History and development of the Bureau of Standards radio beacon experiment station at College Park, Maryland / by Robert W. Beckham

THE HISTORY AND DEVELOPMENT OF THE BUREAU OE STANDARDS
RADIO BEACON EXPERIMENT STATION
AT COLLEGE PARK, MD.

By Roloert W« BecHtiam
December 18, 1936

AN INITIATION REQUIREMENT
FOR

THE MARYLAND BETA CHAPTER OF TAU BETA PI ASSOCIATION

- 1 -

SUMMARY

TKE HISTORY AND DEVELOPl'LIIT OF THE BUREAU OF STANDARDS
RADIO BEACON EXPERIMENT STATION AT COLLEGE PARK, MD.

1. Introduction

2. Early Development

3. The Aura! Radiobeacon

4. The College Park Station

5. Improved Aural Reception

6. Early Double -Modulation System

7. The Tuned-Reed Course Indicator

8. Automatic Volume Control

9. Bent Courses

10. Marker Beacons

11. Simultaneous Radiotelephone & Visual Beacon

12. Reed Indicator of Pointer Type

13. Miscellaneous Developments

14. Conclusion

15. Bibliography

- 2 -
INTRODUCTION

The widespread use of heavier- than- air craft today
renders difficult the realization that mankind's first successful
powered flight in an airplane was made by Willrar Wright just 33
short years ago. The intervening years have witnessed such im-
provements in safety, comfort, speed, and flexibility, that in
1936 the airline is an accepted channel of passenger and light-
freight transportation, rendering rapid and convenient service
on a definite time schedule. This phrase "on a time schedule"
is especially significant, since only "by the maintenance of a
dependable schedule can the airlines attract the patronage of in-
dividuals and concerns with whom time is a major consideration.

As long as the pilot was frequently doubtful of his
ability to ascertain his position with respect to points on the
earth, a time schedule was impossible. Hence, the history of com-
mercial aviation is, to a large extent, the history of efforts
made to overcome the limitations on visibility imposed by weather
conditions.

Quite early in the game it became apparent that any prac-
tical solution of the problem of blind flight must come through the
use, in some form, of the principles of radio. The altimeter, com-
pass, and other standard airplane instruments make the task of
actually maintaining flight while fog-bound a comparatively simple
one. In fact it is quite easy by proper use of the compass to pre-
serve any direction of flight desired. However, the effect of
wind-drift, the extent of which must remain unknown to the pilot,
renders almost impossible the task of arriving at any given poin t.

- 3 -

With no visible landmark against which to check, the increase or
diminution of speed, or the side-drift with respect to the ground
is not determinable. The result might conceivably "be the exhaustion
of fuel in an out-of-v/ay or unlandable area. In any event incon-
venience and delay are met.

The use of the radio links the plane to the invisible
earth so that not only the proper direction, hut also the proper
route in that direction are maintained. This paper will aim to
point out the major developments in the evolution of airway radio-
beacons for this purpose.

Most of the important work on radiobeacons in this
country was carried on by the U. S. Bureau of Standards, in lar^e
part at the airport in College Park. The history of the experiment
station set up at this airport for use in the development is, in
itself, of small interest. Indeed, a chronology of this station
is not available. However, a description of the work done at
College Park and the unfoldment of the system worked out are tanta-
mount to a history of the station. Hence, this paper must deal
largely with the airway radiobeacon as such.

The experiments at College Park consisted of two phases
of approximately equal importance. These are the development of the
airway radiobeacon proper, and that of radio blind-landing aids.
The work was done on these two problems concurrently, but only the
phase of the airway radiobeacon has been treated herein, leaving
the discission of the blind-landing material to some future initiate.
Any difficulty experienced in the writing of this thesis has arisen
entirely from the wealth of data available which had to be sifted
and shortened in order that the paper v/ould not be too voluminous.

- 4 -

It is with the thought in mind of presenting only the most per-
tinent facts that this thesis has "been written.

IOO MJ.

iO r*\.Pi-l. Winp

i "i

true couple t

E".f f«t of winj d«-if t

- 5 -

EARLY DEVELOPMENT

The needs of military flyers during the World War were
the incentive for the first work done in adapting the radio to
airplane guidance requirements. The inherent need of fighting
aircraft is for an extremely flexible system of guidance. There
can be in this type of work no fixed airways such as are needed
by commercial lines. Because of the well known directional proper-
ties of the loop antenna, this was the equipment used in the early
work in the field. The reception characteristic of the loop an-
tenna is in the form of a figure eight, with the long axis in the
plane of the loop. Hence, when the loop is rotated, the strong-
est signals are received when its plane passes through the source
of the signal. In this system a receiving set carried by the plane
picked up the signals transmitted from a ground station on the
loop antenna of the receiving set, and thus gave the proper direc-
tion of the source, vrtiich was located at the home field. If there
were side winds the effect would simply be to shift the plane from
its course and change the direction of the station in relation to
the plane. By continually flying in the direction of the received
signals the pilot would eventually arrive at the landing base, al-
though he might reach it in a very round-about manner (see Eig. 1 }.
A great deal of trouble was experienced with the receiving appara-
tus on the planes, and the system was found to be generally in-
adequate.

Another system which patterns closely after the one just
described was brought forth as a next logical step. It is, in-
cidentally, the one in wide use in Europe, though outmoded in this

- 6 -

country. Each airplane carries a trailing wire antenna, which
has no directional properties. Ground stations, on radio request
from the airplane determine its direction from their loop antennae,
and several radio their findings to the plane, which may then as-
certain its position by tri angulation. This requires two-way
equipment both on the ground and in the plane. Moreover, if a
large volume of flying is being done, the system is inadequate to
take care of more than one plane at a time. In bad flying weather,
when most needed, the system thus becomes jammed.

- 7 -
THE AURAL RABI0BEAC01T

The next important development was undertaken in 1920
by the Bureau of Standards at the request of the Army Air Service.
Cooperating in this work were the Bureau of Lighthouses, and the
Army Signal Corps and Air Service. It is on the work begun at
this time that the entire subsequent experiments were based.
Hence, a thorough -understanding of the principle involved is nec-
essary for an appreciation of the work done at College Park.

Use is again made of the characteristics of the loop an-
tenna. It has been previously noted that the receiving character-
istic is in the shape of a figure eight. The transmitting proper-
ties are similar, the signals being strongest in the direction of
the plane of the coil, and no signals at all being received on a
line through the middle of the eight in a direction of 90 degrees
to the plane of the coil. Thus, in Fig. 2- , the strongest signal
is received on line OB 1 , proportional to OB; no signal on line 00' ;
and the signals received on OA' , OB', OC 1 , are proportional to OA,
OB, OC, respectively.

If two such coils are used at an angle to each other, the
resulting combined transmission characteristic will be as shown in
Fig. 3 . It can easily be seen that the signals will be of equal
intensity on the bisectors of the angles between the coils, and in
no other directions. It may also be seen that the signals become
of unequal intensity very rapidly as the receiver moves off of this
line. Thus, in effect, there are set up 4 narrow sectors of equal
signals from each coil. If power is put on the coils alternately,
a signal is received in such a sector from one antenna, and a moment

- 8 -

later a signal of equal intensity from the other. Thus a plane
"being guided "by such a device would travel to or from the trans-
mitter on an equal signal sector. The moment the plane was blown
off the course by a side wind the pilot would know, because of
the resulting variation in intensity of the two signals.

As developed by the Bureau of Standards originally, the
Morse letters "A" and "T" were sent out, one on each antenna. A
2 kilowatt quenched spark transmitter was used with a double-pole
double- throw switch to change the radio-frequency power from one
antenna to the other. The frequency used was 300 kilocycles. Re-
ception was by headphones, thus giving rise to the name "aural
radiobeacon" . In tests conducted at that time it was found that
the equisignal zone was approximately 1.5 miles wide at a distance
from the source of 31 miles. This was due to the difficulty in de-
tecting small changes of signal intensity by ear. The angle
between the two antennae was 143.5 degrees. These initial tests
were made on signals generated at the Bureau of Standards as re-
ceived on board the Bureau of Lighthouses vessel "Maple" in the
Potomac River just off llathias Point. Many interesting effects
were noted in these tests, but unfortunately a detailed description
of them is beyond the scope of this paper. One, however, is im-
portant enough to merit attention. It was found that unequal cur-
rent distribution in the coil impaired its directive properties.
{See Fig. 4 and Fig. 5 ).

The Bureau had now demonstrated conclusively that the
theory underlying the new invention was borne out in practice.
However, there had as yet been no actual flying tests, because a

- 9 -

"boat had been used in these experiments. Since the work was "be-
ing done in close cooperation with the Army, an Army field was
chosen for adaptation of the method to aircraft, lie Cook Field at
Dayton, Ohio, was selected for this purpose "by the Army engineers.

The refinements upon the original apparatus were numerous
in this new location. The antenna changeover was accomplished auto-
matically and the coils were placed at a 135 degree angle. A 5
kilowatt set was used, transmitting on 300 kilocycles, "but the
connections were essentially as originally set up at the Bureau of
Standards. It was brought out that the equisignal zone of the
crossed loops was shifted by unequal current values in the loops.
Moreover it was noted that in flight the zone was shifted by the
directional receptive qualities of the trailing wire antenna used.
A heavy and short trailing wire, hanging nearly vertical from the
plane, was found to correct this effect.

At this juncture the Bureau of Standards, having ful-
filled the request of the Air Service, dropped from the picture;
and the following four years witnessed development of the radio-
beacon by the Army engineers alone.

Of incidental interest is the fact that at about this
time the term "radio range" came into use to mean any directive
radiobeacon transmitting apparatus.

The chief contributions of the Army engineers during
this 4 year period were in the development of a signal interlock-
ing device and a goniometer for rotating the equisignal zones
through space without moving the antenna loops. Both of these
steps were based on patents issued in 1907; the first to a German,

- 10 -

0. Scheller; the other on the Bellini-Tosi system. The Army men
simply combined the two foreign patents into a system useable in
radio range operation. The exact nature of these improvements
will "be explained in the article on the subsequent work done at
College Park.

Early in 1926 there was before Congress a bill to create
an Aeronautics Branch of the Department of Commerce to supervise
commercial aviation. The proponents of the measure in the Depart-
ment of Commerce requested that the Bureau of Standards submit to
them v/ays in which radio might be used in airway navigation aids
by the proposed Branch. V/hen the bill passed, the Bureau was
called upon to do research work along the lines which it had sug-
gested while the law was pending.

Fig. Z
Look* Antenna C Ira.rdetei'i sT(c

f- cjuisi^ruxl

r Fi 9- 3.

C— bj nr J dwUt*.|.fa« ofiUH

1 50

T=To.t«s l' Square
S" Af>o.rt

VO

X "3

I-

T

GAP

Fig. 4.

^'r*' Current distribution
in loot?

J> 5*.

1^0

^o

1-3

T

1 Hh

TYTT71

GAP

R 9 . 5-

EL^uaJ Co<ire*i"t distribution
1V1 too|=»

- 11 -

THE COLLEGE PARK STATION

A site near the Bureau of Standards was needed for the
research work now necessary. The airport at College Park was
chop en "because it is a large, flat field, sufficiently close to
the Bureau, and free from any wires which might Interfere with the
experiments. The station consisted of a small frame structure with
a tower mounted on top of it on which to place the antennae. It
was equipped with radioheacon, radiotelephone, and radiotelegraph
apparatus. A picture of the station and antennae used may be seen
in Figure 6. This station was "begun in July of 1926, immediately
that the Bureau received its assignment.

At the same time a similar station was set up in the
dangerous mountain area at Bellefonte, Pa. on the New York- Cleve-
land airway. This station was transferred to the Airways Division
of the Commerce Department in 1928 to give radio service to the
airway on which it is placed.

To "begin with, it was desired (1) to improve the design
of the existing apparatus; (2) to replace the aural reception of
the signals with visual reception; and (3) to develop the equipment
so that several intersect in,; courses could be laid out from the
same piece of apparatus.

7)

<_d'

n

f

7-

CP

o

2

Mterf^i^i^^^

- is -

IMPROVED AURAL RECEPTION

The goniometer as used "by the Army, and "based on the
Bellini-Tosi patent was simply a mutual inductance device, each
of whose- two primaries conjoined with the two crossed secondaries
and two crossed antennae to set up a phantom antenna, electrically
equivalent to a loop antenna. By changing the relative positions
of the primary and secondary coils, the plane of the phantom an-
tenna was rotated through space. The primaries were of one or two
turns connected in a tuned circuit fed by the output of the trans-
mitter. A cam- operated relay was used to throw the power from one
primary to the other, such that the signals as heard in the equi-
signal zone interlocked, making one continuous sound. This meant
that the circuit was broken on the radio frequency currents, which
was highly undesirable.

The equipment which was brought out in the Initial work
at College Park was an improvement over this Army apparatus. The
antennae were crossed at 90 degrees and each was fed by a 1 kilo-
watt power amplifier. The goniometer was placed in the circuit
between the antennae and the amplifiers, which were supplied by a
250 watt master oscillator. The primaries were of 32 turns of in-
sulated wire and the secondaries each of 8 turns of heavy litz
wire. The interlocking was all done on the low-power side to
obviate the necessity of breaking the R.F. currents as in the Army
system. Tone modulation was accomplished by exciting the trans-
mitting tube plates from a 500 cycle source.

Cam- operated relays excited the amplifier tubes with the
Morse "K" (-.) and "A" ( .-) , such that the dot of the "A" came

- 13 -

"between the dash and dot of the "IT" and vice versa. Hence a
single signal would be heard in the e qui signal zone. This made
it much easier to determine whether the signals were truly equal
in intensity.

The apparatus operated satisfactorily in flight tests
made over a one year period. However, the aural system of recep-
tion, improved to no matter what degree, has certain inherent
weaknesses. The pilot not only has to listen for the radio range
signals, "but also for the weather and other important information
"being continually broadcast to him. In addition he has the actual
operation of the plane to distract him from his radio duties.
Moreover, the purely human element involved in the proper inter-
pretation of the signals coupled with the distortion by various
kinds of disturbances renders accurate guidance extremely difficult,
Consequently the Bureau cast about for some better reception medium,

- X4 -

EARLY DOUBLE-MODULATION SYSTEM

The answer was at hand in the form of a method of visu-
al reception conceived by E. YJ. Dunmore of the Bureau in June
1926, and described by him in Confidential Bureau Laboratory
Report R-536-12aa. Although Mr. Dunmore did not at the time know
it, the system was patterned after one which had been granted an
early German patent. The idea was original with him, however.

Both antennae were to be supplied with R.F. power at the
same time, modulated by different audio-frequencies. The funda-
mental transmitting circuit as already described was utilized,
with a master oscillator supplying power to two amplifiers which
feed the antennae through a goniometer. This diffei-s from the
self-rectifying method of exciting the amplifier-tube plates as
already in use, in that the amplifiers were excited from a 500 and
a 700 cycle source respectively. The receiving set circuit was so
designed as to separate the modulation frequencies by use of
selector and rejector circuits as shown in Eig. 7. The outputs of
the 500 and 700 cycle transformers were rectified and made to buck
each other through a zero-center micro-ammeter. It can readily be
seen that the ammeter reading would be zero when on the course,
while the needle would vary to the right or left depending on which
side the plane was off course.

A modification of this basic system was also tried with
some degree of success, but neither the one nor the other answered
the basic requirement of simplicity and ruggedness. Moreover, were
the whole transmitting or receiving system dead, the micro-ammeter
reading would still be zero while the pilot might be far from his

- 15 -

course. While the fundamental plan of the double modulation
beacon was sound, it required further research.

Fig 7

S'-jS"tem

- 16 -
THE TUNED-REEI) COURSE INDICATOR

By this time work at the College Park station had "been
in progress for almost two years, and it was in 19S8 that Dunmore,
in collaboration with H. Pratt, R. R. Gessford, and D. 0. Lybrand
developed the now famous reed- indicator. 'While details as to the
developmental steps involved in its perfection are amply available ,
it will "be quite impossible to do more than touch upon these since
a thesis might he written on this piece of equipment alone.

ith the same general transmitting circuit as formerly
used, low frequencies were employed in the excitation of the power
amplifier plates. The receiving circuit output feeds two electro-
magnets which in turn operate two reeds placed in their field tuned
to the modulation frequencies. The reeds vihrate in a manner sim-
ilar to that employed in the frequency meters commonly met with in
electrical laboratories. The reed tips are white against a black
background for easy visibility. (See Fig. 8). When the plane is
on course the reeds vihrate equally and appear simply as two equal
white lines against the dark "background. To prevent confusion to
the pilot, the reed in the direction in which he is off course
lengthens, while the other shortens in length. The unit is en-
closed in a shock-proof mounting and placed in a prominent position
on the instrument hoard of the plane, so that an occasional glance
at the indicator suffices to keep the plane on the course. This
is in contrast to the old aural system which involved the continual
strain and tension of listening for signals whose distortion by
interference might lead to further difficulty. A further advantage

F.g.S

1=

RADIO BLAi^ON

RADIO BCACON

RADIO BEACON

UNEQUAL RELD AMPLITUDES
AIRPLANE. OFF COUROL. TO LE-FT

(OL)

EQUAL REJLD AMPLITUDES
AJRPLANEL ON GOUR3E-

Cb)

UNEQUAL REXD AMPLITUDES
AIRPLANE. OFF COURSE. TO RIGHT

(c)

- 1? -

is that the sharp tuning of the reeds outs down interference to
a minimum. As an illustration of the rugge&ness of the unit it
may "be said that in several airplane crashes it has "been the one
instrument on the "board to come through unscathed.

An ingenious plug- in arrangement was devised so that the
reversal of side of greater deflection resulting from flying to or
from the "beacon, might "be corrected. As now used, a simple opera-
tion keeps the unit constantly set so that the side of greater
deflection is the side off-course.

In the initial radio"beacon for reed operation the modula-
tion frequencies were obtained "by the use of tuning-forks actuated
"by electron- tubes. The tuning-fork outputs were amplified "before
"being used to modulate the 290 kilocycle current in the power
amplifiers. It is very important that the modulation frequency be
kept constant since the reeds are sharply tuned. In order that the
equi signal zone remain fixed it was necessary that the root mean
squared values of antenna current and the wave-forms thereof be the
same in both loops. Moreover, the percent modulation had to be
balanced in the two amplifier circuits. To prevent any coupling
between the amplifier circuits it was necessary to use a great
number of choke-coils and condensers. In order that all these re-
quirements be met, the apparatus needed was too complicated for
satisfaction.

It was realized that a circuit such as shown in Fig. 9
where the modulation is obtained from alternators, would eliminate
the control apparatus formerly necessary. The difficulty in putting
such a circuit into Immediate operation was that constant speed

TO 65 CYCLE- ALTERNATOR

MASTER. ,
OSCILLATOR.
100 WATTS '
890 k. c

INTERMEDIATE
AMPLIFIER.
100 WATTS *

FZ.F. CONTROL

INTERMEDIATE,
AMPLIFIER m
100WATTO

F"«'3 ^

POWELR.
AMPLIFIER,
]000 WATTS

3TATOR

POWER
AMPLIFIER.
1000 WATTS

-5TATOR.

-¥r-

J-^\MMjj

TO 86J CYCLE- ALTERNATOR

SCHEMATIC DIAGRAM OF DOUBLE MQDULATlOH BEACON

FOUR- COURSE. TYPE-

- 18 -

apparatus for driving the alternators was not available. Even
the slight variations in frequency of the utilities' supply
rendered the use of synchronous motors for the purpose impossible.
The great advantage of simplicity to te derived from the arrange-
ment led to further research in adaptation of alternators to modu-
lation frequency supply. A very slight "broadening of the resonant
point of the reeds solved the problem. Also, by coupling the two
alternators on the same shaft, the percent modulation changes
equally in case of speed change.

As now used, the two modulation frequencies are 65 and
86.7 cycles. These are supplied by 6 and 8 pole alternators
respectively. The speed must of course be 1300 revolutions per
minute. The synchronous speed most nearly approaching to this
value is 1800 revolutions. The reduction is accomplished by means
of a chain drive to prevent slippage. This arrangement allows of
3/10 of one percent variation in the bus frequency. In the un-
likely event that power can not be furnished within this tolerance,
a motor must be used which v/ill compensate for the change and main-
tain a more constant speed.

Such a motor was used satisfactorily at College Park in
the operation of the experimental radio range. It was designed
especially for the job by Leeds and Northrup. Part of the output
of an inverted rotary converter is impressed across a frequency
bridge. The bridge operates a galvanometer which in turn controls
a motor-driven rheostat in the converter's shunt field. This con-
trol is maintained through relays. As soon as the frequency of
the converter changes, the circuit operates to compensate the speed

- 19 -

in the right direction. In comparison with the complicated
circuits necessary in the tuning-fork modulated system, this
circuit is fairly simple.

- 20 -
AUTOMATIC VOIUKE CONTROL

flow that the reed-indicator was an established fact, an
incidental difficulty arose which required some research "by V. S,
Hinman, Jr., of the Bureau. It can readily he appreciated that,
since the amplitude of reed deflection is dependent upon the
power output of the receiving set, this amplitude is also a func-
tion of the distance from the radioheacon source. The excessive
vibration of the reeds, where close to the source, might cause
damage to the unit were the volume not controlled in some manner.

The author is not competent to discuss the rather com-
plicated radio circuits involved in the control system. The
general scheme, however, is easily understood. Part of the output
of the radio receiver is rectified "by a copper-oxide rectifier
(See Tig. 10). The resulting pulsating voltage is made as nearly
as possible a direct voltage hy the condensers and iron core in-
ductor of the circuit. Application of this control voltage is
made to the radio-frequency amplifier so as to decrease the
sensitivity of the radio receiver. The control voltage varies
directly with the receiver output. Hence, as the output voltage
tends to increase, the control voltage also tends to increase in
direct proportion thereto. This "bucks down the increasing output
and holds the receiver output constant for varying input pressure
values. Since the output remains constant, the field strength of
the electro-magnets operated thereby also remains the same for
varying distances to the radio range beacon.

A meter measuring the radio-frequency plate current, and
calibrated in miles, may be used to measure the distance of the
plane from the signal source. This may be done because of the

- 21 -

variation of this current with voltage input, which in turn is a
function of the distance.

Y/hen the distance "becomes small, the input voltage rises
and the plate current is low; the curve of voltage and plate cur-
rent tends to "become asymptotic to the voltage coordinate. A
lower current meter provided with a relay-operated shunt for use
at higher currents may be used. This becomes necessary "because
it is only at short distances, when the current is low, that the
meter really has any practical value.

Control

f?aJio Receive*"

Output

Vo l-tage

__!____+

Fig.lO. Au"toma."tic ConXt'ol Cn'r-coft

Resultant
Carrier

Fig. ll.

Heed Awi^li-tode /
due to t-d^, /

"Reecf AmJj/r-tuJ^
due to ij

Fig. 12,

/ deceived fauHcm
x Corre^iij + F.3. l|

Side 3a«d.

s 'de ^a.„^

o lojtion "troLtittri'ittiMg
to* out o+^^i*

Covf»«

"Reed A.«^tocj e
P«J« -to u).

Kee<A Am(?l,tode

F"ig. 14-
"Received ^t { „ c .

or-

BENT COURSES

The radio range as previously described made no provi-
sion for "bending the courses intersecting at the "beacon, to the
already fixed airways. In order that the device "be practical, it
was necessary that this added flexibility "be built into the system.
The work to this end was done on the aural system by F. G. Kear
and W. E. Jackson. Dr. Kear, now a lecturer in the Electrical
Engineering Department at the University of Maryland, formerly was
an assistant physicist at the Bureau of Standards. He was active-
ly engaged at that time in the College Parle experimental work.
Jackson is a radio engineer with the U. S. Bureau of Lighthouses.
These two men collaborated in a report on this subject, released
in August of 1929. The following month saw an adaptation of the
method to the double-modulation beacon by H. Diamond, a radio
engineer with the Bureau of Standards. The principal experiments
were carried on at College Park, but the actual installations
were made at Belief onte, Pa., and Kadley Field, IT. J.

The radiation from each of the loop antennae is at 290
kilocycles; modulated in one case to 86.7 cycles, and in the other
to 65 cycles. The radiation may in each case be broken up into
the two components of a carrier frequency and two side-band fre-
quencies. The carriers are in time phase and of the same frequency,
and hence may be combined. The side-bands cannot. The maximum
intensity of the combined carrier will be in the line bisecting
the angle between the antennae, which is 90 degrees. The non- com-
bining side-bands have their maximum intensity., of course, in the
plane of the producing antennae. (See Fig. 11). The vibration of

- 23 -

the reeds of the receiving set is produced "by the "beating of the
side-"bands with the carrier, giving the reception pattern shown
in Fig. 12. Hence there are only two courses produced, since the
courses at 90 degrees thereto are negligible. In small airports
it is conceivable that the two courses would "be sufficient, but
that these courses would have a 180 degree relationship is not
likely. Hence, some variation must be made in order that the
Deacon be useable.

If one of the in-phase carrier-frequency currents be
placed in time quadrature with the other, a revolving field may
he created, (See Fig. 13), since the loops are in space quadrature.
The reception pattern as set up may he seen in Fig. 14. Four .
courses are now obtained. By varying the current in one of the
antennae, or by utilizing an auxiliary vertical antenna coupled
to the amplifier output, these four courses may be set at arbi-
trary angles other than 90 degrees.

The system as just described theoretically, has been
found in practice to give patterns corresponding quite closely to
the theoretical. Several variations on this scheme have also been
tried with success.

- 24 -

MARKER BEACONS

To serve to indicate dangerous or important locations
on the beacon course, a marker beacon system was devised by the
engineers at College Park. The marker beacon simply operates an
auxiliary reed- indie at or on the instrument board of the plane.
The power of the beacon is low enough so that a signal can only
be received within a 3 or 3 minute time interval, as the plane
passes over.

■Two basic systems were tried. One involved the use of
an open antenna with a coded signal. The Department of Commerce
airway beacon lights are coded to indicate their geographical
position, and it was thought that a corresponding radio signal
might be sent out on the marker beacon. In the use of the open
antenna it is impossible to determine the exact beacon location
unless the plane flies directly over it, but the coding device
offsets to some extent this disadvantage.

The other basic system uses a loop antenna whose plane
is in the course of the main radio beacon. Consequently a mini-
mum, or theoretical zero zone exists at right angles to the main
course at the beacon, because of the figure-of-eight transmitting
characteristic. A coding arrangement is not practicable in this
hook-up, but the location of a line perpendicular to the course
at the beacon is accurately mapped out. Both systems have distinct
advantages. The choice of system in a particular case would be
governed by the conditions of that case.

The radio-frequency used at these beacons is the same as
of the main beacon, modulated at 40 cycles. At the outset it was

- 25 -

decided to use 60 cycle modulation in order that the supply De
obtained from the commercial 60 cycle mains, to cut down expense.
The result was that a 5 cycle flutter appeared in the 65 cycle
reed of the main indicator. This flutter was eliminated "by
changing to 40 cycles. The source is a 4 pole alternator driven
"by a 6 pole synchronous motor operating on the 60 cycle bus.
The reed resonance curve is sufficiently "broad so that minor fre-
quency variations of the commercial system have no effect.

- 26 -
SIMULTANEOUS RADIOTELEPHONE AND VISUAL BEACON

In order that the airplane requirements of small space
and light weight be met with, it "became necessary to design re-
ceiving and sending equipment which 'vould handle "both "beacon
signals and radiotelephone messages simultaneously. It is imper-
ative that the weather reports "be uninterruptedly received no
less than that the "beacon signals "be continuous. Moreover, a
system had to be designed such that a great deal of equipment al-
ready in use might be utilized. The work was done on this adapt-
ation by F. G. Kear and G. H. V/intermute at College lark, and re-
ported on in May of 1931.

The apparatus as boiled down in the abstract of that
report is about as follows:

A 3 kilowatt radiotelephone transmitter operates both
into a non-directive antenna circuit, and also into two loop
antennae through the proper amplifiers and a goniometer. The two
systems are disposed symmetrically with respect to each other, and
in so far as possible, coupling effects are balanced out. Proper
phase relations are preserved by use of a phase-shift unit. A
filter unit used on the receiving set keeps the various frequencies
in their proper circuit?.

Tests made with the equipment have shown it to be free
from interference effects, and in all ways very satisfactory.

It is of incidental Interest that the Bell Telephone
Laboratories collaborated with. Kear and Wintermute in this devel-
opment .

- 37 -
REED INDICATOR OP POINTER TYPE

Mr, Dunmore made a further development in connection
with, his reei indicator, to overcome certain disadvantages ob-
served in its operation. The new device used a zero-center meter
as, it will be recalled, did one of the first double-modulation
indicators described.

The osoillation of the reed is utilized to generate an
alternating current. This is rectified and used to actuate a
zero-center microammeter which gives the course indications. An
extra set of pic]c-up coils is used to create a field for the
generation of the electromotive force by the reeds. The two cur-
rents are made to pass through the meter with opposite polarities
so that a zero indication means on- course.

It will be recalled that a disadvantage inherent in the
early microammeter unit was that there was no indication in the
event that the system went dead, the meter continuing in the on-
course position. This bad feature is overcome in the reed con-
verter by use of a signal-volume indicator in the output of the
oxide rectifiers. It is a 0-500 microammeter connected so that it
is deflected in the same direction by both currents. An indica-
tion of faulty equipment is thus given to the pilot.

The reed converter tunes more sharply and is more adapt-
able than the reed indicator. On the other hand, it is more apt to
get out of adjustment; is heavier, bulkier and more expensive; and
does not operate as well under conditions of interference. Both
types are available at the present time, since they both have good
points in excess of their bad ones.

— 28 —
MISCELLANEOUS DEVELOPMENTS

Under this heading come several of the important pieces
of work, the detailed description of which would either involve
repetition of material already presented, or discussion far "beyond
the limitations of time and space.

THE TWELVE COURSE, THREE REED INDICATOR

First is the design "by Mr. Dunmore of a 12 course, 3 reed
indicator making use ofthe principles already noted, and applica-
ble to use at large airports. The device has the further advantage
that "by the use of 3 modulation frequencies it is possible to in-
terpret the signals to give information to the pilot as to whether
he is on or off course, and hy how many degrees; also, if off,
where the nearest course is, how to get on it, and which way he is
flying on it.

NIGHT EFFECTS

The elimination of an annoying phenomenon known as
"night effects" was accomplished quite late in the history of the
station. The effects were noticeable chiefly at night and in cold
weather, especially in mountainous terrain. Rapid and irregular
variations of the indicated course, exceeding 10 degrees in some
cases, constituted the difficulty. At any appreciable distance
from the beacon the device was rendered absolutely ineffective.
It was found that the night effects were caused entirely hy an-
tenna design, and could be eliminated by use of a so-called
"transmission line antenna", consisting of 4 vertical antennae

- 29 -

on the corner £3 of a square. The phenomenon was completely and
effectively eradicated.

ENGINE SHIELDING

Considerable attention was given to the problem of
properly shielding the ignition of the airplane engine, because
of the sensitivity of the receiving set used. All parts of the
ignition must be enclosed in metal of high conductivity. The
wires are covered with metal tubes, and metal shields are used
throughout. The equipment developed has been made commercially
available.

STATIC-IT COURSE- SHIFT INDICATOR

To provide a means for checking the accuracy of the
indicated courses at the beacon, a station course-shift indica-
tor was brought out which is accurate to within 0.1 of one
degree. The necessary adjustment of the transmitting system may
be made when this instrument shows any variation from standard.

COURSE AND QUADRANT IDENTIFICATION

Lastly, a system was provided so that a pilot near a
port need not be confused by the proximity of the converging
courses. Eefore this development was made, a great deal of diffi-
culty was experienced in orientation close to the beacon. The
difficulty was remedied by the use of coded signals.

- 30 -
CONCLUSION

With the advent of the Roosevelt administration in 1933,
all radiobeacon research was stopped. Orders to this effect were
received on Inauguration Day, March 4, 1933. Since that time the
College Park station has "been dismantled, and the "building: is now
used as the clubhouse for the Y/ashington Air Derby Club.

The developments made, owing to interdepartmental "bick-
erings, have largely fallen into disuse. The aural radio range
is the one now in use, the last visual beacon having been discon-
tinued the week of December 6, 1936. This is a most unfortunate
situation, and it is to be hoped that the Department of Commerce
will soon see fit to change its policy in this regard.

The author wishes to express his appreciation to the
men of the Bureau of Standards for the consideration and coopera-
tion shown him in the preparation of this paper. Photographs and
literature were made amply available, and a real personal interest
was shown.

- 31 -

BIBLIOGRAPHY

PERSONS INTERVIEWED

E. W, Dunmore, Senior Radio Engineer, Bureau of Standards

F. G. Kear, formerly Assistant Physicist, Bureau of Standards
J. H. Bellinger, Principal Physicist, Bureau of Standards.

REFERENCES

1. Development of the Visual Type Airway Radiol) eacon System -
Dellinger, Diamond & Dunmore.

2. A Directive Type of Radio Beacon and its Application to
Navigation - Engel & Dunmore.

3. Design of Tuned-Reed Course Indicators for Aircraft Radio-
"beacon - Dunmore.

4. Automatic Volume Control for Aircraft Radio Receivers -
Hlnman.

5. Applying the Radio Range to the Airways - Kear & Jaclcson.

6. Applying the Double-Modulation Type Radio Range to the Airways
Diamond.

V. A Simultaneous Radiotelephone and Visual Range Beacon for the
Airways - Kear & 'Vintermute.

8. A Course Indicator of Pointer Type for the Visual Radio Range-
Beacon System - Dunmore.

9. A Tuned-Reed Course Indicator for the 4 and 12 Course Aircraft
Radio Range - Dunmore.

10. The Cause and Elimination of Night Effects in Radio Range-
Beacon Reception - Diamond.

11. A Course-Shift Indicator for the Douhle-l'odulation Type Radio-
beacon - Diamond & Dunmore.

- 32 -

12, A Method of Providing Course and Quadrant Identification
with the Radio Range-Beacon System - Dunmore.