Fuzz face sounding dark

[DocBrown]

Hello !

I'm New here and New in the pedal building so I request Some experimented advices.

I build my first pedal yesterday, a Fuzz face clone (I thought it was an easy build but after reading Some infos every where it seems it's not !).

I used the original schematics so the specs are pretty basic. I used silicon bc108 transistors to avoid the germanium voodoo for m'y first build.

As i read that the hfe of Q1should be lower than Q2 I put BC108A for Q1 and Bc108B for Q2.

After a little debugging session my pedal works. First I get Some hiss with the fuzz control at maximum. After à voltage control I found 2.74V on the collector of Q2 so I switched the 8,2k resistor to a 6,6k (the only value that could be right that i had in stock. Now I've 3.9V on the collector of Q2. Not perfect but the closer I can get waiting for a trim pot. No hiss anymore. Other voltages on the transistors pins  are right what I've read I'm supposed to get.

But I'm not satisfied of the sound. The pedal sounds dark like if the tone of my guitar was rolled down. The neck pick-up position is almost no useable for example. Seems also to sound more dirty than expect éd.

I never tried a "True" Fuzz face so it's hard to know if I'm far from the sound I should get or not.

Anyway, do you think this dark sound could Come from the incorrect (even id it's close) bias of Q2 ?

I'd like to stay close to the original circuit so I'd like find what is wrong instead of mod the pedal

Thanks a lot !

[amptramp]

Welcome to the forum!

The Fuzz Face has a low input impedance which interacts with the inductance of the pickup to roll off the highs.  If you want to see what it would be like with a higher impedance, just roll the volume control back on the guitar to add more resistance in series with the pickup.

Silicon was a wise choice as leakage currents are microscopic compared to those of germanium and getting that out of the way makes the design more consistent and easier to debug.

The neck pickup sounds the darkest because it generates fewer harmonics than pickups closer to the bridge.

[DocBrown]

Thanks for the welcome, and your answer.

You're right, if I roll down the volume of the guitar the highs increase. With an other guitar (with humbucker pickups) the sound is less dark.

But, even if the specs of the pick-up make a difference, I still think it sounds darker as it should be.

I think the over dirty sound could Come from the misbiased Q2 but I don't know if it could also cause a dark sound. Or if this really bassy sound is a normal Fuzz face spec....

[j_flanders]

If the capacitor at the fuzz pot (at the emitter of Q2) is missing or wired incorrectly you would get a dark sound.
It should go from wiper to ground.
Also, the more you turn down the fuzz pot the darker it gets. It's brightest when fully up. If that is too much gain, turn down the volume pot on your guitar just a little.

[DocBrown]

Well... I think it's OK but if somebody could check my layout it would probably help.

Here's a layout  pretty close of mine (it's a turret board):



This one is for germanium pnp so I made Some ajustements for my npn bc108 :
- the 20uF and the 2.2 uF are reversed on my board
- the 470ohms resistor is switched to a 330 as on the silicon schematics
- the battery polarity is reverse too (ground is on-, and the + is wired where the- is on this layout.
- the ground is on the unused turret at the bottom right of the board.
Any mistake ?

Thanks a lot !


Édit : here is the schematics of the original silicon arbitrer Fuzz face I'm based on:

[smallbearelec]

I'm sorry that I did not see this post earlier. Ge or Si, the FF circuit benefits hugely from being breadboarded before being committed to solder. You get to tweak bias resistors to compensate for transistor gains, add snubbing caps to reduce harshness and more. Please check out my article:

http://diy.smallbearelec.com/HowTos/BreadboardSiFF/BreadboardSiFF.htm

Happy Experimenting!

[DocBrown]

Thanks a lot for your Link ! So much usefull infos !

So if I sum up all your answer :
- my layout seems correct as nobody smite me for this
- the bassy sound maybe part of the Fuzz face tone and I have to deal with pedal knobs and guitar control to find the sweet spot.
- set the bias of Q2 to 4,5v (3,9 now) to get a better sound
- try different transistors

My transistors May have suffer from soldering/unsoldering 3 times while debugging. Today Q1 is around 170hfe Q2 around 290. I think it was in the recommandations but maybe not. Any suggestion ?

Thanks a lot.

[duck_arse]

also welcome - smiting to follow.

please can we see photos of what you have built. that way we can see if you have twisted the transistor legs into the right places. that layout diagram would confuse most people, showing the same transistor outline with differing pin placings.

[DocBrown]

Eh eh smiting to follow, excellent 

I'll try to post pics of my build but it's extremely difficult to take a photo of the transistors where you can see the mark on the can and the pins at the same time.

I followed the order on the layout, from left to right :
Q1 : emitter, base, collector
Q2: base (linked to the collector of Q1),  collector, emitter.

Thank you all for your help.

[amptramp]

We had a thread a few years back where one poster said that a band doing an arena concert was frantically searching for a 9 volt battery that had been run down enough to operate their Fuzz Face.  The battery voltage and the internal impedance of the battery (which can go up to 50 ohms) were important because the battery impedance sets a feedback value between the two stages.  A Fuzz Face needs the correct battery voltage with the correct impedance, the correct temperature and a limited range of transistor gains, different from each other and if germanium, the correct value of leakage current.  Some people made power supplies with a 50 ohm variable resistor in series so they could dial in the correct power.  And some people add a power supply bypass capacitor to reduce power supply sensitivity to battery impedance.

Some people used a resistor in series with the input so they could avoid the tone sucking that causes the dark tone you mentioned and it acts like the guitar level control has been backed off a bit.  There is an additional design (I forget the name of it) that has an emitter follower driving the Fuzz Face through a resistance so the guitar can see a high impedance even if the controls are set to provide minimum input impedance.  And there are coupling capacitor variations that are selected to get as much bass as possible without turning muddy.

Don't be alarmed by the fact that you don't have the right sound yet - just keep tweaking values until it sounds right.  Then realize the next one you make will sound completely different.

A lot of people like the sound of a Fuzz Face and of course, if Jimi Hendrix used it, it has to make them better guitarists.  It has an easy entry into fuzz even from low inputs that gets more extreme as input level goes up and doesn't have a threshold like diode clipper designs.  It has the sound a lot of people like.  But all the stars have to align for it to sound good and there is no consistency between units or even the same unit over voltage, temperature, power supply impedance and pickup inductance.

[DocBrown]

Thanks for this long and very interresting message.

I've read somewhere that Fuzz faces sound better with non alkaline batteries, now I understand why. My battery is a brand new alkaline Duracell maybe it plays a role somewhere.

Your message help me to realize something I understood a few years ago while I  began to play on a plexi. This amp and the Fuzz face are vintage designed pieces of gear and you can't have the same versatility than with modern gear.

I mean you maybe have a great sound with every setting on the pedal and every pickup of your guitar with a ts9, while there is only one way to get a great sound with a Fuzz face, a sweet spot, an alignment of stars as you said.

I watched a part of the live at woodstock of jimi Hendrix, and I noticed he's on the bridge pick-up of the strat while the Fuzz face is engaged, and turned it off when he goes on the neck position.

Now I realize that my Fuzz face probably doesn't sound too dark and that I just need learn how to find the sweet spot.

I'll tweak it a bit to get the best sound as possible (adding a trim pot for Q2's bias, maybe test other transistors) but I will not go further in mods.

Anyway, even if it's a one sound design it's a good sound, and brings a lot of inspiration.

Thanks for your help.

[DocBrown]

Ok so I made a few tests and every advice you gave me were (obviously) right. But I also discovered there is a light problem with my build :

the sound seems normal until 95% of fuzz. if you pass this limit you hear verry sudently (like if you activate a switch) more noise and a hiss. The problem is that the famous star alignement that make the sound great seems to be with less volume on the guitar but fuzz over 95%...

Do you think the low bias value on Q2 (3.8V) could cause this problem ?

thanks a lot !

[j_flanders]

the sound seems normal until 95% of fuzz. if you pass this limit you hear verry sudently (like if you activate a switch) more noise and a hiss.

Do you think the low bias value on Q2 (3.8V) could cause this problem ?

thanks a lot !

TLDR; No for the bias and the hiss at 100% and dark sound at 95% is normal.

With a 9V supply the maximum output signal is 4,5V peak to peak.
To make sure that nor the bottom nor the top of the wave is clipped off you'd normally bias the collector at 4,5V, right in the middle between 0 and 9V, for maximum up and down swing, to minimize potential distortion caused by hitting the 0V or 9V limits and clipping of the top or bottom peaks of the signal wave.

But in a fuzz face the signal is so heavily amplified that top and bottom of the signal wave already reach far beyond the 0V and 9V limits.
Imagine that if there was enough DC power the top and bottom would go to +100V and -100V (I don't know the exact numbers, could be 50 or a 1000).
But there's only 9V so, of the 100V that it'd like to be, there's 91% chopped off at top and bottom.

In such a case moving the middle point (Q2C) around a volt or two is going to have a lot less impact on the clipping or the resulting tone.
If your signal was 9V then even a couple of milivolts in the 'wrong' direction would be the difference between 100% clean or some distortion (bottom or top clipped off).
But here we're dealing with a signal that wants to be 200V (or whatever big voltage it is in real live) so you're never going to have an ideal middle bias voltage. Bottom and top will be be clipped off so much that changing that bias point will result in changing from 100% symmetrical clipping to very slightly asymmetrical.

Best to stop worrying about Q2C voltage.

About your other 'problem':
Overly simplified (because there is some internal resistance in the transistor), the gain of Q2 is collector resistor divided by emitter resistance.
Fuzz pot fully up, there's 1k 0hm resistance for DC but 0 Ohm resistance for AC/audio (passing the 1k, going through the 22u bypass cap.)
Gain: 8k/0 = infinity
So, turning down the Fuzz pot even just 1 Ohm reduces the gain from infinity to 8000. Turning down 10 Ohm is going from inifinty to 800. That's a huge leap and why it seems like a switch.
It's perfectly normal. My Ge fuzz is bright and hissy when fully up, and a lot darker with the fuzz pot just the tiniest bit down (15 Ohm).
The hiss you hear is because there's so much gain, even for higher frequencies.
Why turning down seems to have more influence on the highest frequencies is explained here:
https://www.diystompboxes.com/smfforum/index.php?topic=123565.msg1169144#msg1169144

[DocBrown]

Thanks for your reply.

So many things that are normal for a fuzz face that could be called "problems" on modern builds. I realize the vintage conception of this unit more and more. I suppose modern Dunlop reissues have an trick to avoid this  high fuzz hiss ?

[DocBrown]

just a few words to keep you in touch. I've lightly moded my fuzz face. I like the fact that's this build is a wild animal, but I also want to have a pedal that I can use almost as easily than a modern one (ie, not to worry about pushing the fuzz pot to 100%).

I study the JH-F1 layout and made the same. My transistors are still BC108 but they're B around 286 hfe for both. Sounds better to my ears.

Anyway, thanks for your help, I'll go to my second build, hoping it will go as smooth as this one.

[amptramp]

I can guarantee your second build will be different from the first.  Consistency is not part of the Fuzz face design.

[DocBrown]

Yep, that's what I learned from all of you. Quite strange at first but pretty fun after !

[DocBrown]

Hummmm a last question about this build gentlemen :

As it is a silicon build it's negative grounded right? So there is no reason that I couldn't add a DC jack to this pedal? I know it will probably sound best on battery and that batteries last for years but I just want to be allowed to choose.

Thanks!

[amptramp]

Hummmm a last question about this build gentlemen :

As it is a silicon build it's negative grounded right? So there is no reason that I couldn't add a DC jack to this pedal? I know it will probably sound best on battery and that batteries last for years but I just want to be allowed to choose.

Thanks!

You use negative ground with NPN transistors and positive ground with PNP transistors regardless of whether it is silicon or germanium.  As I mentioned earlier, there is a sweet spot for battery impedance and the design is so dependent on various things like temperature, voltage, guitar source impedance, load and the phase of the moon, it pays to stabilize these things as much as possible.

[DocBrown]

Oh yes, you're right. My transistors are bc108, NPN so negative grounding. I think I'll make a try.

Thanks a lot.