A MOSFET powered Bazz Fuss :)

[mac]

In a recent post a I saw a schematic having a mosfet biased by a reverse diode.
The tiny leakage current of the diode is enough to turn the mosfet on because its gate needs a few electrons.

OTOH, in a well tuned Bazz Fuss (or Arsenio Novo distortion), the collector voltage Vc must be equal to the base voltage Vb. This is why a darlington BF sounds better than a 2n5088 or BC550, it needs a lot less current across the diode so Vc tends to Vb with a smaller collector resistor. A mosfet is similar to a darlington in this aspect.

Note:
I wrote about it and posted a circuit that it is in my gallery. If you use a simple transistor, add a big resistor at the base and tweak the collector resistor to find the sweet spot.
For example, if you have hfe=100, Rb=1M the tuning condition Vc=Vb gives,
ic*Rc = ib*Rb =ib*hfe*Rc,
Rc = Rb/hfe = 1M/100 = 10k  
And yes, the tuning condition depends on temperature  

So, what if I replace a darlington with a mosfet?
I went to the breadboard, picked a BS170 and it worked fine, much like the darlington version.
But you need a pair of back to back diodes, silicons or leds.
I also used a 100k at the drain with a small change in sound.

If you want a kind of gain control you can add a drain to gate resistor of some mega ohms or less, and/or a source resistor, variable or wired as in a Fuzz Face.

I ran a simulation with an input cap a bit high for my taste, and a power mosfet model. It might roll off more highs than a BS170.
This is the circuit, including a simple guitar mic and cable capacitance.



And the output signal,



Enjoy.

mac

[PRR]

Your images do not show. When I access the URL directly I hit a Microsoft log-in.

Use imgur.com or other public host.

[mac]

I uploaded the files in a private folder in OneDrive, moved them to a public folder and modified the post  
Can you see them now? I can in incognito mode after logging out and closing outlook.

BTW, I can't upload in the Layouts Gallery, I get *error 7*, can't write to disk.

mac

[bool]

Still no-show from your pictures. But yes, a BF mosfet derivative definitely works.

[antonis]

Am I the only one who sees the pictures..

(but I can't see any distortion - perhaps because of the 250mV input sin..)

[mac]

 





mac

[duck_arse]

I see everything twice!

why the 0.001 for Rs?

there is a "can't post to gallery" thread, aron say it's full.

[ggedamed]

[...]
why the 0.001 for Rs?

I think the sim chokes when the resistance is 0.

[mac]

why the 0.001 for Rs?

I think the sim chokes when the resistance is 0.

LTSpice doesn't like 0 resistance.
But the purpose of Rs is to control gain. I ran simulations with different values of Rs.
As I said in the first post, you can put a 2.5K or 5k pot there wired as a variable resistor.

(but I can't see any distortion - perhaps because of the 250mV input sin..)

The input signal is limited and rounded by the diodes. The output looks *like* a sine wave but it is not, it's been distorted, even though the 250mV of the voltage source drops 90% at the *in* node.

mac


mac

[antonis]

The input signal is limited and rounded by the diodes. The output looks *like* a sine wave but it is not, it's been distorted, even though the 250mV of the voltage source drops 90% at the *in* node.

O.K. -  although your screen is at 1mS interval..
(at 500μS or better at 250μS it should be more "realistic"..)

[midwayfair]

You can also put a MOSFET in there instead of the two diodes -- this makes it more asymmetrical like the original Bazz Fuss, but not quite as insanely so.