Cinematographer, Hollywood.

Thomas T. Moulton, Twentieth Century-Fox Film Corporation, Beverly Hills,

Calif.
William H. Offenhauser, Jr., Consultant to Columbia Broadcasting System,

New York.
Lawrence T. Sachtleben, RCA Victor Division, Radio- Corporation of America,

Camden, N. J.
Abraham Shapiro, Ampro Corporation, Chicago.

HONOR ROLL

During the summer the Honorary Membership Committee submitted a recom-
mendation to the Board of Governors proposing that the names of Theodore W.
Case, Edward B. Craft, and Samuel L. Warner be added to the Honor Roll of the
Society in view of their contributions to the technical progress of the motion pic-
ture industry. At a meeting on October 20, the Board approved the recom-
mendation and voted to submit these names to the general membership for rati-
fication.

The proposal was unanimously approved by qualified members present at a
business session held on October 21 during the 60th Semiannual Convention.
The names are listed in the Honor Roll on the back cover of the November
JOURNAL, with other outstanding pioneers of the industry now deceased.

SCROLLS OF ACHIEVEMENT

As a highlight of the 60th Semiannual Convention in Hollywood, eight citations
in recognition of outstanding achievement in the field of sound motion pictures
were presented by the Society at the banquet on October 23. The awards were
made in celebration of the Twentieth Anniversary of Talking Pictures, upon
recommendation of the recently formed Committee on Citations. After unani-
mous approval by the Board of Governors, illuminated Scrolls of Achievement
were prepared and awarded to the following:

Dr. Lee de Forest, in recognition of ...

"His original researches which resulted in the invention of the Audion, a
3-electrode vacuum tube, destined to become a basic element in the development
of telephonic communication, radio, and the sound motion picture;

"His demonstration between 1921-1923 of a method of photographing a
variable-density sound record on motion picture film utilizing the Photion lamp,
and a method of reproducing the photographic sound record in synchronism with
a motion picture by means of the Phonofilm projector which incorporated the
Case Thalafide cell and a multistage Audion amplifier;

"His courage and persistent efforts which made possible the first public showing

522 SOCIETY ANNOUNCEMENTS Vol 47, No. 6

of topical sound motion pictures in the Rivoli and Rialto Theaters in New York
on April 15, 1923, and subsequently in many theaters throughout the world.

"These researches and his pioneering vision of a great industry as described
in a paper before members of this Society in May 1923 are recognized by the
Presentation of this Scroll of Achievement by the Society of Motion Picture
Engineers in this Twentieth Year of the Successful Introduction of Sound Motion
Pictures."

The Scroll was received by Mr. Jack Gaines, of Hollywood, in the absence of
Dr. de Forest.

FIG. 1. Scroll of Achievement presented to Dr. Lee de
Forest.

Bell Telephone Laboratories, Inc., in recognition of ...

"Their fundamental research in the art of communication from which came
the development of sound recording and reproducing equipment;

"Their development of methods of high-quality recording on both disk and
film;

"Their design of equipment that made possible the first commercially suc-
cessful sound pictures.

Dec. 1946 SOCIETY ANNOUNCEMENTS 523

"This Scroll of Achievement is presented by the Society of Motion Picture
Engineers in this Twentieth Year of the Successful Introduction of Sound Motion
Pictures."

The Scroll was received by Dr. Harvey Fletcher, Director of Physical Research.

General Electric Company, in recognition of ...

"Their development of the mirror oscillograph and its application to the prac-
tical recording of sound on film by the variable-area method;

"Their early recognition of the possibilities of sound reproduction and their
technical developments in this art which resulted in greatly improved sound
quality.

"This Scroll of Achievement is presented by the Society of Motion Picture
Engineers in this Twentieth Year, of the Successful Introduction of Sound Motion
Pictures."

The Scroll was received by Mr. S. E. Gates, Resident Officer, Los Angeles.

Metro-Goldwyn-Mayer Studios, in recognition of ...

"The impetus given by them to the design and development of theater speaker
and reproducing systems which has greatly enhanced theater reproduction of
sound ;

"Their origination of wide push-pull recording which forms the basis for present
practice and standards;

"Their initiation of many other methods and devices which have been made
available to the industry and which, through improving the flexibility of produc-
tion operations, have made possible a more complete expression of creative
artistry.

"This Scroll of Achievement is presented by the Society of Motion Picture
Engineers in this Twentieth Year of the Successful Introduction of Sound Motion
Pictures."

The Scroll was received by Mr. Douglas Shearer.

RCA Victor Division, Radio Corporation of America, in recognition of ...

"Their pioneering foresight and ingenuity and that of their predecessor, the
Victor Talking Machine Co., in devising equipment and techniques for recording
sound on disk for sound motion pictures;

"Their uninterrupted research and engineering developments in the field of
motion picture sound which has been a source of continuous improvement in
recordings and reproductions;

"Their development of manufacturing and distribution facilities which has
played a vital role in translating the ideas of scientists and engineers into products
and services for both producers and exhibitors of sound motion pictures.

"This Scroll of Achievement is presented by the Society of Motion Picture
Engineers in this Twentieth Year of the Successful Introduction of Sound Motion
Pictures."

The Scroll was received by Mr. Max C. Batsel.

Twentieth Century-Fox Film Corporation, in recognition of ...

"The pioneering work of the Fox-Case Corporation in the development of
motion pictures with sound on film having quality comparable with that of
sound on disk;

"Their engineering developments which resulted in such innovations as the
first studio wholly designed for commercial sound recording, and the perforated
sound screen;

"Their continuing leadership as producers of sound motion pictures of high
quality which included the first unified sound picture news service, 'Movietone
News' (October 1927) and the first out-of-doors recorded feature picture, 'In Old
Arizona' (December 1928).

"This Scroll of Achievement is presented by the Society of Motion Picture

524 SOCIETY ANNOUNCEMENTS Vol 47, No. 6

Engineers in this Twentieth Year of the Successful Introduction of Sound Motion
Pictures."

The Scroll was received by Mr. Earl I. Sponable, of Movietone News.

Western Electric Company, Inc., in recognition of ...

"Their accomplishments - in manufacturing sound recording and reproducing
equipment speedily and in large quantities to meet the sudden and unprecedented
demand from the motion picture industry to provide facilities to studios and
theaters for the conversion from silent to sound pictures ;

"Their introduction of improved equipment and methods of recording as the
art developed.

"This Scroll of Achievement is presented by the Society of Motion Picture
Engineers in this Twentieth Year of the Successful Introduction of Sound Motion
Pictures."

The Scroll was received by Mr. T. K. Stevenson, Vice-President.

Westinghouse Electric and Manufacturing Company, in recognition of ...

"Their pioneering efforts in the development of sound recording and sound
reproduction for motion pictures; and

"The assistance they have given to engineers in standardization of methods
and equipment in the motion picture art.

"This Scroll of Achievement is presented by the Society of Motion Picture
Engineers in this Twentieth Year of the Successful Introduction of Sound Motion
Pictures."

The Scroll was received by Mr. Charles A. Dostal, Vice-President, San Francisco.

Other citations presented in 1946 by the Board of Governors, upon recommen-
dation of the Committee on Citations, were to Thomas Armat on the occasion of
the Fiftieth Anniversary of the first exhibition of motion pictures in a theater, and
to Warner Brothers in recognition of their pioneering courage and efforts in the
development of sound recording and sound reproduction for motion pictures, re-
ported in the August 1946 JOURNAL.

ENGINEERING SOCIETIES COUNCIL

The Atlantic Coast Section of the Society is a charter member of the recently
formed Engineering Societies Council of New York. As in numerous other cities,
the Council is organized for the purpose of promoting the common interests of
engineers and encouraging cooperation between the various engineering societies.
Other charter members of the Council are:

American Institute of Chemical Engineers.

American Institute of Electrical Engineers.

American Institute of Mining and Metallurgical Engineers.

American Society of Heating and Ventilating Engineers.

American Society of Mechanical Engineers.

American Society for Metals.

American Society of Safety Engineers.

American Society for Testing Materials.

American Society of Tool Engineers.

American Chemical Society.

American Welding Society.

Illuminating Engineering Society.

Institute of Radio Engineers.

Dec. 1946 SOCIETY ANNOUNCEMENTS 525

Two persons are selected by each member society as representatives to the
Council. The present representatives of the Atlantic Coast Section are James
Frank, Jr., and C. R. Keith. While the first meetings have been largely con-
i.vi iK'd with organization, it is anticipated that the Council will soon be taking an
active part in engineering affairs in New York City.

SAMUEL L. WARNER MEMORIAL AWARD

President D. E. Hyndman announced at the 60th Semiannual Convention that
the Board of Governors of the Society had unanimously and enthusiastically ac-
cepted the offer of Warner Bros. Pictures to establish an SMPE Samuel L. Warner
Memorial Award. The award will be a suitably designed gold medal and appro-
priate certificate to be presented annually to any individual contributing an
engineering or technical invention or improvement in the art of motion picture
production, distribution, or exhibition which is considered a recent advance in
the industry.

The first award will be made in 1947, and a committee to formulate rules and
procedure for making the awards has been appointed.

ATLANTIC COAST SECTION MEETING

The problems of maintaining motion picture theater sound equipment were
discussed at the November 13 meeting of the Atlantic Coast Section of the Society
by E. S. Seeley, Chief Engineer of the Altec Service Corporation, New York.
Choosing as his title, "The Contribution of Theater Service to 20 Years of Motion
Picture Sound Progress," Mr. Seeley described a number of conditions which had
to be met during war emergencies and scarce material and parts, and demonstrated
the effect on sound quality resulting from the introduction of noiseless recording,
the objections to 6-cycle and 96-cycle flutter, and the effect on sound quality
owing to deterioration of equipment caused by lack of service.

Mr. Seeley said that national service organizations would play a large part in
the successful introduction of new developments in the future. While these de-
velopments are not commercially available today, they might involve such items
as automatic volume control, stereophonic sight and sound, television, panoramic
or wide-angle sound origin, extended frequency and volume range, and new color
film.

Throughout his presentation Mr. Seeley gave many interesting test film demon-
strations of sound quality and methods of checking equipment.

The meeting, held in the Twentieth Century-Fox Little Theater, New York,
was opened with an enjoyable motion picture short.

MIDWEST SECTION MEETING

A large audience of members and guests of the Midwest Section of the Society
in Chicago heard John A. Maurer describe the Maurer film recording equipment
and new Maurer camera at a meeting held November 14. Mr. Maurer reviewed
the status of sound-on-film recording and related picture problems. Resolution
in sound and picture, as theoretical and practical limits, was discussed and demon-
strated by test films.

526 SOCIETY ANNOUNCEMENTS Vol 47, No. 6

The Maurer equipment was analyzed and salient features elaborated. Of par-
ticular interest was the intermittent which utilizes a long intermittent rest period
instead of registration pins to achieve steadiness. Recording optics are made of
precision polished cylindrical surfaces solely, allowing resolution in excess of avail-
able films. The details of the sound drum stabilizer were analyzed and caused
considerable interest.

The meeting was held in the quarters of the Western Society of Engineers, and
was attended by many members of The Institute of Radio Engineers and The
Acoustical Society of America.

PACIFIC COAST SECTION

A symposium on "Special Equipment" was held at the November 26 meeting
of the Pacific Coast Section of the Society in the ERPD Review Room, Holly-
wood. The speakers were Kurt Singer, of Radio Corporation of America, Philip
E. Brigandi, of RKO Radio Pictures, J. K. Milliard, of Altec Lansing Corp., Carroll
Dunning, of Dunning Color, R. Morgan, of Norman B. Neely Enterprises, and G.
A. Mitchell, of Mitchell Camera Company.

In discussing the need and development of special equipment for the motion
picture industry, Mr. Singer and Mr. Brigandi described a variable dip filter and
its application to studio use. The need for the filter was called to RCA's attention
when an arc whistle was inadvertently picked up on a Technicolor production.
Commutator ripple modulation of the illuminating arcs had been recorded on
scenes whose retaking would have involved a prohibitive cost. The arc whistle was
satisfactorily eliminated by the dip filter and the costly scenes made usable. A
demonstration of the variable dip filter was given.

Two new loudspeakers and a new amplifier were described and displayed by Mr.
Milliard. The smaller of the two loudspeakers is designed to give high-quality re-
production as applied to small portable cabinets such as used in the 16-mm field.
The larger loudspeaker, intended for high-quality monitoring, is designed to give
good high-frequency distribution and comparable high-quality reproduction of
low frequencies. The amplifier was described as a 40-watt beam power unit de-
signed primarily for use in disk recording.

Mr. Dunning demonstrated a new development known as the "Animatic
Projector." The portable equipment synchronized a disk record containing voice
and sound effects with a fratne-by-frame projection of stills from 16-mm film. The
disk turntable actuates the picture changes at regular intervals of several seconds.
The equipment is used currently for sales instruction, visual education and tele-
vision.

A Sorensen a-c line voltage regulator was discussed and shown by Mr. Morgan,
who presented slides of the circuits employed. The characteristics of the d-c volt-
age regulator were also described.

Mr. Mitchell gave a short resume of the features of the new Mitchell profes-
sional 16-mm camera, which was displayed with its accessories.

President-elect Loren L. Ryder and Section officers and managers for 1947 were
introduced to the members and guests.

Lively discussion of the papers presented made the meeting one of the most
interesting held recently in Hollywood.

Doc 1946 SOCIETY ANNOUNCEMENTS 527

AMENDMENT OF BY-LAWS

Proposed amendment of By-Law XIII of the Constitution and By-Laws of the
Society, authorizing the Establishment of Student Chapters, as published on page
268 of the September JOURNAL, was discussed and voted on by qualified members
present at a general business session of the Society on October 21 during the reci-uf
Hollywood Convention. It was unanimously approved.

OFFICERS, GOVERNORS, AND SECTION MANAGERS FOR 1947-1948

As a result of the recent elections, the following is a list of the Officers and
Governors of the Society for terms beginning January 1, 1947:

** President: LOREN L. RYDER

** Past-President: DONALD E. HYNDMAN

** Executive Vice-President: EARL I. SPONABLE

* Engineering Vice-President: JOHN A. MAURER
** Editorial Vice-President: CLYDE R. KEITH

* Financial Vice-President: M. RICHARD BOYER

** Convention Vice-President: WILLIAM C. KUNZMANN
** Secretary: .G. T. LORANCE

* Treasurer: E. A. BERTRAM

Governors from the Eastern and Central Time Zones:

* FRANK E. CARLSON ** ROBERT M. CORBIN

* ALAN W. COOK ** DAVID B. JOY

* PAUL J. LARSEN
Governors from the Mountain and Pacific Time Zones:

"* JOHN W. BOYLE * JOHN G. FRAYNE

** CHARLES R. DAILY * WESLEY C. MILLER

** HOLLIS W. MOYSE

Officers and Managers of the Atlantic Coast Section for terms beginning Janu-
ary 1, 1947, are:

* Chairman: JAMES FRANK, JR.

* Past-Chairman: FRANK E. CAHILL, JR.

* Secretary-Treasurer: H. EDWARD WHITE

Managers: * HERBERT BARNETT ** THEODORE LAWRENCE

* HOLLIS D. BRADBURY * JACK A. NORLING

'* F. J. GRIGNON ** WILLIAM H. RIVERS

Officers and Managers of the Midwest Section effective January 1, 1947, are:

* Chairman: A. SHAPIRO

* Secretary-Treasurer: ROBERT E. LEWIS

Managers: * OSCAR B. DEPUE ** C. E. PHILLIMORE

** WILLIAM C. DEVRY * C. H. STONE

* S. A. LUKES ** R. T. VAN NIMAN

Officers and Managers of the Pacific Coast Section taking office as of January 1,
1947, are:

* Chairman: WALLACE V. WOLFE

* Past-Chairman: HOLLIS W. MOYSE

* Secretary-Treasurer: S. P.. SOLOW

.Managers: * GERALD M. BEST ** F. L. EICH

** A. C. BLANEY * GORDON E. SAWYER

* P. E. BRIGANDI ** N. L. SIMMONS

* Term expires December 31, 1947.
** Term expires December 31, 1948.

528 SOCIETY ANNOUNCEMENTS Vol 47, No. 6

BACK ISSUES OF JOURNAL AVAILABLE

We are passing along to interested members information on the availability
of sets of back issues of the JOURNAL. Many of the issues involved are now out
of stock and cannot be obtained through the Society. Since we have received
requests for such sets from time to time, we are glad to cooperate in offering
these JOURNALS to the membership. Details of price, payment, and shipment
must be arranged direct with the owners concerned.

E. W. Nelson, 4525 Altgeld St., Chicago 39, 111., offers to sell the following:

Issues July 1934 through December 1934 (one vol.), bound in standard library
blue cloth binding, gold lettering, $4.00. Issues January 1935 through De-
cember 1942 (bound as above, one year per vol.), 1935 vol. includes 1930-35
cumulative index, $8.00 each. Unbound issues, January 1943 through De-
cember 1946, $6.00 per year. Entire lot $90.00, f .o.b. Chicago.

Mr. Nelson will arrange to have the single issues bound by the same book
binder, should the purchaser desire. These JOURNALS are in excellent condition,
Mr. Nelson states.

Another set of JOURNALS, beginning with the January 1930 issue through
December 1946, is available from M. W. Palmer, 468 Riverside Drive, New York
27, N. Y. These are single copies and Mr. Palmer reports that they are in good
condition. Details as to price, etc., should be discussed with Mr. Palmer direct.

INCREASE IN MEMBERSHIP DUES

INCREASE IN JOURNAL SUBSCRIPTION RATE

Dec. 1946 SOCIETY ANNOUNCEMENTS 529

CORRECTION

In the paper "Factors Governing the Frequency Response of a Variable- Area
Film Recording Channel," by M. Rettinger and K. Singer, published in the
JOURNAL, 47, 4 (Oct. 1946), the authors request that a change be made on p.
303. The sentence in the eighth line beginning "For this reason. . ." should
read:

"For this reason, the reverberation time at the lower registers is usually longer
than for the extreme high tones. This accentuation or loss may be expressed
by. . ."

We are grieved to announce the death of Joseph E. Robin, Active
member of the Society, on December 21, 1946, in Palisades, New Jersey.

a»Uaye, £difo* oj tlte. Mo&o+i Picture,

f"77ie Technique of Motion Picture Production is the
first unified presentation of modern technical practices
in motion picture production . . . Compact and com-
plete ... In plain terms that any interested layman can
understand. . .

I "This volume is indicated on the desk of anybody who
wants to know about the motion picture and how it is
made."

The Technique of Motion Picture Production

CONTENTS

I Technology in the Art of Producing Motion Pictures

Leon S. Becker

II Cinematography in the Hollywood Studios:

Black and White Cinematography John W. Boyle

Putting Clouds into Exterior Scenes Charles G. Clarke

Technicolor Cinematography Winton Hoch

III Special Photographic Effects Fred M. Sersen

IV Re-Recording Sound Motion Pictures L. T. Goldsmith

V The Technique of Production Sound Recording. .Homer G. Tasker

VI Prescoring and Scoring Bernard B. Brown

VII Illumination in Motion Picture Production

R. G. Linderman, C. W. Handley and A. Rodgers

VIII The Paramount Transparency Process Projection Equipment

Farciot Edouart

IX Motion Picture Laboratory Practices James R. Wilkinson

X The Cutting and Editing of Motion Pictures. .Frederick Y. Smith
XI The Projection of Motion Pictures Herbert A. Starke

Price $3.50*

Each section written by a specialist in the motion picture industry. . . Authentic infor-
mation on various technical problems of motion picture production. ... A useful and
valuable reference for technicians, students, librarians, and others desiring techno-
logical data on the motion picture industry compiled in one volume.

Published for the Society of Motion Picture Engineers by Interscience Publishers, Inc.,
215 Fourth Avenue, New York 3, N. Y.

* 20% discount to members in good standing if ordered through SMPE. Orders must be accompanied
by check or money order, «nd include 2% sales tax if delivered in New York Qty.

JOURNAL

OF THE SOCIETY OF

MOTION PICTURE ENGINEERS

AUTHOR AND CLASSIFIED
INDEXES

VOLUME 47
JULY-DECEMBER, 1946

AUTHOR INDEX, VOLUME 47

JULY-DECEMBER, 1946

Author
AUSTRIAN, R. B.

BACK, F. G.

BAUMBACH, H. L.

(and GAUSMAN, H. E.)
BAUMBACH, H. L.

BLOOMBERG, D. J.
(and WATSON, W. O.)

BOLSEY, J.

BOYER, M. R.

(and WHITE, C. F.)
BRADLEY, J. G.

BUCKINGHAM, W. D.

(and DEIBERT, C. R.)
CUNNINGHAM, T. B.

(and FRAYNE, J. G.,

and PAGLIARULO, V.)
DEIBERT, C. R.

(and BUCKINGHAM, W. D.)
DOHERTY, D.

DONNER, V.

532

Title No. Page

A Complete Motion Picture Pro-
duction Plant for Metropolitan
New York 1 (July) 12

Nonintermittent Motion Picture
Projector with Variable Mag-
nification 3 (Sept.) 248

Zoom Lens for Motion Picture
Cameras with Single-Barrel
Linear Movement 6 (Dec.) 464

Aluminum and chromium as

Gelatin Hardeners 1 (July) 22

An Improved Method for the
Determination of Hydro-
quinone and Metol in Photo-
graphic Developers 5 (Nov.) 403

A New Selsyn Interlock Selection

System 6 (Dec.) 469

Naval Training-Type Epidia-
scope for Universal Projection
of Solid Objects 5 (Nov.) 418

A New Film for Photographing

the Television Monitor Tube 2 (Aug.) 152

A National Film Library — The

Problem of Selection 1 (July) 63

Characteristics and Applications

of Concentrated-Arc Lamps 5 (Nov.) 376

An Improved 200-Mil Push-Pull

Density Modulator 6 (Dec.) 494

Characteristics and Applications

of Concentrated-Arc Lamps 5 (Nov.) 376
The Newsreel — Its Production

and Significance: Editing the

Newsreel 5 (Nov.) 357

The Newsreel — Its Production

and Significance : Women's

Fashions 5 (Nov.) 364

AUTHOR INDEX

Author
Du MONT, A. B.

FRAYNE, J. G.

(and CUNNINGHAM, T. B.

and PAGLIARULO, V.)
GAUSMAN, H. E.

(and BAUMBACH, H. L.)
GORDON, J.

HOLST, B.

HOPPER, F. L.

(and MOODY, R. C.)
HYNDMAN, D. E.

(and MAURER, J. A.)

ISAAC, L. B.
JOHNSON, G. A.
JONES, R. W.

KUDAR, J.
LAWRENSON, H.

LEWIS, C. E.
MAURER, J. A.

(and HYNDMAN, D. E.)

MCGRATH, W. M:

MclNNIS, W.

MESCHTER, E.
MONROE, H. S.

533

Page

Title No.

The Relation of Television to

Motion Pictures 3 (Sept.) 238

An Improved 200-Mil Push-Pull

Density Modulator 6 (Dec.) 494

Aluminum and Chromium as

Gelatin Hardeners 1 (July) 22

The Newsreel — Its Production
and Significance: The Field
Unit 5 (Nov.) 367

The Newsreel— Its Production
and Significance: The Film
Library 5 (Nov.) 365

A Simplified Recording Trans-
mission System 2 (Aug.) 132

The Past and Future Activities
of the Society of Motion Pic- '
ture Engineers 3 (Sept.) 212

Television and the Motion Pic-
ture Theater 6 (Dec.) 482

The Processing Control Sensitom-

eter 6 (Dec.) 474

The Application of Pure Mathe-
matics to the Solution of
Geneva Ratios 1 (July) 55

Optical Problems of the Image
Formation in High-Speed Mo-
tion Picture Cameras 5 (Nov.) 400

The Newsreel — Its Production
and Significance : Foreign .
Editions 5 (Nov.) 361

The High Cost of Poor Projection 4 (Oct.) 295

The Past and Future Activities
of the Society of Motion Pic-
ture Engineers 3 (Sept.) 212

The Newsreel— Its Production
and Significance : Newsreel
Sound 5 (Nov.) 371

The Newsreel — Its Production
and Significance: The News-
reel Cameraman 5 (Nov.) 368

Television Reproduction from

Negative Films 2 (Aug.) 165

Technical Problems of Film Pro-
duction for the Navy's Special
Training Devices 6 (Dec.) 487

534

AUTHOR INDEX

Vol 47, No. 6

Author

MOODY, R. C.

(and HOPPER, F. L.)
MUELLER, W. A.

MURRAY, A. E.

OLSON, H. F.

(and PRESTON, J.)
PAGLIARULO, V.

(and FRAYNE, J. G.,

and CUNNINGHAM, T. B.)
POZNER, W. A.
PRESTON, J.

(and OLSON, H. F.)
RETTINGER, M.

(and SINGER, K.)

RODGERS, W. F.
ROSE, A.

RYDER, L. L.

SEARY, E. G.

(and VARDEN, L. E.)
SHANER, V. C.

(and SPARKS, M. R.)

SINGER, K.

(and RETTINGER, M.)

SLYFIELD, C. O.
SMITH, P. V.

(and STANKO, E.)
SPARKS, M. R.

(and SHANER, V. C.)

STANKO, E.

(and SMITH, P. V.)
VARDEN, L. E.

(and SEARY, E, C.)
WALLER, F.

•Title

No. Page

A Simplified Recording Trans-
mission System 2 (Aug.) 132

Dubbing and Post-Synchroniza-
tion Studios 3 (Sept.) 230

The Photometric Calibration of

Lens Apertures 2 (Aug.) 142

Wide-Range Loudspeaker De-
velopments 4 (Oct.) 327

An Improved 200-Mil Push-Pull

Density Modulator 6 (Dec.) 494

Synchronization Technique 3 (Sept.) 191

Wide-Range Loudspeaker De-
velopments 4 (Oct.) 327

Factors Governing the Frequency
Response of a Variable-Area
Film Recording Channel 4 (Oct.) 299

Motion Pictures Tomorrow 2 (Aug.) 120

A Unified Approach to the Per-
formance of Photographic
Film, Television Pickup Tubes,
and the Human Eye 4 (Oct.) 273

Modernization Desires of a Major

Studio 3 (Sept.) 225

Rapid Test for Ferricyanide

Bleach Exhaustion 6 (Dec.) 450

Application of Methyl Ethyl
Ketone to the Analysis of De-
velopers for Elon and Hydro-
quinone 5 (Nov.) 409

Factors Governing the Frequency
Response of a Variable- Area
Film Recording Channel 4 (Oct.) 299

Tone Control for Rerecording 6 (Dec.) .453

Postwar Test Equipment for

Theater Servicing 6 (Dec.) 457

Application of Methyl Ethyl
Ketone to the Analysis of De-
velopers for Elon and Hydro-
quinone 5 (Nov.) 409

Postwar Test Equipment for

Theater Servicing 6 (Dec.) 457

Rapid Test for Ferricyanide

Bleach Exhaustion 6 (Dec.) 450

The Waller Flexible Gunnery
Trainer 1 (July) 73

Dec. 1946

Author
WALLINGSFORD, A.

WATSON, W. O.

(and BLOOMBERG, D. J.)
WEISS, J. P.

WESTMIJZE, W. K.

WHITE, C. F.

(and BOYER, M. R.)

AUTHOR INDEX

Title

A Film-Splicing and Repair Ma-
chine
A New Selsyn Interlock Selection

No.

535

Page

3 (Sept.) 254
6 (Dec.) 469

System

Sensitometric Control of the

Duping Process 6 (Dec.) 443

A New Method of Counteracting
Noise in Sound Film Repro-
duction 5 (Nov.) 426

A New Film for Photographing

the Television Monitor Tube 2 (Aug.) 152

CLASSIFIED INDEX, VOLUME 47
JULY-DECEMBER, 1946

American Standards Association

American Standards on Motion Pictures, No. 3 (Sept.), P- 258.

Arcs

Report of the Committee on Studio Lighting, No. 2 (Aug.), p. 113.
Characteristics and Applications of Concentrated-Arc Lamps, W. D. Bucking-
ham and C. R. Deibert, No. 5 (Nov.), p. 376.

Atlantic Coast Section (See SMPE Activities and Announcements}
Awards and Citations (See SMPE Activities and Announcements}

Cinematography

Modernization Desires of a Major Studio, L. L. Ryder, No. 3 (Sept.), p. 225.
The Newsreel — Its Production and Significance: The Newsreel Cameraman,

W. Mclnnis, No. 5 (Nov.), p. 368.
Zoom Lens for Motion Picture Cameras with Single-Barrel Linear Movement,

F. G. Back, No. 6 (Dec.), p. 464.

Cinematography, High-Speed

Optical Problems of the Image Formation in High-Speed Motion Picture
Cameras, J. Kudar, No. 5 (Nov.), p. 400.

Color

Rapid Test for Ferricyanide Bleach Exhaustion, L. E. Varden and E. G. Seary,
No. 6 (Dec.), p. 450.

Committee Activities and Reports

Motion Picture Instruction: (Survey of schools, colleges, and universities having
courses of instruction in motion picture engineering), No. 2 (Aug.), p. 95.

16-Mm and 8-Mm Motion Pictures: (Enlarged committee and new scope
defined), No. 2 (Aug.), p. 107.

Standards: (Review of projects under study), No. 2 (Aug.), p. 110.

Studio Lighting: (Light output and levels of various studio lighting equip-
ment), No. 2 (Aug.), p. 113.

Subcommittee on 16-Mm Film Splices: (Review of present standards and new
proposals). No. 1 (July), p. 1.

Television Projection Practice: (Review of subcommittee activities), No, 2

(Aug.), P, 118,
536

CLASSIFIED INDEX 537

Current Literature

No. 2 (Aug.), p. 182; No. 4 (Qct.), p. 353; No. 6 (Dec.), p. 519.

Distribution

Motion Pictures Tomorrow, W. F. Rodgers, No. 2 (Aug.), p. 120.
Synchronization Technique, W. A. Pozner, No. 3 (Sept.), p. 191.
Dubbing and Post-Synchronization Studios, W. A. Mueller, No. 3 (Sept.),
p. 230.

Dubbing (See Sound Recording)

Editing

A Film-Splicing and Repair Machine, A. Wallingsford, No. 3 (Sept.), p. 254.
The Newsreel — Its Production and Significance: Editing the Newsreel, D.
Doherty, No. 5 (Nov.), p. 357.

Education

Report of the Committee on Motion Picture Instruction, No. 2 (Aug.), p. 95.

Engineering Vice-President, SMPE

The Past and Future Activities of the Society of Motion Picture Engineers,
D. E. Hyndman and J. A. Maurer, No. 3 (Sept.), p. 212.

Film, General

A New Film for Photographing the Television Monitor Tube, C. F. White

and M. R. Boyer, No. 2 (Aug.), p. 152.
Television Reproduction from Negative Films, E. Meschter, No. 2 (Aug.),

p. 165.
A Unified Approach to the Performance of Photographic Film, Television

Pickup Tubes, and the Human Eye, A. Rose, No. 4 (Oct.), p. 273.

Film Preservation and Storage

A National Film Library— The Problem of Selection, J. G. Bradley, No. 1
(July), p. 63.

Fixing Baths

Aluminum and Chromium as Gelatin Hardeners, H. L. Baumbach and H. E.

Gausman, No. 1 (July), p. 22.

General

A National Film Library — The Problem of Selection, J. G. Bradley, No. 1

(July), p. 63.

Report of the Committee on Motion Picture Instruction, No. 2 (Aug.), p. 95.
Motion Pictures Tomorrow, W. F. Rodgers, No. 2 (Aug.), p. 120.

High-Speed Photography (See Cinematography, High-Speed]

Illumination

Characteristics and Applications of Concentrated-Arc Lamps, W. D. Bucking-
ham and C. R. Deibert, No. 5 (Nov.), p. 376.

538 CLASSIFIED INDEX Vol 47, No. 6

Illumination, Studio

Report of the Committee on Studio Lighting, No. 2 (Aug.), p. 113.
Modernization Desires of a Major Studio, L. L. Ryder, No. 3 (Sept.), P- 225.

Journal Award (See SMPE Activities and Announcements}

Laboratory Practice

An Improved Method for the Determination of Hydroquinone and Metol in
Photographic Developers, H. L. Baumbach, No. 5 (Nov.), p. 403.

Application of Methyl Ethyl Ketone to the Analysis of Developers for Elon
and Hydroquinone, V. C. Shaner and M. R. Sparks, No. 5 (Nov.), p. 409.

Sensitometric Control of the Duping Process, J. P. Weiss, No. 6 (Dec.), p. 443.

Lenses (See Optics')

Loudspeakers

Wide-Range Loudspeaker Developments, H. F. Olson and J. Preston, No. 4
(Oct.), p. 327.

Midwest Section (See SMPE Activities and Announcements}

Newsreels

The Newsreel — Its Production and Significance: Editing the Newsreel, D.
Doherty, p. 357; Foreign Editions, H. Lawrenson, p. 361; Women's Fash-
ions, V. Donner, p. 364; The Film Library, B. Hoist, p. 365; The Field
Unit, J. Gordon, p. 367; The Newsreel Cameraman, W. Mclnnis, p. 368;
Newsreel Sound, W. M. McGrath, p. 371 ; No. 5 (Nov.).

Obituary

Leon Gaumont, No. 2 (Aug.), p. 189.

J. E. McAuley, F. C. Coates, No. 3 (Sept.), p. 271.

J. E. Robin, No. 6 (Dec.), p. 529.

Optics

The Photometric Calibration of Lens Apertures, A. E. Murray, No. 2 (Aug.),

p. 142.
Optical Problems of the Image Formation in High-Speed Motion Picture

Cameras, J. Kudar, No. 5 (Nov.), p. 400.
Naval Training-Type Epidiascope for Universal Projection of Solid Objects,

J. Bolsey, No. 5 (Nov.), p. 418.
Zoom Lens for Motion Picture Cameras with Single-Barrel Linear Movement,

F. G. Back, No. 6 (Dec.), p. 464.

Pacific Coast Section (See SMPE Activities and Announcements}

Photometry

The Photometric Calibration of Lens Apertures, A. E. Murray, No. 2 (Aug.),

p. 142.
Preservation (See Film Preservation and Storage}

Dec. 1946 CLASSIFIED INDEX 539

President, SMPE

The Past and Future Activities of the Society of Motion Picture Engineers,
D. E. Hyndman and J. A. Maurer, No. 3 (Sept.), P 212.

Processing

Aluminum and Chromium as Gelatin Hardeners, H. L. Baumbach and H. E.

Gausman, No. 1 (July), p. 22.
An Improved Method for the Determination of Hydroquinone and Metol in

Photographic Developers, H. L. Baumbach, No. 5 (Nov.), p. 403.
Application of Methyl Ethyl Ketone to the Analysis of Developers for Elon

and Hydroquinone, V. C. Shaner and M. R. Sparks, No. 5 (Nov.), p. 409.
Sensitometric Control of the Duping Process, J. P. Weiss, No. 6 (Dec.), p. 443.
The Processing Control Sensitometer, G. A. Johnson, No. 6 (Dec.), p. 474.
Rapid Test for Ferricyanide Bleach Exhaustion, L. E. Varden and E. G. Seary,

No. 6 (Dec.), p. 450.

Production

A Complete Motion Picture Production Plant for Metropolitan New York,
R. B. Austrian, No. 1 (July), p. 12.

Synchronization Technique, W. A. Pozner, No. 3 (Sept.), p. 191.

Modernization Desires of a Major Studio, L. L. Ryder, No. 3 (Sept.), p. 225.

Dubbing and Post-Synchronization Studios, W. A. Mueller, No. 3 (Sept.),
p. 230.

A New Selsyn Interlock Selection System, D. J. Bloomberg and W. O. Watson,
No. 6 (Dec.), p. 469.

Projection

Nonintermittent Motion Picture Projector with Variable Magnification, F. G.

Back, No. 3 (Sept.), p. 248.

The High Cost of Poor Projection, C. E. Lewis, No. 4 (Oct.), p. 295.
Naval Training-Type Epidiascope for Universal Projection of Solid Objects,

J. Bolsey, N'/. 5 (Nov.), p. 418.

Projectors

The Application of Pure Mathematics to the Solution of Geneva Ratios, R. W.
Jones, No. 1 (July), p. 55.

Pull-Down Mechanisms

The Application of Pure Mathematics to the Solution of Geneva Ratios, R. W.
Jones, No. 1 (July), p. 55.

Rerecording (Sec Sound Recording)

SMPE Activities and Announcements

(See also Committee Activities and Reports)

540 CLASSIFIED INDEX Vol 47, No. 6

Amendment of By-Law XIII, No. 3 (Sept.), p. 268; No. 6 (Dec.), p. 527.
Atlantic Coast Section:

Meetings, Oct. 16— No. 5 (Nov.), p. 441; Nov. 13— No. 6 (Dec.), p. 525.
Committee Reports, see Committee Activities and Reports.
Dues, Increase in Membership, No. 5 (Nov.), p. 441; No. 6 (Dec.), p. 528.
Engineering Societies Council of New York, No. 6 (Dec.), p. 524.
Fellow Awards in 1946, No. 6 (Dec.), p. 521.
Fifty- Ninth Semiannual Technical Conference:

Address, "Motion Pictures Tomorrow," W. F. Rodgers, No. 2 (Aug.), p. 120.

Citations to Thomas Armat and Warner Brothers, No. 2 (Aug.), p. 124.
Honor Roll, Names to be Listed, No. 6 (Dec.), p. 521.
Journal, Back Issues Available, No. 6 (Dec.), p. 528.
Journal Award, Announcement of 1946 Recipient, No. 6 (Dec.), p. 520.
Midwest Section:

Meetings, June 20— No. 1 (July), p. 92; Oct. 10— No. 5 (Nov.), p. 441;

Nov. 14— No. 6 (Dec.), p. 525.

Officers, Governors, and Section Managers for 1947-1948, No. 6 (Dec.), p. 527. (
Pacific Coast Section:

Meetings, June 10— No. 1 (July), p. 92; Nov. 26— No. 6 (Dec.), p. 526.
Past and Future Activities of the Society of Motion Picture Engineers, The,

D. E. Hyndman and J. A. Maurer, No. 3 (Sept.), p. 212.
SMPE Samuel L. Warner Memorial Award, Announcement of New Society

Award, No. 6 (Dec.), p. 525.

Scrolls of Achievement, Announcement of New Society Awards and 1946 Re-
cipients, No. 6 (Dec.), p. 521.
Sixtieth Semiannual Convention :

Announcements, No. 1 (July), p. 88; No. 2 (Aug.), p. 184; No. 3 (Sept.),
p. 265.

Presentation of Scrolls of Achievement to 1946 Recipients, No. 6 (Dec.),

p. 521.
Subscriptions, Increase in Rate, No. 5 (Nov.), p. 442; No. 6 (Dec.), p. 528.

Sensitometry

Sensitometric Control of the Duping Process, J. P. Weiss, No. 6 (Dec.), p. 443.
The Processing Control Sensitometer, G. A. Johnson, No. 6 (Dec.), p. 474.

Sixteen-Mm Motion Pictures

Report of the Subcommittee on 16-Mm Film Splices, No. 1 (July), p. 1.
Report of the Committee on 16-Mm and 8-Mm Motion Pictures, No. 2 (Aug.),

p. 107.
A New Film for Photographing the Television Monitor Tube, C. F. White and

M. R. Boyer, No. 2 (Aug.), p. 152.

Sound Recording

A Complete Motion Picture Production Plant for Metropolitan New York,

R. B. Austrian, No. 1 (July), p. 12.
A Simplified Recording Transmission System, F. L. Hopper and R. C. Moody,

No. 2 (Aug.), p. 132.

Synchronization Technique, W. A. Pozner, No. 3 (Sept.), p. 191.
Modernization Desires of a Major Studio, L. L. Ryder, No. 3 (Sept.), p. 225.

Dec. 1946 CLASSIFIED INDEX 541

Dubbing and Post-Synchronization Studios, W. A. Mueller, No. 3 (Sept.), p.

230.
Factors Governing the Frequency Response of a Variable-Area Film Recording

Channel, M. Rettinger and K. Singer, No. 4 (Oct.), p. 299; Correction, No.

6 (Dec.), p. 529.
The Newsreel — Its Production and Significance: Newsreel Sound, W. M.

McGrath, No. 5 (Nov.), p. 371.
A New Method of Counteracting Noise in Sound Film Reproduction, W. K.

Westmijze, No. 5 (Nov.), p. 426.
A New Selsyn Interlock Selection System, D. J. Bloomberg and W. O. Watson,

No. 6 (Dec.), p. 469.
An Improved 200-Mil Push-Pull Density Modulator, J. G. Frayne, T. B.

Cunningham and V. Pagliarulo, No. 6 (Dec.), p. 494.
Tone Control for Rerecording, C. O. Slyfield, No. 6 (Dec.), p. 453.

Sound Reproduction

A New Method of Counteracting Noise in Sound Film Reproduction, W. K.

Westmijze, No. 5 (Nov.), p. 426.
Postwar Test Equipment for Theater Servicing, E. Stanko and P. V. Smith,

No. 6 (Dec.), p. 457.

Special Effects

Xaval Training-Type Epidiascope for Universal Projection of Solid Objects,

J. Bolsey, No. 5 (Nov.), p. 418.

Zoom Lens for Motion Picture Cameras with Single-Barrel Linear Movement,
F. G. Back, No. 6 (Dec.), p. 464.

Splicing

Report of the Subcommittee on 16-Mm Film Splices, No. 1 (July), p. 1.
A Film-Splicing and Repair Machine, A. Wallingsford, No. 3 (Sept.), p. 254.

Standards

Report of the Committee on Standards, No. 2 (Aug.), p. 110.
American Standards on Motion Pictures, No. 3 (Sept.), p. 258.

Studios (See Production)

Studio Lighting (See Illumination, Studio)

Synchronization (See Sound Recording)

Television

Report of the Committee on Television Projection Practice, No. 2 (Aug.), p.

118.
A New Film for Photographing the Television Monitor Tube, C. F. White and

M. R. Boyer, No. 2 (Aug.), p. 152.
Television Reproduction from Negative Films, E. Meschter, No. 2 (Aug.),

p. 165.
A Unified Approach to the Performance of Photographic Film, Television

Pickup Tubes, and the Human Eye, A. Rose, No. 4 (Oct.), p. 273.

542 CLASSIFIED INDEX

The Relation of Television to Motion Pictures, A. B. DuMont, No. 3 (Sept.)

p. 238.
Television and the Motion Picture Theater, L. B. Isaac, No. 6 (Dec.), p. 482.

Theaters, General

Television and the Motion Picture Theater, L. B. Isaac, No. 6 (Dec.), p. 482.
Postwar Test Equipment for Theater Servicing, E. Stanko and P. V. Smith,
No. 6 (Dec.), p. 457.

Training Films

The Waller Flexible Gunnery Trainer, F. Waller, No. 1 (July), p. 73.

Nonintermittent Motion Picture Projector with Variable Magnification, F. G.
Back, No. 3 (Sept.), p. 248.

Technical Problems of Film Production for the Navy's Special Training De-
vices, H. S. Monroe, No. 6 (Dec.), p. 487.

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "Journal of the Society of Motion Picture Engineers"

See other formats