For what it's worth I too have noticed this in several different manufacturer's actions. As Manitou described though, it's only at the extreme ppp levels.
I was surprised to read your post about repetition springs effecting touch-weight. Of course as Del mentioned, the spring doesn't compress until let-off anyway, which is way too late to effect the touch. As for the discrepency you mentioned between let-off and drop... again as Del mentioned, the two should be happening simultaneously (This is double escapement). Meaning of course the repetition lever should be contacting the bottom of the drop screw precisely when the jack tender is contacting the let-off button.
We are currently working on reconditioning an action in a Steinway A rebuild we're doing, and are becoming very interested in the finer points of action geometry and design.
We have figured out 3 different ways to measure overall action geometry, but are trying to find a quick and accurate way to do it. Do you have any suggestions?
Simply measuring key travel and hammer travel doesn't seem to be the most accurate way to do it, and measuring each lever from key to hammer seems too time consuming. Do you in your experience have a relatively quick and accurate way to measure the overall ratio of key travel to hammer travel?
Thanks in advance for any input.