Aion Amethyst (Boss DM-2) - Distorted Repeats

[MattG]

I recently built the AionFX Amethyst (Boss DM-2); see my build report.

I touched on a bit of this in the build report, but basically: I found the bias trimpot to be very sensitive, so replaced it with a multi-turn trimpot. I initially biased/calibrated by ear, as I didn't have an oscilloscope. I thought I got it to a pretty good place. I then used the front controls to try and set it similarly to how I have my Boss DM-2w, a fairly quick slapback with only two or three repeats. It sounded pretty good with light strumming. But then I went back and forth between it and the DM-2w, and realized there was fairly noticeable distortion on the repeats.

I assumed my by-ear calibration was to blame, so I ordered a cheap oscilloscope (DSO154Pro). I received the scope last night, and went through the calibration process again. I was a bit impressed with my by-ear calibration, as it was surprisingly close to optimal. I tweaked it a bit anyway. I was also able to do the clock cancellation adjustment, as that requires a scope (previously I just set the trimpot to noon). Though that trimmer only required a minute adjustment.

After all that, I gave it another test - and the distorted repeats are still there. If I play really lightly, I don't notice, so it seems to maybe be a headroom issue?

I used the (substantially cheaper) V3205/V3102 chips and am running at 9v.

First thing I plan to do is repeat the calibration process again. But if that doesn't work, I'm wondering what the next steps would be? The pedal fundamentally works, so I don't think I made any egregious error. From my very rudimentary understanding of BBD delays, the compander and filtering are crucial. Are there particular components in this part of the circuit that, if wrong or out-of-spec would be more likely to cause distorted repeats? Also, the feedback control seems too sensitive to me - it will go into self-oscillation much sooner than my DM-2w (moreso than I would expect as a result of component tolerances).

Hoping some other folks out there have some experience addressing this kind of problem on DIY BBD delays.

 

[owlexifry]

assuming you already have, but if not, might be relevant to check/measure voltages that the chips are seeing?

 

[MattG]

assuming you already have, but if not, might be relevant to check/measure voltages that the chips are seeing?

I'm ashamed to admit it, but I haven't done that yet. Thanks!

 

[MattG]

Small update: the PCB-only documentation does not have voltages. But thankfully he just released a full kit for the Amethyst project, and the kit build documentation does have voltages.

Armed with that info, I went to work with the DMM. Most voltages seemed reasonable to me, except for a few pins on the NE571. I then went to study the schematic to maybe get a hint as to what might be wrong. And then I re-measured the few pins that were off, and this time they looked OK. I kept re-measuring, turning the pedal on and off, messing with the knobs, and still most measurements looked OK. Most likely I managed to botch the measurements on the first pass. However, on the chance that the NE571 has some gremlins inside, I replaced it. I went through the calibration process again.

Now it's definitely better. There's still a bit of distortion on the repeats, but it's no longer blatantly obvious. It still feels like a headroom thing, because when I play softly or only single notes, I can get it to sound nearly identical to my DM-2w. But big chords and or hard-picking will still induce a little distortion on the repeats.

At this point it's hard to say what caused the improvement, since I changed multiple variables (NE571 replacement and re-calibration). But at least it's trending in the right direction. I'm tempted to build another and splurge on the MN3205 and run it at 12v.

 

[blackhatboojum]

When you say “distorted”, it’s hard to really understand what you mean because your definition of distortion might be different from mine. I say that because, the dm-2 is known for having “dirtier” repeats than other analog delays from that time period. It’s part of that circuits charm and just the nature of the beast. In my experience with the dm-2 though, excess grit or dirt on the repeats is indicative of an improperly biased bbd. Another thing to keep in mind, comparing the Aion version to the DM-2W, is like comparing Granny Smith to golden delicious apples. Sure, they’re both apples but they have different tastes. The amethyst is a straight up clone of the original dm-2. The boss DM-2W is not. There are 3 BBDs and 3 clock chips in the dm-2w so, there are different things going on electronically inside the Waza version. I’ve not seen any schematics for the dm-2w so, who knows what changes and/or other improvements boss has made to the circuit. I’m also really curious how the whole standard/custom thing is actually implemented in the dm-2w. Does that switch in standard mode just use one BBD or, does it switch in a resistor that limits the range of the delay pot while using all 3 BBDs? If the latter is the case, then the repeats are going be cleaner sounding because, the load so to speak, is shared by 3 BBDs vs. one. I dunno, hopefully my rambling hasn’t confused you more and some of what I said makes sense.

 

[MattG]

@blackhatboojum - that makes perfect sense and is extremely helpful, thank you!

You spoke exactly to my suspicions, which is that using my DM-2W for comparison probably isn't fair. But it's all I have!

As for the "distorted" repeats: probably a fool's errand to try an describe this verbally, but I never said I wasn't a fool.

Starting with the DM-2W: you said "the dm-2 is known for having “dirtier” repeats than other analog delays from that time period", and I've read similar comments from others in gear forums. I've never heard an actual DM-2, but at least with the Waza version in standard mode, I've never heard anything that I would call "dirt" in the repeats. I can hear that the highs are rolled off, but it sounds very "natural" (which I realize is so subjective as to be meaningless). But it makes me think of the "natural" echo you get if you were to e.g. yell across a big valley.

As for my Amethyst build, the distortion I'm hearing in the repeats doesn't sound like anything I'd ever call "natural". It doesn't sound like any deliberate overdrive or distortion sound I've heard from a guitar from any pedal or amp. (I suppose there could be a niche fuzz that sounds similar to what I'm hearing.) It's more akin to the crackling "static" sounds you get when the radio or old broadcast TV isn't tuned correctly. To put another way - and I'm speculating here, but - I doubt many people would find this particular sound charming; I suspect most people would say "that doesn't sound right". (At least those of us who grew up in the analog broadcast times, who have been conditioned to associate that static sound with something being not right.)

Anyway, I'll continue to play with the bias. With that multi-turn trimmer I installed, I can make what feels like pretty big changes, but see nothing change on the scope.

Thanks again for your thoughts, that's exactly what I was hoping for. Your Noire DM-2B looks absolutely amazing!

 

[blackhatboojum]

You’re welcome and I appreciate the compliments. In regards to what you’ve described with the repeats, I think I know what you’re talking about. I myself describe it as white noise and I hate to say this again but, that’s the nature of the beast with analog delays. I’m not saying there isn’t something wonky going on with your build but, it is a thing with every analog I’ve ever played through in my experience. Some circuits worse than others. For example… the DM-2 has more of that going on than the DM-3 but not as much as the Memory Man in my experience. Again, not saying there isn’t something wrong with your build but keep in mind that these things can be a factor. The OG DM-2 is a barebones analog delay circuit, with minimal filtering, and a BBD pushed beyond datasheet specifications so, there might be some things that a person may not like about it.

If you have the ability, I suggest using an audio probe in conjunction with your o-scope when adjusting the bias and cancel trimmers. The visual and audible reference, at the same time, might do you wonders on getting your build calibrated properly.

 

[Noodles]

Feature, not a bug (I'll see myself out).

I've had two original DM-2 and one DM-3. There was low headroom on the delays. They got pretty washy. I even got my EE buddy to put one on a pro scope and that didn't change. Set to a faux-reverb, I loved the sound from those pedals. Subtlety is their strength.