Roland SH-101 analog synth repair

[ricothetroll]

Hi !
I have this non working Roland SH-101 a friend of mine gave to me. It seems like there is a problem with the VCO : the noise generator is working, along with the VCF, envelope generator, etc... but no VCO sound.

The VCO is a CEM3340, whose datasheet is here :
http://curtiselectromusic.com/uploads/C ... 5_Long.pdf

Here is the service manual for the SH-101 :
http://privat.bahnhof.se/wb447909/dinsy ... SH-101.pdf

I checked pin 8 (ramp) and pin 4 (pulse) with an oscilloscope and I found out the chip is oscillating but very slowly (~1Hz). Then I started to look for a problem in the frequency control circuit. The frequency control in is pin 15 of the CEM3340. It appears to me that here multiple signals are summed : RANGE WIDTH, VCO internal TUNE, CV from Bender board (portamento), VCO user TUNE (cf schematic pg.7 of service manual).
Looking at the block diagram on pg.2, I see that the Keyboard control voltage that goes to the VCO first passes through the PORTAMENTO circuit. It first goes out from IC10 pin1 -> IC9B -> CVout/CVin connectors -> IC4B -> PORTAMENTO -> CEM3340 pin15.

I traced this signal : at pin 4 of IC10 I see ramp signal with an offset of about -7,5V. The frequency of that ramp signal is changes when I play notes on the keyboard. That signal disappears after IC9B.
Should I replace IC9B ? Or am I missing something ? I must admit I have no idea how this ramp signal, whose frequency seems to define the frequency of the VCO, is then converted into a CV...
Best regards.
Eric

[schmidtc]

Welcome to VSE, In general a signal can go into the current domain, so it's not necessarily bad if it disappears on your scope. Is this a fast frequency you're following? super-sonic? I haven't looked at the schematic, but I'd guess from your description that the circuitry around IC9B is going to convert a really fast AC signal with variable pulsewidth into DC (depending on the width) and sum it with the other DC signals. Is it really a ramp wave, or might there be some impedance issue or a slow scope? The CEM most definitely takes DC in control voltages, so at that stage any AC that isn't a modulation source should disappear. What's the resistance from CEM pin 15 to ground when a key is pressed, what's the DC voltage? 3340's go bad all the time, but SH-101's are usually pretty solid. Good luck!