Nicholas J. Carino, Engineering Consultant
The relevant standards related to compressive strength testing are AASHTO T 23 (ASTM C 31) for making and curing specimens, AASHTO T 231 (ASTM C 617) for capping, and AASHTO T 22 (ASTM C 39) for compressive strength testing. This article discusses some recent changes to these standards to reduce interlaboratory discrepancies, especially with high strength concrete.
Background
In the 1980s and early 1990s, field problems were reported where inconsistent results were being obtained between different laboratories testing the same samples of high strength concrete. In response to these problems, ASTM Subcommittee 09.61 on Strength Testing established an ad-hoc task group on high strength concrete to study whether the existing standards could be improved to reduce the likelihood of these inconsistencies. The task group studied recent research and the standards related to specimen fabrication, curing, capping, and compressive strength testing. As a result, some significant changes to ASTM C 31, C 617, and C 39 were adopted by ASTM. Some of these changes are being incorporated into the corresponding AASHTO standards.
Specimen Size and Fabrication
AASHTO T 23-04 and ASTM C 31-03a define the standard cylinder size as 6×12 in. (150×300 mm). However, AASHTO T 23 states that specimens “may be” 4×8 in. (100×200 mm), whereas ASTM C 31 states that 4×8-in. (100×200-mm) cylinders are permitted “when specified.”
The benefits of using 4×8-in. (100×200-mm) cylinders include ease of handling, simpler field curing facilities, and less wear and tear on testing machines. On the other hand, there has been concern over the slightly higher strengths obtained with the smaller size cylinders. A joint NIST/NRMCA study(1) showed that the differences in strength between the two specimen sizes could be reduced by fabricating the smaller cylinders using two layers of concrete instead of three. The task group recommended changing ASTM C 31 to require two layers for fabricating the smaller cylinders. The task group also recommended removing the restriction on using vibration to consolidate concrete with a slump greater than 3 in. (75 mm). There was no evidence that excessive segregation occurs by vibrating high-slump concretes provided limits are placed on the duration of vibration. These recommendations were adopted.
Initial Curing
Another significant modification to ASTM C 31 was to reduce the permitted range of the initial storage temperature of specimens from 60 to 80°F (16 to 27°C) to 68 to 78°F (20 to 26°C) for concrete with specified strengths of 6000 psi (40 MPa) or greater.
Loading Rate
Loading in compression tests is done at a constant rate of movement of the platen relative to the crosshead. The deformation rate should result in a loading rate within specified limits. Before 2004, the loading rate range in ASTM C 39 was 20 to 50 psi/s (0.14 to 0.34 MPa/s). A joint NIST/NRMCA/FHWA study(2) showed that the upper limit of the range resulted in 2.2 percent higher strength than testing at the lower limit. Thus, the task group recommended reducing the acceptable tolerance to one-half of the previous value. The 2004 version of ASTM C 39 requires a loading rate of 35 ±7 psi/s (0.25 ±0.05 MPa/s). In addition, the language related to screw-type machines was deleted and a new note was added, which states that for screw-driven or displacement-controlled testing machines, the operator needs to establish the appropriate rate of movement to achieve the required loading rate. The appropriate rate of movement depends on the size of the test specimen, elastic modulus of the concrete, stiffness of the testing machine, and whether unbonded caps are used.
ASTM C 39-04 also addresses machines with break detectors that stop loading when a prescribed drop in load occurs. To ensure that small momentary drops in load do not stop a test, a drop in the load of at least 5 percent has to occur before the break detector stops the test.
Bonded Caps
An NRMCA study(3) showed that sulfur caps can be used to test high strength concrete without an adverse effect if cap thickness is limited and sufficient aging of the sulfur is allowed before testing. A follow-up study(4) showed that the unconfined cube compressive strength of the capping compound may not be as important as its elastic modulus. The best capping compound is one that has a high elastic modulus.
As a result of these findings, changes were made to ASTM C 617 on the use of bonded caps (see HPC Bridge Views Issue No. 16, p. 3). The cap thickness for cylinders made with concrete stronger than 7000 psi (50 MPa) is now limited to a maximum of 3/16 in. (5 mm) and an average of 1/8 in. (3 mm). To ensure that greater attention is paid to controlling cap thickness, the laboratory is required to check cap thickness on at least three specimens during the day’s testing operations.
The traditional requirement of specifying capping material with a strength at least that of the concrete to be tested was dropped for concrete strengths greater than 7000 psi (50 MPa) and replaced by a performance requirement. A capping material is acceptable if it results in an average cylinder strength that is at least 98 percent of the average strength of cylinders capped with neat cement paste or ground flat. Also, the standard deviation of the capped cylinders has to be less than 1.57 times the standard deviation of the reference cylinders.
Troubleshooting Low Strength Results
High strength concrete is inherently more sensitive to details of specimen fabrication, curing, end preparation, and testing procedure. According to ACI 363.2R,(5) experience has shown that special care is needed for concretes stronger than 8000 psi (55 MPa). Standards-writing committees have attempted to improve testing requirements to reduce discrepancies between laboratories testing the same concrete. When unexpected low strength test results occur, the following details should be investigated:
- Was the cylinder properly fabricated and cured? Weighing each cylinder provides a check of gross errors in specimen fabrication. Check field records of initial curing conditions.
- If bonded caps were used, was the material qualified for the concrete strength? Were the ends of the cylinders sufficiently flat and perpendicular to preclude excessive cap thickness?
- For unbonded pad caps, were the ends of the cylinders sufficiently flat and perpendicular to the cylinder axis? Are the pads of the correct hardness and in acceptable condition?
- Are the loading surfaces of the testing machine plane? Do the dimensions of the spherically seated head satisfy requirements?
- Is the spherically seated head properly lubricated so that it rotates freely upon contact with the cylinder but behaves is a fixed head during loading? Do not use grease as a lubricant for the spherical head.
- Is the loading rate within the requirements? A slower loading rate may produce a lower strength.
- Was the cylinder loaded to its ultimate capacity?
Reference
- Carino, N. J., Mullings, G. M., and Guthrie, W. F., “Evaluation of ASTM Standard Consolidation Requirements for Preparing High-Strength Concrete Cylinders,” High Performance Concrete: Design and Materials and Recent Advances in Concrete Technology, Publication No. SP-172, American Concrete Institute, Farmington Hills, MI, 1997, pp. 733-768.
- Carino, N. J., Guthrie, W. F., Lagergren, E. S., and Mullings, G. M., “Effects of Testing Variables on the Strength of High-Strength (90 MPa) Concrete Cylinders,” High Performance Concrete, Publication No. SP-149, American Concrete Institute, Farmington Hills, MI, 1994, pp. 589-632.
- Lobo, C. L., Mullings, G. M., and Gaynor, R. D., “Effect of Capping Materials and Procedures on the Measured Compressive Strength of High-Strength Concrete,” Cement, Concrete, and Aggregates, Vol. 16, No. 2, December 1994, pp. 173-180.
- Vichit-Vadakan, W., Carino, N. J., and Mullings, G. M., “Effect of Elastic Modulus of Capping Material on Measured Strength of High-Strength Concrete Cylinders,” Cement, Concrete, and Aggregates, Vol. 20, No. 2, December 1998, pp. 227-234.
- “Guide to Quality Control and Testing of High-Strength Concrete,” ACI 363.2R-98, American Concrete Institute, Farmington Hills, MI, 1998, 18 pp.
Editor’s Note
This article is the third in a series that describes tests for use with HPC. Previous articles appeared in Issue Nos. 36 and 37. For additional articles on testing high strength concrete, the reader is referred to Issue Nos. 6, 14, 15, and 16.