Peter and i solved the problem with the fss loop today, but here the long story:
The problem was that the loop was much more stable without the PC connected, so only the PZT of the laser was used as an actuator so far. Already some time ago i thought it might be the wrong sign for the PC and so i tried to change the sign of the loop by changing the two jumpers at the output going to the fast actuator and changing the sign of the error signal, but i never got it to lock in that configuration. So i though the other, previous configuration must be the right one, as it is locking. Later Tara had the same problems. The problem was that if the PC was connected the stability didn't increase and it even seemed to be more unstable using the PC in addition.
So in fact, the sign we used so far was the wrong one. But the problem was that changing the sign of the fast actuator and then flipping the error signal sign didn't work because the other (right) jumper settings don't work ! There is no feedback to the PZT of you change the two jumpers to the other orientation !! That's why it never worked with the right jumper positions.
As we figured out that this might be the problem i've build a BNC adapter changing the sign at the input of the laser and here we go, it's working now So something is wrong with this board . I will figure that out later what exactly is broken on that board.
will measure the ugf of the fss loop tomorrow...