Memorymoog suddenly is corrupted (?)

[LaxSlash1993]

The (?) is because I don't know if that's the right term.

Well, the Memorymoog has been off for a few days, because I was busy with doing something with guitar. It was working pretty good. Well, today I turned it on. First, nothing came out. I turned the program volume knob, and suddenly it had this huge noise sound, with some oscillation on patch 1. Now, keep in mind that the only patches on it were the factory ones. I went to patch 2, the brass one, and it had filter sweeps. First thing I thought was that the battery went. So, before I opened it up, I created the tuning patch in the manual, and saved it to patch 1. I turned it off, turned it back on, and the patch was still there. So, I don't think that it was the battery now (correct me if I'm wrong). I also started fooling around a little bit with patch 2. I noticed a few things - one thing was that in the LFO section, for every patch, both the sawtooth and S&H waveforms were selected. That's impossible, isn't it? And also, when I went to the KB mode for the number of voices to use in unison, it said "0 VOICES." And when I turned the Glide knob, the value that is supposed to be the value of the patch kept changing if I turned the knob. But, only on patch 2. After that, I played a C scale on the lowest part of the keyboard using the tuning patch. The scale was way off, but just for that octave. If I moved up higher on the keyboard, or changed the octave settings, it was fine. When I pressed Auto tune, it said it tuned all six voices, and it fixed that problem, but that's weird, since I thought auto tune doesn't fix the scale. Again, correct me if I'm wrong. A few other things that I noticed was that the only patches effected were patches up to 40, all the patches had no waveforms set for OSC 3, pretty much all of them had the same LFO issue, all the program volume levels were set to a very low level, and the sequences seem to be fine and intact.

So my questions are what would have caused this and is there any thing that I did that may have done this?

By the way, there weren't any power surges that I'm aware of.

[Technician Larry]

Certainly sounds like corrupt data. Low battery voltage could be the explanation even though you successfully wrote and saved a tuning patch. With the power on, the load on the battery is removed and the battery voltage will recover slightly - long enough to retain RAM data for some period of time. The MOD LFO is supposed to enable only one waveform at a time but that is likely a function of the waveform select routine. Once data is corrupted there is no check to verify this so I suspect this is also a result of the data corruption. Number of voices in unison mode error - same explanation. Patch 2 glide mode issue is confusing, though detail in your explanation may not have been stated clearly and consequently misunderstood. I suspect this is probably another symptom of data corruption. Autotune does correct for scaling errors to some extent. Each VCO has an autotune voltage coming from DEMUX. The value of that voltage changes depending on the octave range of the note played.

Corrupt data on an Oberheim OB8 can cause all kinds of strange failure modes that don't go away when data is reloaded. I pull the RAMS and purge the data by inserting the pins in aluminum foil. Then data will load and return the machine to normal operation. I've never seen this on a Memorymoog though and would expect your machine to behave normally with a data load. Short of zapping with a nasty static discharge or connecting to an amp or mixer with a hot chassis (likely resulting in other damage), I doubt that this was caused by anything that you did. First guess would be the battery, second guess would be a RAM with an intermittent problem. One of the most stubborn, therefore memorable problems I ever encountered was an OBXa with an intermittent scratch pad RAM that was mechanically sensitive. The mechanically sensitive part threw off track as it is rare for an IC failure mode to be mechanically sensitive. Frequently thermally sensitive, yes, but when the problem responds to flexing and tapping, I'm looking for something other than a bad IC. I will always remember that one.



Pool of Memorymoogs (and a Quadra) in the shop

[MC]

I've seen that LFO problem and your memory got corrupted. Either the CPU crashed and wrote over patch memory, or your battery is going.

FWIW I just replaced my battery last month, they only last 20-25 years.

The last new Memorymoog left the factory in 1985 - 26 years ago.

[LaxSlash1993]

Yeah, I thought that the glide issue would be hard to understand, but it really is that weird. Basically, if I turn the knob, it will switch between a few different values. And if I don't, it stays. As far as doing the thing with the RAMS and replacing the battery, I'm really hesitant about opening up the Memorymoog without a service tech.

I had it serviced a few months ago, and the battery was replaced I think. Maybe it just got a bad battery.

Nice picture, BTW.

[Technician Larry]

I only mentioned the RAM purge procedure as a relative point of interest, and to repeat or clarify, I only have ever had to do this on OB8s. The previous post mentioned that a CPU glitch might have written over RAM data. Maybe. Hope for low battery voltage!

[RL]

In fact it is a reset problem on all Memorymoogs on power off. Most old batteries have still more than 3 Volts. Only a good reset circuit avoids that problem.
Have fun,
Rudi

[LaxSlash1993]

Well, the glide issue is now showing on all of the patches that were corrupted, and most now show the "0 VOICES." However, the tuning patch that I saved is not showing those issues. If I reload the patches, should I expect it to be fine? So far, the tuning patch is still intact. I'm a little confused as to how the CPU might have glitched when it was off, as well. I hope it's the battery... but the other weird thing is why this would've only happened to some patches and not the others.