Testing the SSM-2044 filter IC

[cincin]

I’m working on a Korg PolySix synth that has at least one dead SSM-2044 (datasheet).

Going through the "Adjustment Procedure" from the service manual, I got to step 4 (2) 2) where I’m supposed to confirm a sawtooth wave coming out of each voices. However, of my 5 remaining working SSM-2044, 3 give me a sort of double peak sawtooth, and 2 only give me the top half of a sawtooth (sharks!). Pin 5 looks especially different on the group of 2 vs the group of 3.

The differences follow the chips when I swap them around from voice to voice, so I’m pretty sure that the SSM-2044 are the root cause.

All 5 of these chips still produce sound, but the group of 2 sounds noticeably different from the group of 3, so it’s not a good setup for a synth.

I’m assuming that the group of 2, which don’t produce as much of a sawtooth wave, are the “more defective” ones, and that the group of 3 are probably good. But that's an assumption, and I would like to find some way of being checking that the replacement chips I order will sound similar to my good ones. I'd also want to make sure that my group of 2 is actually defective and not just different. I wouldn’t resell defective ones, but if they are just different I will. (If those 2 have failed, they have failed in identical fashion, the half sawtooth is the same on both.)

The batch numbers don’t all match on all my chips, but that doesn’t seem to be a factor in how they sound.

Is there a way to test these chips for defects, or partial failures? (voltage drop specs across pins? test jigs?)
A good way to tell if they are actually defective, and to tell fully functional ones?
Is my assumption correct that the 2 half-sawtooth ones are defective (the PolySix is, in itself, a test jig after all)?

Thanks for any insights!

[donlisms]

I have no idea what you're doing, or anything about what you might need to make it work.  I am curious about... you say "of my 5 remaining working", which makes me think you have others that are not working.  Then of those 5, apparently... none of them work correctly, or at the very least, some work right and some don't, because they work differently.  My first thought was... "Your 5 remaining working SSM2044's do not work."

Then I looked for a datasheet, and at least got a general idea of what the thing is and does.  I've used quite a few SSM parts, when they were Burr Brown, and now Analog, but I had never heard of the SSM2044, because apparently that part didn't make it through the acquisition chain.  So it's a VCF, loads of fun, and they are apparently available from a modern second source as the SSI2144, for about $5 each.

Testing the chip, with all those special-purpose pins, seems like the kind of thing where the application circuit is the test circuit.  If it works in it's natural home, it works.  Otherwise, no.  I can't imagine a sort of resistance check or something that would establish the functionality of the chip, though stranger things have happened.  The thing is... at the pins, you only get to see the things connected to the pins, with maybe additional random influences upon those things, but there are plenty of bits in there that isolate the circuitry from the pins, so you can't really see what's up deep inside there.  Or so it seems to me.  And building a test harness would look an awful lot like building a synth module, so... why not use a synth module?

So I would assume, if I was getting funny waveforms with them in place, they are broken, though I would probably look at the input as well as the output to confirm that conclusion.  Or the chip-swapping thing does seem good, too.  If you have another stash of completely broken ones, you could put these partially broken ones in there with them, with labels that say "broken entirely" and "broken partially".

Then, after concluding they're all broken, I would save my money by not eating lunch for a couple of days, and try to buy some brand new ones that work right, and then move on to making music.  Or something like that.  So... testing in place, go/no-go result, and $get some new ones.  I suppose there is a supply of also-partially-broken ones on the auction site, or chips that are nothing like these that have been relabeled by people trying to make a living under difficult circumstances, but all that seems dicey and unwise if new real ones are available for $5 each.

[cincin]

Thanks, yeah I found another post from someone without a PolySix wanting to test those chips, and the suggested test jig was basically a mini synth... a bit involved. And I already have a synth to test with.

To clarify: I have 6 of those chips total, from the 6 voices in the PolySix. 1 is completely dead, 2 give me half sawtooth, and 3 give me a sawtooth with a hiccup.

I'm aware of the modern replacement, but I'm trying to restore this synth for eventual resell, and the vintage synth purists would probably take offence to the new chip being in there. I'd probably have to put 6 of the new ones in there, for it to sound "together". It may be my only viable option in the end, the vintage chips go for $40ea from reputable dealers, and it's hard to tell wether you'll get fully functional ones, as you pointed out.

I appreciate the guidance, thanks!

[donlisms]

Good luck.  This might be one of the times when things just don't go the way you'd like.