Supa Fuzz and TB MkII - what is that input to ground cap doing?

[Bret608]

Hi everyone,

I did a Supa Fuzz a while back (which works great now thanks to help from folks here), and now I'm itching to build more of a straight MkII just to see what the difference is if any. I am wondering about the 0.01uf cap that goes from input to ground on some versions. What is this doing in the circuit exactly? I built the Supa Fuzz version at Tagboard Effects, which has it, but I've seen on other schematics and more than one thread here that the Supa Fuzz actually didn't have this cap.

Any light folks can shed would be much appreciated...

[nocentelli]

A small cap to ground just bleeds off higher treble frequencies from the guitar signal: Fuzz pedals often heavily accentuate treble by generating lots of high order harmonics, so I assume the cap is is intended to reduce the amount of fizzy, harsh high-end added to the guitar sound by the rest of the circuit. It will also dump any RF interference to ground, so you don't get air traffic control coming through your massive Marshall stack.

[Bret608]

I suppose I should leave that cap where it is then! I live pretty close to a radio tower, so some of my other fuzzes make great radios. My Tone Machine picks up this really interesting economics talk show on Sunday mornings. 

[nocentelli]

Are you planning a Ge or Si MKII?  Many modern/boutique versions of the MKII (especially with silicon transistors which tend to be higher gain) add a tone pot or B-C caps to the circuit to tame the treble a bit: If you're planning an authentic OG version, I'd leave the cap in there.

[Bret608]

I was planning GE, most likely 2n1305 as I have some in the right gain ranges. I agree about leaving that cap in. My Supa Fuzz has a mix of 2n217 and 2n281. I think I will leave the cap in there as it is too, since it doesn't seem to be lacking treble!

[Electric Warrior]

I guess it's a schematic of a Marshall made Supa Fuzz with the 10µF electrolytics. They shouldn't make any audible difference compared with the 5µFs.

The cap on input will have a big influence on how it sounds with a wah. Give it a try. Most of the Sola Sound branded MKIIs ("short boards", having been MK1.5s but upgraded to MKII specs) didn't have that cap either, so no difference between a MKII and a Supa Fuzz, really. Originally they were made from the same parts using the same schematic, anyway. The Sola Sound branded version was just discontinued shortly after they started using the cap on input and replaced with the Vox branded version.

Another interesting circuit variation uses 0.015µF caps instead of 0.01µF. They sound a bit thicker. Highly recommended!

Make sure you use a 50µF power filter cap, not 25µF like Sola Sound originally used. With 25µF the circuit is prone to squealing.

[bool]

I suppose I should leave that cap where it is then! I live pretty close to a radio tower, so some of my other fuzzes make great radios. My Tone Machine picks up this really interesting economics talk show on Sunday mornings. 

So THIS is your __money__ fuzz.

[LucifersTrip]

Funny, on a bunch of the ones I built, the only mod I did was put a switch on that input cap. It really makes a big difference, similar to rolling the treble off on the guitar. It makes it quieter (especially when the attack is up on one built with higher gain), of course it rolls off some high end, and makes it "tighter" and less "open" sounding, for lack of better words.

On the other hand, I rarely add the cap to my lower gain ones that are quiet and gate easily.

[Bret608]

Thanks everyone--sounds like this will be worth experimenting with before I box it up for good.

Electric Warrior, I did use a schematic with the 10uf electros. I did use a 47uf electro for the power supply cap.

LucifersTrip, I think I have a high-ish gain set on the board at the moment, I think 73-97-122 for Q1, 2 and 3 respectively. Would that steer things too noisy if I removed it? I do like the idea of switching the cap in or out. Right now it does sound pretty controlled and is not horribly noisy. It only oscillates a bit with the attack on full, but that may be because it's not boxed up yet.

[Davelectro]

I'd keep the cap to ground (it could be lower than 10nf, tho).

Fuzz circuits need some HF taming at the input. An RC filter would be even better.

[LucifersTrip]

LucifersTrip, I think I have a high-ish gain set on the board at the moment, I think 73-97-122 for Q1, 2 and 3 respectively.

you can't tell if it's high gain by the hfe's. the leakages and more so, the C voltages will tell you more.

Would that steer things too noisy if I removed it?

only you can answer that one...

I do like the idea of switching the cap in or out. Right now it does sound pretty controlled and is not horribly noisy. It only oscillates a bit with the attack on full, but that may be because it's not boxed up yet.

remember, there's that optional 100 ohm on Q3E that prevents oscillation when attack is full on. you don't have to put exactly 100 ohm there. just use a 500 ohm trim and turn up till the oscillation stops.

http://forum.musikding.de/cpg/albums/userpics/16602/MKIIOC75schem.png

[Electric Warrior]

The 100Ω was optional (many units that have it have it wired out of circuit) and was only used in the early days of the MKII/Supa. They dropped it pretty quickly and then increased the power filter cap's value to 50µF - which is a better way to deal with that problem imho.
The 100Ω in my OC75 MKII schematic because I wanted it to represent a large number of actual vintage units and cover many circuit variations. Those early units with optional caps and resistors were perfect for this. I guess I should make another schematic that covers the units with 0.015µF caps... I once made a schematic that covered all circuit variations, but it was just too complicated.

With 50µF oscillation/squealing shouldn't really be an issue. You may want to reconsider your transistor selection. May try swapping them around.

[Bret608]

I agree, I am lining up some different transistors to try. I suspect Q2 may actually be higher gain or leakage than is just right for that position. I really like what I've got in Q3 right now, even though on paper it seems like the leakage should be too high. The sound is the best I've gotten off this board yet, and the Q3 collector voltage if much closer to vintage units than I remember from my earlier measurements (about 7.3v with the 8.2k bias resistor). As mentioned, it's around 122 hfe; leakage is actually around 600uA. It still had that slight oscillation/squeal at max attack with a lower gain/leakage transistor in Q3. I'll play around with the 60-80 hfe range and various leakages for Q1 and Q2 and see what happens.

Electric Warrior, where could I find the schematic you mentioned? It sounds like you put a lot of thought into it, so I'd love to have that handy to inform my quest, such as it is.

[Electric Warrior]

600µA is sounds alright for a transistor in that gain range. The OC75 version of the MKII Tone Bender sure works great with plenty of leakage.

You have a very wide spread of gains. That could be a problem. When I put together a MKII I start out with the gains matched closely and then substitute transistors to improve on it. I shape the tone and noise by substituting Q1 and Q3 and set the right ratio of gating/hiss by selecting the right transistor for Q2.

With the stock resistor values I actually find it hard to get the voltages outside the range that can be measured in vintage units. 

I never published the schematic that covered all variants that Sola Sound ever made. It had seven footnotes, which made it very confusing and hard to use. It works as a reference for me, but it's not something I'd like to give other people to use. There's a good chance they'd get it wrong. And I really don't want to spread any more confusion about this circuit.

Instead I decided to publish two schematics, one for the OC75 version and one for the OC81D one, and tried to cover as much ground as possible with them, representing accurately what can be found in actual vintage units, while keeping it simple at the same time.

[Bret608]

I think the schematics are clear and very helpful. Right now I have my Supa Fuzz board like the OC75 version, except with the 10uf electros and with 47uf for power supply and no 100 ohm resistor at the Q3 emitter. What would be the best starting point for hfe and more importantly leakage, given that setup?

I've mainly been trying out 2n1305 and 2n217 transistors. I don't know which of those are closer to OC75 vs. OC81D. Is different tendencies in leakage the main thing that necessitated the different part values in the two versions?

[Electric Warrior]

I'm not sure. The 100k resistors might help with biasing low leakage transistors, but I haven't tried that setup yet as I still have a small stash of OC75.
Some Mullard OC81Ds leak just as much as OC75s, though, so other specs might matter. 

Hfe 90 seems to be right in the middle of the range your transistors fall into, so pick three that measure around 90, audition them and then go by ear.

I'm can't say for sure which gains and leakages need to go where. There are a lot of different setups that may work.
I once got three Valvo AC125s in a lot with plenty of other transistor types. Just three random transistors, but hey happened to work rather well in a MKII, as they have a good amount of leakage. This is the setup I ended up with:

Q1: hfe: 83; 404µA leakage
Q2: hfe: 65; 299µA leakage
Q3: hfe: 80; 388µA leakage

I used the lowest gain for Q2, but in other cases I used the highest gain in this position. It's whatever works, really. With three hfes and three leakages you have six interacting factors that you can work with. It makes coming up with a "formula" rather complicated.

[Bret608]

Thanks again--I've seen a few theories on MkII gain/leakage ranges out there, but never a methodology on just auditioning them by ear as you've provided. I will report back when I get a chance to swap more of my stash in and out of the circuit.

[Electric Warrior]

Listening is a must if you're trying to make a great sounding pedal. 

[Bret608]

Agreed! I do use my ears but am not too systematic with how I deal with the variables. 

[Electric Warrior]

Neither am I. I just mess around until it sounds good to me.