Marshall Guv'nor, what does R6 and C8 do?

[caspercody]

Looking at the schematic from GGG for the Marshall Guv'nor I am wondering if anyone can tell me what R6 (680k) and C8 220pf) do?

I made this circuit up, and I am using a 3 position 3pole switch to switch between three different values for R6 (680k, 470k, or 220K). This was mentioned to me to change this value to get a high, medium, or low gain (respectively). This works, but not to much difference unless I have the gain set low on the pedal. But on this switch I can add different caps that will switch value along with the associated resistor. But not sure how to select which cap value to use with the 470k, or 220k resistor?

Hope this makes sense to what i am asking.

Thanks

[Mark Hammer]

Yep, makes sense.  R6/C8 set the gain and high-frequency rolloff of that stage.  The point where the treble starts to be cut is given by:
freq = 1 / (2 * pi * R * C)

What this means is that if you change the value of R6, C8 will have a different impact.  Stock, that network rolls off high end in that stage starting at around 1060hz.  That sounds low, but remember it is a very shallow rolloff (6db/octave), and the clipping produced by the LED pair will increase the high-frequency content.  If you held C8 constant, but changed R6 to 470k, the rolloff would now start at just over 1.5khz, and with R6=220k, that would move up again to around 3.3khz.

Is that a "bad" thing?  Not necessarily.  Keep in mind that:
a) gain is multiplicative such that reducing the gain of that stage even by a modest amount has a greater impact on total gain
b) the LEDs have a fairly high threshold for clipping such that less harmonic content is generated by them as the gain of the op-amp stage is reduced, making the retention of more treble in that stage not especially punitive,
c) you always have a treble control to use.

Will those presets you're planning sound exactly like a Guvnor but with three different gain settings?  With the exception of the sock 680k value, not exactly.  But that's okay.  You still have the gain control to twiddle.

[caspercody]

I was going to wire the switch so that in one position I will have the original setting (680k/220pf). But I was wondering if I change both the cap and resistor so that I switch to another position it would give me a different sound and yet to the last position another different sound. Or is there really no big difference in the distortion by changing these two components? And using those resistor values (470k, and 220k) what would be a good cap value for each resistor to start with?

I know right now with the cap at 220pf for all three positions and switching just between resistors I get alittle difference in sound but at low gain settings with the gain pot.

Thanks

[slacker]

If you want the frequency response to be about the same with the different resistor values then make C8 the same ratio larger as you've reduced C6 by. I think that works out at something like 330p for 470k and 680p for 220k.

Like you've found I don't think that mod is going to do a lot. At maximum gain the original circuit had a gain of about 3000, this is enough to make the opamps clip and then the diodes add even more clipping. If you lower R6 to 220K the circuit still has a gain of over 1000, so you're still going to get plenty of opamp and diode clipping. As the notes decay they might clean up faster, but it's not going to sound hugely different.
As you reduce the gain the mod will have more of an effect, but I don't think it will give you many more tones than you can get with the original circuit.