Full text of "The laboratory book of mineral oil testing"

See other formats

LIBRARY

UNIVERSITY- OF CALIFORNIA..

Class

THE LABORATORY BOOK

MINERAL OIL TESTING

NET BOOK.— This book is supplied to the
Trade on terms which will not allow of Discount
to the Public.

CHARLES GRIFFIN & CO., LTD.

THE LABORATORY BOOK

MINERAL OIL TESTING

Griffin's Standard Publications*

SECOND EDITION, Revised Throughout. Reset on Larger Page. With Valuable
Bibliography, New Maps, Illustrations, &c.

PETROLEUM AND ITS PRODUCTS.

A Practical Treatise.
By Sir BOVERTON REDWOOD, F.R.S.E., F.I.C., Assoc.R.C.S.

Hon. Corr. Mem. of the Imperial Russian Technical Society ; Mem. of the .American

Chemical Society ; Adviser to the Home Office and to the Corporation of

London under the Petroleum Acts, &c. &c.

SECOND EDITION. With Plates (one Coloured) and Illustrations.

A HANDBOOK ON PETROLEUM.

For Inspectors under the Petroleum Acts.
By Captain J. H. THOMSON,

H.M. Chief Inspector of Explosives.
And Sir BpVERTON REDWOOD.
" Reliable, indispensable, a brilliant contribution." — Petroleum.

In Large Crown 8vo, Cloth. Fully Illustrated. 53. net.

OIL FUEL:

Its Supply. Composition, and Application.
By SIDNEY H. NORTH.

"Every one interested in this important question will welcome Mr. North's excel-
Bent text-book." — Nature. ___^_

In Cloth Boards. Fully Illustrated, is. r.et.

THE PETROLEUM LAMP:

Its Choice and Use.

By Captain J. H. THOMSON,

And Sir BOVERTON REDWOOD.

" A work which will meet every purposs for which it has been written."

Petroleum.

SECOND EDITION. In Large 8vo. Fully Illustrated. 255. net.

OILS, FATS, BUTTERS, AND WAXES:

Their Preparation and Properties, and the Manufacture therefrom

of Candles, Soaps, and other Products.

ByC. ALDER WRIGHT, D.Sc., F.R.S.

Thoroughly Revised, Enlarged, and in Part Re-Written.
ByC. AIN8WORTH MITCHELL, M.A., F.I.C.

"Will be found absolutely indispensable." — Analyst.

SECOND EDITION. Large 8vo. Cloth. Fully Illustrated.

LUBRICATION AND LUBRICANTS.

A Treatise on the Theory and Practice of Lubrication and on the
Nature, Properties, and Testing of Lubricants.

By LEONARD ARCHBUTT, F.I.C., F.C.S.,
AndR. MOUNTFORT DEELEY, M.I.Mech.E., F.G.S.

" Destined to become a classic on the subject." — Industries and Iron,

In Large 8vo. Handsome Cloth. i6s. net.

FIRE AND EXPLOSION RISKS.
A Handbook of the Detection, Investigation, and Prevention of

Fires and Explosives.
By Dr. VON SCHWARTZ.

"A wealth of information oa the chemistry of fire and kindred topics."

Pire and Water.

LONDON : CHARLES GRIFFIN & Co. LTD., EXETER STREET, STRAND.

Abel Petroleum Testing Apparuti

THE LABORATORY BOOK

MINERAL OIL TESTING

JAS. A. HICKS

CHIEF CHEMIST TO SIR BOVERTON REDWOOD

WITH INTRODUCTION BY

SIR BOVERTON REDWOOD

(THIRTY-ONE ILLUSTRATIONS)

LONDON
CHARLES GRIFFIN AND CO. LIMITED

EXETER STREET, STRAND
1906

(s>

\\9

PEEFACE

I HAVE been requested to compile the following notes
on the commercial examination of mineral oil-products
as an outcome of a personal experience of some sixteen
years' continual work on the subject.

I have not endeavoured to write a text-book, but to
give in a concise form such details for the working of
apparatus special to the testing of petroleum and its
derivatives as will enable the analyst who may have
occasionally to examine such bodies, to proceed in a
satisfactory manner. At the same time it has seemed
well to include those ordinary physical tests which
require special modifications for this work.

JAS. A. HICKF.

May 1906.

1 58252

INTRODUCTION

BY SIR BOVERTON REDWOOD
D.SC. (HON.), F.R.S.E.

Adviser on Petroleum to the Admiralty and the Home Office.

THE great importance of standardising the methods
adopted in the examination of samples of commercial
products and in the reporting of the results is un-
questionable, for a large proportion of disputes between
sellers and buyers are directly traceable to misunder-
standings arising from differences of procedure.

To no branch of commerce is this remark more
applicable than to the mineral oil industry, and no
better illustration of it can be furnished than a reference
to the chaotic character of the statements commonly
made in respect of the viscosities of specified oils in the
early days of that industry.

Within recent years much has been done to place
the testing of mineral oil products upon a satisfactory
basis by the adoption of uniform methods, but there is
still, among some of those to whom this class of work
is only occasionally entrusted, a lack of knowledge of
certain of the instruments and processes which others,
whose avocation brings them, wider experience of the

x INTRODUCTION

requirements, have devised or selected as specially suited
to the purposes.

In these circumstances the publication of a manual
designed to convey the requisite information needs no
justification, and I do not know any one better qualified
to undertake the authorship of such a work than Mr.
J. A. Hicks, who has had daily experience of the various
operations in question for the past sixteen years.

Mr. Hicks has written an unassuming book which
cannot fail to prove a most valuable practical guide to
analytical chetrists and others who are called upon to
perform the ciass of work dealt with.

CONTENTS

PAGES

PRELIMINARY 1-2

CHAPTER I.
SPECIFIC GRAVITY.

Hydrometers — Specific gravity bottles — Sprengel

tube — Westphal balance — Sartorius balance 3-12

CHAPTER II.
FLASHING-POINT.

Abel petroleum tester — Petroleum Act, 1879 — Abel-
Pensky tester — Saybolt electric tester —
Tagliabue's pyrometer — Elliot tester — Foster
automatic tester — Granier tester — Pen sky-
Marten's tester — Gray's tester — Open test and
fire test . . . . , . .. . 13-32

CHAPTER III.
VISCOSITY.

Redwood viscometer — Engler viscometer — Engler-
Kunkler viscometer — Saybolt viscometer — Glass
jet viscometers — Coleman-Archbutt viscometer
— Thurston oil tester — Doolifctle torsion vis-
cometer 33-45

xii CONTENTS

CHAPTER IV.
COLOUR.

Wilson's chromometer — Lovibond's tintometer —

Stammer's colorimeter ..... 46-49

CHAPTER V.
SUNDRY APPARATUS.

Pressure of naphtha in closed vessels — Detection of
petroleum vapour — Capillary test — Melting-
point of paraffin scale and wax — Oil in scale-
Estimation of sulphur and water — Calorific
value . . 50_68

APPENDIX ... .... 70-71

... .... 73_74

PRELIMINARY.

THE commercial examination of Mineral Oil Products
may be said to divide itself roughly into that which is
made in connection with legal requirements as to storage
and so forth, and that which aims at satisfying the pur-
chaser as to the quality of the sample for practical use.

The principal tests employed are those for specific
gravity, for flashing-point, and, in the case of lubricating
oils, for viscosity. But further trials are made of some
commercial products manufactured for special purposes,
which may or may not require apparatus additional to
that usually found in the laboratory of the general
chemist.

The following table of the results commonly yielded
by the different products will be of uee as a guide to
what may be looked for : —

Product.

Sp. Gr.
at 60° F.

F. P.

(close)

Viscosity at 70° F.
(Redwood )

B. P.
°F.

K.O at

•F.

Sees.

60° F.

*= roo.

Pentane . .

•624--626

Below

77-99

•650 Gasoline .

•642--64S

o°F.

90-200

Petroleum

Ether . .

•63°-73°

80-300

•680 Spirit

(Petrol) .

•660-700

120-250

Benzoline

(Benzine) .

•690-720

„

130-350

Ligroine . .

715

190-250

Kerosene :

Colour.

American:

(Wilson).

Ordinary

798--802

73-80

2 '6-3-0

High Test .

785-792

IOO-IIO

17-20

MINERAL OIL TESTING

Product.

Sp. Gr.
at 60° F.

F. P.

(close)

°F.

Viscosity at 70° F.
(Redwood.)

B. P.

°F.

Sees.

R.O. at
60° F.

= IOO.

Kerosene .

Colour.

K u ss i ; i u :

(Wilson.)

Ordinary .

•S22--827

85-88

2'I-2'4

High Test .

•823

IOO

2'O-2'2

Mineral Sperm

•825

250

Pyronaphtha .

•865

250

Gas Oils :

American . .

•S55--865

100-150

(and

Russian (Solar

beyond).

oil) . . .

870-'88o

210-240

Lubricating

Oils :

Spindle :

@ 70° F.

American .

varies.

350-37°

180

Russian .

897-898

340-350

330-360

60-66

Engine :

American .

varies.

360-400

500-750

90-130

Russian .

•908 --909

380-400

1150-1300

210-240

Cylinder :

@ 200° F.

American .

varies.

varies

Russian .

9I4--9I5

430-440

i3-J4

Astatki :

@ 70° F.

Russian .

9H--9I3

310-340

1700-2300

310-430

M.P.

Paraffin :

(English).

Scale . .

110-125° ^-

Wax. .

125-130° F.

CHAPTER I.
SPECIFIC GRAVITY.

THE determination of the specific gravity of petroleum
products requires special care in recording the tem-
perature of working, owing to the high coefficient of
expansion possessed by these bodies. It is usually
taken at 60° F. ( 15*56° 0.), or is corrected to that
standard by the following coefficients.

Spirits lighter \ -00048 to -00040 for i° F. dependent
than Kerosene J on the volatility of the sample.
Kerosene, -00040.

Gas Oils, -00036.

Lubricating Oils, "00034.

These figures include the correction for the expansion
of the glass vessel used in the determination, and are
correct when used to adjust to the standard from the
temperature of determination ; but to correct the
specific gravity of the oil itself from one temperature
to another a coefficient O'OOOO I higher must in all cases
be taken.

Hydrometer. — For thin oils the hydrometer is
commonly considered accurate enough for most purposep
if it has a sufficiently open scale. A form such as tha<:
shown in Fig. I is an improvement on that commonly
employed. The bulk of the liquid necessitated by its
use (12 ounces) makes it advisable to stir well with the
thermometer before and after the reading is taken.
Such a hydrometer is conveniently made with a range
of O'O2, and this being distributed over a scale six

4 MINERAL OIL TESTING

inches in length makes readings possible to the fourth
place of decimals with fair accuracy. When first
employed it should be standardised with oils whose
specific gravity has been exactly determined in the
specific gravity bottle ; and these oils should be of such
densities as to afford readings generally distributed
over the whole range of the scale, for the
correctness of the instrument commonly
varies with different densities. The hydro-
meters may conveniently read from 780 to
•800, from -800 to '820, from '820 to '840,
from -840 to -860, and from '860 to -880,
respectively.

Baume and Twaddell Hydrometers,
— The scales of most hydrometers in use
in this country are marked in terms of
what is known as "absolute specific gravity,"
that is they show readings that compare
the weight of a unit volume of the sample
with that of a unit volume of water. Un-
fortunately, other scales are also in exist-
ence, notably those of Baume and Twaddell,
which are quite arbitrary in their divisions.
To convert Twaddell degrees to specific
gravity, multiply by 5 and deduct the pro-
duct from 1000. To find the specific gravity
equivalent of Baume degrees, add 130 and
divide 140 by the sum. The first of the
following tables gives the absolute specific
Hydrometer, gravity equivalents of the Baume degrees
required for oils and spirits, and was ob-
tained by the above formula.

The second table, which is in use in the United
States, will be seen to differ considerably from the first,
but it must be borne in mind that in some makes of the
Baume hydrometer the factor here given as 1 30 is as
high as 135.

FIG. i.—

SPECIFIC GRAVITY

TABLE I.

Baum£.

Sp Gr.

Eaume.

Sp. Gr.

Baum<5.

Sp. Gr.

I 'OOOO

•8187

•6931

•9929

•8139

•6897

•9859

•8092

•6863

•9790

•8046

•6829

•9722

•8000

•6796

•9655

7955

•6763

•95«9

•7910

•6731

•9524

•7865

•6699

•9460

•7821

•6667

•9396

5°

7777

•6635

'9333

7734

•6604

•9272

7692

•6573

•9211

7650

•6542

•9J5i

•7609

'6512

•9091

•7568

•6482

•9032

7527

•6452

•8974

7487

•6422

•8917

7447

•6392

•8861

7407

•6363

•8805

•7368

•8750

7330

•8696

•7292

•8642

7254

•8589

7217-

•8537

7180

•8485

7143

•8434

7107

•8383

"joj i

•8<33

7035

•8284

•7000

•8235

•6965

MINERAL OIL TESTING

TABLE II.

Baume.

Sp. Gr.

3aum£

Sp Gr.

Baum6.

Sp. Gr.

Baum£.

Sp. Gr.

I 'OOOO

3°

•8755

5°

7794

7025

•9930

•8702

7752

•6990

•9861

•8650

7711

•6956

•9791

•8597

7670

•6923

•9722

'8544

7628

•6887

•9658

•8492

7587

•6856

'9594

•8443

7546

•6823

'953°

•8395

7508

•6789

•9466

•8346

7470

•6756

•9402

•8299

7432

•6722

'9339

•8251

7394

•6689

•9280

8204

7357

•6656

•9222

"8157

7319

•6619

•9163

•8110

7281

•6583

•9105

•8063

721-3

•6547

•9047

•8017

7205

•6511

•8989

•7971

7168

•6481

•8930

7927

7133

'6451

•8872

•7883

7097

•6422

•8814

•7838

7061

•6363

Specific Gravity Bottle. — The ordinary Specific
Gravity Bottle with a drilled stopper, having a capacity
of 50 grammes, is the one generally em-
ployed where greater accuracy is required
than is possible with a hydrometer. But,
of course, where the sample is of less
volume than 5O-gramrne bottles of 10
grammes or 25 grammes can be adopted,
if a less accurate result is sufficient.

A narrow-stemmed thermometer,
marked in half degrees Fahr., is neces-
sary, and temperatures should be read to a quarter of
a degree F.

The bottle is filled with the sample in question, and
the thermometer inserted. When the air-babbles have
totally dispersed, and the temperature has become con-
stant (the latter must be ascertained by stirring with
the thermometer from time to tirne), the thermometer is

FIG. 2. — Specific
Gravity Bottle.

SPECIFIC GRAVITY 7

removed, the space left by it being filled up with a few
drops of oil from a small bulk which has been standing
in close proximity to the bottle.

The stopper is then put in and pushed well home
with a slight rotary motion, the excess of liquid which
flows through the drilled hole is carefully wiped away
with a piece of filter-paper, and the surface of the oil
is left exactly on a level with the top of the stopper.
From the time the thermometer is removed 'to this
point the bottle must not be handled in the slightest,
and the hands should not approach it to a greater
extent than is necessary, only the tips of the fingers
being used for operating. The bottle is now carefully
wiped with a soft cloth and weighed at once on a
delicate balance. The tare of the bottle and stopper,
which has been ascertained previously, is deducted
from the weight found, and the remainder is divided by
the water-contents of the bottle at 60° F., which has
been also very exactly determined beforehand. The
quotient is the specific gravity of the sample at the
temperature of working. The tare and water-contents
of the apparatus should be checked from time to time,
distilled water, of course, being employed.

In the case of Spirits it is essential that great care
be taken that none of the liquid is lost by the partial
evaporation of the portion which may expand out of
the bottle during wiping. This is best avoided by
adjusting in a beaker of water slightly warmer than
the atmosphere of the laboratory, so that after the top
of the stopper is wiped and the bottle is removed from
the bath the spirit contracts slightly down the drilled
hole, and the bottle can be manipulated safely. This
use of a water-bath also obviates the awkward variations
of temperature caused by the evaporation of spirit
which may have got on to the outside of the bottle.
With Lubricating Oils of moderate viscosity it is
sufficient to follow the above directions, but when
working with those of a thicker nature the sample
must be thoroughly warmed before it can be poured

8 MINERAL OIL TESTING

into the bottle, which, when full, should be placed in
an oven for some time to ensure the entire absence of
air. When this is certain, it is taken out and allowed
to cool slowly, and finally placed in a beaker of water
that also holds the thermometer. If this be not done
and the thermometer be put in the bottle in the usual
way, it is almost impossible to avoid the air, which
passes in when it is withdrawn, being trapped by the
oil and vitiating the result.

Some Cylinder Oils present a peculiar difficulty
when cooling, owing to the sample solidify-
ing in the neck before contraction has
ceased in the body of the bottle, and an air
passage is forced through the solid, leading
to a bubble in the interior. This can be
most satisfactorily avoided by cooling
slowly, and, on any signs of solidification
becoming apparent, by stirring the oil in
the neck with a warm wire until contrac-
tion is complete. With such samples very
great care must be taken when placing the
stopper in position, or the bottom of the
bottle will be forced out.

Regnault Bottle.— When the com-
Gravity mon specific gravity bottle contains very-
Bottle thick liquids and the drilled stopper is
inserted, danger of bursting is incurred.
With the Regnault Bottle this is avoided, as the
contents are adjusted to a mark on the neck.

The best method, of working, perhaps, with very
thick oils is to fill the bottle with a warmed portion
of the sample, and when cold to remove some of the
contents and thoroughly clean the neck to a point a
millimeter or so below the mark. Placing the bottle
then in a beaker of water and very gradually warm-
ing will bring the meniscus up to the line, and a
thermometer in the water will give the required
temperature. Very viscous oils frequently leave a
hollow core of air down the neck in cooling, and as

SPECIFIC GRAVITY 9

this often means the formation of an invisible air-
bubble, the oil in the neck must be kept warm until
the bottle has reached a stationary temperature. This
may be done by stirring with a warm wire. Removal
of the surplus oil can be conveniently accomplished
with a strip of stiff filter-paper, folded down the middle
and cut diagonally at one end, and the cleaning, with a
roll of the same.

The Regnault Bottle being entirely closed by its
stopper makes it very convenient for use with spirit, as
after the stopper is inserted no
evaporation can take place. In
such use the liquid may, of
course, be adjusted to the con-
tents mark directly, and with-
out cooling and re-warming as
with cylinder oils.

Sprengel Tube. — An in-
genious modification of the
bottle consists of a U-tube
with horizontal extremities of capillary tubing. It is
particularly useful for the determination of specific
gravities at elevated temperatures, as the
peculiar form enables it to be suspended in a bath,
leaving the ends projecting for adjustment.

To fill, remove the closing caps, replace one with the
filling tube, and connect a few inches of rubber tubing to
the other. Invert the U-tube, letting the filling tube pro-
ject downwards into the sample, which, if very viscous,
may have been warmed. Then exhausting the air by
suction of the rubber will cause the first limb of the
U-tube to be filled. As soon as this is complete turn the
tube right side up, without removing the filling tube
from the sample, but revolving it upon its ground-glass
connection. Proceed with the exhaustion until the oil
has completely filled the apparatus. Turn the filling
tube up so that it stands uppermost, immerse the U-
tube in a bath with a thermometer, and raise to the
required temperature. Leave the whole until the

FIG. 4. — Sprengel Tube.

MINERAL OIL TESTING

temperature is uniform, then remove the filling and
rubber tubes, also the surplus oil from the ends of the
capillaries, replace the caps and weigh after wiping. A
platinum wire hook will be found useful as a support on
the balance.

Westphal Balance. — The Westphal Balance is of
use for oils the viscosity of which is not sufficiently

FIG. 5. — Westphal's Specific Gravity Balance.

high to interfere with the free movement of the plummet
or float. Stirring with an accurate thermometer before
and after each reading and a careful observation of the
size of the different riders and their respective positions
on the beam, will give figures trustworthy to about two
or three in the fourth place of decimals. It is best to
disregard the thermometer contained in the plummet, as
the scale is not sufficiently open. The largest sized

SPECIFIC GRAVITY

rider represents 'I when hung on the notch marked
" i " ; when on notch " 2," -2 and so on. The other
riders represent -01, 'ooi, and -oooi according to their
size, and all are usually supplied in duplicate.

To adjust the balance, the foot is turned round until
the levelling screw is under the plummet end of the
beam, and with the plummet hanging in air the screw
is turned up or down until the needle-point on the

FIG. 6. — The Sartorius Specific Gravity Balance.

counterpoise bob vibrates to an equal extent on either
side of the corresponding point on the frame.

Sartorius Balance. — The Westphal Balance has
been modified by several designers. It will be necessary
here to notice only that of Sartorius.

The general notes as to the use of Westphal's
Balance apply, with the addition that in the ex-
amination of oils it will be, perhaps, advisable to dis-
card the double cylinder and to substitute one of the
usual shape. The readings obtained are more definite

12 MINERAL OIL TESTING

with this instrument than with the less expensive form,
owing partly to the use of a larger plummet. A
thermometer divided into half-degrees Fahrenheit should
be employed and read to quarter-degrees.

Very Small Samples. — The specific gravity of
very small samples can be ascertained with a fair
approach to accuracy by mixing alcohol and water to
such a density that a drop of the oil placed in the
mixture will show no tendency to sink or rise, and
by then taking the specific gravity of the mixture by
one of the above methods.

The addition of alcohol to water generates heat, and
time must accordingly be given for the temperature of
the mixture to approach that of the room after each
addition of either liquid. The temperature at which
the position of the drop is stable must be read, and the
density of the mixture ascertained at that temperature.

f UNIVERSITY )

CHAPTER II.
FLASHING-POINT.*

THE flashing -point, or flash-point, of an oil is the
empirical temperature at which it gives off sufficient
vapour to ignite momentarily
on the introduction of a flame
or spark, when the oil is heated
at a given rate, in an apparatus
of given construction and di-
mensions, and a defined igni-
ting agent is applied in a given
manner.

The " flashing-point " of an
oil will therefore vary with
the different instruments used
in its determination. The re-
sults yielded by these instru- M
ments may be divided into two [I
classes, " close " and "open,"
the former being given by the
forms of apparatus having a
lid to the testing-cup, and the
latter by those which have none.
To these two classes must be
added the " fire-test," which
differs from them by showing
the temperature at which the va-
pour is evolved so rapidly as to
continue burning when ignited. FlG< 7._Abel Petroleum

Abel Petroleum Tester. Tester.

— The legally recognised

*Kedwood, " Petroleum," ed. 1906, p. 545.
Thomson and Kedwood, " Handbook on Petroleum," ch. v. and vi.

14 MINERAL OIL TESTING

method of determining the flashing-point of an oil in
the United Kingdom is that designed by Sir Frederick
Abel and embodied in the Petroleum Act of 1879.
The manipulation of the Abel Tester is there set forth
as follows :

PETEOLEUM ACT, 1879.
(42 & 43 Viet. c. 47.)

Mode of Testing Petroleum so as to ascertain the Tem-
perature at which it will give off Inflammable Vapour.

The test apparatus should be placed for use in a
position where it is not exposed to currents of air
or draughts.

The heating vessel or water bath is filled by
pouring water into the funnel until it begins to
flow out at the spout of the vessel. The tempe-
rature of the water at the commencement of the
test is to be 130° F., and this is attained in the
first instance either by mixing hot and cold water
in the bath, or in a vessel from which the bath is
filled, until the thermometer which is provided for
testing the temperature of the water gives the
proper indication ; or by heating the water with
the spirit lamp (which is attached to the stand of
the apparatus) until the required temperature is
indicated.

If the water has been heated too highly, it is
easily reduced to 130° by pouring in cold water
little by little (to replace a portion of the warm
water) until the thermometer gives the proper
reading.

When a test has been completed this water bath
is again raised to 1 30° by placing the lamp under-
neath, and the result is readily obtained while the
petroleum cup is being emptied, cooled, and refilled
with a fresh sample to be tested. The lamp is then
turned on its swivel from under the apparatus,
and the next is proceeded with.

FLASHING-POINT 15

The test lamp is prepared for use by fitting it
with a piece of flat-plaited candle-wick,* and filling
it up with colza or rape oil up to the lower edge of
the opening of the spout or wick table.

The lamp is trimmed so that when lighted it
gives a flame of about o* 1 5 of an inch in diameter ;
and this size of flame, which is represented by the
projecting white lead on the cover of the oil-cup,
is readily maintained by simple manipulation from
time to time with a small wire trimmer.

When gas is available it may be conveniently
used in place of the little oil lamp, and for this
purpose a test-flame arrangement for use with gas
has been devised, which may be substituted for
the lamp.

The bath having been raised to the proper tem-
perature, the oil to be tested is introduced into the
petroleum cup, being poured in slowly until the
level of the liquid just reaches the point of the
gauge which is fixed in the cup. t In warm weather
the temperature of the room in which the samples
to be tested have been kept should be observed in
the first instance, and if it exceeds 65°, the samples
to be tested should be cooled down (to about 60°)
by immersing the bottles containing them in cold
water, or by any other convenient method.

The lid of the cup, with the slide closed, is then
put on, and the cup is placed into the bath or
heating vessel. The thermometer in the lid of the
cup has been adjusted so as to have its bulb just
immersed in the liquid, and its position is not under
any circumstances to be altered.

When the cup has been placed in a proper
position, the scale of the thermometer faces the
operator.

* The description of wick known as Field's night-light candle-wick
has been found most suitable.

f Great care must be taken to prevent the oil being splashed
against the sides of the cup and the formation of air-bubbles.

16 MINERAL OIL TESTING

The test lamp is then placed in position upon
the lid of the cnp, the lead line or pendulum,*
which has been fixed in a convenient position in
front of the operator is set in motion, and the
rise of the thermometer in the petroleum cup is
watched.

When the temperature has reached about 66°
the operation of testing is to be commenced, the
test name being applied once for every rise of one
degree in the following manner : —

The slide is slowly drawn open while the pendu-
lum performs three oscillations and is closed during
the fourth oscillation.

NOTE. — If it is desired to employ the test appa-
ratus to determine the flashing points of oils of
very low volatility, the mode of proceeding is to be
modified as follows : —

The air chamber which surrounds the cup is
filled with cold water to a depth of ij inches, and
the heating vessel or water bath is filled as usual,
but also with cold water.t The lamp is then placed
under the apparatus and kept there during the
entire operation. If a heavy oil is being dealt
with, the operation may be commenced with water
previously heated to 120°, instead of with cold
water.

In the Petroleum Bill of 1883 the following remarks
occur on the use of the pendulum : —

The first oscillation is from a to 5.
,, second „ ., „ b to a.
„ third „ „ „ a to &.

„ fourth „ „ „ b to a.

The opening of the slide commences the moment
the pendulum leaves position a in the first oscilla-

* Pendulum twenty-four inches long.

f A far preferable method, although not in accordance with the
terms of the Act, is to pour water into the air chamber to a depth
of ^inch, and to ina.inta.in the outer bath at 130° as usual.

FLASHING-POINT 1 7

tion and is steadily continued while it performs the
first, second, and third oscillations, so that the slide
is fully open when, in the third oscillation, the
pendulum has reached position b. The slide is
kept open for an instant and then quickly shut,
the moment of its being quite closed again being
coincident with the return of the pendulum to
position a at the end of the fourth oscillation.

A clock having a 24-inch pendulum is useful for
those who are continuously making the test.

In every case in which the test is repeated, a fresh
portion of the sample must be used.

Although in English law there is no allowance for the
barometric pressure prevailing at
the time the test is made, it must
be borne in mind that a difference
in the height of the barometric
column does effect a considerable
alteration in the result of the test.

The alteration thus brought about
is shown in the table on page 18,
which gives results as determined
by the Abel instrument.

The use of the Abel Tester with ?IGL 7a'

liquids containing solid matter in solution or sus-
pension (such as paints, rubber solution, or certain
classes of crudes) is misleading, as, owing to the
sluggish flow of the convection currents, the figure
recorded by the thermometer by no means represents
the temperature of the portion giving off vapour.
Several means have been suggested for overcoming
this difficulty, but, strictly speaking, none is at present
acknowledged by law in this country. The Inflammable
Liquids Bill of 1891 contained a provision for such
cases as rubber solution, in respect of which it was
directed that about a tablespoonful should be placed
in the cup, the air-bath filled to a depth of ij inches
with water, and the water-bath kept at 76° F. The

MINERAL OIL TESTING

^ w
o <&
K &

s ^

Q S
> «

H g
U O

O tf

o <:

FLASHING-POINT 19

tests were to be made every fifteen minutes. Redwood
suggests the addition of a thermometer with a small
cylindrical bulb fixed in a perpendicular position in the
cover of the oil-cup, so that the bulb is only a tenth of
an inch from the side of the cup. This has been adopted
by the Indian Government for use with Burma Crude.
The following Official Memorandum, although it deals
with the examination of metal-polishes especially, is of
the first importance in this connection : —

MEMORANDUM.
Petroleum Acts.

It has recently come to our knowledge that
attempts are being made by officers of local autho-
rities to apply the legal test for petroleum to
samples of liquid metal-polish, and we have satisfied
ourselves by experiment that the results thus
obtained are sometimes entirely misleading, as
they do not represent the temperature of the
portion of the liquid from \7hich inflammable
vapour is being evolved.

In carrying out the test prescribed by the Petro-
leum Act, 1879, the sample under examination is
slowly heated in a closed cup, and the temperature
is indicated by a thermometer the bulb of which
is immersed in the liquid in the centre of the
vessel. In these circumstances, the heat com-
municated to the sample through the walls of the
cup creates in such a liquid as petroleum convec-
tion currents, and through the circulation thus set
up the temperature of the contents of the cup is
equalised and the thermometer correctly indicates
the temperature at which inflammable vapour is
evolved by the liquid.

On the other hand, if the sample contains solid
matter in suspension, as is the case with the liquid
metal-polish in question, the formation of convec-
tion currents is interfered with, and the surface

20 MINERAL OIL TESTING

of the liquid from which inflammable vapour is
evolved acquires a higher temperature than that
of the portion in contact with the bulb of the
thermometer. Thus, as we have ascertained ex-
perimentally, the thermometer may indicate a
temperature of 59° F., when the temperature
of the surface is 83° F., and a sample may be
erroneously reported as having a flash-point below
the legal limit of 73° F., when the true flash-point
is far above the limit.

No doubt this would have been provided for
when the Act was passed if at that time the need
for applying the test to such substances had been
foreseen, but it was not until judgment in the case
of the " London County Council v. Holtzapfel's
Compositions Company (Limited)" was given in
1 899 that mixtures containing petroleum were held
to be petroleum within the meaning of the Acts.

In our handbook on Petroleum, published in
1901, we referred, on page 90, to the necessity for
a stirrer in the oil-cup when the test specified in
the Petroleum Act is employed for the testing of
paints and other substances containing petroleum,
and when opportunity occurs for a revision of the
law this addition will doubtless be legalised.

We are, however, of opinion that, in the mean-
time, authorities charged with the administration
of the Petroleum Acts should be made aware of
the circumstances we have referred to, and that
testing officers should take steps to ascertain the
true flash-point of samples of liquid metal-polish
or other substances which are not thoroughly
liquid, and, therefore, cannot be satisfactorily
tested in precise accordance with the directions
given in the schedule to the Act. In many in-
stances it may be possible to obtain a sample of the
petroleum used in the substance, or the solid matter
present in the sample can be removed by straining
or filtration, care being taken to avoid loss of the

. — Section of modified Abel Petroleum Tester
showing Plunging Agitator,

22 MINERAL OIL TESTING

more volatile constituents by evaporation, when
the separated liquid can be tested in the prescribed
manner. The liquid should not, however, be
separated by distillation, as this operation may
yield a distillate of lower flash-point than that of
the petroleum with which the mixture was made.
For guidance in determining whether there has
been any infraction of the law the sample may also
be tested in an apparatus provided with an efficient
stirrer. In any case of doubt as to the true flash-
point of the material, we would suggest that
reference should be made to his Majesty's Inspec-
tors of Explosives, at the Home Office, who will be
prepared to give advice as to ihe course which
should be adopted.

J. H. THOMSON, Captain, H. M. Chief
Inspector of Explosives.

BOVERTON KEDWOOD, Adviser on
Petroleum to the Home Office.

April 30, 1904.

With reference to the stirrer suggested in the pen-
ultimate sentence, it may be mentioned that, some
time ago, the writer designed a form of lid in which
the standard dimensions of the Abel Tester were
adhered to, but the position usually occupied by the
small ivory bead was taken by a plunging agitator, as
illustrated, and very satisfactory results have been
uniformly obtained with this modification. As it is
not possible to get this apparatus standardised by the
Board of Trade, each operator should himself test in it
samples of known flashing-point to see that it gives
correct results.

For the examination of samples that are insufficient
in bulk to fill the cup to the correct height, it is a
useful procedure to pour what there is into the cup and
then slowly to add water or mercury until the oil

FLASHING-POINT

(which of course remains on the surface) reaches the

required height.

Abel-Pensky.* — To avoid the personal error

likely to be introduced in the opening of the slide of

the Abel instrument, the German Government has

adopted a clock-work

movement for this

operation. The test-

flame is applied by

pressing the trigger,

and the mechanism is

rewound after each test

by turning the knob

(Fig. 9). As used in

Germany the Abel

Pensky gives a flash-

ing-point some 3° F.

higher than the Abel

as standardised by the

Board of Trade, but in

India it is adjusted to

give results in accord-

ance with that appa-

ratus.

Saybolt Electric

Tester. — The use of

this tester is confined

almost entirely to the

United States. It was

adopted in 1879 by the

New York Produce Ex-

change. The official directions for its use are as

follows : —

" Fill the metal-bath with water, leaving room
for displacement by the glass ^ cup. Heat Qthe
water until the bath thermometer indicates IOO F.,
at which point remove the lamp. Fill the glass cup

* Redwood, " Petroleum," p. 566.
Thomson and Redwood, " Handbook," p. 93.

Petrotoum

24 MINERAL OIL TESTING

with oil to the top line indicated by the rim surround-
ing cup, which is one-eighth of an inch below the
top edge of the cup. See that there is no oil on
the outside of the cup, nor upon the upper level
edge, using paper to clean cup in preference to
cotton or woollen material. See that the surface
of the oil is free from air-bubbles before first flash
is produced. Lift the cup steadily with left hand,

FIG. 10. — Saybolt Electric Tester.

and place in the bath. Suspend the thermometer
with the bulb of same immersed just from view
under the surface of oil. Adjust the flashing bar
and immerse the battery zincs in fluid. Try for
first flash every degree until the same is obtained.
Attain flash by producing spark with one stroke
of the key. The stroke on the key should be such
as in telegraphy is used to produce what is called
a dot, that is, a short, quick stroke. The first
flash produced from 110° test oil is generally
obtained when the temperature of the oil has
arrived at 90°. The temperature of the bath at
100° (as per note above) will carry the oil to about
90°, or, in other words, to about the first flashing-
point, without the aid of a lamp. When the

FLASHING-POINT 25

thermometer in the oil indicates 90°, introduce
lamp under the bath, and do not remove until the
operation is finished.

" The temperature of oil when placed in bath
should not be lower than 55°, nor higher than 70°
F. The flashing-bar must be free from oil before
adjusting for test. Draughts of air must be ex-
cluded from the apartment wherein tests are made.
Oil of 1 10° and upwards shall (after first flash) be
flashed at 95, 100, 104, 108, no, 112, 115. Oil
of 120° and upwards, after first flash, at 100. 105,
110,115,118,120,122, 125. Oil of 130° and up-
wards every 5° after first flash until burning-point."

It will be noticed that the results furnished are of the
" open-test " and " fire-test " order.

Tagliabue's " Pyrometer."* — Thisapparatns for
Open and Close Tests was patented in America in 1 862
and is used as follows : —

Kemove the cover, and take out the oil-cup. Fill
the water-bath with water to within 2 inches of the
top. Replace oil-cup and fill with the oil to be tested
to within f-inch from the top, then put on the cover,
and secure it in position by turning. Light the spirit-
lamp under the bath and remove it when the ther-
mometer reaches about 20° F. below the supposed
flash in or- point. When the lamp has been removed,
press down the brass knob on the top of the cover,
which will open the valves and admit air to the instru-
ment and the vapour to the dome. Insert a very small
lighted taper into the dome through the slot, and if the
flash-point has been reached a slight puff will occur.

Failing this, replace the lamp and warm at the rate
of 2° or 3° F. a minute, remove lamp, open valves, and
apply taper until the "puff" is obtained. The reading
of the thermometer at which this occurs is the flashing-
point of the oil.

* Kedwood, " Petroleum," p. 576.
Thomson and Redwood, " Handbook." p. 104,

26 MINERAL OIL TESTING

To ascertain the burning-point, replace the lamp,
and, when the temperature has risen another 8° F.,
remove it, swin^ back the cover by the handle, and
pass the lighted taper quickly across the oil ; if the
burning-point has been reached the oil will ignite. If
not, replace the cover and the lamp and continue
testing every 3° F. More than one testing of a sample
should be performed, fresh oil and cold water being
used, and all tests succeeding the first will probably
give lower results, as the instrument can be more care-
fully watched.

The results furnished by the Tagliabue are not
always satisfactorily concordant.

Elliot Tester.* — This is a modification of the
Wisconsin State Tester, from which it differs chiefly in
the substitution of a glass cover for a metal one. It was
suggested by Professor Arthur Elliot, and was adopted
by the State Board of Health of New York in 1882.

The Elliot Tester has a metal water-bath with a
capacity of 20 fluid ounces, and a copper test-cup
requiring 10 fluid ounces to fill it in the manner
prescribed.

Above the cup proper is a vapour-space of larger
diameter. The vapour-space is covered by a glass disc
through which passes a thermometer held by a cork,
and a hole f inch wide for the insertion of a gas-jet
-J inch long (a test-flame of burning waxed twine may
be employed). The oil-cup being removed, the water-
bath is filled to a mark and the cup is replaced. Oil is
poured into the cup to a point ^ inch below the flange
between the oil-chamber and the air-space, without
allowing the walls of the latter to become oily. The
glass cover is placed in position with the bulb of the
thermometer just covered by the oil. A small Bunsen
or spirit-lamp is lighted underneath, and the rate of
heating regulated to 2° F. per minute.

The flash -torch is introduced at every 2° rise, to

* Redwood " Petroleum," p. 577.

Thomson and Redwood, " Handbook," p. 107.

FLASHING-POINT 2 ;

about halfway between the oil and the cover, and this
is done from 85° to 95° F., when the lamp is removed
and the testing continued at every degree to 1 00° F.
After that point the lamp is replaced and the test
applied every 2° F.

The cover may be removed and the thermometer
suspended in the oil in order to ascertain the fire-test
of a sample. In this case the rate of heating is not to
exceed 10° F. a minute.

Foster Automatic Tester.* — In Ohio an appa-
ratus is used in which a wick dipping into the sample
gives a small flame constantly burning at an opening
indicated by the extinction of this flame by the slight
in the cover of the cup, and the flashing-point is
explosion occurring. The oil-cup is exactly filled to
a gauge-mark, the water-bath is half filled, the rate of
heating is 2° F. a minute, and the wick is lit at
100° F.

Granier Tester. f — This is another " automatic "
tester, on somewhat similar lines to the Foster, with the
addition that the oil is heated by a copper wire passing
from the test-flame into it.

With both these automatic testers the results obtained
are so unsatisfactory that lengthened directions for use
would not be warranted.

The foregoing instruments are of little use for tem-
peratures over 150° F. And, owing to the employ-
ment of soft solder in the construction of some of them, a
word of warning is necessary against their being heated
at all strongly. To the knowledge of the writer an
Abel cup has been wrecked in this way by more than
one careless operator.

The following are designed for use with heavier oils.

Pensky- Mar tens Tester .J — The apparatus most

* Thomson and Kedwood, " Handbook," p. 112.

f Ibid. p. 1 1 6.

j Redwood, "Petroleum," p. 593.

Thomson and Redwood, " Handbook," p. 124.

Archbuttand Deeley, "Lubrication and Lubricants," p. 179 (ed.
1900).

MINERAL OIL TESTING

generally employed for oils of high flash-point is the
Pensky-Martens instrument. The cup is of the same
dimensions as that of the Abel, and the lid has an

arrangement for the appli-
cation of the test-flame in
a similar manner to the
one adopted with that
instrument, but a modifi-
cation enables the test to
be applied by turning a
non-conducting button.
Before determining the
closed test, the cup and
lid with the attached stir-
rers are very thoroughly
cleansed from any oil re-
maining from a previous
sample. By forcing the
manipulating button up-
wards, the revolving plate
on the cover may be re-
moved, and the cleaning
facilitated. If the oil last
under examination was
of a much more volatile
nature than the sample
in question, it may be
well to reject the result
of the first test made
with the heavier oil, and
to use this experiment

as a means of completely freeing the cup from any
foreign vapour.

When thoroughly cleansed the cup must be filled
to the line inscribed, the lid put on so that it is
well " home," and the cup placed in the air-bath. The
heating, by means of a Bunsen under the air-bath for
at least 50° below the flashing-point, is to be at the
rate of 10° a minute, and the test is to be applied at

FIG. ii.— Pensky-Martens Tester

FLASHING-POINT 29

every 2° F. rise during the same time as the Abel
Tester. During the whole experiment the stirrers
must be kept revolving with a steady, continuous
motion, but during each actual application of the test-

FIG. 12.— Gray's Tester.

flame it is usual to cease agitation. Care should be
taken that in stirring none of the oil is thrown up on
to the lid of the cup, but no difficulty need be ex-
perienced here if the stirring be neither jerky nor too
rapid.

3°

MINERAL OIL TESTING

A nitrogen pressure thermometer, reading to
700° F., is useful for cylinder oils, but for other lubri-
cants an ordinary mercury thermometer reading to
550° F. usually suffices. The results furnished by the

Pensky-Martens Tester ap-
proximate very closely to
those given by the Abel
Tester at those tempera-
tures at which it is possible
to use either apparatus.

Gray's Tester.* — A
form of heavy-oil tester
closely resembling the Pen-
sky-Martens is the Gray's.

The chief variation con-
sists in the means for rota-
ting the stirring-vanes and
for applying the test-flame.
Both these operations are
performed from a non-con-
ducting button fixed on
the end of a horizontal shaft,
which also carries one of a
pair of bevelled wheels, and
is pierced by a short pin.
The other bevelled wheel is
fixed on the upper end of
the spindle to which the
vanes are attached, and by
turning the button slowly
by means of its handle the
oil is kept gently agitated.
In making a test, the
horizontal shaft is slid back

(a certain amount of "play" being allowed it by the
supports) until the pin engages with a projection

* Kedwood, " Petroleum," p. 595.
Thomson and Kedwood, " Handbook," p. 127.
Archbutt and Deeley, " Lubrication and Lubricants," p. 181

FIG. 1 3. — Section of Gray's
Tester.

FLASHING-POINT 31

attached to the sliding cover of the lid, then, by grasp-
ing the button itself and turning it firmly, the cover is
opened.

The same general remarks apply to the Gray's Tester
as to the Pensky-Martens, and the results furnished by
the Gray's should be concordant with those obtained
with the other apparatus.

Open-Test and Fire -Test. — These tests, in con-
junction with the Close-Test, give some idea as to the
homogeneous character of the sample under ex-
amination.

It is possible, for instance, to detect the presence of
a small percentage of a light oil in a lubricant by the
wide range covered by the three tests. Very often, in
such cases, if the proportion of light oil be not too
great, by allowing the cupful to cool after the fire-test
is taken and then repeating the whole experiment, a
very great difference between the first and second close
tests, a smaller difference between the open-tests, and
a practical agreement between the fire-tests will confirm
such an inference. But it must always be remembered
that this treatment will invariably cause some alteration
in the lower figures.

The usual way of taking the open- and fire-tests is to
continue the heating when the close-test has been
ascertained, and, removing the lid of the cup, to hang
the thermometer from a retort-stand with its bulb im-
mersed about a quarter of an inch below the surface of
the oil. At every 2° F. a test-flame of gas, T3T of an
inch in diameter, is passed across the surface of the oil
slightly below the level of the cup, taking care that it
does not come in contact with the oil. The heating
remains at the same rate of 10° F. a minute. That
point at which a flicker of flame covers the whole of the
surface is noted as the open-flash, and the fire-test is
the temperature at which the oil vapours continue to
burn until the next application of the flames is due.

The two higher tests may also be performed by
heating the oil in a porcelain crucible of about two

32 MINERAL OIL TESTING

inches diameter in a sand-bath. The crucible is filled
to about a quarter of an inch from the top, and the
level of the sand and oil should be in the same plane.

In the absence of coal-gas for the test-flame a useful
substitute can be provided by hydrogen or air which
has been passed through a Woulffs bottle containing
cotton wool saturated with light petroleum spirit.

CHAPTER III.
VISCOSITY.*

THE " body " or viscosity of an oil is most commonly
measured by its rate of flow through an orifice of
certain dimensions. The results obtained are, of course,
entirely comparative, and these are only of use as a
guide in the valuation of lubricants when compared
with those given by samples of the same nature and of
known value for any specific purpose.

Redwood Viscometer .f — The form of visco-
meter adopted almost universally in this country is
that known as the Redwood.

The instrument consists of a silvered brass oil-cylin-
der, furnished with an agate jet, and surrounded by a
copper bath. A copper tube, closed at the lower end,
projecting at an angle of 45° from the side of the bath
near the bottom, provides a means of heating the bath
liquid, and by the use of a revolving agitator, which
forms part of the apparatus, the heated liquid rising
from the copper tube can be uniformly distributed
through the bath. But for temperatures not far re-
moved from that of the room in which the work is
being done, constancy is most easily attained by the
addition of small quantities of water heated in a small
beaker.

The agitator carries a thermometer, to indicate the
temperature of the bath. The oil-cylinder is furnished

* Archbutt and Deeley, " Lubrication and Lubricants."

Redwood, "Petroleum," p. 597, 614.
f Redwood, J. Soc. Chem. Ind., March 1886, "Petroleum," p. 600.

OF THE

K nwiVFRSlTY

MINERAL OIL TESTING

with a stopper, consisting of a small brass sphere
attached to a wire, the sphere resting in a hemispherical
cavity in the agate jet. A short standard attached to

the oil-cylinder car-
ries a clip to support
a thermometer in the
oil. The diameter of
the stem of this ther-
mometer must be ap-
proximately a quar-
ter of an inch. Inside
the oil-cylinder, and
at a short distance
from the top, is fixed
a small bracket, ter-
minating in an up-
turned point, which
forms a gauge of the
height of the oil-level.
The instrument is
supported on atripod
stand provided with
levelling screws.

The bath is filled
with a suitable liquid
to a height roughly
corresponding with
the point of the
gauge in the oil-
cylinder. Water an-
swers well for tem-
peratures up to 200°
Fio. 14. — Eedwood Viscometer. J1., and for higher

temperatures a heavy

mineral oil may be used. The liquid having been
brought to the required temperature, the oil to be
tested, previously brought to the same temperature, is
poured into the oil-cylinder until the level of the liquid
just reaches the point of the gauge. A narrow-necked

VISCOSITY

flask, holding 50 c.c. to a point marked on the neck, is
placed beneath the jet in a vessel containing a liquid of
the same temperature as the oil. The ball-valve is
then raised, a stop-
watch at the same
time started, and the
number of seconds
occupied in the out-
flow of 50 c.c. noted.
It is of the greatest
importance that the
oil-cylinder should be
filled exactly to the
point of the gauge
after inserting the
thermometer, and
that the given tem-
perature should be
precisely maintained
during experiment, a
difference of J° F.
making an appreci-
able alteration in the
viscosity of some oils.
It is also essential
that the oil should be
quite free from dirt
or other suspended
matter, and from
globules of water, as
the jet may be par-
tially obstructed by
them. If the oil-cylin-
der requires to be wiped out, paper rather than cloth
should be employed, as filaments of the latter may be
left adhering. When oils are being tested at tem-
peratures much above that of the laboratory a gas
flame is applied to the copper heating-tube, and the
agitator kept in gentle motion throughout the experi-

FIG. 15. — Section of the Redwood
Viscometer.

36 MINERAL OIL TESTING

ment. The jet should be carefully examined before
the apparatus is used, and, if necessary, should be
cleansed by passing a piece of soft string through it.
The apparatus should be adjusted by means ot the
levelling screws, so that a spirit-level placed on the top
of the oil-cup shows it to be horizontal.

The viscosities at 70° and 140° F. are usually suffi-
cient for ordinary lubricating oils, and for cylinder oils
those at 200° and 250° F.

The results are generally stated in terms of Rape Oil
at 60° F., this being taken as equal to 100.

When the instrument was first designed, a large
number of rape oils were tested in it, and the average
of these tests was found to give 535 seconds.
Therefore the number of seconds obtained must be
multiplied by IOO and divided by 535*. In addition, a
correction is to be made for the influence of the different
specific gravities on the rate of outflow, and the figure
obtained from the division by 535 is therefore to be
multiplied by the specific gravity of the oil at the tem-
perature of the experiment and divided by -915 (Sp.
Gr. of Rape Oil at 60° = "915)^ The result will give
the viscosity in terms of Rape Oil at 60° F. multiplied
by IOO.

The Redwood viscometer requires about 6 fluid
ounces of oil for the performance of a test.

* The rape oil of the present day does not, as a rule, give a figure
so high as this, but, for the sake of uniformity, the original figure 535
is adhered to.

t The most rapid method of calculating the viscosity in terms of
rape oil is to multiply the number of seconds by the specific gravity
at the temperature of the experiment, and to divide the result by
489*525 (= 535 x '915). For the purpose of this division a copy of
the following table will be useful :

489-525 x i = 489*525

489-525 x 2 = 979 -050

489-525 x 3 = 1468-575

489-525 x 4 = 1958-100

489*525 x 5 = 2447-625

489-525 x 6 - 2937-150

489-525 x 7 = 3426-675

489-525 x 8 = 3916-200

489-525 x 9 = 4405725

VISCOSITY 37

Owing to the peculiar " lag" in the complete altera-
tion of viscosity shown by many viscous oils when
heated or cooled, the sample for examination should
be kept for twenty-four hours at the temperature of
working.

For very exact determinations it is recommended
that the test should be performed in a cupboard, at the
same temperature as the viscometer whenever this is
possible, thus overcoming the otherwise serious difficulty
introduced by the cooling of the lower part of the
agate. When the oil is very fluid this drawback is not
felt so much, as the rapidity of the stream keeps it at
a more uniform temperature. In those cases where it
is necessary to keep a burner under the heating arm
throughout the test, care should be taken that the hot
gases from the flame do not impinge on the under side
of the agate.

Some misunderstanding has been caused by the
mention of 25*5 seconds as the time taken for the out-
flow of 50 c.c. of distilled water at 60° F. It has been
found in the manufacture of the Kedwood Viscometer
that it is practically impossible to drill a jet which will
exactly give this figure with water, and which will at
the same time give a result with more viscous liquids,
agreeing with that obtained with the original apparatus.
So that, although 2*55 seconds was shown by distilled
water in the original, it is considered better to neglect
this when the viscometers are standardised, and to rely
only on the results yielded by different oils.

Engler Viscometer.* — In the En^ler instrument
the jet is a metal tube, and the water, or oil-bath, is
carried underneath the oil-cup with the object of
keeping the whole length of this tube at the same
temperature as the rest of the apparatus. The oil-
cup is covered by a loosely-fitting lid, and the jet is
closed by a plug of hard wood passing through this
lid, so that it can be opened without uncovering

* Redwood, " Petroleum," p. 602.
Archbutt and Deeley, " Lubrication and Lubricants," p. 144.

38 MINERAL OIL TESTING

the oil. A ring-burner under the bath serves foi1
heating.

To standardise, the water-bath is filled with water
and warmed to 20° 0., the oil-cup is filled with distilled

water to the height of the
three points, and is levelled
by placing pieces of card
under one or more of the
three feet until each point
is just immersed in the
liquid. The temperature
as indicated by both ther-
mometers is brought to
20° C. and the time noted
for the outflow of 200 c.c.
The experiment is com-
pleted three times, and the
average of the three re-
sults, which should not
differ by more than half a
second, is taken. If the ap-
paratus is correctly made,
the mean will be between
50 and 53 seconds. The
whole number nearest to
this average is taken as the
basis of future calculations.
The oil to be tested is
poured into the inner cup
until level with the points,
and stirred until the ther-
mometer indicates the re-
quired temperature. The lid is then put on and 200 c.c.
are run into the measuring flask.*

The test should be repeated, and two results should
not differ by more than one per cent. The resulting
average number of seconds is divided by the seconds

* The 240 c.c. mark on the measuring flask which is supplied,
shows the capacity of the oil-cup to the height of the points.

FIG. 1 6. — Erigler Viscometer.

VISCOSITY 39

taken by water, and the quotient gives what Engler
calls the " specific viscosity " of the sample at the
temperature of the experiment. The flask may be
placed in a bath at the same temperature as the oil-cup,
to avoid the error due to cooling.

The Engler Viscometer has been found in practice to
show roughly about 170 seconds to be equivalent to
loo seconds in the Redwood. 240 c.c., which the
Engler Viscometer requires for working, are equal to
nearly 8J fluid ounces.

Engler-Kunkler Viscometer.* — Owing to the
absence of stirrers in the Engler Viscometer variations
in temperature during the experiment are difficult to
avoid, particularly at high temperatures. To meet this
drawback, Engler and Kunkler have designed an
envelope of metal which covers the entire apparatus,
and can be heated by a flame placed underneath, thus
forming an oven, which is so constructed that the
experiment can be conducted from the exterior without
opening.

To operate, the flask is placed in the oven on the
stand below the plate holding the viscometer, then the
plate and viscometer are also put in position, and the
cover of the oven is firmly fixed. (The marks on the
plate, viscometer, and cover must agree in position with
that on the side of the oven.) Next, the thermometers
are inserted, the outer being on a level with the oil-cup
and the inner at the bottom of it, and the tube for filling
is put in so that it dips into the lip of the oil-cup. The
stirrer is lowered into the cup, and the whole apparatus
levelled by the plumb-line at the side. The oven is
heated until the outer thermometer registers the re-
quired temperature. While this temperature is being
reached, the can supplied with the apparatus is filled
with the sample and heated to half a degree or so above
the requisite temperature, the contents being stirred
meanwhile. The can being then full to the height of

* Redwood, "Petroleum," p. 605.

Archbutt and Deeley, " Lubrication and Lubricants," p. 145.

plug with-

40 MINERAL OIL TESTING

the indicator, the oil is quickly poured into the visco-
meter through the filling-tube until all the points in
the viscometer are just covered.

The stirrers are worked backwards and forwards, and
as soon as the temperature is constantly correct they

are lifted out of the oil
and the
drawn.

The hole in the
cover left by the plug
should be closed with a
cork.

Saybolt Visco-
meter.*— A form of
viscometer employed in
the United States is
that known as the
Saybolt. It consists of
a water-bath of large
capacity and an oil^cup
holding less than 'the
Kedwood or the Engler
instrument.

The jet is of metal.
As the volume of liquid
flowing out is measured
while in the oil-cup no
in a cool external measuring

FIG. 17. — Saybolt Viscometer.

error due to contraction
flask can arise.

In use, the outside bath is filled with water at the
required temperature (there is no arrangement for
heating), and the sample to be examined is poured into
the cup until it overflows into the gallery at the top, a
cork being inserted into the tube at the bottom of the
apparatus into which the jet delivers. The oil is stirred
with a thermometer, and when at the right temperature
the thermometer is removed, the displaced oil returning

* Eedwood, " Petroleum," p. 604.
Archbutt and Deeley, " Lubrication and Lubricants," p. 145.

VISCOSITY 41

to the cup from the gallery. All surplus oil iu the
gallery is then extracted with a pipette. The cork is
removed from the bottom tube at the same time as a
stop-watch is started, and the watch is stopped on the
surface of the oil appearing at the window of the
oil-cup.

An oil showing 100 seconds for 50 c.c. in the Red-
wood Viscometer should show roughly about 56 seconds
in the Saybolt when tested in the manner described.
The Saybolt Viscometer requires nearly three fluid
ounces for operation.

Glass Jet Viscometers. — The three viscometers
already described are the best-known forms of jet
viscometers which had their forerunners in ordinary
glass pipettes that were filled to a mark and emptied
to a mark. These primitive instruments were unsatis-
factory for many reasons, but especially because only
those results obtained with the same pipette were
comparable, and only very -rough attempts were made
to regulate their temperature ; moreover they were
liable to fracture.

Coleman-Archbutt Viscometer.* — The least
unsatisfactory form of glass viscometer, perhaps, is
Archbutt's improved form of the viscometer first
used by Coleman about 1869. It consists of a glass
pipette, contained in an outer water-jacket with a
funnel for pouring in hot or cold water, a tube for
running off water, and a stirrer for thoroughly mixing
the water in the jacket. The neck at the lower
end of the jacket is made narrow, a rubber stopper
not more than ^ in. thick is fixed in it, and the
jet of the efflux tube projects through the stopper
only y\ in., and does not extend quite to the end of the
neck. Thus the oil in the tube is surrounded by the
water in the jacket until it has reached nearly to the
end of the jet, and the temperature is maintained
constant until the oil has passed out of the tube. The

* Archbutt and Deeley, " Lubrication and Lubricants," p. 141,

42 MINERAL OIL TESTING

jet is protected from change of temperature as well as
from fracture by being contained entirely within the

FIG. 1 8. — Coleman-Archbutt Viscometer.

neck of the jacket. The efflux tube is narrowed for a
short distance above the jet, and four circumferential
marks are etched upon it. The lowest or zero mark is
rather above the middle of the narrow portion ; the

VISCOSITY 45

other three marks are on the wide portion, and divide
the tube above the zero mark into capacities of 25 c.c.,
50 c.c., and 100 c.c. The volume of oil used for a test
may therefore be varied according to the viscosity and
the quantity available. But the tube must be separately
standardised from each mark, as the times of efflux of
the different volumes bear no simple relation to each
other. The temperature is maintained constant by
pouring hot or cold water into the jacket, through the
funnel, and running off the excess as often as required,
using the stirrer frequently. The temperature is
indicated by a thermometer immersed in the water.
The oil, having been brought to about the desired
temperature, is poured into the efflux tube, where the
final adjustment is made by stirring with a thermometer,
which is removed before the oil is run out. The jet is
closed by a plug of soft wood, which is now removed
and replaced by the finger, and the level of the oil is
adjusted exactly to the mark- which it is desired to run
it from. It is then allowed to flow out, and the time
occupied in reaching the zero mark is measured by a
stop-watch, and compared with the time occupied by the
standard oil to flow out under exactly similar circum-
stances. When it is desired to make a determination
at 212° F., the temperature of the water is gradually
raised to about 180° F., by pouring in very hot water,
and then the funnel is removed and replaced by a tube
connected with the metal boiler, and stearn is blown
in until the water boils. A short bent glass tube is
provided for the escape of steam. The length of the
jet is about I in., and the diameter is such that 100
c.c. pure rape oil at 60° F. take about ten minutes to
flow out.

Thurston's Oil Tester.* — Efforts have been made
to reproduce in the laboratory the actual conditions
under which the lubricant is used in practice. But all
such methods of testing are of somewhat doubtful
value, inasmuch as the bearings are in such perfect
* Kedwood, "Petroleum," p. 621.

44 MINERAL OIL TESTING

condition that the requirements of practice are not
fulfilled.

Such a machine as that of Thurston, the best-known
of the type, requires power to drive it, and it must be
possible so to control the power used as to regulate the

FIG. 19.— Thurston's Oil Tester.

speed within any required limit. A measured quantity
of the oil is introduced into the journal and the load
under which the oil is to be used applied by means of
the spring in the pendulum. The spindle being
revolved at the required speed, the friction-reducing
power of the oil is inversely measured by the angle
through which the pendulum is moved. The rise of
temperature as indicated by the thermometer during
a given period affords a further indication of the value
of the sample.

VISCOSITY 45

Doolittle's Torsion Viscometer.* — Yet another
principle is embodied in the Torsion Viscometer, namely
the retarding effect of the oil between two surfaces,
one of which is fixed, the other having1 had given to it
a rhythmical motion. The needle attached to the wire
is set to the zero of the circular scale, and the wire is
twisted through exactly 360° and released. The reading
at the end of the first " swing " is noted, the next stop
is ignored, and the third stop again noted. The whole
operation is then repeated, reversing the direction of
the twist, and the average difference between the first
and the third stops gives the retardation due to the
" body " of the sample.

* Redwood, " Petroleum," p. 612.

CHAPTER IV.
COLOUR.

THE general appearance of mineral oils has a decided
influence on their sale, and although a light or dark
colour is to a great extent a matter of prejudice or
sentiment, the analyst is very often called on to state
the character of the sample in this respect.

The colour of burning oils is commercially taken
with the Wilson Chromometer and that of lubricants
with the Lovibond Tintometer.

Wilson's Chromometer.* — This instrument con-
sists of two similar tubes, 16 inches in length, closed at
each end by a screw cap carrying a stout glass disc.
Light is reflected upwards through the tubes by a
mirror, then twice reflected by two pairs of prisms, and
thus brought into an eyepiece. One of the tubes is
filled with the oil to be tested, and beneath the other,
which is empty, a disc of stained glass is placed.

On looking through the eyepiece, the field is seen to
be divided by a sharp line formed at the juncture of
the two pairs of prisms, the two halves of the field
being tinted respectively with the colour of the oil and
that of the Standard. An accurate comparison can
thus be made. The four Standards used are known as
" Water White," "Superfine White," " Prime White," and
" Standard White " ; and glasses of these degrees of
colour prepared from Mr. Robert Redwood's Standards
are provided with the apparatus. A third tube similar

* Redwood, " Petroleum," p. 544.

COLOUR

to the other two is useful in order to directly compare
two samples with each other, without making the tube
which holds the standard-glass oily.

Lovibond's Tintometer.* — The two-inch cell is
filled with the sample which has been cleared by slightly
warming (a high temperature will darken the oil), and
is placed in one side of the tintometer. Standard glasses

* Redwood, "Petroleum," p. 592.

MINERAL OIL TESTING

of Mr. Lovibond's series " 500 " are then put in the
slots provided in the other side until a match is obtained
with light reflected from the milk-glass reflector when
viewed from the eyepiece. It is advisable to repeat the
experiment by reversing the sides for the sample and

FIGS. 21 and 22. — Lovibond Tintometer.

glasses, and the two readings thus obtained should
agree to within one or two per cent.

Stammer's Colorimeter** — This is largely used
on the Continent for the colour-measurement of oils, and
has the advantage of showing from one standard the
exact length of the column of liquid that matches it.
The tube for the sample is depressed by means of the
thumbscrew to its lowest position and removed for
filling ; it is then replaced. The box containing the
prisms (similar to those of the Wilson and having the
same effect) is lifted off, the standard -glass is placed in
the cell in the left-hand tube, and the box is replaced.
On now looking through the eyepiece and adjusting
the mirror at the base, a similar view is obtained to that

* Redwood, " Petroleum," p. 544.

COLOUR

in the Wilson. The thumbscrew at the rear of the
apparatus is then turned until the two halves of the

FlG-. 23. — Stammer Colorimeter.

field appear of equal colour, and the reading on the
scale in millimetres is recorded as the colour of the
oil.

CHAPTER V.
SUNDRY APPARATUS.

Pressure of Naphtha in Closed Vessels.*—

The high co-efficient of expansion possessed by petro-
leum spirit and its high vapour-tension
may generate considerable pressure in its
containing vessel on a slight rise in tem-
perature. It is often necessary that an
idea should be formed as to the amount
of strain likely to be imposed on such
vessels by their contents in order that a
sufficient factor of safety may be allowed
in their strength. The method devised
for this purpose by Captain Thomson (his
Majesty's Chief Inspector of Explosives)
and Sir Boverton Redwood involves the
employment of a glass tube 6 inches long
by i inch in diameter joined at its upper
end to a short piece of J-inch tube, and
at its lower end to a capillary tube, which
is bent up to a height of some 30 inches.

On to the short tube is firmly wired a
short piece of rubber pressure-tubing,
which can be closed with a screw clip
close to the glass. Mercury is poured
into the i-inch tube up to the mark near
the bottom, and nine-tenths of the re-
maining space is filled with the sample
to be tested. (The upper mark shows the
correct height.)

The tube is then placed vertically into water main-
* Thomson and Eedwood, "Handbook," p. 130.

FIG. 24. — Ap-
paratus for
Testing the
pressure of
Petroleum
Vapour.

SUNDRY APPARATUS 51

tained at 50° F., and the level of the spirit exactly
adjusted when the temperature is constant. The
screw clip is then firmly closed, and the apparatus is
plunged into another bath of water which has been
previously heated to 100° F.* The highest position
attained by the mercury in the capillary tube is
marked, and its vertical distance above the lower mark
on the i -inch tube recorded. The temperature of the
bath is maintained at 100° F. for half an hour, and
then the height of the mercury is again noted. This
second reading should not exceed 24 inches if the
spirit is to be contained in the vessels recommended.

Detection of Petroleum Vapour.f — In view
of the enormous volume of mineral oil products
stored and conveyed every year, the number of men
engaged in handling it, and the costly nature of the
installations and vehicles employed, the use of every
possible precaution on the part of those responsible
becomes a primary duty.

In many cases this care by no means ceases when
the oil has passed out of hand, for with the lighter
liquids, naphtha and kerosene, their evolved vapours
remain for some considerable time after their removal,
so that before any fire can be allowed to approach the
empty spaces for cleaning, repairs or other purposes,
means must be taken to prove the perfect dispersion of
such gases.

Moreover, the same precautions are, of course, often
required in confined spaces adjacent to such liquids,
into which the vapours may have found their way.

The mode of testing most commonly adopted is that
known as the Clowes-Redwood system, and depends on
the occurrence of a " cap " of ghostly appearance over

* If the water be well above the top of the rubber tubing any
leakage will be visible.

f Clowes and Redwood, " Detection of Inflammable Vapour."
(Crosby Lockwood & Co.)

Kedwood, "Petroleum," p. 588.

Thomson and Redwood, "Handbook," p. 17.

52 MINERAL OIL TESTING

a non-luminous flame burning in the contaminated
atmosphere.

In a paper on the " Transport of Petroleum in Bulk,"

FIG. 25. — Apparatus for detecting Petroleum Vapour.

read before the Institute of Civil Engineers (Proceed-
ings, OX VI. (1893-4), Part II.), SirBoverton Eedwood
thus described the working of this test : —

" In the use of the apparatus the first step is to
connect the hydrogen-cylinder with the lamp,
taking care that the unions are screwed up gas-
tight. The sliding chimney of the lamp being
raised about half-way, the gas is then cautiously

SUNDRY APPARATUS

turned on at the cylinder, the regulating valve on
the lamp being left open, and a light is applied to
the hydrogen jet. The valve on the hydrogen

FIG. 26. — Apparatus for detecting Petroleum Vapour.

cylinder is then adjusted so as to give a flame
rather more than 10 millimetres (04 inch) in
length, and the lamp-chimney pushed down until

54 MINERAL OIL TESTING

there is an opening of only about a quarter of an
inch in height at the bottom. This opening is left
for the supply of air to the hydrogen flame during
the few minutes occupied in the warming of the
chimney. As soon as the moisture which at first
condensed upon the cold glass has evaporated the
lamp is ready for use, and, assuming the collecting
vessel to have been already charged with the
sample to be tested, and connected with the lamp,
all that remains is for the observer to completely
close the sliding chimney of the lamp, adjust the
hydrogen flame by means of the regulating-valve
on the lamp, so that the tip of the flame is only
just hidden when the eye of the observer is on a
level with the bottom of the window, place his
head under a cloth such as used by photographers,
so as to exclude light, and, as soon as his eyes
have become sufficiently sensitive, turn on the tap
of the collecting cylinder, and carefully observe
what takes place in the lamp chimney.

" The tap may at once be turned on fully, as the
construction of the outlet and inlet orifices pre-
vents the sudden rushing out of the contents of
the cylinder, and the sample will be gradually
delivered into the test-lamp during a period of
more than two minutes, which is ample time for
noting the effect. The rate of delivery is, of
course, a gradually diminishing one, but this is
not found to be attended with any inconvenience,
the conditions being the same in each experiment.
" In this way a proportion of vapour, considerably
below that which is required even for the pro-
duction of an inflammable mixture, and still lower
than that which is needed to give an explosive
atmosphere, may be detected by the formation of
a flame-cap of greyish blue colour, which, though
faint, is easily seen, especially after a little
practice.

" With an increase in the quantity of vapour,

SUNDRY APPARATUS

56 MINERAL OIL TESTING

the flame-cap first becomes much better defined,
though it is not greatly augmented in size, and
then considerable enlargement of the cap occurs,
this condition being arrived at before the atmo-
sphere becomes inflammable.

" In taking a sample of the air in a tank the
collecting vessel may be used in the tank if the
proportion of vapour present is known to be small ;
but even in such cases it is better to employ a
short suction-tube, the open end of which can be
placed at the lowest point in the tank, where most
vapour would probably be found. If, on the
other hand, the atmosphere of the tank is sus-
pected to contain so much vapour that there would
be danger of its producing insensibility when taken
into the lungs, and especially if the compartment
is entered through a small man-hole, it would
obviously be most improper that anyone should be
sent into the tank, and in that case the sample
should be taken by the use of a long suction-tube
reaching to the bottom."

If a sample of atmosphere gives no cap at all with
this apparatus, absolute security may be felt in giving
permission for work with naked lights in such air. It
is important that the operator should assure himself
that the connections from the pump to the interior of
the lamp are quite free ; also that all the baffles, &c.,
be periodically taken out and dust removed, as the test
is rendered worse than useless unless a free current of
the atmosphere passes through the lamp.

Capillary Test.* — A method of measuring the
capillary power of kerosenes and of lamp-wicks has
been designed Toy Mr. Eobert Redwood.

A trough of considerable capacity is supported in a
cupboard, which must be maintained at a uniform
temperature, at a height which permits of the shorter

* Redwood, "Petroleum," p. 542.

SUNDRY APPARATUS 57

ends of the bent wicks dipping into it and the longer
ends hanging outside over small beakers. The oil in
the trough is thus allowed to ayphon over, and the
relative rate of flow gives the measure required. The
results are comparative only.

For the testing of wicks the procedure is as

FIG. 28. — Redwood Apparatus for Capillary Test.

follows : Fill the trough to within a quarter of an inch of
the top and bring the cupboard and oil to 70° F. Cut the
wicks to be compared to exactly the same length (say
7 inches), and draw across them a light line in ink at
the same distance from one end (say 3 inches). Then,
having soaked them in the oil to be used and having
removed all surplus oil by slight pressure, hang them
on the highest of the three glass rods provided, and
pass their ends inside the lower rods, adjusting the
ink marks so that they exactly correspond with the
top of the upper rod.

Take as many small beakers as there are wicks to be

58 MINERAL OIL TESTING

examined, and place them in front of the trough ready
for use. Then put the wooden holders of the glass
rods on the ledges at the end of the trough, allowing
the short legs of the wick-syphons to dip into the
oil. Note the exact time that the first drop falls
from each wick, and remove each beaker at the end
of thirty minutes from the time the first drop fell into
it. The several weights of oil the beakers contain will
then be a relative measure of the oil-passing capacity
of each wick.

For testing oils, a trough divided into compart-
ments must be used, and wicks of uniform character
must be employed. In order to eliminate errors likely
to be introduced by unavoidable differences in the wicks,
the experiment may be repeated several times, using
fresh wicks, and the average results given.

Melting-Point of Paraffin Scale and Wax.—
The melting-point of the solid products of crude petro-
leum is taken commercially by two methods, known as
the " English " and "American " tests.

In performing the first, the scale or wax is put into
a test-tube 4 inches long by J inch in diameter and
• melted in a water-bath. When complete liquefaction
has taken place the tube is removed and its contents
continually stirred with a delicate thermometer, the
bulb of which is well covered by the sample. It is
important that this stirring should be confined to the
centre ot the tube as much as possible, as contact with
the walls of the tube may chill the bulb below the true
temperature of the liquid. The fall in the mercury is
arrested at about the same time that the liquid becomes
cloudy, and an exact reading (to the nearest quarter of
a degree F.) is taken at the first check in the cooling
and recorded as the English Melting-point.

The American test usually gives results 2\° — 2j° F.
higher than the English. The melted sample is
poured into a warm metal dish hemispherical in
shape and of 3f inches diameter. This is placed on
a stand and a round-bulb thermometer (bulb | inch

SUNDRY APPARATUS 59

in diameter), supported in it with three-quarters of
its bulb immersed, the whole being placed within a
glass screen to avoid draughts. On carefully watching
the sample, patches of film will be presently observed
on its surface, and the temperature shown by the
thermometer is noted at which theso patches touch the
bulb.

FIG. 29. — Hydraulic Press.

It will be observed that both the methods in general
use give, really, the solidifying-point rather than the
melting-point.

Oil in Scale.* — Five hundred grains of the
powdered sample are weighed into a glass dish and
transferred to a press. On each side of the scale is a
circular calico press-cloth, and outside the cloths
enough circular pieces of blotting-paper to absorb the
oil expressed.

The diameter of the press-cup is 5| inches, and a
total weight of 9 tons is applied for five minutes. The
press may be of hydraulic power such as that figured.

* Redwood, "Petroleum," p. 631.

60 MINERAL OIL TESTING

On the expiration of the time mentioned, the press
cake is again weighed together with the glass dish,
and the carefully removed scrapings from the cloths.
The loss in grains divided by five represents the oil
percentage of the sample. The press and scale must
be kept at 60° F. for the time of the experiment.

Estimation of Sulphur. — After numerous ex-
periments carried out under the instructions of Sir
Boverton Redwood with very varied methods it is
clearly evident that while one method may be suited to
one class of oil and another to another class, the Bomb-
Calorimeter is the most trustworthy apparatus for
employment with all oils. Full details as to the use
of this instrument are given under the head of
Calorific Value (p. 65).

Estimation of Water.— Oils that lose an
inappreciable percentage on being heated to 230° F.
may have their water contents estimated by weighing,
cay 25 grammes, into a glass dish, heating to that
temperature on a sand-bath, and stirring continuously
with a thermometer, until bubbles of steam cease to
form. The sample is then allowed to cool, is re-weighed,
and the loss of weight in grammes multiplied by four
represents the percentage of water.

In oils having more volatile constituents the water is
best determined by subsidence in a Sutherland bulb
(see Fig. 30).

The sample is weighed into the bulb, and the stopper
tied in, and covered with a piece of rubber tissue to
avoid the entrance of condensed steam. The bulb is
then placed in a bath kept at about 180° F. until no
more water settles out. Oils of high viscosity may be
diluted with kerosene to hasten the subsidence.

Calorific Value. — The question of .the thermal
efficiency of oils is occupying the attention of the
petroleum specialist to an increasing extent, in view
of the steadily extending use of liquid-fuel for in-
dustrial purposes. Th« apparatus yielding the most
satisfactory results with a minimum of trouble is

SUND/Y ^AI^|S^^6\ 6 1

undoubtedly some form of "bomb calorimeter " such as
that employed by M. Berthelot. Tk^high cost of this
particular instrument, owing to the amount of platinum
used in its construction, often places it beyond the
reach of the chemist, but a very good substitute has
been designed by Mahler, and may be safely recom-
mended.

The folio wing free translation from the French of the
pamphlet supplied by the makers gives
full guidance as to its manipulation.

The calorimeter is essentially com-
posed of a bomb B, a calorimeter D, an
insulating jacket A, and an agitator S.
The bomb has a capacity of about
650 c.c., and the walls are 8 m.m.
thick. This capacity assures complete
combustion of the sample by providing
a decided excess of oxygen.

The bomb, made of specially forged
steel, is nickel-plated outside, and has
an inner coating of enamel* to resist
the corrosive action of the nitric acid
which always forms during the com- pIG> ^a
bustion. The bomb is closed by a screw Sutherland Bulb,
lid fitting tightly on a lead washer.
The cover carries a screw valve, which permits the
introduction of oxygen. j* Through the cover (or lid)
a platinum elecurode E (well insulated) is carried into
the interior of the bomb.

Another rod of platinum is also fixed, supporting the
capsule plate E, where the sample to be tested is
placed.

The sample is ignited by contact with the iron wire

* The specific heat of steel is 0*1150; this specific heat has been
determined by M. Matignon, at the College of France (laboratory of
M. Berthelot). The specific heat of enamel is 0*2045, also found by
M. Matignon.

f The small nut on the top of the milled head is provided to keep
in position the screw valve, when the tube is screwed tight, by means
of a ipanner.

62 MINERAL OIL TESTING

spiral F (of known weight), which is connected at the
desired moment to an electric current of about 12
volts and 2 amperes. A simple mechanical combina-
tion works the helicoidal agitator, and allows the
operator to give a regular stirring movement without
trouble. On the right side of the stand carrying the
manometer is placed a screw valve, which can easily
be adjusted to allow the slow introduction of the oxygen
into the bomb. This valve avoids the necessity of using
the valve on the oxygen cylinder, and is much more
easily adjustable to give a slow supply of oxygen.
M. Mahler uses the cylinders of compressed oxygen as
figured, and as a suitable pressure for the combustion of
i gramme of oil is 25 atmospheres, a cylinder contain-
ing 1000 litres of gas at 120 atmospheres would be
sufficient for about 60 experiments.

Weigh one gramme of the substance to be tested
into the capsule (7, adjust the iron wire spiral used for
igniting the sample. After the capsule has been firmly
fixed on its supporting rod, screw on the lid of the
bomb very firmly by gripping the latter in the vice
provided and using the large spanner. Connect the
screw valve of the bomb to the tube leading from the
pressure gauge (manometer), taking care to have the
screw tap (or valve) on the manometer stand closed.
Now open the tap of the oxygen cylinder 0, and then
very gently open the screw tap on the manometer
stand, and allow the gas to flow until it registers a
pressure of 25 atmospheres on the manometer. Now
close the manometer screw tap, the screw tap of oxygen
cylinder, also the screw tap on the cover of the bomb ;
and disconnect the tube from the bomb. It is recom-
mended that the substance to be tested should not be
in too fine a powder, as the current of gas entering
may blow it out of the capsule. To avoid this care
should be taken to allow the oxygen to enter the bomb
slowly.

Place the bomb in the calorimeter A, then put the
thermometer T, and the agitator S in position, and

SUNDRY APPARATUS 63

pour in the water which has been previously measured.
Leave the apparatus so assembled for a few minutes,

stirring at intervals so as to ensure a somewhat equable
temperature before commencing observations. Then,
when all is ready, the stirring is performed con-

64 MINERAL OIL TESTING

tinuously, and a careful note taken of the temperature
every minute for about five minutes, so as to fix the
law which the thermometer follows before the sub-
stance is ignited. Now ignite the substance by con-
necting one wire of battery to the platinum electrode E,
and the other wire to some part of the screw tap
Take note of the temperature half a minute after the
ignition, also at the end of this minute and continue to
take the thermometer readings every minute until they
commence to show a regular fall in temperature. This
is the maximum temperature to be noted. Continue
the readings of thermometer for another five minutes,
so as to fix the law the thermometer follows after the
maximum temperature has been reached.

The following are the rules for correction of calcula-
tion : (i) The law of the decrease of temperature
observed, after the maximum, represents the loss of
heat of the calorimeter before the maximum for
any one minute, on the condition that the average
temperature of this minute does not differ by more
than one degree from the maximum temperature.
(2) If the temperature of the examined period differs
by more than one degree, but by less than two degrees
from that of the maximum, the figures which represent
the law of decrease at the moment of the maximum,
diminished by 0.005, give the correction required. The
two preceding remarks suffice in all cases. Moreover, it
may be taken that without altering the precision of
the operation, the law of variation followed during the
first half of the minute during which the ignition took
place is that which existed at the beginning of that
minute. During the whole time of the observation the
operator ought to take care to see that the agitator
acts regularly.

When the observation is finished, first open the
screw tap, then the bomb itself. Now wash the in-
terior of the bomb with a little water, so as to collect
the acid formed during the explosion. The nitric acid
is estimated by titration with alkali, and now we are

SUNDRY APPARATUS 65

in possession of all the elements for the calculation, for
since the calorific power Q is the whole,

Q = A (P + P1) — (0-23 »? + i-6/)

A = the difference of the corrected temperature.

P = the weight of water in the calorimeter.

P' — the equivalent in water of the bomb and its accessories.

j> = the weight of the nitric acid (HNO3) found.

p' = the weight of the small iron spiral.

0-23 = the heat caused by the formation of one gramme diluted

nitric acid,

i '6 = the heat caused by the combustion of one gramme of iron.

If it is a question of testing oil, no notice need be
taken of the small quantity of sulphuric acid which
results from the oxidisation of the sulphur of the
sample, and which will be estimated in the titration, as
nitric acid. The error is in reality negligible in a
commercial estimation. But one will notice that the
bomb gives a means of estimating the sulphur, which
is entirely oxidised and transformed into sulphuric
acid.* In this case it is better to burn two grammes
under 30 atmospheres ; it is unnecessary to take note
of thermometer reading. In the case of a trial with a
-substance with little hydrogen in it, coke for example,
:so little water is formed that the quantity is insufficient
to dissolve the acids. It is then necessary to put at the
bottom of the bomb a few c.c. of water, which one
must take into account in the calculation.

Proceed just the same for a liquid as for a solid.
Bat if the liquid emits appreciable vapours at ordinary
temperatures it is better to weigh the amount taken in
a thin capsule with slender points, through which,
passes the fuse of iron wire. When the capsule is in-
troduced into the bomb, care must be taken to break
these points, so as to allow the oxygen to make contact
with the liquid.

The method of finding the calorific power of gases
is as follows. Carefully measure the empty space in

* One can calculate the heat given out by the formation of the
•diluted sulphuric acid H.,SO4, that is to say Ocal73 per gramme of
HjSO,

66 MINERAL OIL TESTING

the bomb, fill it first of all with gas, now empty, and
definitely introduce the gas under atmospheric pres-
sure at the temperature of the laboratory, then add the
oxygen, and proceed the same as for solids or liquids.

The determination of the calorific power of gases
offers a peculiar difficulty. Care must be taken not to
mix the gas with such a quantity of oxygen that the
mixture ceases to be explosive. For ordinary lighting
gas, 5 atmospheres of oxygen are sufficient. For the
gas of Siemens' generator not more than half an atmo-
sphere, measured by the mercury manometer, must
be taken.

Determination of the Equivalent in Water of the
System. — To find the term of correction representing
the exact equivalent P of the system in water the most
simple 'way is to make a double experiment as follows : —
Burn in the bomb a known weight, one gramme, of
naphthalene for example, with 2300 grammes of water
in the calorimeter. Next burn one gramme of naph-
thalene with only 2100 grammes of water in the
calorimeter. We have, then, two equations which
between them eliminate the calorific value of the
naphthalene, and the water equivalent can be deduced
from the difference in the two sets of figures. Care
must be taken to weigh the naphthalene after it is half
melted, otherwise, if it were not fused, some of it would
be scattered by the current of oxygen, and would not
be burnt.

Example of the Determination of Calorific Value. —
The combustible was a sample of colza oil : its analysis
gave : — Carbon, 77' 182% ; hydrogen, 11711%; oxygen
and nitrogen, 11*107%. The weight taken was one
gramme. The calorimeter contained 2200 grammes
water. The equivalent in water of the bomb and its
accessories was 48 1 grammes.*

The apparatus having been assembled as described,
a few minutes were allowed to elapse, so as to establish

* The equivalent in water, 481 grammes, was found by a special
method giving directly the exact heat value of the system.

SUNDRY APPARATUS 67

a uniform temperature, the stop-watch was started,
and the thermometer readings recorded as below : —

Preliminary Period.

0 minutes ..... 10-23°

1 „ 10-23°

2 „ 10-24°

3 • 10-24°

4 10-25°

5 „ 10-25°

A, = 10-25° - 10-23° = 0.004

Ignited by means of electrodes.

Period of Combustion.

5j'lmiDutes . . 10*80°

6 UL- „ . . 12-90°

7\ !„ • 1379°

,, . . 13-84° (maximum)

Last Period.

9 minutes 13-82°

10 13-81'

11 „ I3'8oc

12 ..... 1379'

13 ... 137^

AT= ^ - 1378° = Q.QI2

The thermometrical observations were stopped. The
variation of the temperature was : —

13-84° — 10-25° = 3-59°

Take note of the following corrections. The system
had lost, during the minutes (7, 8) (6, 7), a quantity
of heat =

13-84° - I3-780 x 2 = Q.OI2 x 2 = 0-024 *

* Law of cooling after the maximum.

68 MINERAL OIL TESTING

During the half minute (5|, 6) a quantity of heat =

(0-012 - 0-005) 2 = 0-0035

and during the half-minute (5, 5-J) it gained

10-25° - 10-23° x i = 0.004 x .t = 0.0020*

It follows that the net loss during the minute (5,6) was-
0*0035 — 0-002 = 0*0015.

Altogether the system lost during the experiment a
quantity of heat = 0-024 + 0*0015 = 0*0255, which
must be added to the 3'5Q° already found.

The variation of the corrected temperature is then
3*615°, neglecting the fourth place. The quantity of
heat observed is therefore (2*200 + 481) x 3*615 =
9^691815. Take 9^6918.

To get the required result we must subtract from
this : —

(1) The heat caused by the for-
mation of 0*13 gr. of nitric acid

HN03, titrated volumetrically . 0*13 x 0*23 = o cal 0299

(2) The heat of the combustion of

0*025 gr. of iron thread . . 0*025 x i'6 =• o cal 0400

To be subtracted . o cal 0699

The final result is then : —

9cal 6918 - ccal 0699 = gcal 6219.

or for one kilogramme of oil 9*62ical9.t

* Law of variation of temperature at the moment of the minimum.

f Note that the dimensions of the apparatus are such that one-
can, without difficulty, arrange once for all so that in all experiments
the insignificant corrections cancel one another. If x is the correc-
tion due to loss of heat during the operation, it will not be necessary
to make the corrections if

0-237; + i '6/ = a* (P + P').

2/ is at the disposition of the worker within certain limits ; P equally
so. Evidently, then, it is possible to make the equation sufficiently
true. Thus the calorific value of colza oil is 9621*9 cal. Simply
multiplying 3-59 by 2681 gives 9624, which approximates to within
2 in 9000.

APPENDIX

TABLE FOR THE CONVERSION OP CENTIGRADE DEGREES
I>TO FAHRENHEIT DEGREES.

0 C.

O
32

i
34

2
~

3
37

7
45

8
46

9
~48~

20
30
40

68
86
104

52
70
88
1 06

72
90
1 08

55
73
9i
109

57
75
93
in

59
77
95
IJ3

61
79
97
"5

63
81

99
117

100

118

84
1 02

120

122

124

126

127

129

*3«

133

135

136

138

60
70
80
90

I4O
158
I76
194

142
1 60
178
196

144
162
1 80
198

x«
163
181
199

147
165

183
20 1

149
167

185
203

151
169
187
205

J53
171

189

207

154
172
190
208

156
174
192
210

100

212

214

216

217

219

221

223

225

226

228

I2O
I30
I4O

230
248
266
284

232
250
268
286

234
252
270
288

235
253
271
289

237
255
273
291

239
257
275
293

241

259
277

295

243
261
279
297

244
262
280
298

246
264
282
300

150

302

304

306

307

3°9

3i3

315

316

318

1 60

170

i So
190

320

338
356

374

322
340
358
376

324
342
360
378

325
343
361

379

327
345
363
38i

329
347
365
383

33i

349
367
385

333
35i

369
387

334
352
370
388

336

354
372
390

200

392

394

396

397

399

401

403

405

406

408

2IO
220
230
240

410
428
446
464

412

43°
448
466

414
432
450
468

415
433
45i

469

417
435
453
47i

419
437
455
473

421
439
457

475

423
441

459

477

424
442
460
478

426

444
462
480

APPENDIX

TABLE FOE THE CONVERSION OF CENTIGRADE DEGREES
INTO FAHRENHEIT DEGREES — (continued)

° c.

•

250

482

484

486

487

489

491

493

495

496

498

260

270
280

290

500
518
536
554

502

520
538
556

504
522
540
558

505
523
54i
559

507
525
543
56i

509
527
545
563

5n
529
547
565

513
53i

549
567

5H
532
550
568

5i6

534
552
570

300

572

574

576

577

579

581

583

585

586

588

310
320
330

340

590
608
626

644

592
610
628
646

594
612
630
648

595
6'3
631
649

597
6^5
633
651

599
617
635
653

601
619
637
655

603
621
639
657

604
622
640
658

606
624
642
660

350

662

664

666

667

669

671

673

675

676

678

360

370
380
390

680
698
716
734

682
700
718
736

684
702
720
738

685

703

721

739

687
705-

723
74i

689

707
725
743

69 1

709

727
745

693
711
729

747

694

712
730
748

696
7H

732
750

400

752

754

756

757

759

761

763

765

766

768

410
420

43°
440

770
788
806
824

772
790
808
826

774
792
810
828

775
793
811
829

777
795
813
831

779
797
815
833

78i
799
817
835

783
80 1
819
837

784
802
820
838

786
804
822
840

45°

842

844

846

847

849

851

853

855

856

858

460
470
480
490

860
878
896
914

862
880
898
916

864
882
900
918

865
883
901
919

867
885

903
921

869
887
905
923

871
889
907
925

873
891
909
927

874
892
910
928

876

894
912

930

500

932

934

936

937

939

941

943

945

946

948

OF THE

UNIVERSITY

INDEX

ABEL Flashing-Point, 13

with stirrer, 22
-Pensky Flashing-Point, 23
Act of 1879 (Extract), 14
Astatki, Characters of, 2

BALANCE, Sartorius, 1 1

Westphal, 10
Barometric Pressure, Influence of,

on F.-P., 18
Baume Hydrometer, 4

Tables, 5, 6.
Benzoline, Characters of, i

CALORIFIC value, 60
Calorimeter, 61
Capillary Test, 56
Chromometer, Wilson, 46
Clowes-Redwood Vapour Detec-
tion, 51

Coleman-Archbutt Viscometer, 41
Colorimeter, Stammer, 48
Colour, 46
Cylinder Oil Specific Gravity, 8

DOOLITTLE Torsion Viscometer, 45

ELLIOT Flashing-Point, 26
Engler Viscometer, 37

-Kunkler Viscometer, 39
Expansion, Coefficients of, 3

FIRE Test, 13, 31
Flashing-Point, 13

and Barometric Pressure, 18

Abel, 13

with stirrer, 22
-Pensky, 23

Elliot, 26

Foster, 27

Granier, 27

Gray, 30

Flashing-Point, Pensky -Martens, 2 7

Saybolt, 23

Tagliabue, 25

of Viscous Liquids, 17
Foster Flashing-Point, 27

GAS Oil, Characters of, 2
Gasoline, Characters of, i
Granier Flashing-Point, 27
Gray Flashing-Point, 30

HYDROMETER, 3
Baume, 4

Tables, 5, 6
Twaddell, 4

INDIAN Government F.-P., 19, 23

KEROSENE, Characters of, i

Specific Gravity, 3
Kunkler Viscometer, Engler-,

LIGROIN, Characters of, i
Lovibond Tintometer, 47
Lubricating Oils, Characters of, 2
Specific Gravity, 3, 7

MAHLER Calorimeter, 61

Melting-point, 58

Mineral Sperm, Characters of, 2

OIL in Scale, 59

Open Test (Flashing-Point), 13, 31

PARAFFIN, Melting-point, 2, 58
Pensky Martens Flashing-Point, 27

Flashing-Point, Abel-, 23
Pentane, Characters of, i
Petroleum Act (1879), 14
Pressure of Spirit Vapour, 50
Pyrometer, Tagliabue's, 25
Pyro naphtha, Characters of, 2
F

INDEX

REDWOOD Viscometer, 33

Vapour Detection, Clowes-,

Regnault Srjecific Gravity Bottle,
8

SAETORIUS Balance, n
Saybolt Flashing-Point, 23

Viscometer, 40
Scale, Melting-point, 2, 58

Oil in, 59

Small Samples, 12, 22.
Specific Gravity, 3

Bottle, 6

Regnault, 8

Hydrometers, 3

Small Samples, 12

Sprengel Tube, 9
Spirits, Characters of, i

Pressure of Vapour, 50
Sprengel Tube, 9
Stammer Colorimeter, 48
Sulphur, Estimation of, 60, 65
Sutherland Bulb, 60

TAGLIABUE Flashing-Point, 25
Thurston Lubricant Tester, 43
Thomson Vapour Pressure Tube, 50
Tintometer, Lovibond, 47
Torsion Viscometer, Doolittle, 45

VAPOUR Detection, 51

Pressure, 50
Viscometer, Coleman-Archbutt, 41

Doolittle, 45

Engler, 37

-Kunkler, 39

Redwood, 33

Saybolt, 40

Thurston, 43

Torsion, 45

Viscous Liquids, Flashing-Point
of, 17

WATER, Estimation of, 60
Wax Melting-point, 2, 58
Westphal Balance, 10
Wick Testing, 57
Wilson Chromometer, 46

Printed by BALLANTYNE 6° Co. LIMITED
Tavistock Street, London

BAIED & TATLOCK (LONDON), LTD.

LIST OF APPARATUS REQUIRED FOR OIL-TESTING, &c. AS

MENTIONED IN THE " LABORATORY BOOK OF MINERAL

OIL-TESTING," BY JAS. A. HICKS,

All apparatus detailed below is required for a full equipment :
the quantity will vary according to requirements

No. 32

No. ii

No. 24

SPECIFIC GRAVITY.

Item.

1. Hydrometer, specific gravity .

2. Hydrometer, thermometer for

3. Hydrometer Jar .....

4. Specific Gravity Bottle with drilled stopper

5. Thermometer for Specific Gravity Bottle

6. Regnault's Specific Gravity Bottle

7. Sprengel's Tube

8. Water Bath for Sprengel tube

9. Westphal's Specific Gravity Balance
10. Sartorius' Specific Gravity Balance

Page
4
4
4
6
6
8
9
9
10
ii

FLASHING POINT.

II. Abel's Flash-Point Apparatus 13

Ii a. Thermometers for ditto 13

lib. Abel's Flash-point Apparatus with Agitator (Fig. 8) . 21

76 APPARATUS AND CHEMICALS

Item. Page

12. Abel-Pensky's Flash-point Apparatus .... 23
- 1 2a. Abel-Pensky's Apparatus, modified by Sir Boverton

Redwood ......... 23

13. Saybolt's Electric Tester 23

14. Tagliabue's Pyrometer , 25

15. Elliot's Tester . .26

1 6. Foster's Automatic Tester ..... 27

17. Granier's Tester . . . . . . . .27

1 8. Pensky-Martens' Tester 27

19. Gray's Flash-Point Apparatus 29

20. Retort Stand with clip . . . . . . 31

21. Porcelain Crucible 31

22. Sand Bath 32

VISCOSITY.

24. Redwood's Viscometer ....... 33

24a. Stopwatch ......... 33

25. Engler's Viscometer 37

26. Engler-Kunkler's Viscometer 39

27. Saybolt's Viscometer . . . . . . 41

28. Coleman-Arch\)utt's Viscometer . . 41

COLOUR.

29. Pipette ....... 41

30. Thurston's Oil-Tester ... ... 43

31. Doolittle's Torsion Viscometer ..... 45

32. Wilson's Chromometer ...... 46

33. Lovibond's Tintometer , -47

34. Stammer's Colorimeter ....... 48

SUNDRY APPARATUS.

35. Apparatus for testing the pressure of Petroleum Vapour 50

36. Clowes-Redwood's inflammable Vapour and Gas De-

tector 51

37. Redwood's Apparatus for Capillary test . . 57

38. Beakers . 57

38a. Test-tubes . 58

39. Metal Basin 58

40. Thermometer ......... 59

41. Balance .......... 59

42. Set of Weights for ditto 59

43. Glass Dishes . . . 59

44. Hydraulic Press ........ 59

45. Sutherland's Bulb ........ 61

46. Bomb Calorimeter (Mahler's,) 61

BAIRD AND TATLOCK, (LONDON) LTD.

LABORATORY FURNISHERS, SCIENTIFIC INSTRUMENT MAKERS,
PURE CHEMICALS AND REAGENTS, &c.

Warehouse : i4^jft^4i^Mr,iJiftttpn Garden, London, E.C.
Fat.

RETURN CIRCULATION DEPARTMENT

TO— •*> 202 Main Library

LOAN PERIOD 1
HOME USE

ALL BOOKS MAY BE RECALLED AFTER 7 DAYS

Renewals and Recharges may be made 4 days prior to the due date.

Books may be Renewed by calling 642-3405

DUE AS STAMPED BELOW

SENT ON ILL

NOV 2 8

U. C. BERKELEY

]

UNIVERSITY OF CALIFORNIA, BERKELEY
RMNO. DD6 BERKELEY, CA 94720 @$

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 "The laboratory book of mineral oil testing"

See other formats