Tone Bender Mk I - attack pot

Started by Arangonix, February 09, 2021, 05:50:31 PM

Previous topic - Next topic

Arangonix

Hi guys,

maybe you'll drop me an advice one more time ;) I came back to a project that I left over 6 months ago and it is beloved/hated TB Mk I.

First thing first, if I remember correctly I used Sola Sound schematic:



In my version I used russian GT309B trannies, which I have dozens. Q1 = 80hfe, Q2 = 123hfe, Q3 = 83hfe.
In Q1 I have stock resistors + I added 470k resistor between base/collector to fake some leakage. This trannies don't have leakage nearly at all. With fresh 9V battery my voltages are: E = 3.70V, B = 3.27V, C = 9.16V.
In Q2 I also have stock resistors but I used a bit bigger, 2k2 resistor between base/pot. It gives me (pot min/max): E = 0V/0V, B = 0.04V/0.25V, C = 9.15V/5.63V.
In Q 3 I have trimmer to base, rest is stock + I have 100k resistor between base/collector to fake some leakage. I keep 7V in collector, the rest are: E = 0V, B = 0.15V.

Overall, it works fine, I should say very good. Classic Ziggy Stardust vibe with plenty of sustain and ocasional gateing. What bothers me here is attack pot. If I have it on 0, the effect doesn't produce any sound (even with Level pot fully open), only some overdriven amp hiss. If I opens it up to more less 1 - 1.5 (yup, I have knobs with scale :D) it wakes up, giving quiet, farty sound of heavilly missbiased trasistor. If I go a bit higher, 1.5 - 2, it jumps out with beautifull, roaring distortion. After that the attack pot works fine, brightening and tightening up the sound while opening it up.

I must say I've never had original MK I nor a decent clone. But I think attack pot shouldn't works like that. Any suggestions?

antonis

#1
Start by making R4 5k or so..

Alternatively, you can add a 5k resistor in series between R4 and R5 wiper or between R5 and GND..
"I'm getting older while being taught all the time" Solon the Athenian..
"I don't mind  being taught all the time but I do mind a lot getting old" Antonis the Thessalonian..

Rob Strand

#2
Tweak R4 like Antonis says until you get about 0.2V to 0.3V *across* Q2's collector resistor (R6).

With Attack full Q2's collector voltage should end-up around 5V.

Typically voltage *across* R8 maybe 1.2V.

I can't remember the exact numbers but those should be close.
Send:     . .- .-. - .... / - --- / --. --- .-. -

Reg18

#3
I can't add any advise as I'm an absolute novice, but this thread did inspire me to bread board a Mk1 based on the schematic above and it sounds great with a 5k instead of 50k.
What makes the sound/gain change so much between transistors. I went through a few and the sound varied dramatically, I settled on a Q1 AC128, Q2 2SB267, Q3AC128.
Tried some Toshiba 2SA53, Mullard GET102 and some others but some barley even got much volume or fuzz at all. Is it the hfe that is the main determining factor, I dont really understand the leakage side of things.


antonis

Quote from: Reg18 on February 10, 2021, 01:36:29 AM
it sounds great with a 5k instead of 50k.

You probaly mean in series with 50k..  :icon_wink:

5k instead of 50k for R5 results into Q2 Base sitting about 128mV lower than GND..
(it can't fully turn-on Q2 unless it exhibits a VERY high leakage current..)
"I'm getting older while being taught all the time" Solon the Athenian..
"I don't mind  being taught all the time but I do mind a lot getting old" Antonis the Thessalonian..

anotherjim

Is that Solasound schematic deliberately showing a 2 cell (3v) battery or just a sketchy representation for a 9v?
1965 so it might be 2 cells.

duck_arse

RG has dissected the Zonk machine in this link, which is very very similar to the mkI circuit.

http://www.geofex.com/Article_Folders/zonkmach/zonkbst.pdf
You hold the small basket while I strain the gnat.

Rob Strand

QuoteIs that Solasound schematic deliberately showing a 2 cell (3v) battery or just a sketchy representation for a 9v?
1965 so it might be 2 cells.
The Tone Benders were 9V.  The early Maestro's were 1.5V and 3V.
Send:     . .- .-. - .... / - --- / --. --- .-. -

Reg18

#8
Quote from: antonis on February 10, 2021, 05:47:14 AM
Quote from: Reg18 on February 10, 2021, 01:36:29 AM
it sounds great with a 5k instead of 50k.

You probaly mean in series with 50k..  :icon_wink:

5k instead of 50k for R5 results into Q2 Base sitting about 128mV lower than GND..
(it can't fully turn-on Q2 unless it exhibits a VERY high leakage current..)

No I tried both, in series with 50k but ended up with a 5k pot instead. I have a 20k trimmer bias Q3 and a 5k resistor on Q2 so maybe that's how I got it working, it sounds great though!
I'm new to pedal building and thought that's what you meant, I guess this is how we learn, I'll switch it back and see how it sounds.

iainpunk

Quote from: anotherjim on February 10, 2021, 07:29:29 AM
Is that Solasound schematic deliberately showing a 2 cell (3v) battery or just a sketchy representation for a 9v?
1965 so it might be 2 cells.
not sketchy at all, its a correct IEEE notation of a battery. IEEE always draws that style batteries independent of voltage, or number of actual cells. even single cells are drawn like that, since the american single cell notation is used here as an ideal voltage source symbol.
the whole drawing is IEEE correct, except for the resistor symbols (i prefer american resistor symbols over real resistor symbols as well.)

cheers
friendly reminder: all holes are positive and have negative weight, despite not being there.

cheers

Arangonix

So, antonis, with 5k6 resistor in R4 it starts to produce proper sound a bit earlier, but still if I have attack pot on 0 it's off. Bad thing is it starts to sound a bit shrill. When I went further, with 6k8 or 10k resistor, it had louder sound with attack pot on 0, but it was rather nasal and farty. I had to keep my guitar's vol maxed. If I turned a guitar down to 8 or 7, sound became very quiet. With 6k8 and 10k resistor the effect produces a lot of upper midrange shrill, rather unpleasant sound it was.
If pot was "grounded" with 5k6 resistor or so, it works kinda same.

Funny thing. If I have my "stock" resistors in Q2, while turning up the attack pot, Q3's collector voltage (set to 7V) suddenly drops to 3.42V (with pot around 1 - 1.5), after that it starts to grow up to 7V (when attack pot reach it's maximum). Should this works like that? :)

I noticed that if I force a bit more leackage in Q3 and then set Q3's collector voltage to 8V, it changes nearly nothing while turnig attack pot up/down. Didn't check the sound with that configuration yet.

Rob Strand

QuoteFunny thing. If I have my "stock" resistors in Q2, while turning up the attack pot, Q3's collector voltage (set to 7V) suddenly drops to 3.42V (with pot around 1 - 1.5), after that it starts to grow up to 7V (when attack pot reach it's maximum). Should this works like that? :)

I noticed that if I force a bit more leackage in Q3 and then set Q3's collector voltage to 8V, it changes nearly nothing while turnig attack pot up/down. Didn't check the sound with that configuration yet.
Assuming no wiring or connection issues, the things than can cause that are oscillation, noise even RF.   You can see if the behaviour stabilizes when you set the volume control on the guitar.   It's better if you have a dummy shorted 6.5mm jack you can plug into the input socket.  Another way is just soldering a short at the input.
Send:     . .- .-. - .... / - --- / --. --- .-. -

Arangonix

I have it on breadboard right now. My experiment change it's overall sound a bit (brighter, longer sustain, a bit more gating), but attack pot still works like before.

Arangonix

Ok, guys. I did many different tweaks and I think I solved the problem. It was easier than I thought :) My R4 resistor is 10k now, I also change the pot from linear to logaritmic. It works from very begining of the scale now, giving me a regulation from 8.88V to 5.68V. Overall, it makes the effect a bit brighter but in norm.

Drop me an idea how to reduce trebly noise a bit. Ceramic/silver mica cap to ground will be good? Or maybe between collector/base?  Which tranny should have it?

iainpunk

to reduce brightness, i think you might want to experiment with miller cap across Q2 base to Collector, i suspect somewhere between 10p to 100p, but if that's still to bright, you might want to go up to 1n

cheers, Iain
friendly reminder: all holes are positive and have negative weight, despite not being there.

cheers

Arangonix