Quadrafuzz

[zpyder]

I'm interested in the quadrafuzz to see how it recombines the 4 frequency ranges after it fuzzes them each... I've been to paia but that schem there seems to come out looking like microwaved used toilet paper.

Anyone know of a better one?

thanks,
zpyder

[Paul Marossy]

There's a full schematic the Anderton's book entitled "DIY Projects For Guitarists" that is probably much more legible.

[Mark Hammer]

I photocopied the article when it appeared in Guitar Player, and it wasn't that clear originally, largely because it was reduced to fit the printed page.  When I scanned it later (which is most likely what PAiA did too), the results were not that hot.  My scan *might* be a little clearer, though.  I'll see if I can dig it up.

More importantly, though, that design struck me as a little unfinished in some ways.  In particular, although the general idea behind band-splitting and separate fuzzing with LEDs is a sound one, the original design tended to ignore the fact that different bands present different amounts of acoustic energy, and that the amount of gain needed to bring the bass up to clipping levels with LEDs would necessarily be different than the level needed to bring upper treble to the same point.  Sadly, the original used identical gain/clipping stages for each band, and I imagine this led to it being less satisfying than many had hoped.  It IS easily fixed, though.

The original simply mixes the four bands down to mono in a very conventional way.  Where it gets a little exotic is that it provides patch-points along the way to the mixer so that you can feed each band out separately if the mood struck you.  That was not a critical aspect of the functioning of the circuit, though, and a perfectly acceptable version could be made without those jacks and break-points.

[Paul Marossy]

I could scan a copy of the schematic from DIYPFG if you would like. I can email it to you.

[StephenGiles]

It might be better to scan an enlarged photocopy, but of course it's all down to the quality of the original print.

[phaeton]

More importantly, though, that design struck me as a little unfinished in some ways.  In particular, although the general idea behind band-splitting and separate fuzzing with LEDs is a sound one, the original design tended to ignore the fact that different bands present different amounts of acoustic energy, and that the amount of gain needed to bring the bass up to clipping levels with LEDs would necessarily be different than the level needed to bring upper treble to the same point.  Sadly, the original used identical gain/clipping stages for each band, and I imagine this led to it being less satisfying than many had hoped.  It IS easily fixed, though.

You know, I had wondered about that.  I haven't seen the DIYPFG schematic, but I did see a hand-drawn schematic on the web awhile back (about a year ago, when pretty much everything was going over my head), but haven't been able to find it since.  The PAIA schematic is pretty hard to follow, yes.

I've experimented with chopping up the signal and distorting each piece seperately, then mixing it back together.  I have a lot of issues like that (unbalanced segment levels, ect), which I've resolved a number of ways, but I was curious how 'experts' like CA solved that in the Quadrafuzz.  I guess he didn't.

If it's not too much to ask Paul, I'd like to take a peek at that schem too.  Only if it's no trouble, that is.

[Paul Marossy]

If it's not too much to ask Paul, I'd like to take a peek at that schem too.  Only if it's no trouble, that is.

I'll scan it later on today. Those of you who want a copy of it, send me a PM with your email address.

[cd]

What CA did to resolve the unequal levels was simple... distort the crap out of them   Each bandpass stage has 30x gain, which coupled with a variable gain input stage (the "attack" control - which goes from roughly 2x to 250x gain), makes for easy distorting of the signal.  May not be ideal but it greatly simplifies the Quadrafuzz's already complicated setup.

[soggybag]

I was looking at this schematic on the PAIA site. It sounds like a very interesting idea. One thing that I noticed is it uses a bipolar power supply. This has kept me away from many PAIA guitar projects. For some reason I feel guitar stuff should run on a 9V; I have built all of their synth stuff.

I noticed the power supply looks like it uses a +/-5V regulators. It seems then that it could be run from a 9V supply.

[Mark Hammer]

Well, you can see it, such as it is, here: http://www.paia.com/quafzsch.gif  It IS a little fuzzy but if you have any graphics package that will "sharpen" images that will work fine.

You will note that R6 through R9 and R37 through R40 (the gain-setting resistors in each of the 4 clipping stages) are identical values. and the high-end rolloff caps in each of those fuzz stages is identical in all but one of the stages.  In fairness, this design was published when we were ALL a lot younger and less experienced.

[cd]

I noticed the power supply looks like it uses a +/-5V regulators. It seems then that it could be run from a 9V supply.

With the usual 7805 and 7905 you'd need 14V minimum.  Two 9Vs would do it.

However the Quadrafuzz is all simple opamp gain stages, it could adapted to a 9V, 4.5V, and 0V supply quite easily.  Not with the existing PAIA PCB mind you, but if one were building from scratch it wouldn't be an issue.  The filter depths might be different though.

[Paul Marossy]

From what I remember, there's more than a few opamps in that circuit. I would think that batteries would not last that long with this circuit...

[phaeton]

Well, you can see it, such as it is, here: http://www.paia.com/quafzsch.gif  It IS a little fuzzy but if you have any graphics package that will "sharpen" images that will work fine.

You will note that R6 through R9 and R37 through R40 (the gain-setting resistors in each of the 4 clipping stages) are identical values. and the high-end rolloff caps in each of those fuzz stages is identical in all but one of the stages.  In fairness, this design was published when we were ALL a lot younger and less experienced.

Well, yeah... it's fuzzy coz it's a quadrafuzz....






True about 'we were all a lot younger'.... I notice a lot of old designs are like that.  Neat and elegant in their simplicity, but can *always* use improvements in the way of noise reduction, level balancing, etc.  Lots of CA designs are like this, but one must remember it was a toadilly different era.

I don't hold anything against anyone.   

[jimbeaux]

Here are pics of my QuadrFuzz build.

http://s79.photobucket.com/albums/j133/Jimbo_99/

I did use break-out jacks to tap into the individual fuzz outputs.

Have split the fuzz outputs to separate amplifiers, and the stereo image sounded great.

[cd]

From what I remember, there's more than a few opamps in that circuit. I would think that batteries would not last that long with this circuit...

Nah it can be done with three quad opamps, use TL074s and two batteries would last a good long time.  Especially if you leave off the FX loop and electronic switching.

[Paul Marossy]

Nah it can be done with three quad opamps, use TL074s and two batteries would last a good long time.  Especially if you leave off the FX loop and electronic switching.

Well, I guess that wouldn't be as bad as I thought.

[Mark Hammer]

"Running" and running for a long time are two different things.  I can't see this as drawing less than 15-20ma even in reduced form.  That is certaily *usable* on batteries, just not for months at a time.

On the other hand, if one was willing to forego a couple of the optional things, like the split quad outputs, send/receive loop, and the input/output options (line level etc.), this is exactly the sort of thing that could be pretty easily built into a 1590DD and have some sort of internally regulated bipolar supply fed from a wallwart with an AC output.  Indeed, if PAIA were smart, the market seems just about right for something like that.

Okay, here is a hastily editted, cleaned up, and "reduced" version of the circuit: http://img.photobucket.com/albums/v474/mhammer/moddedQUAFZSCH.png
Note the following changes:
- all electronic switching omitted.  This assumes the input and output would be switched via the usual DPDT or 3PDT stompswitch
- guitar level in, and guitar level out.
- no separate routing of individual frequency bands.
- no separate send/return loop for tonal processing.

Essentially this becomes a distortion unit with more tonal tailoring than usual, rather than the studio distortion "thinkbox" it was designed/conceptualized as in rackmount version.

Things to change:
1) R2 (upper right hand corner) should probably be 10k rather than 2k2.
2) R39 should probably be bumped up to 150k. R38 should be bumped up to 220k and R37 up to 270k.  This will even out the amount of clipping in each band so that they all clip roughly the same amount.
3) Depending on your tastes, C4/5/6 may need to be reduced slightly to compensate for the increased resistances in #2.
4) If this was to be a stompbox, then there should be a DC-blocking cap at the input just after the 10k resistor.  A value of .01 is probably fine.  And because of that, there needs to be a 1-2.2M terminating resistor on the input to de-pop the switching.

The overall Fuzz level is optional.  The four individual band levels also help to set the overalllevel but then you lose the proportions.  Your choice.  If you need the panel space, the overall level control can be omitted.

[zpyder]

Thanks for all the replies...

I'm mostly interested in the method with which the signals are recombined.  I would like to create a similar device using effects other than fuzz.  Actually, I would first like to create a box that simply a) splits the signal into two frequency ranges, b) goes out to an fx loop for each, and c) recombines them for output.

I wonder what is required to recombine the signals after they've been introduced to seperate effects loops?

thanks!
zpyder

[cd]

I'm mostly interested in the method with which the signals are recombined.

Garden variety opamp summer.  Textbook circuit.

http://www.allaboutcircuits.com/vol_3/chpt_8/8.html

[Dave_B]

Here are pics of my QuadrFuzz build.

I don't think I've ever seen that much RG-174 in one place.