OBX-a DAC calibration - IC #64 and other fun

[defog]

Hey Guys,

I recently did some work on an OBX-a (way too many problems to describe in one post), and I'm finally at the home stretch. Just a few things left to do.

EDIT: Fixed the first and second issues mentioned here - More appeared though

[First is that the unit goes about 12 cents flat overall when octave down, narrow, and OSC2 switches on the bender board are turned on (slightly lower with all 3 at the same time). I get the -1v 0v +1v on my meter dead on when the octaves are engaged. Voltages on the bender (Pin 1 of IC #8 ) are correct as well when centered, as well as offset. The whole unit just goes flat when they're activated. Something is pulling down the voltage where it shouldn't be, but I wanted to make sure this wasn't normal behavior or that I'm not supposed to calibrate everything to 1 oct down or something.]

The second biggie is that I need to know where IC #64 is located. It seems as if my synth is an octave too high on the initial note, which I'm sure has something to do with the DAC trimmer set up. This keyboard has the Kenton Midi control board installed (however, the trimmers on it adjust the midi input only and not the built-in keyboard scaling/range). Probably just needs to be adjusted lower. This is a later model unit, btw.

Thanks

[defog]

Figured out where it was pulling the voltage down. Loose wire connection for the Kenton Midi to power. Not enough juice to it and was affecting the tuning. Now the switches no longer pull it out of tune. Going to try to tune the oscillators up again tonight and I'll post details later.

[defog]

EDIT: It Works fine in Manual Mode, so it was just the old patches that were off since it has been recalibrated.

It stays in tune now with C after auto tuning since the re-soldering of the loose wire to the midi board.

However, Octave up switch above middle C is useless on OSC 1. it seems like the whole keyboard is an octave or two too high. Anything above C5 on OSC 1 is in dog hearing range. Where is IC 64 to adjust the DAC? Help!

[defog]

And the fun continues....

Got it to stay in tune and the range of octaves and all that fun stuff are good now. Only I'm getting errors with saving patches (crashes) and the 4-pole switch doesn't seem to do anything. Doh! Time to trace the signal path/logic ICs again...

[defog]

The issue has been narrowed down to the top 4 voices not responding to the 4 pole switch. CEMs are in good shape, as I've swapped them, so it is something on the logic board causing the issue. Scope time.

[Scot Solida]

What was the source of the problem with the octave being too high? I'm trying to get mine into shape, too and have the very same problem.

[defog]

[Scot Solida]

Ah, okay. On further inspection, I think mine had more to do with screwing up the initial calibration. I'm quite interested in how your OBXa turns out. You are giving me hope!

[defog]

[Scot Solida]

Mine would hold patches for a bit, but then they'd disappear - even with a new battery. I ended up installing one of those magnetic RAM boards from Electrongate and it has been solid since. However, there are still quite a few issues with it. One of the voices has a filter problem. Also, many buttons have dead LEDs. The buttons work, but they don't light. Also, the left-hand controllers are screwy and undependable - the pitch bend often does not spring back to the proper pitch. Plus, the dip switches to turn the voices on an off are faulty.

However, it works just enough that I keep procrastinating the repairs - I'm not even sure I can do them on my own. I may have to send it out and I fear the cost of that!

[defog]

[Scot Solida]

Thanks for the information... that is most encouraging! The 4-pole filter on one voice is always open... One of the trim pots for that voice is broken, so I have been assuming that was the culprit. I have a replacement that I have not put in. However, I will do as you suggest to make sure I'm not looking in the wrong place (though the pot certainly needs to be replaced either way).

That's good to know about the bender! The OBXa seems to think it's in tune (and it is, even if it might be an octave high at this time). I'll give that a try too. That left hand section had other problems if I recall, but I can't remember them at the moment.

Thanks to your posts, I am now anxious to get the thing on the table and open it up.

[defog]

Definitely replace the trimmer on the voice card regardless. The trimmers they used on them are pretty shitty. And considering you can get 100 multi-turn ones for like $10 on eBay, it would be worth it to do it for all of them. It would definitely improve your tuning stability. Just bumping the keyboard around with the original ones can knock it out of tune (literally). I just bought a bunch off eBay and plan on replacing all of them. Already did all of the electrolytic caps on them, so I'm assuming it'll take just as long (about 2 hours) for all of them.

Check this out. It doesn't have enough of all the values you need, but you get the point:
http://www.ebay.com/itm/Potentiometer-Assorted-Kit-12valueX5-variable-resistor-/200695764178?pt=LH_DefaultDomain_0&hash=item2eba6654d2#ht_7081wt_905[/url]

[defog]

Hello again,

I'm still trying to hunt down the bug in my OB-Xa that is causing the lower Motherboard/Voice Board to not produce 4-pole filtering. As I wrote before, 4-pole filtering does work on the top board (I wrote bottom initially which was incorrect), and the switch does turn on-off. It is just not being enabled/recognized by the bottom board. If anyone can lend a hint as to what controls the enabling of it on the bottom voice boards, that would be extremely helpful to me.

Here is what I've done to this, so that duplicate efforts aren't made:

- Replaced ALL electrolytics on all boards (including voice boards)
- Replace all socketed OpAmps, OTAs and Logic Chips (TL0XX, CA3080s, 4000 series, the few 74xxx chips)
- Traced leads on the bottom motherboard and cables, as well as every other ribbon cable on the synth - all check out ok
- Replaced voltage rectifiers on the power board,
- Calibrated trimmers on power board, bender board, and all voice cards (except 4-pole on the bottom ones)

* I still may replace the remaining 4000 series logic chips that are non-socketed, but I'd rather not do all of that work, even though it would probably prevent some future issues. I'll replace them down the road if necessary.

Since it is working on the top voice board, I'd suspect that a logic IC is faulty with switching it on for the bottom. However, I'm having trouble following the schematic as to what enables/disabled 4-pole filtering on the bottom. It appears is though it runs into the bottom motherboard/voice board from the bottom logic board, which then is connected to the top board, and the pot board for the switching.

Here is a link to the service manual.

http://www.lazyblueoctopus.com/oberheim_obxa_service_manual_3.pdf

On the Mother Board schematic, it appears as though "2p4p" comes from point C25 on the logic boards to point B3 on the voice motherboard. There is also "VCFP" (which seems to be the switching input) listed on that schematic, coming from point C7 to D5 on the voice motherboard. Point C7 comes from the lower control/logic board -15v point, which connects to A10 and A9. It looks like points A10 and A9 are located on the modulation board, but I'm probably reading it wrong.