Moon LNDer overdrive distortion - depletion mode MOSFET pedal design

[POTL]

hi Jonny Tell us about your impressions and experiences. I see the advantages of the LND150 over the Jfet in that there is no need to adjust the offset, apparently the minimum variation in parameters. How is the LND150 doing with the usual MOSFet problems? I noticed that you used a zener for static protection, is this transistor protected? I did not see the input capacitor in the first stage, usually such transistors do not work without a capacitor at the gate. What can you say about the difference in sound compared to regular Jfet? I was disappointed with the sound of the BS170 / 2N7000 due to their capacity.

[Fancy Lime]

hi Jonny Tell us about your impressions and experiences. I see the advantages of the LND150 over the Jfet in that there is no need to adjust the offset, apparently the minimum variation in parameters. How is the LND150 doing with the usual MOSFet problems? I noticed that you used a zener for static protection, is this transistor protected? I did not see the input capacitor in the first stage, usually such transistors do not work without a capacitor at the gate. What can you say about the difference in sound compared to regular Jfet? I was disappointed with the sound of the BS170 / 2N7000 due to their capacity.

The datasheet is your friend:
https://www.microchip.com/wwwproducts/en/LND150

The LND150 has an internal gate ESD protection diode, so no need for that extra Zener you'd use on BS170s. Input capacity is 10pF max. The LND150 is a depletion MOSFET and therefore needs to be biased like a JFET, not like a enhancement MOSFET (e.g. BS170). Depletion devices in self bias configuration are biased to ground like triodes and therefore do not need a DC blocking cap at the input.

Andy

[POTL]

thanks Andy it remains to understand if there are any advantages in sound (frequency and dynamic range). I know the LND150 has been successfully used in tube amplifiers, but is there any benefit from it at 18-27 volts, will it give better results than regular Jfets?

[Fancy Lime]

For most pedal purposes, you can probably treat the LND150 as if it was a JFET. It can take very high voltages but that is not really relevant for us, is it? It has tighter specs than many JFETs, which is good for consistent design. But some high quality JFETs have tighter groupings still, although these are getting harder to find. Noise may or may not be on par with JFETs. I think availability is a factor, the LND150 seems to be in active production and is quite cheap. Apart from that, it's mostly taste. Do you like the sound better than that of a certain JFET?

Andy

[POTL]

I'm not sure if I can tell one jfet from another by sound. however, I remember when I collected clean preps or tweed, they sounded cleaner and better on the 2N7000, the sound was warmer (thanks to the Miller capacity, I did not know about its existence then) and more spacious. Jfet (I tried several different ones and did not notice much difference) sounded more toned and flat, the clean sound was cold, but the medium and high gain (preamps with 3-4 amplification stages) were definitely more readable and had a pleasant Chug, in contrast to the loose sound and coarse grain 2N7000. I just wondered if the LND150 would be somewhere in the middle, sounding good in a clean and distorted sound, since it's used in real tube amps?

[phasetrans]

Hey Johnny,

If I wanted to graft the moon LNDer onto the Rosie, do you think that scaling parts to hold similar quiescent current as the 9V is the right choice?

Any idea of where the gate bias should end up in that circumstance?

Thanks in advance,

[jonny.reckless]

If I wanted to graft the moon LNDer onto the Rosie, do you think that scaling parts to hold similar quiescent current as the 9V is the right choice?
Any idea of where the gate bias should end up in that circumstance?

It would be pretty easy to do. I'd raise the drain resistors to 10k, and adjust the source resistor so that the drain sits around 2/3 of the supply.

[jonny.reckless]

hi Jonny Tell us about your impressions and experiences. I see the advantages of the LND150 over the Jfet in that there is no need to adjust the offset, apparently the minimum variation in parameters. How is the LND150 doing with the usual MOSFet problems? I noticed that you used a zener for static protection, is this transistor protected? I did not see the input capacitor in the first stage, usually such transistors do not work without a capacitor at the gate. What can you say about the difference in sound compared to regular Jfet? I was disappointed with the sound of the BS170 / 2N7000 due to their capacity.

Here are some general observations:

JFETs (J112 / J113) - tricky to bias due to variations in VGSoff. Low noise. Good sound - quite tube like especially if you bypass the source with a capacitor. Great for light crunch all the way to heavy distortion. Typically you'll need 3 or 4 cascaded gain stages to get decent modern distortion. I really like the sound of JFET preamps for guitar. I think you'd be hard pressed to tell a well designed JFET preamp from a tube preamp in a double blind test. I used JFETs for the Boba FET and Rosie amps (both available on this site).

MOSFET (2N7000 / BS170) - high input capacitance (especially high Miller cap in a bypassed common source gain stage), so you need a buffer on the input to avoid treble loss. Lots of hiss! High gain, and an aggressive, crunchy / grainy sound. Can be brittle unless you carefully control treble response. Can work well for aggressive high gain distortion sounds. You don't need more than 3 cascaded to get high gain, but you'll end up with about 12dB more hiss than with a JFET equivalent. I used BS170s for the Little Jim distortion pedal.

LND150 (depletion MOSFET) - these are quite interesting. They have low input capacitance and (in my experience) low noise like a JFET. They are internally protected against gate ESD. You bias them using self bias, and they seem to be more consistent VGSoff unit to unit than JFETs. They work great at 9V, and also at tube HT voltages so they complement preamp tubes nicely. The tone is pretty smooth and "tube like". They're pretty cheap and readily available in TO92. Because they are a MOSFET, there is no gate current when forward biased, so you don't have to design for blocking distortion. I was able to come up with the Moon LNDer circuit quite quickly after playing with them for a couple of days. I still think I have a marginal preference for the tone of a JFET preamp, especially under break up / mild overdrive conditions, but I definitely think there is scope for more people to come up with good sounding designs with these. I'm surprised they are not more widely used in guitar circuits given their numerous advantages. I'm planning a preamp "amp in a box" using these, I just need to find the time to finish the PCB, and most of my free time is currently being spent on the firmware for my upcoming reverb module (which I will post on this site if I ever get it finished ).

[Ben N]

Hi, Johnny, got my LND150s in and i'm going to have to give this a go on perf.
What did you use for all those 10uf caps? The schematic doesn't indicate polarity. The bypass caps are obvious enough, but there is also a 10uf coupling cap into the tone stack, and IIRC you like MMCCs as an alternative to electrolytics or too-big films.

[rankot]

What did you use for all those 10uf caps?

He used MLCC AFAIK (see previous posts).

[phasetrans]

If I wanted to graft the moon LNDer onto the Rosie, do you think that scaling parts to hold similar quiescent current as the 9V is the right choice?
Any idea of where the gate bias should end up in that circumstance?

It would be pretty easy to do. I'd raise the drain resistors to 10k, and adjust the source resistor so that the drain sits around 2/3 of the supply.

Thanks Jonny. A question, the schematic as it exists in this thread already has 10K on the drains? Did you mean some other value?

Also you try bypassing the source(s) while doing your tinkering?

[rankot]

MOSFET (2N7000 / BS170) - high input capacitance (especially high Miller cap in a bypassed common source gain stage), so you need a buffer on the input to avoid treble loss. Lots of hiss! High gain, and an aggressive, crunchy / grainy sound. Can be brittle unless you carefully control treble response. Can work well for aggressive high gain distortion sounds. You don't need more than 3 cascaded to get high gain, but you'll end up with about 12dB more hiss than with a JFET equivalent. I used BS170s for the Little Jim distortion pedal.

Why didn't you use input buffer for this? Too much treble?

[Dreadneck]

I am very much thinking about building this wonderful pedal!

A question though. It has the potential to be a very versatile pedal, it is incredibly even and smooth.
Would it be possible to add "grit" to it? Excuse the term I did not find a better word... What I mean is like a switch or knob that would make it break up in a nasty raw way. Preferably in the mids.
I was thinking a switch or knob so that it could go back easily to the clean smooth distortion/overdrive it has.

If that is possible it would make it the ultimate pedal for my needs.

I will try to find good examples of what I mean and post it here.

[iainpunk]

I am very much thinking about building this wonderful pedal!

A question though. It has the potential to be a very versatile pedal, it is incredibly even and smooth.
Would it be possible to add "grit" to it? Excuse the term I did not find a better word... What I mean is like a switch or knob that would make it break up in a nasty raw way. Preferably in the mids.
I was thinking a switch or knob so that it could go back easily to the clean smooth distortion/overdrive it has.

If that is possible it would make it the ultimate pedal for my needs.

I will try to find good examples of what I mean and post it here.

you could experiment with crossover diodes, a clipping pair in line with the signal (instead of going to ground), i would put this in between the 1st and 2nd stage, but experimenting with the placing would be a good idea.
crossover gives a certain growl which i personally really love, but it also creates a gate-effect, which threshold is determined by the diodes Vf

cheers

[bowanderror]

Has anyone seen depletion-mode MOSFETs used as voltage controlled resistors, like in the Orange Squeezer? It looks like it should be possible, but the behavior in the ohmic region looks a bit different to that of JFETs.

[Fancy Lime]

Has anyone seen depletion-mode MOSFETs used as voltage controlled resistors, like in the Orange Squeezer? It looks like it should be possible, but the behavior in the ohmic region looks a bit different to that of JFETs.

Mind the body diode, though.

Andy

[jonny.reckless]

MOSFET (2N7000 / BS170) - high input capacitance (especially high Miller cap in a bypassed common source gain stage), so you need a buffer on the input to avoid treble loss. Lots of hiss! High gain, and an aggressive, crunchy / grainy sound. Can be brittle unless you carefully control treble response. Can work well for aggressive high gain distortion sounds. You don't need more than 3 cascaded to get high gain, but you'll end up with about 12dB more hiss than with a JFET equivalent. I used BS170s for the Little Jim distortion pedal.

Why didn't you use input buffer for this? Too much treble?

I did. The Little Jim uses BS170s for the gain stages and J112 as a source follower input buffer. I found you don't need one with the LND150s as their capacitance is much smaller than the BS170.

[jonny.reckless]

I am very much thinking about building this wonderful pedal!

A question though. It has the potential to be a very versatile pedal, it is incredibly even and smooth.
Would it be possible to add "grit" to it? Excuse the term I did not find a better word... What I mean is like a switch or knob that would make it break up in a nasty raw way. Preferably in the mids.
I was thinking a switch or knob so that it could go back easily to the clean smooth distortion/overdrive it has.

If that is possible it would make it the ultimate pedal for my needs.

I will try to find good examples of what I mean and post it here.

you could experiment with crossover diodes, a clipping pair in line with the signal (instead of going to ground), i would put this in between the 1st and 2nd stage, but experimenting with the placing would be a good idea.
crossover gives a certain growl which i personally really love, but it also creates a gate-effect, which threshold is determined by the diodes Vf

cheers

An other way get some "grit" is to experiment with the DC bias voltage of the final gain stage, you get a much fuzzier sound when the bias is closer to ground that the typical 2/3 of VDD.

[box]

Hi All,
my suggestion of small modifications of the circuit.Sorry for  the hand drawing
Regards,
box

[box]

Hi,
I'm sorry but I need to update some small changes to the drawing.
Regards,
box