Moog Music Forum

whinylittlerunt wrote:

Markyboard wrote:Also does pushing the autotune button do anything or is it locked as well?

Locked up as well. Unless I remove U5, then I can hit it and it works for about 10 seconds until it all locks up again.

I'm guessing that removing the Op Amp breaks the closed loop sample and hold circuit where the CPU is writing the value it has stored in part of it's timing cycle and sampling the parameter for updates in another part. I wouldn't make too much of this observable.

I also just realized I was probing the Z80CTC, not the CPU - Glad I looked at the schematics! I want to recheck that tonight and make sure things look right.

I still feel like the problem is at the DMUX somewhere.

If I probe something and it goes high on the scope to the point of disappearing, should I assume that it's still good or is that also an indication of a failed or failing chip?

EDIT: MB I just saw your response about the chip removal. Cool... I just ordered a ton of chips, pretty much everything to cover the DMUX board in the event I need it, plus a couple Z80s to keep on hand.

whinylittlerunt wrote:I also just realized I was probing the Z80CTC, not the CPU - Glad I looked at the schematics! I want to recheck that tonight and make sure things look right.

I still feel like the problem is at the DMUX somewhere.

If I probe something and it goes high on the scope to the point of disappearing, should I assume that it's still good or is that also an indication of a failed or failing chip?

Cant answer that last question - too many variables. But I agree it's still looking like the DMUX. At this point it looks like the CPU isn't seeing the fact that a button is pushed or a pot is turned...OR... it seeing it but when it trys to send an updated value or state to the DMUX it's not getting read. All the front panel information gets back to the CPU via the MUX BUS signal going through the U6A comparator. So again, try to look at that MUX BUS signal and the address and the inhibit lines of the 4051s.

And trust me, this can be difficult to troubleshoot even if I had the board in front of me.

I'm getting readings on U6, both the comparator line and the MUX line. Just for giggles I pulled the chip and fired it up, and sure enough I get no front panel leds (except for the octave switch on the mod wheel). I tried replacing the chip with a TL082 and it made no difference (though I don't know if the chips I have are good). I'm also following the same reading to the connector on the digital board. I also reseated the CPU and tried the TP reset again but again, nothing. All 4051s on the DMUX are getting reading on the INB lines. Voltages still look right in various places I check.

An interesting thing happened though. After I reseated the CPU, I fired it up and noticed I had no more noise on boot and functions started working, except no audio. Then it froze again. A minute or 2 later. I thought I was getting somewhere but then realized my 15v line came out of the power supply. So I plugged it back in and was back to where I left off. When I shorted the board I was testing the voltages at the caps on the DMUX. I am still getting correct voltage everywhere but would I be right in presuming whatever I blew out was on the 15v rail?

Is it possible I just damaged the CPU somehow? I'm getting readings from it (reset, busrq, etc), just don't know what I'm looking for. I do have my new parts coming in the next day or two so if I have to start throwing stuff at it, so be it...

OK - let back track here for a second. I watched your posted video again and wonder did you power it off at the end? In other words if you leave it on does it stay on with the same LEDs on each time and the character display showing MOOG? If so than I think your original problem is still present. In other words your probe shorting may have done something downstream affecting the audio (or not), but it's unrelated to the lock-up condition. Could be your intermittent problem elevated to a permanent problem which on the bright side should make it easier to troubleshoot

This being the case maybe your rework on the power supply didn't really fix anything - certainly worth a closer inspection of what you believed to be a cracked solder joint. Have you checked all the secondary power signals including the sense lines? With this train of thought I would look carefully at the Digital Board reset circuitry and battery back-up circuit. Also good descriptions in the service manual - may be of help. By the observable symtoms it appears that things are stuck in Reset mode. Again it would be helpful to know whether all LEDs come up the same each power up .

I'm starting to run out of ideas here Mike -trying to do this remotely is very difficult. Also (and as I said I'm not the scolding type) I highly encourage you to not swap out parts randomly. I'm betting there are a few very knowledgeable regulars here lurking and rightfully shaking their heads because they've been here too many times! This board is 35+ years old. Those socket connectors suck even when they were new, and they oxidize along with the ribbon cable connectors. Every time you pull a part you risk causing another problem compounding the issue. This make it impossible to troubleshoot logically. The chances of you getting lucky are not good. If you end up bringing to a repair shop your're potentially looking at much higher cost. Well, I never said I wasn't the lecturing type . But seriously- just trying to help.

I may be off line for a few days- After a replacement Mother Board install Microsoft decided last night to automatically upgrade me to Window10 Pro (was Home edition) and now my Access code doesn't match up. Got to do a clean install. Damn - I should have activated right away...UGGGH

MB,

The video I posted was after I dropped the lead and zapped something. That noise it makes is what happened as soon as I hit that cap or whatever I sparked against, and has not come back from it.

When I originally got the synth, the description was that it would work for maybe 2-3 minutes and then shut down. I confirmed that yes, it works about 2-3 minutes and the keyboard worked fine, audio, all patches etc... until something overheated or tripped and then locked up. It would not "shut down" but it definitely froze.

I started by checking the voltages based on the steps in the service manual under the power supply troubleshooting. I got 15v, and then when it said to check for 5v I probed it and saw I was getting about 3v and dropping. Then it asks you to pull P131 and check the pins themselves for voltage but I was reading 0.00v, only later to find out that the connector description may be backwards in the manual... either way, I still wasn't measuring 5v on the pins anywhere. I was getting about 12v up to the transistor before the regulator and then nothing. So when I pulled the power supply, I looked carefully around the back and noticed the shunt resistor on the 15v side was loose and not making a good connection. so i soldered it up and put it all back, and the synth powered up and I measured 5v at the point where I was supposed to, and it stayed on... hours I let it run, Never failed.

So I pulled the supply and re-capped it, I replaced the 5v regulator and socket while I had it out just because I had pulled and re-seated it a few times during my troubleshooting. I also replaced the trimpots. I put it back in and again, everything worked and stayed running.

It was only when I went to start calibrating, in fact re-calibrating the voltages to be sure, I slipped with my red probe and hit either one of the adjacent caps or something else causing it to wig out like you hear in the video. I powered it off and then powering it on puts me right back to where I'm at now. I pulled the DMUX board and checked it for damage underneath because I got nervous. Then I changed out the caps on that board. Like 4 of them. I re-assembled everything and no changes.

When I power it on, it wigs out and and stays frozen. I think *sometimes* if not all the time the RATE led will flash. The octave switching buttons and LEDS will always work. The sound will go away after a few seconds and then it remains on and frozen. As mentioned before, if I pull certain components, I can work the front panel when I first power it on, just for a few seconds...

I don't know what else to say, I may have to send it off at this point. I will see if all the same LEDs come on and stay on when it's powered on, I didn't explicitly look for that, I just figured they were all the same.

Sorry to offend anyone who knows these machines and thinks I'm just breaking it more - I'm really trying to resolve the issue myself but I don't know if i'm looking at a DMUX board issue or a digital board issue... so far everything I probe is getting some sort of activity, albeit some very little but it's there (but that could be my scope settings too).

Ok thanks Mike - I have a better understanding of your short history with it which helps. I have to ask ....any chance you live within reasonable driving distance of Washington DC.?

I’m in CT- but it’s not out of the question...

PM me.

Guess what...

On a whim I pulled the RAMs and shorted them in foil...

Everything is working! Holy *** Just need to reload the factory data.

I came across a post on another forum, don't even remember which one, earlier today where it described almost the same symptom as what I was having (the long sweep, frozen LEDs) and someone suggested corrupt data. This had happened to my OB-8 once before after I recapped the digital board and I didn't know what to do... had to bring it in for service and for all I know, they did the same thing and fixed it.

NOW... I'm going to calibrate it CAREFULLY. Interestingly enough, I'm getting 5 voices tuned right off the bat, whereas before I was seeing 0, then would creep to 2 max. Coincidence? I dunno. But, I'm happy!

Shorting RAM in foil? That’s a new one on me. But whatever, glad it’s working.

I have heard of that trick from respected sources, have not tried it myself.

MC wrote:I have heard of that trick from respected sources, have not tried it myself.

Well now I'm really curious. I Googled it but the only info of possible value was ESD related. There must be some latch-up condition with this RAM that remains even when powered down. Some how the aluminum foil discharges whatever potential remains. Always up for adding to my bag of tricks so thanks Mike and Mike.

Glad I could be the guinea pig!

I suppose replacing those chips would also have fixed the problem if we never thought of that trick.

I finally finished the calibration and now have a very nice Memorymoog to play!

Getting the voices in tune. Took a long time but I followed the manual backwards, so I actually did the full voice calibration by tuning each oscillator on each voice to the correct frequency.


Once I got to the software tuning section, I knew I'd hit the dreaded upside-down characters and needed a solution that wasn't going to break my stones. I didn't have a mirror. But, I came up with a pretty novel idea... I got out my DV video camera and ran it live pointing at the values. I hooked it up to a spare flat-panel TV and flipped the TV upside down


The software tuning went pretty quick, and I only had to adjust the range trimpot. I was thinking that it was possible, having done all the extended voice tuning ahead of time, that I may have expedited the software tuning since the values were maybe now closer? i don't know, but I got every voice in range fairly quickly. I think I only had one that was 7F all across, but every one of them had a 7F value in there and the other 2 values were right in line.


I now have a very nice, very playable AND very stable (it seems so far) early #SN Memorymoog. I must have played it until 2 in the morning. I feel very accomplished for having taken this on, and I think I'm at the point of leaving it alone unless something funky crops up. This isn't an instrument I want to tear down a lot.

Here she is, under my Roland SH-2:

Love a happy ending. Oh, and this story’s good too