Tone pot in Brian Wampler's modded MXR Distortion +

[fishfude]

Hi folks,

I was looking at Brian Wamplers article on modding the MXR Dist + and his choice of tone pot confuses me! It's the 1M in the bottom right of the schematic here:



I would expect a cap in parallel with a resistor but it seems it's in series with C5, unless it is working with C4, in which case why have C5?

Is C4 for the output maybe, and if so is it not redundant?

Cheers,

ff

[rousejeremy]

C5 and the 1meg pot are forming a hi pass filter.

[duck_arse]

with C4 a dc blocking cap.

high pass? I thought low pass/ high cut.

[induction]

Yes, it's a high-cut. Same design as a guitar tone knob.

[GibsonGM]

You could think of the 1M as 'slowing down' C5's shunting of highs to ground.  If it weren't there, you'd have 100% high cut all the time.  The more resistance the pot offers, the less high cut you get....yup, just like the one on your guitar.

[Mark Hammer]

Quite honestly, I think most builders of this exact circuit will end up saying that the tone control doesn't really "work".  Stock, a Dist+ will have a .001uf cap directly to ground, where the tone control is.  That cap value, in tandem with the 10k resistor just ahead of the diodes, will yield a shallow rolloff starting just under 16khz.  Raising the cap value to 2200pf from 1000pf will drop that rolloff point down to around 8khz (with the Tone pot set to 0 ohms).  Given that most guitar amps use speakers that start to lose high end around 6khz, that change in tone is not going to be terribly obvious.  You will be able to hear the effect of rotating the pot, but what you'll get is small variations in fizz, rather than any wholesale change in "warmth" or "bite".

Instead, I'd suggest one of Jack Orman's variations on the SWTC :  http://www.muzique.com/lab/swtc.htm

You will note that both of Brian's suggest switches incorporate more bass, vs less bass.  However, there is no means, other than the suggested tone control, to remove fizz from the signal.  And with the signal being double-clipped, and having a potential max gain of 1001x, there will be fizz.

The solution is simple: stick a 47pf cap in parallel with the gain pot.  When the gain is set low, there will be no treble rolloff.  But as the gain is increased, the amount of treble cut will also increase.  At max gain, the treble rolloff will begin around 3.4khz.  That may seem a little low, but keep in mind that a) it's a shallow rolloff with plenty of treble retained, and b) once that signal gets clipped again by that second pair of diodes, more treble will be added (that you can either keep or trim back with a tone control).

[Pojo]

I actually think that method of tone control can be very effective, but using different values for me. I would try R6 at 15k, C5 at .01uF, and a 20k or 25k linear (avoid audio taper unless you want 0-50% of the rotation to barely do anything) pot instead of the 1M and tweak from there. You should end up with a much more effective control then what's in the schem as is.

And as Mark said, the SWTC's from Mr. Orman are great options too.

[Mark Hammer]

I actually think that method of tone control can be very effective, but using different values for me. I would try R6 at 15k, C5 at .01uF, and a 20k or 25k linear (avoid audio taper unless you want 0-50% of the rotation to barely do anything) pot instead of the 1M and tweak from there. You should end up with a much more effective control then what's in the schem as is.

Eminently sensible.

[pappasmurfsharem]

What is the difference between a High Cut and Low Pass?

Also could you reverse the cap and resistor for a "Low Cut"?

[J0K3RX]

Useful..

Calculator
http://www.muzique.com/schem/filter.htm

[nocentelli]

What is the difference between a High Cut and Low Pass?

Also could you reverse the cap and resistor for a "Low Cut"?

High cut is basically low pass: The cap and pot cut highs and only allow lows to pass to the output.

An easy way to get low cut/high pass is to have a large value output cap, followed by a much smaller cap in series after the first cap to the output: A pot wired as a variable resistor in parallel with the smaller cap will bypass the cap for full low end when resistance is zero, and as the pot is turned and resistance increases, the lower frequencies are progressively rolled off.

[Gus]

Useful..

Calculator
http://www.muzique.com/schem/filter.htm

This can mislead you.  You need to know what is before and after the RC

[Gus]

Have you seen this web site?  It has been on the web a long time.  The home page has the launch date as may 96 http://www.gmarts.org/index.php?go=100
http://www.gmarts.org/index.php?go=217

[rousejeremy]

with C4 a dc blocking cap.

high pass? I thought low pass/ high cut.

I get those things messed up all the time

[fishfude]

Thanks guys these are all amazing responses.

Yes! The SWTC is what I want here I think

[Mark Hammer]

I whipped something up a year ago (since sold, so I'm going from memory) that was similar to what is illustrated in the schematic, although I had the diodes in the feedback loop fixed, and the 2nd set after the output as the switchable option.  A couple of bits of useful advice:

1) The 2nd set of diodes won't do much for you (i.e., re-clip) unless the op-amp output tends to exceed the total forward voltage of the 2nd set of diodes.  So, consider making the first set some combination with a higher forward voltage than the second set.  That could be a matter of simply selecting some from the same type with higher Vf for the first pair.  So, if you had a pile of 1N914s that ranged between 520 and 640mv forward voltage, you'd use 4 at the upper end of the range for the diodes in the feedback loop, and 4 in the lower range for the 2nd set.  The difference in Vf  between the sets will determine how much of a re-clip effect there is.  Note that re-clipping at a lower threshold will reduce your output level.

2) Nobody likes to hear harmonics of harmonics of harmonics.  So it is VERY advisable to stick a small cap in the feedback loop of the op-amp, such that there is not excessive high-freq content to be re-clipped by the 2nd set of diodes, and place undue burden on the tone control to tend to that much fizz; particularly since the tone control is still a puny 6db/oct rolloff.

I will take a shot ion the dark and suggest that you go with a 150pf cap value.  With gain set to max (pot = 1M), this gives a treble rolloff starting around 1060hz.  That might seem low, but a) it IS only 6db/oct, and b) the clipping adds harmonic content from what is there.  Given your use of a 1k ground-leg resistor, there is a potential gain of 1001x, which is pretty darn high, so once you go below around 200k pot resistance, you'd have an overdrive more similar to a cranked Dist+, and rolloff starting around 5.3khz, which should yield plenty of bite and sizzle, but not so much that your tone control downstream will have insufficient effect.

[fishfude]

I was actually going to leave out the 1st set of diodes out of it and just have the diodes on the output. I've rigged up a 12 way switch with 12x combinations of diodes. Are there any problems with leaving the clipping until this late?

Thanks for the tip on the feedback cap!

ff

[Mark Hammer]

12-way?    Do you think you will actually make use of that?  I find 2 or 3 choices enough; especially when the drive control has a wide-enough range.  Certainly, you may go through a bigger variety of diode arrangements than 2 or 3 when tryng to settle on what you really like.  BUt once you find out what you like, I'm pretty confident that 9 or 10 of those dozen positions are going to end up being neglected.

[fishfude]

Ha, yes, it is a little excessive, but I wanted to do it as an experiment and since all the extra diodes add up to less than a Euro, I thought why the hell not. I also had a sneaking suspicion that the different clipping points at similar gain, would have a different effect than just having the same clipping point at different gains levels.

I'm stuck on something else in Brian's Schematic. Here is the original Dist+ schematic:



In his mod schematic he changes R2 and C3 to ''[get] the same high gain frequency response, but more gain potential.''

I thought R1 and R4 were responsible for the gain.

How does changing R2 and C3 result in the same frequency response when C3 is the component responsible for the filtering? Does R2 play a part too?

Cheers,

ff

[Digital Larry]

This is a non inverting stage.  R1 has hardly anything to do with the gain.

If it were inverting then yes the gain would be -(feedback resistor)/(input resistor).

See http://sparksandflames.com/p72.html