Fuzz Central MKII BIAS MOD

[Jujo]

Hello,

This is my 1st post and my first time trying to mod a pedal. Please don't hurt me...

So I've been messing with this mod- http://fuzzcentral.ssguitar.com/mkII.php

I've tried at 10k and a 25k trimmer in place of the 8k2 resistor as recommended- "I would recommend that you replace the 8K2 resistor on the collector of Q3 with a 20K trimpot which will allow you to adjust Q3s collector voltage to -4.5VDC"

The trimmer seems like it is only adding a fixed resistance. When I turn the trimmer I don't hear any difference until the very last part of the range and then the signal cuts out altogether. What I want to know is if this bias mod should have an audible effect as I turn the trimmer or does it just change the voltage to Q3 but it doesn't really change the sound???

The fuzz is surely more gated and spitty (which is what I wanted) but I was hoping a bias trimmer would allow me to dial in the level of gate and spit. Can anyone help me out on this?

[mac]

The sound of both TB and FF do really change when you vary Q3 collector voltage. This is the major drawback of both circuits. Instead of a trimmer I suggest you to use a pot to correct for variations in temperature that sends the out voltage up and down constantly.
If your transistors are leaky then maybe a 25k pot is not enough.

Check the connection of your trimmer. Maybe you connected pin 1 & 3, should be 2 & 1 -- or -- 2 & 3.

mac

[Jujo]

Mac-

Thanks for the reply. Did you mean to say that the FF and TB does not really change? You typed the "do really change". I just want to be clear about what you mean. Also the trimmer lugs are connected properly- 1 and 3 are not connected.

If the trannies are leaky, why would going to a larger value trimmer correct the problem? I do have a 50k that I bought just in case. Maybe I'll try it but I was afraid that a 50k trimpot wouldn't let any sound come through. The higher the value of the trimpot, the more gated and spitty the fuzz will get right? That's why I'd like to be able to adjust the sound with the trimmer. But maybe I can't do that. Maybe the sound does not change. Please let me know.

THanks!

[foxfire]

the size of the trim pot doesn't really matter as long as it's value is large enough for it to do what it is supposed to do. in other words you could use a 100k and it wouldn't make a difference. going bigger just give you more adjustment.

[Jujo]

That's the thing though. I don't hear any adjustment when I turn the trimmer. The trimmer only seems to work as an on/off switch. I want to know if I should hear the bias changing or what?

[tungngruv]

Also the trimmer lugs are connected properly- 1 and 3 are not connected.

If I'm looking at the right schem at fuzz central, lugs 1 & 2 should be connected to Q3's collector and lug 3 to the 1k2/.01uF junction. On all the fuzz pedals I have built, that's how they are always connected. Also, in my limited experience I have noticed that when biasing a fuzz, at one extreme of the trim rotation it will almost die out, then get very sputtery, the within a small distance of that, it suddenly fuzzes the right way. I have also used 10k and 25k and both work fine. 

[mac]

Thanks for the reply. Did you mean to say that the FF and TB does not really change? You typed the "do really change". I just want to be clear about what you mean. Also the trimmer lugs are connected properly- 1 and 3 are not connected.

I meant FF and TB, and many other circuits using Ge change with temperature, so the sound. (I speak one language, spanish, and very bad )

If I'm looking at the right schem at fuzz central, lugs 1 & 2 should be connected to Q3's collector and lug 3 to the 1k2/.01uF junction.

The schematic is confusing, I avoid drawing pots this way. The trimmer is just a variable resistor in this circuit. If you connect lug 2 and lug 1, then lug 3 is unconnected. Or if you connect lug 2 and lug 3, then lug1 should be not connected. BTW, use a linear trimmer. For example, connect lug 1 to q2 collector and lug 2 to the out cap and the other resistor.

If you use an external pot to correct for temperature run away, do this. First, find the resistance that makes q2 collector voltage near 4.5v. Sustract 470R or 1K2 resistor. The pot you need is one that at the middle position makes 4.5v at vc2.

mac

[yeeshkul]

Mac what about the input 10n cap (to the ground) you recommend - did you see it inserted in a real MKII?

[mac]

Mac what about the input 10n cap (to the ground) you recommend - did you see it inserted in a real MKII?

Well, the TB II generates a lot of treble content. To some years, it is like having the speaker inside your brain. But it depends on the transistors you use. My silicon version uses 2n2369 and the 10nf input cap is a must. But my Ge version uses 2sd352 matsushitas power transistors which have a transition freq at 10khz. They are dark so the 10nf is not necessary. Trully vintage sound. Other Ge I've tried need the input cap. I guess that the guitar tone, or a switch/pot maybe used to adjust the input cap.
I never saw the guts of one. Most schematcs I've seen, and threads, show the 10nf.

mac

[slacker]

I've just built a silicon version using 2n3053s and without the 10n to ground it suffers from terrible radio interference (unless you like French radio that is).
Some of the pictures here look like they've got the 10n judging by the number of caps in them. Phillip Bryant from Fuzz Central also claims the originals had it.

[yeeshkul]

Thanks Mac. I am tweeking my first TB at the moment and have some troubles.
First of all, i use Matsushitas (i guess we both like them ) 2SB175 of the folllowing gains Q1(83), Q2(83), Q3(110). I measured the gains carefully (waited 5 minutes to let the tranny breathe before measuring it ... by the GEO circuit indeed).
When i checked the sound out it was surprisingly nice and smooth, but when i turned down volume on my guitar it became a nightmare - it was really awful. That made me to check the bias, because it sounded like Q3 was too close to saturation, not letting the silent sounds go through untouched.. And it was - Vc(Q3) was about 8.5V.
The trouble is that i inserted a 100k(!!) pot and it wasn't enough to get -4.5V. Maximum is -5.6V when the pot is 100k...
Isn't it strange? Shall i tweak any other resistor?

[yeeshkul]

aslo the Fuzz pot doesn't realy have much influence ... but that may be the wrong bias again ....

[mac]

If your 2sb175s are like mine, most likely, then:
1. As you increase leakage , q3 collector resistor also must be increased to get 4.5v. Did you measure the leakage? Mine are in the range 100-200ua.
2. If you are using the standard 100k at q2, and you have around 200ua at q2, then you may need 100k to set 4.5v at q3 collector. Try using a 47k at q2. Many use this value as it is supposed to sound less gated. With my 250ua 2sd352 I had to use this value and near 20k at q3.

Use this to estimate
http://geocities.com/guitarfxs/exes/biascalculator.zip
or
http://geocities.com/guitarfxs/bias/main.html

mac

[yeeshkul]

Great! Mac thanks! How big influence has the Fuzz knob on your TB? Because mine is quite helpless, it doesn't seem to work much .

[yeeshkul]

I am a bit confused now . Where should be the -4.5V bias voltage? Right on the Q3 collector leg or at the spot between "8k2" and 470 (output junction) .

Also, when i'm re-adjusting Vc (on the Q3 collector leg) from -4.5 to -7.5 (by 100k pot), the volume goes up and it actually sounds fairly good (a bit more juicy). At -4.5, the volume seems to be more quiet than when bypassed ... 

And one more thing. I am not sure what is the first transistor exactly for. Both Q1 and Q2 have grounded emitters. Which means their bases are biased by a little voltage on B-E. The positive half-wave the first tranny amplifies is reversed to negative half-wave and sort of attenuated by the second tranny ... 

Just to sum up my questions:
1. How much the 1k pot really influence the fuzz (in case of FF it works really well, but not here)
2. What is the role of Q1
3. Why do we try to get -4.5V on collector leg considering the output comes from the "8k2" - "470" junction which has much higher bias voltage due to the "8k2" resistor?

[mac]

1. As in the FF the drive pot does not go from full gas to clean, but IMHO it works better in the TB than in the FF. Most of he time I set the drive pot to min, and it sounds good.
I remember to have changed q1 biasing to make it a buffer, and with this mod the drive pot cleaned the sound completely.
Maybe the higher resistors values at q3 you're getting has something to do with it.

2. Q1 is one more stage making some fuzz. If you have an audio probe, or simply put a 0.1uf cap at q1 collector and then to the amp, you'll notice some fuzz. It's "similar" to a Big Muff stage, but as Ge are leaky you may not need the feedback resistor from C to B. Nevertheless, you can put a pot from C to B and play with the collector voltage. I like to set it at 8v. The rule is this: as you have more leakage the output voltage goes down, and with very low leaky Ge and Si the output voltage tends to vcc. That's why Si does not work without the feedback resistor. So a good Ge, a one with almost no leakage, is not good at q1, but ideal for q2, and q3 could be anything as long as the gain is around 100.
Depending on the circuit leakage an be a good or a bad thing.

3. As you need a large resistor to get 4.5v at q3 collector, not at the junction of the 0.47k and the 8.2k, the ratio 0.47k/R, (R=the value to get 4.5v) is very small compared to the standard value, 0.47k/8.2k=0.0573. If you look at the Rangemaster schematic you'll see that the siganl is taken anywhere between the transistor collector and vcc. The more you are closer to vcc the less signal you get.
Consider this, the transistor is making the voltage swing, so it is at its collector where you need 4.5v DC, because the AC signal ?V will make the collector voltage vary from vcc/2 ± ?V. As you walk away from the collector more resistance is on the signal path sending the volume down.

Do you have the circuit on the breadboard, or it is already soldered? I ask because you can reduce the the 100k at q2 to 47k, which will make q3 resistor go down.

mac

[yeeshkul]

Hey Mac, thanks for the brilliant answer. That was really much more than i expected!

I've changed the resistor 100k -> 47k, have chosen a new tranny which let me use the bias resistor 36k - and that is finally not an insane value 

I also added a switch to FF mode. Both TB and FF mode play sooooo good! 

The Fuzz knob works perfectly in FF mode and just slightely in TB mode, but whatever ... hehehe

All the transistors for Q3 i'd picked had leakage around 190uA and Hfe between 100-120. Only one of them though (the last one) let me put smaller bias resistor than 100k.
As for the Q1 - you mentioned that a leaky tranny can be used - i mean do we prefere leaky device here? If i understand you well then leaky one causes lower collector voltage ...

[mac]

As for the Q1 - you mentioned that a leaky tranny can be used - i mean do we prefere leaky device here? If i understand you well then leaky one causes lower collector voltage ...

Exactly. I mean 150-300uA of leakage. More than this and the collector voltage can drop a lot. Keep vc1 around 7.5-8.3v and you're fine. Of course you can experiment with lower/higher vc1 values with the aid of a pot from B to C, or just another trannys.

Maybe you like to try a 2n2369 at q2 or some other low gain silicon. An hybrid Ge/Si sound with amost no temperature issues. With a 100k at q2 C expect 5-7k at q3 C.

I liked your idea a FF/TB switch. The input and drive pot caps are different but can easily be solved.

mac