[sdiy] Sequential Conundrum (no, that's not a type of drum machine)

[greg montalbano]

There is a type of Zen koan which, upon repeated contemplation, will
drive one insane;  I'm beginning to think this Prophet 600 I've been 
wrestling with would be a contender...

The voice board has six individual voices, each containing two CEM 
oscillator chips and one CEM 3372 VCF/VCA chip.  There were two dead 3372s, 
which I pulled and replaced.  Now, the two voices with the new chips are 
responding to VCA control voltages, but NOT to the VCF controls.
OK, maybe it's a bad socket or bad filter circuitry, right?

--Wrong; if I switch one of the old working 3372s with the new ones, then 
the voice will work just fine;  and now the voice with the NEW 3372 is not 
responding to filter control.
Ah, you say:  it must be that the new chips are faulty.

-- Nope; I built a test circuit (from the 3372 data sheet), and found that 
the two new chips perform just as well as the four old ones ... in the test 
circuit; as soon as I put them back in the Prophet, it's back to filter 
non-function.

The only other factor that could be involved is the tuning routine -- upon 
power-up, before the voices become active, the Prophet runs a tune 
procedure that, if it encounters any oscillators or filters that it 
considers to be un-tunable, scales them such that they are disabled.
But WHY would it be having trouble tuning the new chips, if the new chips 
are  functional?

I have heard un-documented rumors over the years about the unreliability of 
later CEM chips;  does anybody here have any real info regarding the 3372s?

Can anyone suggest anything else I might be overlooking here?
At this point, I'll try just about anything....

Thanks.

~GMM
______________________________________
"Remember, class:  there are no stupid questions -- just stupid people."
                   Mr. Garrison