Page 159 - Handbook of Adhesives and Sealants

Standard Test Methods 129

standards are presented in Table 4.1 for adhesives and in Table 4.2
for sealants. A more detailed listing of test specifications and stan-
dards may be found in Appendix C. The properties reported by sup-
pliers of adhesives and sealants often reference ASTM standards.
A word of caution should be noted here. Standard test data are suf-
ficient to compare strengths of various bonding systems. They can be
used to compare relative effectiveness of different adhesives, surface
treatments, curing schedules, and so forth. They can also be used to
separate and quantitatively define the many variables that ultimately
determine the performance of a joint. However, standard test results
cannot be readily translated into specific strength values for an actual
production joint. Actual joints generally have a complex geometry that
is significantly different from the standard test specimen geometry.
Under specific operating environments and depending on the kinds,
frequency, and severity of the stress that the joint actually sees in
service, an adhesive or sealant may perform spectacularly better or
worse than what is represented by ASTM tests on a supplier’s data
sheet.
The most noticeable differences between standard test results and
the results from an actual joint in service are due to several factors.

1. Joint design is seldom the same.
2. The mode and application of stress loading are usually different
and more complex in practice.
3. Environmental aging is usually less severe in service (laboratory
tests tend to ‘‘accelerate’’ aging so that the testing can be completed
in a reasonable time). However, the effects of the actual service
environment are generally more complex. For example, when in
service, the joint may be simultaneously exposed to cyclic stress,
cyclic temperature, and humid environments.
4. Laboratory test specimens are usually made in ‘‘controlled’’ envi-
ronments. This control pertains to the equipment, the cleanliness
(less weak boundary layer opportunity in the lab), and the person-
nel (training, care, and awareness).
5. The sample population is limited with actual production parts be-
cause of expense. Even with laboratory specimens, a full design-
of-experiment statistical process is difficult to achieve because of
the many production variables that can affect the joint strength.

Thus, the most reliable test is to measure the strength of an actual
assembly under actual operating conditions. Unfortunately, such tests
are often expensive or impractical. The next best method is to measure
the strength of an actual assembly under simulated operating condi-