OMS 404 multiladder not filtering.

[revtor]

I’ve got an old multiladder here recently that doesn’t seem to be doing any filtering. The six and 12 DB outputs pass signal from the three inputs while the 18 and 24 DB outputs do not pass a signal. Does anyone have a schematic for this or any thoughts based on the schematic as to what to look for?


Thanks a lot!
Steve

[revtor]

FYI, it’s a June 2000 PCB built on a legit 2u MOTM panel with the Oakley green pots. The soldering looks decent.

[Synthbuilder]

Hi Steve,

Does anyone have a schematic for this or any thoughts based on the schematic as to what to look for?

If you send me an e-mail I can send you the schematic. My e-mail can be found here:

http://www.oakleysound.com/contact.htm

If the resonance control is working correctly, the most likely cause of the outputs not working is a fault in the wiring between the board and the sockets in question.

Tony

[revtor]

Thanks Tony!

I will check the operation of everything once again. Hopefully just a miswired pot or socket. I’ll email you for the schematic.

Thanks again,
-Steve M

[Synthbuilder]

I will check the operation of everything once again.

There's a very high possibility that the pots will be in a poor state. Contrary to the advertised specifications those early green Omeg pots did not stand the test of time. As soon as I started getting reports of crackly pots I quickly moved over to Spectrol pots which required every PCB to be redesigned.

Omeg Eco pots can be replaced with standard 16mm Alpha or ALPS pots. However, the mounting bush is much smaller than the bigger Omeg pots which means the new pots need to be carefully aligned to the panel to make sure the pot sits in the middle of the panel hole. For the pots mounted to the PCB you will find that the original Omeg pot brackets are not suitable for the smaller Alpha pots. Like the panels, the hole on the bracket is too large. Some people just solder the pots to board and throw away the brackets - which should be fine if the unit is not going to be moved around a lot. The other option is to get new Alpha pot brackets from me. These will fit the new pots and also fit into the PCB. Unfortunately, the pot's pins will now not exactly match up to the position of the three solder pads on the PCB. So the pot can either be mounted with its pins sticking upwards and away from the board surface and then hand wired to the PCB, or the pot legs can be gently bent to fit the holes in the PCB.

None of these methods of mounting is ideal - although I tend to use new brackets and hand wire the upside down pot to the board. I don't feel comfortable about bending the pot legs.

[revtor]

Hi Tony,
I got the build doc and schematic from another wiggler here (yay!!)

I noticed that there was no wire and resistor connecting the switching tab of the feedback jack socket to pin 1 of U12 so I put that in place, still no change. There’s no output from either the 18 or 24 DB sockets. I reflowed all the joints around U12, U18, and U20 and the 18/24 dB sockets and their wiring- no change.

Checking voltages at the IC’s, everything okay except u18 pin 4 not getting +15. Reflowed it again, nothing. jumpered it to +15, still 18/24db jacks are silent. All resistors and connections have continuity to their pins of U18 and U12, from ICs to jack sockets.
Swapped U18 with U14 and behavior is the same. (These are the only two socketed IC’s)

The six and 12 DB sockets pass signal just fine (12db socket is lower volume) but overall, no filtering action no resonance, no CV control or anything.

I’d love to get this working! if not I can probably return it. But it’s a great fit for my little 5U case..

Thanks for any ideas
Steve

[revtor]

could a bad pot keep the filter from working at all?

[Synthbuilder]

could a bad pot keep the filter from working at all?

Probably not. And those Omegs normally don't stop working completely, they just go horribly noisy.

First thing to check is that 0V is getting to pins 2 and 3 on the power header. With the module plugged in, but powered down, the resistance between pins 2 and 3 should be close to zero ohms. Also the resistance from pin 2 on the power header of the Multiladder to any other module's jack socket's mounting nut should also be close to zero ohm.

What concerns me most is that U18 didn't have any +15V. That suggests a tracking fault, ie. a broken copper trace somewhere on the board. That was pretty rare 20 years ago, and impossible with the tested boards I do now. However, one tracking fault on the board could mean others. Do you know if this board ever worked?

Tony

[revtor]

could a bad pot keep the filter from working at all?

Probably not. And those Omegs normally don't stop working completely, they just go horribly noisy.

First thing to check is that 0V is getting to pins 2 and 3 on the power header. With the module plugged in, but powered down, the resistance between pins 2 and 3 should be close to zero ohms. Also the resistance from pin 2 on the power header of the Multiladder to any other module's jack socket's mounting nut should also be close to zero ohm.

What concerns me most is that U18 didn't have any +15V. That suggests a tracking fault, ie. a broken copper trace somewhere on the board. That was pretty rare 20 years ago, and impossible with the tested boards I do now. However, one tracking fault on the board could mean others. Do you know if this board ever worked?

Tony

Hi Tony, I had a few hours today to spend looking at this and probing around.
I got this module recently in a bundle and so I don’t know the history of it.. for me, it has never worked.

Yes, pins two and three of the power header are electrically connected and the sockets ground tabs all are connected to each other through the circuit board and down to the power supply. I’m pretty confident that that is not the issue.
I started looking at individual components today and realized that there were 33 pF capacitors instead of 33 nF in the ladder. Closest I had was 22 nF, so I swapped them in, and now no signal gets through at all. All of the other capacitors values are correct, and all of the resistors that I checked today have been correct as well. I reflowed about 75% of the circuitboard at this point.

As far as I can tell, the input mixer circuit works fine because signal is getting on to pin 2 of U19. After finding a signal there, I can’t seem to find it anywhere else. Mind you, I’m just using a wire through a capacitor into my amplifier as my probe.. I could grab a scope though, if I had some guidance on where/what to check for.
Both TL074 chips have been replaced, since they are in sockets and I have a bunch.

I ran the tuning pot up and down through the range and that did not bring anything to life.

Thanks all, for any direction or help!

Steve

[Synthbuilder]

... Closest I had was 22 nF, so I swapped them in, and now no signal gets through at all.

Interesting. That suggests the ladder is filtering but the control current is too low. The control current is derived from the exponential convertor based around U13 and U15. Check the output voltage (with respect to ground or 0V) on the CV summing amplifier with the voltmeter or scope on pin 1 of U13. When you wiggle the frequency control on the front panel the voltage should change from around -12V to +12V. Check the voltage across R106. This should also vary but over a much smaller range, something like +200mV to -200mV.

If all is well check pin 6 on U13. It should be 0V. If not replace U13.

[revtor]

All right Tony, voltage on pin one varied from a little over 14 to -13.7 as the freq pot was turned.

Negligible voltage across R106, that did not change with the frequency pot.

Pin six of u13 stuck at 13.7 V.

Replacing u13 now.

[revtor]

New TL072 in (a socket this time).... same readings.

Pin one reads 14.3 (freq CCW) to -13.8 (freq CW)
Pin six reads 13.9v

V across R106 spans .025v to -.025v as freq pot is turned.
EDIT- noticed that R106 was 1K so I changed it to a 10 K and now the voltage across is .24 to -.24

Both tune and v/oct trimmers are centered.
R116 R117 R118 are correct values.

Edit again:
The pin one voltage reading reaches -13 V or so around noon on the freq potentiometer. So not a linear sweep at all. I took the module apart and replaced the green Omeg with a 50k linear alpha but that behavior stayed the same.

And I want to thank you again for your time here Tony, this module is officially at this point not worth my time anymore, but now I’m just curious and don’t wanna give up.

[Synthbuilder]

The pin one voltage reading reaches -13 V or so around noon on the freq potentiometer. So not a linear sweep at all.

That's odd. Is R138 100K, R137 2M2, and the trimmer 10K?

As an aside this section is not my best design work. A better set of values would have R115 as 47K and R116 as 270K. This would allow the full range of input voltages to be summed by the left hand side of U13 without pin 1 clipping which it does at the moment. If you want to change this you will now notice the voltage on pin 1 moves from +7.5V to +7.5V. That said, this isn't the reason your module is not working at the moment.

Pin six reads 13.9v

I'm pretty sure that this is our problem. Since it's not the op-amp it is likely to be either that D5 is around the wrong way or broken, or that the NPN pair (U15) is not working. Now U15 is a rare part these days and fake 2SC1583s were even around twenty years ago. It is possible to cludge in there a couple of BC549s or other general purpose NPN transistors, eg. 2N3904. Just match up their leads (emitter of the replacement transistor to match the 2SC1583's emitter leads) to fit into the holes left by the removed 2SC1583.

[revtor]

Thanks for those ideas Tony, will definitely put them to use in the next couple days.
Will update the resistors, new diode. Then if it’s still not working replace the transistor package with 549’s.
The “tune” trimmer in my board is 100k and v/oct is 10k.. (matching the schematic) should I change tune to 10k?

Fingers crossed. !!

[Synthbuilder]

The “tune” trimmer in my board is 100k and v/oct is 10k.. (matching the schematic) should I change tune to 10k?

No that's fine. I should have said that the scale (V/oct) trimmer had to be 10K. If that one had been 100K, that may have been the source of the weird behaviour with the pin 1 voltage.

[revtor]

The “tune” trimmer in my board is 100k and v/oct is 10k.. (matching the schematic) should I change tune to 10k?

No that's fine. I should have said that the scale (V/oct) trimmer had to be 10K. If that one had been 100K, that may have been the source of the weird behaviour with the pin 1 voltage.

Okay, so I changed those resistors and put in a new 4148 diode.. -we now have signal passing and filtration!!! 6, 12, and 18 dB outputs have a definite strength difference as they should.

But ..
No resonance. Any patching into the feedback input has no effect either. Traced the whole feedback scheme and everything seems okay. The IC is getting power, reflowed it all, played with the fbk trimmer, Still nothing.
And the 24db output is still dead. Traced and reflowed it as well ..all nets have continuity, parts seem correct. Silent output.

So forward motion which is good, but I have uncovered more issues... ugh.
Thoughts anyone?
Tony I could post the schematic if you wouldn’t mind, others could take a swing at this too...

[Synthbuilder]

I could post the schematic if you wouldn’t mind, others could take a swing at this too...

Please don't.

No resonance. Any patching into the feedback input has no effect either.

Do you have a scope or an audio probe? An audio probe is well worth building. See here:

http://diy-fever.com/misc/audio-probe/

If so take a look at the signal on pin 1 of U17. This should show the resonance signal either from the four pole output or the feedback socket. Since the four pole output is silent, plug something into the feedback socket and see if it gets to pin 1 of U17. It will be quite small, a 5V signal into the feedback socket will turn into a 0.3V signal here.

That same signal should also be present at the negative end of C32 when the resonance/feedback pot is turned up full. If not, it is possible the pot is dead.

If there is a signal it should should also show on the 1, 2, and 3 pole outputs. The filtering will sound like the usual low pass filter action. The feedback input is essentially another audio input.

For the four pole output being silent:

Swap the ICs in positions U18 with U14 and see if the problematic output changes. If the four pole output now works, the IC previously known as U18 is partially dead. Swap U16 with U12 and see if the problematic output changes. If the four pole output now works, the IC previously known as U18 is partially dead.

With the power off, measure the resistance between pins 3 and 5 of U18's socket. If it's less than 100 ohms replace U18 or U20. If it's close to zero, then check for board shorts around C24.

Power up again. What's the voltage (with respect to ground) at pin 1 U12? It should be close to 0V (+/-0.5V). What's the voltage across pin 1 and pin 7 of U18? Again this should be close to 0V.

[revtor]

Amazing! Thank you Tony. Scheme is safe don’t worry.

Will work through this today.

[revtor]

......no resonance. Any patching into the feedback input has no effect

If so take a look at the signal on pin 1 of U17. This should show the resonance signal either from the four pole output or the feedback socket. Since the four pole output is silent, plug something into the feedback socket and see if it gets to pin 1 of U17. It will be quite small, a 5V signal into the feedback socket will turn into a 0.3V signal here.

-yes, signal does appear on pin 1 of U7 when fed to the feedback socket.


That same signal should also be present at the negative end of C32 when the resonance/feedback pot is turned up full. If not, it is possible the pot is dead.

-Yes the signal does appear at the negative end of C32, and its level varies with the feedback pot.

If there is a signal it should should also show on the 1, 2, and 3 pole outputs. The filtering will sound like the usual low pass filter action. The feedback input is essentially another audio input.

-Yes the signal does appear on the 1, 2 and 3 pole outputs. When I patch the 18 DB output into the feedback socket and I turn it all the way up I do get some noticeable resonance. So I think the core of the problem lies with the 24DB output being silent.


For the four pole output being silent:

Swap the ICs in positions U18 with U14 and see if the problematic output changes. If the four pole output now works, the IC previously known as U18 is partially dead.

-swapped U18&14, no change. (Those ic’s Are new as of a few days ago)

Swap U16 with U12 and see if the problematic output changes. If the four pole output now works, the IC previously known as U18 is partially dead.

-new TL072’s for U16 and 12 now in sockets, no change, output 4 still mute.


With the power off, measure the resistance between pins 3 and 5 of U18's socket. If it's less than 100 ohms replace U18 or U20. If it's close to zero, then check for board shorts around C24.

-resistance btwn pins 3&5 of U18 measures 12Meg ohms.

Power up again. What's the voltage (with respect to ground) at pin 1 U12? It should be close to 0V (+/-0.5V).

-pin 1, U12 measures -13v. (!). The resistors around U12 are all correct value

What's the voltage across pin 1 and pin 7 of U18? Again this should be close to 0V.

-btwn pin 1 and 7 reads +13.7 v

In addition to the above..
Looking over and measuring everything for the 4000th time, I noticed that with no inputs and all dials turned down all the way, the fourth pole output sitting at -13.2v, third pole is at -5v, second at -2v and the first pole at -3.5v. I swapped U18/14 and 12/16 again and these negative V’s stay the same..

[Synthbuilder]

I noticed that with no inputs and all dials turned down all the way, the fourth pole output sitting at -13.2v, third pole is at -5v, second at -2v and the first pole at -3.5v. I swapped U18/14 and 12/16 again and these negative V’s stay the same..

Interesting. The ladder is unbalanced. The current running down each side of the ladder is different but it should be the same. This means that the voltage across each capacitor is not close to zero, which it should be.

Since it gets worse the nearer the top of the ladder you go, I'd suggest that U20 is not working correctly. However, before you replace it, with the power off, measure the resistance between pins 2 and 7 of U20. Measure on the pins of the chip and not the pads on the board. There should be a resistance of zero ohms as the two pins are joined together. Do the same with pins 1 and 8. Again, these two pins should be joined together.

[revtor]

Hmmm. Resistance btwn U20 1 and 8 and 2 and 7 are both zero ohms. At the chip and at the board. The board traces are there and seem solid.
When powered up, the voltage across the ladder caps is:
C24:
C25:
C26:
C27: all zero volts!

[Synthbuilder]

...all zero volts!

Bizarre given that we have the outputs at such odd voltages. So all those zero volts suggest the ladder is balanced - although your meter will have an effect to drag them down a bit. This initially suggests that all the differential amplifiers are up the spout. But given that you're replaced the op-amps and we have audio signals from the lower three outputs suggests the diff amps are working just fine. So I'm still thinking that the ladder itself is not properly balanced somehow.

What's the voltage at pin 2 of U20? It should be around 8.9V.

Try swapping the SSM2210 packages. Does that make a difference to the voltage seen at the 4 pole output amplifier (pin 1 of U12) or the voltage seen at pin 2 of U20?

Does moving the BAL trimmer make a difference to the voltages found on the outputs?