A round robin study using the CSA O112.9 standard was conducted with six participating test laboratories, three in Canada and three in the U.S., with the support of the CSA Wood Adhesive Sub-Committee. The tests included block shear (dry, vacuum-pressure, and boil-dry-freeze), delamination, and condition B1 creep. Test specimens were prepared using Douglas-fir as substrate and four adhesives (phenol-resorcinol formaldehyde, melamine-urea formaldehyde with 80% melamine, melamine-urea formaldehyde with 40% melamine, and catalyzed polyvinyl acetate).
Precision and bias statements (repeatability and reproducibility limits) were developed for the block shear test (strength and wood failure) and delamination test for O112.9. The statistical procedures of ASTM E 691 were found to be inadequate for the block shear test, so appropriate statistical procedures were identified and applied. However, a refinement of the E691 method was used to analyze the delamination test. The creep data was not analyzed because it showed huge variability across the laboratories. The rest of the data will be analyzed in the future.
In the block shear test, the repeatability estimates obtained from the round robin test tend to overestimate those of O112.9, that is, the outcomes from using the O112.9 test procedures would actually vary less than those indicated by the limits derived from the study.