DOD 690 VCO Question

[CheapPedalCollector]

Yes, it going to be a pain whatever way you go.   

If you want to make a board that plugs into *any* pedal then it has to look like a SAD512D from outside, get the LFO/VCO from the existing board, be compatible with existing input output circuits.

Something that occurred to me was to use more modern devices but it seems they are only available in 256 stage or 1024 stage.    Probably need to operate the BBD at 5V or 9V.   You would need more circuit but the idea would be to use SMD to keep the size down.   Unfortunately the modern BBDs are still through-hole.   Two 256 stages is going to be bigger than a MN3004.   If you used a 256 stage divider the clock down by it would work but the quality wouldn't be great.  You could add more filtering but that may limit the circuit you can plug in to - and the filtering will change the sound.   Going for a 1024 and doubling the clock is going to be a headache since you would need to put in a PLL to double the clock.   Other options would be to make mods to the VCO but that going to be pedal specific and doubling the frequency is unlikely to keep the original behaviour.   All messy and intrusive options.

Yeah I think the SAD512D devices will just all eventually stop working and that will be that, unless someone manages to clone the chips or make something better that works the same enough to be usable. Sad state of affairs really, these units sound really good when they work.

Additionally I've found another design flaw with this and the 670 Flanger, and possibly other of their effects from this time period. They often couple the clock filter after directly, and I found that modern a opamp of the same type will not function at these tiny voltage levels that the SAD512D can output. I suppose this was a cost cutting measure as the ECG datasheet for it always has a darlington connected to the output. I think the reason these pedals fail and the MXR/EH ones tend not to, is because that opamp is drawing too much current through the SAD512D itself. Now that I look at the design, it's really kind of dumb in that regard.

This explains why the schematics floating around have that 22K load resistor crossed out and changed to 100k. All the performer series ones have it 100k too and run at 15v as well. I've done that on two pedals now and I'm getting about 2.5v on the following stage now instead of 1.2, and the circuit begins to function as it should. Unfortunately one of the BBD chips I bought got damaged by this current draw before I did that mod and now has distortion Maybe LM358 would be a better choice than 4558 or 4559 here? Is there another op amp that can take 15v and has a low current/voltage requirement? TL022 perhaps?

I think designing a drop in for MN3005 is not a problem for the 680, I already found some nice pin headers on mouser, the Aries 16-600-21 should do nicely with a side caddy board for the MN3005, and simple enough to design. I'm going to do that and have some prototype boards made and buy a few xvive MN3005 and see how it works. I notice the 680 also has a 22K load resistor so that might need to be changed, I need to look at the MN3005 datasheet and see what it wants.

[Rob Strand]

Yeah I think the SAD512D devices will just all eventually stop working and that will be that, unless someone manages to clone the chips or make something better that works the same enough to be usable. Sad state of affairs really, these units sound really good when they work.

When the time comse some of the messy solutions may be better than having no pedal!

Additionally I've found another design flaw with this and the 670 Flanger, and possibly other of their effects from this time period. They often couple the clock filter after directly, and I found that modern a opamp of the same type will not function at these tiny voltage levels that the SAD512D can output. I suppose this was a cost cutting measure as the ECG datasheet for it always has a darlington connected to the output. I think the reason these pedals fail and the MXR/EH ones tend not to, is because that opamp is drawing too much current through the SAD512D itself. Now that I look at the design, it's really kind of dumb in that regard.

This explains why the schematics floating around have that 22K load resistor crossed out and changed to 100k. All the performer series ones have it 100k too and run at 15v as well. I've done that on two pedals now and I'm getting about 2.5v on the following stage now instead of 1.2, and the circuit begins to function as it should. Unfortunately one of the BBD chips I bought got damaged by this current draw before I did that mod and now has distortion Maybe LM358 would be a better choice than 4558 or 4559 here? Is there another op amp that can take 15v and has a low current/voltage requirement? TL022 perhaps?

The Flanger schematic I posted raises the resistor value to 100k for the R5106 device.  I would have to compare the datasheet for the SAD512D and the R5106 to see if a larger resistor makes sense.  What could have happened is DOD found problems like you have with 22k.  Then later on went they had to change to a R5106 they decided to use a 100k to fix the problem as part of the changes - it's hard to second guess this stuff.

My evaluation of the problem is the DOD design isn't great!  It's asking for trouble, you can't expect audio op amps to bias down at 1.2V.   There will be no opamp swing left even if it biases.   

It seems the MXR microflanger also DC couples the opamp off the BBD.  They use 47k on an SAD512D.  That's going to raise the bias point to a better place.  That unit is only a 9V supply, for 15V the ideal opamp bias point would be higher.

The output load resistor on the BBD supposed to have a lot slack since the BBD outputs are JFET buffers.  However, I'm sure some resistor values work better than others.   You can use quite low value resistor loads.   Changing the load resistor affects the output level of the BBD.   The "Darlington" you see isn't a Darlington, it's a current mirror the idea is it helps combine the two outputs from the BBD.   In this case the output loads are quite low.   More the point though the output taken from the transistor collector is AC coupled to the opamps.  That means the opamp biasing doesn't have to follow the BBD biasing.  So that don't have the DC couple problem anyway.

The root cause of the problem is DC coupling the BBD to the opamp.   The BBD works fine with a low DC output bias but hardly any common audio opamps will work down that low.  If you AC couple the opamp from the BBD then it avoids the problem and lets both the BBD and opamp be biased optimally.  For the opamp that's around Vcc/2, however that means you would have to add some biasing resistors to the opamp.

So you can only make the best of a bad situation and tweak the resistor on the output of the BBD to get put the opamp bias at a better voltage.

When I did the calculations for the Flanger it was evident the LM324 opamps on the LFO and following shaper stage were getting push too much into the rails if you use the crossed-out values (on the schematic I posted).  The newly written values made a lot more sense, as if someone came in a fixed the evil, although it's still running close to the rails.   

When I looked at your cleaned up schematic, I was surprised to see you re-instated all the crossed out values,

https://www.diystompboxes.com/smfforum/index.php?topic=126281.0

Is the unit really made with those crossed-out values, ie the values on you cleaned-up schematic?

[CheapPedalCollector]

The Flanger schematic I posted raises the resistor value to 100k for the R5106 device.  I would have to compare the datasheet for the SAD512D and the R5106 to see if a larger resistor makes sense.  What could have happened is DOD found problems like you have with 22k.  Then later on went they had to change to a R5106 they decided to use a 100k to fix the problem as part of the changes - it's hard to second guess this stuff.

I have a 565 Chorus and 575 Flanger and they also have 100K resistors and came with SAD512D, I don't think putting an R5106 in them is a good without lowering the voltage to 12V.

When I looked at your cleaned up schematic, I was surprised to see you re-instated all the crossed out values,

https://www.diystompboxes.com/smfforum/index.php?topic=126281.0

Is the unit really made with those crossed-out values, ie the values on you cleaned-up schematic?

Yes they are accurate to 2 real units. When I get a schematic, I want it to be accurate, and I don't want random scribbles all over it that makes it hard to read. I can make another with the changed values, but as I said I don't think it's safe to run a R5106 at 15v, and the altered schematic is a bit misleading in that regard and could cause people to damage their working unit. It would have been better if they just added *100K or similar next to the values instead of scribbling things out.

[Rob Strand]

I have a 565 Chorus and 575 Flanger and they also have 100K resistors and came with SAD512D, I don't think putting an R5106 in them is a good without lowering the voltage to 12V.

Yes, good point.  I don't keep as many of these fine details in my head these days.   The funny thing is the example in the R5106 datasheet has the chip at 15V and the rest of the circuit at 13V, yet the chip is spec'd at 13V max.   The 15V is a mistake for sure.

Something else I noticed in the R5106 datasheet is it actually says a load resistor of less than 100k is not recommended, as the distortion increases.    Reading in between lines I suspect the SAD chips have different FETs on the outputs.

Yes they are accurate to 2 real units. When I get a schematic, I want it to be accurate, and I don't want random scribbles all over it that makes it hard to read. I can make another with the changed values, but as I said I don't think it's safe to run a R5106 at 15v, and the altered schematic is a bit misleading in that regard and could cause people to damage their working unit. It would have been better if they just added *100K or similar next to the values instead of scribbling things out.

I generally try to make schematics accurate as well, or at least I write errata notes for the errors - I must have hundreds of files for schematic fixes.

Interesting your markup has the real values.   Electronically the circuit behaves quite different with those crossed out values on the schematic I posted.  I used the newly written values in my calculations - they look more correct from an electronics perspective.  The crossed out values give different VCO frequency estimates - as the LFO output is different and that affects the range of voltages at the VCO input.

To me it's totally unclear what the markups on the original schematic mean - there's no context.  They imply two versions or a fix (but by who?).  It also implies it's *not* new vs old.   Does the 100k vs 22k *only* apply to R5106 vs SAD512D?  or did someone mod the whole unit at some point and make a whole heap of changes.   *Sooo* many loose ends.    The schematic is useful for a guide but not for documenting what it is!

In the past when I'd come across stuff like this I'd trying to find many PCB pics to judge what versions were out there (no DIY clones allowed as they tend to use Web schematics).   Your real units essentially remove all the markups on the schematic.

[CheapPedalCollector]

I generally try to make schematics accurate as well, or at least I write errata notes for the errors - I must have hundreds of files for schematic fixes.

Interesting your markup has the real values.   Electronically the circuit behaves quite different with those crossed out values on the schematic I posted.  I used the newly written values in my calculations - they look more correct from an electronics perspective.  The crossed out values give different VCO frequency estimates - as the LFO output is different and that affects the range of voltages at the VCO input.

To me it's totally unclear what the markups on the original schematic mean - there's no context.  They imply two versions or a fix (but by who?).  It also implies it's *not* new vs old.   Does the 100k vs 22k *only* apply to R5106 vs SAD512D?  or did someone mod the whole unit at some point and make a whole heap of changes.   *Sooo* many loose ends.    The schematic is useful for a guide but not for documenting what it is!

In the past when I'd come across stuff like this I'd trying to find many PCB pics to judge what versions were out there (no DIY clones allowed as they tend to use Web schematics).   Your real units essentially remove all the markups on the schematic.

Yes I became very frustrated trying to my original unit due to the inaccuracies and changes. I wasn't sure if mine had been modded or not, so I ended buying another one that needed repair because the symptoms it had were that of the caps being bad, and it was. Then I had another unit to compare to and they were identical other than I had changed that 22k resistor to 100k, which made the opamps I replaced bias up properly with a new SAD512D I purchased. I'm sure there are revisions of it, I just haven't seen one yet.

I'll go ahead and post the gut shots here too, why not. I have replaced the caps in both units, and one unit has more done to it because it came to me in a worse state, missing knobs, etc.

[Rob Strand]

Yes I became very frustrated trying to my original unit due to the inaccuracies and changes. I wasn't sure if mine had been modded or not, so I ended buying another one that needed repair because the symptoms it had were that of the caps being bad, and it was. Then I had another unit to compare to and they were identical other than I had changed that 22k resistor to 100k, which made the opamps I replaced bias up properly with a new SAD512D I purchased. I'm sure there are revisions of it, I just haven't seen one yet.

I'll go ahead and post the gut shots here too, why not. I have replaced the caps in both units, and one unit has more done to it because it came to me in a worse state, missing knobs, etc.

Well well, very interesting.

I did a quick eyeball of unit/pic #1.   Your updated schematic which re-instates the crossed out values matched.
There is one more detail:
- On your schematic, the input opamp has a 2x22k, 10uF, 100k.
  However, as much as I can work out this should be:
  2x1M in place of the 22k's,  no 10uF,
  100k is not present (replace 100k with a wire so the center of the 2x1M's goes to the opamp + input)

   If you look at the schematic I posted with the crossed out values. It says old units have this exact circuit.

So what I think is going on is the crossed out values and the 2x1M etc represent the old model.

I didn't finish going over unit/pic #2 but there seems to be a good agreement with the newly marked values
on the schematic I posted.

So I guess that explains the mystery of the two sets of values on the schematic. Newly marked = new and crossed out = old.

Not sure if both units originally had the 22k on the output of the BBD.

That also means the two units you have are different.  One new and one old.
The pots on unit #1 are R1378034 = week 34, 1980
The pots on unit #2 are R1378127 = week 27, 1981

All starting to make more sense now.  I suspect the two units sound a little different.

There is one thing hanging.  The pre-marked up schematic has old values but 2x22k + 10uF+100k, as if there is a third version.   An old schematic cannot know about the new circuit configuration in this region!

[CheapPedalCollector]

The other unit did originally have a 22k instead of 100k on the output of the SAD512D, but the opamp would not bias with that so I changed it.

They indeed sound different, I sold the "newer" one because I liked the sound of the older one better.

Oops, thought I had pics of the circuit boards, guess I forgot to take some. The wires are fragile and I don't want to take the one I still have apart again.

[Rob Strand]

The other unit did originally have a 22k instead of 100k on the output of the SAD512D, but the opamp would not bias with that so I changed it.

They indeed sound different, I sold the "newer" one because I liked the sound of the older one better.

Oops, thought I had pics of the circuit boards, guess I forgot to take some. The wires are fragile and I don't want to take the one I still have apart again.

That pretty much ties up all the loose ends.  I'll do another pass of the calcs with the "old" values.
Another DOD pedal with a quirky history!  There are so many.

FWIW, the R870 circuit is more like the new version of the 670.


[For future readers the thread morphed into a Flanger discussion after Reply #20]


Errata on your updated schematic:   For the old units the 100k + 3k3 on wiper of the width pot should be 100k + 10k.

[CheapPedalCollector]

I still have the 690 torn apart, I'll take some pics of that too.

I was thinking, isn't one of the Panasonic chips 2x512? could just use half of it I guess. Would be kind of a waste, but a solution none the less.

:EDIT: Added pics

I'll take pics of the circuit board too the next time I have it out.

I did 2 changes, the 100K resistor on the output of the SAD512D, which is directly left of the socket, it was 22K stock. The other is the 22K resistor just below the trim pot which is the bias resistor for the output side of the NE571. Strangely this didn't change the bias on the circuit, and it seems a bit low to me, and perhaps the distortion is coming from the NE571 as its only about 5.81 volts with a 15 volt supply. Possibly similar to the FX90 problem I encountered.

The electrolytic caps are replaced by Nichicon 35V ones as they were all out of spec and some had very high ESR/dissipation. The 78L15 was replaced. The Coolaudio NE571 sounds better than the original to me (more treble), so I will leave it in. I will put the original LM324 back when I get some other parts I ordered as I just swapped it while troubleshooting. I socketed the 324, CD4007 and the two op amps, which I replaced with some NOS RC4559s I have as the originals (Ti 4558) were dead shorted. I also replaced the SAD socket because the ones they used are really cheap and often the flux they used (smells like honey) corrodes stuff and turns it green.

Unfortunately the SAD512D may be damaged and has quite a bit of distortion, but it sounds really nice otherwise, and it could be the NE571 making this distortion as I noted above.

I just got a new switch for it as the original was burned up, probably what killed the pedal. I still have the original cable and for once no one cut the ground prong off. I have a DC jack in there temporarily so I can power it up with 18v DC. I will be restoring the AC functionality, this is only temporary.

[Rob Strand]

First off I have some new info regarding the DOD 690 Chorus.   There are at least two versions:
- An early version, before approx mid 1980, which uses an SAD1024 BBD chip (1024 stages)
  I suspect very much this version corresponds to the 21ms delay spec in the DOD brochure.
  No schematic for this from what I can see.
- A later version, after mid 1980, which uses an SAD512D (512 stages).   That's the one you have.
(- A possible third version with an R5106.   Not sure about this ATM, hard to know if units have SAD512D subs.)

So screwed over by Dod variants again 

I was thinking, isn't one of the Panasonic chips 2x512? could just use half of it I guess. Would be kind of a waste, but a solution none the less.

Yes, there was a very early 16 pin device, MN3001, 2x512 stage.  There was also the SAD1024, also 2x512 stages.  AFAIK, none of these have reliable sources.

:EDIT: Added pics

I'll take pics of the circuit board too the next time I have it out.

Cool.

I did 2 changes, the 100K resistor on the output of the SAD512D, which is directly left of the socket, it was 22K stock. The other is the 22K resistor just below the trim pot which is the bias resistor for the output side of the NE571. Strangely this didn't change the bias on the circuit, and it seems a bit low to me, and perhaps the distortion is coming from the NE571 as its only about 5.81 volts with a 15 volt supply. Possibly similar to the FX90 problem I encountered.

I couldn't confirm that 22k resistor was in fact the biasing resistor for the NE571 on the BBD output (expandor).   The PCB traces are hard to follow.  Are you confident that's the correct resistor?

With the FX90, the supply was 9V and the 10k resistor incorrectly biased the NE571 output to 6.6V (or whatever it was) and we increased the 10k resistor to 39k to *lower* the bias voltage.

I'm not sure if your 5.81V is before or after the resistor change.    However, the original 10k resistor looked good to me for a 15V supply.  It should bias the NE571 output to 6.6V (same as original FX-90 voltage).  If you wanted to raise that voltage a tad you could use 8.2k, however it's probably not going to give you much improvement - as the BBD is the bottle neck (maybe even the opamp bias, as discussed earlier).   Lowering the resistor will raise the bias voltage - opposite to what we did with the FX-90

If you are getting distortion the most likely place tweak is the BBD bias pot.   However, if the BBD output is clean *and* the filter output (at the opamp) is clean, then maybe there is a fault around the NE571.   You can get distortion if the cap on pin 1 of the NE571 isn't connecting or the value is too small.

[CheapPedalCollector]

First off I have some new info regarding the DOD 690 Chorus.   There are at least two versions:
- An early version, before approx mid 1980, which uses an SAD1024 BBD chip (1024 stages)
  I suspect very much this version corresponds to the 21ms delay spec in the DOD brochure.
  No schematic for this from what I can see.
- A later version, after mid 1980, which uses an SAD512D (512 stages).   That's the one you have.
(- A possible third version with an R5106.   Not sure about this ATM, hard to know if units have SAD512D subs.)

So screwed over by Dod variants again 

I don't think there is an R5106 version, I have never seen those used until late Performer series in the Flanger or Chorus. There is also many circuit revisions on those, I have at least 3 different 565 boards and 2 different 575 boards. I have seen two different 585 boards so far too, one with SAD4096 and one with MN3005. The 680 has at least two versions as well, one with R5101 (the older one, which is odd) and the newer one (which I have as well) is SAD4096.

I did 2 changes, the 100K resistor on the output of the SAD512D, which is directly left of the socket, it was 22K stock. The other is the 22K resistor just below the trim pot which is the bias resistor for the output side of the NE571. Strangely this didn't change the bias on the circuit, and it seems a bit low to me, and perhaps the distortion is coming from the NE571 as its only about 5.81 volts with a 15 volt supply. Possibly similar to the FX90 problem I encountered.

I couldn't confirm that 22k resistor was in fact the biasing resistor for the NE571 on the BBD output (expandor).   The PCB traces are hard to follow.  Are you confident that's the correct resistor?

With the FX90, the supply was 9V and the 10k resistor incorrectly biased the NE571 output to 6.6V (or whatever it was) and we increased the 10k resistor to 39k to *lower* the bias voltage.

I'm not sure if your 5.81V is before or after the resistor change.    However, the original 10k resistor looked good to me for a 15V supply.  It should bias the NE571 output to 6.6V (same as original FX-90 voltage).  If you wanted to raise that voltage a tad you could use 8.2k, however it's probably not going to give you much improvement - as the BBD is the bottle neck (maybe even the opamp bias, as discussed earlier).   Lowering the resistor will raise the bias voltage - opposite to what we did with the FX-90

If you are getting distortion the most likely place tweak is the BBD bias pot.   However, if the BBD output is clean *and* the filter output (at the opamp) is clean, then maybe there is a fault around the NE571.   You can get distortion if the cap on pin 1 of the NE571 isn't connecting or the value is too small.

I am sure it's the correct resistor, but you're right I was tired and got it backwards. I'll try 8.2k or lower, I would think it should be around 7.5v. However it was 5.81 with either resistor which is a bit strange. I don't want to use my desoldering gun and soldering iron on the board too much, they are really cheap, but not as bad as EH boards. I'll try my decade box first and see if I can get a correct voltage out of it.

As for the electrolytics they are all new nichicons, either low esr audio grade in the signal path or 105C 5000 hour ones for power supply.

I do believe it's the SAD512D that's having problems, a shame, that was a new chip I bought, the other one was good that I put in my 670 Flanger. It's possible it just doesn't like this circuit, I don't know. I found these all to be really finicky unlike the rest of the hundreds and hundreds of pedals I've repaired over the last 35 years.

[Rob Strand]

I don't think there is an R5106 version, I have never seen those used until late Performer series in the Flanger or Chorus. There is also many circuit revisions on those, I have at least 3 different 565 boards and 2 different 575 boards. I have seen two different 585 boards so far too, one with SAD4096 and one with MN3005. The 680 has at least two versions as well, one with R5101 (the older one, which is odd) and the newer one (which I have as well) is SAD4096

I think you are right, the R5106 units I saw could be 565/575.  The available schematics for those show an SAD512, which should be an SAD512D.   If there was a change to use the R5106 it's probably in 1982, and there are 1983 units as well.

I looked through some easy to get to notes and it turns out I actually traced a DOD 565A back in 2003 - 20 years ago now, eek!   That one was 1983.  It used yet another circuit with an MN3007!  As I recall this one was called a stereo chorus.



Oh, read the notes on the schematic.  The unit I traced was built with the filters caps swapped!

Clearly DOD are making changes by the year!

I am sure it's the correct resistor, but you're right I was tired and got it backwards. I'll try 8.2k or lower, I would think it should be around 7.5v. However it was 5.81 with either resistor which is a bit strange. I don't want to use my desoldering gun and soldering iron on the board too much, they are really cheap, but not as bad as EH boards. I'll try my decade box first and see if I can get a correct voltage out of it.

Good to know.   It's odd the voltage isn't changing, even suspicious.

I do believe it's the SAD512D that's having problems, a shame, that was a new chip I bought, the other one was good that I put in my 670 Flanger. It's possible it just doesn't like this circuit, I don't know. I found these all to be really finicky unlike the rest of the hundreds and hundreds of pedals I've repaired over the last 35 years.

It's possible the clock shaping network on the original SAD512D's was very robust, that was the idea, but some of the recent day copies might use a weaker circuit making them sensitive to the clock pulse shape.  The DOD VCO produces a *very* narrow negative clock pulse which is perhaps too narrow for the replacement chips.   Moreover the clock-pulse barely swings to 0V as the CMOS chip is being pushed.  The reason being is the new chips might handle the max clock frequency but only for square clock signals - they didn't consider that some circuit use narrow clocks, which the original SAD512D's handled.

Something you can try is to increase the 330R resistor on the VCO.   You might need to go as high as 3k3 ohm but the idea is to change it just enough to get it working then add a bit of safety margin.   Start with say 1k, if no good try 2k2 then 3k3.  If 1k works try 680R.

It would help if you could listen to the audio to confirm exactly what point (BBD, Filter, Expandor) the audio becomes distorted.

[CheapPedalCollector]

Thank you for the schematic! I would love any more you might have.

I can take pics of the other boards I have, I can't be bothered to trace them. However be warned I've cannibalized the boards for caps and chips. I bought them in a lot, and I would have rebuilt them, but the pots and knobs and harness was missing for each of the four I got cases for, and nothing came with the delay board. I really didn't feel like making new harnesses with molex connectors. So they have been added to the parts bin.

Finding errors in their schematics and production units doesn't surprise me at all. I feel they were kind of like EH and just using whomever they could get, and filling boards with whatever they had a lot of the time. I suppose I can ask Tom Cram about that, but he stopped posting on the FB group after DOD was bought by Cort. I really wish the companies didn't gate keep all this stuff and release it to the public after all these years. I used to be able to get Schematics and parts from Dunlop but not anymore. BOSS recently stopped letting dealers purchase parts too unless they are external and now require you to send stuff to authorized repair centers.

So before I changed the resistor back, and rewired the AC power switch with a the one and removed my temporary DC jack, I took some pictures of wiring and the bottom side of the board. Posting them here for posterity. I haven't finished cleaning the flux off the board yet, apologies.







I feel adapting the circuit to MN3007 is a bit more trouble than it's worth and it wouldn't sound the same at all. The SAD512D was a unique device only needing a single clock input and having 512 stages only. If I can't get it working, I'll just buy some more chips or wait until I get another donor unit that isn't worth repairing. I'm intending to keep them for my collection, so I'm not in any real hurry.

[Rob Strand]

Thank you for the schematic! I would love any more you might have.

No problem.   I honestly don't know what I have.   Finding that one was a surprise but as soon as I opened up the file I remembered doing it and some of the things that came up at the time.   I've got a large stash of stuff but I deliberately made it convenient to access so I don't mess around with it.    I've got a CD around my desk from 2010 but I know it's nowhere near half of it.

I can take pics of the other boards I have, I can't be bothered to trace them. However be warned I've cannibalized the boards for caps and chips. I bought them in a lot, and I would have rebuilt them, but the pots and knobs and harness was missing for each of the four I got cases for, and nothing came with the delay board. I really didn't feel like making new harnesses with molex connectors. So they have been added to the parts bin

My days a numbered tracing PCBs.   I might end up doing one or two a year but its more the interest in something particular that draws me in rather than me chasing after it.

Finding errors in their schematics and production units doesn't surprise me at all. I feel they were kind of like EH and just using whomever they could get, and filling boards with whatever they had a lot of the time. I suppose I can ask Tom Cram about that, but he stopped posting on the FB group after DOD was bought by Cort.

I think a lot of pedal manufacturers worked the same way.  MXR seemed to be different, the way they did things gives me the impression some those guys came out of different industries.

I really wish the companies didn't gate keep all this stuff and release it to the public after all these years. I used to be able to get Schematics and parts from Dunlop but not anymore

Yes, it's very inconvenient for repairs.  Lets face it, no matter what companies do if someone is keen enough they will trace it and even make clones of it.

BOSS recently stopped letting dealers purchase parts too unless they are external and now require you to send stuff to authorized repair centers.

Is that what's happened with Boss.  I used to think their support was great.

So before I changed the resistor back, and rewired the AC power switch with a the one and removed my temporary DC jack, I took some pictures of wiring and the bottom side of the board. Posting them here for posterity. I haven't finished cleaning the flux off the board yet, apologies.

Yes, good to capture these things.

Something else that gets under the radar is the transformer voltages and the DC resistances of the transformer windings.   Small transformers are a headache since the regulation is poor a 12Vac transformer can measure 15Vac with no load.  Someone measures a 12Vac transformer at 15Vac then buys a 15Vac transformer and that produces 18Vac.  Then of course the internet propagates all the errors.

I feel adapting the circuit to MN3007 is a bit more trouble than it's worth and it wouldn't sound the same at all. The SAD512D was a unique device only needing a single clock input and having 512 stages only. If I can't get it working, I'll just buy some more chips or wait until I get another donor unit that isn't worth repairing. I'm intending to keep them for my collection, so I'm not in any real hurry.

At this point in time I agree, it's going to take some work.   Yes, the SAD512D was target towards making things easy and packing circuits in a small form factor.   It didn't last long since it got replaced by the R5106, which is similar and was popular for a while as well.   Boss and Ibanez tended to stay with the Panasonic devices - they had a culture to stick with the Japanese parts.

As a side point.  When I skimmed over a few random DOD schematics it looked like some of the units that converted to the MN3101/MN3007 etc era chips tried to emulate the old VCO.   No idea how close they work but you could see some intent there.

[CheapPedalCollector]

No problem.   I honestly don't know what I have.   Finding that one was a surprise but as soon as I opened up the file I remembered doing it and some of the things that came up at the time.   I've got a large stash of stuff but I deliberately made it convenient to access so I don't mess around with it.    I've got a CD around my desk from 2010 but I know it's nowhere near half of it.

I've been around lurking for ages, but I had a house fire in 2011 and lost a lot of stuff. I've recovered a lot through the way back machine, but I have no idea what I don't have anymore too. I still have quite a bit and friends who have shops send me stuff on request when I need it.

My days a numbered tracing PCBs.   I might end up doing one or two a year but its more the interest in something particular that draws me in rather than me chasing after it.

You and I both, I used to trace everything when I was young, but now I mostly want to make my own things again, so I'm more learning how things work to the best of my ability. I have a good working knowledge of a lot of things, the engineers really made it easy to do a lot of things. I think more in terms of features and how I want things to sound.

I think a lot of pedal manufacturers worked the same way.  MXR seemed to be different, the way they did things gives me the impression some those guys came out of different industries.

I agree, MXR are my favorite effects from the old times. They clearly knew how to design and layout properly and made stuff to last, and I don't find anything they made sounding bad, breaking down very often, and they tend to be quiet too without any corner cutting. At least the best that they could at the time with the tech that was available.

Is that what's happened with Boss.  I used to think their support was great.

Yes I am extremely disappointed at this decision by their management. It's almost as if as soon as right to repair passed here in the USA, they immediately started with the gate keeping. I've never been a fan of their products, but my friends and customers are, and I wanted to continue repairing them for them. Now when I run out of certain parts like jacks and pots, that's not going to be possible anymore as I dislike substituting.

Something else that gets under the radar is the transformer voltages and the DC resistances of the transformer windings.   Small transformers are a headache since the regulation is poor a 12Vac transformer can measure 15Vac with no load.  Someone measures a 12Vac transformer at 15Vac then buys a 15Vac transformer and that produces 18Vac.  Then of course the internet propagates all the errors.

I usually figure a transformer is rated 3-5 volts usually above the regulated voltage, unless I know the spec of a part.

At this point in time I agree, it's going to take some work.   Yes, the SAD512D was target towards making things easy and packing circuits in a small form factor.   It didn't last long since it got replaced by the R5106, which is similar and was popular for a while as well.   Boss and Ibanez tended to stay with the Panasonic devices - they had a culture to stick with the Japanese parts.

As a side point.  When I skimmed over a few random DOD schematics it looked like some of the units that converted to the MN3101/MN3007 etc era chips tried to emulate the old VCO.   No idea how close they work but you could see some intent there.

I think I can adapt this chorus very easily to R5106 by just changing the vreg to a 78L12 and popping the chip in. The 100K resistor on the output should be enough. I have a limited source for those and I will do that as a last resort.

[Rob Strand]

I've been around lurking for ages, but I had a house fire in 2011 and lost a lot of stuff. I've recovered a lot through the way back machine, but I have no idea what I don't have anymore too. I still have quite a bit and friends who have shops send me stuff on request when I need it.

It's a happened to a few people here.  I really need merge all my stuff and make two copies of *everything* - I used to make three copies of everything but started losing interest.   If you ever get stuck just ask, I might remember if I have specific things.   The post 2010 to 2018 stash is very inconvenient to access ATM.

I think I can adapt this chorus very easily to R5106 by just changing the vreg to a 78L12 and popping the chip in. The 100K resistor on the output should be enough. I have a limited source for those and I will do that as a last resort.

A good back-up plan just the same.  You will have to re-bias the NE571 perhaps even the compandor (input) side as well.

[CheapPedalCollector]

It's a happened to a few people here.  I really need merge all my stuff and make two copies of *everything* - I used to make three copies of everything but started losing interest.   If you ever get stuck just ask, I might remember if I have specific things.   The post 2010 to 2018 stash is very inconvenient to access ATM.

Thank you Rob, you're a champion.

A good back-up plan just the same.  You will have to re-bias the NE571 perhaps even the compandor (input) side as well.

Indeed, and I know how now easily enough thanks to your guidance.

[CheapPedalCollector]

Follow up on this.

The new Coolaudio V571Ds sound a bit different than the Sygnetics ones, they have more top end and they also have a bit more distortion it seems. I swapped an older chip in and the distortion in the background went away. I'm not sure if this is bias related, or the chip masks aren't as good quality, or different thd caps are needed. I noticed the one I just fixed for a friend had tantalum caps in those like the FX90 delay I fixed did, I'm wondering if ESR is very sensitive here.

I also had to replace the SAD512D and it works wonderfully now. I think I may tweak the output resistor to be 51k or so, and I may mess around with getting rid of the direct coupling to the 4559s with a 1uf cap and a resistor to VDD, say 470K or so. Hopefully it won't affect the sound and will keep the SAD512D working longer as it won't be supplying any current to the 4559. I can say that my RC4559s will bias up with just 2.4v, while any other modern 4558 or 4559 will not, which is also interesting.

I also noticed if you remove the SAD512D to test voltages it might be a bad idea, as the following opamp stage then has a few mv on the + input and a full 15 volts on the - input and output. I don't think that's good for the chip, and I remember reading something from RG Keen about that.

I may make similar mod to my 670 Flanger and 680 Delay once I get a working SAD4096 for that.

This dive also helped me fix a friends 690 Chorus in just a matter of a couple hours, so it was worth it.

[Rob Strand]

Follow up on this.

Good work!

I'll have to check the RC4559 datasheets to see if there's a reason the thing won't bias.   The output stage should be able to swing down to 2.4V, even if the swing is poor.  So I don't think it is that.  I'm suspicious it's got something do with the common mode range of the inputs.