Little Jim - a Marshall inspired MOSFET distortion pedal design

[Vivek]

One other thing I was wondering about: the 100 ohm resistor (R15) in the PSU input stage, does it serve any purpose for the dynamic response (sagging) of the rest of the circuit? Or is it purely for keeping out the PSU noise?

Rutger

I think simple answer is No, not for sag


It appears that the word "Sag" has two meanings

A) Like an Amp with an overload on its power supply. Hence initial part of note is higher volume and cleaner, but as voltage drops, the note gets compressed, more distorted. After a very short time, the voltage recovers and than we can have same story for the next notes

B) Like a permanent drop of voltage normally seen in a fuzz circuit, leading to the circuit operating at a different point. There is no "recovery" since the voltage does not go back to normal.


Hence here the resistor is only for hum and ripple rejection by increasing the time constant of the filtering capacitor.

But maybe different sounds are possible by changing this resistor to higher values (Idea B)

[Vivek]

This is the Little Jim distortion pedal. It's inspired by and named after Jim Marshall, and attempts to capture that bright, crunchy, raucous sound of a cranked Marshall amp, which is also deceptively complex and rich with harmonics and timbral movement.

Thank you for posting your project, Sir Jonny !!!

A) I am interested in studying inter-stage frequency response and transfer functions to be able to achieve a particular sound.

I request you to post these graphs if possible

Alternatively, if your files are compatible with LT SPICE, maybe I know enough to be able to make these graphs.


B) Did you try to aim for particular inter-stage frequency response and transfer function while designing this circuit ? Please help me to understand how the various capacitors were calculated.

Thank you !!!

[GibsonGM]

I've built a few with JFET's like this, but without the buffer.  I have to hand it to you Jonny, really nice job; other attempts I've made weren't quite so smooth and responsive    This thing wails!  Must be...Mosfet booster power! 

I can't say yet if overdriving my amp is contributing, tho I've tried it on very low levels to avoid this, with great results.   Modded some stuff...2N5457 for buffer with a 100k to bias, and since not selecting the BS170s by hand, dialing them in at the drain w/4.7k source R's.   All the same, nice little Marshall-emulating project!   I look forward to boxing it and running it thru my Astro Sim.   Thanks for sharing!

[LightSoundGeometry]

I just tossed a bottle of etchant away thinking I was going to ltspice a fabbed board    now I want to do this ! but this is the crazy sheeite that you can do shrinking stuff down



modified to say I miss radio shack ..there is a void where I am for electronics ...would a brick and mortar work or people just buy online? considering SAFB is right next to me there would be a lot of hobbyist ( those air force guys like them rc planes and drones) , radio guys , there is some bloke in waterloo, a few towns over, who builds amps etc ..seems like a good idea to open a store lol. if radio shack was open still, I would be in there in the morning to get etching and a copper clad ..if I print the file will the layout be sized automatically?

[rutgerv]

Hi everyone,

The more I play with this circuit, the more I like it! For my own learning process, I experimented with the original Little Jim circuit and tried a couple of modifications. Here are my findings (and some new questions raised, that perhaps someone knows the answer to already):

1) tried the alternative biasing method with a 1M from gate to ground and 1M from gate to drain (like in the Box of Rock and La Grange circuits previously mentioned). Can confirm what Jonny said before: certainly no improvement. Overall brightness didn't suffer too much, but the sound became more lifeless and less immediate. I suspect it influences the dynamics of the circuit somehow.

2) I tried running the circuit at a higher (and lower) voltage and learnt that it has a dramatic effect on brightness. At 12V the circuit is much brighter (too much for my taste). I was hoping to incorporate the circuit in a preamp that I was planning to run on 12V. If anyone knows a way to bring the circuit back to it's original 9V while running it from 12V, I'd be happy to hear! Otherwise I apply some filtering

3) Again, trying to learn from the alternative (not necessarily better) approach in the Box of Rock and La Strange, I tried some alternative gain structures. Instead of connecting C3 to ground, I connected it to a 330 ohm resistor. This made a big change. Overall gain of the circuit drops (but for my taste it has plenty left), the low-end becomes a bit more lively and there's more audible compression on peaks. Actually I very much liked this tweak! The only down-side (on my breadboard) was a higher amount of hum/noise. I struggle to understand why, because I was expecting the exact opposite from lowering the gain in the last stage. Perhaps it's just my breadboard layout? Or is there a more fundamental reason (increased influence of a dirty PSU?). I also tried a similar modification with a 100 ohm resistor in the stage before the final gain stage, but that didn't make such a dramatic difference.

4) I see many MOSFET pedals carry protection zener's near the MOSFET's, and was wondering it would be wise to include any in the Little Jim if I put it to its final board, but can't think of any reason why a boxed pedal could suffer from voltage spikes beyond 60V on the MOSFET gates.

Rutger

[11-90-an]

I was hoping to incorporate the circuit in a preamp that I was planning to run on 12V. If anyone knows a way to bring the circuit back to it's original 9V while running it from 12V, I'd be happy to hear!

LM7809 voltage regulator?

[Vivek]

Hi everyone,

2) I tried running the circuit at a higher (and lower) voltage and learnt that it has a dramatic effect on brightness. At 12V the circuit is much brighter (too much for my taste). I was hoping to incorporate the circuit in a preamp that I was planning to run on 12V. If anyone knows a way to bring the circuit back to it's original 9V while running it from 12V, I'd be happy to hear! Otherwise I apply some filtering

Rutger

I can understand how harmonics can change in a circuit with a change in supply voltages, due to different bias operating points leading to different clipping.

Do you feel that the brighter sound at 12V was due to change in clipping due to difference in bias operating points ?

[Marcos - Munky]

Finally boxed mine. The etching didn't came out too good because I tried a different paper for the tone transfer. But I liked the result anyway.

[rankot]

What if you use LND150? They are lateral MOSFET, maybe they have less noise than BS170?

[jonny.reckless]

What if you use LND150? They are lateral MOSFET, maybe they have less noise than BS170?

They're depletion mode MOSFETs so you'd have to bias them with a negative gate voltage, or use source biasing like you would for a JFET, so the circuit would have to be different. They have a much lower input capacitance and a much higher channel resistance than the BS170. It might be worth trying to substitute them in a few JFET based preamp circuits and seeing how they sound, but my guess is you'd have to specifically design a circuit around them. Depletion mode MOSFETs with high channel resistance are quite unusual.

I'll buy some and see if I can come up with a circuit to use them. Watch this space

[rankot]

This guy built JCM800 emulation using LND150. https://www.ssguitar.com/index.php?topic=2006.msg15734#msg15734

[jonny.reckless]

This guy built JCM800 emulation using LND150. https://www.ssguitar.com/index.php?topic=2006.msg15734#msg15734

I listened to those clips, it's a great sounding preamp but neither of them sound remotely like my JCM800 2x12 4211 combo. Everybody quotes, copies or models the JCM800 series, and just about everybody gets the gain structure and voicing wrong. Most of the distortion tone of those amps comes from two things: asymmetric clipping in the DC coupled cathode follower driving the tone stack, and cutoff clipping in the phase inverter stage. There's some transformer compression with the output stage too and power amp harmonics which relies on the EL34 screen grid kinks. I actually built a direct copy of a JCM800 master volume head once, with the exact same component values, tube types, and Dagnall electronics UK wound transformers, and people who heard it said it lacked gain and definition The truth is that the JCM800 is quite a primitive, low gain, inflexible amp, that sounds magic for blues and vintage rock, but needs a stomp box or two to make it really sing, and only comes alive when you overdrive the 100 watt power stage at ear splitting volume. I've owned several of them (original Bletchley manufactured from the 80s) and to get a decent tone out of them I always had to put a drive pedal in the front, and use quite a bit of EQ in the mix.

I was never trying to build a clone, that's more runoffgroove's thing. The Little Jim is far too simple and primitive for that. I was just going for a saturated tone with some harmonic motion and a nice bright cutting distortion tone. That guy's preamp sounds great but has about 10x the component count! It looks like he's using 2SK216 not LND150?

[Vivek]

the gain structure and voicing wrong.
a saturated tone with some harmonic motion and a nice bright cutting distortion tone.

I am trying to learn a bit about emulating Amps by copying the gain structure, voicing, harmonic motion of each stage.

Please help me to understand your target of gain structure, voicing, harmonic motion of the stages of Little Jim and how you achieved it.

Thank you !!!!

[jonny.reckless]

the gain structure and voicing wrong.
a saturated tone with some harmonic motion and a nice bright cutting distortion tone.

I am trying to learn a bit about emulating Amps by copying the gain structure, voicing, harmonic motion of each stage.
Please help me to understand your target of gain structure, voicing, harmonic motion of the stages of Little Jim and how you achieved it.
Thank you !!!!

The circuit design of Little Jim is very simple.
The input stage is a unity gain buffer designed to present high impedance and low capacitance to the guitar pickup.
The first gain stage is partially bypassed, so that its frequency response has a pole around 160Hz and a zero around 3kHz. This allows for a bright crunchy sound due to accentuating the higher harmonics of the guitar frequency range.
The second and third gain stages are fully bypassed for maximum gain.
The drains are biased at DC to around 2/3 VDD to provide asymmetric clipping with a fat rounded top half wave and a heavily clipped bottom half wave. This exploits the quadratic nature of the field effect transistor for plenty of even and odd harmonics.
The filter stage rolls off the sharp high frequencies introduced by the clipping stages for a more rounded tone.

All amps and distortion circuits follow these general principles to some degree: filter the bass, overdrive, clip, then filter the treble and / or boost the bass.

[Vivek]

Thanks !!!!

Is it necessary to design the circuit so that goes into symmetrical clipping for larger signals ?

Please help me understand the "Harmonic Motion". Is that something that needs to be especially designed in, or is a natural byproduct of having multiple clipping stages ?

[rankot]

This guy built JCM800 emulation using LND150. https://www.ssguitar.com/index.php?topic=2006.msg15734#msg15734

I listened to those clips, it's a great sounding preamp but neither of them sound remotely like my JCM800 2x12 4211 combo. Everybody quotes, copies or models the JCM800 series, and just about everybody gets the gain structure and voicing wrong. Most of the distortion tone of those amps comes from two things: asymmetric clipping in the DC coupled cathode follower driving the tone stack, and cutoff clipping in the phase inverter stage. There's some transformer compression with the output stage too and power amp harmonics which relies on the EL34 screen grid kinks. I actually built a direct copy of a JCM800 master volume head once, with the exact same component values, tube types, and Dagnall electronics UK wound transformers, and people who heard it said it lacked gain and definition The truth is that the JCM800 is quite a primitive, low gain, inflexible amp, that sounds magic for blues and vintage rock, but needs a stomp box or two to make it really sing, and only comes alive when you overdrive the 100 watt power stage at ear splitting volume. I've owned several of them (original Bletchley manufactured from the 80s) and to get a decent tone out of them I always had to put a drive pedal in the front, and use quite a bit of EQ in the mix.

I was never trying to build a clone, that's more runoffgroove's thing. The Little Jim is far too simple and primitive for that. I was just going for a saturated tone with some harmonic motion and a nice bright cutting distortion tone. That guy's preamp sounds great but has about 10x the component count! It looks like he's using 2SK216 not LND150?

He actually has two designs, the one I linked to is using LND150s. http://milas.spb.ru/~kmg/files/projects/jcm800mv/fet/lnd150/jcm800mvfSch.pdf

I posted it just as a reference to you, if you decide to try those depletion mode lateral MOSFET.

[rutgerv]

He actually has two designs, the one I linked to is using LND150s. http://milas.spb.ru/~kmg/files/projects/jcm800mv/fet/lnd150/jcm800mvfSch.pdf

Wow, look at these voltages! . Might as well use tubes then!

[jonny.reckless]

I'm guessing the diodes in the gate circuit are to simulate grid current when the grid is positive biased. Interesting idea, but needing the high voltage makes it unsuited to a simple (or cheap) pedal design - generating 300V from 9V is not practical with a simple boost converter, you need a coupled inductor with multiple windings to get that kind of voltage ratio (or multiple stages of boost conversion). Personally if I was going to all that trouble I'd just use ECC83s and be done with it

Back in the 90s at Trace Elliot we made a hybrid guitar power amp using a tube topology, with a pair of push pull high voltage MOSFETs, driven by a MOSFET phase splitter, driving an output transformer which coupled to the speaker. I think it ran on around 400VDC. I can't remember the name of it now, it was one of Clive Button's designs, but it responded surprisingly like a traditional push pull tube power amplifier when overdriven.

I've been using JFETs in tube like guitar circuits since the mid 90s, and it works pretty well. You don't need to copy the schematic exactly, just generally get a similar circuit topology and then voice it to suit FETs. The same rules apply: reduce the bass (or boost the mids / treble) early in the signal path, progressive clipping, moving DC bias points, some form of asymmetry to give even order harmonics, and a lowpass filter at the end to remove unpleasant high order harmonics. I like the sound and feel of multi stage clipping circuits, you get more "swirl" and "harmonic motion" than with a single diode-op-amp type clipper, due to the fact the interstage DC bias points are modulated by the signals as they go into and out of saturation. Provided you don't get blocking distortion, it sounds rich, complex and more "amp like". I've posted a couple of similar amp designs on here before, the Boba FET and the Rosie amp, which use this approach with JFETs.

https://www.diystompboxes.com/smfforum/index.php?topic=118627.0
https://www.diystompboxes.com/smfforum/index.php?topic=123572.0

[jonny.reckless]

@Ranko, I got some LND150s today from Mouser and started to play around with them. They are interesting. Much lower saturation current and higher drain resistance than even most JFETs. I've got some basic ideas for a distortion pedal using them, the common source gain stage looks quite usable on the scope when correctly biased. When I've pulled a complete design together I'll post again here, probably in a couple of weeks. It's going to be quite similar to Little Jim, but I'd like to add separate bass and treble controls.

[teemuk]

Back in the 90s at Trace Elliot we made a hybrid guitar power amp using a tube topology, with a pair of push pull high voltage MOSFETs, driven by a MOSFET phase splitter, driving an output transformer which coupled to the speaker. I think it ran on around 400VDC. I can't remember the name of it now, it was one of Clive Button's designs, but it responded surprisingly like a traditional push pull tube power amplifier when overdriven.

"Reactor" it was called. IIRC, the phase inverter was actually opamp based and quite similar to that in Music Man hybrid amps, with the exception of featuring zener diodes in the feedback loop that clipped the signal. (Could have been the primary clipping mechanism, IIRC).
Didn't have high supply voltage either, it was pretty much in the range of generic SS power amps with similar output power rating (e.g. 80V rail-to-rail). No need for high supply voltage because the output MOSFETs had low impedance and there was no need for similar current-voltage transform as in tube amps with high plate output impedance.
Speaking of which, IIRC the amp also featured the typical current feedback scheme in order to increase the output impedance artificially.

There was also a 30W Reactor amp with just two high power opamps running bridged and no OT.

Gallien-Krueger made several transformer-coupled SS amps in the early 1970's (with current-starved PI for clipping and "power scaling") and Fender had a patent for one as well (although I don't know if they ever employed the idea in practice). I recall Trace Elliot had an European patent as well. Most importantly, these designs were similar to tube amps by topology, not just OT in output of a follower output like in 70V line systems and such.