Vintage Transistors

[Ruptor]

I see that there is interest in AC128 transistors for the FuzzFace and I remembered I had a couple of matched pairs for use in amplifiers. It is funny how memory works with associations because remembering I had AC128s somewhere I remembered I bought packets of obsolete components when I was a kid. I didn't find the AC128s I was after in the first place but I did find one of the packets and guess what it is a bunch of AC128s and another bunch of glass transistors that I found on the NET are probably OC45 or OC44 types. There is about 25 of each but whether they are any good or not I don't know but the diode junctions measure OK. Anyway I could use a couple and make an original FuzzFace although after reading about the variation with them any working unit would be a one off.

[Electric Warrior]

Making a good sounding fuzz face is nowhere near as hard as some people make it sound. Just give it a try.

[Ruptor]

I picked up the FuzzFace spice circuit from the NET and altered it to use an FET that I have lying around. The simulated circuit worked giving interesting wave shapes so I rummaged through my used bits and found the components and an old piece Veroboard. That was a mistake since the old board started to lift all the tracks and after building it I had to find all the breaks that took hours. Anyway when I finally made it work it didn't give any fuzz until the input was 1V so I went back to the simulator and realised I had set 1V input when fiddling about but hadn't put it back to 100mV. The circuit works as simulated when I stick a 1KHz sine wave in to it but it doesn't have enough gain for the normal low level guitar input signal.
I was interested to see what the FET circuit would sound like compared to YouTube demos of original FuzzFaces but it wasn't going to happen as is. Are there any pictures of the waves out of an original FuzzFace? It would be nice to know what represents a good fuzz unit.

[antonis]

It would be nice to know what represents a good fuzz unit.

IMHO, a "good" fuzz unit also needs a "good" guitar pickup and a "good" power amp with a "good" speaker and - of course - a "good" ear, so we have to determine and stipulate the "good" term..

i.e. is a Ge fuzz good for a solid state amp..??
     is a Fet fuzz good for a tube amp..??

[EBK]

It would be nice to know what represents a good fuzz unit.

IMHO, a "good" fuzz unit also needs a "good" guitar pickup and a "good" power amp with a "good" speaker and - of course - a "good" ear, so we have to determine and stipulate the "good" term..

i.e. is a Ge fuzz good for a solid state amp..??
     is a Fet fuzz good for a tube amp..??

Don't forget "good" graphics and a "good", preferably suggestively lewd, name.

Although, if the name is "good" enough, graphics are unnecessary and probably best avoided. 

[Ruptor]

So reading between the lines nobody has any idea what a good FuzzFace should sound like or what the waves look like.  After 50 years of effects pedals I thought there might be a NET page that showed some waveforms of what sounds better. I am sure Jimi Hendrix and the engineer involved with him had an some idea.

[EBK]

So reading between the lines nobody has any idea what a good FuzzFace should sound like or what the waves look like.  After 50 years of effects pedals I thought there might be a NET page that showed some waveforms of what sounds better. I am sure Jimi Hendrix and the engineer involved with him had an some idea.

Have you read through this? http://www.geofex.com/article_folders/fuzzface/fftech.htm
It might give some additional parameters to consider for your model.

[Ruptor]

Yes I have read that and it is interesting about how the transistor gain affects things but there are no wave pictures to relate to the sounds.
I have a solution for my Mosfet version of the FuzzFace that gives rounded edges at low gain and sharper edges with more signal that seems to be variable with the gain capacitor pot in the emitter of the second transistor. Well that is what the simulator says but what it does when I build it I shall find out later.

[EBK]

Does this help at all? http://www.electrosmash.com/fuzz-face
It doesn't provide a comparison of what "good" vs. "not as good" look like, but it does show waveforms and lists the SPICE parameters used.

[anotherjim]

Now that someone asks. Many here have posted screenshots of fuzz waves both real and simulated in various topic threads, but I'm not aware of a dedicated page for such things.
But Geofex does show this, and that makes the point does it not?

You may have to use neck pickup and tone rolled off (bassiest tone) to see a wave like that.

[EBK]

Here is a page with some scope traces: http://www.reocities.com/sunsetstrip/studio/2987/britface.html

[antonis]

Just take in mind that any simulator's waveform is quite different from the actual one.. 

[Ruptor]

Just take in mind that any simulator's waveform is quite different from the actual one.. 

Yes that is what it used to be like but not now I think. I wanted to get deep in to LTSpice and see how useful it is so I decided to use the FuzzFace as a training example because I wanted one and it is easy to build. Having satisfied my self that the original PNP based circuit looked good I put FETs in the simulator, found the best resistors, then I built it and it did exactly what the simulator did but as I said I made a mistake with the input level on the simulator so there was not enough gain for a guitar. A classic case of the model only being as good as the input data.  Anyway I have modified it some more on the simulator and updated the actual circuit that now will work with a guitar input. The input impedance is 100K but I could raise it to 1M if it still loads the guitar so that is the first improvement over the original. The FET design has opened up the possibility of adding another Pot to alter the bias so it can be made to clip on either top or bottom first. It looks like I don't need a volume control so it will still have two Pots but the bias Pot will replace the volume. The unit only needs to output the same level as the guitar puts out so I can't see the need for a volume control on the unit but it could be fitted if needed.
I put a signal generator on the input with a scope on the output and it shows the signal level output is the same for frequencies above 50 Hz dropping down to half as 15 Hz approaches but I am not using it for a bass. The constant signal from the generator let me see how changing the fuzz Pot affected things and how the bias Pot could move the wave between upper and lower clipping levels.
I shall see what it sounds like later and see if it is crap or not but the waves look good to me.

[thermionix]

a "good", preferably suggestively lewd, name.

I thought I had come up with the greatest fuzz name ever..."Whisker Biscuit"...and the plan was to have the fuzz control on the left labelled "Whisker", and the volume on the right labelled "Biscuit"...but then I saw that someone (was it ROG?) has already used that name for a pedal.  Bummer.

[Ruptor]

When I first plugged it in the fuzz and overall volume was low but there seemed to be punch when the sound entered and it immediately went flat like someone turned the volume down. At first I thought it was a ropey Pot on the fuzz control because when I touched it the thing came to life but it turned out to be a break in the old vero board where the track was leaving the board. Anyway once that was sorted it sounds pretty good to me. The Fets give it a crisper feel I think well compared to my 1970s Big Muff Pi it sounds more lively. It might be because the guitar signal isn't being loaded that helps as well. Whether it sounds like an original FuzzFace I don't know but I like it and that is what counts. It is also silent of hum, absolutely none at all, so the heavy but simple supply filtering has paid off. I will have to get some new pots and a bypass switch before I put it in a box but it proves LTSpice is exceedingly useful and has saved me a lot of time because it is easier to delete wires and do circuit mods on the screen than in the real world.

[pinkjimiphoton]

a perfect sounding fuzzface is impossible to describe or attain.
for a fuzzface to be right, it needs to be germanium. pnp or npn doesn't matter at all.

unfortunately, the gain does. too high, and it sounds like poo. too low, and it doesn't fuzz.

the magick gain numbers are all over the web... 70-100ish for q1 and 90 to maybe 120 for q2. even then, i'd advise biasing trimmers.

a starve control is also very useful.

a good fuzzface should clean up spectacularly as you roll back your volume knob... to the point where way down low your guitar sounds almost crystaline.

my advice is dime the fuzz and volume, and control it with the guitar knobs.

a proper fuzzface sound works best with an amp that is breaking up. an overdrive pedal placed after it will work well too. a fuzzface into a clean amp isn't particularly very good sounding.

remember that the fuzzface in a sim is NOT gonna get you the results you need to hear. your fingers, the guitar, the fuzzface and the AMP are all equal parts of the equation. the sim can be real useful, but it can't let you hear what it is.

i f'd around a few years back with a mosfet fueled fuzzface, i called it the sterno face if memory served.

try germanium in q1 and a vn2222a mosfet for q2. for extra weirdness parallel it with another ge q, so the ge and mosfet are wired paralell.
it is a pretty cool sound once biased correctly.

google up "britface" too, cuz with ge you may need to add a diode to help keep the bias stable. fuzzfaces get farty and don't work at all when they get cold, and when they get hot they get crazy over fuzzy. the diode trick between e and b can really help.

people say bias to 4.5v. i say use your ear and bias it to wherever it sounds best to your ears. you should be able to get a full and complex range of sounds from crystal clear to edgy to crunchy to scream to all out muddy filth right from your guitar knobs.

if you add a simple low pass filter to the input that can be useful too. but it all comes down to what you like.

me, it;s gotta clean up good, and it's gotta have some balls. a good fuzzface should be able to provide way above unity gain.

good luck man!

like others have said... its not THAT hard to get one built. to get one that sounds right can take a few years sometimes . but don't give up.

rock on