>> In short, it’s a moral duty to bring it back to life.
Yes it is.
>> 1) start with power supply, replace old electrolytics cap. Are there any other parts that you recommend to change?
Because my unit was in the same situation, and because I couldn't find drop-in replacements, I removed and reformed the big electrolytics over a 4 day period and got them down to a leakage rate of 8 uA, better than many new ones. I wouldn't just fire them up after sitting so long, I'd reform the layer. I'd replace all the other electrolytics (use high ripple spec on the rectifiers and low impedance spec on the decouplers), the power transistors (the 4 big ones and the 6 little ones) and the two 4558 op amps.
2) go to the five boards (KAS, SH, TKC and the two TSB) under the keyboard:
- replace all the electrolytics Yes Go to 25 volt units instead of the 16 volt units installed, especially on the OE1, OE2, and PRA boards in the rack
- replace all the CMOS (maybe installing new sockets) Yes, using the expensive sockets with built-in decoupling caps
- add decopuling cap for the IC (based on your experience is it better to buy “built-in sockets with caps” or solder the caps separately?) As above, but the built-in socket type are for the 4000-series chips only.
Given that I am still defining the plan, do you also recommend these operations (or just some of them) for these five boards?
- replace old op amp Yes, for those items not in the audio signal chain, and any 4558's that are in the audio signal chain.
- replace old bjt Do not do this. You risk losing the synth's character
- replace all the ceramic disc capacitor (with dipped ceramic) No need to do this unless they are physically damaged or have corrosion on the leads.
- replace all the "famous" tantalum capacitor (with polystyrene or polypropylene or aluminum electrolytics, same value obviously) There's only one or two in the synth, but sure, replace those too. Use a heat sink on the leads when doing so to prevent loss of capacitance, tant's don't like heat.
- replace all the mylar capacitor with polypropylene or polystyrene Absolutely not, you risk losing character bit time, especially on the filters. Only replace if you have a bad one, and then try to find an old good one. Take my advice from experience.
- replace the few old trimmers That's a lot of work and expense, I'd only do it as required, not as a course of practice.
Going to the boards behind the keyboard..
Regarding all M Boards:
- replace all the electrolytics Yes. Use audio grade on the filters and in the signal path, low impedance for the large decouplers
- replace all the old trimmers too? As above, only as required.
Regarding the two KBC Boards:
- replace all the electrolytics Yes.
- replace all the CMOS (maybe installing new sockets and decopuling cap for the IC (like the boards under the keys) Yes to both. There is some decoupling on here already, but the additional ones at the chip can't hurt.
Of note:
Be ready to find some dead ig00151 amplifiers on the M boards if you're not getting a signals. Mine (which was a basket case, admittedly) needed (45) ig00151 chips in the end. Yes, that's 45 chips, but that's not normal. Still, I had those along with one dead oscillator, one dead wsc, two dead filters, and one dead filter envelope chip. The only one on the M boards that didn't have a dead one was the amplifier envelope chips. You have to mentally either commit to owning a truly restored vintage CS80 including all the time and expense that entails, or give it to someone else that will. I made the decision to commit and have not regretted it once, quite the contrary, it's a stunningly beautiful instrument to play when hitting on all cylinders.
Also, be aware that the 151's can 'sort of' work, that is, they pass a signal, but are not controlling the envelopes properly or you can't get a full output level as you should. When everything is working properly, you should be able to calibrate the M boards per the manual. If you can 'almost' get a setting on a trimmer but run out of travel, something's wrong (and it could be the trimmer, have seen that too, there's a couple of failure modes for those).
There's a lot of errors in the manual. A major one of note I just spent a lot of time with is the VCA levels for the last two output chips on the M boards. The adjustment procedure calls out 400 mV and 80 mV respectively. That cannot possible be correct, you will have lots of 50/60 cycle hum in the output and a lot of noise in the OE effect section.
I tried several different combinations and settled on 900 mV and 250 mV pressing C4 with an 8 feet setting and a sine wave input and am happy with it. If you want a bit more bite and sizzle, the 900 mV can be increased a bit, maybe to 1200 mV. Over 1400 you will start getting some distortion on certain intense patches. I like a more mellow sound, and found the 900 mV setting to be a good compromise between output level, hum, and character.
I added led's on the M boards using the key trigger signal along with a regular diode and resistor. That mod was invaluable. I also added little hard point loops on the six main test points on the boards for grabbing them with a spring-loaded scope hook probe. WELL worth the effort.
Best of luck,
Scott
synthRodriguez
