TEST METHODS 14

REVISED FLAMMABILITY TEST METHODS

FOR AEROSOL PRODUCTS
Recommendations of Subcommittee on
Consumer Hazards and Safety Aerosol Division of CSMA
1. Abandon the Open Drum Test as a meaningful guide
in establishing flammability hazards.
2. The following changes and modifications in the
Flame Projection Test, Tagliabue Open-Cup Flash Point
Test, and Closed Drum Test:
Flame Projection Test
1. Equipment required. The test equipment consists of a
base 8 inches wide, 2 feet long, marked in 6 inch intervals.
A rule 2 feet long and marked in inches is supported
horizontally on the side of the base and about 6 inches
above it. A paraffin candle approximately 1 inch in
diameter and of such height that the top third of the flame
is at the height of the horizontal rule, is placed at the zero
point in the base.
2. Procedure. The test is conducted in a draft-free area
that can be ventilated and cleared after each test. Condition
the dispenser to 70° ± 1° F. Shake the dispenser before
test. Hold the dispenser upright unless label states
otherwise. Place the dispenser at a distance of 6 inches from
the flame source. Spray for 4 seconds (one observer noting
the extension of the flame and the other operating the
dispenser) through the top third of the flame and
essentially parallel to the rule. The height of the flame
should be approximately 2 inches. The normal bending of
the flame is part of the recorded distance. Take 3 readings
for each test and average. As a precaution do not spray
large quantities in a small, confined space. Free space of
previously discharged material.

Tagliabue Open-Cup Flash Point Test

1. Equipment. ASTM Designation: D 1310-55T. Issued
1954; Revised 1955. "Flash Point of Volatile Flammable
Materials by Tag Open-Cup Apparatus". The apparatus
proposed for use is the new Tag Fischer open-cup flash
point apparatus, with the addition of some means,
preferably an open type of vessel to contain dry ice, to chill
the aerosol unit.
2. Procedure. The aerosol unit, filled as for use, is
chilled to a temperature of about 25° F. below zero and
also the flash cup and the bath solution (brine or glycol).
The chilled formulation is transferred to the test apparatus
and the cup filled to the conventional level using the
mechanical leveling device furnished with the apparatus.
The test liquid is allowed or caused to increase in
temperature at a rate of about 2° F. per minute and the
test flame taper passed across the cup at intervals of 2° F.
until the sample reaches +20° F. or until the test sample
has evaporated completely.
This procedure is not applicable to products in which
the presence of a solid portion prevents the transfer of a
uniform sample to the cup at -25° F.
Closed Drum Test
1. Equipment required. The apparatus consists of a
55-gallon open head drum which has been modified as
follows:
(1) a closure is fitted over the open head (see note No.
i);
(2) a circular opening 1 inch in diameter is bored
through the base, about 2 inches from the edge, in such a
position that when the drum is on its side the hole will be
at the top.
(3) a metal base 9 inches long, 2 inches wide and at least
1/16 inch thick is used as a candle support (see drawing
above);
(4) a paraffin candle approximately 1 inch in diameter
and at least 3 inches in height;
(5) optionally a 6-inch square opening is cut through the
center of the base of the drum and securely covered with a
piece of safety glass.
2. Procedure. Lay the modified drum on its side under
conditions where the temperature is between 60° and 80°
F., but preferentially as close to 70° F. as possible. Stand a
paraffin candle in the drum on a special metal base half-way
from each end of the drum. (See drawing above) Condition
the dispenser to 70° + 1 0 F .
Light the candle on the base and secure closure. (When
the film closure is used the candle should be ignited by
means of a taper through the 1 inch circular opening).
Shake the dispenser and hold upright or, if necessary, in
such a position that the liquid contents can be sprayed
directly into the drum. As quickly as possible place the
dispenser at the one-inch opening and spray into the drum,
directing the spray toward the center of the opposite end
until an explosion takes place, or for a period up to 60
seconds, whichever occurs first. After each test, open the
drum to clear the atmosphere. Clean the drum of any
residues which might affect future tests. Repeat the test
twice as before using the same dispenser if possible. If size
limitations make it necessary to use more than one
dispenser, then not more than one dispenser shall be used in
the performance of any one test.
The results of three tests are averaged.
Note L A hinged lid or a l/2 mil thickness polyamide
film can be used as the closure.
Polyamide Film: The film is stretched over the open end
of the drum and held in place by a strong rubber band, that
CANDLE BASE HINGED LID
•long bottom of drum
will stretch one inch when a one pound weight is hung from
its lowest point, when around the drum. A one-inch slit is
cut vertically in the film beginning at a point 2 inches from
the top of the drum.
The polyamide film must be drawn taut over the
opening
TENTATIVE METHOD FOR DETERMINING
THE RELATIVE FLAMMABILITY OF
AEROSOL FOAM PRODUCTS
Introduction
Flammability is one of the more important criteria of an
aerosol product. The Flame Propagation and Closed Drum
Test procedures, which work so well with most aerosols,
cannot be made to reliably assess the flammability of those
products which are dispensed as stable foams. The present
method is designed to fill this void.
Foam products may be sprayed or extruded. They may
be of the stable type, semi-stable variety or quick-breaking
compositions. Flammability may be brought about through
the use of sufficient quantities of flammable or combustible
liquids, flammable propellents, or both at once. It is
because of these variations that no single method has
appeared to be suitable for all foam products. Rather, a
dualistic method, based upon two complementary
procedures, is required.
The first of these procedures, known as the "Tower
Test," is used to assess the rate at which a fixed volume of
foam, with a specified surface area, can generate gas; and
whether this gas is capable of forming a flammable mixture
with air. In the second procedure, the "Trough Test," a
flame is touched to the foam surface to determine if the
composition is capable of flashing or sustaining a fire.
Apparatus
1. Apparatus for the Tower Method: The dimensions
and other specifications are provided in FIGURE ONE. The
equipment consists of a base section, designed to hold a
measured amount of product, over which is fitted a
cylindrical chimney pierced with a vertical row of small
holes. The suggested material is aluminum, although brass
and stainless steel have also been used.
2. Apparatus for the Trough Method: The dimensions
and other specifications are provided in FIGURE TWO. The
equipment consists of a simple flat trough, preferably
formed from stainless steel sheet, of about No 16 Gage
(0.062") thickness.

Procedure for the Tower Method

1. The aerosol can is brought to equilibrium
temperature by holding for at least 45 minutes in a water
bath at 70° F. Tests are run for each of two "use
conditions." The first test is conducted on a new can, and
the second test on the same can after it has been discharged
to 20% of the stated net weight of the contents of the can
plus the weight of one test cupful of product. If there is
insufficient material in one can for both tests, two different
cans for each test may be used. In order to more nearly
reproduce "use conditions," tests are not conducted on the
partially used cans until one hour after the cans have been
discharged. Cans should be kept at room temperature (70°

2. The sample can is shaken briefly and then discharged

in the normal use position into the test cup. Care should be
exercised when charging the test cup so that the air pockets
are not occluded by the foam, thereby reducing the
apparent volume of test material used. An adapter can be
used to facilitate charging the test cup with foam. Figure 1. Apparatus for Tower Method

3. When the foam sample is in the test cup, allow five to

ten seconds, counted from the time the cup has been filled,
for the foam to expand fully before proceeding to the next
step.
4. Level the foam by drawing a spatula across the top of
the test cup.
5. The tower unit, with a single length of strippable tape
adjusted to seal its apertures, is set in place over the foam
filled cup.
6. The apparatus is left undisturbed in a draft-free area
for a five-minute holding time before proceeding to the test
measurement.
7. After the five-minute holding time, and again without
moving the apparatus, the protective tape is stripped from
the touchholes at a uniform rate so that each hole is
exposed at two-second intervals.
8. As each touchhole is exposed, starting from the top
^15 hole, the flame of a small lighted gas jet is brought to
the aperture. The flame should just touch the metal at the
lower edge of the hole. Do not insert the flame into the
hole.
Procedure for the Trough Method
1. The test samples used for this procedure are the same
samples that are used for the Tower Test. The flame
propagation tests are run in between the two sets of tower
tests; i.e. after the intial tests, but before the cans have been
discharged to 20% of the stated net weight of the contents
of the can plus the weight of one cupful of product.
2. The aerosol can is brought to equilibrium in a water
bath at 70° F.
3. The sample foam is shaken briefly and then
discharged into the test trough. Care should be taken when
charging the test trough, so that air pockets are not
occluded by the foam, thereby reducing the apparent
volume of the test material used.
4. When the trough has been filled, allow the foam to
expand for five seconds before proceeding to the next step.
5. Quickly draw the edge of a large spatula across the
top of the trough to level the foam.

9. A positive result is indicated when a flash occurs in

the tube, and the hole number from where the flash occurs 6. Three tests are conducted for each foam sample. The
is recorded. Reported results should be the average of three first test, immediately after the foam has been leveled in
tests for each sample. the trough; the second test, after a two-minute holding

NOTt:
W£i£> ALL WM*
$e£AX AU fHjer £OM££$
Figure 2. Apparatus for Trough Method
time; and the third, after a five-minute holding time. The
trough is left undisturbed and in a draft-free area for the
designated holding times between tests.
7. At the end of the holding time, a small lighted gas
flame is brought to the surface of the foam in the area to
one-half inch from the end of the trough.
8. A positive test is indicated when the flame is
sustained or propagated by the propellent gas or foam
material after withdrawal of the flame.
Presentation of test results
A positive result for the Tower Test is reported as the
average of three tests for each product. For example:
Test No. 1 Positive; at hole number 7.
Test No. 2 Positive; at hole number 9.
Test No. 3 Positive; at hole number 6
Average: Positive. Hole number 7.
A positive result for the Trough Test should be reported
in terms of these observations:
P Indicates propagation of flame.
12 Depicts distance of travel in inches.
SM Indicates flame is sustained on surface.
RT Shows flame is not sustained, but travels to
end of trough and returns toward the
starting point. Numbers may be added to
indicate number of inches on return.
Typical positive results might then be recorded as:
P-3 Showing that the flame traveled three inches
toward the far end of the trough and then
went out.
P-14/RT-7 Indicating that the flame traveled to the far
end of the trough and then returned about
halfway back before extinguishing itself.
P-14/SM Flame travels across trough and continues to
burn.
These typical results of each test can be compared as
follows:
Tower Test: Touchhole No. 9 indicatesv a product more
flammable than Touchhole No. 6
Trough Test: P-14/SM is worse than P-14/RT-7 which, in
turn, is more flammable than P-3
Discussion
Because of the difficulty in relating the results of these
tests to the degree of hazard involved in the transportation,
warehousing or end uses of these foam type products, the
two procedures are not to be used in order to determine if a
product is to be called "Flammable" as distinguished from
"Non-flammable." They are presented simply as the means
by which the relative flammability of foam products may
be assessed.
The procedures have been most thoroughly evaluated on
simple shaving cream aerosols, and are certified only for
application to this particular product type. In addition, the
methods have been applied to a variety of stable foam
items, such as upholstery cleaner, charcoal lighter, metal
polish and hair shampoo, with good results. The procedures
have not been evaluated with thermal foams, foaming gels,
quick-breaking foams, anhydrous foams and reticulated
gels; so that any conclusions made concerning the relative
flammability of these products must be made according to
the investigator's own recognizance.
FLASH POINT DETERMINATION
Excerpts from Regulations Issued For
The Federal Hazardous Substances Act
(1) Extremely flammable and flammable contents of
self-pressurized containers — (1) Extremely flammable
contents. Contents of self-pressurized containers are
"extremely flammable" if when tested by the method
prescribed in f 191.15, flashback (a flame extending back to
the dispenser) is obtained at any degree of valve opening
and the flashpoint, when tested by the method described in
1191.16, is less than 2O0F.
(2) Flammable contents. Contents of self-pressurized
containers are "flammable" if when tested by the method
described in fl91.15 a flame projection exceeding 18
inches is obtained at full valve opening or a flashback (a
flame extending back to the dispenser) is obtained at any
degree of valve opening.
(m) Substances that generate pressure. A substance is
hazardous because it "generates pressure through
decomposition, heat, or other means" if:
(1) It explodes when subjected to an electrical spark, or
to percussion, or to the flame of a burning paraffin candle
for 5 seconds or less; or
 (2) It expels the closure of its container, or bursts its
container, when held at or below 130° F. for 2 days or less;
or
(3) It erupts from its opened container at a temperature
of 13O0F. or less, after having been held in the closed
container at 13O0F. for 2 days.
(4) If it comprises the contents of a self-pressurized
container
fl91.13 Tentative method of test for flashpoint of
volatile flammable materials by Tagliabue
open-cup apparatus1 > 2
Scope
1. (a) This method describes a test procedure for the
determination of open-cup flashpoints of volatile
flammable materials having flashpoints below 175° F.
(b) This method, when applied to paints and resin
solutions which tend to skin over or which are very viscous,
gives less reproducible results than when applied to
solvents.

Outline of Method
2. The sample is placed in the cup of a Tag Open Tester,
and heated at a slow but constant rate. A small test flame is
passed at a uniform rate across the cup at specified
intervals. The flashpoint is taken as the lowest temperature
at which application of the test flame causes the vapor at Figure 1
the surface of the liquid to flash, that is, ignite but not
continue to burn.
(d) Glass test cup. Glass test cup (Fig. 2), of molded
Apparatus clear glass, annealed, heat-resistant, and free from surface
defects.
3. The Tag open-cup tester is illustrated in Fig. 1. It
consists of the following parts, which must conform to the (e) Leveling device. Leveling device or guide, for proper
dimensions shown, and have the additional characteristics adjustment of the liquid level in the cup (Fig. 3). This shall
as noted: be made of No. 18-gage polished aluminum, with a
(a) Copper bath, preferably equipped with a constant projection for adjusting the liquid level when the sample is
level overflow so placed as to maintain the bath liquid level
1/8 inch below the rim of the glass cup.
(b) Thermometer holder. Support firmly with ringstand
and clamp.
(c) Thermometer. For flashpoints above 40° F., use the
ASTM Tag Closed Tester Thermometer, range of +20 to
+230° F., in 1° F. divisions, and conforming to
thermometer 9F. of ASTM Standard E 1. For flashpoints
below 20° F., use ASTM Thermometer 33F. The original
Tag Open-Cut (Paper Scale) Thermometer will be a
permissible alternate unit January 1, 1962. It is calibrated
to -20° F.

The Food and Drug Administration has obtained permission

from the American Society for Testing Materials, Philadelphia, Pa.,
to reprint this method in these regulations. The text has been
slightly modified, for practical reasons.
2
ASTM Designation: D 1310-59T, issued 1954; revised
1955,1956,1959.This tentative method has been approved by the
sponsoring committee and accepted by the American Society for
Testing Materials in accordance with established procedures, for use
pending adoption as standard. Suggestions for revisions should be
addressed to the Society at 1916 Race St., Philadelphia, Pa. Figure 2

FIGURE 3 — Leveling device for adjusting liquid level in test cup.
added to exactly 1/8 inch below the level of the edge or rim
of the cup.
(O "Micro," or small gas burner of suitable dimensions
for heating the bath. A screw clamp may be used to help
regulate the gas. A small electric heater may be used.
(g) Ignition taper, which is a small straight, blow-pipe
type gas burner. The test flame torch prescribed in the
method of test for flash and fire points by Cleveland Open
Cup (ASTM designation: D 92) is satisfactory.
(h) Alternative methods for maintaining the ignition
taper in a fixed horizontal plane above the liquid may be
used, as follows:
(1) Guide wire, 3/32-inch in diameter and 3 1/2 inches
in length, with a right-angle bend 1/2-inch from each end.
This wire is placed snugly in holes drilled in the rim of the
bath, so that the guide wire is 5/8-inch from the center of
the cup and resting on the rim of the cup.
(2) Swivel-type taper holder, such as is used in ASTM
METHOD D92. The height and position of the taper are
fixed by adjusting the holder on a suitable ringstand
support adjacent to the flash cup.
(i) Draft shield, consisting of two rectangular sheets of
noncombustible material, 24 inches x ( 2 8 inches, are
fastened together along the 28-inch side, preferably by
hinges. A triangular sheet, 24 inches x 24 inches x 34 inches
is fastened by hinges to one of the lateral sheets (to form a
top when shield is open). The interior of the draft shield
shall be painted a flat black.
Procedure
4. (a) Place the tester on a solid table free of vibration,
in a location free of perceptible draft, and in a dim light.
(b) Run water, brine, or water-glycol solution into the
bath to a predetermined level, which will fill the bath to
1/8-inch below the top when the cup is in place. An
overflow is permissible for water-level control.
Standardization
8. (a) Make determinations in triplicate on the flashpoint
of standard paraxylene and of standard isopropyl alcohol
which meet the following specifications:
(i) Specifications for p-xylene, flashpoint check grade.
p-Xylene shall conform to the following requirements:
Specify gravity: 15.56° C./15.560 C., 0.860 minimum,
0.866 maximum.
Boiling range: 2° C. maximum from start to dry point
when tested in accordance with the method of test for
distillation of industrial aromatic hydrocarbons (ASTM
designation: D 850), or the method of test for distillation
range of lacquer solvents and diluents (ASTM designation:
D 1078). The range shall include the boiling point of pure
p-xylene, which is 138.35° C. (281.03° F.).
Purity: 95 percent minimum, calculated in accordance
with the method of test for determination of purity from
freezing points of high-purity compounds (ASTM
designation: D 1016), from the experimentally determined
freezing point, measured by the method of test for
measurement of freezing points of high-purity compounds
for evaluation of purity (ASTM designation: D 1015).
(ii) Specifications for isopropanol, flashpoint check
grade. Isopropanol shall conform to the following
requirements:
Specific gravity: 0.8175 to 0.8185 at 20° C./200 C. as
determined by means of a calibrated pycnometer.
Distillation range: Shall entirely distill within a 1.0° C.
range which shall include the temperature 80.4° C. as
determined by ASTM method D 1078.
Average these values for each compound. If the
difference between the values for these two compounds is
less than 15° F. (8.5° C.) or more than 27° F. (16° C.),
repeat the determination or obtain fresh standards.
(b) Calculate a correction factor as follows:
X = 92-A
Y = 7 1 -B
y +Y
Correction= — —
2
Where:
A = Observed flash of p-xylene, and
B = Observed flash of isopropyl alcohol.
Apply this correction of all determinations. Half units in
correction shall be discarded.
Precision
9. (a) For hydrocarbon solvents having flashpoints
between 60° F. and 110° F., repeatability is ±2° F. and the
reproducibility is 5° F.
(b) If results from two tests differ by more than 10° F.,
they shall be considered uncertain and should be checked.
The calibration procedure provided in this method will
cancel out the effect of barometric pressure if calibration
and tests are run at the same pressure. Data supporting the
precision are given in Appendix III of the 1956 Report of
Committee D-I on Paint, Varnish, Lacquers and Related
Products, Proceedings, Am. Soc. Testing Mats., Vol. 56
(1956).
1191.14 Method for determining extremely
flammable and flammable solids.
(a) Preparation of sample —(1) Granules, powders, and
pastes. Pack the sample into a flat, rectangular metal boat
with inner dimensions 6 inches long x 1 inch wide x
one-fourth inch deep.
 (2) Rigid and pliable solids. Measure the dimensions of
the sample and support it by means of metal ringstands,
clamps, rings, or other suitable devices as needed, so that
the major axis is oriented horizontally and the maximum
surface is freely exposed to the atmosphere.
(b) Procedure. Place the prepared sample in a draft-free
area that can be ventilated and cleared after each test. The
temperature of the sample at the time of testing shall be
between 68° F. and 86° F. Hold a burning paraffin candle
whose diameter is at least 1 inch, so that the flame is in
contact with the surface of the sample at the end of the
major axis for 5 seconds or until the sample ignites,
whichever is less. Remove the candle. By means of a
stopwatch, determine the time of combustion with
self-sustained flame. Do not exceed 60 seconds. Extinguish
flame with a CO2 or similar non-destructive type
extinguisher. Measure the dimensions of the burnt area and
calculate the rate of burning along the major axis of the
sample.
(c) Firmly support the thermometer vertically halfway
between the center and edge of the cup on a diameter at
right angles to the guide wire, or on a diameter passing
through the center of the cup and the pivot of the taper.
Place so that the bottom of the bulb is 1/4-inch from the
inner bottom surface of the cup. If the old Tagliabue
thermometer is used, immerse to well cover the mercury
bulb, but not the wide body of the thermometer.
(d) Fill the glass cup with the sample liquid to a depth
just 1/8-inch below the edge, as determined by the leveling
device.
(e) Place the guide wire or swivel device in position, and
set the draft shield around the tester so that the sides form
right angles with each other and the tester is well toward
the back of the shield.
(f) If a guide wire is used, the taper, when passed, should
rest lightly on the wire, with the end of the jet burner just
clear of the edge of the guide wire. If the swivel-type holder
is used, the horizontal and vertical positions of the jet are
so adjusted that the jet passes on the circumference of a
circle, having a radius of at least 6 inches, across the center
of the cup, at right angles to the diameter passing through
the thermometer, and in a plane 1/8-inch above the upper
edge of the cup. The taper should be kept in the "off
position, at one end or the other of the swing, except when
the flame is applied.
(g) Light the ignition flame and adjust it to form a flame
of spherical form matching in size the 5 32-inch sphere on
the apparatus.
(h) Adjust heater source under bath so that the
temperature of the sample increases at a rate of 2 ±0.5° F.
per minute. With viscous materials this rate of heating
cannot always be obtained.
Initial Test
5. Determine an approximate flashpoint by passing the
taper flame across the sample at intervals of 2° F. Each pass
must be in one direction only. The time required to pass
the ignition flame across the surface of the sample should
be 1 second. Remove bubbles from the surface of the
sample liquid before starting a determination. Meticulous
attention to all details relating to the taper, size of taper
flame, and rate of passing the taper is necessary for good
results. When determining the flashpoint of viscous liquids
and those liquids that tend to form a film of polymer, etc.,
on the surface, the surface film should be disturbed
mechanically each time before the taper flame is passed.
Recorded Tests
6. Repeat the procedure by cooling a fresh portion of
the sample, the glass cup, the bath solution, and the
thermometer at least 20° F. below the approximate
flashpoint. Resume heating, and pass the taper flame across
the sample at two intervals of 5° F. and then at intervals of
2° F. until the flashpoint occurs.
Reporting Data
7. The average of not less than three recorded tests,
other than the initial test, shall be used in determining the
flashpoint and flammability of the substance.
fl91.15 Method for determining extremely
flammable and flammable contents of
self-pressurized containers
(a) Equipment required. The test equipment consists of
a base 8 inches wide, 2 feet long, marked in 6-inch intervals.
A rule 2 feet long and marked in inches is supported
horizontally on the side of the base and about 6 inches
above it. A paraffin candle 1 inch or more in diameter, and
of such height that the top third of the flame is at the
height of the horizontal rule, is placed at the zero point in
the base.
(b) Procedure. The test is conducted in a draft-free area
that can be ventilated and cleared after each test. Place the
self-pressurized container at a distance of 6 inches from the
flame source. Spray for periods of 15 seconds to 20 seconds
(one observer noting the extension of the flame and the
other operating the container) through the top third of the
flame and at a right angle to the flame. The height of the
flame should be approximately 2 inches. Take three
readings for each test, and average. As a precaution do not
spray large quantities in a small, confined space'. Free space
of previously discharged material.
fl91.16 Method for determining flashpoint of
e x t r e m e l y f l a m m a b l e contents of
self-pressurized containers
The apparatus used is the Tagliabue Open-Cup
Flashpoint Apparatus as described in 1191.13. Some means
such as dry ice in an open container is used to chill the
pressurized container. The container, the flash cup, and the
bath solution of the apparatus (brine or glycol may be
used) are chilled to a temperature of about 25° F. below
zero. The chilled container is punctured to exhaust the
propellant. The chilled formulation is transferred to the test
apparatus and tested in accordance with the method
described in f 191.13.
CSMA NOTE: This test is taken from ASTM Designation
D-1310-SST Issued 1954; revised 1955. "Flash Point
Volatile Flammable Materials by Tag Open-cup Apparatus".
The test is considered to terminate when the sample flashes,
reaches +20° F. or evaporates completely.

TENTATIVE METHOD FOR MEASURING
"A-D" DIMENSION OF AEROSOL CAN VALVES
INTRODUCTION
A simple device is presented for the routine
measurement of the "A-D" valve dimension. This distance
is defined as the standard for stipulating dip tube depth. It
is the centerline dimension from the top plane of the
mounting cup curl to the far end of the diptube; (Figure 1).
For quality control purposes it is desirable to have a
method available for the rapid measurement of dip tube
depth. This is especially important where large numbers of
measurements must be taken. The present device eliminates
cumbersome hand manipulations, parallax errors and
similar difficulties. It provides a means for measuring dip
tube depths with excellent accuracy at speeds of from
about 10 to 14 per minute.
METHOD
The measuring guage is produced from an aluminum
cylinder machined to a length of 8.717" and an outside
diameter of 1.216". A complete description is given in the
accompanying drawing; (Figure II). This is for a left-handed
guage. A right-handed one may be produced by machining a
mirror-image of the gauge bar itself and locating the ruler
on the opposite quadrant flat.
For individual accurate "A-D" measurements the valve is
inserted so that the dip tube slides down the centerbore of
the gauge. With the mounting cup held snugly in place
against the gauge end, the exact dip tube depth is read off
the steel rule.
For repetitive or quality control measurements the valve
is inserted as before, using a gauge where the steel rule has
been temporarily marked in pencil to show the tolerance or
range of acceptance. Each valve may then be rapidly
checked to show compliance with the "A-D" specification.
The pencil marks may then be removed by hand rubbing.
Do not use ball point pens or other marking devices where
the lines will have to be removed with solvent or abrasive.
This can also remove the black lettering and line fillers of
the ruler itself, rendering it much harder to read. With only
a slight loss in accuracy the tolerance marks may be made
upon the opposite quadrant flat. For some operations
permanent marks may be scribed on this flat with a suitable
tool.
DISCUSSION
The true measure of dip tube depth is a function of the
depth of protrusion into the aerosol can. This will be
"A-D" VALVE DIMENSION
Figure 1. "A-D" Valve Dimension
affected by relative dip iuoe curvatures and by variations in
the ratio of can diameter to total height — factors which
cannot be anticipated by any simple gauging device. The
present measurement compensates for variations in cup
depth, cup contour, body length and upper reference
position of the dip tube. Operations such as crimping the
valve and pressurizing the aerosol container will shorten the
"A-D" dimension slightly, but these effects are fairly
reproduceable and compensations could be made if
necessary.
If the steel ruler is attached to the gauge bar with a high
degree of precision, the device may be expected to function
with an accuracy of about ±0.012". Where the ruler is
attached with a metal adhesive, the inset end of the gauge
bar should end at 0.033" by extension of the cut offend of
the ruler. If it is too long the end should be machined
down. Where the ruler is attached by means of several
screws, it is desirable to use roundhead machine screws, set
through holes in the ruler large enough to allow for precise
positioning of the scale before final tightening.
Only one assumption is made in the construction of the
gauge: that the vertical distance through the curl of the
mounting cup, including the thickness of the cup gasket,
totals 0.0333". Studies made with an Ames No. S-4469 dial
micrometer or comparable equipment will show that this
crown dimension usually holds true to within about
±0.004" on a 2-sigma or 95% confidence level. The
variation is due mainly to differences in cup gasket
thickness. It is not possible for the average operator to read
the. ruler this accurately; normal reproduceability being
about ±0.0065" with a perfectly sheared dip tube. The
variation may be neglected for all practical purposes.
Figure 2. Measuring Gauge
 The average dip tube is not cut off at precisely a right
angle, nor is the cut end smooth and free from
imperfections. Often the center-line distance from the short
sector to the long sector is as much as 0.006". The slight
residual curvature of the dip tube, as it reposes within the
gauge bar, may require the averaging of high and low
readings on the oblique surface. Polyethylenes stretch easily
and should never be straightened with any method which
involves tension. Polypropylenes and nylon dip tubes are
less easily stretched.
The distortional effects of valve crimping and can
pressurization have been mentioned. Hot tanking also
produces an irreversible distortion of about 0.0025". To
these vectors are added the swelling effect of the aerosol
formula and the consideration that the cut end of the dip
tube is never horizontal, but slanted towards a point on the
bottom chime of the container. From all this it is apparent
that dip tube measurements with accuracies better than
±0.016" serve no real purpose in quality control operations.
The usual specification for dip tube depth is supplied by
valve manufacturers at L±0.063".
A gauge bar has several applicational limitations. It
cannot be used to provide an accurate "A-D" measurement
for ferrule type valves*, co-dispensing valves, extra-large
dosage metering valves, or those with unusual mounting cup
designs, such as the turret-top styles. Double dip-tube valves
and those with dip tubes having an outside diameter greater
than about 0.242" cannot be handled. The "A-D"
dimension itself is often found to be inapplicable to these
unique valves.
CREDITS
The Modern Machine Shop, Inc., at 123 North Hazel
Street, Danville, Illinois 61832, has had experience in the
manufacture of a number of these gauge bars and may,
therefore, be considered as a reliable source.
We are indebted to the British Aerosol Manufacturers
Association (BAMA) and the Federation of European
Aerosol Associations (FEA) for providing the original idea
for the construction of this instrument.
ALTERNATE DEVICE FOR MEASURING THE
DIP TUBE DEPTH OF AEROSOL VALVES
This device consists basically of 3 machined parts made
of plexiglas. In addition, there is a scale glued to the "scale
block"; a foam bottom pad may be added for non-slip
characteristics.
The valve to be measured is inserted into the device,
placing the actuator/stem end to the higher end of the
device. The slide bar is then pushed firmly to the left to
"seat" the mounting cup against base end. The reading may
then be taken off the scale at the end of the dip tube, giving
the "A-D dimension" from top of mounting cup to dip
tube end.
Note that this measuring device can be used to measure
valves of various types, including ferrule-type valves. It
would not be applicable to co-dispensing or large-tank
metered valves; undue distortion would occur with double
dip tube valves or large O.D. dip tubes.
*Reference is made to an attached alternative device for measuring
dip tube depths. It may be used for "A-D" measurements of ferrule
type valves as well as for aerosol can valves.
PROCEDURE FOR EVALUATION
OF COMBING CHARACTERISTICS OF
HAIR SPRAY FORMULATIONS
(Editor's Method)
SCOPE
This test evaluates, subjectively, the adhesion, flaking
and general effect on combing characteristics of aerosol
formulations on hair.
Two undesirable effects seen in some formulations on
hair:
1. Flaking
This condition manifests itself in the appearance of
small polymer particles distributed throughout the
hair.
2. Snarling
This appears to be due to general tackiness or,
conversely, extreme brittleness. In the case of brittle
polymers, the resin ruptures during combing and
produces small barbs of polymer protruding from the
hair shafts. Both the tacky and brittle polymers resist
combing and increase the possibility of entanglement
of individual hair shafts.
PROCEDURE
1. Ten inch hair swatches are prepared according to the
description given in the "Curl Retention Procedure".
2. All testing; i.e., spraying, drying and combing is
performed at 72°F./50RH and run on three swatches
for each formulation.
3. The test formulation is applied to a rotating (20
RPM) hair swatch by a 5-second aerosol discharge at a
distance of 7 inches.
4. The swatch is allowed to dry, undisturbed, for 30
minutes.
5. Comb the length of the hair piece 5 times (at a rate of
3"/second) with the fine teeth of a comb (13-7/8"
teeth/inch): Record evaluations.
6. Repeat steps 3 to 5 until the formulations under test
show severe flaking or are difficult to comb.
7. Perform test on 3 swatches for each formulation
variable.
EVALUATION
A. Combing Rating
None to slight resistance to combing —1
Moderate resistance distributed at different points
along length of hair -2
Moderate to severe resistance throughout length of
of hair -3
 Definite snarling of hair, resisting combing at Motor and rheostat
different points along length of hair ~4 Shampoo (J. H. Breck)
Definite snarling of hair throughout length of Interval timer
hair, completely resisting combing —5
B. Microscopic Examination Rating PROCEDURE
Good adhesion —1 A. Preparation of Curl Swatches
Small amount of resin particles -2 1. Separate hair into small swatches of approximately 2
Appearance of polymer fragments protruding from grams in weight.
resin film - increased flaking (moderate) -3 2. Secure hair sample 1 inch from root end by winding
A more general loss of adhesion from hair shafts —4 with cotton thread.
Severe Flaking -5 3. Fold over at root end and secure the hair into a small
Note - With microscope ratings of 3-5 the flaking is usually loop with thread (Figure 1). Leave approximately 6
apparent upon visual examination of the hair swatch inches of loose thread to secure identification tag.
without magnification. 4. Wash hair swatch in a 10% solution of shampoo. Then
rinse thoroughly with distilled water.
5. Comb to untangle the hair shafts.
C. Flaking Rating - Visual Inspection 6. Cut the hair swatch to measure 10 inches in length
None - very slight visible -1 from the looped root end. Dry 1/2 hour at 14O0F.
Slight -2 7. In order to prevent loss of individual hair shafts
Moderate -3 during subsequent combing, glue the root end at the
thread windings with Duco cement
Moderate - severe —4
Severe -5 B. Preparation and exposure of Test Samples
1. Hang a hair swatch from a motor chuck. Secure the
Note - Also report differences in gloss among the end of the tress with a Swingline binder clip to
formulations under test. prevent motion of the hair during spraying. The
motor should be adjusted with a rheostat so that the
shaft rotates at 20 RPM.
2. With the hair swatch rotating, hold the aerosol
DETERMINATION OF THE container at a distance of 7 inches and spray for 10
seconds. Direct the spray uniformly throughout the
HIGH HUMIDITY CURL RETENTION length of the tress.
3. Remove the binder clip and bring fingers gently down
OF AEROSOL HAIR SPRAYS the full length of the hair, squeezing off any excess
resin solution.
(Editor's Method)
4. Comb hair twice, using the wide teeth (9-7/8"
teeth/inch) of a 7 1/2" nylon comb. Roll curl on a
1/2" diameter Teflon mandrel. Remove curl from
mandrel and secure with a "Klippee".
The curl retention properties of polymeric hair spray 5. Dry in a 14O0F oven for 30 minutes and then allow
resins are measured at 90% RH over a period of three hours. curl to condition at 720F at 50% RH overnight.
The change in percentage retention versus time may be 6. Remove "Klippee" and gently bring the curl down by
plotted to illustrate differences among formulations. The unwinding into a spiral with a glass rod.
procedure lends itself to statistical analysis of formulation 7. Place the unwound curls at random on Plexiglass
variables. retention boards (Figure 2). Take initial reading.
8. Set retention boards into humidity chamber and
record length of curl every 30 minutes over a period
of three hours (Figure 3).
EQUIPMENT
12" Remi, Blue String, European, brown hair C. Calculations
Barber shears Calculate percentage curl retention by:
7 1/2" nylon combs
Alligator and paper clips L-Lt
White cotton thread, size # 8 Curl Retention (%) = x 100
Identification tags L - LO
"Klippees" hair clips
Where L = Length of hair fully extended
3" x 1/2" Teflon mandrel Where L0 = Length of hair before exposure
140° oven - Forced air Where Lt = Length of hair after exposure
Swingline binder clips
Plexiglass retention board Note - Determine retention of 10 replicates of each
Humidity Cabinet (precision to ± 2% RH at 7O0F.) variable.
 DETERMINATION OF THE PRESSURE (OR VACUUM) IN AN AEROSOL CONTAINER
(Editor's Method)

SCOPE
The method is satisfactory for all aerosols having a
pressure of 180 psi-g. or less, provided they have a valve to
which the gauge assembly can be connected.
APPARATUS
Apparatus requirements will vary according to the
pressure (or vacuum) in the container, the design of the
aerosol valve, and the accuracy desired. Gauge assemblies
should be selected in accordance with the following table:
PROCEDURE
1. Condition test container by upright immersion in water
bath for 30 minutes.
2. For pre-pressurized gauges, pressurize to 2 or 3 psi.
greater than anticipated test container pressure.
3. Remove container from water bath, actuate briefly,
remove button or foam spout, shake for five seconds,
and apply the gauge assembly firmly to the valve to get a
good seal.

Pressure Range Required

Gauge Pressure Accurate Less Accurate
Type of Gauge Range; psi-g Determination; psi-g. Determination; psi-g.
Inspector's O" to 30" Hg0 O" to 30" Hg0 O" to 30" Hg0**
General Purpose* O" to 30" Hg0 O" to 30" Hg0
Inspector's O to 60 O to 40 O to 40**
General Purpose O to 60 30 to 60
Inspector's O to 160 30 to 100 30 to 100**
General Purpose O to 160 30 to 160
Inspector's O to 300 80 to 180 60 to 180**
General Purpose O to 300 60 to 300

*General Purpose Gauges may be used with or without the pre-pressurization assembly.
**Inspector's Gauge is used without the pre-pressurization assembly.

For the accurate readings the U. S. Inspector's Gauge is
threaded to a coupling, then a toggle valve and finally to an
adapter made to match the valve of the test unit. For less
accurate readings the Inspector's Gauge may be used
without the pre-pressurization assembly.
For the less accurate readings a General Purpose Gauge
may also be used, either with or without the
pre-pressurization assembly. If pre-pressurization is not
desired, connect the gauge directly to the valve adapter. In
this case the gauge should be either two-inch or three-inch
diameter, to minimize the size of the Bourdon tube.
Individual gauges or complete assemblies may be
purchased from several sources, including Builder's Sheet
Metal Works, Inc., 108 Wooster Street, New York, N. Y.
10012; or General Kinetics Division, 451 Stephens Street,
SW, Atlanta, Ga. 30302. The Inspector's Gauge with
pre-pressurizing assembly cost about $69.29 in 1972. The
General Purpose Gauge with adapter cost about $9.70
during 1972.
Pressure gauges should be checked against standards on a
daily basis, using pure propellent as the pressure source.
A water bath, normally maintained at 70° ± 0.5 0 F., is
required for accurate work. For less accurate studies the
water bath may have a tolerance of ± 1.00F.
4. Actuate the valve by moving the gauge assembly against
it. Open toggle valve and agitate assembly to bring gauge
pointer to maximum reading. Tap gauge face lightly.
Re-actuate several times and take the average of the
readings.
For best readings at elevated temperatures, leave test
container in water bath while taking readings.
5. Clean out any product forced into valve system by
applying gauge to a can containing a P-12/ethanol
mixture.
REPORTING
Accurate gauge readings should be reported to the
nearest + 0.5 psi. Less accurate readings may be reported to
the nearest + 1.0 psi.
Test methods in this chapter supplement discussions
elsewhere in this text. The reader is referred also to the
CSMA Aerosol Guide, Seventh Edition, 1981.
 FORMI
FURNITURE POLISH EVALUATION
INDIVIDUAL RATINGS FOR 4.3

TEST PANEL APPLICA TION NO. 1 TEST PANEL APPLICA TION NO. 2
Properties Left Center Right Left I Center Right

TEST PANEL APPLICA TION NO. 3 TESTPANEL APPLICATION NO. 4

Properties Left Center Right Left Center Right

Rating Scale: 0-5 NOTE: Designate position of product (X or C) in box

5 - Excellent 3 - Good 1 - Poor designating position on test panel; i.e., Left, Center, or
4 - Very Good 2 - Fair O - Complete Failure Right.

TENTATIVE METHOD FOR
FURNITURE POLISH EVALUATION
Subcommittee B (R. J. Thimineur, Chairman) of the
Scientific Committee of the Waxes, Polishes and Floor
Finishes Division, (C.S.M.A.) approved the following
method in New York City, December, 1970.
1. Fun ct io nal D efin itio n
1.1 A furniture polish aids in cleaning, improving the
appearance, and protecting the furniture finish from such
common damaging effects as finger marking, food stains
and other surface deposits that are commonly encountered
with furniture.
2. Scope
2.1 This test method is intended to define the
properties to be tested, the apparatus to use and the
comparisons of product performance. It is recognized that
considerable discretion exists among formulators and
marketers of furniture polish on what properties or
performance characteristics are best for their products. This
test method will be flexible to honor this fact within the
confines of the furniture polish definition above.
2.2 The methods of testing are subjective and
empirical in order to conform to the basic characteristics of
the industry and to allow flexibility in testing.
3. Apparatus and Materials
3.1 Sample of polish to be tested
3.2 Sample of control polish. The control polish is
selected subjectively for comparison to the test polish. It
may be a competitive product, a modified formulation of
the test polish, etc. The one stipulation is that the control
polish be of the same or similar type as the test polish. For
example, if the test polish is an aerosol emulsion polish, the
control should be an aerosol emulsion polish. It would not
be meaningful to select a paste or liquid product as a
control for comparison with an aerosol emulsion test
polish.
3.3 Test Substrates. The test substrate should be one
for which the test polish is intended. The test surface
should be in good physical condition, not badly cracked,
scratched, or otherwise damaged so as to interfere with
evaluation of polish properties. The minimum test surface
area for each sample should be 200 sq. in.
3.4 Polishing cloth. The same type of polishing cloth
should be used with each sample tested. Separate cloths are
to be used for each sample. Materials such as washed cheese
cloth, rumple cloth, flannel, cotton diaper cloth, and
non-woven fabrics are suitable for this purpose. Felt or
paper should not be used.
3.5 Cleaning solvent. Aliphatic solvents with KB
values less than 38.
3.6 Eye droppers and tap water.
4. Procedure
4.1 Cleaning of Test Substrates. An aliphatic solvent
having a KB value less than 38 should be used to clean the
test substrate thoroughly. Paper towels may be used to
apply the solvent to the surface and to wipe it clean. New
towels should be used each time.
4.2 Application of Polish. Assuming the control
polish or the test polish is a commercially available product,
follow the directions on the container so far as is possible.
When in doubt on the method of use, the directions for
similar products may be used. Equal volumes of control and
test polish should be used to avoid excessively thin or heavy
coats of polish. One or two applications may be used
depending on the substrate and the discretion of the tester.
The same number of coats must be used for both the test
polish and the control polish.
4.3 Placement of Polishes. A controlled randomized
method of laying out the test (x) and control (c) polishes is
represented as follows:
Test Panel Left Center Right
1 C X C
2 C C X
3 X C X
4 X X C
These four positionings should be written on tags and
drawn randomly by each of the four who apply the polishes
Four persons are involved with the application and
evaluation of the polishes. Each of the four persons take
turns applying the polishes while the other three rate all
properties except application properties. The persons
applying the polishes rate ease of use and other application
properties. This means there will be only four readings on
application properties. The three rating the polishes do not
observe the application because they rate properties of each
polish "blind."
5. Evaluation
Five minutes following the application of the polishes
evaluation is begun. In each case comparison is made
between the test polish and the control. All or any number
of the following properties are evaluated:
5.1 Ease of Rub-Up to Maximum Gloss. During
application of the polishes note the time and ease with
which each product develops maximum gloss.
5.2 Gloss. Evaluated as depth of gloss.
5.3 Uniformity. Observe the surface for streaks,
unpolished dry spots and general uniformity.
5.4 Film Clarity. Observe the clearness or sharpness of
an object's image in the polished surface. Overhead lights,
face, hand, or other objects may be used for reflection. This
test may be eliminated for low luster surfaces that do not
possess mirror-like finishes.
5.5 Smear and Mar Resistance. Smear is the degree of
oiliness or greasiness after the polish is rubbed-up to the
desired polish appearance. Mar is the degree of film damage
resulting from a glancing blow to the polish substrate.
Check smear by making a design such as an "S" with one's
finger. A glancing blow with one's knuckles or soft object
such as a book or magazine may be used for determining
the degree of mar.
5.6 Film Healing. Observe the length of time required
for the smear or mar in 5.5 to disappear from the polish
film.
5.7 Rebuff ability. Observe the ease and completeness
of repairability when the smears and mars are buffed with a
polishing cloth. The amount of physical effort and length
of time required is noted.
5.8 Cleaning. Observe the ease of removal of old
polish films as well as common soiling materials such as
dust, grease, oils, finger marks, beverage stains, etc. This
may be done either in the lab or observed during actual use
trials of the products. In the lab, removal of old polish may
be determined by applying multiple coats (10-20
applications) and determining polish build-up. A polish
showing little build-up would be rated a good cleaner for
old polish. Other materials, such as greases, oils, sugar
solutions, etc., should be tested on an individual basis.
5.9 Water Spotting. At least two hours after
application of the polishes, place at random to the polished
surfaces several spots of water about the size of a penny.
Allow the water to remain on the surface for 5-minutes,
15-minutes, 30-minutes, and 1-hour. At precise intervals,
blot the water with a paper towel or other absorbent
material. DO NOT WIPE! Observe the presence and degree
of film damage. Other materials such as milk, coffee, juice,
alcoholic beverages, etc., may be used to supplement the
water test.
5.10 Gloss Retention. Observe the degree of gloss of a
freshly applied polish film compared to that of an aged
polish film.
5.11 Dust Attraction is measured by carefully wiping
the test surface to remove all dirt and dust. Place the test
substrate in the place of your choice to accumulate dust.
Check dust build-up on the panel after 24, 48, and 72
hours, and after one week.
6. Report
6.1 All properties should be rated 0-5. A value of 5
equals excellent and O equals complete failure. Values in
between are various degrees between these extremes.
6.2 Form I should be used to record the raw data.
Form II should be used to summarize and compare the raw
data. The following calculation provides a rating factor for
each property tested:
F = Rating factor for test polish
FQ - Rating factor for control polish
^property= Sum of all readings of a specific property
for the test polish.
/"<
^property = Sum of all readings of a specific property
for the control polish.
n = Number of observations.
X property _ C property
h fi^C n
MILITARY STANDARD SAMPLING
PROCEDURES MIL-STD-IOSD
Many excellent sampling tables exist for acceptance
inspection by attributes. Each of the tables differs
from all others in various ways. Some emphasize select-
ing the plans by Acceptable Quality Level, AQL, the
maximum percent defective that can be considered
satisfactory as a process average; others by Average
Outgoing Quality Limit, AOQL, the maximum Average
Outgoing Quality, AOQ, which can be accepted.
Sampling tables provide different inspection levels; single,
double, and multiple types of sampling plans; and normal,
tightened, and reduced degrees of inspection. Regardless
of their characteristics, each set of tables possesses plans
for which the Operating Characteristic, or OC, Curves
and AOQ Curves have been plotted, or for which these
curves may be determined. If a person keeps these
curves in mind and recognizes their value, any of the
tables may be used efficiently and satisfactorily.
In this chapter we shall discuss the Military Standard
Sampling Procedures and Tables for Inspection by At-
tributes, or MIL-STD-105D, the international designation
of which is ABC-STD-105. These tables will be discussed,
not because they are necessarily the -best" tables, but,
rather, because they are gaining the widest acceptance
at the present time and are representative of most all
sampling plans.
These standards have undergone several revisions since
they were first adopted by the Department of Defense
in September 1950. The present revision, MIL-STD-
105D, 29 April 1963, has been approved by the Depart-
ment of Defense and is mandatory for use by the Depart-
ments of the Army, the Navy, the Air Force and the
Defense Supply Agency. The large number of defense
contracts awarded since the original adoption may,
in large part, account for the wide acceptance of the
standards by industry as well as the Government.
A. Preliminary Considerations
As with any sampling plan, one of the first considerations
is to select the characteristic to be checked, and to provide
an adequate test procedure.
B. Selection of an AQL Value
After the characteristic has been defined, MIL-STD
105D requires the selection of an AQL. This necessitates
a classification of particular characteristic as critical,
major, or minor. The usage of the particular character-
istic must be known and considered with respect to the
possibility and consequences of failure in assembly and
use. If the characteristic is of a critical nature, a low
AQL must be selected. If it is of a non-critical nature,
a higher AQL would be indicated. Where a classification
of characteristics has been made by specifying them
as critical, major, minor, the selection of an AQL value
can be greatly simplified.
C. Normal, Tightened, and Reduced Inspection
MIL-STD-105D provides a specific set of rules as to
the use of nermal, tightened, and reduced inspection,
including continuing one and switching from one to
another.
The following instructions are taken from MIL-STD-
105D.
INITIATION OF INSPECTION. Normal inspection
will be used at the start of inspection unless otherwise
directed by the responsible authority.
CONTINUATION OF INSPECTION. Normal, tight-
ened or reduced inspection shall continue unchanged for
each class of defects or defectives on successive lots or
batches except where the switching procedures given
below require a change. The switching procedures shall
be applied to each class of defects or defectives inde-
pendently.
SWITCHING PROCEDURES
NORMAL TO TIGHTENED. When normal inspec-
t.on is in effect, tightened inspection shall be instituted
when 2 out of 5 consecutive lots or batches have been
rejected on original inspection (i.e., ignoring resubmitted
lots or batches for this procedure).
TIGHTENED TO NORMAL. When tightened inspec-
tion is in effect, normal inspection shall be instituted
when 5 consecutive lots or batches have been considered
acceptable on original inspection.
N O R M A L TO REDUCED. When normal inspection
is in effect, reduced inspection shall be instituted provid-
ing that all of the following conditions are satisfied:
a. The preceding 10 lots or batches (or more, as indicated
by the note to Table V I I I ) have been on normal
inspection and none has been rejected on original
inspection; and
6. The total number of defectives (or defects) in the
samples from the preceding 10 lots or batches (or such
other number as was used for condition -a*' above)
is equal to or less than the applicable number given
in Table VIII. If double or multiple sampling is
in use, all samples inspected should be included,
not "'first" samples only; and
c. Production is at a steady rate; and
d. Reduced inspection is considered desirable by the
responsible authority.
REDUCED TO NORMAL. When reduced inspection
is in effect, normal inspection shall be instituted if any
of the following occur on original inspection:
a. A lot or batch is rejected: or
b. A lot or batch is considered acceptable under certain
conditions of reduced inspection when the sampling
procedure may terminate without either acceptance
or rejection criteria having been met. In these
circumstances, the lot or batch will be considered
acceptable, but normal inspection will be reinstated
starting with the next lot or batch.
c. Production becomes irregular or delayed; or
d. Other conditions warrant that normal inspection
shall be instituted.
DISCONTINUATION OF INSPECTION. In the event
that 10 consecutive lots or batches remain on tightened
inspection (or'such other number as may be designated
by the responsible authority), inspection under the pro-
visions of this document should be discontinued pending
action to improve the quality of submitted material.
It should be noted here that many of the rules estab-
lished for the use of MIL-STD-105D by the Government
have been modified for use in industry. The foregoing
rules have often been modified or changed, but the pro-
spective user of these plans should benefit by restating
the above conditions with his own modifications to insure
a definite plan for changing from one type of plan to
another.
D. Selection of an Inspection Level
The MIL-STD-105D tables provide for varying the
discriminatory power of the sampling plan by selecting
one of four special inspection levels, S-I, S-2, S-3, S-4,
or one of three general inspection levels, I, II, III. In
each case the higher the number the more discriminatory
the plan. The more discriminatory the plan the larger
the sample size. The OC Curves and certain tabulated
values for them are given.
Inspection level II is usually selected.
E. Single, Double, and Multiple Sampling
For most AQL values and inspection levels, the user
has the opportunity to select single, double, or multiple
sampling. The selection of the type of sampling plan
will depend upon:
1. Psychological likes and dislikes,
2. Difficulty of administering and using the various
kinds of sampling,
3. Average amount of inspection required by using
the various kinds of sampling, and
4. Difficulty in selecting samples.
are accumulated. If the accumulated number of defec-
tives is equal to or less than the second acceptance number,
the lot should be accepted. If the accumulated number
of defectives is equal to or greater than the second rejection
number, the lot should be rejected. In multiple sampling
this process is repeated until a decision is reached.
It is observed that a decision is reached with one sample
in single sampling, with two or less samples in double
sampling, and with seven or less samples in multiple
sampling. However, the size of each sample decreases
as one goes from single to multiple sampling, so that,
in general, a smaller average amount of inspection is
necessary for multiple than for single sampling.
The disposition of any defectives found in the samples
will depend upon the nature and magnitude of the defect
and upon the agreement made between the producer
and consumer.
G. An Illustrative Example

However, for any given AQL and inspection level, the OC Suppose we have agreed on an AQL of 1.5°0 for a
curves will be nearly the same regardless of the kind of certain characteristic, a shipment of 1500 pieces is received,
sampling. and we are to determine appropriate sampling plans
based on general inspection level II, normal inspection.
From Table I, we obtain the sample size code letter K.
F. Selecting a Sampling Plan
From Tables II, III, and IV, we obtain:
After the preceding decisions have been made, a sampl- Single Double
ing plan can be selected for any given lot when the lot Sampling Sampling
size is known. A sample size code letter, dependent
upon the lot size and the inspection level, is obtained Sample Sample Sample
from Table I. (See page 342.) Size Ac Re Number Size Ac Re

The appropriate sampling plan, sample size(s) with 125 5 6 First 80 2 5

acceptance and rejection number(s), can then be obtained Second 80 6 7
from Tables IMV.
Multiple
In a single sampling plan, Table II, a sample of the
Sampling
prescribed size is drawn at random from the lot, the
characteristic checked, and the number of defectives Sample Sample
determined. If the number of defectives found in the Number Size Ac Re
sample is equal to or less than the acceptance number,
First 32 4
the lot should be accepted. If the number of defectives
Second 32 1 5
found in the sample is equal to or greater than the rejection
Third 32 2 6
number, the lot should be rejected.
Fourth 32 3 7
In a double or multiple sampling plan, Table III or Fifth 32 5 8
Table IV, a first sample of the prescribed size is drawn Sixth 32 7 9
at random from the lot, the characteristic checked, and Seventh 32 9 10
the number of defectives determined. If the number
of defectives found in the first sample is equal to or less
than the first acceptance number, the lot should be ac- H. Additional Features of MIL-STD-W^D
cepted. If the number of defectives found in the first
sample is equal to or greater than the first rejection Although the AQL is commonly thought of as a percent
number, the lot should be rejected. If the number of defective, it can also be interpreted as the number of
defectives found in the first sample is greater than the defects per 100 units. The single sampling plans are
first acceptance number and less than the first rejection particularly adaptable and have AQL values ranging
number, a second sample of the prescribed size is drawn from 0.010 to 1000 defects per 100 units. This permits
from the lot. The characteristic is checked, and the the use of these tables where the number of defects is
number of defectives found in the first and second samples a better measure of quality than is the percent defective.
 TIN PLATE FABRICATED AEROSOL CANS

202 Diameter
202 x 200 202 x 406 202 x 700
202 x 214 202 x 509 202 x 708 METHOD FOR DETERMINING DELIVERY RATE
202x314 202x608
207.5 Diameter OF AEROSOL PRODUCTS
207.5x413 207.5x605 207.5x708
207.5x509 207.5x701 207.5x713
211 Diameter
211 x407.5 211 x 5 1 0 211 x 6 1 2
211 x 4 1 3 211 x604 211 x713 Introduction
The delivery rate of an aerosol dispenser is determinable
300 Diameter by measuring the quantity of material expelled through the
300 x 709
valve in a given time. The exact duration of discharge and
the temperature of the dispenser must be carefully
Dimensions included in Industry Specifications to Date: controlled for good reproduceability.
A. Inside Diameter of 1" Cup Opening. Delivery rate tests assist in evaluating one aspect of valve
B. Outside Diameter of 1" Cup Opening. performance and are considered a prerequisite to both
C. Height of Curl Opening above Double Seam. biological and storage testing. Biological and clinical tests
D. Height over Double Seam. are often made with one or two dispensers selected from a
group similiar in all respects except for delivery rate. When
E. Overall Height of Container. it is desirable to select dispensers with equal delivery rates,
F. Height Between Double Seams. two or three tests should be performed on each dispenser.
G. Thickness of Curl Around the One Inch Opening. In the case of storage tests a single delivery rate test is
normally performed at each examination period to conserve
the contents and extend the life of the dispenser.
Equipment
Water-bath, maintained at 70° ±0.5°F. (80° ±0.5° F.
for insecticides, to match Peet-Grady test requirements),
with a screen or perforated metal shelf one inch above the
bottom of the bath.
Stirrer — air or electric.
Balance, one-tenth gram scale.
Stop watch or electric timer.

Procedure
Remove the protective cover, paper label and other
detachable materials from the dispenser, except the button.
If a foam spout is used, remove, cut away all non-essential
plastic, and replace.
Activate the valve for a few seconds, remove any valve
cup impingements, and weigh the dispenser to the nearest
0.05 gram. Place the dispenser on the shelf in the water
bath, which is at the test temperature of 70° ± 0.1° F.
Keep the dispensers in an upright position, spaced one
inch apart and covered with one inch of water. Circulate
the water with the mechanical stirrer. Hold for 15 minutes
in the case of essentially anhydrous products and for 25 to
30 minutes for viscous or water-based formulations.
Remove the dispenser. Actuate the valve for a given time
(preferably 10 seconds), preferably into an exhaust hood.
Dry the dispenser with a cloth or towel and use a blast of
compressed air to remove moisture from the mounting cup
and can seams, in the case of metal containers. Reweigh the
dispenser and compute the difference in weight.
 The results of each test are reported as delivery rate in
grams per second and calculated as follows:
Weight loss in grams
Delivery Rate =
10 Seconds
Repeat the above procedure if duplicate tests are made
on dispensers.
Discussion
The reproduceability of results by this method is
considered to be about ±1%. Precision is about ±2%.
Any air, trapped in the product and outage space, may
be expected to have an elevating effect upon delivery rates.
The effect decreases as the can is emptied. The rate of
decrease is much more rapid for dispensers with vapor-tap
valves. Even without air, minor distillation effects will
occur within the dispenser as it is used up, and these will
have a small depressing effect upon delivery rates.
The variation in valve orifice diameters may be as high as
±10%. Because of this, delivery rates may vary as much as
±15% from can to can.
Generally, the product should be shaken before any
actuation. Reference should be made to label instructions.
In the case of thixotropic, rheopectic and certain other
emulsion types the degree of agitation will have a
considerable effect upon delivery rates.
The delivery rate of freshly prepared aerosols will not
usually be the same as that after aging. This is due to gasket
swelling, changes in elastomer hardness and other factors.
NOTH: About 0.03% of sodium nitrite may be added to
the water bath to prevent the minor rusting of
some tinplate containers.