Testing of material is essential to gain practical knowledge of how materials react under various situations. The ultimate goal of any test is to enable the making of decisions that provide the best economic results. In practice, two general methods of testing are used.

DIRECT TESTING:

The only test that supplies absolute information about a workpiece or a material is a test of the particular property of interest conducted on that part itself. In this method of direct testing, an attempt is made to use the materials under the exact conditions of practical use, and the test may be concerned with a product, a process or both. Direct testing is usually time consuming and for the results to have statistical significance, often requires compilation of data from many test samples. The procedure is necessary, however, for those cases in which simpler methods are not available and in which sufficient historical information has not been accumulated to permit correlation between the attribute about which information is desired and some other measurable factor.

INDIRECT TESTING:

Indirect testing involves the use of such a correlation, such that accurate knowledge of the relationship between the two factors must exist. The ability of grinding wheels to resist the centrifugal forces imposed in use is directly tested by rotating them at higher speeds than those of actual use. Such a test indicates that the wheel strength is sufficient for normal use with some safety margin.

An indirect test that is sometimes used for the same purpose can be performed by rapping a suspended wheel to cause mechanical vibrations in the sonic range. A clear tone indicates no cracks.

A danger of indirect testing is that the conclusions depend on the assumption that the correlation between the measured factor and the critical factor exists under all conditions. The rapping test for grinding wheels does not give any real indication of strength unless knowledge of the wheel's history permits the assumption that with no cracks it has sufficient strength for use.

DESTRUCTIVE TESTING:

A large number of direct tests are destructive. These are also dangerous because the assumption must be made that those materials not tested are like the ones for which test information has been obtained. A portion of weld bead may be examined for quality by sectioning it to look for voids, inclusions, penetration, bond and metallurgical structure by visual examination. By this operation, this portion of the bead has been destroyed, regardless of the quality that was found, the only knowledge acquired about the remaining portion of the weld comes from an assumption that it is similar to that examined because it was made under the same conditions.