Prediction of compressive strength by modified splitting tests: a case of concrete cylinders. (General Notes).
In the standard splitting test (ASTM Ch96-90), the load is applied over the width of the packing strip, which is standardized as 15 mm. Using plates made of steel and having a radius matching that of the cylinder, this bearing width may be increased to any value up to the cylinder diameter. Figure 1 shows the experimental setup. The compressive failure stress for the different concrete grades was then plotted against the ratios of cylinder diameter to the increased bearing widths. Figure 2 shows the plot. Comments are later made about the predicted compressive strengths from the modified splitting tests as shown in Figure 2 and the conventional compressive strengths tests.
Five different plate sizes with different bearing widths were used in this work. The bearing widths were 15, 20, 40, 60 and 80 mm. The plates were extended to the full length of the concrete cylinders. Each concrete cylinder was 100 mm in diameter by 200 mm in length. For every concrete grade and bearing width studied three tests were conducted. The concrete grades studied were namely: 45, 65, 90 and 100 Mpa (megapascals). Specimens for all tests were stored in water for 28 days prior to being tested.
The compressive load was applied at a standard rate of 12 Mpa per minute (30 ksi/sec) over the bearing width and along the length of the cylinder. The compressive failure load was recorded for each case.
It was observed from the experiments that for lower grades of concrete, failure was achieved by a sudden silent drop in the load. For the higher grades of concrete, it was also sudden but violent.
The results from the modified splitting tests are shown in Figure 2. The stress values were obtained by dividing the failure compressive load by the bearing area for each test. The plot shows a non-linear relationship between compressive failure stress and diameter/bearing width. The generalized equation obtained from the data is given by:
C = [K.sub.l][f.sub.r] + [K.sub.2]/(d/w)
C = Compressive strength (28 days)
[K.sub.l] = Restraint factor ([approximately equal to] 0.90)
[f.sub.r] = Standard splitting strength based on d/w ratio of 7
[K.sub.2] = Strength constant obtained from the best- fit curve
d/w = Diameter/Bearing Width
Conventional 28-day compressive strength tests conducted on concrete cylinders respectively yielded 35, 62, 84 and 92 Mpa for 45, 65, 90 and 100 Mpa concretes. The modified splitting tests using plates yielded 43, 62, 85 and 95 Mpa. The 7-day and 14-day average compressive strength using conventional compressive tests were also recorded. The results were respectively 28, 50, 67, 74 and 32, 56, 76, 83 Mpa.
DISCUSSION AND COMPARISON OF EXPERIMENTAL RESULTS
It is noticeable from Figure 2 that when the d/w [less than or equal to] 1, the compressive stresses are disproportionately large for all grades of concrete studied. This is because the increased end restraint induced by the interaction of the concrete with the steel during loading increases the compressive load and in turn reduces Poisson's ratio. Furthermore, when d/w = 2 = vertical height/bearing width (ratio for the conventional compression tests for cylinders), the compressive stress values from the modified splitting tests in Figure 2 are approximately 43, 62, 85 and 95 Mpa. By comparison, the values determined from the conventional 28 days compressive tests are 35, 62, 84 and 92 Mpa. The values obtained from the modified splitting tests are equal to or higher than the corresponding values from the standard compression tests. This difference could be accounted in the shape of the bearing area. In the standard compression tests, the specimen is a symmetrical column, whereas in the modified splitting test s, it is a rectangular block of concrete under compression.
The use of plates for the modified splitting tests to study the compressive strength of concrete cylinders leads to rectangular hyperbolic relationships between the diameter/width ratio and the modified splitting strength (compressive strength), for a wide range of concrete grades. Compressive strength values derived from these relationships correlate well with, but are generally in excess of those determined from conventional compression tests. This shows that a compressive strength value may be derived from any standard cylinder splitting test simply by dividing the failure load by a figure that depends on the cylinder size alone.
Adopting a different viewpoint concerning a particular test (the new test method) can be an effective method of obtaining a simple analysis by making comparison with an already familiar method of testing.
[FIGURE 2 OMITTED]
MF at: M-Faruqi@tamuk.edu
ASTM. 1993. Annual Book of Standards, Vol. 4.02, Concrete and Aggregates. ASTM, Philadelphia, Pennsylvania, 845 pp.
|Printer friendly Cite/link Email Feedback|
|Author:||Faruqi, M.; Sai, J.|
|Publication:||The Texas Journal of Science|
|Date:||Feb 1, 2003|
|Previous Article:||The goldstripe darter (Etheostoma parvipinne) and its tolerance to low pH in an east Texas pond. (General Notes).|
|Next Article:||Characterization of the Alamo Creek Basalt, Big Bend National Park, Texas.|