Mosfet booster

[jaytee]

I've been playing around with the idea of getting a fetzer valve type thing with a mosfet. This is what I have at the moment



I measured a few valves(tubes) to get the high frequency -3db point which was 25-26kHZ.
As it turns out with a 33k input resistor for the 2N7000 -3db is around 25kHZ so no extra cap was needed. I don't know how much this varies I've only tried it with this one mosfet.
I had no idea how to apply the formula at ROG for the source resistor IDSS being a different thing for JFETS. Does anyone know how to do this? I ended up guestimating it from simulation trying to get matching curves and harmonics. The resistor seems like a compromise and is more difficult to get the right value as the gain gets higher.
The gain from this is around 8 times.
It has a nice sweet sound and overdrives my amp quite nicely
The only gripe I have about it is it starts to clip with my guitar full up. I wanted something that would stay clean so I might have a look at making an 18 volt version or maybe adding a gain control to this one. But then the source resistor would have to change.
Any comments are welcome.
 

[petemoore]

 Looks good, what is the drain bias voltage?
  that kind of setup works great, I had similar boost stage w/mosfet, but made R5/R3 a pot, it's wiper would feed the large resistor to allow biasing the drain to 1/2v or whatever you choose.

[jaytee]

The drain is about 5.8v. I think I was lucky with the FET because it came out just how I simulated it. The bias adjustment sounds like a good idea.

[markm]

Nice work jaytee, I love little MosFET circuits like this one!
I just thought I'd give you the thumbs up and compliment you on your work, since everyone else is! 

[Gus]

Yes mosfets are fun.  A cap at the 1 meg to bias string might help.  The source leg can be a resistor or a cap bypassed resistor or a setup like yours.

It took some time but people seem to trying this circuit

http://www.diystompboxes.com/pedals/mosfet.gif

Not a big deal something I built for a harsh sounding solid state amp I had one time built with what I had at the time extra power mosfets and parts.  Can be fun with bass.  I would change the 100K vol to a 10K vol.

Standard mosfet biasing(I used a regulated supply that is why the two bias resistors)  I run the POWER mosfet at higher that "normal" current because of using a supply.  The first power mosfet gate cap is what can take the edge off when feeding a harsh sounding amp.  So like yours using the series 33K to the gate cap of the 2n7000.

  Now one needs to think about noise.

Other fun stuff a book from about 1980 has mosfet circuits one is a single ended transformer out 4 watt jfet mosfet amp IIRC.  I been looking for my copy for the ISDN number It was something like  VMOS circuit Blackstone publishers?

[Barcode80]

throw a couple of Ge diodes on the output and i bet you'd have a nice sounding OD!(i might try this  )

[jaytee]

Hello fellas thanks for the comments. I was after a clean boost really, something sweet and clean because I like the way it made my amp sweet and dirty  I made up this 18 volt version that I like much better. It just pushes my amp getting the valves rattling. I don't seem to be able to get it to clip which is what I wanted. The gain is around 10 and it gives out 12v peak before clipping. I gave it a name which seems customary. Just another boost I guess but it was enjoyable working on it 
The pot should be 100k LOG not LIN.

[jaytee]

Yes mosfets are fun.  A cap at the 1 meg to bias string might help.  The source leg can be a resistor or a cap bypassed resistor or a setup like yours.

It took some time but people seem to trying this circuit

http://www.diystompboxes.com/pedals/mosfet.gif

Not a big deal something I built for a harsh sounding solid state amp I had one time built with what I had at the time extra power mosfets and parts.  Can be fun with bass.  I would change the 100K vol to a 10K vol.

Standard mosfet biasing(I used a regulated supply that is why the two bias resistors)  I run the POWER mosfet at higher that "normal" current because of using a supply.  The first power mosfet gate cap is what can take the edge off when feeding a harsh sounding amp.  So like yours using the series 33K to the gate cap of the 2n7000.

  Now one needs to think about noise.

Other fun stuff a book from about 1980 has mosfet circuits one is a single ended transformer out 4 watt jfet mosfet amp IIRC.  I been looking for my copy for the ISDN number It was something like  VMOS circuit Blackstone publishers?

Hello Gus. I'll certainly try the cap at the bias string it can only help things. I think the vol pot could be lower, 100k is what was handy. Maybe 50 or 20k.
The 33k resistor is chosen to give the same high frequency roll off as a tube input stage in an amp. The usual thing having around 34k to the grid of the tube. The resistor gave me the same roll of as I measured on a tube so I'm guessing the capacitance must be about the same as a 12AX7.
The unbypassed source resistor is meant to give just enough degeneration to move the square law mosfet transfer characteristic to the 3/2 law of a triode. A triode being naturaly more linear than a mosfet. So its approximating the response and sound of an input tube stage. An input stage won't go into clipping though with a normal guitar input so I didn't want it clipping. Also it won't clip like a tube when it does.
With the 9 volt version I came to a value of 100 ohms which looked about right. I upped it to 150 ohms just to reduce gain a bit. Runoffgroove has a formula for this to apply to a jfet but I don't know the maths for a mosfet. I got it by simulating the mosfet comparing the curve to a 12AX7 and doing some fourier analysis the see what the harmonics were like.  So I guess it depends how accurate the models are. Like the fetzer valve idea but with more gain. That's why I have a fixed gain on it. On the 18 volt version I just sized the resistor to get the gain I wanted so it will be more linear than the 9 volt one. I have a feeling the source resistor is related to the gm and Vt of the fet. Maybe a fet with lower gm would be usefull to play with.
So hopefully it's like putting an extra tube gain stage in front of the amp to drive it harder.