Differential fuzz???

[Alpha579]

Hey all,
Whats the theory behind this differential fuzz idea? The Boss Blues Driver does a similiar thing with JFets. How does it work?
http://www.muzique.com/schem/diffuzz.gif
http://home.hawaii.rr.com/stompbox/Bd-2_1.jpg

[Transmogrifox]

The BD-2 is slightly different in that it doesn't actually use the JFETs for distortion.  That BD-2 set-up i what an op amp looks like inside.

The differential fuzz overdrives and saturates a differential pair amplifier.

In both cases, the differential pair is used.  The theory is that the resistor at the junction of the two emitters sets the current through the two transistors, and if they're well matched, they both have equal current when the voltage on the base of both transistors is the same.  In fact, it doesn't even matter what that voltage is, as long as it's the same.  The difference of the outputs does not change.  This idea is known in op amps as 'common mode rejection'.  As with any system, this common mode rejection is not perfect.  An op amp biases the differential pair with a current mirror so that the voltage at the transistor bases has even less effect on the biase current through the pair.

So here's how it works:  In theory, the ideal differential pair is biased with a current source that does not change (realized with a large-valued resistor in differential fuzz), so the total current between the two is the same, no matter what.  When there is a difference between the voltages on the bases of the two transistors, one of the pair begins to turn off while the other turns on, so when the voltage difference gets large enough, one transistor is sinking ALL of the bias current...reaching a maximum limit.

This usually happens at a differential voltage of 100 mV, so you can see how the differential fuzz works now.  The first 2N5088 works as a high-gain preamp to drive the differential pair well beyond +/- 100mV differential voltage.   I haven't analyzed it to see if the differential pair also saturates, but if it does, it happens well after the differential pair limits out.

So basically, there is a limited amount of current divided between the two 2N3906 transistors, and changing the voltage at one of the bases controls how much of the current is going through either one of the transistors, and when it maxes out, the signal is clipped, and when it turns off, the signal is clipped--->Differential fuzz

[Ge_Whiz]

From another angle, the differential amplifier or 'long-tailed pair' configuration has both a high input impedance, and a softer clipping characteristic than single transistors. In my experience, the distortion is more readily adjusted from a soft overdrive to full fuzz, although this conguration on its own will not give you either JFET overdrive, nor 'metal'.

I haven't built the diff fuzz circuit that appears around the Internet, but the 'Guitar' magazine "Tube Bender" uses a similar circuit. While not producing my favourite sounds, it is the most versatile distortion box that I have.

And no, I won't share the circuit as there are a couple of guys trying to eke out a living from it. But the 'Differential Fuzz' circuit is similar.

[Alpha579]

lol, i used to have that mag. (the one wiv the two riviera amps in it, right?), but i lost it at school(still reckon it was nicked  :evil: ...).

[swt]

there's a schem at keeley's site of the fuzzhead, his version of the fuzz face, that uses a diff distortion. It's interesting and i guess it sound great.

[Phorhas]

Well how about this - putting a real tube in that OA configuration and sue it as the heart of a circuit in the liking of a TubeScreamer, or SonicDistortion... or any OA based pedal... (and maybe, just maybe - it'll even help the TubeWorks TubeDriver sound goood... or at least, deasent LOL )

So what do ya think... anyone up to try?

[Alpha579]

Ooo, thats sounds insanely hard...might have to try that...

[brett]

Hi.  RE the differential distortion, could someone explain the 0.01uF cap?  Is it to ground some highs?  What is the 470k resistor doing?  Does it promote differences in the base voltages of the 2 2N3906s?

Sorry, but the explanation above was as clear as mud to me. :oops:

[rocket]

Hi. RE the differential distortion, could someone explain the 0.01uF cap? Is it to ground some highs? What is the 470k resistor doing? Does it promote differences in the base voltages of the 2 2N3906s?

the 470 k resitor provides the bias voltage to the 2nd transisor in the pair.
the differntial pair amplifies the differential voltage between the 2 PNP transistors' bases.
as the 0.01 Cap shorts the signal at the 2nd base to ground, otherwise there would be no difference.

oh - the explanation still seems muddy :wink:

[brett]

Thanks for thje explanation - I'm slowly understanding it.

By the way, I played around with a differential d circuit today.  It seems to work a lot better if driven harder.

I used a 10uF cap to bypass the emitter resistor on the 2N5088/2N3904, and reduced the 470k resistor to 47k.  That sounded a LOT better to my ears.  It still had some of that post-attack "bloom" in harmonics, but didn't muffle the sound as much as with the 470k resistor.

I'll post a schematic sometime next week.

cheers

[petemoore]

That schematic just 'looks kool' because of the Symmetry in it.
 I look foreward to seeing posts on this one!