Basterdised Fatpants2/Overdrive w/ Mosfet clipping. What you guys think?

[hammerheadmusicman]

So, in an effort to learn some more, i thought i'd try to hack a few things together. I had this idea of taking the fatpants v2, and feeding it in to an overdrive.

So.. i did it, it may be all kinds of wrong, hence posting it on here, constructive criticism/help/advice welcomed with open arms .

So it's a fatpants, minus the volume control, i put a 50k resistor inline instead as that should be roughly around the 12 o'clock (maybe - no way to test things at the moment) as i want it as a preamp, with the fat control, but not the volume (and want it not to boost, or maybe a touch, depending on how it affects the overdrive circuit). Then it feeds in to a TS style dual op amp overdrive, the first half having diode clipping. The second half having one of Jack Orman's modded TS9 tone controls, and in the middle i put some mosfet clippers on a swicth before the tonestack.


This is really just an experiment, and to try and do some learnin' whilst i have the next two months away from building.

Thanks Dudes

George

[hammerheadmusicman]

Oh yeah, if there are any silly mistakes, it's because i am extremely tired as i was doing this, just couldn't put it down! Will check through it tomorrow.

[midwayfair]

Your MOSFETs will clip as normal silicon diodes in that arrangement. You need a blocking diode on each one to prevent it from reverse conducting. (Yes, both. The full drive etc. do it wrong.)  I'm on a mobile so it's hard to link to the required reading, but if you take a look through my mossy Sloth build doc, I link to a thread explaining it in overwhelming detail.

I don't think you should omit the input buffer. You may get some unusual behavior from the Fatpants side of things, and you won't get any of the 'son of the screamer' benefits on the ts side.

[kothoma]

You don't need a DPDT for SW1. A SPDT on the path to the clippers is all you need.
But you could use a DPDT that way to switch between different clippers.

I think your TS has lost the mid hump. Is this intended?

If you use only the left MOSFET you get asymmetrical clipping to about -0.7V...2V depending on the type.
It clips harder at -0.7V and softer at 2V.

If you use both as in your drawing then you get a combined clipping to about -0.7V...0.7V much like two Si diodes but switching slower and having higher capacitance than your common 1N4148.
So this is not wrong per se if you're after that sound but not that much different from antiparallel 1N4148s.

You could use the two MOSFETs in (anti-)series to get high swing symmetrical soft clipping.

You could add diodes with low forward voltages (Ge, Schottky) in series with the MOSFETs as midwayfair suggested to stop them conducting in direction of the body diode. This would give you high swing symmetrical soft clipping, too.

[hammerheadmusicman]

I will probably put a light on to indicate the MOSFET clipping mode being on. When you say diodes in series do you mean like the 8th example on this link?

http://www.muzique.com/lab/zenmos.htm

Thanks

George

[kothoma]

I will probably put a light on to indicate the MOSFET clipping mode being on. When you say diodes in series do you mean like the 8th example on this link?

http://www.muzique.com/lab/zenmos.htm

Well, no. That's for Zener diodes and is much like example 2 of that article (antiserial). But great Zener ideas here!

Here's how: http://www.muzique.com/schem/shaka5.gif i.e. the diodes need to point the other way round.

[marauder]

RG Keen posted this on DIY.

"There are two ways to get "diode connected out of a MOSFET.

One is to use the substrate diode, which always allows current to flow backwards to the way it flows from drain to source. This diode is a normal silicon diode characteristic, consisting as it does of just the substrate isolation tubs.

The second is the one we usually want from MOSFETs. It consists of connecting the gate to the drain, and then using the drain-source as the diode in the normal current flow direction. Connecting the control node to the incoming power node (as in base to collector in bipolars) is the way to get the "diode characteristic" of the control node. It's done in bipolars to get a more-ideal diode characteristic than you get from either of the normal junctions. But this is the connection you want: gate to drain, then drain and source as the diode. Don't clip any leads off the MOSFET.

There is a problem here; the reverse/substrate diode prevents this "diode" from blocking in the reverse direction. In one direction it electronically looks like the amplified-diode MOSFET you want, and in the other it's an ordinary silicon diode. You can use an external diode in series with the drain or source to block the substrate diode from conducting, or if you're using two MOSFETs, you can connect them so the substrate diodes themselves are what allow current flow in the "right" directions each way."

Found this when building a Zendrive, that didnt connect them correctly either, might as well use a silicon diode.

i need to try the real mosfet clipping method, it will clip at the mosfets gate threshold voltage, higher than a LED for some mosfets.

[kothoma]

The second is the one we usually want from MOSFETs. It consists of connecting the gate to the drain, and then using the drain-source as the diode in the normal current flow direction.

Yep.

In one direction it electronically looks like the amplified-diode MOSFET you want, and in the other it's an ordinary silicon diode.

That's why you get asymmetrical clipping to -0.7V...~2V on a single MOSFET with G+D connected.
(Or ~-2V...0.7V if you turn it around, of course.)

You can use an external diode in series with the drain or source to block the substrate diode from conducting,

See the Shaka5.

or if you're using two MOSFETs, you can connect them so the substrate diodes themselves are what allow current flow in the "right" directions each way."

That's putting them anti-serial (see ex. 2 of http://www.muzique.com/lab/zenmos.htm).

Found this when building a Zendrive, that didnt connect them correctly either, might as well use a silicon diode.

Hm, every schematic I've seen so far the Zendrive does it right. My clone too.
But the OCD/EgoDriver look "wrong" with only one diode. Or is it intended?

[midwayfair]

That's the RG post I was referring to last night. Here's the whole thread (RG's original post is quoted, and then he goes into more detail later, plus discussion from others):
http://www.diystompboxes.com/smfforum/index.php?topic=90474.0

The Zen does it right, at least on the schematics I looked at.

The Fulldrive pedals don't. (They in fact waste two MOSFETs and a Germanium diode.

[hammerheadmusicman]

Thanks dudes, will give it am in depth read, apart from the MOSFET clippers, what do you think to the rest of it, or even the concept as a whole..

[kothoma]

VR=9V?
And shouldn't pin 3 get biased too? After a coupling cap?
And the clipping should go to VR, or you need decoupling here?
Not sure about this DC biasing business.

Edit: Don't have an opinion on the overall concept, yet.
But what are your design goals?
I remember you saying your Fatpants is always on?
So wouldn't is make sense to put it together with other always-on-stuff, say a noise gate?

[RobA]

That's the RG post I was referring to last night. Here's the whole thread (RG's original post is quoted, and then he goes into more detail later, plus discussion from others):
http://www.diystompboxes.com/smfforum/index.php?topic=90474.0

The Zen does it right, at least on the schematics I looked at.

The Fulldrive pedals don't. (They in fact waste two MOSFETs and a Germanium diode.

Can you point me at the schematic of the Zen drive you have? Because, when I look at the schematics, they all look like they do it wrong to me. They all look to have the series diode with the cathode pointing in to the source of the MOSFET. But the body diode of the 2N7000 goes from source to drain and the MOSFET clipping should be D+G to S. So, you should need the series diode to block the body diode and let the signal through the D+G to S path. Right?

Moreover, when I breadboard what I see in the schematics, the clipping is always at the level of two series diodes and not the MOSFET plus a series diode. If I put the arrangement in the way that seems to follow from what Keen says, to me that goes with the cathode attached to the G+D connection and then the complete MOSFET diode would be anode of diode to source of the MOSFET. When I breadboard this arrangement I get what seems to me to be the clipping levels I would expect from the MOSFET clipper inline with a series diode 2.3V to 3.0V or so.

Given that, the arrangement of diagram 2 here http://www.muzique.com/lab/zenmos.htm would seem to be the most sensible assuming you didn't mind the slightly higher forward voltage that you get because of the 0.6V drop of the body diode. Has anyone actually tried this in a circuit?

Of course, this all assumes I understand the point of the MOSFET clipper and that's far from sure because all of the schematics/breadboarding have me confused.

[midwayfair]

Can you point me at the schematic of the Zen drive you have? Because, when I look at the schematics, they all look like they do it wrong to me.

The one here:
http://revolutiondeux.blogspot.com/2008/05/zendrive-project-at-fsborg.html

I'm pretty sure it's the same as Bean's schematic, too.

[kothoma]

They all look to have the series diode with the cathode pointing in to the source of the MOSFET.

...using a n-channel MOSFET like 2N7000 or BS170.

But the body diode of the 2N7000 goes from source to drain and the MOSFET clipping should be D+G to S.

According to Jack Orman in http://www.muzique.com/news/mosfet-body-diodes/ it's the other way round:

The top body diode can now conduct from drain to source [...]

Now I'm curious. Need to read up more.

Edit: And I did.
R.G. Keen in http://www.diystompboxes.com/smfforum/index.php?topic=38581.0 :

One way it conducts from source to drain, and looks like an ordinary silicon diode; that's the substrate diode. The other way it conducts from drain to source, and is the "MOSFET diode" caused by the gate making the channel conduct.

and consequently:

N channels need the diode pointing (anode ->|- cathode) in the direction from drain to source.

Looks like R.G. Keen contradicts Orman, and your breadboard proves Keen.
So the Zen and Shaka5 only use the body diode in series with a Ge diode? Funny, I like my Zen as it is.
Maybe the higher forward voltages wouldn't work well in a 9V build with non-rail-to-rail op-amps anyway?
I'm far away (spatial and temporal) from my breadboard and mental gymnastics is hard in summer...

Edit2:

Given that, the arrangement of diagram 2 here http://www.muzique.com/lab/zenmos.htm would seem to be the most sensible assuming you didn't mind the slightly higher forward voltage that you get because of the 0.6V drop of the body diode. Has anyone actually tried this in a circuit?

I've tried example 5 (two Zener in anti-series). That's the same idea. I tried 2.7V and 3V Zeners, not in a to-ground-distortion configuration but in a feedback-loop-overdrive manner, much like in a Tech21 character series pedal. You need a rail-to-rail opamp for this, or the opamp clips way before the diodes.

Anyway, the best thing is: using two MOSFETS or two Zeners in anti-series you don't have to think much...

[RobA]

Thanks for the link Jon. That does line up with all the others that I've seen.

...
According to Jack Orman in http://www.muzique.com/news/mosfet-body-diodes/ it's the other way round:

The top body diode can now conduct from drain to source [...]

...

I think he may just be using conduction in the inverted sense there. His diagram shows it the right way around (the bright red diode between body/source connection and the drain). Also, this aligns with the diagram of the body diode on the Fairchild 2N7000 spec sheet.

From there, just looking at the flow form the body diode indicates to me that the blocking diode has to be oriented the opposite direction to the body diode.

So the Zen and Shaka5 only use the body diode in series with a Ge diode? Funny, I like my Zen as it is.
Maybe the higher forward voltages wouldn't work well in a 9V build with non-rail-to-rail op-amps anyway?

Yep. The next step in looking at this is breadboarding a simple distortion to try it and the first thing I thought was that I'll have to go to more headroom to get past the distortion in the op-amp and to the clipping levels of the anti-series mosfets. So, I'm doing it as split-rail. It might even need a couple of gain stages to get it to the right level relatively clean before the clipping and then a stage to knock it back down to a reasonable level for output.

I'm fiddling with it now and I'll report back once I hear something.