Identify DM-3 layout for me

[Mark Hammer]

Trying to catch up on partially-finished projects this long weekend, I resurrected a DM-3 clone that I had started (thanks again for the chips, Sean!  I made 4 boards, and don't need the others, so call me if you want one).  Finally wired it up, tweaked the trimpots, and fired it up.  It appears to work fine except for one fatal flaw: the echo is much less robust than I would expect, and the recirculation is pretty much absent.  There is always the very real possibility that *I* screwed up somehow, but I don't seem to have any documentation connected with the PCB layout I used, and the parts are not legended on the layout drawing in any manner corresponding to the original Boss schematic.

So, anyone know the origins of this layout, and whether it has been tested and confirmed? (or appended for that matter)

My deep appreciation to the person who drafted it....whoever you are (I gotta start taking notes on these things).

[zeeman]

Sorry, I have not seen this layout.

What documentation do you have showing how to adjust the trim pots?

zeeman

[Mark Hammer]

Sorry, I have not seen this layout.

What documentation do you have showing how to adjust the trim pots?

zeeman

I don't have any.  I just know how to adjust them.  The 22k is adjusted until one gets the strongest and cleanest delay signal.  The 10k is adjusted until the clock whine is minimized, and the 100k is adjusted to get the best tradeoff between delay and audio quality.

[zeeman]

Sorry, I have not seen this layout.

What documentation do you have showing how to adjust the trim pots?

zeeman

I don't have any.  I just know how to adjust them.  The 22k is adjusted until one gets the strongest and cleanest delay signal.  The 10k is adjusted until the clock whine is minimized, and the 100k is adjusted to get the best tradeoff between delay and audio quality.

Thanks.

I take it you do not recall if this layout was verified - correct?

Have you traced the layout to the schematic and checked for errors?

zeeman

[Dan N]

Sorry, I don't recognize that layout.

Fooling around with it I can see two things-

The 47K resistor between pins 1 and 6 of the 4558. With the original FET switching, I think that guy would be disconnected by an open gate when the effect was in, um, effect.

The cap going to the center of the Intensity pot should be 0.1 (not 1uf). It's probably a tant in the Boss pedal.

These are the only things I could find.

[quarara]

http://www.diystompboxes.com/smfforum/index.php?topic=63923

[Mark Hammer]

Thanks!

I used the 2SC945 (as noted on the original schematic) on mine since I had a bunch of them sitting around.  I needed to slip some insulation on the middle pin and twist two of the pins around to fit them on the layout.  Possible that's where my problem is.  I replaced a few with some 2N2222s that I have, but there was no improvement in wet levels.  I have some BC550s sitting around too.  I may try those out and see.

Exasperating.  I'd like to add a few more tonal options and box this up, but until the level thing gets settled, it's simply a board with stuff hanging off of it.

Ry asks at the end of the linked-to thread: "What's the difference between the DM-2 and the DM-3?"

The answer is that, on the audio path, not a whole heckuva lot.  The 2nd issue DM-2, like the DM-3, also uses an MN3205.  The principal difference on the audio path is that the DM-3 includes a FET input stage, before the bipolar transistor stage.  Other than that, the major differences are in the control of the clock circuit (The DM-3's is a little more involved, presumably to achieve better dialability, and perhaps stability, of delay time), and the regulation of supply voltage.

[MetalGuy]

Hi,

This is my layout. Actually I used an already existing layout of DM-2 and adapted it for DM-3. Check out below how it happened:


http://www.diystompboxes.com/smfforum/index.php?topic=57928.0

http://www.diystompboxes.com/smfforum/index.php?topic=63923.msg507190#msg507190

[Mark Hammer]

First, thanks for the layout.  Easy to etch and stuff.  And, as noted, it worked first time out.  It just doesn't "work" as expected.

The confusing part for me is the way the power is implemented, and which transistors are which.  I gather what you have on the board is some sort of compound of the DM-2 and DM-3?  The DM-3 schematic shows 7 transistors in the circuit, excluding all of those used for switching purposes.  Your layout has 5. The JFET input buffer is clearly missing, but beyond that, I'm a little confused and too backed up to do the detective work.

Please don't take this at all as an insult, but given that it is something you posted a while back, (and whose details may not immediately spring to mind as a result), I'm wondering if the best course of action might not be to liberate the components from this board and apply them to an AD-3208?  To the best of my knowledge, that board works exactly as expected.  Moreover, since I have a pair of V3205s, I could squeeze some serious delay out of it.

Not unless you have some ideas about what the problem might be.  Just for the record, you have built this successfully in the sense that you CAN get multiple repeats that are clearly audible?  Again, not meant as an insult.  I am deeply appreciative of the effort you put in.  It's just that I have such a backlog of projects, I'm looking for the path of least resistance.  Any guidance you can provide is appreciated.

[MetalGuy]

It just doesn't "work" as expected.

Honestly I don't even know how it is expected to work because I've never used DM-2 or DM-3 before.

The DM-3 schematic shows 7 transistors in the circuit, excluding all of those used for switching purposes.  Your layout has 5. The JFET input buffer is clearly missing, but beyond that, I'm a little confused and too backed up to do the detective work.

In order to simplify the layout input FET was omitted intentionally - I used a "regular" transistor input buffer similar to the ones found in many pedals.
Direct output transistor was also omitted and the 7.5V regulator one was replaced by a zener diode that's why there are only 5 transistors left.

Please don't take this at all as an insult

I would appreciate any comments especially those helping to improve layout and sound.

Just for the record, you have built this successfully in the sense that you CAN get multiple repeats that are clearly audible?

If you can suggest some settings I'll try to record a sample.

[Mark Hammer]

"Expected" simply means that one can dial in an echo that is almost as loud as the clean signal, and that one can also dial in more than a single audible repeat.  In other words, what one can get from basically any analog delay line.  At present, it behaves as if both the Intensity and Echo controls are turned down about 3/4 or more.  The absence of the FET input stage is no great loss in my case since I know the pedal will get a buffered signal, which means it was a wise decision on your part.

I thought I had a bunch of BC550s lying around, but it was actually some 2N4401s and BC549s.  Still, I think I'll replace the ECB units I have in there with some EBC units just tobe certain that the "crossed leg" devices aren't screwing things up.  That, and the obligatory joint reflow and check for PCB solder-side errors/flaws, and I'll get back to you.  In the meantime, thanks for your willingness to help.   

[MetalGuy]

Below is a quick sample. Chain is Dr. Boogey -> Marshall cab simulator -> DM3 clone -> PC line in.

http://www.musicuploader.org/MUSIC/4899671239892974.mp3

Settings are:

Time - 2 o'clock
Repeats 12
Mix 12

I don't know if it's normal but when turning Time past 3 o'clock it starts feeding back.

I hope it helps. Sorry for the crappy playing.

[Mark Hammer]

Thanks.  yep, that's exactly what I'm not getting....at any setting.

[Mark Hammer]

An update.

Yesterday I finally started probing the signal level at different points in the circuit.  At first, I thought it was a bad NE570 compander chip because the output from the compressor side was a little over 150mv when I strummed, yet the output of the expandor side was only 6mv or so for that same strum.  Changed the 570 and no change.

So, I started probing other signal levels.  The signal coming out of the BBD was also in the 150mv range, so that was good.  Clearly, the "problem lay somewhere between the BBD and the expandor half of the 570.  I had changed most of the transistors from C945 or 732 to some metal-can 2N2222As....except for one, which is a plastic MPS2222a.  Hmmm, I wonder if it has the same pinout?  Lets unsolder that puppy and confirm.  Oops, the E and C pins are opposite to what I thought they were.  Re-installed that transistor the right way, and bada-boom-bada-bing, echo and runaway feedback all present and accounted for. 

So, the layout is confirmed to work by yet one more satisfied customer.  Personally, I found the 100k trimpot associated with the clock circuit more difficult to set by ear than either the bias or BBD output-balance trimpots.  I also found there was simply too much bass in the wet signal for my tastes so I replaced the 1uf cap going to the Echo and Intensity pots with an 82nf unit. helps a lot with managing potential overloading of the BBD by recirculated signal too.

So, now I can say it: Thanks, Metalguy!

[MetalGuy]

Hmmm, I wonder if it has the same pinout?  Lets unsolder that puppy and confirm.  Oops, the E and C pins are opposite to what I thought they were.  Re-installed that transistor the right way, and bada-boom-bada-bing, echo and runaway feedback all present and accounted for.

I suspected such a minor problem but it was up to you to find out.

As I mentioned before when Time pot is past 3 o'clock I'm getting sort of feedback similar to the one when Repeats pot is turned to the max. Do you have the same going on with your build?