A circuit potted in SMT components will barely feel it.

Pablo, I want to underline this statement. If you use leaded components the following can result:

http://www.reliantlabs.com/art_HaltC...ilure.html

But also certain SMD quartzes, especially the lower clocking ones, can be sensitive to g-forces:

http://www.euroquartz.co.uk/pdf/cx-3h-sm.pdf

Here, maximum g-forces of 10g survival (vibration) and 1500g (peak) are specified.

Beside these destructional effects of high g-forces, change of frequency is only very small, and for most purposes totally necglectable. As you can find at:

http://www.ieee-uffc.org/freqcontr...gaccel.htm

df/f is about 10**-10/g to 10**-9/g. So, at 121g less than 1ppm frequency change is to be expected.

I'm aware of the fact, that in the above publishings g-forces are meant to be applied in vibrational mode, which is different from constant rotational acceleration. But due to the fact, that a loaded generator will show fluctuations of rotational speed, also vibrational modes must be taken into account.

Kai