T 22 METHODS OF SAMPLING AND TESTING

Standard Method of Test for

Compressive Strength of Cylindrical Concrete Specimens

AASHTO DESIGNATION: T 22-921

(ASTM DESIGNATION: C 39-861)

1. SCOPE1 for Verifying the Load Indi- maximum, but preferably after an interval
cation of Testing Machines of 12 months.
1.1 This test method covers 5.1.1.2 On original installation or
determination of compressive strength of 3. SUMMARY OF METHOD relocation of the machine.
cylindrical concrete specimens such as 5.1.1.3 Immediately after making repairs
molded cylinders and drilled cores. It is or adjustments which may in any way
limited to concrete having a unit weight 3.1 This method consists of applying a affect the operation of the (weighing)
in excess of 50 lb/ft3 (800 kg/m3). compressive axial load to molded system or the values displayed, except for
cylinders or cores at a rate which is zero adjustments that compensate for the
1.2 The values stated in inch-pound units within a prescribed range until failure
are to be regarded as the standard. weight of tooling, or specimen, or both,
occurs. The compressive strength of the
1.3 This standard may involve hazardous specimen is calculated by dividing the 5.1.1.4 Whenever there is reason to
materials, operations, and equipment. maximum load attained during the test by doubt the accuracy of the results, without
This standard does not purport to address the cross-sectional area of the specimen. regard to the time interval since the last
all of the safety problems associated with verification.
its use. It is the responsibility of whoever 5.1.2 Design-The design of the
uses this standard to consult and establish 4. SIGNIFICANCE AND USE machine must include the following
appropriate safety and health practices features:
and determine the applicability of 4.1 Care must be exercised in the 5.1.2.1 The machine must be power
regulatory limitations prior to use. interpretation of the significance of operated and must apply the load
compressive strength determinations by continuously rather than intermittently,
2. REFERENCED DOCUMENTS this method since strength is not a and without shock. If it has only one
fundamental or intrinsic property of loading rate (meeting the requirements of
concrete made from given materials. Section 7.5), it must be provided with a
2.1 AASHTO Standards: Values obtained will depend on the size supplemental means for loading at a rate
T 23 Making and Curing Concrete and shape of the specimen, batching, suitable for verification. This
Test Specimens in the Field mixing procedures, the methods of supplemental means of loading may be
sampling, molding, and fabrication and power or hand operated.
T 24 Obtaining and Testing Drilled the age, temperature, and moisture
Cores and Sawed Beams of 5.1.2.2 The space provided for test
conditions during curing. specimens shall be large enough to
Concrete
4.2 This test method may be used to accommodate, in a readable position, an
T 67 Load Verification of Testing determine compressive strength of elastic calibration device which is of
Machines cylindrical specimens prepared and cured sufficient capacity to cover the potential
T 126 Making and Curing Concrete in accordance with Methods T 23, T 24, loading range of the testing machine and
Test Specimens in the T 126, T 231, and ASTM C 873. which complies with the requirement of
Laboratory 4.3 The results of this test may be used as ASTM E 74.
T 231 Capping Cylindrical Concrete a basis for quality control of concrete
Specimens proportioning, mixing, and placing NOTE 1-The type of elastic calibration device
most generally available and most commonly
2.2 ASTM Standards: operations; determination of compliance used for this purpose is the circular proving
with specification; control for evaluating ring.
C 683 Test Method for Compressive
effectiveness of admixtures and similar
and Flexural Strength of Con-
uses. 5.1.3 Accuracy-The accuracy of the
crete Under Field Conditions
testing machine shall be in accordance
C 873 Test Method for Compressive with the following provisions:
Strength of Concrete 5. APPARATUS
Cylinders Cast in Place in 5.1.3.1 The percentage of error for the
Cylindrical Molds loads within the proposed range of use of
5.1 Testing Machine-The testing machine the testing machine shall not exceed ±-
E 74 Methods of Calibration of shall be of a type having sufficient 1.0 percent of the indicated load.
Force-Measuring Instruments capacity and capable of providing the
5.1.3.2 The accuracy of the testing
rates of loading prescribed in Section 7.5.
machine shall be verified by applying
1
Except for Section 5.2.1.3, Section 6.2, Section 7.4, 5.1.1 Verification of calibration of the five test loads in four approximately
and the ANNEX permitting the use of neoprene caps testing machines in accordance with equal increments in ascending order. The
with steel extrusion controllers for compressive Method T 67 is required under the difference between any two successive
strength testing of 6 in. diameter by 12 in. concrete
cylinders, this test method is identical to ASTM C
following conditions: test loads shall not exceed one-third of
39-861. 5.1.1.1 After an elapsed interval since the the difference between the maximum and
previous verification of 18 months minimum test loads.

11
T 22 METHODS OF SAMPLING AND TESTING

5.1.3.3 The test load as indicated by the NOTE 2-It is desirable that the bearing faces the radius of the sphere. The diameter of
testing machine and the applied load of blocks used for compression testing of the sphere shall be at least 75 percent of
computed from the readings of the concrete have a Rockwell hardness of not less the diameter of the specimen to be tested.
than 55 HRC.
verification device shall be recorded at 5.2.2.3 The ball and the socket must be
each test point. Calculate the error. E, and 5.2.1 Bottom bearing blocks shall so designed by the manufacturer that the
the percentage of error, Ep, for each point conform to the following requirements: steel in the contact area does not
from these data as follows: permanently deform under repeated use,
5.2.1.1 The bottom bearing block is
E=A-B specified for the purpose of providing a with loads up to 12000 psi (82.7 MPa) on
Ep = 100 (A - B) / B readily machinable surface for the test specimen (Note 5).
where: maintenance of the specified surface NOTE 5-The preferred contact area is in the
conditions (Note 3). The top and bottom form of a ring (described as preferred
A = load, lbf (or N) indicated by the
surfaces shall be parallel to each other. "bearing" area) as shown in Figure 1.
machine being verified, and
The block may be fastened to the platen
B = applied load, lbf (or N) as of the testing machine. Its least horizontal 5.2.2.4 The curved surfaces of the socket
determined by the calibrating dimension shall be at least 3 percent and the spherical portion shall be kept
device. greater than the diameter of the specimen clean and shall be lubricated with a
5.1.3.4 The report on the verification of a to be tested. Concentric circles as petroleum-type oil such as conventional
testing machine shall state within what described in Section 5.2 are optional on motor oil, not with a pressure type
loading range it was found to conform to the bottom block. grease. After contacting the specimen and
specification requirements rather than 5.2.1.2 Final centering must be made application of small initial load, further
reporting a blanket acceptance or with reference to the upper spherical tilting of the spherically seated block is
rejection. In no case shall the loading block when the lower bearing block is not intended and is undesirable.
range be stated as including loads below used to assist in centering the specimen.
the value which is 100 times the smallest The center of the concentric rings, when
change of load that can be estimated on provided, or the center of the block itself
the load-indicating mechanism of the must be directly below the center of the
testing machine or loads within that spherical head. Provision shall be made
portion of the range below 10 percent of on the platen of the machine to assure
the maximum range capacity. such a position.
5.13.5 In no case shall the loading range 5.2.1.3 The bottom bearing block shall
be stated as including loads outside the be at least 1 in. (25 mm) thick when new,
range of loads applied during the and at least 0.9 in. (22.5 mm) thick after
verification test. resurfacing operations, except when the
5.13.6 The indicated load of a testing block is in full and intimate contact with
machine shall not be corrected either by the lower platen of the testing machine,
calculation or by the use of a calibration the thickness may be reduced to 3/8 in.
diagram to obtain values within the (10 mm).
required permissible variation.
NOTE 3-If the testing machine is so designed FIGURE 1 Schematic Sketch of a Typical
5.2 The testing machine shall be that the platen itself can be readily maintained Spherical Bearing Block
equipped with two steel bearing blocks in the specified surface condition, a bottom
with hardened faces (Note 2), one of NOTE-Provision shall he made for holding
block is not required.
the ball in the socket and for holding the entire
which is a spherically seated block that
unit in the testing machine.
will bear on the upper surface of the 5.2.2 The spherically seated bearing
specimen, and the other a solid block on block shall conform to the following
which the specimen shall rest. Bearing requirements: 5.2.2.5 If the radius of the sphere is
faces of the blocks shall have a minimum 5.2.2.1 The maximum diameter of the smaller than the radius of the largest
dimension at least 3 percent greater than bearing face of the suspended spherically specimen to be tested, the portion of the
the diameter of the specimen to be tested. seated block shall not exceed the values bearing face extending beyond the sphere
Except for the concentric circles given below: shall have a thickness not less than the
described below, the bearing faces shall difference between the radius of the
not depart from a plane by more than sphere and radius of the specimen. The
0.001 in. (0.025 mm) in any 6 in. (152 Diameter Test Maximum Diameter of least dimension of the bearing face shall
mm) of blocks 6 in. in diameter or larger, Specimens in. (mm) Bearing Face in. (mm) be at least as great as the diameter of the
or by more than 0.001 in. in the diameter 2 (51) 4 (102)
sphere (see Figure 1).
3 (76) 5 (127)
of any smaller block; and new blocks 5.2.2.6 The movable portion of the
4 (102) 6 ½ (165)
shall be manufactured within one-half of bearing block shall be held closely in the
6 (152) 10 (254)
this tolerance. When the diameter of the 8 (203) 11 (279) spherical seal. but the design shall be
bearing face of the spherically seated such that the bearing face can be rotated
block exceeds the diameter of the NOTE 4-Square bearing faces are permissible, freely and tilted at least 4° in any
specimen by more than ½, in. (13 mm), provided the diameter of the largest possible direction.
concentric circles not more than 1/32 in. inscribed circle does not exceed the above
(0.8 mm) deep and not more than 3/64 in. diameter. 5.3 Load Indication:
(1.2 mm) wide shall be inscribed to 5.3.1 If the load of a compression
facilitate proper centering. 5.2.2.2 The center of the sphere shall machine used in concrete tests is
coincide with the surface of the bearing registered on a dial, the dial shall be
face within a tolerance of ± 5 percent of provided with a graduated scale that can

12
T 22 METHODS OF SAMPLING AND TESTING

be read to at least the nearest 0.1 percent 6.2 Neither end of compressive test Test Age Permissible Tolerance
of the full scale load (Note 6). The dial specimens when tested shall depart from 24 h ±0.5 h or 2.1 percent
shall be readable within 1 percent of the perpendicularity to the axis by more than 3 days 2 h or 2.8 percent
indicated load at any given load level 0.5° (approximately equivalent to a 7 days 6 h or 3.6 percent
28 days 20 h or 3.0 percent
within the loading range. In no case shall difference in height of 1/16 in. for a 6 in.
90 days 2 days or 2.2 percent
the loading range of a dial be considered diameter cylinder). The ends of
to include loads below the value which is compression test specimens that are not
100 times the smallest change of load plane within 0.002 in. (0.050 mm) shall
which can be read on the scale. The scale be capped, sawed, or ground in
shall be provided with a graduation line accordance with T 231 or if the ends
equal to zero and so numbered. The dial meet the requirements of A6, then
pointer shall be of sufficient length to neoprene caps with steel controllers may
reach the graduation marks; the width of be used instead of capping. The diameter
the end of the pointer shall not exceed the used for calculating the cross-sectional
clear distance between the smallest area of the test specimen shall be
graduations. Each dial shall be equipped determined to the nearest 0.01 in. (0.25
with a zero adjustment which is easily mm) by averaging two diameters
accessible from the outside of the dial measured at right angles to each other at
case, and with a suitable device that at all about mid-height of the specimen.
times until reset will indicate to within 6.3 The number of individual cylinders
one percent accuracy the maximum load measured for determination of average
applied to the specimen. diameter may be reduced to one for each
ten specimens or three specimens per
NOTE 6-As close as can reasonably be read is
considered to be 1/50 in. (0.5 mm) along the arc day, whichever is greater, if all cylinders
described by the end of the pointer. Also, one- are known to have been made from a
half of the scale interval is about as close as single lot of reusable or single-use molds
can reasonably be read when the spacing on which consistently produce specimens
the load indicating mechanism is between 1/25 with average diameters within a range of
in. (1 mm) and 1/16 in. (1.6 mm). When the 0.02 in. (0.51 mm). When the average
spacing is between 1/16 in. and ⅛ in. (3.2 diameters do not fall within the range of
mm), one-third of a scale interval can be read
0.02 in. or when the cylinders are not
with reasonable certainty. When the spacing is
1/8 in. or more, one-fourth of a scale interval made from a single lot of molds. each
can be read with reasonable certainty. cylinder tested must be measured and the
value used in calculation of the unit FIGURE 2 Sketches of Types of Fracture
5.3.2 If the testing machine load is compressive strength of that specimen.
indicated in digital form, the numerical When the diameters are measured at the
reduced frequency, the cross-sectional 7.4 Placing the Specimen-Place the plain
display must be large enough to be easily
areas of all cylinders tested on that day (lower) bearing block, with its hardened
read. The numerical increment must be
shall be computed from the average of face up. on the table or platen of the
equal to or less than 0.10 percent of the
the diameters of the three or more testing machine directly under the
full scale load of a given loading range.
cylinders representing the group tested spherically seated (upper) bearing block.
In no case shall the verified loading range
that day. Wipe clean the bearing faces of the upper
include loads less than the minimum
and lower bearing blocks and of the test
numerical increment multiplied by 100. 6.4 The length shall be measured to the specimen and place the test specimen on
The accuracy of the indicated load must nearest 0.05 D when the length to the lower bearing block. Carefully align
be within 1.0 percent for any value diameter ratio is less than 1.8, or more the axis of the specimen with the center
displayed within the verified loading than 2.2, or when the volume of the of thrust of the spherically seated block.
range. Provision must be made for cylinder is determined from measured The spherically seated block is to be
adjusting to indicate true zero at zero dimensions. rotated immediately prior to testing. to
load. There shall be provided a maximum
assure the freedom required in Section
load indicator that at all times until reset
7. PROCEDURE 5.2.2.6.
will indicate within one percent system
accuracy the maximum load applied to 7.5 Rate of Loading-Apply the load
the specimen. continuously and without shock.
7.1 Compression tests of moist cured
specimens shall be made as soon as 7.5.1 For testing machines of the screw
practicable after removal from moist type, the moving head shall travel at a
6. SPECIMENS
storage. rate of approximately 0.05 in. (1.3
mm)/min when the machine is running
7.2 Test specimens shall be kept moist by idle. For hydraulically operated
6.1 Specimens shall not be tested if any
any convenient method during the period machines, the load shall be applied at a
individual diameter of a cylinder differs
between removal from moist storage and rate of movement (platen to crosshead
from any other diameter of the same
testing. They shall be tested in the moist measurement) corresponding to a loading
cylinder by more than 2 percent (Note 7).
condition. rate on the specimen within the range of
NOTE 7-This may occur when single use 7.3 All test specimens for a given test age 20 to 50 psi/s (0.14 to 0.34 MPa/s). The
molds are damaged or deformed during shall be broken with the permissible time designated rate of movement shall be
shipment, when flexible single use molds are tolerances prescribed as follows: maintained at least during the latter half
deformed during molding or when a core drill
of the anticipated loading phase of the
deflects or shifts during drilling.
testing cycle.

13
T 22 METHODS OF SAMPLING AND TESTING

7.5.2 During the application of the first 10. PRECISION Figure Al. Other modes of manufacture
half of the anticipated loading phase a may be used such as steel casting or
higher rate of loading shall be permitted. machining in one piece from round stock
10.1 The precision of this method has provided the inside diameter, minimum
7.5.3 Make no adjustment in the rate of not yet been determined, but data are
movement of the platen at any time while wall thickness, and minimum bearing
being collected, and a precision statement surface thickness comply with the
a specimen is yielding rapidly will be included when it is formulated.2
immediately before failure. dimensions shown in Figure Al. All
bearing surfaces of the controller, both
7.6 Apply the load until the specimen inside and outside, shall be machine
fails, and record the maximum load ANNEX
planed to within 0.002 in. (0.05 mm).
carried by the specimen during the test. During use, the outside bearing surface
Note the type of failure and the Compressive Strength of Cylindrical shall be maintained free of gouges, or
appearance of the concrete. Concrete Specimens Using Neoprene dents larger than 0.010 in. (0.25 mm) in
Caps depth or 0.05 in.2 (32 mm2) in surface
8. CALCULATION area. Protrusions of any kind will not be
allowed.
Al. SCOPE
A4.2 Caps made from neoprene
8.1 Calculate the compressive strength of
meeting the requirements of ASTA D
the specimen by dividing the maximum A1.1 This method covers the procedure 2000. line call-out M2BC514A14B14.
load carried by the specimen during the for compressive strength testing of 6 in. The dimensions of the caps shall be 6 ⅛
test by the average cross-sectional area (152 mm) diameter by 12 in. (305 mm) in. (155.6 mm) in diameter and ½ in.
determined as described in Section 6 and concrete cylinders using neoprene caps (12.7 mm) thick.
express the result to the nearest 10 psi (69 with steel extrusion controllers.
kPa). A4.3 A compression testing machine
Provisions are made for alternate
meeting the requirements of the main test
8.2 If the specimen length-to-diameter reusable cap systems which utilize other
method.
ratio is less than 1.8, correct the result materials for pads and extrusion
obtained in Section 8.1 by multiplying controllers than neoprene and steel.
the appropriate correction factor show n A5. PRECAUTIONS
in the following table:
A2. REFERENCED DOCUMENTS
A5.1 Concrete cylinders tested with
LD: 1.75 1.50 1.25 1.00 neoprene caps rupture more intensely
Factor: 0.98 0.96 0.93 0.87 A2.1 AASHTO Standard:
than comparable cylinders tested with
T 231 Capping Cylindrical Concrete sulphur-mortar caps. As a safety
NOTE 8-These correction factors apply to Specimens
lightweight concrete weighing between 100 precaution, the cylinder testing machine
and 120 lb/ft3 (1,600 and 1,920 kg/m3) and to A2.2 ASTM Standard: should be equipped with a protective
normal weight concrete. They are applicable to D 2000 Rubber Products in cage.
concrete dry or soaked at the time of loading. Automotive Applications A5.2 The loading rate for some testing
Values not given in the table shall be
machines may Nave lo be adjusted when
determined by interpolation. The correction
factors are applicable for nominal concrete using neoprene caps.
A3. SIGNIFICANCE AND USE
strengths from 2000 to 6000 psi (13.8 to 41.4
MPa).
A6. TEST SPECIMENS
A3.1 Use of neoprene caps should he
considered as a suitable alternate for
9. REPORT compressive strength testing. Alternate The test specimens shall be that as
reusable cap systems must be verified in detailed in the Main test method,
accordance with paragraph All through modified as follows:
9.1 The report shall include the A12.
following: A6.1 Each end of the concrete cylinder
shall be plane within ⅛ in. (3.2 mm)
9.1.1 Identification number, across any diameter; i.e., there shall be no
A4. APPARATUS
9.1.2 Diameter (and length, if out side depressions in the concrete surfaces
the range of 1.81) to 2.2D), in inches or which are deeper than ⅛ in. (3.2 mm).
millimeters, A4.1 Two steel extrusion controllers Cylinders which do not meet this tolerant
9.1.3 Cross-sectional area, in square shall be used. An acceptable configura- shall not be tested unless the surface
tion for extrusion controllers is shown in irregularity is first corrected.
inches or square centimeters,
TABLE Al Tolerances A6.2 Neither end of compressive test
9.1.4 Maximum load, in pounds force
Controller I.D. -0” +1/32” specimens when tested shall depart from
or newton, 1
Controller wall - /16” +0” perpendicularity to the axis by more than
9.1.5 Compressive strength calculated Any tolerance not specified -1/64” +1/64” 2.0° (approximately equal to a difference
lo the nearest 10 psi or 69 kPa, in height of 3/16 inches for 2 6-inch
9.1.6 Type of fracture, if other than the diameter cylinders). Cylinders not
usual cone (see Figure 2), 2
meeting this tolerance shall not be tested
See “Concrete Strength in Structures.- by D. L. unless this irregularity is first corrected.
9.1.7 Defects in either specimen or Bloem, ACI Journal, March 1968, especially Table 3.
caps, and p. 185, for possible guidance as to the level of
reproducibility of concrete strength measurements
9.1.8 Age of specimen. that may he expected.

14
T 22 METHODS OF SAMPLING AND TESTING

FIGURE Al

NOTE-Metric equivalents for Figure Al are listed below.

Inches ½ 1 6 6 1/8 6 3/16 7¼ 8 12
mm 12.7 25.4 152.4 155.6 157.2 184.2 203.2 304.8

A7. PROCEDURE A8. CALCULATION A11. ALTERNATE REUSABLE

CAP SYSTEMS
The procedure followed shall be that as The compressive strength shall be
detailed in the Main test method, calculated as described in Section 8.1 of A11.1 Commercial systems are
modified as noted below: the main test method. available which utilize reusable caps
A7.1 Place an extrusion controller, manufactured from materials other than
containing a neoprene cap, on the top and neoprene and extrusion controllers of
A9. REPORT metals other than steel. An alternate
bottom surfaces of the concrete cylinder.
With the neoprene caps in contact with reusable cap system may be used
the concrete cylinder, carefully align the The report shall contain al] the items provided the following criteria,
axis of the specimen with the center of noted in Section 9 of the main test manufacturer. and agency (or purchaser)
thrust of the spherically seated block. method. verifications are satisfied.
Bring the bearing blocks of the machine
in contact with both of the extrusion A12. CRITERIA FOR
A10. PRECISION AND
controllers. ACCEPTANCE OF
ACCURACY
A7.2 No loose particles shall be ALTERNATE REUSABLE
trapped between the concrete cylinder CAP SYSTEMS
and the neoprene caps or between the A10.1 Testing variation associated with
bearing surfaces of the extrusion neoprene caps is no higher and possibly
controllers and the bearing blocks of the lower than that associated with sulfur- A12.1 The system must utilize reusable
test machine. mortar caps. Based on a study, caps retained within extrusion
coefficients of variation associated with controllers. Tolerances on all bearing
A7.3 The same surface of the neoprene surfaces of the extrusion controllers shall
neoprene caps were within the range (0.0
cap shall bear on the concrete cylinder meet Section A4.1.
to 4.0 percent) and considered to
for all tests performed with that cap. Each
represent excellent control. A12.2 Reusable caps shall provide a
neoprene cap shall not be used to test
more than 100 cylinders. The life of A10.2 Neoprene caps should be minimum of 0.5 in. (12.7 mm)
alternate pads must be verified by the considered as an acceptable substitute for compressible thickness over the entire
agency (or purchaser) in accordance with sulfur-mortar caps without correction for bearing area. The manufacturer shall
paragraph 12.4. apparent strength differences. supply reusable caps which are uniform
in dimensions and physical properties.

15
T 22 METHODS OF SAMPLING AND TESTING

A12.3 Manufacturer Verification of testing for capping method x batch will strength and variability for sets of field
Alternate Reusable Cap Systems: be randomized. One set of new pads shall cylinders manufactured from the lame
A12.3.1 The manufacturer shall be used for all tests within each strength samples of concrete; the reusable cap
provide results of a two-factor factorial level. At the option of the manufacturer, system shall be compared to T 231. The
experiment with capping method as the one set of new pads may be used for all paired t test shall be used at a level of
main factor and concrete batch as the tests at all strength levels; but sequence significance of 0.10.
secondary factor. For the estimation of of testing for capping method x batch x A12.4.2 A minimum of 30 sets of
variances for error, interaction among strength level cells shall be randomized. cylinders shall be compared for
factors, or for factors, the cap system A12.3.3 Specific statistical compressive strength for one evaluation
factor shall be considered a fixed variable equivalencies to be demonstrated by the with one set of reusable pads; however,
and the batch factor shall be considered a manufacturer shall be as follows: the number of sets may be increased (to
random variable. The experiment shall be A12.3.3.1 The analysis of variance of determine usable life) if the reusable caps
repeated at three levels of concrete each strength level shall show no have not developed visible damage such
strength at 28 days, representing a range significant effect of capping method. as splitting, gouging, or permanent
of average strength of 2000 psi to 6000 compression set in the bearing area in
psi (2000 psi, 4000 psi, and 6000 psi A12.3.3.2 The estimate of average thirty repetitions. The evaluation shall be
recommended). The F test level of difference in strength between capping repeated at least once with a new set of
significance for testing significance of methods utilizing the cell averages for reusable caps. Job control cylinders of
difference among variances shall be 0.05. each capping method X batch x level of two or more per set cast from the same
strength cell shall not be significantly sample may be used. If cylinders are cast
A12.3.2 In conducting the experiment. different from zero. The paired t test at a
a minimum of eight batches of four from successive batches in a laboratory,
level of significance of 0.10 shall be selection of cylinders shall be
cylinders per batch shall be prepared at used.
each level of strength, yielding two randomized among batch x capping
replicates for each capping method x NOTE-One reference describing factorial method.
batch combination. Within each batch, experiments is: "Fundamental Concepts in the A12.4.3 The agency (or purchaser)
equal numbers of cylinders shall be Design of Experiments" by Charles R. Hicks. shall reject a reusable cap system if the
randomly chosen for testing by the Holt, Rinehart, and Winston. in-house evaluation shows significant)
reusable cap system and by T 231. The difference in compressive strength o¡
sequence of testing for capping method x A12.4 Agency (or Purchaser) variability as compared to T 231.
batch x level of strength cells shall be Verification of Alternate Reusable Cap
Systems: A12.4.4 The agency (or purchaser)
randomized. If laboratory facilities do not shall reserve the right to reject a reusable
permit fabrication of all cylinders within A12.4.1 Prior to implementation of a cap system if the number of possible
one day, cylinders for one complete reusable cap system, the agency (or repetitions or life of a set of caps is not
strength level shall be fabricated on each purchaser) shall conduct an~ in-house acceptable to the agency (or purchaser).
of three successive days; sequence of evaluation comparing compressive

16