Using MosFETs or JFETs in a "stock" NPN Fuzz Face

[culturejam]

I did a search but couldn't find exactly what I was looking for, so.........

I recently built a NPN Fuzz Face with a GGG pre-made PCB. Love the sound with Ge trannies. But then I saw the ROG Multi Face and was intrigued by using MosFETs and JFETs in the circuits.

I put two 2N7000s in the sockets. Nothing but hum. I switched them out in every possible combination. Nothing but hum. I played around with the trimmer. Hum.

Looking at the two layouts/schematics (FF vs. MultiFace), to me they don't look all that different.

What am I missing here? Will it just not work in a NPN Fuzz Face?

Thanks

[soulsonic]

Here's a design from Jack Orman that uses a MOSFET.
http://www.muzique.com/schem/fuzzface4.gif

[culturejam]

Here's a design from Jack Orman that uses a MOSFET.
http://www.muzique.com/schem/fuzzface4.gif

Is that your way of saying that it won't work for me to use MosFETs in a stock Fuzz Face? 

[mojo_hand]

If you did it with something like a suitable J201, the second stage would work, with a little trim pot twiddling.  The first stage seems considerably more doubtful to me, since features (like the feedback loop) were designed to work with bipolars.

[culturejam]

If you did it with something like a suitable J201, the second stage would work, with a little trim pot twiddling.  The first stage seems considerably more doubtful to me, since features (like the feedback loop) were designed to work with bipolars.

Sweet! I'll give that a try. Thanks!

[R.G.]

Think about bias.

[culturejam]

Think about bias.

That's deep, R.G.    I'll read up on the difference in biasing bjts and FETs. Maybe there's an article on that subject on your site....


Mojo_hand:

I put the MosFET in Q2 and it worked great. I adjusted the bias, but it sounded best almost exactly where the bias for the Ge tranny was. Thanks.

Interestingly, on one end of the trimmer spectrum, there is a very low-speed oscillation thing happening. Almost sounds like a phaser!

[JDoyle]

That's deep, R.G. 

It's more than deep, it is the entire reason your circuit didn't work. 

I'll read up on the difference in biasing bjts and FETs. Maybe there's an article on that subject on your site....

Google search 'JFET biasing' or look up JFET on wikipedia, or any beginning EE textbook. There is an enormous difference between the two.

[DougH]

Dude, what's that little metal box with the plastic thingie plugged into it? Do I really need that? It won't work without it for some reason. I'll do a search on it tomorrow...


( )

[culturejam]

Dude, what's that little metal box with the plastic thingie plugged into it? Do I really need that? It won't work without it for some reason. I'll do a search on it tomorrow..

You know, I'm not *trying* to ask stupid questions. I already searched on what I was previously asking about and couldn't find anything that was totally relevant (as far as I could tell). I read several FF articles trying to figure this one out before I posted.

Trust me, asking a question here was the the last thing I wanted to do, and your response is exactly why.

[DougH]

No offense intended. I was just having a little fun. Note the green smileys.

Seriously, though- with a transistor circuit, bias is almost as important of an issue as power. If the bias is not set up correctly, you may as well not have any power either as the circuit will not work (as you found out). I would suggest reading up on biasing techniques for transistors, JFETs, and MOSFETs. You can substitute MOSFETs/JFETs for bipolar junction transistors if you tweak the bias accordingly.

[Gus]

IIRC J.G. had a BJT MOSFET circuit on the web years ago.

[culturejam]

No offense intended. I was just having a little fun. Note the green smileys.

No problem. Sometimes the "Internets" can ruin the subtleties of a good joke (even with emoticons). RG's early response reminded me of a Zen koan (ie - Two hands clap and there is a sound; what is the sound of one hand?), which is why I said "that's deep." The statement was so simple and yet the depth of it is astounding. I suppose that joke didn't go over either.

Seriously, though- with a transistor circuit, bias is almost as important of an issue as power. If the bias is not set up correctly, you may as well not have any power either as the circuit will not work (as you found out). I would suggest reading up on biasing techniques for transistors, JFETs, and MOSFETs. You can substitute MOSFETs/JFETs for bipolar junction transistors if you tweak the bias accordingly.

Thanks. That's quite helpful. I "sort-of" understand bias...in a very basic, generic sense. Just playing around with trimmers is teaching me a lot, but it is good to understand the underlying theory. I'll get to reading and see if I can stir up an epiphany. 

[Dragonfly]

Think about bias.

I NEED to make that a bumper sticker !

It applies to way more than just Mosfets, Jfets, etc !

[culturejam]

It applies to way more than just Mosfets, Jfets, etc !

See!? I told you guys it was deep! 

[vortex]

Here's a less "deep" answer.
Check out Tim Escobedo's "Many Faces of Fuzz"

http://www.geocities.com/tpe123/folkurban/fuzz/fuzzyfaces.html

[mac]

Correct me if I'm wrong. A fet at Q1 need a resistor at the source with a bypass cap because the gate is at about 0.5v, being the gate resistors the 100k and the drive pot.

mac

[JDoyle]

Correct me if I'm wrong. A fet at Q1 need a resistor at the source with a bypass cap because the gate is at about 0.5v, being the gate resistors the 100k and the drive pot.

mac

Bingo! Q1 is biased by the voltage at the emitter/source of Q2 so if the source of a Q1 JFET is connected straight to ground, the gate will ALWAYS be higher and the gate-source diode will be biased on, effectively shunting your signal to ground without changing the conduction state between drain and source...

A way to keep the gain and get the same effect as adding a source resistor bypassed by a cap, would be to put a diode in series with the source, raising it's voltage without adding negative feedback (or very little) as a resistor. You will also get a different type of distortion as the diode's effective resistance changes with the current through it...

Jay Doyle

[culturejam]

Bingo! Q1 is biased by the voltage at the emitter/source of Q2 so if the source of a Q1 JFET is connected straight to ground, the gate will ALWAYS be higher and the gate-source diode will be biased on, effectively shunting your signal to ground without changing the conduction state between drain and source...

A way to keep the gain and get the same effect as adding a source resistor bypassed by a cap, would be to put a diode in series with the source, raising it's voltage without adding negative feedback (or very little) as a resistor. You will also get a different type of distortion as the diode's effective resistance changes with the current through it...

I think something might be starting to click in my noggin. I'm going to look at a few more schematics and reread the chapter on FETs in my electronics book.

Thanks for the help.

[petemoore]

  So how do you or don't you get the jfet source above gate in FF Q1 position ? [forget Q2 for now?]
  First off take the 'ground clamp' from it, perhaps introduce a resistor or resistor w/bypass cap here ?
  Then see what the voltages look like with a resistor [<1k? .. >2k7??]..use 5k or 10k pot { between Q1E and ground, should lift source bias.
  [I don't really understand but] Q1 input [base/gate] isn't tied to ground, so maybe it floats up when the source bias is raised, making it difficult to bias Jfet there.
  A separate bias string for the source maybe ?
  I went back to using bipolar transistors.