There are more hydrogen embrittlement testing methods as follows.

(5) Delayed failure test using the actual part

In order to obtain the final confirmation on whether the plated part is likely to break due to the hydrogen occluded during the plating processes, or the baking process' effectiveness of hydrogen embrittlement removal, a delayed failure test using the actual part must be performed.
This is also necessary for failure analysis when they occur, as well as when there are changes in part designs and intended applications for the part.
Although the problems regarding failures are complex and the delayed failure tests may not always reveal all problems, confirmation tests are still needed for problem analysis.
A good measure in devising the delayed failure tests using the actual part is to consider the individual application specifics and design an appropriate scheme. Failures due to hydrogen embrittlement occur suddenly by the presence of occluded hydrogen, and at a location of notched feature where tensile stresses concentrate. It is best to replicate this condition.
An ideal method is the one adopted by the aircraft industry. If the part has a notched feature, a static load equal to 70% (90% if not notched) of the part's ultimate tensile strength is applied. If the part does not fail this delayed failure test within 200 hours, the part is deemed as passed.

An evaluation method using Delta Gage is introduced below. A plated test specimen is placed in a vise of a hydrogen embrittlement test system, and the vise is closed and the test specimen is flexed to 90~95% of the distance where the test specimen actually failed. The condition is held static and the time it took for the test specimen to fail was measured. There is a clear correlation between the hydrogen embrittlement rate measured by the Delta Gage and failures as results are shown in [Table 1]

[Table 1] Delayed failure test results