Mosfets as Clippers...

[v8pete]

Hi folks - so, other than the added headroom gained by using a Mosfet's internal anti-parallel diode as a clipping diode, assuming that this is slughtly greater than a 'regular' silicon diode say - although I can't really see why that would be, as it is just a silicon diode anyway,  does it really give any extra 'mojo' to the sound ? I've read about it supposedly having a larger capacitance, but is this really true ? - can't imagine that in the case of someting like a 2N7000 that it's really that significant at audio ? If you take something like a Zendrive, and stick in a couple of 1N4148s in place of the Mosfets, do you think that you can really tell the difference?? Interested to hear what you think !

[MikeH]

Welcome!  Try it out and see... I think there's a big difference.  Regardless  of whether it's just silicon, I think it has more to do with forward voltage.

[v8pete]

Hi - great forum BTW! Sorry about my spelling, but it's getting late here & I've had a few more beers than planned already :-) Sound-wise, I can't really hear a lot (any?) difference - the only way to do this properly would be the so-called doulble-blind test, but that's too much hassle... I just wanted to see if there is any real 'engineering' reason why they should acutally sound different. As far as I can see, the body-diode is just another silicon diode, that's probably going to behave very much like a 1N4148 at the kind of foward currents we're taling about. Now if Mosfet action came in to this in the reverse direction, I'd see what all the fuss was about, but as far as I can tell, this can't happen as the other clipping device (eg the other parallel Mosfet or Mosfet +diode) will clamp well before the threshold voltage is reached for the other device - and anyway, something like the Zendrive uses diodes in series with the Mosfets too, so these will be reverse-biased in any case?! I'm just trying to use a bit of science to sort out the problem, as its hard relying on ones ears all of the time :-)) I could well be missing something here, so I'd be very interested in hearing whether people have any strong views. But basically, at the moment, I can't see any difference between a Mosfet clipper & one made with silicon diodes - also you could always 'pad' with some additional series resistance to get the same volt drop at a given forward current.

Pete

[R.G.]

The body diode is indeed just another silicon diode junction.

The diode-connected characteristic of the device with the gate shorted to drain is quite another matter. That's the one that's interesting.

[v8pete]

I agree that it's a nice asymmetric clipper, if the circuit in which it's used allows it to conduct in the forward direction...but in something like the massively-hyped Zendrive, this kind of action isn't able to happen...so in the latter case, I can't see that it's doing that prevents it from being replaced by a simple silicon diode?

[WGTP]

http://www.diystompboxes.com/smfforum/index.php?topic=38581.0

Check this out for lots of info.   

[JHS]

I replaces the MA150 diodes in a TS-clone with 2N7000 for testing, but only used the body diode for the clipping (app. 0,7-0,8V compared to the 0,6V of the MA150). The gate was not connected, a bit different than the hookup in the OCD. Generated that typical MOSFET sound with those unnatural highs, similar to a IBZ Mos-Screamer, SHO, or other MOSFET-Boosters. A bit more headroom and rawer OD-sound, sometimes quite useful. Sounds really good if a AM150 is paired with a 2N7000, a bit more asym. clipping not so extrem trebly like 2 2N7000. A Ge-diode in series with one of the M-Fets smoothes the sound, a bit more musical sounding compared to 2 M_Fets w/o this diode.

I swapped the new BS170 in a SHO-clone for a very old one w/o body diode some time ago, IMHO the old one sounded more musical and don't break up so harsh when overdriven.

As an alternative to the 1N915/4148 try 1S113 and 1S1588, the 1S113 sounds a bit fuzzy, the 1588 really musical.

JHS

[R.G.]

Never, ever leave the gate of a MOSFET open. The impedance is so high that it will pick up stray RF and 60Hz hum from the air as well as more mundane things like the voltage gradient in the air between two copper terminals at different voltages. All this is so inconsistent based on the variability of air as a conductor that it's unpredictable. Always tie the gate SOMEWHERE. Even more important with CMOS logic.

[v8pete]

Thanks guys - some interesting points & information !

Cheers,
Pete

[rhdwave]

Never, ever leave the gate of a MOSFET open. The impedance is so high that it will pick up stray RF and 60Hz hum from the air as well as more mundane things like the voltage gradient in the air between two copper terminals at different voltages. All this is so inconsistent based on the variability of air as a conductor that it's unpredictable. Always tie the gate SOMEWHERE. Even more important with CMOS logic.

In the case of using a MOSFET for clipping in say a tubescreamer in the diode section.  How would you connect it? Where would be a good place to hook the gate to?

Much thanks!

[rhdwave]

Sorry, i actually saw there was a link in the thread which gave me all the info i was looking for.  Peace...

[dano12]

FWIW, I think the Zendrive is more interesting because of part choices that do pre and post filtering, not so much the clippers. The clippers are a mostfet/schottky asym pair.

[mat]

FWIW, I think the Zendrive is more interesting because of part choices that do pre and post filtering, not so much the clippers. The clippers are a mostfet/schottky asym pair.

What do You mean by that, could You explain more of the 'part choices that do pre and post filtering'

Matti T.