ABSTRACT
This chapter discusses concrete strength variability due to variations of cement from a single source.
It is well understood that there are significant differences in strength of Type I cements from different sources. How variable is the strength of different cement shipments from the same source? One of the most exhaustive studies to address this question was conducted by Walker and Bloem (1958) at the NRMCA/ NSGA (National Sand and Gravel Association) Joint Research Laboratory. Cement samples were secured from each of five sources every 2 weeks from October 1955 to October 1956 and stored in sealed containers. A sixth cement, consisting of a blend of five cement brands from the Washington area, was thoroughly mixed at the start of the program and stored in sealed containers for use as a control. Three principal series of tests were conducted at different times. The first series involved standard mortar strength tests (ASTM C109) on all samples; the second, concrete tests on selected samples; and the third, concrete and mortar tests on selected samples. In the first series, five mortar batches were made on different days with each cement sample, including the control; each round, including all sources, was made on the same day. All work was performed by the same operator. Nine cubes were molded from each batch for strength tests in triplicate at 3, 7, and 28 days of age. So, for a given cement sample at each age, 15 mortar cubes were broken (3 from each of 5 batches mixed on different days). The series involved testing of approximately 7000 2 in. (50 mm) mortar cubes. In the second series, concrete batches with a cement factor of 517 lb/ yd3 (307 kg/ m3) and mixed to a constant slump of 3 to 5 in. (75 to 100 mm) were made using samples of cement that had produced the highest, median, and lowest mortar strengths from each of the five cement sources in Series 1. For each cement sample, three concrete batches were made on different days. Concrete cylinders of size 4 × 8 in. (100 × 200 mm) were tested in triplicate at 3, 7, and 28 days. Only Series 1 and 2 are discussed below. Series 3 findings were in line with Series 1 and 2.
