Modding a Super Fuzz for bass?

[Snufkinoob]

I've built a few of these but want to add a little extra flexibility to make it more bass-friendly. Something like the Wounded Paw Attack Goat, (though I've not seen the guts of one)

I'm guessing that adding a blend pot would be the way to go, but I've never tried that before. Is it just a pot at the input or am I thinking of something else?

Also, IIRC I've read that removing/trimming the messy upper-octave apparently helps too. I'm sure the Wattson EFY (square box version) has something like this going on for it's 'octave' pot (in addition to the common 10k trimpot connected to the two 22k's) But how would that pot be wired up? Or a simple toggle to go to the two extremes of that pot?

Ta!

[Snufkinoob]

I've found IvIark's mini blend, but I've read that the input cap should be increased to 1uf or more if using a bass, but what if you want it to be usable for both guitar and bass? Toggle both values or simply omit the cap altogether so all of the signal is getting through?

[anotherjim]

Something in that scheme may be a clue to a built in method of adding blend. I'm wondering why that output pot is labelled "balance"? Could it's grounded end be re-routed to a dry signal feed? it may be as simple as linking the bottom of the balance pot (disconnect the ground) to the top of the expander pot. You may still want a  volume control, so that 100k pull-down on the output becomes a pot - wiper to output jack.
Whoever designed the original circuit must of had a lot of 10uF caps to use up 

[Gargaman]

I've found IvIark's mini blend, but I've read that the input cap should be increased to 1uf or more if using a bass, but what if you want it to be usable for both guitar and bass? Toggle both values or simply omit the cap altogether so all of the signal is getting through?

As a beginner I've learned that increasing the input cap on circuits let more bass get in. Works like that for the output cap, but then you control the bass out. I've learned also that controlling the low frequencies (getting in and out) is the key to get fuzz and distortion to sound good. I've seem a lot of projects that use two input capacitors, switched by a SPDT. If you set up the right values on those capacitors you may use one SPDT position to 'guitar in', and the other to 'bass'.
I believe that this 10uF as the input is letting all frequencies to get in.

[anotherjim]

Indeed,  here the 10uF input is huge, so not actually a problem for bass. None of the coupling caps are too small for bass.

[Gargaman]

Hey, Jim
how would you calculate the frequency cutoff on coupling caps (in and out)?
I'm sure it would depends in each particular case, but does the cap affects the frequency response only by itself?
I think I should google for capacitive reactande, right?
Let's say, in a Fuzz Face, does the 2.2uF starts to roll off bass in wich point?
sorry if I'm away from the subject of the thread.
regards

[Snufkinoob]

I believe that this 10uF as the input is letting all frequencies to get in.

Should do. A lot of people are happy with the stock circuit on bass. Honing the Super Fuzz to be a little better is more about trimming back the upper harmonics/octave (though not technically an octave IIRC?) so it's just a little leaner, and less messy for bass while retaining the original tone. And it's something in addition to the 10k trimpot that can be put in between the 22k's at Q3&4 (that's not on this schematic). It might just be the 10k-to-9v at Q3&4 wired as a pot but I think I tried that in the past with no result.

Regarding cap values and bass, it's the input cap of the blend circuit that I'm curious about. Not sure if I should increase it or omit it altogether. I want to be able to plug in a guitar or bass and have no adverse effects and have the blend work correctly for both.

EDIT: Can't modify my original post for some reason... I'm going to try out a tone pot rather than a toggle switch to blend rather than switch between the two 'modes'. At the moment I'm putting a 56nf in place of the 100nf, then to a 10k pot wiper and a lug to ground. I think that's right, BUT do I then connect the 47k/10k junction to the 1nf/10k point?

[anotherjim]

Gabriel, the coupling caps are making RC high pass filters, you have to know the R value also, which can take a bit of working out with discrete circuits like this.
Take the cap off the Expander pot wiper.
There is some series resistance before the cap - I take the easy way. The preceding stage is an emitter follower so has very low impedance. If the expander pot is maximum, I can ignore what comes before the cap.
After the cap there are 2 resistor paths. One to supply positive via 220k and the other to supply negative via 150k. For an AC signal, which is what matters, positive and negative supply are the same destination, because they are shorted together (AC only) via the supply/battery capacitance. So you have to calculate the parallel value of those 2 resistors, I'm going to guess that total at less than the smallest (150k) and call it 100k (actually, it's 90k if you work it out) - so 100k and the 10uF.
Using this tool...
http://sim.okawa-denshi.jp/en/CRhikeisan.htm
I enter R=100k and C=10u and click calculate.
Cut off is about 0.16Hz!  It will pass everything. Try C=47n. A smaller value that doesn't need to be polar, but enough for a low-b bass.
But wait, caps have wide capacitance tolerances and low-end cut off elsewhere in the chain might already be cutting the same spot or higher. Better be safe. Will a 100nF cost more or take more space? Probably not, so I'd go for 100nF.
Following the same method, A FuzzFace input at 2.2uF would seem to be low enough for bass


Have to admit, that with distortion I've yet to make my mind up about what and where is best to filter. There are some on this forum that have spent more time tuning distortions than I ever will, so hopefully they will suggest something.
As bass often uses full range amplification, it's not so easy as for guitar, which can be expected to have known limits of top and bottom response (generally 100Hz-5KHz) that your circuit can exploit. If pushed, I'd suggest filtering low end out of the distortion AFTER the distortion and high end BEFORE AND AFTER distortion - then mixing with the clean signal. Here reducing the cap after the clipping diodes might be a good spot to clear the low - it will be a lot smaller than 10uF - say 22nF or thereabout. For pre-distortion hi cut, a small cap (100pF?) on the phase splitter  transistor base-collector? There is already a tone control, but a cap across the clippers might help cut fizz - 22nF or less?

[Gargaman]

Thanks for the effort, mate!

After the cap there are 2 resistor paths. One to supply positive via 220k and the other to supply negative via 150k. For an AC signal, which is what matters, positive and negative supply are the same destination, because they are shorted together (AC only) via the supply/battery capacitance.

Man, this is really enlightening 
Can I say that, in this case, the frequencies below 0.16Hz are being shunted to ground/positive, while all above go to the next transistor?
If I intend to calculate on the Fuzz example, will the input cap interacts with the components before the transistor, i.e. the 100k resistor and the fuzz pot? Will the frequencies be shunt to the closest ground point? I presume that the transitor has a insignficant resistance and will not come into this calcs... Would the path of 33k to power supply be taken into account?
The way signal (AC) interacts with power (DC) in our circuits is still not obious to me, in many ways (as I just asked in another thread about an AB with a LED).
Since I started buildind I try to 'visualize' what happens on it. You helped taking another step forward. I guess 
Sorry about my ignorance, but those are legitimate doubts
regards

[anotherjim]

Yep, AC paths can be obscure. The trick is to recognise which are the significant elements to make a useful guide calculation. My plan is to fudge it on the safe side to get a starting point for trial and error if I want to deliberately control the tone, otherwise most coupling caps want to be all-pass and its safe to make it 2 or more times larger than calculation.

High or low pass RC filters, the cap is a frequency dependent resistance. It forms a voltage divider with the fixed  resistance.
For the high pass, there is in order...
A: Source impedance - before the cap and in series with it. Could be a guitar which you can count as 10k, and op-amp output count as zero, a common emitter transistor count the same as the collector resistor.
B: The capacitor impedance at the frequency of interest.
C: Load impedance - which might be a deliberately chosen fixed resistor or whatever the following circuit presents or the parallel combination.
So it can be a 3 part voltage divider. Say a the frequency you want to check the cap impedance is 10k. If source and load is 10k, then 1/3rd of signal will make it past the cap (1/3 lost in source, 1/3 in cap). At a higher frequency where the cap is worth less that 1k, nearly 1/2 signal makes it through. To get nearly all the signal through, the load needs to be about x10 the source - 100k.

Fuzz Face is intended for direct passive guitar connection, so you have 10k source and the 100k bias resistor. What else is there will have an effect with the base current and feedback, but for practical purposes, you can ignore it. Since the resistor is 100k, it's large enough to ignore the 10k source too - do the calculation - what difference in cut-off frequency between 100k and 110k?

The filtered out signal is lost as a voltage drop across the capacitor and is subtracted from the input signal voltage. It doesn't get shunted away.
Low pass filter has the capacitor as a ground connection after some source & fixed resistor and before some following load - you can think of that as shunting signal above the cut-off away to ground, but also think voltage divider. As a divider, when the cap impedance is close to zero, then ALL the signal voltage is dropped across the preceding impedance so nothing is left to amplify or whatever in the following circuit.

[petemoore]

 If distortion guitar signal is like a file, distorted bass signal is a bit more like a rasp.
The less frequent [yet still 'sharp', quick] diode conduction begins to sound more like ''clicking'', ''buzz'' or ''raspy'' when speaker cone excursions are at low enough frequencies that you can begin to see and visualize their rate of movement.
  An nexpensive boost/eq pedal = easy filter-adjustment experiments allowing quick alteration and for sampling various flavors fed to your clippings, the settings will be a graphic display of dersired taste. The graphic eq's F plot could then be used to guide boost/cut pedal mods.
  A split of low frequency bypass, run parallel to the effect [outputs then mixed] can be considered.