I have a tascam 32 so I know what you are talking about, how it overshoots and then finally reaches zero. The fostex's don't do that. what is happening is that when you hit rewind, it doesn't get enough "juice" to get the tape moving, until you lift the right tension arm, then it gets going and you can let go.
I am thinking it is either
-a mechanical adjustment of the tension arm springs (even though this is done via eyeball and eyeballing it it looks ok)
-tape tension (which all measures correct)
-count pulse duty (is as close to 50% as I can eyeball, tolerance is 45%-55% but on the scope I have no idea how one would even see that closely)
-fast wind speeds:
manual reads:
"Hold down the FF (RWD) button. Adjust R502 (REEL LOW) on the system control PCB so that wave length at test point 1 becomes 16, +0-1msec. "
OK the action when you hold down the FF or RWD button is identical to what is happening when you just try to rewind.
What I understand the manual to mean is that test point 1 should be 16 msec, or .16 seconds, so with my horizontal range switch set to .05 s/cm, the wavelength should measure 3.2 cm.
"Wind tape to the beginning and enter the FF mode. Adjust R503 (REEL HIGH) so that is is 0.78 ~ 0.82 msec."
I took this to mean .008 seconds, or with horizontal range switch set to .01 s/cm, .8 cm.
I made these adjustments to the best I could eyeball on the oscilloscope.
Does this look to be ballpark ?