MOSFET Driver after other pedals (impedance question)

[lrgaraujo]

As I've mentioned before, I have very little understanding of the concept of impedance so take everything with a grain of salt.
Running some tests on LTspice, I've noticed that, when you plug a guitar straight to the input (assuming my pickup simulation is correct) there's a lot of treble roll off because of the low input impedance. However, when you feed the input from an ideal source (a buffer might act similarly), there's almost no treble cutting, which might be resulting in a lot of treble feeding the circuit down the road.
If that's the case, I'm not 100% sure Bean's suggestion will solve your problem.
Then again, I might be completely wrong here

 

[allanshookphoto]

As I've mentioned before, I have very little understanding of the concept of impedance so take everything with a grain of salt.
Running some tests on LTspice, I've noticed that, when you plug a guitar straight to the input (assuming my pickup simulation is correct) there's a lot of treble roll off because of the low input impedance. However, when you feed the input from an ideal source (a buffer might act similarly), there's almost no treble cutting, which might be resulting in a lot of treble feeding the circuit down the road.
If that's the case, I'm not 100% sure Bean's suggestion will solve your problem.
Then again, I might be completely wrong here

That is interesting. In that case, it is actually working "properly" when fed from a buffer (or a pedal that is "on") and it has a noticeable treble cut when coming straight from the guitar.

 

[allanshookphoto]

I would seriously just consider the advice from bean and what I said: snip out R2, swap R3 for 1M. At this point I'm rather sure it's just faulty design.

So, snipping out R2 would be eliminating the 1M pulldown resistor. From my understanding, that resistor is only there to eliminate the pop from turning on/off the switch. And, changing the R3 10K resistor for a 1M would change how the non-inverting op-amp stage is biased. I don't know enough about it to know if or how that would change the input impedance. I have a lot of research to do. Thanks for the help.

 

[lrgaraujo]

That is interesting. In that case, it is actually working "properly" when fed from a buffer (or a pedal that is "on") and it has a noticeable treble cut when coming straight from the guitar.

have you tried it with a buffer? a boss pedal, for example?

IMO 'properly' is a trick word to use here, because I'd say it works properly when it sounds like it should, but I see what you mean. I'm not sure that's what's going on, but it might be a case similar to those fuzzes (like a fuzz face) and wahs that like to be first in chain

 

[allanshookphoto]

have you tried it with a buffer? a boss pedal, for example?

IMO 'properly' is a trick word to use here, because I'd say it works properly when it sounds like it should, but I see what you mean. I'm not sure that's what's going on, but it might be a case similar to those fuzzes (like a fuzz face) and wahs that like to be first in chain

Well, technically I have not tried it with a buffer/buffered bypass pedal (Boss). However, any pedal that is "on" (and has a buffered output) is a buffered pedal, so in that sense, yes, I have tried it with buffered pedals. I have put it after my Bad Stone (turned on), and my Joyo Sweet Baby (also on), and that is when the high end gets crazy. if I turn both of those pedals off (both true bypass), the crazy high end goes away, and the MOSFET Driver sounds quite nice, very balanced and pleasant OD effect.
Yes, I agree with you, it does seem to be like a Fuzz Face style pedal that really needs to be first in chain. I was hoping there are others who built the MOSFET Driver who could confirm this for me.

 

[jesuscrisp]

So, snipping out R2 would be eliminating the 1M pulldown resistor. From my understanding, that resistor is only there to eliminate the pop from turning on/off the switch. And, changing the R3 10K resistor for a 1M would change how the non-inverting op-amp stage is biased. I don't know enough about it to know if or how that would change the input impedance. I have a lot of research to do. Thanks for the help.

For R2 to act as a proper anti-pop pulldown resistor, it would need to be right at the input, before R1 and C1. The original Tube Driver (see below) doesn't have it, probably as it may not be needed with the switching it has but don't quote me on that.

R3 sets the input impedance basically and making it 1M is pretty standard, especially with a JFET input op amp like the TL072. R2 in this schematic doesn't make sense overall and seems like a remnant of the original Tube Driver, which has bipolar power supply and therefore biases the op amp to ground. For some reason, whoever drew up the schematic decided to leave it in and also make the bias resistor 10K, which gives you rather atrocious input impedance. The way the Mosfet Driver is biased is a mistake, R3 SHOULD DEFINITELY be 1M.

Changing R3 to 1M, snipping out R2 and ideally putting a 1M pulldown resistor at the input will fix the issues and make this pedal actually act like it should, or at least closest to what it should do. While we're at it, R4 shouldn't really exist at all and R6 could be jumpered as well.

It SHOULD NOT BE a Fuzz Face style pedal that needs to be first in chain to sound decent.

EDIT: Just noticed the EQ pots are logarithmic, not linear in the original. That could also make especially the treble pot's taper much less usable.

 

[allanshookphoto]

For R2 to act as a proper anti-pop pulldown resistor, it would need to be right at the input, before R1 and C1. The original Tube Driver (see below) doesn't have it, probably as it may not be needed with the switching it has but don't quote me on that.

R3 sets the input impedance basically and making it 1M is pretty standard, especially with a JFET input op amp like the TL072. R2 in this schematic doesn't make sense overall and seems like a remnant of the original Tube Driver, which has bipolar power supply and therefore biases the op amp to ground. For some reason, whoever drew up the schematic decided to leave it in and also make the bias resistor 10K, which gives you rather atrocious input impedance. The way the Mosfet Driver is biased is a mistake, R3 SHOULD DEFINITELY be 1M.

Changing R3 to 1M, snipping out R2 and ideally putting a 1M pulldown resistor at the input will fix the issues and make this pedal actually act like it should, or at least closest to what it should do. While we're at it, R4 shouldn't really exist at all and R6 could be jumpered as well.

It SHOULD NOT BE a Fuzz Face style pedal that needs to be first in chain to sound decent.


View attachment 51698

Yes. That makes sense, and seems right to me. Making R3 a 1M, and moving the R1 1M to the input would essentially make it identical to the Rat input stage. I'm thinking that this circuit board need a re-design/new version with these changes.
Thank-you to everyone for this very useful discussion! Much appreciated.

 

[Coda]

For R2 to act as a proper anti-pop pulldown resistor, it would need to be right at the input, before R1 and C1. The original Tube Driver (see below) doesn't have it, probably as it may not be needed with the switching it has but don't quote me on that.

R3 sets the input impedance basically and making it 1M is pretty standard, especially with a JFET input op amp like the TL072. R2 in this schematic doesn't make sense overall and seems like a remnant of the original Tube Driver, which has bipolar power supply and therefore biases the op amp to ground. For some reason, whoever drew up the schematic decided to leave it in and also make the bias resistor 10K, which gives you rather atrocious input impedance. The way the Mosfet Driver is biased is a mistake, R3 SHOULD DEFINITELY be 1M.

Changing R3 to 1M, snipping out R2 and ideally putting a 1M pulldown resistor at the input will fix the issues and make this pedal actually act like it should, or at least closest to what it should do. While we're at it, R4 shouldn't really exist at all and R6 could be jumpered as well.

It SHOULD NOT BE a Fuzz Face style pedal that needs to be first in chain to sound decent.

EDIT: Just noticed the EQ pots are logarithmic, not linear in the original. That could also make especially the treble pot's taper much less usable.

View attachment 51698

Post of the Day right here...

 

[Feral Feline]

Good looking gutshot, @allanshookphoto, clean build.

Look forward to hearing about it after the mods/corrections.

 

[Feral Feline]

I find it very useful to evaluate impedance with LTspice. Simulation of the input stage of the mosfet drive results in the following:
View attachment 51684

Cheers.

I've yet to try LTspice. Lots of basics to learn still before I tackle that. I'll get there eventually.

 

[lrgaraujo]

Cheers.

I've yet to try LTspice. Lots of basics to learn still before I tackle that. I'll get there eventually.

I'm no master by any means, but I find it very interesting. A bit of a learning curve, but nothing terrible.