This is interesting
Power up the E3646A
Channel 1 unreg message on screen and 0V output
Channel 2 works as intended
Connect Channel 1 and Channel 2 in parallel
Unplug Channel 1 and Channel 2
Channel 1 works as normal for a while, like 5 to 10 minutes
Channel 2 works as normal
Repeat from the Parallel connection
Can even power cycle to do a Calibration...
The driver for Q503 seems to have all theirs bits in order +12V1 and -12V1
My next step is to investigate the circuit driving Q109
I need a rest, I'll continue this tomorrow, if anyone of you have an idea, let me know... I am looking into Electrolitic caps used as a delay or something...