The Brown-Duvel moisture tester and how to operate it

Historic, archived document

Do not assume content reflects current
scientific knowledge, policies, or practices.

b2

- Issued February 1926
Washington, D. C. Vv Slightly revised February 1936

THE BROWN-DUVEL MOISTURE TESTER
AND HOW TO OPERATE IT

By

D, A. COLEMAN and E, G. BOERNER, Grain Investigations, Grain Division
Bureau of Agricultural Economics

CONTENTS

Page
Description of the Apparatus cs . . s . . s . s o . e. J es e e s es e s . . ee e. . ° 3
SOTistrUctiOlt SPSCHICALIONS (cts, och ue lc) oo code w!'w felts cee 6 * oilers so s- 6 4
Description of Accessories . . 2... 2.22 2 ee ee a clislisl a Lewes Son en enam cise. < 5
Supply Pipesfor Gas-Heated Moisture Testers . . 2... 2 eee cee eee eee 15
Ramer M Teemicrere, WAMISEOCE TOSE gfe io rea, a ot eal ws 'q le oh eb 16: 0 0 eles ae 17
PeeWee best Uitterent Subsiatices . 5 5s 6c ic 03 we Be ee Bes ee 8 8 18
Swatdlardizinefheivoisture tester, 0.2 s,4\0 «= <-¢ oes os ee «<< © 20
Variations in Heating Time Caused by Variations in Gas Pressure. ........--. 23
Variations in Heating Time and Moisture-Test Result Caused by Pesition and Condition
CE OSE Gg RS RE Ne ee ee ee oe a ae a 28
Variations in Heating Time Caused by Variations in Thermometer-Bulb Immersion . . 29
Influence of Cooling Time on Moisture-Test Results. . . 2... 2 2 ee ee eee 30
MME PARE Re a Ree ae. Su ated nad) inec, oes eve wc aleve whete ce a) % 31
Volume ef Oil versus Moisture-Test Result ........-. circ chet eites cnc aan 33
Brown-Duvel Moisture Testers Equipped with Electric Heaters. . 2... 2. eee 33
Tests with Alcohol and Gasoline Burners. . . 2... 2 eee we eee ee es 39
How to Choose Extinguishing Temperatures . . 2... 2 0 2 2 se oe ee eee 39
Drawing and Handling Samples. . . 2.2... ee ee eee SE oo bo) ois 41
Special Points for Consideration. . . 2... 22 ee EMcmcmciet aileale| el /el ones. on ae
a
f

CHICAC

j GENERAL FIELD
UNITED STATES
GOVERNMENT PRINTING oFFICE | Bk VAY 4
WASHINGTON
1936

| U.S, DEPARTMENT

PR cae

WAR ATABEA

Rian oe

30, I q
TEADOI f
= ab

OF AGR 5

URE ;

UNITED STATES DEPARTMENT OF AGRICULTURE

Issued February 1926
Vv Slightly revised February 1936

Washington, D. C.

THE BROWN-DUVEL’ MOISTURE TESTER AND HOW TO OPERATE IT

By D. A. CoLEMAN’® and E. G. Boerner, Grain Investigations, Grain Division,
Bureau of Agricultural Economics

CONTENTS
Page Page
Description of the apparatus_-__---_- 3 | Influence of cooling time on moisture-
Construction: specifications——_____-_—— 4 + test. Lesulige = = eee 30
Description of aecessories____-_-__-~~_ 5) Priming: tesis=3-) 2. —.. Bae ae 31
Supply pipes for gas-heated moisture Volume of oil v. moisture-test result_ 33
(Pgh a eee ee eee als; Brown-Duvel moisture testers equipped
How to make a moisture test-__----~ 17 withtelectric heaterss22-2- S25. = 33
How to test different substances______ 18 | Tests with alcohol and_ gasoline
Standardizing the moisture tester____ 20 DUUEMCRS seth eg Se _ Ses 39
Variations in heating time caused by How to choose extinguishing tempera-
variations in gas pressure____-_-~--~_ 23 Tunes ee see oe ee a ae 39
Variations in heating time and mois- Drawing and handling samples—_-_-_~_ 41
ture-test result caused by position Special points for consideration______ 42
and condition of wire gauze_______ 28

Variations in heating time caused by
variations in thermometer bulb im-
mersion sso e S22 Sh a Ss Ses ats 29

The official grain standards of the United States specify the
water-oven method as the basic official method for determining the
moisture content of corn and the air-oven method for all other
grains. The standards permit, however, the use of any device and
method which give results equivalent to those obtained by the official
methods. When the Brown-Duvel device is used for the purposes
of grain inspection, the operative instructions contained in this bul-
letin are essential for its efficient use.

Since the method was first developed, it has been further perfected
for determining the moisture content of rice, of the more important
seeds, of flour and meal, and certain other commodities. Modifications
have been made both in the apparatus and in the methods for making
tests and more specific instructions have been recently developed
for so making the moisture tests as to accomplish greater accuracy
and uniformity in the results with the apparatus.

1This apparatus was originally invented by J. W. I’. Duvel and Edgar Brown, United
States Department of Agriculture, who hold Public Service Patent No. 848616 on the
apparatus,

2 Credit is given John H. Cox and H. C. Fellows, Grain Division, Bureau of Agricul-
tural Economics, and to H. B. Musser, formerly scientific assistant in Grain Standardi-
zation, Bureau of Agricultural Economics, for some of the experimental data contained
in this bulletin.

41523°—36——_1

bo

BULLETIN 1375, U. S. DEPARTMENT OF AGRICULTURE

Electrically-heated moisture testers have lately come into general]
use. Their Operation and efficiency have been studied and compari-
sons drawn between them and the standard gas-heated machines.
Wiring specifications have been worked out for electrically-heated
machines and piping specifications for gas-heated machines.

30th the construction of the apparatus and the methods of mak-
ing the tests have been made as simple as practicable, so that reliable
tests can be made by any careful worker, even though he has not had
the benefit of special chemical- laboratory training.

In making the tests it is of the utmost importance that the operator
be a careful and honest worker, who can be depended upon to follow
the instructions given for making the tests and who will report re-
sults correctly.

Fic. 1.—Brown-Duvel moisture tester equipped with gas burners and gas-pressure
regulator

The instructions for making moisture tests given in this bulletin
are applicable only when used in connection with the moisture tester
herein described, specified, and illustrated, and do not apply to modi-
fied forms of moisture testers.

This apparatus was originally described by J. W. T. Duvel and
E. Brown in 1907 in Bureau of Plant Industry Bulletin 99, “A
Quick Method for the Determination of Moisture in Grain.” Later
this publication was reissued as Bureau of Plant Industry Circular
72, under the title, “A Moisture Tester for Grain and Other Sub-
stances and How to Use It.”

With the use of a Brown-Duvel moisture tester a single moisture
determination on grain can be made in 25 or 30 minutes, and with
a 6-compartment tester six tests can be made in approximately the
same time as for a single test. In commercial work, where a large

number of tests are to be made, one man and a helper, with the use

THE BROWN-DUVEL MOISTURE TESTER 3

of three 6-compartment machines, can readily make 200 or more
tests in a day of eight hours, an average of less than two and one-
half minutes for each test. The rapidity with which moisture
determinations can be made by this method makes it especially
valuable for testing many substances other than grain which have
a large percentage of

water, are difficult to MOISTURE FESEER:

grind, or contain such
large quantities of vola-
tile oils that determi-
nations based upon dif-
ferences in weight dur-
ing drying are unre-
liable. In addition,
this method obviates
the necessity of grind-
ing samples, thus elimi-
nating the loss of water
which always accom-
panies the grinding of
material of relatively
high moisture content.

No. § one hole
rvbber stopper

DESCRIPTION OF THE
APPARATUS

The apparatus con-
sists of a heating
chamber divided into
compartments so that a
number of samples can
be tested at the same
time, a tank for cold
water through which
the condenser tubes
pass, and a_ suitable
stand for supporting
the heating chamber
and cold-water tank,
together with numer-

ous accessories, includ-
ing thermometers, dis-

pale a bolt
tillation flasks, and til |

graduated measuring Sectional view of the official moisture fester, __
eylin d ers. Showing the various parts properly connected for vse.

ig Te | Hos Rubber le
=e Sopp.
Oey

Figure 1 shows an mie eae section of Brown-Duvel moisture tester,
: showin e various part roperly connected f
external view of a asa FONE CAE: Gilk oeteok

standard 6-compart-

ment moisture tester set up ready for use. Figure 2 is a cross section
through one of the compartments of the tester, showing the position
of the flask and the flask support within the compartment, the ad-
justment of the thermometer, and the proper connections of the vari-
ous parts.

4 BULLETIN 1375. U. S. DEPARTMENT OF AGRICULTURE

Figures 1 and 2 show testers equipped with burners for using gas
as fuel. The form of construction, however, is such that alcohol
or gasoline burners, or electric heaters can be used, provided the di-
rections herein given are carefully followed.

CONSTRUCTION SPECIFICATIONS

The heating chamber is 3 feet 214 inches long, and is divided
into six compartments, each 514 inches square and 61% inches deep,
inside measurements. The compartments are separated by air spaces
one-half inch wide. Each compartment is lined with ;3;-inch asbestos
sheeting (transite board).

The heating chamber is made of heavy, galvanized iron through-
out. The sides extend 714 inches below the chamber on the back and
the two ends to form a stand, and are rolled at the bottom for firm
footing. The front side extends only 11% inches below the chamber,
and to this side is attached a rack 11/4, inches wide extending the full
length of the chamber which is provided with holes for holding
thermometers when not in use. The heating chamber has a hole 3
ee in diameter cut in each end, in line with the burners. (See

gs)

A galvanized angle iron seven-eighths inch wide is riveted around
the inside of each compartment as a rest for the wire gauze used
over the gas flame, on which the flask seat is placed. Each compart-
ment has a 214-inch round mica window for observation during
test.

Each compartment is provided with an asbestos-lined (transite
board) cover with a 114-inch opening in center to allow for pro-
jection of neck of flask when closed.

Each compartment has a cut in back five-eighths to three-fourths
inch wide and seven-eighths inch deep on center line of compart-
ment, to allow for projection of stem of flask which leads to con-
denser tank.

A copper condenser tank 4 inches wide and 12 inches deep extends
the full length of the heating chamber. This is mounted on the
base back of the chamber.

The tank has a 14-inch inlet pipe at the bottom and a 34-inch over-
flow pipe three-fourths inch from the top. The tank has six open-
ings in the bottom so arranged as to coincide with the center line of
the compartments for condenser tubes; these openings are large
enough to take a No. 5 rubber stopper and are reinforced by a cop-
per plate 114 inches in diameter sweat to the bottom.

The base for the heating chamber and condenser tank is 3 feet
334 inches long, 1114 inches wide, and 111% inches high. The legs
are made of 214 by 21% by 1% inch angle iron and tied across back
and ends by 1 by 1 by 1% inch angle iron. Tie angles are mitered at
corners and set inside of legs to form seat for chamber and tank.

The end legs are tied 214 inches from bottom by a %4 by 34 by %&
inch angle iron. From this angle to top angle, two 14-inch round
rods are set 114 inches apart center to center, between which the
gas line pipe is passed and a lock or clamp nut is placed on pipe on
either side of rods for clamping pipe into position, making same
adjustable up and down.

THE BROWN-DUVEL MOISTURE TESTER 5

The center of gas line should be 314 inches from front of base.

The gas pipe should be three-fourths inch inside diameter.

Pet cocks for gas burners are mounted on pipe line, the center of
which coincides with centers of compartments.

Figure 8 is a working drawing of the Brown-Duvel tester,
equipped with gas heaters.

DESCRIPTION OF ACCESSORIES

THERMOMETERS

Tt is of the utmost importance that the thermometers be of extra
quality and not the ordinary grade of chemical thermometers which
are commonly furnished with apparatus of this general character.
The success of this method of making moisture determinations de-
pends largely upon the accuracy with which the temperature read-
ings are made, and any thermometer showing an error of more than
one-half of a degree at any of the prescribed points should not be used
unless such errors are known and provided for in the readings.

It is also necessary that the mercury bulbs of each of the ther-
mometers be of approximately the same length, so that uniformity
can be had as to the depth to which the thermometers are immersed
in the oil, the correct depth being four-fifths immersion, as shown
in Figure 2.

The thermometer used in this apparatus should be made of suitable
glass properly annealed for use at high temperatures, and should be
free from defects which would prevent making accurate readings.

The graduations of the thermometer should be in whole degrees
from 0° to 220° C., and it should have a white background on the
stem. Errors in scale graduations should not exceed 1° at any point
and between 170° and 200° C. should not exceed 0.5°. The diameter
of the stem of the thermometer should be nine thirty-seconds inch,
so that it will properly fit into a No. 5 one-hole rubber stopper. The
length of the mercury bulb should be approximately three-fourths
inch (19 mm.) and its diameter should be approximately 5 mill-
meters.

Variation from the dimensions given above and pictured in Figure
4 should be not greater than 0.25 millimeter on the stem and
bulb diameters and one-sixteenth inch (1.6 mm.) 1m the bulb length.

Manufacturers should be required to furnish certificates of cor-
rection with all thermometers showing errors in calibration and
graduation exceeding 0.5° at 175° and 200° C.

A standard thermometer for use in the Brown-Duvel moisture
tester is illustrated in Figure 4.

GRADUATED MEASURING CYLINDERS

The full-sized graduated cylinders for measuring the moisture
content of the grain being tested should have a capacity of 25 cubic
centimeters.

The graduates should be made to contain the indicated volume at
20° C., and the graduations in whole cubic centimeters and either
one-tenth or two-tenths of a cubic centimeter should be legible and
conspicuously and permanently marked on one side of the graduate.

BULLETIN 1375, U. S. DEPARTMENT OF AGRICULTURE

‘e ‘DI

,

ne Oe
co Sa

snjviedde off JO UOTNAJsUOD po[IeJep SuTMOYS ‘10}S9} OINYSlouL JOANCG-UMOI_ JO WOPJooS [VUIPLjLsUO'T

pr ; |

Y31S3L AYNLSION

wseddoje

10f Pafl4p 212M
2240, U2Bd Of Yb
{0 UlOYOY Of {POMS
80/0 420002 bS F/
Uodt boy JX/%/

peyof4ad aiid |
ol// p2asZ

/

z

0 }
-2ap,2/ | {fH
~9L/M,% } IT Gajaas sojsagso
- Luo Z ee | . 9z00b aliV
~26 OZy |
Vly da000D |

Lb dy JOP 130009
SOlSAGEOD 7,

UO4I paziubs/on - : ! j
gouy Uepoom | “Uol{bjyued 104 do? fo : Foe ; — vr bas cere aamaeee Shion

asim ysbual Seoy Fd

THE BROWN-DUVEL MOISTURE TESTER q

All graduation marks should be perpendicular to the axis and par-
allel to the base and to each other. The graduation marks should
be clear and distinct and uniform in character and should be etched
or engraved and should not exceed 0.38 miliimeter in width. Blown

Ae

OL LU LU LY
Q)

pee

|
Not less than 11"(2290m)—-—>
7%" -—__—-.

LE

WS
| mS 5 1.
Fie. 4.—Standard thermometer with Fic. 5.—Standard moisture-test-
the proper shape and dimen- ing graduate, 25 cubic centi-
sions for use in the Brown-Duvel meters capacity

moisture tester

or pressed graduation marks should not be used. ‘The clear inter-
val between graduation marks should be not less than 1 millimeter.
The value of the main graduation marks should be plainly desig-
nated, as shown in Figure 5, The tolerance to be allowed in excess

S$ BULLETIN 1375, U. S. DEPARTMENT OF AGRICULTURE

or deficiency in the capacity of the graduates should be not greater

than 0.05 cubic centimeter at the points 5, 10, 15, 20, and 25 cubic

centimeters. The base of the graduates should be sround to rest

evenly on a flat surface at right angles to the axis of the diameter.

The eraduates should be cylindrical in form, 75g inches high and
eleven-sixteenths inch inside di-
ameter, as shown in Figure 5.
They should be of good quality
glass, thoroughly annealed, clear,
transparent, of uniform but not
excessive thickness, and free from
bubbles and streaks.

In using the graduates where
100 grams of the material being
tested are used for the test, the
number of cubic centimeters of

, Water which collects in the gradu-
Yate corresponds to the percentage
of moisture originally contained

in the sample.

In making the moisture deter-
mination, a small quantity of oil—
usually less than 0.5 cubic centi-
meter—is carried over into the
measuring cylinder or graduate
and collects on the surface of the
water, so that the moisture read-
ings should be made at the bottom
of the meniscus between the oil
and the water. Should the oil
and the water not separate read-
ily, the graduated cylinder should
be whirled by rolling quickly be-
tween the two hands.

QO £ GQ O

fe)

SPECIAL GRADUATE

wm a £f§ TW Oo VN WO ©

In making moisture tests of oats
it is necessary, on account of the
bulkiness of the sample, to use
only half the quantity of sample
ordinarily used for other grains,
so that a_ special oraduate of

| fn smaller capacity 1s necessary. In

Prove a using the special graduates in

ie, 6 Specialisnoistare-testingeradu- connection with tests in which

ate, 16 per cent capacity, for use in only 50 grams of the material to
testing moisture content of oats and

other lightweight material be tested are used, the number of

cubic centimeters of water which

collects in the graduates corresponds to only one-half of the moisture

contained in a full-sized sample of 100 grams, but the graduates are

so calibrated that the results are read in direct percentages. All the

requirements noted for the manufacture of the regular 25 cubic centi-

meter graduate obtain for this one also, except that the dimensions

16

THE BROWN-DUVEL MOISTURE TESTER 9

for this graduate are 75g inches high, nine-sixteenths inch inside
diameter, with 114 inches of the top portion flared to eleven-six-
teenths inch in diameter, as shown in Figure 6.

After each test the cylinders must be cleaned and dried, which
can best be done with a swab or test-tube cleaner with a small
sponge attached to one end, as shown in Figure 7. Satisfactory
cleaners can also be made as needed by twisting cotton
waste, strips of cheesecloth, or some other similar sub-
stance about one end of a piece of fairly heavy wire.

If the graduate is not thoroughly clean, water from
the following test will hang along the sides, making a
correct reading exceedingly difficult; if the graduate is
not thoroughly dry, the following test will of course
show an inaccurate result.

DISTILLATION FLASKS -

The distillation flasks, with a capacity of approxi-
mately 1,000 cubic centimeters, should be made of the
best grade of resistant glass and should be well an-
nealed to withstand sudden changes in temperature
without breaking. The necks of the flasks should be
sufficiently heavy to stand tight corking. Figure 8
shows one of these flasks in which the dimensions of the
different parts are given in both centimeters and inches.
Flasks showing wide variations in the dimensions
are useless. It is therefore recommended that in order-
ing flasks it be definitely stated that they must be in
accordance with the specifications given in this bulle-
tin. In case the dimensions of the flasks, as well as
other parts of the apparatus, are not approximately as
herein specified, they should not be accepted or used.
The most important points regarding the distillation
flasks are the diameter of the flask, the dimensions of
the neck, the distance from the bottom of the flask to ~
where the outlet tube is formed, and the angle of the ©
outlet tube. If the distance from the bottom of the flask
to the outlet tube is not correct, the flasks will not rest F'¢. %-— Swab

, ° for cleaning
in the compartments of the tester in the proper manner. graduated

This condition makes it impossible to place on correctly = Giinders >
the cover of the compartment.

It is recommended that a plaster of Paris mold, as illustrated in
Figure 9, be made of a flask of the proper dimensions and that all
flasks be tested in this mold before being accepted from the manu-

facturer.

YI INCHES LONG

COPPER DISTILLATION FLASKS

Numerous requests for information are received from time to time
as to the advisability of using copper instead of glass flasks for mak-
ing moisture determinations by the method described in this circular.
The results of investigations by J. H. Cox, of the Grain Division,
Bureau of Agricultural Economics, have shown that copper flasks
may be used in the tester, although glass flasks have proved more
satisfactory for eee: work. In making moisture determinations

41523°

BULLETIN 13875, U. S. DEPARTMENT OF AGRICULTURE
10

of corn and oats, the same treatment is required in the use of copper
flasks, when spun from 16-ounce sheet copper and used in the tester
described in this bulletin, as is required when the tests are made in

100C..GC.

Fig. 8.—Standard single-walled distillation flask, 1,000 cubic centimeters capacity

glass flasks, the essential feature being to extinguish the flame
promptly when the thermometer reaches a temperature of 190° for
corn and 195° C. for oats. In using copper flasks in this machine

Fic. 9.—Plaster of Paris mold for determining the accuracy in size and shape of
distillation flasks

THE BROWN-DUVEL MOISTURE TESTER 11

for making moisture determinations of the classes of Hard Red
Winter, Hard Red Spring, Durum, and Red Durum wheat, how-
ever, the flame should be extinguished when the thermometer reaches
a temperature of 185° and not at 180° C., as is required when using
glass flasks, otherwise the treatment should be the same. Correct
temperatures for use with copper flasks in making moisture deter-
minations of the other grains, including Soft Red Winter and White
wheat, have not been determined.

SPECIAL DOUBLE-WALLED DISTILLATION FLASK FOR TESTING FLOUR AND MEAL 2

A special double-walled flask designed for use in connection with
the Brown-Duvel tester for making moisture determinations of flour
and meal has the same outside dimensions as the regular distillation
flask which is illustrated in Figure 8. The double-walled flask can

Fic. 10.—Special double-walled distillation flask for testing moisture content in
flour and meal

be made of either glass or copper. It is illustrated in Figure 10.
The inner flask has a capacity of approximately 900 cubic centimeters
and the space between the two walls should hold not less than 250
nor more than 300 cubic centimeters.

If the flasks are made of copper, the thickness of the copper be-
fore it is spun should be 0.022 inch, or 16 ounces to the square foot.
The copper flasks will have to be made in two sections and the sec-
tions soldered together in the middle with a very hard solder. The
soft solder commonly used by plumbers is not suitable for this work.
Success with these flasks has been attained only when they were
soldered together with a silver solder. The neck of the flask must be

° This flask was developed by J. H. Cox of the Grain Division, Bureau of Agricultural
ee ake flask and its manner of construction are fully described in Department
ulletin No.

12 BULLETIN 13875, U. S. DEPARTMENT OF AGRICULTURE

of but one thickness of copper, for if it is too heavy it will melt the
rubber stoppers.

If the flasks are made of glass, they should be made of the best
grade of resistant glass and well annealed, and the necks should be
sufficiently heavy to stand tight corking.

When 150 cubic centimeters of oil is poured in between the two
walls, the top of the oil should be about halfway up the sides of the
flasks. If the flasks do not meet these specifications they should not
be used.

FLASK SUPPORTS

The flask supports used in this apparatus consist of an asbestos
ring of 14-inch transite board, 414 inches in diameter, incased in
a galvanized-iron frame 514 inches square, the iron frame being
flanged on either two sides or on all four sides to a depth of 114

Sr
rie

ae

Huss!

Se
on"
4
nee
o
“

Ky mil
. | |

Fic. 11.—Flask support upon which the distillation flask rests when in place in
the moisture tester

inches. Each corner of the casing is cut in such a manner that
a %-inch epening is formed. The flask support is shown in Figure
11. This form of support exposes the bottom of the flask uniformly
to the action of the heat, and at the same time is so made that a
sufficient quantity of heat comes in contact with the upper part of
the flask.

The asbestos ring in the top of the support should be cut so that
the bottom of the flask when in place on the support will be not less
than three-eighths inch above the asbestos center of the wire gauze,
as shown in Figure 2. The support should be 11% inches high.

WIRE GAUZE

The wire gauze in the bottom of each compartment between the
flask and the flame should be made of iron wire, from 0.016 to 0.020

|

a a

THE BROWN-DUVEL MOISTURE TESTER 13

inch in diameter, with 20 wires to the inch and with a 2-inch as-
bestos center, as shown in Figure 12.

The asbestos center insures a uniform distribution of heat and adds
greatly to the life of the wire. It can readily be put in by mixing
asbestos cement with water and rubbing it into the mesh of the gauze.
The asbestos center should always be kept in good condition in order

r to prevent the flame from playing directly on the bottom of the flask.

—= ————— ———
eneasnessna REE SR SRCRERSKRERAERRE SESE’
RERSRSRene :
PR BR! ek BRAEERSLARERSEARESRERS E ‘
BETRORERSHARSRGRERSHLA SR: SRGRERERERSR SESE Selpe nlos Sade bon Sad
BR at SR BRE Bee ‘ i CEB ME Be
RERSRER BASE CRERERERSEERERER
SEERA KRERAGORARER SSAA RTEAMEARAe Re S:
~ eR, |

ad bod sche Sonne dente toad
REKSRELSLARERRAKHAKSAD A!

BSRSRELLELREUSR

RERCR BH:
SRELLELLLCARIAKREKASARARSARLIKRASVAKSRS
2 SLES 8 2° ~~ “SE ee

RSRSRATARERERS SBP
= = =

ie
rH

j

Bene
rH

ae
nat!
TR

ieles
Bokel
iH

s
3
2
=
£
=
2
R

i HEH ;
Hh

sassess
Daneneunn Cs
sein siiitesstase:

oti tet ee
eo

ssassestssseaase
seeeeeseaes

pre ;
BB
peaeasacagas
aseseotatetaaeaese

Fig. 12.—Wire gauze, with asbestos center
CONDENSER TUBES

The glass condenser tubes are 34 centimeters in total length, 0.70
centimeter in diameter, and the upper 3 centimeters flanged to an
inside diameter of 2.1 centimeters so that it just fits a No. 3 stopper.
The tubes should be so adjusted in the No. 5 rubber stoppers that,
when the latter are firmly pressed into the holes in the bottom of
the cold-water tank, the tops of the tubes will be approximately one-

}

fourth inch above the top of the tank.
RUBBER STOPPERS

The N 0. 9 one-hole rubber stoppers which carry the thermometers
and the No. 3 one-hole rubber stoppers used on the side tubes of the
flasks should be of such quality as will withstand comparatively

14 BULLETIN 1375, U. S. DEPARTMENT OF AGRICULTURE

high temperatures and resist the action of the oil. Stoppers con-
taining a large percentage of pure gum are not desirable, as they
soon swell and become unfit for use. Before attempting to put any
of the glass parts through the holes in the rubber stoppers, they
should be moistened with oil or water, thereby lessening the chances

of breaking the tubes and cutting the
wa 2./ cm, hands. A good grade of stopper

should last through 400 tests.
OIL FOR THE MOISTURE TESTS

In making moisture tests a good
grade of mineral engine oil must be
used. The oil must be free from water,
should have a specific gravity of about
0.9000, or 25° Baumé, should have a
viscosity at 20° C. of from 342 to 513
seconds Saybolt. Its flash point in open
cup should be between 350° to 400° F.
Fire point should not be over 450° F.
The higher the viscosity, the greater
the danger of boiling over, and for this
reason cylinder oils should not be used.

AUTOMATIC OIL-MEASURING DEVICE

The automatic oil-measuring device
consists of a container to which is at-
tached an automatic measuring gauge.
The gauge must be sufficiently accurate
to measure and deliver not less than 150
cubic centimeters and not more than
155 cubic centimeters of testing oil in

one draft. The device is operated by
70 C.M. an upward pressure of the neck of the
flask against an automatic catch on the
gauge. The cover for the container
should have a strainer sieve for strain-
ing all the oil used in the tests. One
form of the device is illustrated in
Figure 14.

OIL-RECOVERY AND OIL-STORAGE CANS

I4 CIN, Tora/ feng? fi

The oil-recovery and oil-storage cans

Ie are made similar to 5-gallon milk cans.
== The storage can is equipped with a

Fic. 13.—Condenser tube standard faucet at its base. The oil-
recovery can is supplied with a strainer funnel, the sides of which
are made of heavy-weight tin. The hopper of the funnel should be
12 inches in diameter at the top, 7 inches deep, with an opening 5%%4
inches at the base. The collar should be 2% inches wide, having an
outside diameter of 534 inches. A brass mesh filter screen of fine
dimensions should be soldered over the opening of the funnel where
the collar joins the hopper. The top edge of the funnel should be
finished with a 34,¢-inch roll. These devices are shown in Figure 15.

THE BROWN-DUVEL MOISTURE TESTER 15
BALANCE FOR WEIGHING SAMPLES

Though not a part of the moisture tester, a balance is necessary
for weighing the samples of grain to be tested. A balance of simple
construction, having an accuracy and sensitiveness of not less than
0.1 gram, has been found satisfactory. Balances similar to the type
shown in Figure 16 will meet all requirements.

Whatever the type of balance used, it should rest on a firm sup-
port, preferably a heavy shelf securely fastened to a solid wall of

Fic. 14.—Automatic oil-measuring device

the building. Such a shelf will greatly facilitate the keeping of it
in balance, which is absolutely essential for rehable work.

SUPPLY PIPES FOR GAS-HEATED MOISTURE TESTERS

Tt is absolutely necessary that a plentiful supply of gas be on hand
to operate a moisture tester. The only way to assure this is to have
gas pipes of sufficient size. The gas pipe on a single 6-compartment
tester should be at least one-half inch inside diameter and prefer-

16 BULLETIN 1375, U. S. DEPARTMENT OF AGRICULTURE

ably three-fourths inch. Cases are known where from two to four
machines have been attached to a %%-1inch supply lne, and in
every case serious errors have resulted in the moisture-test results.
The errors were the result of an insufficient supply of gas to the
burners when more than one machine was operated at the same
time.

The installation of gas supply pipes for moisture machines de-
pends to a large extent upon where the moisture machines are to be
located in the laboratory. They should be installed in a place where
they will not be exposed to air currents. If
the moisture machines are placed so that they
are all in the same direction, the size of the
gas pipes or manifolds on each machine should
each, in a general way, be of a different size.
For example, if four machines are to be placed
in a line, the machine nearest the supply line
should have a manifold 114 inches inside di-
ameter, the machine next in line should have a
1-inch manifold, the next machine a 34-inch
manifold, and the last machine also a 34-inch
manifold. Of equal importance is the capac-
ity of the pipe line from the meter. Under
no consideration should this be smaller than 1
inch, and for a set-up as described above this
pipe should be not less than 114 inches.

In some laboratories it is not possible to
place all the machines in one line. In such
cases they are usually placed in a formation
at right angles to the supply line. Under
these circumstances the main supply line
should be larger than in the set-up mentioned.
With testers in an angular position, each ma-
chine should be supphed by a 34-inch pipe,
and these should each be tapped into a 214-
inch main-line pipe. It is of further de-
cided advantage to eliminate as many

ga atisigie oes gb angles, turns, and tappings as possible on the

main supply line; that is, the meter line. No
particular advantage appears to accrue from leading a supply
pipe to both ends of the manifold.

BURNERS

Tt is not essential to have any particular type of burner for gas-
heated machines. Almost any burner having a diameter of 20 milli-
meters and 130 millimeters in height will suit the purpose. The
important thing concerning any burners that are used is that, after
being once adjusted, they stay adjusted to give off a uniform quan-
tity of heat. When Meker burners are used the burners should be
equipped with wing-nut set screws placed on the burner to anchor
the air valve to any desired position. When Bunsen burners are
used they should be equipped with set screws for both the air and
gas valves.

THE BROWN-DUVEL MOISTURE TESTER 17

HOW TO MAKE A MOISTURE TEST

The method of making moisture tests with a Brown-Duvel tester
consists of heating the whole kernels of the various grains, rice, seeds,
spices, or the flour, meal, broken grains, or other material to be tested
in a mineral oil having a flashing point much above the boiling point
of water, in condensing the water which distills off, and in collecting
and measuring the moisture in a suitable graduate. The method is
so simple that the tests can be made by any careful worker who is

. eee

H | |
|
:

>

Fic. 16.—Balance for weighing samples to be tested for moisture content

able to follow simple instructions, but it is essential for uniformly
accurate results that the tests be made in strict accordance with the
details of the method here described.

After properly mixing the bulk sample to be tested, carefully weigh
the desired quantity for the moisture test and empty immediately
into the distillation flask, to which add the oil and then shake, with a
slight whirling motion, until the two become well mixed. Grasp the
neck of the flask in one hand and hold it in such a manner that when
the No. 5 rubber stopper carrying the thermometer is inserted it can
be readily determined that the mercury bulk of the thermometer is

41523°—36——_3

18 BULLETIN 1375, U. S. DEPARTMENT OF AGRICULTURE

exactly four-fifths immersed in the oil, as shown in Figure 2. Place
the flask in the proper compartment of the tester and proceed in a
similar manner with the remaining samples.

The correct adjustment of the thermometer is a factor which must
not be ignored. If the mercury bulb of the thermometer is too near
the bottom of the flask, the percentage of moisture will be too low;
and if the mercury bulb is placed too high, the percentage of mois-
ture will be too high. For the same reason it is important that there
be uniformity in the length of the mercury bulbs of the thermom-
eters, which, as previously stated, should be approximately three-
fourths inch.

As the thermometer is being inserted, glance along the side tube of
the flask to make certain that it has not become stopped by the lodg-
ment of some particle while the flask was being emptied or filled.
If the side tube is not open to permit the free escape of the rapidly
forming steam the pressure during the heating will become sufficient
ie on out the stopper and thermometer or possibly to burst the

ask.

When the flasks have been filled and placed in the compartments of
the tester, connect the side tubes of the flasks by means of the No. 3
rubber stoppers with the thimbles of the glass condenser tubes which
extend down through the cold-water tank, so that the moisture which
is liberated from the grain or other substance will be condensed and
then collected in the graduated cylinders beneath the tank. Figure 2
shows the correct manner of adjusting the various parts of the
apparatus.

Place the cover over the flask, then ignite the gas, the heating
capacity of which has already been adjusted, as described on page
22. When the desired temperature is reached, extinguish the flame
quickly, after which the thermometer will show a slight gradual
increase in temperature, approximately 5 to 6°. Let the thermometer
recede to. 160° C. before making a reading. Before reading, remove
the covers and then disconnect the flasks from the condenser tubes
to allow the small quantity of moisture which sometimes collects at
the base of the No. 3 rubber stoppers to drop into the graduated
measuring cylinders. The percentage of moisture which has collected
in the graduates is read beneath the layer of oil on top of the water.
To guard against possible error, it is desirable to make duplicate
tests of all samples, and if there is no appreciable variation, to take
the average of the two readings as the correct percentage of moisture.

While the contents of the flasks are still hot, remove the ther-
mometer, take the flask by the side tube and after giving a slight
whirling motion, invert quickly, emptying the contents into a suit-
able strainer, so that the oil can be recovered for further use.

When the flasks are not in use, keep them in place in the com-
partments and make all connections the same as for a test. In using
a new flask for the first time, or when the machine has been idle over
24 hours, “run” a preliminary sample previous to making a regular
test so that all the flasks will be in uniform condition.

HOW TO TEST DIFFERENT SUBSTANCES

Detailed instructions have been worked out for making moisture
determinations of a limited number of substances, as follows:

THE BROWN-DUVEL MOISTURE TESTER 19

Barley.—Use 100 grams of grain and 150 cubic centimeters of oil,
and extinguish the flame when the thermometer registers 190° C.

Barley malt—Use 100 grams of malted barley and 200 cubic
centimeters of oil, and extinguish the flame promptly when the ther-
mometer registers a temperature of 168° C. The flame should be
adjusted so that it will require approximately 20 minutes to reach
the prescribed temperature.

Buckwheat.—Use 100 grams of grain and 150 cubic centimeters of
oil and extinguish the flame when the thermometer registers 195° C.

Corn (maize) —Use 100 grams of grain and 150 cubic centimeters
of oil, and extinguish the flame when the thermometer registers
190° C. This method, however, can not be used with pop corn.

Corn meal.—Use the double-wall flask with 100 grams of the corn
meal and 150 cubic centimeters of oil in the inner flask and 150
cubic centimeters of 011 between the walls. Extinguish the flame when
the temperature reaches 175° C. The oil in the inner flask should
reach 175° C. in 26 minutes.

Corncobs.—Use 50 grams of cob cut in pieces that can be easily
removed from the flask and 250 cubic centimeters of oil, and extin-
guish the flame when the thermometer registers 190° C.

Cottonseed.—Use 50 grams of seed and 150 cubic centimeters of
oil and extinguish the flame when the thermometer registers 190° C.

Distillers’ dried grains—Use 50 grams of distillers’ dried grains
and 200 cubic centimeters of oil, and extinguish the flame promptly
when the thermometer registers a temperature of 190° C. Special
care should be taken to see that the oil and the distillers’ dried grains
are thoroughly mixed before beginning the test.

Emmer—Use 100 grams of grain and 150 cubic centimeters of oil
and extinguish the flame when the thermometer registers 190° C.

Flaxseed —Use 100 grams of seed and 150 cubic centimeters of oil,
and extinguish the flame when the thermometer registers 175° C.

Grain sorghums.—Use 100 grams of grain and 150 cubic centi-
meters of oil, and extinguish the flame when the thermometer regis-
ters 195° C.

Oats, Feed oats, and Mixed Feed oats—Use 100 grams of grain
and 150 cubic centimeters of oil and extinguish the flame when the
thermometer reaches 195° C. With oats, especially those that are
light and chaffy, special care should be taken to insure thorough
mixing of the oil and grain. In extreme cases it may be necessary
to add an extra 50 cubic centimeters of oil and to heat with a slower
flame in order to reduce foaming to a minimum. Coincident with
the use of a 100-gram sample, the use of the special oat graduate is
discontinued and the regular-sized graduate is used in its place.

fice (unhulled) —Use 100 grams of grain and 150 cubic centi-
meters of oil, and extinguish the flame when the thermometer regis-
ters 200° C.

Milled Rice (head, second head, and screenings) and Brown Rice.—
Use 100 grams of grain and 150 cubic centimeters of oil. Place a
glass-wool or mineral-wool pad 3 inches in diameter and 1-14 inch
thick in. the bottom of the flask. Care should be taken not to
displace the wool pad while placing the oil and grain into the flask.
Cut off the heating unit when the thermometer registers 200° C

20 BULLETIN 1375, U. S. DEPARTMENT OF AGRICULTURE

Brewers Rice—Use 100 grams of brewers rice and 150 cubic centi-
meters of oil. Place a glass-wool or mineral-wool pad 4 inches in
diameter and 14-14 inch thick in the bottom of the flask. Care
should be taken not to displace the wool pad while placing the oil
and grain into the flask. Cut off the heating unit when the ther-
mometer registers 200° C.

Rye.—Use 100 grams of grain and 150 cubic centimeters of oil and
extinguish the flame when the thermometer reaches 185° C. With
the moisture tester properly standardized and the moisture test made
with dockage-free grain, there appears to be no necessity for the use
of the glass-wool pad as described in the earlier editions of Bulletin
1375, on page 20. However, thermometer guards or cages to mini-
mize superheating may sometimes be employed to advantage.

Shelled peanuts.—Use 100 grams of shelled peanuts and 150 cubic
centimeters of oil, and extinguish the flame when the thermometer
registers 175° C.

Wheat.—Classes Hard Red Winter, Hard Red Spring, Durum,
and Red Durum. Use 100 grams of grain and 150 cubic centimeters
of oil, and extinguish the flame when the thermometer registers
180° C. No change from procedure heretofore used.

Wheat.—Classes Soft Red Winter, and White. Use 100 grams of
grain and 150 cubic centimeters of oil and extinguish the flame when
the thermometer registers 190° C. ,

Tabulative specifications and special points for consideration are
given in the summarization on page 42.

STANDARDIZING THE MOISTURE TESTER

The earlier directions for operating the Brown-Duvel moisture
tester called for adjusting the flame so that it would require about
20 minutes to reach the temperature prescribed for the substance
being tested. This required the constant attention of the operator,
with frequent adjustments of the air and gas valves on the burners,
while the moisture tests were being made. At intervals the tests
would be running too rapidly, at other times not fast enough. Sev-
eral years of observation have shown that these recommendations
have frequently been overlooked. Often when tests are started, no
further attention is given them until the flame is extinguished when
the temperature prescribed for the substance being tested is reached,
regardless of whether it took the right number of minutes or not.
This leads to irregular results.

Errors resulting from incorrect heating time are illustrated in
Figure 17, wherein the heating time has been varied with tests on
different portions of the same sample, with the result that the
moisture obtained from a sample of wheat varied by as much as 0.4
per cent. The data further showed that, other things being equal,
as the heating time of the moisture test increased beyond the correct
number of minutes, there was a corresponding decrease in moisture-
test results. Furthermore, if the heating time was fast, the moisture

THE BROWN-DUVEL MOISTURE TESTER ae |

tests showed not only too much moisture but were more irregular
than when the tests were made at a slower heating time. Moreover,
careful investigations in this field showed that for certain moisture
contents, particularly low-moisture contents, 20 minutes was too long
to heat the sample, whereas the reverse was true if the moisture con-
tent was high.

It therefore became a necessity in order to obtain accuracy and
uniformity of results, to standardize the heating time of each in-
dividual burner. The standard chosen was 20 minutes, the same as
in the earlier method. In these experiments, however, the standard
heating time was made specific for only one moisture content;
namely, 18 per cent, and the grain was specified as corn. This per-
centage of moisture was chosen because it was known that at this
figure the moisture in corn is uniformly distributed, and this figure
is halfway between the maximum and minimum grade requirements
for the Federal grades for corn.

J/2 Minutes
SERIES 15.8% Moisture

44447 777 747

SAtsttalttttt » 4477

ALA GAS SALAS Lh py
A 2 44/477 4

16.8 7, Moisture
47 7 Fa 4477 47 4 444 SIGS
/ 4/7 = AAA
See cays eee CO MINUTES Gat Tt ttTTLTEET
TWO Z Zz Z£422£24272.
| 16.6 7. Morsture
SICAL IALSA PLAST SILAILALASASAS 7 7 7
44744 44°74 26 Minutes 27770 O OTE OCIS OCCT OL tle Lelst tts lls slestsses se
S4OLINAS AAT SOOPLLPLLALEFEAFTFTAATELAALAAALALOAAAAAASAASAAE
THREE a SALLI LSA GS SALAS SASLALSAECLL ve 4
| 16.5 7, Morsture
7 SAA AAAS 7 77
44444 SOCLTIY Mi 4 AAA AAA
SASS ARAAT LL {nu CSAIL ILOALTTLLALALLELALLOLOSDALAEOLLPEOLEOAPLOAAAAAALLAA ALOE
FOUR ZLZLLZRL LLZLLALALLLLLILALALLLLLLLL ELLE z

| 16.47. M oi1sture

Heating Time WE 0 isture Test Result

Fic. 17—Effect of heating time on the moisture test result on a sample of wheat

It is apparent that grain that is drier than this (18 per cent) will
take less than 20 minutes to reach 190° C., the proper cut-off temper-
ature for corn, and likewise that grain having moisture in excess of
18 per cent will take a somewhat longer time than 20 minutes,
roughly one-half minute for every 1 per cent of moisture over and
above 18 per cent. In other words, in place of the earlier method of
trying to manipulate the burners to make them consume 20 minutes
of time to reach a specified temperature, regardless of the moisture
content of the material being tested, a standard of heat has been de-
fined equivalent to that which will drive the moisture out of 18 per
cent corn in exactly 20 minutes. Realizing that one would have to
know accurately beforehand whether he had 18 per cent corn in
order to standardize his tester; and, knowing that this type of corn
is not always easy to obtain, a simple method for standardizing the
gas flame has been developed that gives equivalent results and at the
same time is easy of application. This method is generally referred
to as the “ oil test.”

22 BULLETIN 1375, U. S. DEPARTMENT OF AGRICULTURE
OIL TEST FOR STANDARDIZING GAS FLAMES

To standardize the heating time for making the moisture test pro-
ceed as follows: Place 450 cubic centimeters of oil (the regular
moisture testing oil) of room temperature into the flasks of each
compartment of the tester. This oil should be accurately measured
in a cylinder and not by three trips of the oil measuring device.
Insert the thermometer and regulate the mercury bulb so that it is
completely immersed and the top of it just flush with the surface of

y

?
vine LLY
ete
le fase
fe

LT)
aay fatty
THY]

KAY AD LTH MIT} H

Ms

Fic. 18.—Bristol recording gauge chart of gas pressure over a 24-hour period at
the Grain Research Laboratory at Washington, D. C. The chart also shows
the uniformity of the gas pressure obtained by use of a gas pressure regulator
attached to the moisture tester

the oil. Connect the flasks with the condensing tube, replace the
covers, light the gas, and note the time. Apply heat until the tem-
perature of the oil reaches 153° C. above the original temperature
of the oil, and again note time. The time elapsed should be 20 min-
utes. If more than 20 minutes is required to reach 153° C. above
the original temperature of the oil, the holes in the bases of the
burners should be slightly opened; if less than 20 minutes, the holes
should be closed somewhat. The holes may be closed by lightly
tapping the top of the base with a small hammer or any blunt instru-
ment. They may be opened with a suitable reaming tool, such as a

g

THE BROWN-DUVEL MOISTURE TESTER 23

fine brooch, used by jewelers. Trial tests should be continued until
the required time for heating all compartments does not vary mor

than one-half minute from 20 minutes; that is, not less than 1914
minutes or more than 2014 minutes, should be necessary to raise the
temperature of the 450 cubic centimeters of oil to 175° C.

The only way in which this check test differs from the making of
an ordinary moisture test is that 450 cubic centimeters of oil are
used instead of 150 cubic centimeters, and that the thermometer is
adjusted so that the top of the mercury bulb is just flush with the
surface of the oil rather than being four-fifths immersed.

If an automatic gas-pressure regulator is used, and the correct
heating time for each burner has been adjusted, as described on
page 22, a uniform heating time for all compartments is assured.
If, however, a gas governor is not available, it will be necessary for
the operator to fix in his mind the flame necessary to heat the tester
in the proper time, and, in case the gas pressure varies, to com-
pensate for this by adjusting the keys and air valves at the base of
the burner.

VARIATIONS IN HEATING TIME CAUSED BY VARIATIONS IN GAS
PRESSURE

Originally it was found difficult to maintain a standard heating
time. In making a study of the reason for this, several factors ap-
peared. The gas pressure on the supply hne of a moisture tester
varies at different periods of the day, at different seasons of the
year, for different cities, and with the number of burners in opera-
tion. From Figure 18, which is a Bristol recording gauge chart of
gas pressure on the supply line of the moisture tester installed in the
grain chemical research laboratory at Washington, D. C., it will be
seen that the gas pressure at this laboratory varies from 6.7 inches
to 5 inches, or a range of 1.7 inches during the working period of
the day. Table 1 gives the gas pressure that was found on the
supply line of moisture testers at 13 field offices of Federal grain
supervision :

TABLE 1.—Variation in the gas pressure on the supply lines for moisture testers
located at different cities

Average Average
gas gas
pressure pressure
on on
Federal grain supervision office located pu Federal grain supervision office located supply
at— F at— F
moisture | moisture
testers in || testers in
inches inches
head of head of
water | water
LETIES OD 0y3 2s ee a ee ee | Ae On W Whines DOS <= 28-252 5.1 5. 5 5.6
wipeeiaes i Tig td ae 7 Ou Milwatikce. 85 25 es | 5.0
“DCT; ee eee a APS Omahinwecet <2 earns eked SN Stel eect 4.6
STEUD TEED | 7 a el ia Od gee ANS 1 | S20) ea ES SS ee Ree er 5.6
aL Se en ee ae oe a ee ioe Stee OUIS hom 00S ee on Soe Le Zeige 4.0
MMRERESONAS 95 2s ne 2 a Bo OMG 02 oars re 8 ee 9.0
ESAS (GAL. 22-2 seo 2 a ESS | 4.0 ||

24

BULLETIN 1375, U. S. DEPARTMENT OF AGRICULTURE

TABLE 2.—Variation in the gas pressure in the supply line on a moisture tester
when a varying number of gas burners are in operation

Gas Gas
pressure pressure
on on
Number of burners in operation supply Number of burners in operation aveply
inches inches
head of head of
water water
Oo 22. Ce ee eer eet AAG Wed 8 8 ee ek . eA Peo AS ae Se 2.3
| Pe arene OSES SS ite td a ao ca | he! en nr es ee ees 2.0
ee et ee ee a fet PARE A 5A 1 SOP eh noe a lent te oe al A os ie 13
As 2 4 eee) oes Pe eee Fee Fro) | itl be Se ote ee ee eS ee NS 2
Be a en 2.6
|

Table 2 gives the results of tests of the gas pressures on the supply
line of a 12- -compartment moisture tester with varying numbers of
burners in operation.

Other things being equal, a low gas pressure will result in a longer
heating time and in a lower moisture test, and a high gas pressure
will result in a shorter heating time and in a higher moisture test.
From tests made, data for which are given in Table 3 and illustrated

Eitinutes Correct Heating Time

10.5% ii asia

= RExK ao xx
eee alex /
xxx xx x

an
Correct Moisture Test

GAS PRESSURE ,.ERROR.
3.6 INCHES
~*~

3.2 INCHES

a or

2.2 INCHES

eEErECre ‘77 Minutes. BEETLE EEE ESET EEE
A XK XKKK KKK KKK KKK KKK MK RK AA KK RAR

1.2 INCHES

V4.7 7 Moisture

KA KK KKK RAK KKK KK KARR
MRA KK KKARRKK KK KKM

*x

KXKKX Ks 2/7 Min 4
METHOD perenne 7 Minutes RAK KKKKKKKAK KKK KKK A
WITH GAS

STANDARD gee ae 7
REGULATOR

CORRECT [°
SK ESF =

Heating Time BBS Moisture

Fig. 19.—Effect of gas pressure on the heating time and the moisture test result

W773 Moisture

in Figure 19, it was found that when the burners of the compart-
ments were set to give the correct heating time at the minimum gas
pressure on the supply line, which in this case was 1.2 inches, the
heating time was reduced and the moisture test result was increased
very appreciably when the gas pressure was increased. It will be
noted that the moisture test ‘result increased in direct proportion to
the increase in the gas pressure.

THE BROWN-DUVEL MOISTURE TESTER 25

TaBLeE 3.—Variations in gas pressure affect the heating time and cause varia-
tions in moisture test results

{Excessive pressure results in too fast heating time and too high moisture test. Burners adjusted to
1.2 inches gas pressure]

| Average error
from correct
method

Average Maximum Minimum Range

Mois-

| |
‘Heating Mois- lteating Mois- |Heating Mois- |Heating Heating! Mois-

time ture time ture time ture time ture time | ture
eee Eee Caer gst ry
Inches Min. |Per a, Min. Per cent; Min. |Per cent] Min. |Percent| Min. |Per cent
pe ee ee 17 Tu 18 11.9 16 11.6 2 0. 3 0 0
(Sb es eee 13 12:4 14 12.4 12 11.8 2 0. 6 —4 4
ie es eee 11 12-3 11 1 Ay 10 12. 0 1 0. 7 —6 6
2. i ae 10 12.6 10 13. 0 fe) 12-1 2 0.9 —7 e)
Correct (1.2 inches)
with gas regulator
eantrole #2. -- 17 be 17 11.9 16 11.6 1 O53) |s--2ee2|-=

NotEe.—Data given in this table summarize 18 tests for each zgas-pressure series on the same sample of
wheat.

Conversely, if the burners are adjusted to give the correct heat-
ing time at a higher gas pressure on the supply lines and the gas
pressure decreases, a corresponding increase in the time of heating
and a decrease in moisture results. Such variation in pressure must
be overcome to give accurate results.

GAS PRESSURE REGULATOR

Variations in gas pressure can be overcome by the use of a gas
pressure regulator, commonly called a gas governor. One type of
gas governor is shown in Figure 1, attached to a moisture tester,
and diagrammatically in Figure 20.

In operation, the gas pressure regulator, when once correctly ad-
justed, will deliver gas to the burners of the tester at a constant pres-
sure regardless of pressure variations on the gas supply line, or of
variations in the number of burners in operation simultaneously ;
whereas, without the use of a gas pressure regulator, it is necessary
for correct and uniform moisture test results to adjust the gas valve
which regulates the gas supply (pressure) to the burners of the
moisture tester at different periods of the day and to adjust the
flame for every burner still in use each time that one of the burners
is cut off.

Table 4 gives the correct size of gas pressure regulator for moisture
testers having varying numbers of burners. The table also gives
the various sizes of gas supply lines to be used and the minimum gas
pressure on the supply lines at which the regulators will operate.
The size of regulator and pipe lines for moisture testers should con-
form to the specifications given in the table to insure against possible
overtaxing of the gas supply.

The regulator may be installed at any place on the gas supply line
between the meter and the moisture tester, although | it is desirable
to place it as close to the tester as is possible. A gas pressure regula-
tor, however, is not capable of compensating for changes in the
British thermal units, or for changes in the density of the gas sup-
plied the consumer. If these change, the heating time will “change,
and the burners in the moisture tester will have to be restandardized.

°6 BULLETIN 1375, U. S. DEPARTMENT OF AGRICULTURE

TABLE 4.—Correct sizes of gas pressure regulators for moisture tester or testers
having various number of burners, varying sizes of gas supply lines, and the
minimum gas pressure on the supply line at which the regulators will
operate

Size of regulator
Required | Minimum

| size of pressure

Number of burners Inside supply pipe] at which

diameter | Trade desig- (inside regulator
of inlet nation diameter) |will operate

and outlet
Inches Inches Inches

Diss doe ee Se Pee ee ee ee 34| 5-light______- % 0.3
Aes Se see ee ee ee eee ee A ees a 6 dot se" 34 .3
6225) 2osc ol ee ee eee 3422-3 doz: 245 34 .3
49 ce os ee ee ee eee ae 1 | 10-lights- 1 3
18: Suto bee. ee Se ee eee A Wise dol === uf ao
DAE eS ee ee 114] 30-light__ __- 1% a
B02 22 22-2 ole ee ee ee ee LEA eae Gonaka: 14 -3

NotE.—Gas pressure measured in inches head of water.

HOW TO ADJUST A GAS PRESSURE REGULATOR

After the pressure regulator has been installed on the supply line
as near to the tester as possible it must be adjusted to give a standard
heating time. The following method of procedure in making such
an adjustment is necessary :

(1) Overhaul the moisture tester thoroughly by:

(a) Replacing all cracked or burned-off wire gauzes with new
ones and adjusting them so that the flame will strike the asbestos
center of the gauze.

(>) Adjust the asbestos ring of the flask rest so that the bottom of
the flask is approximately three-eighths inch above the wire gauze.

(c) Take down each burner and clean it thoroughly. In this oper-
ation special care should be given to the cleaning of grids in the top
of the burner, the needle hole in the burner base, and the burner
valve.

(7) Test all gas connections for leaks, and if any are found tighten
or replace the fittings.

(2) Determine the gas pressure on the supply line of the tester or
testers when all burners are in operation, by the use of a pressure
gauge connected to the supply line and which will register gas pres-
sure in inches head of water. To do this, throw the regulator out of
operation by pressing down the valve in the top.

(3) With the pressure gauge still connected to the supply line and
all burners in operation, release the valve of the regulator.

(4) Place a weight on the valve that will set the regulator to
deliver gas at a pressure of approximately 2.5 inches less (as deter-
mined by readings of the pressure gauge) than the gas pressure as
determined in item 2 above.

(5) Make a preliminary heating time test (oil test) on every com-
partment of the tester or testers, with the flasks containing 450 cubic
centimeters of oil alone and the mercury bulbs of the thermometers
completely buried in the oil, as described on page 22. If the 450
cubic centimeters of oil in all compartments heat to 175° C. in less
than 20 minutes, place a lighter weight on the valve in the regulator.
Repeat the tests, adjusting the weight on the regulator after each

THE BROWN-DUVEL MOISTURE TESTER Dip

trial until one or more compartments show the correct heating time.
If, on the other hand, the 450 cubic centimeters of oil show a heating
time of more than 20 minutes to reach 175° C., the needle holes in
the bases of the burners should be opened as explained in item 6
following. Repeat the oil test, opening or closing the needle holes
after each test until one or more compartments have the correct heat-
ing time. For all heating time trial tests use 011 at room temperature,
and in every instance the machines should be given time to cool
thoroughly before the next trial test is made.

GAS PRESSURE REGULATOR

Screw cover Air vert in cover

Upper float

Upper float casting

4

ehhh heh deheh AAAAM ALARA REALE AL LAAIIAT

RM HOoo os

Mercvry sea/ 32 ag
Upper f/oar. He EY)

AES

MA KG

Sw

Lower tloar —
and va/ve

PITTI 72

Mercury sea/
Lower Float

Ss
=.
Rs
=
=
a
ma
=
~
im
ce
=
=
=
we
=
=

Fic. 20.—Sectional view of one form of gas pressure regulator

(6) When the preliminary adjustments have been made so that
one or more burners of each moisture tester will heat the 450 cubic
centimeters of oil to a temperature of 175° C. in 20 minutes, the
remaining burners should be adjusted, with the corrected burners as
a standard, to show the correct heating time. These adjustments are
made by opening or closing the needle hole in the base of the burner.
If the burner is too fast, the hole may be closed by lightly tapping
the top of the base with a small hammer or any blunt tool. If the
burner is too slow, the hole may be opened with a suitable reaming
tool, such as the fine brooch used by jewelers. |

28 BULLETIN 1375, U. S. DEPARTMENT OF AGRICULTURE

(7) Trial tests should be repeated ‘until the heating time for all
compartments does not vary more than half a minute “from 20 min-
utes; that is, the heating time should not be less than 1914 minutes
or more than 2014 minutes to raise the temperature of the 450 cubic
centimeters of oil to 175° C.

VARIATIONS IN HEATING TIME AND MOISTURE-TEST RESULT
CAUSED BY POSITION AND CONDITION OF WIRE GAUZE

The wire gauze used in the bottom of the tester compartment will
cause variations in heating time, with consequent errors in the
moisture test result when the wauzes are burned off, cracked, have

Correct Heating Time
And

15 minutes Moisture Test

11.9% moisture

FA KARE AAALA AIT GAAAAAAARAAA AAA AAAI A AAA
COCPPOPOPPEEPEELELOEEESE
At 18% minutes 7 sooo ieee estes ees SOCATAL

$0 AAAAA SASS SLIP AS SSS APH, AAA AMAAAA AAA

GAUZES 2
IN GOOD 113 22°%e
CONDITION | Seon

cess orton aR
|
:

CF4FALAGE CRICILAECEEL LEILA L,
VARIATION &724: IZ minutes $2332

- [4% MIN. AAA
GAUZES
IN GOOD Eee ara
CONDITION 170

BUT

INCORRECTLY | nance __ TRAM

RIATI N_ POVAOOEPOFTPPEEFLAAAEALGE
VARIATIO 324/62 minutes 275 0etee

- 1% MIN. PP VLA AAA

0.6%

pede ating = MB Average Ry Minimum = [TT Maximum

moisture moisture moisture
Results of 6 tests for each type of Gauzes using the same sample of wheat

Fig, 21.—Effect of the condition and position of the wire gauze in the moisture
tester on the moisture test result

holes burned in the asbestos center or in the gauze itself, or when
they are incorrectly placed.

Data given in Table 5 and illustrated in Figure 21 show that the
use of wire gauzes that are burned off, cracked, have holes burned
in them, or are incorrectly placed will cause a faster heating time
than the standard time, and an increase in the moisture test “result
over the correct result. The results also show that the use of dam-
aged or incorrectly placed gauzes will cause an appreciably wider

range between results of tests than when gauzes are in good condi-
tion and in correct. position.

THE BROWN-DUVEL MOISTURE TESTER 29

TABLE 5.—Damaged or incorrectly placed wire gauzes cause errors in moisture
test results in testing wheat

[Wire gauzes that are damaged or incorrectly placed in the compartments of the moisture tester affect the
heating time of tests and cause variations in moisture results]

Condition of wire gauze

Hood ano cor- ood, ee Baeed shes ood pug cor-
rectly placed striking edge of |tos burned offand} rectly placed
Compartment No. (check) asbestos center | holes in gauze) (check)
Heating | Mois- | Heating | Mois- | Heating | Mois- | Heating | Mois-
time ture time ture time ture time ture
Minutes | Per cent} Minutes |Per cent) Minutes |Per cent) Minutes | Per cent
ESS Ae wt Ls a eee 18%) 13.1 174%} =13.3 e1A 11353 18 13. 0
Jp ae Es IS eee eee eal 18 13.3 ily/ 13. 4 1614) 13. 6 18%) i. 2
eee sere! Se tt ee IBY 1322 184%} 13.6 1734) 13.5 18 1322
7 ee ee ee ee 18% 13. 4 17% Neve 15 13.7 184 | tS.0
Te 58S oh ee ae 184| 13.3 15%} 12.9 15 ile} al 1844 13) 2
encase eee ee mee 18 ley 164| 13.1 17 13:2 18 13:1
PRG ETAC Ger es xa ae 184%| 13.2 17 1323 16%} 13.4 1814 13.2
RUAN Oke at ete 8 eee i 0.3 234 0.8 234 0. 6 X% 0.3
AVEFAGEe GELOF ILOHT, Checks! .6)- === -2--~2|.--2222- —14%| +0.1 sl Alls Se 0 ed aa cea el aes Te

VARIATIONS IN HEATING TIME CAUSED BY VARIATIONS IN
THERMOMETER-BULB IMMERSION

Data in Table 6, illustrated in Figure 22, show that a deeper
immersion than the correct depth of four-fifths immersion of the
mercury bulb of the thermometer in the oil will cause a faster
heating time and a lower percentage of moisture, and that a
shallower depth of immersion will cause a slower heating time and
a higher percentage of moisture.

To insure correct moisture test results from the standpoint of this
factor, the depth of immersion of the mercury bulb of each ther-
mometer should be carefully noted and adjusted to the correct four-
fifths immersion for each test.

TABLE 6.—Variations in the depth of immersion in oil of the thermometer
bulb will affect the heating time and cause errors in the moisture-test
result

Average varia-

Pa wi tion from cor-
Average Maximum Minimum Range rect (four-fifths)

, immersion
Bulb immersion

Heat- | iiois- | He | mois- | 2&8 | Mois- | Hea | mois- | He®™ | mois-

ing ing ing ing ing :
TG ture dime ture tante ture aye ture time ture
Min- Per Min- Per Min- Per Min- Per Min- Ber.
: utes cent utes cent utes cent utes cent utes cent
Bulb touching sur- 19 10.5 1914} 10.6 181% 10.4 1 0.2 234, +0.4
Sblrtaceraeet er ee) ©
Bulb one-fifth im-
RCT SC Oe es 174) 10.3 1734; 10.4 ee O53 34 0.1 1 +0. 2
Bulb one-half im-
MICTSCGHS! 2 8 i == 1634; 10.2 live 10.3 16% 10.0 4% 0.3 4 +01
Bulb four-fifths im- |
mrerse dit. 2 8 = is _ 164% 3810.1 1634] 10.2 16 | 10.0 34 Os ee ae os
Bulb totally im-
Miersedeee 2 e 16 10. 0 164| 10.2 154% 9.9 34 0.3 ~A) ee

1 Correct position.

30 BULLETIN 1375, U. Ss. DEPARTMENT OF AGRICULTURE

Ci t Hear TN
BULB 9.8Minutes orrec see re ime
IMMERSION 90% Moisture Moisture Test
(TOO FAST BY =z
TOUCHING +2.8 MIN. ice catwate
SURFACE |ERROR=+04%
TOO FAST BY 5
ve +1.0 MIN.
ERROR=+0.2%
TOO FAST BY x os
¥, +0.5 MIN. nen L038 Minutes«
ERROR=+0.1% &
CORRECT FOR ¥xr
a THIS SAMPLE
CORRECT
(CORRECT) MOISTURE
TOO SLOW BY
TOTAL -O.3MIN. .
IMMERSION ERROR=-0.1% ‘ 10.0%, Morsture

Tota/ range tor eres time 3./ min

Morsture =0.5%
Exxxs] Heating Time Ps 255] Mossture

22.—Effect of variations in the depth of immersion of the thermometer
bulb in the oil on the moisture-test result

INFLUENCE OF COOLING TIME ON MOISTURE-TEST RESULTS

Fic.

When the flame under the moisture testing flask has been extin-
guished, the material in the flask must be allowed to cool before
making the moisture reading, for the reason that moisture will
continue to distill off the material in the flask for several minutes
after the flame has been extinguished. In case the moisture reading
is made before the temperature has dropped to a point at which
no further moisture will pass into the graduate, the result will be
inaccurate in that it will indicate less moisture than is actually pres-
ent. The rate at which the grain and oil in the flasks cool after the
flame is cut off is also important. Too rapid cooling will cause low
moisture results, and if the compartment covers are removed too soon
low results will also prevail. The thermometer reading should drop
to 160° C., a point where no further accumulation of moisture takes
place, before the reading is made. Table 7 and Figure 23 sum-
marize the errors that result if the moisture readings are made pre-
vious to the thermometer receding to 160° C.

TABLE 7.—/nsufficient cooling time causes low moisture-test results

|

Covers removed

10 minutes after | 5 minutes after Praca ae,
After tests had | cut-off tempera- | cut-off tempera- patrons
cooled to 160° C. | ture was reached | ture was reached oF naa (ther-
(correct method) (thermometer (thermometer mnowmiereranadte
reading 180° C.) | reading 186° C.) 190° C.) 8
Cooling ‘ Cooling +. | Cooling | y7,;.. | Cooling oat
time to ee time to roe time to nas time te os
IBOCIG. || Gags, |) LOU Fe PE Aere Cc. 160° C. |
Minutes | Per cent| Minutes | Per cent! Minutes | Per cent} Minutes | Per cent
AVVerage tc 58 258 ee ee 1844; 18.3 16144) 18.0 15%) 17.9 84 17.5
AVE ARUN Sse eee eee 19 18. 4 ily 18.1 15% 18. 0 9 17.6
Miaininmitan ee ee 17 18.1 16 17.9 1444) 17.7 7% 17.3
(ange .foe- sot he se ee ee 2 0.3 1 0. 2 14 0.3 1% 0.3
Average variation from cor-
rectmethod: ==. 4=)- "2 IL eee —1%4| —0.3 — 234) —0.4 —10 —0.8

LE TT TS a Re AD > eR

a i

THE BROWN-DUVEL MOISTURE TESTER 31

Correct Cooling Time

COMPARTMENT and.
COVERS REMOVED Correct Mo/sture Test Result
AFTER TESTS a2 BEES

COOLED TO 160°C. __} TIME

(THERMOMETER vy = CORRECT

READING 160°C.) TEST

10 MINUTES AFTER VARIATION SxzE™="* a aceaee Saarinen eee

CUT-OFF TEMPERATURE -174% MIN. sinnssiaessneiiis

WAS REACHED =

(THERMOMETER ERROR

READING 180°C) “0.3%

ee ee euxERKY

1574 STRIUIES. Dektiaes

Ru MEKR RRR REKK
=x =x

CUT-OFF TEMPERATURE | 9% win,
WAS REACHED

(THERMOMETER
READING 186°C)

IMMEDIATELY AFTER a

117.9% morsture

5 MINUTES AFTER a

CUT-OFF TEMPERATURE |_i9 qin, 644 min.
WAS REACHED
(THERMOMETER
READING [90° C)

5 minutes
16.75% moisture

Cooling Time

Fic. 23.—Effect of cooling time on the moisture-test result

pig Mo/sture Test Result

PRIMING TESTS

In making a moisture test with the Brown-Duvel tester, all of the

moisture which is distilled from the grain is not obtained in the grad-
uate during the first test with a new flask. Some of it stays behind
on the walls of the glassware. To accomplish accurate results, it is
therefore necessary to make a preliminary or priming test in all the
compartments of a new moisture tester, or when new distillation
flasks are used, before a regular test is made. This is also true when
a machine has been standing idie or standing open for any appreci-
able length of time. The length of time a tester may stand idle, after
which it will be necessary to ‘make a priming test before the regular
test is made, depends somewhat upon the state of the machine when
not in use.

The data, which are summarized in Tables 8 and 9, and shown
graphically in Figures 24 and 25, show that a failure to make a

15.9% Morsture

PRIMING TEST CORRECT
MADE GEST. 16.62% moisture
CORRECT METHOD (CHECK)

TESTER STANDING | ERROR
2 HOURS -0.02%
Comes... Eo SOR
CLOSED TESTER STANDING | ERROR — Qyssssss SSSA
BUT 5 HOURS OE of AS TR RARER ERR
5 JERE i710) ‘seca Ganarme mai L____.- ASS
Satta TESTER STANDING | ERROR NS
20 HOURS -0.20%
MADE NSS

TESTER STANDING
40 HOURS

ERROR 16.33% moisture
-0.29%

15.9% Mossture

Fig. 24.—Necessity of making a priming test before making moisture test after
the moisture tester, with ‘all connections closed, has been standing idle for
any appreciable length of time is here shown

$2 BULLETIN 1375, U. S. DEPARTMENT OF AGRICULTURE

priming test before a regular test, after the machine has stood idle or
disconnected for an appreciable length of time, will cause low mois-
ture-test results, the amount that the results are low depending upon
the time that the tester has stood idle. The method for avoiding
errors from this source is to connect the apparatus, as for use, im-
mediately after every test is completed, and to make a priming test
before making a regular test every time new glassware is used or
when the tester has stood idle for any considerable length of time.

TABLE 8.—Failure to make a priming test before the regular test, after the

machine has been standing idle with all connections made, causes a low
moisture-test result

No priming test made

Priming
test made |
by correct Tester Tester Tester Tester
method standing standing standing standing
(per cent 2 hours 5 hours 20 hours 40 hours

moisture) | (per cent (per cent (per cent | (per cent
moisture) | moisture) | moisture) | moisture)

Average: (23> i354 eee eee 16. 62 16. 60 16. 48 16. 42 16. 33
Maximum." 3 22 hae are ee 16.7 16.7 16.7 16.5 16.5
Minimum 5 ee ee ee ee 16. 4 16. 5 16.3 16.3 16.1
Range: 4 2S ar ee eee 0.3 2 4 Me 4
A verage error from correct method-______--]_._.-----_.- —. 02 —. 14 —.20 —.29

Taste 9.—Failure to make a priming test before the regular test, after the
machine has been standing open with no connections, causes a low moisture
result

No priming test made
Priming :
| test made |
by correct | Tester Tester Tester | ‘Tester Tester
method standing standing | standing standing | standing
(per cent 1 hour 3 hours 6 hours | 20 hours 40 hours
moisture) (per cent (per cent (per cent | (per cent (per cent
moisture) | moisture) | moisture) | moisture) | moisture)
Avene 28552 185 eet te pe 10. 57 10. 50 10. 50 10. 38 | 10. 33 10. 15
Maximise ee 10. 6 10. 6 10. 6 1085 10.5 10. 4
Minimise ee eee 10.5 10. 3 10. 4 10. 2 10.0 10.0
Rangese2: tees Ae be ee wal .3 7% .3 5 4
Average error from correct
Tmephods =< sy Aa ee ee —.07 —.07 —.19 —.24 —. 42
9.8% Moisture
“PRIMING TEST | CORRECT
MADE TEST 10.57% moisture

CORRECT METHOD |! (CHECK)

;

Ses es ee ee Terror Y

TESTER STANDING | ERROR

| HOUR ! 0.07%

SSS Sa SSE SSR ree (70.07%
CONNECTIONS | tester STANDING | ERROR
LEFT 3 HOURS 1 -0.07%
OPEN) 27) 6 vit) os ae Se ee ---------4
AND TESTER STANDING ! ERROR
NO 6 HOURS 1 -0.19%
PRIMING Tio an Loe Se Sy ae Sl SR So Se DO
TEST TESTER STANDING | ERROR Ae Pegg ons con a ey
MADE 20 HOURS “0.24% f.4/O:3F% MOISTUre sree eres

erie en Pues. «06, nels Malte, Ce ess Kas

TESTER STANDING : ERROR
40 HOURS 1 -0-42%

9.8% Moisture

aeodesanma =

Fic. 25.—Errors resulting from failure to make a priming test before a regular
test, after the tester has stood idle and disconnected for one hour or longer,
are here shown

THE BROWN-DUVEL MOISTURE TESTER 33
VOLUME OF OIL VERSUS MOISTURE-TEST RESULT

Variation in the volume of oil used in making a moisture test
may cause variation in the moisture-test result. Five series of 12
moisture tests each were made on the same sample of corn, using
varying volumes of oil in each series.

The data, as shown in Table 10, indicate that there is an increase
in the moisture-test result when less than 150 cubic centimeters of oil
is used but show no appreciable variation in the moisture-test result
from using slightly more than 150 cubic centimeters of oil.

TABLE 10.—A decrease in oil volume below 150 cubic centimeters causes an
inerease in the moisture-test result; an increase in oil volume above 150
cubic centimeters will only slightly effect the moisture-test result

Volume of oil used

|

l15e.c. | 185¢.¢. Cones 165¢.c. | 185¢.c.
(per cent | (per cent (per cent | (per cent

moisture) | moisture) ae moisture) | moisture)
TAP Ceres eee, nae ab ee a | 19. 03 18. 98 18. 86 18. 85 18. 83
iy Reece t) o 0) ae De a a ee 19. 2. 19.1 19. 1 19.1 19.1
Mlaaviebeti et eS Wad FED AS ee Ot 18.9 18.8 18. 7 18. 6 18. 7
pre eee e fe 8 Se ee .3 3 5 ot
Average variation from standard 150 c. ¢. |
Spriesiees ca ey Whe Ta Se Bee opin +.17 72 | eae a eee —.01 | —. 03

BROWN-DUVEL MOISTURE TESTERS EQUIPPED WITH ELECTRIC
HEATERS

Reports are frequently received stating that it is difficult to ob-
tain the same moisture-test results when making comparative mois-
ture tests on grain with the gas-heated and the electrically-heated
Brown-Duvel moisture testers. In most instances, the reports show
that the moisture tests on the electric machines run lower than the
tests on the standard gas machines, although there are instances at
some of the markets where the tests on the electrically-heated tester
give moisture-test results higher than the results obtained when made
on the standard gas tester, some of them reported as being higher
by over 2 per cent.

To determine the efficiency of the electrically-heated tester and
the reasons for its failure always to check results obtained on the
standard gas machine, investigations were conducted for the pur-
pose of studying and comparing the factors which influence mois-
ture-test results as made with the electrically-heated tester. These
investigations covered such points as the types of heaters now em-
ployed (in 1925), their heat capacity, their efficiency in transmitting
heat to the moisture flask, their durability, the nature and uniformity
of the electric service supplied, loss of current due to conditions of
wiring, and how fluctuations or lack of voltage and current will
affect the wattage or heat developed. Finally, how lowered voltage
affects the time of making the tests and how this time factor in-
fluences the moisture-test result.

34 BULLETIN 1375, U. S. DEPARTMENT OF AGRICULTURE
ELECTRICAL TERMS

Attention is called to some facts that should be generally known
about electricity before an electrically heated moisture tester is in-
stalled. To illustrate some of the factors involved, consider elec-
tricity as water. The size of the stream of water flowing through
a pipe would be indicated in “amperes” for electricity. The pres-
sure of the water, or the speed with which it flows, is measured in
“volts” for electricity. The quantity of water passing through, or
the gallons of water flowing, is measured in “ watts” for electricity.
The term “kilowatts” merely means 1,000 watts by the regular
metric system. The term “kilowatt hour,” means a kilowatt quan-
tity of electric current used for an hour’s time to light or blow
breezes or fry an egg or sweep the floor or run a moisture machine,
or to do anything else.

The large quantity of current consumed by relatively small elec-
trical heating units is generally not appreciated by the ordinary
layman not versed in electricity. Take for example, one 6-compart-
ment moisture testing machine heated by electricity. Each compart-
ment requires 5.4 amperes of current at 110 volts pressure or 594
watts, making 3,564 watts for the whole machine. This amount of
energy is sufficient to ight 125 ordinary 25-watt electric lamps. It
would generate mechanical power at the rate of almost 5 horse-
power.

TYPES OF HEATERS NOW IN USE

The electric heaters in use in 1925 are of two general types. One
of these general types consists of a sheet-metal box 5 by 5 inches
square, 114 inches deep, the upper side of which is open except for
a heavy wire netting which allows the developed heat to rise into
the compartment of the tester. The heating elements or resistance
coils of this type of heater stretch between two porcelain barriers.
within the box. The heaters are fastened to the outer edge of the
tester compartment with hinges and are held in place under the com-
partment by a spring metal clip at the rear. On the floor of the
heater is a thin sheet of asbestos which prevents the resistance coils
from touching the metal boxes and which functions at the same time
as a means of keeping some heat from being lost to the outside. As
a rule, these heaters are wired to take a voltage or pressure of 110
or 220 volts and the resistance wire within the heater is made of
such diameter and length as to take a current of 4.6 amperes for
the 110-volt current or 2.8 amperes for the 220-volt circuit. Ex-
pressed differently, they are so wired that they will develop 500
watts of heat energy per heater. It will be shown later, however,
that 500 watts will not develop sufficient heat to give a standard heat-
ing time of 20 minutes.

Experiments have demonstrated that because of the excessive heat
leakage from this type of heater, caused by lack of insulating mate-
rial at the base and sides of the heater, not more than 25 per cent of
the possible heat energy developed by the heater is transmitted to
the compartment of the tester when this heater is in operation.
Furthermore, since this type of heater has very little heat-retaining
capacity, it has to be used on a constant and uniform electric circuit

THE BROWN-DUVEL MOISTURE TESTER 35

where the specified voltage and amperage is always at the maximum
point. As the heat energy developed is proportionate to the voltage
supplied, a lag or drop in the voltage for a few minutes will be
readily noticed when this type of heater is used, inasmuch as there
is no provision for retaining any heat for an emergency period, as
would be the case if the inside of the heaters were well insulated with
asbestos or fire clay.

Another heater of this general style, but of shghtly different de-
sign, is used to some extent. This is identical in all respects with
the one just described, with the exception that both the top and
bottom of the box are absent and the coils are suspended between
mica supports within the walls of the box. Necessarily less of the
possible heat generated is transmitted to the compartments of the
tester by this type of heater than with the one described before.
These heaters are being gradually eliminated from service, and for
moisture-testing purposes should be eliminated entirely.

The second type of heater in general use consists of a reinforced fire-
clay block 414 by 4% by 1 inch deep with recessed portions on the
upper side for installing the heating coils. These heaters are wired
for 110 or 220-volt circuits, but the resistance coils are of such a size
that they take only 3.2 amperes of current, and therefore are possible
of developing only 350 watts of energy. Because of their fire-clay
structure, much more of the heat energy is transmitted to the com-
partments of the tester from this heater than from the other type.
The heater swings on an improvised hinge attached to the inside of
the back of the compartment and is raised and lowered by means of
a cam attached to a rod.

ELECTRICAL ENERGY FOR STANDARD HEATER

From a theoretical standpoint, enough heat will be supplied from
a heater properly constructed and delivering 300 watts of energy
to operate a moisture tester properly, providing none of the heat is
lost during the interval of the test. But, as a practical matter, the
electric heaters commonly used to-day in moisture testers must be
wired to deliver approximately twice this amount of heat energy, or
600 watts, on account of the excessive heat leakage which they
permit.

WIRING FOR ELECTRICALLY-HEATED MOISTURE MACHINES

Observations have shown that although these heaters are wired to
take either a 110 or 220-volt circuit, it is not always that the specified
voltage is delivered in every installation to the switch board of the
moisture tester. Observations have shown that the voltage supplied
is usually below 110 or 220, apparently for two reasons.

In the first place, the necessary voltage is not delivered to the
moisture-testing laboratory by the electric service companies. This
is notably true at certain large terminal or export elevators which are
located from 1 to 10 miles from the generating plant. Moreover,
many isolated plants furnish their own electrical energy, and the
required 110 volts are not always available. As a general rule, large
city service is rather regular, although there are notable exceptions.

86 BULLETIN 1375, U. S. DEPARTMENT OF AGRICULTURE

Electric service is usually delivered to large terminal elevators at
high voltage, such as 550, 1,100, or 2,200 volts, and is then cut down
by means of a transformer in the power plant to the subvoltage re-
quired. Instances of this are cuts to 440 volts, as used by some ele-
vators to run the belts, and 110 or 220 for lighting or other low-power
service. A number of instances have been noted where the voltage
received is considerably under that expected, in which case the differ-
ence is automatically carried forward when a subdivision of the
original voltage is made at several points. Tests on electric circuits
before entering the heaters of moisture testers have given figures
varying from 95 to 115 volts on 110-volt circuits.

There always is an additional drop in pressure or in voltage as
the heaters are put into operation, provided nothing is done to keep
the voltage bolstered up to the point specified on the heating appa-
ratus. If one will think of a large tank of water of definite volume
with a series of spigots turned on, one after the other, it will soon
become apparent that the rate or speed at which any spigot, or series
of spigots for that matter, will flow is decidedly influenced by the
number of other spigots running at the same time. This is more
and more noticeable as the tank empties. If the tank receives a
renewed supply of water as fast as it runs out at the base, the rate
of flow will be constant. Electric pressure is influenced in the same
way. All other interfering factors eliminated, if the voltage remains
constant, the rate of flow cf electric energy or the watts will also
remain constant.

The second important factor regarding drops in voltage is found
in the wiring conditions from the power station or from the trans-
forming box to the machine. Wiring circuits installed in buildings
are usually designed to carry current for a certain number of electric
lamps, usually not more than ten 100-watt lamps. In such case this
number, or the equivalent current, should not be exceeded. TZ'o use
an ordinary lighting circuit for a four or six compartment moisture
tester is not only entirely unsatisfactory as far as moisture testing
is concerned, but zs a very dangerous fire risk. The danger comes
from excessive heat developed in the line by forcing too large a cur-
rent of electricity over a small wire. This feature is taken care of
by the national underwriters, who specify certain maximum allow-
able current strength for all wires used in interior wiring.

Table 11 shows the allowable current for both “rubber” and
“ other than rubber ” insulated wires. The values for wires of various
sizes are given in the first two horizontal columns of the table. These
values should never be exceeded. The current and voltage at which
an electrically heated moisture machine is designed to work is usually
given on a plate attached to the machine as it comes from the factory.
The vertical column in Table 11, headed “Amperes required by
machine,” gives the amperes required by one of the common types of
electrically heated testers, of two, four, and six compartments.

The application of the table as to length of circuit can best be
shown by an example. Assume a 6-compartment moisture testing
machine operated on 220 volts, located so that it will require a
circuit 40 feet long. In Table 11, under the subheading “ For 220
volt circuits,” read along the horizontal line starting with six com-

THE BROWN-DUVEL MOISTURE TESTER 37

partments, amperes required, 13.8. A size 14 wire is large enough
to carry the current if it is “other than rubber insulated,” but 30
feet of this wire will give a line drop in voltage of 1 per cent so
that it would not be desirable to use that size. Size 12 requires 48
feet to give a 1 per cent drop, hence, would be the size to select.
Should it be desired to decrease further the drop in voltage, any
larger size wire could be used, as for instance a No. 6 wire, and the
drop determined by dividing ‘40 by the length for 1 per cent drop,
or 193, giving practically 0. 2 per cent drop.

It is important that all wiring for electrically-heated moisture
machines be done by a competent electrician. Fiven then he should
be impressed with the fact that the voltage at the machine must be
kept as nearly constant as possible, whether one compartment or
several dozen are operating.

TABLE 11.—Wiring for electrically heated moisture machines

Size of wire B. & S. gauge

Amperes allowed, insulation other than rubber --_---------------- 16 23 82 46 65 77
Amperes allewed; rubber covered: —2 2.) - =~ -22.2. 2228.22 s-_-=--- 12 17 24 33 46 54
Amperes
peared Length of circuit for 1 per cent drop
machine
On 110-volt circuit: Feet | Feet | Feet | Feet | Feet | Feet
PRCOMIIAR I AHONLS. == es oe Po oe oo aL 9. 2 22 36 57 91 | 144 182
SICOrmipnr LMIGnis: 2555 ete eet est ee ret ahs ar a (ee 18 28 45 72 91
SieCOTIATRNCTES © oo es Foe 8 See PARAS | eae eee ee 19 30 48 60

On 220-volt circuit:

PAC ATIATICN GS = = 52 28 OSS 8 4.6 OOM ease) 228 se eee. oe eee
a: CPCI TEES AES V0 Es ee ts ee ee eee 9.2 45 a2 \¢ TAA P82) | P290F 22202.
Greompariatents* Oo. 4 Ty ts. eyed 2.8 4 ls ek 13.8 80 48 76 | 121-} - 193 244

Values in italic figures refer to ‘‘insulation other than rubber’’; all other values refer to either method of
insulation.

EFFECT OF LOW VOLTAGE ON MOISTURE-TEST RESULTS

Investigations have shown that there is a wide variation in the
electric pressure available at various markets. Studies were made
to determine what influence that fluctuation or drop in voltage has
on the heating time in making moisture tests with testers equipped
with electric heaters and its effect on the moisture-test result. Ex-
periments to determine the influence of a drop in voltage on the
heating time were made at the electric measurement division of
the United States Bureau of Standards. The results obtained from
these tests are given in Table 12. The table shows the maximum,
minimum, average, and range in results of all series of tests on both
the electric testers and the gas-heated testers. The table also shows
the minus or plus variations, in time of heating and percentage of
moisture obtained for all voltage series on the electric tester, from
the same series of samples w hen tested on the standard gas-heated
tester.

38 BULLETIN 1375, U. S. DEPARTMENT OF AGRICULTURE

TABLE 12.—Influence of low voltage on the moisture-test results

110-volt circuit: | (1)
Voltages 2 <2 8. 90 95 100 105 110 112 TG). oy ee
AT perage.-. 22.222 24. 4 25. 6 26. 8 28. 4 29. 6 30. 4 SIS OMe e s.e eee
Wattage: 2235.) _ “eet 2, 196 2, 432 2, 680 2, 982 3, 256 3, 420 3,060) 3-2? 2 ee
220-volt circuit:
Voltages = 22545. sr ts3 180 190 200 210 220 225 2307) 292s feat ss
Amperage-__- SP ote 1 DA 12.8 13. 4 14. 2 14.8 152 15. 5
WattagOn- sehen 2, 196 2, 432 2, 680 2, 982 3, 256 3, 420 3, 000) |. See See we
Maximum:
Time (minutes) _______ 331% 31 271% 25 23 21% 201%
; Moisture (per cent) ___ 16.9 17.0 Uzi pk 762 ala} i763} V7E5
Minimum:
Time (minutes) --_-_-- 30144 281% 2334 2234 2034 194% 174% 19
Moisture (per cent) - _- 16. 6 16.8 16.9 16.9 U7. 1s VAP 17.3 17.3
Average: :
Time (minutes) -___-_- 31% 2914 2614 24 22 2014 19 19%
Moisture (per cent) _ __ 16. 8 16.9 17.0 ital tenes 17.3 17.4 17.4
Hang: Gane
ime (minutes) --__--. 3 24% 234 214 2144 2 3 3
Moisture (per cent) _ __ 0.3 0.3 0.3 0. 3° 0. 2° 0.1 Oz ic

Average variation, stand-
ard gas tester:

Time (minutes) --_---- 12 10 634 44 2% 34 —Volti i sees

Moisture (per cent) __- —0.6 —0.5 —0.4 —0.3 —0.2 —0;.54|\ Brace alae ee

1 Results obtained with standard gas tester.

The results obtained in these experiments with a Brown-Duvel
moisture tester, electrically heated, operated at varying voltages,
show that as the voltage and wattage increase to optimum condi-
tions the time of heating decreases, and that the percentage of mois-
ture obtained increases. It shows, also that the heating time is de-
pendent upon the wattage developed and delivered by the heater. A
drop in voltage of over 5 per cent will cause a decrease in wattage
sufficient to cause appreciable variations in moisture-test results and
these results will always be low.

The experiments also demonstrated that the results obtained with
the electrically-heated tester can be made to check with the results
obtained with a gas-heated tester by means of wiring the electric
heaters to such a capacity that each heater will develop approxi-
mately 600 watts of energy per heater, or 3,600 watts per 6-compart-
ment machine. To attain these results on a constant pressure, 110-
volt circuit, the heaters must be wired to carry 5.4 amperes of cur-
rent. If the voltage is not constant at 110 but fluctuates, as for in-
stance from 103 to 95 volts, it will be necessary to rewire the heaters
so as to produce a current which will develop 600 watts per heater.
For illustration, the amperage necessary on a 95-volt circuit is 6.2 am-
peres and on a 103-volt circuit it is 5.8 amperes. If the voltage is
fluctuating badly, the average voltage will have to be ascertained
for the circuit on which the moisture machine is to be operated and
the heaters overwired; that is, so wired as to give more wattage than
is necessary, and the excess cut out by means of a point rheostat.
This, of course, necessitates the constant use of a voltmeter, ammeter,
and rheostat on the line.

HOW TO TELL IF THE ELECTRIC HEATER IS WORKING AT STANDARD CONDITION

By the combined use of a voltmeter and an ammeter it can readily
be determined whether the proper wattage is being developed for
the type of heaters used. As these instruments are not always avail-
able, however, the oil test (p. 22) is the best test to apply. The

THE BROWN-DUVEL MOISTURE TESTER 39

standard heating time is the same regardless of the sources of heat
energy, whether gas, electricity, alcohol, or gasoline.

If, on making a check test with the oil-test method, all heaters do
not rise uniformly within one-half minute of 20 minutes, they are
not standard and will give either high or low moisture-test results,
and the. services of a competent electrician are needed.

To make an electric-heated machine check with the gas-heated ma-
chine, the electric heaters of the box type in which the upper side is
open, if they are giving a standard output of heat, should be lowered
after making the moisture test. If they are not giving a standard
amount of heat, direct comparison with a standard gas tester is the
only way in which one can determine in what position the heater
should be left. When that type of box heater is used in which the
top and bottom casing of the box is absent, the heater should be left
in position to assure check results with the gas-heated testers. If this
type of heater is lowered from its operating position when the cur-
rent is turned off, heat is lost too rapidly from the compartment after
the completion of the test.

The data obtained from these experiments show that the maxi-
mum difference between the results obtained on an electrically-heated
machine, properly adjusted and carefully operated, and on a stand-
ard gas-heated tester should be very slight, with a voltage drop of
not over 5 volts. Results, therefore, which are in excess of this are
due to faulty operation of the machine, such as not waiting for the
temperature of the oil in the flask to drop to 160° C. before making
a reading; use of faulty equipment, including incorrectly gradu-
ated thermometers or measuring cylinders and spongy rubber stop-
pers; inaccurate weighings, or lack of priming tests.

TESTS WITH ALCOHOL AND GASOLINE BURNERS

Moisture tests can be made with a Brown-Duvel moisture tester
heated with alcohol or gasoline burners, but such tests can not be
made with the satisfaction attending the use of gas or electric-heating
units. In making tests with alcohol or gasoline burners, the flame
is very irregular and needs frequent adjustment of the burner valves
during the progress of the test.

When the burners are used, they may be standardized by placing
them at a distance of 234 inches from the wire gauze and the flame
adjusted to a point where it will develop heat enough to raise the
temperature of the oil and grain in the flask to 120° C. in 10 minutes
and then to give a steady rise in temperature beyond this point of
from 6° to 8° per minute, until the cut-off temperature is reached.

HOW TO CHOOSE EXTINGUISHING TEMPERATURES

The Brown-Duvel method has not yet been developed for testing
a number of commodities for which the method is applicable. Ques-
tions are frequently asked regarding the procedure necessary to de-
termine the proper extinguishing temperatures for different com-
modities.

The extinguishing temperatures for testing the moisture content
of various commodities with the Brown-Duvel moisture tester are

40 BULLETIN 1375, U. S. DEPARTMENT OF AGRICULTURE

secured by checking the results obtained with the moisture tester
against results obtained by making moisture determinations by dry-
ing to constant weight in the common type of double-walled oven
filled with water maintained at the boiling point, except that the
treatment for flaxseed is determined by checking with samples dried
to constant weight in a 28-inch vacuum at a temperature of 99 to
100°C.

To determine the proper extinguishing point for the Brown-Duvel
method, the following procedure. takes place : :

In the case of corn, the grain is heated in a water-jacketed oven
at the temperature of boiling water, 99° to 100° C., until no further
loss of weight occurs in 24- hour intervals. This usually takes about
96 hours. This loss in weight is considered moisture and is known
as the standard moisture test. One hundred gram portions of the
same grain are then treated in the Brown- Duvel way, the only vari-
able being the temperature at which the flame is extinguished. The
cut-off temperature is gradually changed through 5° or 10° units
until the moisture test result by the Brown-Duvel method is the
same as the moisture test obtainea with the water-oven method.
This temperature is then tried out in check tests on a large number
of samples of corn having different moisture contents and further
comparisons made with results as determined by the water-oven
method.

TABLE 13.—Comparison of the moisture machine method tests on a sample of
corn at extinguishing temperatures of 160°, 170°, 180°, 190°, and 200° C. with
oven test of the same sample

Percentage of moisture

Test No. Machine tests
Oven
test |
160°) C. 170° CG. 180° C. 190° C. 200° C.
fase, BEA. BE ee Ee ee 14. 24 12.4 12.8 13. 6 14.0 14.8
a ae a ee a ee 14. 21 1 AR ls yal 13: 5 14.0 14.8
BP Acie ee st ods oe ee eee ee ee els oe ES 14. 27 W253 12.9 13. 6 14.1 14.6
Aid Pee SSE BA Sn at ok De pg he es ee ee 12. 4 Beal 13. 4 13.9 14.5
ee SE Ce a es Ee Ae eee 12. 5 1351 13d5 14. 2 14.6
Gaia eee Se ee ee ee ee 1253 13.0 13. 4 14. 2 14.6
SA VeETabe! : seis. soir sees * soy [Ase eset 14. 24 12. 4 13. 0 13.°5 14.1 14.7
Variation from oven els, eee =1.8 —1.2 —0.7 —0.1 +0. 5

}

Table 13 illustrates how the preliminary extinguishing tempera-
ture was chosen for corn.

The cut-off temperatures are varied and comparisons continue to
be made with results obtained with the water-oven method until the
correct extinguishing temperature is found and this temperature is
finally chosen as the. proper cut-off temperature. The extinguishing
temperatures that have been established to date for various com-
modities are listed in Table 14.

To determine the proper cut-off temperature for any commodity
for which the test is applicable but for which the cut-off temperature
has not yet been determined, proceed in a manner similar to that
described for determining the proper cut-off temperature for corn.

THE BROWN-DUVEL MOISTURE TESTER 4]

DRAWING AND HANDLING SAMPLES

The sampling of grain is not a part of the moisture test itself, but
an accurate sample “of the lot of grain to be tested is essential for
correct moisture tests. To obtain a representative sample is some-
times difficult, but the necessity of sparing no trouble in this respect
is imperative for correct results.

Grain in cars, elevator bins, cargoes, and other similar places is
seldom uniform in moisture content throughout the quantity, and
among other variations often shows less moisture at the surface
than in the body of the grain. In the case of corn in cars, it is not
uncommon to find differences of several per cent between the mois-
ture content of the layer of corn immediately at the surface and that
of the body of the grain directly beneath. In an experimental ship-
ment of three lots of corn contained in a single bulkheaded railroad
car in which special care was taken to see that each lot of grain was
loaded uniformly throughout, moisture tests at destination, 15 days
after shipment, ‘showed 14.1, 14.3, and 14.1 per cent, respectively,
for the corn from the surface of each of the three lots, as compared
with 17.3, 16.9, and 19.2 per cent, respectively, for the samples taken
with a grain trier from the body of the grain 2 to 314 feet below
the surface, the latter percentages being practically the same as at
the time of shipment.

TABLE 14.—Summary of specifications for testing grain and other substances

Oil in Weight Extin-
Substance flask of grain | guish the
in flask |flame at—

Grains:

Hard Red Winter, Hard Red Spring, Durum, and Red Che: Grams at CL
nT eSi cee PURE 2 0 a Se Oe ee SS ee os 150 100 180

Solved Winter and Whites +=. 5 ee eerie SS Se 150 100 190
SUG Gt brett) ce ee ee ee eS er ee. ee a 150 100 190
Oats cea Oats and Mixed: Heed Oats: >_222_) -=-3. 5.28 = se 150 100 195
inten seeks ERE ee eee eee oe oe ee ee oe Oe eee 150 100 185
GTAP ENSONS HGS hg BPS Se hs Be tees ee Ee PS 150 100 195
ES rape nes eed oD es DE 2 2 ne Se et See 28 150 100 190
IB Dekwiedbe nose Le nS oh ee et ee eet ee 150 100 195
TR aS Cees eg oe hk ee el A Se 150 100 175
IB rratad Tp t eee se oe Peet Pee 1 eee eS ee eee 150 100 190
FEE EES aS) a ee ek ee eee a 3 150 100 200
SEGUE MiGs gett SS Ba Oe ae ee eee ee eee 3 150 100 200
EE ELE go eat es Sea Be eS ee ce ee ee 3 150 100 200
AS TENGNS aT GGro nee ee ea ae ae ee eye 4150 100 200
Eas ee ee ee oe ee 3150 100 200
LRG) ENG) TRUE) te TE AEN EP ae al eee RS Det ee ee ee, Sa ee ee ee 150 100 200

Other substances

Commies: Seek see A, OR ee ee oh ee SS eee ee ee ee 250 150 190
WOpRONSECE CEs ee oie SES se Fe eee 150 3 50 190
Shelled peanuts 150 100 175
sear ERy POOF ok ge SS a en ee ee ee ee 200 | 100 168
ens EgeTOM COT ANS S22 - Sa: ee ee 22) See ek 200 | 1 50 190
Cornpmigalits 120_ ete sweetie eet ete eos ATCA a A es 5 150 | 1 50 6 175

1 Use special graduate which is one-half the volume of the regular graduate. However, the regular
graduate may be used by doubling the moisture-test reading.

2 In making tests of these classes of rice the distillate is sometimes cloudy. This in no way affects the
accuracy of the test and may be disregarded.

3 Use glass-wool pad 3 inches in diameter and 4 inch thick in bottom of flask.

4 Use glass-wool pad 4 inches in diameter and 14 inch thick in bottom of flask.

5 Use double-wall copper flasks with 150 cubic ‘centimeters of oil in the inner flask and 150 cubie centi-
meters between the walls.

6 Oil and meal in inner flask should reach a temperature of 175° C. in about 26 minutes.

492 BULLETIN 1375, U. S. DEPARTMENT OF AGRICULTURE

For correct results, it is essential that the same care which is used
in obtaining the sample be observed in handling the sample after it
is drawn from the grain or after 1t is weighed for the test. If ex-
posed to the air, samples will lose moisture in accordance with the
percentage of water in the sample and the condition of the atmos-
phere. An experimental test with 100 grams of wheat containing
14.5 per cent of moisture showed a loss of 0.20, 0.35, 1.10, and 1.90
per cent when ex-
posed to the air of
the laboratory for
periods of one-half,
1, 4, and 26 hours, re-
spectively. A second
lot of wheat, contain-
ing 16.1 per cent of
moisture, lost mois-
ture to the extent of
0.50, 0.75, 1.75, and
3.20 per cent, respec-
tively, during the
same periods under
identical treatment.
Figure 26 shows the
type of metal can
and moisture-proof
sample bag used by
the offices of Federal
Grain Supervision
for the purpose of
obtaining samples for
moisture tests and
other grading factors.

The sampling de-
vices and the methods
used for obtaining
correct samples of
grain are fully de-
scribed in United

Fic. Aue ease a uae! can and moisture-proot sample States Department
bag used by ces of Federal Grain Supervision for | = : r
handling samples of grain to be tested for moisture of Agriculture Hand
content. ane our ble bag is not Gest Me oe Mali book, Ue Noe G. S: i
purpose unless the moisture test can be made within \T
an hour or two from time of sampling Form No. 90.

SPECIAL POINTS FOR CONSIDERATION

(1) The moisture tester should be installed in a place where it
will not be exposed to strong air current.

(2) The standard tester is equipped for heating with illuminating
gas.

(3) The wire gauze with asbestos center should be kept in good
condition, and so adjusted that the flame plays directly in the center
of the asbestos.

THE BROWN-DUVEL MOISTURE TESTER 43

(4) The bottom of the flask should be not less than three-eighths
inch above the wire gauze.

(5) The column of mercury in the thermometer should be con-
tinuous; if broken, it should be shaken down.

(6) The sample should be thoroughly mixed before weighing for
tests; and unless the test is to be made immediately upon its ar-
rival in the office, it should be placed in an air-tight container.

(7) Tests should be made in duplicate, and if duplicates vary
over 0.3 per cent, another test should be made.

(8) The thermometers should be so adjusted that four-fifths of
the mercury bulb is submerged in the grain and oil after the grain
has been placed in the flask. (See to the adjustment each time. Do
not guess.)

(9) Correctly graduated thermometers and graduates should be
used.

(10) Mushy rubber stoppers must not be used as they absorb some
of the moisture that should pass into the graduates.

(11) Each graduate should be cleaned and dried before using
for a test. (Do not let them show any moisture in the bottom or
along the sides.)

(12) Oil should not be used directly from the previous test. Used
flasks should be emptied into a large storage can and never directly
into the oil-measuring device.

(13) A good circulation of cold water should be maintained
through the condenser tank.

(14) The heating apparatus should be so adjusted that the re-
quired temperature is reached in 20 minutes. A longer time will
give results too low and a shorter time, too high.

(15) If the moisture content of the sample is high so that there
is a tendency to boil over, the flame should be lowered until a con-
siderable portion of the water is distilled over.

(16) The heat should be cut off at the exact temperature prescribed
for each grain.

(17) After the flame is extinguished, a shght gradual rise in the
temperature is to be expected. A sudden increase or sudden decrease
in temperature of several degrees indicates that the flame was too
intense during the latter part of the heating, and the test should be
repeated.

(18) Covers and thermometers should not be removed until the
temperature recedes to 160° C.

(19) After the temperature has fallen to 160° C. or lower, the
thermometer is disconnected and then the delivery tube.

(20) The percentage of moisture in the graduated cylinder should
be read after all the drops clinging to the sides of the graduates
have been shaken down. The reading is taken beneath the layer of
oil on top of the water.

(21) Results of tests need not be expressed more closely than 0.1
per cent.

(22) If the water which distills over is discolored, the substance
has evidently been burned and the test should be repeated. (Note
exception to this in the case of rice.)

(23) When machine is not in use, thermometers should be kept
connected in the flasks and the flasks connected with the distilling
tubes in the same manner as for making a test.

44 BULLETIN 1375, U. S. DEPARTMENT OF AGRICULTURE

(24) Before making a test in a new flask, or before using a ma-
chine that has not been in use for a 24-hour period, a test should be
made on a preliminary sample so that all the flasks will be uniform
in condition.

(25) Scales should be placed on a firm support and care should
be taken that they are in balance before making a weighing.

(26) The specific directions given above for making tests do not
apply to modified forms of testers.

ADDITIONAL COPIES
OF THIS PUBLICATION MAY BE PROCURED FROM
THE SUPERINTENDENT OF DOCUMENTS
U.S. GOVERNMENT PRINTING OFFICE
WASHINGTON, D. C.
AT
10 CENTS PER COPY

V

Pa