Oberheim Matrix 6 Troubleshooting

Discussions about anything analog, digital, MIDI, synth technology, techniques, theories and more.

Oberheim Matrix 6 Troubleshooting

Postby telmnstr » Mon Aug 21, 2017 6:16 am

Howdy!

I recently scored one of the keyboards on my "check list" of want to owns, an Oberheim Matrix 6.

I purchased it from a local who alerted me to the fact that one of the voices was bad, no worries I tinker and repair things.

It's throwing me for a bit of a loop though.

Voice 6, it's not totally out but sound different -- one of the functions is strange. I found that by turning DCO2 off on a patch where all other voices go silent, the bad voice still sounds.

I swapped the CEM chips and the issue stays on voice 6, so that indicates it's not the CEM chip as far as I can tell.

I've already socketed and replaced the 4051 ICs on that voice which act as decode logic, which originally I thought was going to fix my issue but nope.

Any thoughts? I compared the resistors in voice 5 with 6 and they read the same in-circuit.

Looking at the schematics I don't totally grasp it yet. I see that each voice has a bunch of lines coming into them carrying data, but haven't quite grasped how that is demultiplexed into it.

I'm guessing the 40 pin ribbon cable is address and data bus, given all the other support components on the voice board.

Anyone troubleshoot one past CEM chip swaps?
telmnstr
Newbie
Newbie
 
Posts: 6
Joined: Mon Aug 21, 2017 5:57 am

Re: Oberheim Matrix 6 Troubleshooting

Postby Mooger5 » Mon Aug 21, 2017 10:42 pm

There´s an ongoing thread about the Matrix 1000 in a similar state of trouble of misterious origin. That both synths have a failing voice 6 is probably no coincidence. Prior to replacing the mux ICs, did you notice if one the chips was from a different manufacturer or had a different sufix than the others?

You said you compared the values of the resistors; did you do the same for the capacitors? I reckon by this time replacing the capacitors would be a nobrainer.

Try comparing the in and outgoing voltages on both chips (from the offending voice and from a voice that works). An oscilloscope would be best for the job.

The Matrix 6 being editable makes it easier to troubleshoot. The oscillators are always on. By closing the VCF and setting the VCF ENV to 0, there ought to be no sound going out even when the AMP VCA is fully open.
If there is still sound at the output, and knowing the VCAs are trimless i.e. there is no CV feedthrough, I think the sound source might be extraneous to the chip. Assuming of course every voltage rail and ground connection is within spec...
If there is no sound, the problem is most probably in the CV for the AMP VCA.
Herrare umanum est.
Mooger5
Expert Member
Expert Member
 
Posts: 1288
Joined: Tue May 06, 2008 5:11 pm
Location: Lisbon

Re: Oberheim Matrix 6 Troubleshooting

Postby telmnstr » Wed Aug 23, 2017 2:23 am

Much thanks for the advice! I'm starting to dig into it this evening.

I just confirmed that all the ICs on all of the voices in my Matrix 6 (keyboard model) are similar. There is no different chip on voice 6 (I have the original ones I removed.)

I have not compared the capacitors. I have an ESR tester ... somewhere. I need to dig it up. As I recall they don't work so well in-circuit.

In the arcade world, replacing all electrolytic capacitors is normal. Replacing poly and tantalum caps is not normal. I notice on the circuitry around the voice chips there are a lot of poly looking caps. Will prod at them, and also try to get some comparisons with the oscope of ... things.

----

Okay, played with the scope for a few minutes (Chinese digital unit, analog one is buried downstairs.)

I notice the ground seems to be pretty noisy, so swapping the electrolytics might clean that up. Or maybe that is just noisy on any PSU when looking at it with the DSO.

With DCA2 set to off, all voices go silent except 6. So it's not listening to that instruction, for some reason. Some patches voice 6 sounds off because certain paremeters are ignored, others sound fine because it doesn't use those parameters. Most sound off.

I tried the other setting suggested under VCA setting it to zero and that muted all voices.

Pretty wild.
telmnstr
Newbie
Newbie
 
Posts: 6
Joined: Mon Aug 21, 2017 5:57 am

Re: Oberheim Matrix 6 Troubleshooting

Postby Mooger5 » Wed Aug 23, 2017 10:35 pm

telmnstr wrote:
I just confirmed that all the ICs on all of the voices in my Matrix 6 (keyboard model) are similar. There is no different chip on voice 6 (I have the original ones I removed.)


Oh. OK, thanks.

I have not compared the capacitors. I have an ESR tester ... somewhere. I need to dig it up. As I recall they don't work so well in-circuit.


ESR meters do work in-circuit. Try it.

In the arcade world, replacing all electrolytic capacitors is normal. Replacing poly and tantalum caps is not normal. I notice on the circuitry around the voice chips there are a lot of poly looking caps. Will prod at them, and also try to get some comparisons with the oscope of ... things.


If the M6 has tantalums, replace them as when they fail they usually short.

I notice the ground seems to be pretty noisy, so swapping the electrolytics might clean that up. Or maybe that is just noisy on any PSU when looking at it with the DSO.

With DCA2 set to off, all voices go silent except 6. So it's not listening to that instruction, for some reason. Some patches voice 6 sounds off because certain paremeters are ignored, others sound fine because it doesn't use those parameters. Most sound off.

I tried the other setting suggested under VCA setting it to zero and that muted all voices.

Pretty wild.


I think the noisy ground is typical of cheap DSOs. I must remember of attaching a ferrite ring around the power cord in mine, in case it´s picking up just mains noise.

What is DCA2?

If closing the VCA (same as setting Sustain to zero, right?) mutes all voices, then it´s a problem of the Gate not closing on voice 6. Problem is, there should be no Gate signal to speak of, IMHO, since the CV is done by software. It´s somewhere between the processor and the mux.
However, I´ve had a look at the M6 Service Manual and well it´s a bad scan, but I think there are several logic chips that say Gate #2, #3 and so on, or something? I couldn´t find a #6 however... This is where you start to admire the Roland, Moog and SCI service manuals...
BTW there are a number of calibration procedures. Have you zeroed the DAC already?

Good luck!
Herrare umanum est.
Mooger5
Expert Member
Expert Member
 
Posts: 1288
Joined: Tue May 06, 2008 5:11 pm
Location: Lisbon

Re: Oberheim Matrix 6 Troubleshooting

Postby telmnstr » Thu Aug 24, 2017 6:53 pm

I have zeroed the DAC, it was spot on but re-did it anyways then re-calibrated.

I need to brush up on analog synth architecture.

I will check all of the caps for shorts tonight, and try to find my ESR meter. I never had great luck with it but this might prove it useful.

I looked at the data sheet for the CEM chip and interestingly the implementation in the Matrix 6 is very close to the example in the CEM datasheet as far as the multiplex ICs, DAC, etc.

I looked at the Matrix 1000 schematic last night and the voice section looks close so that is a bit of a help just seeing it all laid out on one page and readable.

More poking will happen.
telmnstr
Newbie
Newbie
 
Posts: 6
Joined: Mon Aug 21, 2017 5:57 am

Re: Oberheim Matrix 6 Troubleshooting

Postby Mooger5 » Thu Aug 24, 2017 10:55 pm

CEM3396 datasheet, page 13, second paragraph. The resistor in series with a capacitor. If that cap fails, there will be an offset at the output. Pin 21. If the10K resistor checked OK, the 4.7u cap should be next.
Herrare umanum est.
Mooger5
Expert Member
Expert Member
 
Posts: 1288
Joined: Tue May 06, 2008 5:11 pm
Location: Lisbon

Re: Oberheim Matrix 6 Troubleshooting

Postby telmnstr » Thu Sep 07, 2017 1:31 pm

Mooger5 wrote:CEM3396 datasheet, page 13, second paragraph. The resistor in series with a capacitor. If that cap fails, there will be an offset at the output. Pin 21. If the10K resistor checked OK, the 4.7u cap should be next.


Okay I replaced the 4.7uF 25v electrolytic cap at C609. Checked the resistor as well against a neighbor voice. The resistor checks out OK, capacitor didn't make a difference.

I wish I understood the control path to the voice, would like to be able to swap the data going to the voices to make sure the problem isn't somewhere on the digital side. I will have to look at the Matrix 1000 schematic. Maybe the 74HC32P ICs are the decoder for that.
telmnstr
Newbie
Newbie
 
Posts: 6
Joined: Mon Aug 21, 2017 5:57 am

Re: Oberheim Matrix 6 Troubleshooting

Postby Mooger5 » Thu Sep 07, 2017 4:51 pm

So the problem here is not a matter of VCA leakage, it´s with oscillator 2 still sounding when set to off! I must have mixed up both threads. Sorry for the inconveniance!

There is a balance control in the CEM3396 that crossfades the DCOs. You will notice that on the datasheet. I think the MUX that controls them is labeled U704 but the service manual scan for the M6 is hard to read. To confirm it, see where pin 14 of the CEM chip for voice 6 leads to.

EDIT- Or just ignore the above. Crossfading doesn´t totally silences the voices. So you must be referring to Voice Select set to OFF on DCO2. There is a control for that in the CEM3396 but this part of the M6 scan is definitely unreadable. You can only tell that pin 6 of the CEM3396 leads somewhere, probably a transistor.

Good luck!
Herrare umanum est.
Mooger5
Expert Member
Expert Member
 
Posts: 1288
Joined: Tue May 06, 2008 5:11 pm
Location: Lisbon

Re: Oberheim Matrix 6 Troubleshooting

Postby Mooger5 » Fri Sep 08, 2017 8:25 pm

http://www.cedos.com/datasheets/cem3396pdf.pdf on page 8 under Waveform Selection: if converter A is osc1 and converter B is osc2, then for both oscs to be "off" the voltage at pin 6 must be between +0.5 and +2. If a value between -1.5 and +0.5V reaches the CEMs osc1 turns off while osc2 remains on! Since all pins 6 are linked, because the on/off controls for the oscs are common, zero volts at voice 6 due to a broken trace could very well be the cause.
Herrare umanum est.
Mooger5
Expert Member
Expert Member
 
Posts: 1288
Joined: Tue May 06, 2008 5:11 pm
Location: Lisbon

Re: Oberheim Matrix 6 Troubleshooting

Postby telmnstr » Sat Sep 09, 2017 3:45 am

Mooger5 wrote:http://www.cedos.com/datasheets/cem3396pdf.pdf on page 8 under Waveform Selection: if converter A is osc1 and converter B is osc2, then for both oscs to be "off" the voltage at pin 6 must be between +0.5 and +2. If a value between -1.5 and +0.5V reaches the CEMs osc1 turns off while osc2 remains on! Since all pins 6 are linked, because the on/off controls for the oscs are common, zero volts at voice 6 due to a broken trace could very well be the cause.


Before work, keyboard cold I swung by and checked continuity across pin 6 on CEM3396 on voice 6 versus 5 and 4, it beeped. Sigh. Went to work. Came home, hooked up hook test lead to ground and decided to measure voltage. Voice 6 was way off! Low and behold, pin 6 read no continuity to the neighbor chip! So I pulled the PCB out, looked at the trace side and it looks pretty mint. I sort of re-flowed the solder point (A friend has borrowed my Hakko desoldering tool at a bad time ;-) and that didn't help. Continuity from pin through PCB. Bit the bullet and soldered a jumper wire from voice 6 pin 6 to 5, and that cured ... things.

It still doesn't sound identical in all things, but that definitely cured some things. Which is strange. What I will do next is check every pin on the chip looking for continuity issues and dig into the parameters to see if I can figure out which parameter isn't being picked up now.

Strange that there is a crack in the trace or cold solder joint, the board looks very good. I do notice the CEM sockets look like they have some flux and stuff where the rest of the PCB doesn't. It does have some spots that look like re-work.

Much thanks, your knowledge of synthesis is impressive! Will report next findings.
telmnstr
Newbie
Newbie
 
Posts: 6
Joined: Mon Aug 21, 2017 5:57 am

Re: Oberheim Matrix 6 Troubleshooting

Postby telmnstr » Sat Sep 09, 2017 4:02 am

Calibration brought it in line! Now it seems 100% ! I bow to you!! (And hopefully this thread helps others!)
telmnstr
Newbie
Newbie
 
Posts: 6
Joined: Mon Aug 21, 2017 5:57 am

Re: Oberheim Matrix 6 Troubleshooting

Postby SkyWriter » Fri Sep 14, 2018 12:16 am

Tele, wasn’t there a way to get a laser in there too? :-)

My matrix-6 still works. My 6R though is a bit noisy. Too much to fix around here sometimes...
SkyWriter
Newbie
Newbie
 
Posts: 1
Joined: Wed Dec 17, 2008 9:34 pm


Return to General Synthesizers

Who is online

Users browsing this forum: Bing [Bot] and 12 guests