Is my VPE cursed?

[Spav]

Having had a perfectly functioning synth for just a few days it's now sick again. On Friday I went into the music room and turned kit on, I usually leave it for 15-20 minutes before playing. I went downstairs and made a quick lunch, but then the hospital called, my wife was out of her small op and I could go and collect her, so I got myself ready and left but forgot to turn off all the kit. The rest of the day was taken up looking after Lizzy, but just before bed, I remembered I had turned everything on in the music room earlier in the day, so I went to turn everything off.

As normal I cannot help myself but have a quick play when I get in there, to my horror the VPE was massively out of tune and the keyboard was not tracking well at all. I turned it off and thought I would look again the next day.

The problem was still there when I turned back on the following day. It had been left on for about 11 hours the previous day. This is England and it's summer so it rained for most of the time so the temp was only 15-16C outside and no heating on in the house so overheating shouldn't be a problem, all the other kit has stayed on many times for similar periods without problem.

I initially retuned all oscillators and their scaling, things didn't improve, but it did prove the oscillators are staying in tune and they are tracking the octave selection switch across the entire range 32' to 1'. I then played via MIDI and the weird scalling was still there. After checking the intervals i noticed the following, F2 to A#3 track okay. A#3 to B3 almost no change in pitch. B3 to C6 the tracking is almost exponetial in nature. One other strange feature, when pressing a note above B3 then below A#3 there is a 1-2 second delay before the Oscillator pitch changes (it steps after this time does not glide). All patches are fine Triggers, Filters, VCA's all working as they should be.

I have checked the CV coming from the digital board and the keyboard is generating linear 960mV per octave across the whole keyboard. However around the oscilator generator circuit I can see the problematic signal and I and guessing the oscillators are doing exactly what they are being told, the problem is somewhere between the CV input from the digital board and the CV input to the Oscillators.

I wasn't expecting to have to dig further into this machine for a while, but at least I will know just about every corner of this thing when I have finished!!!

[RL]

Hi Spav,
I remember you fixed a few bugs on the analog board.
If there are more problems after one or two weeks it's probably a destroyed CMOS chip. Maybe when you fixed it you forgot the static instructions. I have this problem often after a Memorymoog upgrade and sometimes I have to change a CMOS chip after a week or so.
Rudi

[Spav]

Strapped to ground and worked on an antistatic surface.

I may have another look later today, but we have the British Grand Prix and Murry in the Wimbeldon Finals today, so it's a beer packed day here

[thealien666]

IMHO, circuits and chips failing in cascade over time are a very strong indication that there's something wrong, and possibly intermittent, with the power-supply. Bad electrolytic caps in a switching PSU can cause strange and damaging behavior, that is sometimes difficult to catch. If you have access to an ESR meter, I would check if any of those caps in the PSU are showing higher than normal ESR values. (taking extra care of discharging them first of course).

As you may, or may not know, electrolytic caps can start failing due to high ESR, without showing any external signs of damage or even noticeable changes in their capacity value. And can cause other components around them (electrically) to fail prematurely. That's especially true with critical circuits like switching PSU. When they do start to bulge or leak, it's too late anyway. ESR tends to get higher with age, usually. But there are instances where bad manufacturing can cause premature failure even in relatively recent devices.

TIFWIW.

Al.

[Spav]

I am getting closer, I managed to get a couple of hours yesterday to take another look at this. I am suspecting something in the glide circuit is not right, I have not yet identified enough of the control signal points to decide what is going on in there yet. But there is good linear tracking going into the glide circuit and the problem is present on the output /input to the oscillator control circuits. The virtual earth of the glide contol amplifier is mis-behaving and the output of the glide control amplifier is switching to rail part way through the linear tracking.

On the up side I have a fantastic spreadsheet with all the control points, device connections and signals that I have identified to date