Tonebender MkII Questions

[Big Monk]

After playing through my recent build with the Small Bear Flat Hats, I have a few questions for the collective:

1.) I used an "Easy Face" style pre-gain pot on the inout of the circuit. It seems to work pretty well and rolls a bit of the "hair" to tighten things up. My concern is whether it would work better between Q1 and Q2 (where it would be on a Fuzz Face style circuit)? It seems to have a more limited range than I remember on the Gagan circuit.

2.) I was coming off a Si Tonebender circuit (the RDV unit) and installed a switch for changing the output capacitor from 0.01 uf to 0.1 uf. It is subtle, but very nice and actually presents itself as more of a gain boost than a bright switch as I intended. I'm wondering if switching the Attack pot capacitor would be of interest here as well/as an alternative.

3.) I'm entertaining the thought of installing some NOS Mullard caps I have in my stash. In particular, I have some Radial caps in Tonebender values and some Tropical Fish in the same values. I try to stay objective about components and not fall into the "mojo" trap but I really like this circuit and anything that can take it up a notch is welcome. Anyone have any experience with using vintage caps in this circuit.

4.) I'm wondering what effect dropping the input cap down to about 2.2 uf and increasing the Attack pot cap to 10 uf might have? Cancel each other out?

Side note: The 1kC pot for the Attack is wonderful. Very much improved the range of the control.

[garcho]

Those NOS caps can crack, become microphonic, will have corroded leads, and values that can be way off from what they're marked as. Other than that mojo, no bonus.

Post some links to schematics, most of us either aren't going to know what you're referring to or it's something we built years ago and can't recall. Don't make people you're asking for advice from have to waste time googling things, like small bear flat hats, which I couldn't find any mention of.

Your questions could be answered by the breadboard, and then you won't take a chance on f'ing up your germanium mojo with the iron. I'm a huge fan of breadboarding so I can't help but say something

[Big Monk]

Those NOS caps can crack, become microphonic, will have corroded leads, and values that can be way off from what they're marked as. Other than that mojo, no bonus.

Post some links to schematics, most of us either aren't going to know what you're referring to or it's something we built years ago and can't recall. Don't make people you're asking for advice from have to waste time googling things, like small bear flat hats, which I couldn't find any mention of.

Your questions could be answered by the breadboard, and then you won't take a chance on f'ing up your germanium mojo with the iron. I'm a huge fan of breadboarding so I can't help but say something

To be clear, this is a working circuit. I use sockets most of the time anyway so no iron on the Ge stuff anyway. Let me ask more detailed questions with a schematic to eliminate any confusion:




I used a Tonepad MkII PCB, which is detailed by the schematic above. I used NPN Germanium transistors. The type isn't all that important for my questions.

1.) I currently have the pre-gain pot on the input of my circuit. My question was would it be better between C2 and the base of Q2, where it would typically sit in the "Easy Face" circuit? Just curious if anyone had any experience with that control.

2.) I'm currently switching the output cap on a DPDT from the stock 0.01 to 0.1 and it gives a nice gain boost. My question was would switching C3 from stock to about 10 uf be a better alternative? Just curious if anyone had switched that cap before.

3.) I'm not a huge proponent of NOS except for transistors so I was just curious.

4.) I'm finding myself using my Rangemaster clone at a touch below unit gain (it has a master volume) with the blend control backed off so I may have answered this myself. A smaller input cap or a switch to lifted cap to ground on the input may be in order. I used R7 in the schematic above for the input gap to ground.

[idy]

The schematic shows PNP.

[Big Monk]

The schematic shows PNP.

Right. I just reversed the polarized connections and the power. Easy peasy.

Tough crowd!

Just to reiterate: I'm not looking for troubleshooting but rather just curious as to how I might improve the tone in subtle ways or add some versatility.

Maybe only people who are familiar with the Tonebender circuit will be interested?

[garcho]

1.) I currently have the pre-gain pot on the input of my circuit. My question was would it be better between C2 and the base of Q2, where it would typically sit in the "Easy Face" circuit? Just curious if anyone had any experience with that control.

Input impedance can be a bit of a tough piece of meat to chew with these old transistor designs. Why not put the "pregain" pot after the input buffer like what you were saying, and then just use your guitar's volume knob instead of the pregain pot the way you have it now? I assume this pedal will be one of the first things in the chain, yeah?

2.) I'm currently switching the output cap on a DPDT from the stock 0.01 to 0.1 and it gives a nice gain boost. My question was would switching C3 from stock to about 10 uf be a better alternative? Just curious if anyone had switched that cap before.

C3 is for Q3's AC gain. C2 along with the output volume forms a HPF. Lowering C2 brings the cutoff lower, leaving more "bass" in. If you have a 100k volume pot and a 10N cap, cutoff is 160Hz. If it's a 100n cap, cutoff is 16Hz. That's an important range for guitar.

[Big Monk]

1.) I currently have the pre-gain pot on the input of my circuit. My question was would it be better between C2 and the base of Q2, where it would typically sit in the "Easy Face" circuit? Just curious if anyone had any experience with that control.

Input impedance can be a bit of a tough piece of meat to chew with these old transistor designs. Why not put the "pregain" pot after the input buffer like what you were saying, and then just use your guitar's volume knob instead of the pregain pot the way you have it now? I assume this pedal will be one of the first things in the chain, yeah?

2.) I'm currently switching the output cap on a DPDT from the stock 0.01 to 0.1 and it gives a nice gain boost. My question was would switching C3 from stock to about 10 uf be a better alternative? Just curious if anyone had switched that cap before.

C3 is for Q3's AC gain. C2 along with the output volume forms a HPF. Lowering C2 brings the cutoff lower, leaving more "bass" in. If you have a 100k volume pot and a 10N cap, cutoff is 160Hz. If it's a 100n cap, cutoff is 16Hz. That's an important range for guitar.

All great points. I think I may get into the box this weekend and move the pre-gain control and install some sockets on the board so I can tweak the caps.

You make a good point on the pre-gain pot, although I have been using it to set the maximum amount of gain for the circuit and then riding the volume for variation. It sort of works to change the quality of clean up on the volume knob as well. I may choose a smaller value pot there and see how the new placement works out. I have my way first in the chain.

Thanks for the advice and I'll tweak accordingly.

[mozz]

 I would try the pot swap. I know sometimes it's needed in a fuzzface and where you are thinking would be just the right spot.
As to the 10uf, i think if you went 22uf-33uf you won't hear any difference, but i have read 47uf works a little bit, but you might not need that deep boost.

As for the tropical fish, i would use them if you have the right values. As was said, they may be bad but i doubt it if they were stored properly. There is a lot more to capacitors then just a marked value and tolerance. I can bet most people can't hear a 20% tolerance difference. There was one pedal with a polystrene output cap? Mind fog, can't remember the name. If you have a decent LCR meter, a polystrene cap has one of the lowest dissipation factors you can get. Only thing that gets close is a silver mica. The silver mica is not available in all higher values as a polystrene. I can bet with exact value .01uf caps, you would hear the difference between a ceramic, a disc, a poly and a film cap.

On the other hand, some amps used cheap disc capacitors, so using a 1% or 2% quality there might change the sound enough to make it worse. Goes both ways.

Just measured a tropical fish and some others 0.1uf, 1khz
lowest dissipation value seems to be higher voltage but not always. Maybe using those 250v caps like they used to in a 9v pedal does matter.

description/value in .uf / dissaption
cheap yellow axial /.100/.00022
parts express       /.099/.00049
yellow block        /.104/ .00071
tektronix             /.100/.00110
orange drop        /.098/.00313
good all             /.100/.00318
plastic foil         /.103/.00353
film                  /.105/.00365
trop fish            /.089/.00368
mustard            /.105/.00430
cheap disc        /.063/.00976
cheap disc        /.095/.01302
cheap disc 5zu  /.086/.01780
cheap disc       /.097/.01794

[Big Monk]

So I just opened up because I realized I had never taken voltage measurements. I took the values provided by Small Bear at face value and did not inquire as to the bias point.

So I measured with Q1, Q2 and Q3 in the sequential order from Small Bear:

Source Voltage = 9.22 Vdc

Q1: C = 9.13 Vdc, B = 0.02 Vdc, E = 0.00 Vdc (Hfe = 63, Leakage = 21 uA)

Q2: C = 0.85 Vdc, B = 0.14 Vdc, E = 0.00 Vdc (Hfe = 80, Leakage = 33 uA)

Q3: C = 2.59 Vdc, B = 0.86 Vdc, E = 0.70 Vdc (Hfe = 144, Leakage = 71 uA)

So Q3 collector voltage shocked me a bit. I have obviously heard of 4.5 volts as a bias point for Q3 in Modern a Tonebender kits/circuits, etc. but I was under the impression something closer to 8 Vdc was considered nominal.

One thing to consider is that I don't have the "vintage" 470 ohm resistor for R5, but rather a 1k, which while boosting output volume, also seems to act in concert with R6 to supply the bias voltage for Q3.

So I swapped Q2 and Q3 and measured again with the order of Q1, Q3 and Q2:

Source Voltage = 9.22 Vdc

Q1: C = 9.13 Vdc, B = 0.02 Vdc, E = 0.00 Vdc (Hfe = 63, Leakage = 21 uA)

Q2: C = 0.63 Vdc, B = 0.12 Vdc, E = 0.00 Vdc (Hfe = 144, Leakage = 71 uA)

Q3: C = 4.89 Vdc, B = 0.67 Vdc, E = 0.47 Vdc (Hfe = 80, Leakage = 33 uA)

I also swapped in another Q3 from a previous General Electric transistor set of the same type used here and got roughly the same values shown in the second set above. 

Why the ramble and data? One of my only criticisms of the pedal as it stands right now is that it seemed uncharacteristically "wooly" and "bassy" considering it's a Tonebender MkII circuit. Even factoring in my use of humbuckers, it still did not seem to have the brightness and treble of a Tonebender.

The tone gained more high-end and bite with the transistors swapped. Now I am wondering whether the Small Bear resistors were meant to bias Q3 to 4.5 Vdc and if upping the voltage would be beneficial. I may have to play with R5 and R6 to hit that right.

[Big Monk]

Just an update. After some resistor changes, I settled on 7.43 V on Q3c.

It was like pulling a wet blanket off. Increased treble, crunch and overall volume with blooming sustain. Great feedback as well.

Also, pregain control was improved wonderfully, with better taper and control over the f8nal bit of hair on the top end.

I may swap some transistors around to see how things sound but I am very happy.

[Steben]

Just my 5 cents, but ild always put the higher hfe on q3 position instead of q2.

[Big Monk]

Just my 5 cents, but ild always put the higher hfe on q3 position instead of q2.

I have a resister order coming in and I'm going to play around more with it. I modified the board with sockets a few days ago so I could experiment.

Right now, if I put Q3 in the Q2 slot, the pedal is quieter but the voltage goes up to a point I'm not all that fond of, i.e. I start to get undesirable gating/raggedness at full volume.

So, I agree that for now, the higher Hfe Q3 is staying in the Q3 slot, but I have a feeling that after the upcoming resistor swaps, I'll have Q3 in the Q2 slot permanently because it's quieter and rebiasing will reduce the voltage a touch.

[Electric Warrior]

Just my 5 cents, but ild always put the higher hfe on q3 position instead of q2.

I have a resister order coming in and I'm going to play around more with it. I modified the board with sockets a few days ago so I could experiment.

Right now, if I put Q3 in the Q2 slot, the pedal is quieter but the voltage goes up to a point I'm not all that fond of, i.e. I start to get undesirable gating/raggedness at full volume.

So, I agree that for now, the higher Hfe Q3 is staying in the Q3 slot, but I have a feeling that after the upcoming resistor swaps, I'll have Q3 in the Q2 slot permanently because it's quieter and rebiasing will reduce the voltage a touch.

I think the entire point of the MKII design may have been to make the circuit more usable when it's gating. The MK1.5 was biased too hot, so it quickly got farty sounding in the warm time of year. The additional stage in the MKII keeps the signal over the threshold for longer, so a bit of gating sounds real good. My vintage units sounded fine during the summer, even when it was around 30°C.

[Big Monk]

Just my 5 cents, but ild always put the higher hfe on q3 position instead of q2.

I have a resister order coming in and I'm going to play around more with it. I modified the board with sockets a few days ago so I could experiment.

Right now, if I put Q3 in the Q2 slot, the pedal is quieter but the voltage goes up to a point I'm not all that fond of, i.e. I start to get undesirable gating/raggedness at full volume.

So, I agree that for now, the higher Hfe Q3 is staying in the Q3 slot, but I have a feeling that after the upcoming resistor swaps, I'll have Q3 in the Q2 slot permanently because it's quieter and rebiasing will reduce the voltage a touch.

I think the entire point of the MKII design may have been to make the circuit more usable when it's gating. The MK1.5 was biased too hot, so it quickly got farty sounding in the warm time of year. The additional stage in the MKII keeps the signal over the threshold for longer, so a bit of gating sounds real good. My vintage units sounded fine during the summer, even when it was around 30°C.

I think the threshold for these transistors is a touch lower by my ear. 7.98 V was a bit too much and while they sound awesome at 7.43 V, I think I want to find a sweet spot in between if it exists.

The process is fun anyway and worst case I go back to 7.43 V. Thanks again for info. It's been a night and day difference.

[Electric Warrior]

7.43 seems rather low, most vintage units measure over 8V, but the voltages don't tell me exactly what it sounds like, anyway. Gating will generally increase as Q2C goes down/Q3 goes up (depending mostly on Q2's leakage and thus the temperature), but it can still be a little unpredictable. My Supa Fuzz doesn't clean up nicely today, but as temperature increases it doesn't necessarily clean up increasingly worse. I guess it may depend on how the transistors are matched and how the individual transistors drift at different rates? Sometimes it cleans up nicely even when it hasn't been getting colder. At very high temperatures, it does gate quite a bit, though.

My Vox MKII cleans up perfectly today (it behaves a lot more predictably than the Supa). The voltage on Q3C is still higher than what you currently prefer. This depends a lot on the exact value of the attack pot, though, there's a quite a bit of tolerance here.
Q2C sure is very high in this unit. It sounds rather similar to the Supa. A little brighter and hissier, but the character is similar and both sound huge. 

Vox MKII, OC75
@21 °C
Battery: 9.7V
Q1 C -9.04V B -0.04V E 0V
Q2 C -0.26V B -0.08V E 0V
Q3 C -7.84V B -0.26V E -0.18V

Marshall Supa Fuzz, OC75
@21°C
Battery: 9.38V
Q1 C -8.83V B -0.03V E 0V
Q2 C -0.17V B -0.08V E 0V
Q3 C -8.32V B -0.17V E -0.09V

[Big Monk]

7.43 seems rather low, most vintage units measure over 8V, but the voltages don't tell me exactly what it sounds like, anyway. Gating will generally increase as Q2C goes down/Q3 goes up (depending mostly on Q2's leakage and thus the temperature), but it can still be a little unpredictable. My Supa Fuzz doesn't clean up nicely today, but as temperature increases it doesn't necessarily clean up increasingly worse. I guess it may depend on how the transistors are matched and how the individual transistors drift at different rates? Sometimes it cleans up nicely even when it hasn't been getting colder. At very high temperatures, it does gate quite a bit, though.

My Vox MKII cleans up perfectly today (it behaves a lot more predictably than the Supa). The voltage on Q3C is still higher than what you currently prefer. This depends a lot on the exact value of the attack pot, though, there's a quite a bit of tolerance here.
Q2C sure is very high in this unit. It sounds rather similar to the Supa. A little brighter and hissier, but the character is similar and both sound huge. 

Vox MKII, OC75
@21 °C
Battery: 9.7V
Q1 C -9.04V B -0.04V E 0V
Q2 C -0.26V B -0.08V E 0V
Q3 C -7.84V B -0.26V E -0.18V

Marshall Supa Fuzz, OC75
@21°C
Battery: 9.38V
Q1 C -8.83V B -0.03V E 0V
Q2 C -0.17V B -0.08V E 0V
Q3 C -8.32V B -0.17V E -0.09V

You've hit on a couple of things I've been trying to observe and adjust for. Take the following with a grain of salt because I feel there is room for improvement in the in between voltages between 7.43 and the 7.98 I observed initially:

1.) Clean up is pretty at 7.43 V, although I noticed it's hanging in a little fuzzier for longer than I would ultimately like. I'd say with Attack on full and volume between 8-10 on the guitar, I get very subtle differences. Between 5-8 on the guitar I get more shades of distinct cleanup.

2.) Given the resistor values I have in now for biasing (Q2c is 110k, Q3c is 8.2k (with a 1k between them instead of the nominal 470 ohm), and the Q3e to Q2b feedback resistor at 82k) I think swapping Q2c down to 100k (the nominal value for low leakage transistors like mine) and the feedback resistor to 100k should get me in the sweet spot.

3.) I've taken the hiss I'm experiencing as just taking the good with the bad. The overall increase in brightness, crunch, sustain and note bloom is well worth the above average hiss. I'm hoping that swapping in my Q3 into the Q2 slot (which was quieter when I tried it) plus the resistor changes will be ideal.

Overall, it sounds incredible. If anyone every needed a reason to avoid 4.5 V on Q3c, let this be the endorsement.

[Electric Warrior]

Overall, it sounds incredible. If anyone every needed a reason to avoid 4.5 V on Q3c, let this be the endorsement.

Interestingly, when I built my first MKII, I started out with trim pots on every transistor and thought that it sounded thin and buzzy at 4.5V, which is the exact opposite of what you experienced

Not sure how it would sound with only Q2's collector resistor tweaked to get 4.5V on Q3C. Increasing Q3C's resistor sure reduced the output volume a lot. This causes the volume issues many people experience. Stock MKIIs and Fuzz Faces have plenty of volume.

[Big Monk]

Overall, it sounds incredible. If anyone every needed a reason to avoid 4.5 V on Q3c, let this be the endorsement.

Interestingly, when I built my first MKII, I started out with trim pots on every transistor and thought that it sounded thin and buzzy at 4.5V, which is the exact opposite of what you experienced

Not sure how it would sound with only Q2's collector resistor tweaked to get 4.5V on Q3C. Increasing Q3C's resistor sure reduced the output volume a lot. This causes the volume issues many people experience. Stock MKIIs and Fuzz Faces have plenty of volume.

It could be down to how the 2N169s are reacting in this circuit. In fact, I could actually see how some people might have liked the 4.5 V sound. It definitely was not thin and buzzy like your experience with the NOS transistors that actually existed in vintage units. The output was definitely lower and it resembled a tamer version of the vintage Big Muffs I've heard.

Personally, I was not a fan of it and wondered if I had done something wrong when wiring and populating the circuit. It quite literally was like I was at the edge of another dimension where some was playing a proper Tonebender MkII: muffled, lower volume, but you could hear how it should sound.

It looks like even if lower voltage sounds are not constant across all NOS devices, at the very least higher voltages, varying somewhere from 7.5-8.0 V on Q3c, are almost universally the recipe for the classic MkII sound.

[Electric Warrior]

It looks like even if lower voltage sounds are not constant across all NOS devices, at the very least higher voltages, varying somewhere from 7.5-8.0 V on Q3c, are almost universally the recipe for the classic MkII sound.

There's no doubt at all that they were set up with Q3C far from 4.5V.

I once calculated average voltages, based on a couple of sets of OC81D and OC75 MKII voltages:

supply: 100%
Q1C 92%
Q2C 1.82%
Q3C 89%

[Steben]

One of the advantages of a little gating is noise reduction in silent phases