Prophet 2000 update!

So, my main keyboard controller / sample-based instrument, the mighty Prophet 2000 from Sequential Circuits (happy birthday, Dave!) has sadly not been functioning properly since late last year. It appears to be suffering from an electronics issue of a kind that could only surface in a weirdo digital / analog hybrid wunderbox of this vintage (circa 1985). The good news is that the underlying sound generation circuitry is intact and in good working order: that's not where the problem is. Not only that, but it is still currently plugged in, turned on and still capable of selecting among presets and all parameter settings. That means the CPU still works as well.

But it won't make any sound.

Boo! But why? Well, (adjusts propeller hat and weighty collection of pocket pens), fairly early on I surmised it must have something to do with the concurrent lack of response from the Parameter Value knob. You know, the one adjustable universal parameter value knob that was supposed to make programming synthesizers easier by simplifying the interface; the one famously introduced on the Yamaha DX-7 in 1983 and much derided by many actual synthesists since... Yes, that one! Well, it stopped making parameter value changes, but bafflingly (and this really threw me at first) the Parameter Value *buttons*, both increment and decrement, were still 100% operational! You can only imagine my confusion.

Not only this, it turned out the circuitry associated with the knob was also operational, and the knob control itself was fine. Nevertheless it wasn't sending a changing train of pulse widths to the CPU according to the instantaneous knob position per the design spec. This involved long hours staring at the schematic and carefully poking around the boards with a voltmeter and a homemade logic probe. What I eventually discovered was that certain control signals from the CPU were not reaching their intended destinations due to corrupted logic circuitry further downstream. Repairing the issue will involved carefully tearing the bad chips out of there and replacing them with new ones. Fortunately the old Proph is mostly standard CMOS and TTL logic with only a few custom chips; none of those are thankfully at fault here. Once the chips are finally replaced the remaining front panel controls multiplexed on the same lines will again become operational and the machine will begin to work properly again.

So why exactly doesn't it make any noise if it's plugged in, turned on and the sound generation circuitry is intact? Essentially it's because the volume control which, like the parameter value knob, is not an ordinary volume pot but a voltage divider sampled by an ADC, is stuck at "0" and can't be moved. That's why I said this was a problem somewhat unique to 1985. The balance control is stuck permanently at 100% left map also, which means half the sounds can't be heard even if (as I eventually did) you change the volume back to some value other than "0" via MIDI. Unfortunately, and likely due to space constraints in ROM, the Prophet 2000's left map / right map balance can't be changed via MIDI. I know because I tried! It was a good idea though...

I'll keep everyone informed of course.