Dumbloid  [traced]

General documentation, gut shot, schematic links, ongoing circuit tracing, deep thoughts ... all about boutique stompboxes.
Manfred
Transistor Tuner
Posts: 1074
Joined: 05 Apr 2009, 00:42
Has thanked: 918 times
Been thanked: 657 times

Post by Manfred »

Based on Bob:
"The answer is blowin' in the wind." :?

Ripdivot
Resistor Ronker
Posts: 318
Joined: 19 Jan 2008, 20:15
Location: Canada
Has thanked: 8 times
Been thanked: 68 times

Post by Ripdivot »

1oldsidewinder wrote:What I really find curious are the posts where guys have tried every possible combination of clipping diodes and it apparently hasn't helped. It seems like converting the pedal to diode clipping would solve the problem if the rattle was indeed a product of op-amp distortion. If someone was to replace the accent control network with a 1k resistor and add a pair of LEDs to the first stage, it would be a CJOD with a slightly different voicing; a pedal not known for *farty* decay. If I were to pursue this, I'd continue to look at the many other changes the designer made. Either that, or build a CJOD and voice it like the Dumbloid.
I gave this a try, 1k in place of the attack pot and LEDs in the feedback loop of the first stage and it makes a huge improvement as far as the fizzy decay goes. The problem is the circuit looses a lot of its vocal quality although it still sounds really good. The fizz is still there a tiny bit but it's totally usable IMHO. I'll make a recording in the next couple days.

Ripdivot
Resistor Ronker
Posts: 318
Joined: 19 Jan 2008, 20:15
Location: Canada
Has thanked: 8 times
Been thanked: 68 times

Post by Ripdivot »

By the way I left the single LED in the second stage. It only conducts when you turn the tone pot way up.

1oldsidewinder
Posts: 42
Joined: 01 Mar 2011, 15:37
Has thanked: 14 times
Been thanked: 5 times

Post by 1oldsidewinder »

Ripdivot wrote:
1oldsidewinder wrote:What I really find curious are the posts where guys have tried every possible combination of clipping diodes and it apparently hasn't helped. It seems like converting the pedal to diode clipping would solve the problem if the rattle was indeed a product of op-amp distortion. If someone was to replace the accent control network with a 1k resistor and add a pair of LEDs to the first stage, it would be a CJOD with a slightly different voicing; a pedal not known for *farty* decay. If I were to pursue this, I'd continue to look at the many other changes the designer made. Either that, or build a CJOD and voice it like the Dumbloid.
I gave this a try, 1k in place of the attack pot and LEDs in the feedback loop of the first stage and it makes a huge improvement as far as the fizzy decay goes. The problem is the circuit looses a lot of its vocal quality although it still sounds really good. The fizz is still there a tiny bit but it's totally usable IMHO. I'll make a recording in the next couple days.
I'll be looking forward to hearing it. I've built some CJOD's but never with the Dumbloid voicing. The 'loid already has a 1k minimum gain resistor in series with the accent pot so you're hearing the same thing as accent set at zero. Makes sense the 3mm LED's response is weak as it's bypassed by the 1k resistor. More resistance gives it a greater role as accent is increased.

Manfred
Transistor Tuner
Posts: 1074
Joined: 05 Apr 2009, 00:42
Has thanked: 918 times
Been thanked: 657 times

Post by Manfred »

I'll be looking forward to hearing it. I've built some CJOD's but never with the Dumbloid voicing. The 'loid already has a 1k minimum gain resistor in series with the accent pot so you're hearing the same thing as accent set at zero. Makes sense the 3mm LED's response is weak as it's bypassed by the 1k resistor. More resistance gives it a greater role as accent is increased.
A further point is the applied 3mm LED.
The manner of clipping is dependent on the LED nominal current,
The knee voltage of about 1,8 Volts is exceeded at 20 Milliamps using a standard LED
and at 2 Milliamps using a low current LED.
Who knows which LED type is applied?

I suppose the idea behind that circuit was to imitate the clipping of a warm biased triode tube stage.

1oldsidewinder
Posts: 42
Joined: 01 Mar 2011, 15:37
Has thanked: 14 times
Been thanked: 5 times

Post by 1oldsidewinder »

Ripdivot wrote:By the way I left the single LED in the second stage. It only conducts when you turn the tone pot way up.
Ripdivot, would it be too much trouble for you to also do a clip with the LED lifted? Aside from leaving the 3mm LED in place, you've pretty much reduced the Dumbloid to a YATS with what are now common mods. If it still rattles without the LED, it might point to other areas we've overlooked.

Ripdivot
Resistor Ronker
Posts: 318
Joined: 19 Jan 2008, 20:15
Location: Canada
Has thanked: 8 times
Been thanked: 68 times

Post by Ripdivot »

1oldsidewinder wrote:
Ripdivot wrote:By the way I left the single LED in the second stage. It only conducts when you turn the tone pot way up.
Ripdivot, would it be too much trouble for you to also do a clip with the LED lifted? Aside from leaving the 3mm LED in place, you've pretty much reduced the Dumbloid to a YATS with what are now common mods. If it still rattles without the LED, it might point to other areas we've overlooked.
First off I added a 1k resistor in place of the attack pot so it's like the pot was down most the way but not all. So to be clear you would like to hear one clip with 2 LEDs in the first stage and the stock LED in the second stage and the another clip with the LED in the second stage removed? I don't mind doing the clips at all.

1oldsidewinder
Posts: 42
Joined: 01 Mar 2011, 15:37
Has thanked: 14 times
Been thanked: 5 times

Post by 1oldsidewinder »

Ripdivot wrote:
1oldsidewinder wrote:
Ripdivot wrote:By the way I left the single LED in the second stage. It only conducts when you turn the tone pot way up.
Ripdivot, would it be too much trouble for you to also do a clip with the LED lifted? Aside from leaving the 3mm LED in place, you've pretty much reduced the Dumbloid to a YATS with what are now common mods. If it still rattles without the LED, it might point to other areas we've overlooked.
First off I added a 1k resistor in place of the attack pot so it's like the pot was down most the way but not all. So to be clear you would like to hear one clip with 2 LEDs in the first stage and the stock LED in the second stage and the another clip with the LED in the second stage removed? I don't mind doing the clips at all.
Thanks for the clarification. Ideally, I'd like to hear it with just a single 1k and no LED like a TS. That would either point a finger at the second stage mods or eliminate them.

Ripdivot
Resistor Ronker
Posts: 318
Joined: 19 Jan 2008, 20:15
Location: Canada
Has thanked: 8 times
Been thanked: 68 times

Post by Ripdivot »

1oldsidewinder wrote:
Ripdivot wrote:
1oldsidewinder wrote:
Ripdivot wrote:By the way I left the single LED in the second stage. It only conducts when you turn the tone pot way up.
Ripdivot, would it be too much trouble for you to also do a clip with the LED lifted? Aside from leaving the 3mm LED in place, you've pretty much reduced the Dumbloid to a YATS with what are now common mods. If it still rattles without the LED, it might point to other areas we've overlooked.
First off I added a 1k resistor in place of the attack pot so it's like the pot was down most the way but not all. So to be clear you would like to hear one clip with 2 LEDs in the first stage and the stock LED in the second stage and the another clip with the LED in the second stage removed? I don't mind doing the clips at all.
Thanks for the clarification. Ideally, I'd like to hear it with just a single 1k and no LED like a TS. That would either point a finger at the second stage mods or eliminate them.
OK, I'll do one clip as is and one the way you would like to hear it. It's seems the farty decay problem is caused by the attack pot being turned up. I have tried every diode combination I could think of in both stages separately and at the same time and even no diodes and none of that worked. I always had the attack pot up at about 50% though.

mmolteratx
Degoop Doctor
Posts: 510
Joined: 19 Dec 2009, 02:50
Has thanked: 95 times
Been thanked: 587 times

Post by mmolteratx »

Well, increasing the value of the attack pot increases the gain of the second stage, so not surprised that it exacerbates the problem.

Ripdivot
Resistor Ronker
Posts: 318
Joined: 19 Jan 2008, 20:15
Location: Canada
Has thanked: 8 times
Been thanked: 68 times

Post by Ripdivot »

mmolteratx wrote:Well, increasing the value of the attack pot increases the gain of the second stage, so not surprised that it exacerbates the problem.
Yes I agree, it's obvious that the attack pot increases the second stage gain which in turn causes the op-amp to clip harder which makes the farty decay worse. As I said above though when you remove the attack pot it changes the voicing quite a bit. It still sounds good but it looses a lot of the vocal quality that it has in its stock form.

1oldsidewinder
Posts: 42
Joined: 01 Mar 2011, 15:37
Has thanked: 14 times
Been thanked: 5 times

Post by 1oldsidewinder »

This is kind of interesting. I did a little shoot-out last night between a Dumbloid-mod SD-1 and a GGG Red Llama clone and was able to get very similar tones, using the guitar's tone control to trim the highs a bit on the RL. The D-mod Sd-1 was a little more sensitive to pick attack but the RL had a nicer decay. While I never considered the RL to be a "D" pedal, a tweaked RL/TSF might be an option for someone who likes the Dumbloid's basic character but is put off by it's quirks.

Ripdivot
Resistor Ronker
Posts: 318
Joined: 19 Jan 2008, 20:15
Location: Canada
Has thanked: 8 times
Been thanked: 68 times

Post by Ripdivot »

Ripdivot wrote:
1oldsidewinder wrote:
Ripdivot wrote:
1oldsidewinder wrote:
Ripdivot wrote:By the way I left the single LED in the second stage. It only conducts when you turn the tone pot way up.
Ripdivot, would it be too much trouble for you to also do a clip with the LED lifted? Aside from leaving the 3mm LED in place, you've pretty much reduced the Dumbloid to a YATS with what are now common mods. If it still rattles without the LED, it might point to other areas we've overlooked.
First off I added a 1k resistor in place of the attack pot so it's like the pot was down most the way but not all. So to be clear you would like to hear one clip with 2 LEDs in the first stage and the stock LED in the second stage and the another clip with the LED in the second stage removed? I don't mind doing the clips at all.
Thanks for the clarification. Ideally, I'd like to hear it with just a single 1k and no LED like a TS. That would either point a finger at the second stage mods or eliminate them.
OK, I'll do one clip as is and one the way you would like to hear it. It's seems the farty decay problem is caused by the attack pot being turned up. I have tried every diode combination I could think of in both stages separately and at the same time and even no diodes and none of that worked. I always had the attack pot up at about 50% though.
Sorry this took so long but hear are the sound clips I promised. Same setup as my previous clip as far as gear goes.

Part 1 - Attack pot shorted out, 2 x 3mm red LED's in first stage, stock LED in second stage.

Part 2 - Attack pot shorted out, 2 x 3mm red LED's in first stage, LED removed from second stage.

The pedal is still quite dynamic but not as dynamic as the stock version. The farty decay is much better, not perfect but much better. The vocal like voicing of the stock version is not quite there but is still sounds very good, especially when cranked up.

https://soundcloud.com/ripdivot/dumbloid-modded-leds

1oldsidewinder
Posts: 42
Joined: 01 Mar 2011, 15:37
Has thanked: 14 times
Been thanked: 5 times

Post by 1oldsidewinder »

It's pretty clear that the second stage LED isn't doing much at the minimum accent setting of 1k (stock TS value). Why not put the first stage LED's on a switch so you could get both the "stock" Dumbloid sounds as well as the smoother CJOD tones from your pedal. You could even do a 3-way like the Landgraff or Boiling Point; silicon array on one side, LED on the other, and diode lift in the center position. That would make for a really versatile YATS. Thanks again for the clips!

Ripdivot
Resistor Ronker
Posts: 318
Joined: 19 Jan 2008, 20:15
Location: Canada
Has thanked: 8 times
Been thanked: 68 times

Post by Ripdivot »

1oldsidewinder wrote:It's pretty clear that the second stage LED isn't doing much at the minimum accent setting of 1k (stock TS value). Why not put the first stage LED's on a switch so you could get both the "stock" Dumbloid sounds as well as the smoother CJOD tones from your pedal. You could even do a 3-way like the Landgraff or Boiling Point; silicon array on one side, LED on the other, and diode lift in the center position. That would make for a really versatile YATS. Thanks again for the clips!
I thought about doing this pedal with switching options but I have other pedals that I like better so I probably won't go any further with it. It was fun to see what I could get out of it though. Glad you found the clips useful.

Jonotron
Posts: 12
Joined: 03 Aug 2013, 10:33
Has thanked: 1 time

Post by Jonotron »

Any advantage in having the 22uf output cap as opposed to the usual 10uf? More low end? Seems quite a high value..

Ripdivot
Resistor Ronker
Posts: 318
Joined: 19 Jan 2008, 20:15
Location: Canada
Has thanked: 8 times
Been thanked: 68 times

Post by Ripdivot »

Jonotron wrote:Any advantage in having the 22uf output cap as opposed to the usual 10uf? More low end? Seems quite a high value..
It makes no difference really. With the 22uf the corner frequency is .723hz and with a 10uf it would be about 1.6hz. Both are well below guitar frequencies.

Jonotron
Posts: 12
Joined: 03 Aug 2013, 10:33
Has thanked: 1 time

Post by Jonotron »

I don't think even a church organ can hit those frequencies! Thanks for that info :D

ipm
Posts: 11
Joined: 06 Jun 2012, 23:23
Been thanked: 1 time

Post by ipm »

So, no new ideas?

MoonWatcher
Diode Debunker
Posts: 721
Joined: 28 Jul 2008, 13:27
Has thanked: 37 times
Been thanked: 43 times

Post by MoonWatcher »

Ripdivot wrote:1 - Attack pot shorted out, 2 x 3mm red LED's in first stage, stock LED in second stage.

2 - Attack pot shorted out, 2 x 3mm red LED's in first stage, LED removed from second stage.

The pedal is still quite dynamic but not as dynamic as the stock version. The farty decay is much better, not perfect but much better. The vocal like voicing of the stock version is not quite there but is still sounds very good, especially when cranked up.
Sorry that I stopped following this discussion. I think I wrote off the Dumbloid as being just a crude and accidental design, and probably minimized any potential good qualities it might have. I think part of this probably stems from the expectations of the average person here versus the average person at tgp. Maybe our expectations aren't very different, but the way we view them is probably fundamentally different.

If it's really that the stock version seems to have some quality that gets nullified with mods, why not try raising the supply voltage?

I know that at this point something like a charge pump might not be a worthy or fun experiment, but it could be what gives the smallest (if any) compromise.

I think the biggest trick will be determining what voltage will retain the biggest percentage of stock values, but minimize the decay to the point where it's subtle enough not to notice it - unless you're purposefully listening for it.

So that means that it's not just to add a charge pump - the supply voltage will have to be adjustable - the best results might come at 9.5v, or 10v, or basically something less than 18v.

I had another idea. While it would be a big deviation from stock, it seems that no one has yet to try clipping diodes shunted to ground. I'm not talking about anything traditionally strong like with a Guv'nor or something, but I think using LED's with a higher conduction point could work out well. I thought a big part of the Suhr Riot design was not that clever until someone re-analyzed the clipping diodes, and found that high-intensity blue types were used. I put one of these in a personal Guv'nor build, and found it to be very good sounding in a counter-intuitive way. You can even keep most of the stock Guv'nor tone circuits and have a much more open and what is probably considered to be vocal quality. There was another thread here where someone mentioned to increase the tone pot values, and I think I did that as well. But I'm hoping that tone circuit issues with the Dumbloid are kind of a moot point, or they are actually enjoyed despite any quirky accidental design.

So if stock really does sound best except for the farty decay, I think adjustable supply voltage could nix a lot of that with no need to play with clipping diode variations (in negative feedback loops) at all. Is there going to be perfection? I doubt it. As has been mentioned many times, this is really the nature of almost any op amp. Diodes seem to nullify the issues of amplifier clipping at the transitional point using traditional supply voltages. And 9v is actually on the low side if there are no clipping diodes to aid as clamps to the rails, right?

So the obvious 2nd step would be clippers to ground with an obnoxiously large threshold before clipping. Like some of the smarter designs I've seen, you can even use LED's at the input - so not exactly as for clipping, but for processing the peaks of the signal that is input. You could even do some calculations if you know the output (in mv) of your pickups - you'd want something right at the precipice of their range. Some would say that this only deals with the largest signals, but I think you need to consider the entire range for best results, and just be gentle with what you process away.

The next step is to determine if you need shunted clipping on the output of just ic1a, or also ic1b. If using the stock attack mechanism that essentially makes the tone circuit also a gain circuit, I think it's clear that you'd need to try shunted clipping at both. I also wouldn't be surprised if ic1a would benefit from more clamping, and ic1b from less. Just start with some high-intensity blue types or something like them, and work up or down from there.

It might also help to establish some "base case," so you can keep yourself from going mad. Try and decide on how much nasty decay you can live with. If the final results are slightly above or below that, I'd personally declare victory and then just get on with life again.

I think I'll end with this - I built |V|ark's vero layout of the Tech 21 XXL some time back. That one has no clipping diodes either. It works pretty well at circumventing the decay issues, probably in part because of the warp control. But even with that set to where it is basically negligible, the decay isn't really a big deal. But what I found to make it sound 200 times better was simply increasing the supply voltage, and then it became quite good. The circuit design itself may not be very appealing compared to more recent overdrive or distortion designs, especially whatever seems to be occurring with the Dumbloid.

So it seems like this boils down to 3 choices:

1) handle the stock limitations (as Ripdivot and others have done) and try to retain the qualities through primarily trial-and-error

2) increase the supply voltage - but make it adjustable - so not just 18v

3) address the clipping on the output of the op amp stage(s) with mild clipping setups to ground

Maybe the best results could be obtained with a combination of 2 and 3, but that depends how much time and trouble someone wants to go to.

Post Reply