3. Choosing tests and sizing test specimens, if necessary

Unlike mechanical properties, which are fairly predictable, electrical performance is more complex to tackle and, unless you imagine extravagant over-dimensioning, will have to be assessed taking into account the relative dimensions of the objects (as seen in the previous step).

It's therefore an appreciable paradox to have to know more or less the dimensioning of an object for which we're trying to define, through this study, what that dimensioning will be!

We are therefore faced with the following choice:

• Or the electrical problem is considered only as a "subsidiary issue", in which case the object (prototype, for example) will only be tested to guarantee these capacities after the mechanical studies have been carried out and the object produced, even if this means having to go back on either...