Standard Method of Test for

Soundness of Aggregate by Use of Sodium

Sulfate or Magnésium Sulfate

AASHTO Désignation: T 104-99 (2007)

1. SCOPE

1.1. This method covers the procédure to be followed in testing aggregates to deteraiine their
résistance to disintegration by saturated solutions of sodium sulfate or magnésium sulfate. This is
accomplished by repeated immersion in saturated solutions o f sodium or magnésium sulfate
followed by oven drying to partially or completely dehydrate the sait precipitated in permeable
pore spaces. The internai expansive force, derived from the rehydration of the sait upon re­
immersion, simulâtes the expansion of water on freezing. This test method fumishes information
helpful in judging the soundness of aggregates subject to weathering action, particularly when
adequate information is not available from service records of the material exposed to actual
weathering conditions. Attention is called to the fact that test results by the use of the two salts
differ considerably and care must be exercised in fîxing proper limits in any spécifications that
may include requirements for these tests.

1.2. The values stated in SI units are to be regarded as the standard.

1.3. This standard may involve hazardous materials, opérations, and equipment. This standard does
not purport to address ali o f the safety concerns associated with its use. It is the responsibility o f
the m er o f this standard to consult and establish appropriate safety and health practices and
determine the applicability o f regulatoiy limitations prior to m e.

Z REFERENCED DOCUMENTS

2.1. AASHTO Standards:

H M 92, Wire-Cloth Sieves for Testing Purposes
■ M 231, Weighing Devices Used in the Testing o f Materials
■ R 16, Regulatory Information for Chemicals Used in AASHTO Tests
■ T 27, Sieve Analysis o f Fine and Coarse Aggregates

2.2. ASTM Standards:

■ C 670, Practice for Preparing Précision Statements for Test Methods for Construction
Materials
■ E 100, Spécification for ASTM Hydrometers

TS-1C T 104-1 AASHTO

3. APPARATUS

3.1. Sieves— With square openings o f the following sizes conforraing to M 92, for sieving the samples
in accordance with Sections 5, 6, and 8:
4.75 mm (No. 4) 63 mm (21/2 in.)
4.00 mm (No. 5) 50 mm (2 in.)
2.36 mm (No. 8) 37.5 mm ( l '/ 2 in.)
1.18 mm (No. 16) 31.5 mm (1 XU in.)
600 jim (No. 30) 25.0 mm (1 in.)
300 nm (No. 50) 19.0 mm CU in.)
150 jjm (No. 100) 16.0 mm (5/8 in.)
12.5 mm (Vz in.)
9.5 mm (% in.)
8.0 mm (5/i6 in.)

3.2. Containers fo r Samples— Sieves 203.2 mm (8 in.) in diameter for each separate size fractions of
aggregate during test. Used, out-of-tolerance sieves according to M 92, in acceptable condition,
may be used as containers (Note 1).

3.2.1. Coarse Aggregate—2.36-mm (No. 8) Size.

3.2.2. Fine Aggregate—250- jam (No. 60) Size.

Note 1— Sieves 203.2 mm (8 in.) in diameter may be substituted with containers which permit
free access of solution to the sample and drainage o f the solution from the sample without loss of
aggregate. Substitution with such containers may affect results. Referee testing, comparison
testing, or testing o f aggregate to be used in critical applications must be performed using sieves
203.2 mm (8 in.) in diameter.

3.3. Apparatus forîm m ersing Samples in Solution (optional)—When necessary, apparatus for
holding the sieves containing the sample for immersing into the solution shall be constructed in
such a manner to permit free access of the solution to the sample and drainage of the solution from
the sample.

3.4. Temperature Régulation—Suitable means for regulating the temperature o f the samples during
immersion in the sodium sulfate or magnésium sulfate solution shall be provided.

3.5. Thermometer—A thermometer covering the recommended temperature range for solutions during
test and readable to 0.1°C (0.2°F).

3.6. Temperature Recorder—A unit capable of recording solution temperature a minimum of once
every 10 minutes for the duration of the test with an accuracy o f 0.3°C (0.5°F).

3.7. Balance—The balance shall have sufficient capacity, be readable to 0.1 percent o f the sample
mass, or better, and conform to the requirements o f M 231.

TS-1c T 104-2 AASHTO

3.8, Dtying Oven—The oven shall be capable o f being heated continuously at 110° ± 5°C (230° ± 9°F)
and the rate of évaporation, at this range o f temperature, shall be at least 25 g/h for 4 hours, during
which period the doors of the oven shall be kept closed. This rate shall be determined by the loss
of water from 1-liter Griffin low-form beakers, each initially containing 500 g o f water at a
temperature of 21° ± 2°C (70° ± 3°F), placed at each corner and the center o f each shelf of the
oven. The évaporation requirement is to apply to ail test locations when the oven is empty except
for the beakers of water.

3.9. Specific Gravity Measurement—Hydrometers conforming to the requirements o f ASTM E 100, or

a suitable combination of graduated glassware and balance, capable of measuring the solution
specific gravity within ±0.001.

4. SPECIAL SOLUTIONS REQUIRED

4.1. Préparé the solution for immersion of test samples from either sodium or magnésium sulfate in
accordance with Section 4.1.1 or 4.1.2 (Note 2). The volume o f the solution shall be at least five
times the solid volume o f ail samples immersed at any one time.
Note 2— Some aggregates containing carbonates of calcium or magnésium are attacked
chemically by fresh sulfate solution, resulting in erroneously high-measured losses. If this
condition is encountered or is suspected, repeat the test using a filtered solution that has been used
previously to test the same type o f carbonate rock, provided that the solution meets the
requirements o f Sections 4.1.1 and 4.1.2 for specific gravity.

4.1.1. - Sodium Sulfate Solution—Préparé a saturated solution of sodium sulfate by dissolving a reagent
grade o f the sait in water (Note 5) at a temperature o f 25°C (77°F) minimum. Add sufficient sait
(Note 3) of the anhydrous (Na2S 0 4) form to ensure not only saturation but also the presence of
excess crystals when the solution is ready for use in the tests. Thoroughly stir the mixture during
the addition o f the sait and stir the solution at frequent intervals until used. To reduce évaporation
and prevent contamination, keep the solution covered at ail times when access is not needed.
Allow the solution to cool to 20.3 to 21.9°C (68.5 to 71.5°F). Again stir, and allow the solution to
remain at the designated temperature for at least 48 hours before use. Prior to each use, break up
the sait cake, if any, in the container, stir the solution thoroughly, and determine the specific
gravity of the solution. When used, the solution shall have a specific gravity not less than
1.154 nor more than 1.171. Discard a discolored solution, or filter it and check for specific gravity.
Note 3—For the solution, 215 g of anhydrous sait per liter o f water are sufficient for saturation at
22°C (71.6°F). However, since this sait is not completely stable and since it is desirable that an
excess o f crystals be present, the use of not less than 225 g of the anhydrous sait per liter of water
is recommended.

4.1.2. Magnésium Sulfate Solution—Préparé a saturated solution of magnésium sulfate by dissolving a

reagent grade o f the sait in water (Note 5) at a minimum temperature of 25°C (77°F) (Note 4).
Add sufficient sait (Note 4) of either the anhydrous (M gS04) or the crystalline (MgS04*7H20 )
(Epsom sait) form, to ensure saturation and the presence of excess crystals when the solution is
ready for use in the tests. Thoroughly stir the mixture during the addition of the sait and stir the
solution at frequent intervais until used. To reduce évaporation and prevent contamination, keep
the solution covered at ail times when access is not needed. Allow the solution to cool to 20.3 to
21.9°C (68.5 to 71.5°F). Again stir, and allow the solution to remain at the designated temperature
for at least 48 hours before use. Prior to each use, break up the sait cake, if any, in the container,
stir the solution thoroughly, and determine the specific gravity of the solution. When used, the
solution shall have a specific gravity not less than 1.297 nor more than 1.306. Discard a discolored
solution, or filter it and check for specific gravity'.

TS-1c T 104-3 AASHTO

 Note 4— For the solution, 350 g o f anhydrous sait or 1230 g of the heptahydrate per liter o f water
are suffîcient for saturation at 23°C (73.4°F). However, since these salts are not completely stable,
with the hydrous sait being the more stable of the two, and since it is desirable that an excess of
crystals be present, it is recommended that the heptahydrate sait be used and in an amount o f not
less than 1400 g/L o f water.
When preparing the saturated solution of magnésium sulfate, the sait crystals will dissolve more
readily if the crystals are added in small amounts and the water temperature is higher than 35°C
(95°F) due to the chemical cooling effect during préparation.
Note 5—Distilled water shall be used in referee or comparison testing.

4.2. Barium ChJoride Solution—A 0.2 molar solution o f barium chloride (41.6 g o f BaCl2 per liter of
solution) to determine the presence o f sodium or magnésium sulfate in the wash water.

5. SAMPLES

5.1. Fine Aggregate—Fine aggregate for the test shall be passed through a 9.5-mm (3/g-in.) sieve. The
sample shall be o f such size that it will yield not less than 100 g of each o f the following sizes,
expressed in terms o f the following sieves:
Passing Sieve Retained on Sieve
9.5 mm (3/8 in.) 4.75 mm (No. 4)
4.75 mm (No. 4) 2.36 mm (No. 8)
2.36 mm (No. 8) 1.18 mm (No. 16)
1.18 mm (No. 16) 600 |im (No. 30)
600 jim (No. 30) 300 jim (No. 50)

5.1.1. Should the sample contaîn less than five percent o f any o f the sizes specified in Section 5.1, that
size shall not be tested.

5.2. Coarse Aggregate— Coarse aggregate for the test shall consist o f material from which the sizes
finer than the 4.75-mm (No. 4) sieve have been removed. The sample shall be o f such a size that it
will yield the amounts indicated in Table 1.
Table 1— Coarse Aggregate Sample
Sieve Size Mass, g
63 mm to 37.5 mm (27; in. to 172 in.) 5000 ± 300
Consisting of:
50-mm to 37,5-mm (2-in. to 17;-in.) material 2000 ± 200
63-mm to 50-mm (272-in. to 2-in.) material 3000 ± 300

37.5 mm to 19.0 mm (17; in. to % in.) 1500i 50

Consisting of:
25.0-mm to 19.0-mm (1 -in. to 3/4 -in.) material 500 ±30
37.5-mm to 25.0-mm ( l 'A-in. to 1-in.) material 1000 ±50

19.0 mm to 9.5 mm (V4 in. to *’/5 in.) 1000 = 10

Consisting of:
12.5-mm to 9.5-mm (7:-in. to V s-in.) material 330 ±5
19.0-mm to i2.5-mm (V^-in. to 7;-in.) material 670 ± 10

9.5 mm to 4.75 mm (3/s in. to No. 4) 300 ±5

TS-1c T 104-4 AASHTO

5.2.1. Should the sample contain less than five percent o f any o f the sizes specified in Section 5.2, that
size shall not be tested. When a combination of sizes is specified for the test portion and one o f the
sizes specified is less than five percent o f the sample, reduce the test portion by the applicable
mass specified in Section 5.2 for the size not available.

5.2.2. When testing large rock (broken stone, ledge rock, cobbles, and boulders for use as riprap, channel
lining, etc.), obtain the test portion by crushing, splitting, or sawing the larger sample pieces. Test
only those pieces in the 37.5 to 19.0 mm (1 '/2 to 3/4 in.) and 63 to 37.5 mm (2'/2 to 1V2 in.) size
fractions when size réduction is by crushing or splitting. Test pieces in the 63 to 37.5 mm (2]/2 to
1 ’/2 in.) size fraction when size réduction is by sawing.
Note 6—When the rock contains bedding planes, crushing or splitting will usually cause the rock
to break on those bedding planes. Sawing will usually leave the bedding planes intact, and thus
more likely to split the particles due to growth of the sait crystals in the soundness test.

5.2.3. When testing large rock (to evaluate a potential source) which will be subsequently crushed to
produce aggregate, obtain the test portion by crushing the larger sample pieces. Test pieces only in
those sizes that will be included in the produced aggregate, but ignoring any material fïner than
4.75-mm (No. 4) sieve or coarser than 63-mm (2V2-in.) sieve.

5.2.4. When the fïnished aggregate material will contain particles coarser than 63 mm (2*/2 in.), such as
aggregate for use in mass concrete, crush the material coarser than 63 mm (2V2 in.) and distribute
the material among that in the range of 63-mm (2]/2 in.) to 4.75-mm (No. 4) sieves. Discard
material finer than 4.75-mm (No. 4) sieve.

5.3. When an aggregate to be tested contains appréciable amounts o f both fine and coarse material,
having a grading with more than 10 percent coarser than the 9.5-mm (3/s-in.) sieve and, also, more
than 10 percent finer than the 4.75-mm (No. 4) sieve, test separate samples o f the minus 4.75 mm
(No. 4) fraction and the plus 4.75 mm (No. 4) fraction in accordance with the procédures for fine
aggregate and coarse aggregate, respectively. Report the results separately for the fine aggregate
fraction and the coarse aggregate fraction, giving the percentages of the coarse and fine size
fractions in the initial grading.

6. PREPARATION OF TEST SAMPLE

6.1. Fine Aggregate—Thoroughly wash the sample of fine aggregate on a 300-jim (No. 50) sieve, dry
to constant mass at 110° ± 5°C (230° ± 9°F), and separate into the différent sizes by sieving, as
follows: Make a rough séparation of the graded sample by means of a nest of the standard sieves
specified in Section 5.1. From the fractions obtained in this manner, select samples of sufficient
size to yield 100 g after sieving to refusai. (In général, a 110-g sample will be sufficient.) Do not
use fine aggregate sticking in the meshes of the sieves in preparing the samples. Weigh samples
consisting o f 100 ± 0.1 g out of each of the separated fractions after final sieving, record the
masses o f the test samples, and place in separate containers for the test.

6.2. Coarse Aggregate—Thoroughly wash and dry the sample of coarse aggregate to constant mass at
110° ± 5°C (230° ± 9°F) and separate it into the différent sizes shown in Section 5.2 by sieving to
refusai. Weigh out quantities o f the différent sizes within the tolerances of Section 5.2 and
combine them to the designated total mass (Note 7). Record the masses of the test samples and
their fractional components. In the case of sizes larger than 19.0 mm (J/4 in.), record the number of
particles in the test samples.
Note 7— The fractional components of each sample may be placed in separate containers if so
desired but is not required. If separate containers are used, the two sizes must be combined for the
calculations for Section 8.1.2 (Table 2).

TS-1C T 104-5 AASHTO

Table 2— Suggested Form for Recording Test Data (with Illustrative Test Values)
M ass o f T e st Percerttage P a ssin g W eig h te d
G ra d in g o f O rig in al F rac tio n s B efo re D e sig n a te d S ie v e P e rc e n ta g e
S iev e S iz e S am p ie, P e rc e n t T e st, g A fte r T est Loss
Soundness Test o f Fine Aggregate
M inus 150 jim 5
300 51m to 150 nm 12
600 nm to 300 nm 26 100 4.2 1.1
1.18 mm to 600 jim 25 100 4.8 1.2
2.36 mm to 1.18 mm 17 100 8.0 1.4
4.75 mm to 2.36 mm It 100 1J.2 1.2
9.5 mm to 4.75 mm 4 11 .2 " 0.4

Totais 100 5

Soundness Test o f Coarse Aggregate

63 mm to 50 mm 2825 g
63 to 37.5 mm 20 4783 4.8 i.O
50 mm to 37.5 mm 1958 g _
37.5 mm to 25.0 mm 1012g "
j- 37.5 t o î 9.0 mm AS 1525 8.0 3.6
25.0 mm to 19.0 mm 513g .
19.0 mm to 12.5 mm 675 g '
j- I9 .0 to 9 .5 m m 23 1008 9.6 2.2
12.5 mm to 9.5 mm 333 g .
9.5 mm to 4.75 mm 12 298 11.2 i.3

Totals 100 8

The percentagc loss (11.2 percent) o f the next smaîler size is used as the percent âge loss for this size, since this size ccmta ins less than five percent o f the original
sample as received. See Section 10.1.3.4.

7. PROCEDURE

7.1. Storage o f Samples in Solution—Immerse the samples in the prepared solution o f sodium sulfate
or magnésium sulfate for not less than 16 hours nor more than 18 hours in such a manner that the
solution covers them to a depth o f at least 12.5 mm (l/2 in.) (Note 8). Cover the containers to
reduce évaporation and prevent the accidentai addition o f extraneous substances. Maintain the
samples immersed in the solution at a temperature o f 20.3 to 21.9°C (68.5 to 71.5°F) for the
immersion period.
Note 8—Suitably weighted wire grids placed over the sample in the containers will permit this
coverage to be achieved with very light-weight aggregates.

7.2. Diying Samples Âfter Immersion—After the immersion period, remove the aggregate sample from
the solution, permit it to drain for 15 ± 5 minutes, and place in the dtying oven. The temperature of
the oven shall have been brought previously to 110° ± 5°C (230° ± 9°F). Dr)' the samples at the
specifîed temperature until constant mass has been achieved. Establish the time required to attain
constant mass as follows: with the oven containing the maximum sample load expected, check the
mass losses of test samples by removing and weighing them without cooling, at intervais of two to
four hours; make enough checks to establish required drying time for the least favorable oven
location (Section 3.6) and sample condition (Note 9). Constant mass will be considered to have
been achieved when mass loss is less than 0.1 percent of sample mass in four hours of drying.
After constant mass has been achieved, allow the samples to cool to 20 to 25°C (68 to 77°F)
(Note 10), when they shall again be immersed in the prepared solution as described in Section 7.1.
Cooling may be aided by the use of an air conditioner or fan. Temperature of the material shall
be checked by thermometer or other acceptable means before placing the material in the
soaking solution.
Note 9—Di^'ing time required to reach constant mass may vary considerably for several reasons.
Efficiency o f drying will be reduced as cycles accumulate because of sait adhering to particles
and, in some cases, because of increase in surface area due to breakdown. The différent size

TS-1c T 104-6 AASHTO

 fractions of aggregate will have differing drying rates. The smaller sizes will tend to dry more
slowly because o f their larger surface area and restricted interparticle voids, but this tendency may
be altered by the effects o f container size and shape.
Note 10—Expérience has shown that sample temperatures significantly différent than the
solution temperature o f 21.1°C (70.0°F) may change the temperature o f the solution temporarily,
thereby causing a change in sait saturation even though the solution retums to 2Î.1°C (70.0°F) for
most of the soaking period. This may cause erroneous results.

7.3. Number o f Cycles—Repeat the process o f altemate immersion and drying until the required
number o f cycles is obtained. Preferably, the test shall be performed continuously until the
specified number o f cycles is obtained. However, if the test must be interrupted, leave the samples
in the oven, at 110° ± 5°C (230° ± 9°F) until the testing can be resumed.

7.4. Review the temperature record from the recording unit. Verify solution temperature limits were
not exceeded.

8. QUANTITATIVE EXAMINATION

8.1. Make the quantitative examination as follows:

8.1.1. After the completion of the final cycle and after the sample has cooled, wash the sample free from
the sodium sulfate or magnésium sulfate. Wash by circulating water at 43° ± 6°C (110° ± 10°F)
through the samples in their containers by introducing hot water near the bottom and allowing the
water to pass through the samples and overflow. The thoroughness of washing shall be checked by
obtaining a sample of rinse water after it has passed through the samples and checked with
0.2 molar barium chloride. Further washing is required if sample becomes cloudy upon addition of
the barium chloride solution, In areas where the water gives a reaction with barium chloride other
analytical means shall be used to assure thoroughness o f washing. In the washing opération, the
samples shall not be subjected to impact or abrasion that may tend to break up particles.

8.1.2. After the sodium sulfate or magnésium sulfate has been removed, dry each fraction of the sample
to constant mass at 110° ± 5°C (230° ± 9°F). Sieve the fine aggregate over the same sieve on which
it was retained before the test, and sieve the coarse aggregate over the sieve shown below for the
appropriate size o f particle. For fine aggregate, the method and duration of sieving shall be the
same as were used in preparing the test samples. For coarse aggregate, sieving shall be by hand,
with agitation sufficient only to assure that ail undersize material passes the designated sieve. No
extra manipulation shall be employed to break up particles or cause them to pass the sieves.
Determine the mass of the material retained on each sieve and record each amount. The différence
between each of these amounts and the initial mass of the fraction of the sample tested is the loss
in the test and is to be expressed as a percentage o f the initial mass for use in Table 2.

Sieve Used to
Size o f Aggregate Determine Loss
63 mm to 37.5 mm 31.5 mm ( l '/4 in.)
(2 '/2 to I '/: in.)
37.5 mm to 19.0 mm 16.0 mm (s/8 in.)
(IV2 to 3/, in.)
19.0 mm to 9.5 mm 8.0 mm (s/i6 in.)
(3/,, to V* in.)
9.5 mm to 4.75 mm 4.0 mm (No. 5)
(3/s in. to No. 4)

TS-1c T 104-7 AASHTO

9. QUALITATIVE EXAMINATION

9.1. Make a qualitative examination o f test samples coarser than 19.0 mm f /4 in.) as follows (Note 11):

9.1.1. Separate the particles o f each test sample into groups according to the action produced by the test
(Note 11).

9.1.2. Record the number o f particles showing each type of distress.

Note 11—Many types of action may be expected. In général, they may be classified as
disintegration, splitting, crumbling, cracking, flaking, etc. While only particles larger than
19.0 mm (3/4 in.) in size are required to be examined qualitatively, it is recommended that
examination of the smaller sizes be made in order to determine whether there is any evidence
of excessive splitting.

10 . REPORT

10 . 1. The report shall include thefollowing data (Note 12):

10 . 1 . 1 . Mass of each fraction o f each sample before test.

10. 1.2 . Material from each fraction of the sample finer than the sieve designated in Section 8.1.2 for
sieving after test, expressed as a percentage of the original mass of the fraction.

10.1.3. Weighted average calculated from the percentage of loss for each fraction, based on the grading of
the sample as received for examination determined by using T 27 or, preferably, on the average
grading o f the material from that portion of the supply o f which the sample is représentative
except that:

10.1.3.1. For fine aggregates with less than 10 percent coarser than the 9.5-mm (3/g-in.) sieve, assume
sizes finer than the 300-jim (No. 50) sieve to have zéro percent loss and sizes coarser than the
9.5-mm (3/8-in.) sieve to have the same loss as the next smaller size for which test data are
available.

10.1.3.2. For coarse aggregate with less than 10 percent finer than the 4.75-mm (No. 4) sieve, assume sizes
fmer than the 4.75-mm (No. 4) sieve to have the same loss as the next larger size for which test
data are available.

10.1.3.3. For an aggregate containing appréciable amounts o f both fine and coarse material tested as two
separate samples as required in Section 5.3, compute the weighted average losses separately for
the minus 4.75-mm (No. 4) and plus 4.75-mm (No. 4) fractions based on recomputed gradings
considering the fine fraction as 100 percent and the coarse fraction as 100 percent. Report the
results separately giving the percentage of the minus 4.75-mm (No. 4) and plus 4.75-mm (No. 4)
material in the initial grading.

10.1.3.4. For the purpose of calculating the weighted average, consider any size in Section 5.1 or 5.2 that
contain less than five percent of the sample to have the same loss as the average o f the next
smaller and the next larger size, or if one o f these sizes is absent, to have the same loss as the next
larger or next smaller size, whichever is present.

TS-1c T 104-8 AASHTO

10.1.3.5. For large rock tested according to Section 5.2.2 the weighted average (if more than one size
fraction is tested) shall be the arithmetic mean of the loss on the fractions tested.

10.1.3.6. For large rock tested according to Sections 5.2.3 or 5.2.4, the weighted average shall be based on a
sample grading conforming to the middle o f the spécification to which the aggregate will be
produced, or the actual grading as produced. If the spécification grading includes fractions larger
than 63-mm (2V2in.) sieve, assume such sizes have the same percentage loss as the 63 to 37.5 mm
(2 V2 to 1 ‘/2 in.) fraction.

10.1.4. Report the weighted percentage loss to the nearest whole number.

10.1.5 . In the case of particles coarser than 19.0 mm (3/4 in.) before test: (1 ) the number of particles in
each fraction before test, and (2) the number of particles affected, classified as to number
disintegrating, splitting, crumbling, cracking, flaking, etc., as shown in Table 3.

Table 3— Suggested Form for Qualitative Examination (with Illustrative Test Values)
Q ualitative Examination o f Coarse Sizes
Particles Exhibiting Distress ^ j^
Splitting Crum bling Cracking Flaking Particles
Sieve Size No. Percent No. Percent No. Percent No. Percent Before T est
63 mm to 37,5 mm 2 7 — — 2 7 — — 29
37.5 mm to 19.0 mm 5 — — 2 — — — — 50

10.1.6. Kind o f solution (sodium or magnésium sulfate) and whether the solution was freshly prepared or
previously used.

10.1.7. Method of producing particles for test, when reduced from large pieces as described in
Sections 5.2.2, 5.2.3, or 5.2.4.
Note 12—Table 2 , shown with test values inserted for purpose of illustration, is a suggested form
for recording test data. The test values shown might be appropriate for either sait, depending on
the quality o f the aggregate.

11. PRECISION

11.1. For coarse aggregate with weighted average sulfate soundness losses in the ranges o f 6 to 16
percent for sodium and 9 to 20 percent for magnésium, the précision indexes are as indicated, in
Table 4 (Note 13).

TS-1c T 104-9 AASHTO

Table 4— Précision Indexes
Différence
Coefficient of Between Two
Variation (ls%), Tests (d2s%),
percent" percent o f average"
Multilaboraiory:
Sodium sulfate 41 116
Magnésium sulfate 25 71

Single-Operator:
Sodium sulfate 24 68
Magnésium sulfate 11 31
These numbcrs rcprescnt, respectively, the (Is %) and (d2s%) limits as describcd in ASTM C 670.

Note 13—The values in the précision statement are based on testing according to this
method prior to révision in 1991, The révisions in 1991 are believed to improve the précision
of the method.

TS-1c T 104-10 AASHTO