Foxx Guitar Synth Wah

[DiscoFreq]

I have one of these that needs some repair... 

What's wrong with it?

I think it was a problem with some of the settings (LFO/envelope/...) not sweeping the filter as they should, but I should check again, it's already 2 years since I tried it...

[digi2t]

Well Mark, it's a shame that the pedal pot is being held in a vise for the time being, and not inside a wah pedal. I think being able to reverse the sweep would be a hoot, apart from all the other bells and whistles on this puppy.

As for a Mutron foot control mod, using the Foxx and Mutron schematics, we can see that we can graft the pedal and LFO part in fairly easily to the Mutron. Maybe a jack in the side of the Mutron, and a rebuilt wah with the extra circuitry on a small board that plugs in? I think, Plug n Play of sorts. Am I out to lunch here?

This thing has really sent my imagination on a magic carpet ride  .

I already have the empty wah shell ready for it (move over Ludwig Phase II build ). As soon as I finalize everything on the board, it'll be vero and build time again.  

[Mark Hammer]

The wah-antiwah thing is relative to how one intends to "express" with a given filter type.  Certainly bandpass works "best" with forward providing the most emphasis, since that's where all the harmonic content lives.  Lowpass tends to work a bit betterwith foot-forward opening up the filter, though closing the filter with foot-forward is useful too.  Personally, I found that sweeping the cutoff downward with foot forward worked better in hipass mode, since that would result in the guitar having "more balls" the more bandwidth you gave it, and more bandwidth meant flipping the sweep in the other direction.

Direction reversal for autosweep is an entirely diffeerent thing than foot sweep, because there is a sort of "psychic link" between verbal expressive intent (and wahs/filters are always a translation of what you want to do with your mouth into something done with your foot) and how you want to move your foot.  EXpressive intensity is inextricably linked to moving forward, not backward.  Or at least that's how MY brain/foot works.

The downside to having switchable wah/anti-wah sweep is that the pot MUST be linear, or else you end up with a whack job taper trying to go in the other direction.  Gotta sweep the same way i the one direction as the other.  That may not work for everyone's tastes, although I suspect that the sonic difference between a state-variable filter and the oddball bandpass of a Cry-baby obliterates any memory of taper preference in a wah.

[Mark Hammer]

Don't forget to give yourself a notch-filter setting (HP+LP), in addition to a variable Resonance control.  The schematic shows a two-position setting for Q, using a 100k and parallel 22k resistor.  NO reason why that can't be a 15k in series with a 100k pot.

[digi2t]

Don't forget to give yourself a notch-filter setting (HP+LP), in addition to a variable Resonance control.  The schematic shows a two-position setting for Q, using a 100k and parallel 22k resistor.  NO reason why that can't be a 15k in series with a 100k pot.

Thanks for the heads up Mark. It's already there on the board, the hi/lo, and the Q. The Q is a 12K resistor with a 100K pot. I figured why have a switch, when I can have a pot instead. According to R.G.'s Neutron article, he states that the Mutron had a 150K pot. He recommend using a dual gang 100K in series, but I found 200K pots on EBay, so I'm going with that. I must say though, the 100K does a nice job. They will be part of the finished product, for sure.

As for the voltage divider to balance the up and down modes, the 250K is OK, but it's still out a touch. I have a 500K trimmer, so I might try to cheat. A pair of 470K resistors across the lugs will give about 326K. I'll see if it does the trick.

[Mark Hammer]

I keep reminding folks that upward and downward sweeps on autowahs are not simply the direct opposite of each other.  The feel and starting/ending points of each generally need to be a little different for the most "musical" sounds.  All the Mu-Tron derivatives tend to acknowledge this by adjusting the start point for downward sweep a bit.  Of course, that doesn't mean the adjustment suits our individual tastes.

Do you have a copy of either EPFM 1 or 2 or the Super Tone Control project?  I can scan it for you if need be.

The notch setting produces something functionally equivalent to a Phase 45, so with the LFO moving it around, you get yourself a simple phaser for free.

Finally, the über-version would have either a send/receive loop between the input stage and filter, so you could sense the envelope prior to external processing, or a dedicated clipping circuit to feed the filter something more harmonically rich....just like the Q-Tron+ does

[digi2t]

Do you have a copy of either EPFM 1 or 2 or the Super Tone Control project?  I can scan it for you if need be.

No, and yes please.

Finally, the über-version would have either a send/receive loop between the input stage and filter, so you could sense the envelope prior to external processing, or a dedicated clipping circuit to feed the filter something more harmonically rich....just like the Q-Tron+ does

Funny you bring that up. Today I found an article by Beigel and Randall on Envelope Controlled Filters, and lo and behold, there's a section on adding a loop. Very well written for the noob like me. I'm sold on the loop.

I did try J.C. sensitivity control tonight, but it didn't really blow my skirt up. The effective range was very small, and nearly all at the end to the dial. I'll stick to the resistors. Speaking of which, I err'd when I said previously that the Mutron had two 22K's next to the up/down switch. It's a 12K and 22K, like the Foxx. My copy of the Mutron diagram wasn't the cleanest, and I found a cleaner one today, showing up my erroneous statement. My apologies.

The "cheated" trimmer works better than the 250K, or 500K. A 500K, with two 470K resistors across the lugs, does the job better. It seems to dial in a more balanced response down versus up. I'm assuming because the resistances on either side of the center lug are closer to the original specs than before. I've got it at 235K/65K, versus the 150K/82K spec'd, but it sounds very good. 330K trimmers are available, but not in multi-turn Bourns, which I prefer. I'll just add the resistors to the vero when I get there. I just have to decide on where I like my sweep right now. It seems that every day I tweak that damn trimmer a bit. I'm just trying to spread the love between the HP and LP I guess.

I'm going to try the loop next. Afterall, I have to shoe horn this puppy into a Crybaby. Besides, like the Skyripper, I prefer to have the option of what I want to plug into the loop. Good for people who get bored easily  . If that works well, then I'm drawing this puppy up, and building it. I'm putting a fuzz box on the tarmac  .

[Mark Hammer]

If you run out of space in the chassis, you can simply do what so many mixers do: use a single stereo phone jack for both send and return, like a sort of "mini-snake".

Alternatively, just use a single op-amp stage where the loop would go, and plunk a Distortion+ type circuit in there.  Realistically, once the filtering is applied, the "character" of whatever is placed in there won't really shine through all that much, so there's no sense in driving yourself crazy about picking the "right" on-board source of extra harmonic content.  Really what you're looking for is a simple clipping stage to intensify the filtering effect that you can toggle in and out.  A Dist+ with a trimpot for adjusting gain and another to yield clip/no-clip volume balance, is pretty much all you really need.

[digi2t]

there's no sense in driving yourself crazy about picking the "right" on-board source of extra harmonic content.

I get the feeling that you're starting to get to know me. I think that's why I'm leaving it somewhat open ended by going with a loop.

The mini snake idea reminds me of some pictures of older Schaller wah pedals that i came across recently. It wasn't for stereo or anything like that, they just had a jack, and a cable for the in/out. I don't think it was very practical though, if one wanted the pedal further away from the amp or instrument. It intrigued me for the longest time as to why my Schaller had a hole in the casing, right next to the switch. I thought it was a manufacturing thing, but I found out why recently when I saw the pictures.

[digi2t]

Well, I decided to put down what I have so far. I also "modernized", including the MAX1044 that I'm using, the trimmer to balance the up/down modes, and the use of VTL5C3 Vactrols. The "Pedal/Envelope" switch is modded now to a DPDT to kill the +9 volts to the oscillator, effectively doing away with any possibility of ticking in the audio. I did not use the 20K and 220K resistors in the filter section as per the Foxx. Instead, I went with the Mutron diagram, and sent these two pins to ground.

This makes things a lot clearer than that other "chicken scratch" diagram.



I haven't added the effect loop yet, but I haven't tested that yet, and I can just update this diagram later. This drawing will give you a working effect though. It sounds pretty good on the breadboard. I forgot to mark down what my trimmers are reading though. I'll put that down when I update, since it will give a good starting point for tuning if anyone else decides to build this.

I'm testing the effect loop tomorrow, and from what I read in the Beigel/Randall article, I think I should be splicing it in between the buffer and the VCF. I just not sure if it should be all before, or after the 4.7K resistor. Or, if "IN" should be before the resistor, and "OUT" after the resistor. I guess testing will tell.

I am debating whether I should make this true bypass, or just leave it as is. I don't know how it will "play" with other effects right now, so I'll leave it in stock trim for the time being. It's easy enough to convert to T.B. if need be.

*edit* - After a good nights sleep, I thought of looking at the Meatball circuit. Shows very clearly where the effects "send/return" should be.

[pinkjimiphoton]

well done again, dino...

are you gonna do a vero of this one, too?  i hope?

[digi2t]

OK, version 2...



This one has the effects loop, and the output volume as well. I figured the volume may come in handy if you're using the effects loop, to match up with any other effect one might decide to use. Also changed the 3.3uF cap to 4.7uF tantalum, since that's what is on my board. 3.3uF works fine as well, but at the time, I didn't have any. I'm sticking with the 4.7uF. Also, the 10uF cap going to the detector is now a 1uF.

It was kind of bugging me that I had these open ended connections insofar as the loop is concerned, but I don't seem to be picking up any noise when nothing is plugged in. Besides, I lifted this from the Meat Sphere, which seems to have an overdrive in it already. Since the Foxx doesn't, it doesn't really have anything to loop around. Then I thought about the Skyripper, and I noticed that if you're not in Ripper mode, with nothing in the loop, the connections are open ended, so it should be fine. I'm speculating here, so, Mark, you can chirp in anytime now  

All in all, the loop works fine. I tried it with a Geiger Counter, and an Uglyface. The result was quite satisfactory. The filter really goes to work on the signal. The 10K resistor fits in nicely to keep the loop, and the filter, at more or less even gain when bypassing the external effect. There is a slight loss of "wonkiness" (in the higher frequencies) when the effect in the loop is bypassed, but I can live with it. Nature of the beast, I guess.

On to vero-land now  

[beatnik]

great job! will build this very soon

thanks for sharing

[Ronan]

Dino, there might be a typo on Z2-2, both inputs are tied together. Another thing that has me baffled is how the flange outputs are any different to the HP and LP outputs, except for the series resistors 47K and 68K. Other than that, looks like you got it sorted. From the vids you did, I think there will be some really nice fat rich wah sounds come out of this one! Very tempting.

[digi2t]

Good catch Ian. I also spotted that, plus a mistake above Z2-3, with an extra line between the diodes. I've corrected it. I'm removing the other schematic as well, it had the same errors. Besides, the v.2 drawing has most all the elements of the original drawing, plus some added goodies. If anyone wants to build a Foxx in stock trim (or an upgraded Mutron?), it's an easy retro-grade.

BTW, I'm convinced, that "brown turd" on the pot is the 2.2uF capacitor. I did a bit of research, and they are available in metal film, box and axial form. I'm not sure whether it being non-polar really makes a difference in this case (for the LFO), but I do know that the metal film has a tighter tolerance at 5%. Going to bed less stupid, again  .

As for the "flanging", I'm only using the term that's on the original diagram. It ain't no flanger. It's more useful in LFO mode. I think the resistors are cutting back a bit on the filtering, so more of the guitar signal cuts through. More like a background shimmer of sorts. The bigger resistor on the LP makes sense, since the bottom end cuts through more than the HP. Personally, I debated keeping this, but I didn't want to massacre the original circuit either. I prefer to keep the idiosyncrasies that made these pedals special, "ahead of their time", or just plain stupid, for their day. That's the mojo factor I guess, it reminds us from whence we came.

Back to the vero board... and more head busting. I've got to fit this into a Crybaby  

[Mark Hammer]

The "flanging" (cough, cough) setting is simply the notch produced when highpass and lowpass outputs are combined.  The Anderton Super Tone Control does this.  While it certainly can be used in foot-control mode, it tends not to be as noticeable an effect as the other 3 modes, simply because it retains much more than it filters out.  If you feed it a very harmonically rich source, it becomes more noticeable, though.  Used in LFO mode, it is functionally equivalent to a Phase 45, that produces a single notch.

The notch that IS produced occurs because of the manner in which the highpass and lowpass functions are produced.  because neither of those are particularly steep, and they don't overlap, they leave a small zone where stuff falls below the corner frequency of the highpass, but above the corner frequency of the lowpass, producing a moving midscoop.

I draw your collective attention to the difference between the Mu-tron and Meatball with respect to how the two variable resistances in the filter are swept.  The Meatball uses a 120k fixed resistor in parallel with the LDR in the first section and the standard 220k for the second.  I remember a discussion way back when we were trying to decipher the Meatball, where someone noted that this tended to produce a sweeter, more musical, change at the highest points in the sweep.  Never tried it myself, but iof you still have it on the breadboard...

Finally, why is the return in the loop brought back to the inverting input in the re-draw?  One would think that the send/receive loop would interrupt the 4k7 path to the second op-amp.  I can see why one might be interested in futzing with sum/difference things, but the loop is intended to provide a pre-processed signal that doesn't interfere with envelope detection.

[digi2t]

Finally, why is the return in the loop brought back to the inverting input in the re-draw?

Hmmm... major brain fart (or too many night shifts!) on my part. As I mentioned before, I was looking at the Meat-Sphere, and the blinders went on  . As the old adage goes about assuming... don't! Thank God guys like you are around to spank us out of our temporary stupor  . I did find that it sounds better with the loop behind the resistor, especially with high gain devices like an Uglyface. I tried all the different configurations (pre, mid, post), and what is on the drawing worked best for me. Redraw, coming up!



As for the variable resistances, it so happens that it's still on the breadboard, and I was using 500K trimmers sub'd in for the 220K resistors on both LDR's. I went on a hunch early on that this might be a place to fool around with the circuit. I did try the Meatball settings, along with a slew of different combinations, but in the end, they always found their way back to 220K. It especially didn't produce any positive results with a high gain device in the loop, I was getting some howling in the sweep when trying lower resistance levels.

Here's a short video of the effects loop in action;

Sorry about the sound quality, I should have turned the amp down a touch. I only used the H.P. and B.P., resonance at about 50%, but the filtering really pops the dirt right in the Chicklets. I like. I keep.

[digi2t]

Alrighty then... here's a vero. If anyone would like to peruse it for errors, I'd greatly welcome it. I'm still going over it myself, but sometimes, things get blurry after a while  



Although it's a bit big, it can still comfortably reside inside a Crybaby, sideways. If anyone can see a way to shrink it more, please pipe up, but 2 Vactrols take up board space.

There are off-board connections, but I didn't put them on the layout just yet. The schematic fills in those blanks. Once the vero is verified, I'll add them to the layout, to consolidate things.

I also changed the name to the "Muttoxx", seeing how Foxx is back in business. Besides, there's more Mutron than Foxx in this thing anyway. Hence, a "mutt". Other name recommendations are heartily welcomed. God know there are people out ther with a lot more imagination (or time) than me  .

*EDIT* - Just spotted a missing cut under R23. Noted, and corrected on future issue.

[Ronan]

Hi Dino, just doing a check on the vero, and you may have already found it, but it looks like pins 9 and 10, and 11 and 12 are mixed up on the upper TL074. (10 and 12 should go to gnd, not 9 and 13). I can't find the second 10uF cap on the MAX1044. The ENV. SENS POT 2 at the top left needs to move down one row. I ordered some TL074, will try it with discrete LED's and LDR's instead of vactrols, and see how it goes.

Edit: oh yeah, its boxing day over here now, hope y'all have a great Christmas day!

[digi2t]

Hi Dino, just doing a check on the vero, and you may have already found it, but it looks like pins 9 and 10, and 11 and 12 are mixed up on the upper TL074. (10 and 12 should go to gnd, not 9 and 13). I can't find the second 10uF cap on the MAX1044. The ENV. SENS POT 2 at the top left needs to move down one row. I ordered some TL074, will try it with discrete LED's and LDR's instead of vactrols, and see how it goes.

Edit: oh yeah, its boxing day over here now, hope y'all have a great Christmas day!

Thanks a bunch Ian. Nice of you to take a look. At some point, I can't see the forest for the tree, so it's nice to have an objective eye. The second cap for the MAX is located more towards the center of the board, to the right of the chip. Your observations have been duly noted (and greatly appreciated!), and I'll make the necessary corrections.

I'm going to try and put together 2 LDR's, and an LED, into a package. I gutted an FRN type fuse, cutting the ends off. It's a nice, 3/8" diameter, black fiberglass tube body, which I'll encapsulate the LED and LDR's inside. That way I'll save the Vactrols for something else. I'll take pics of the process, if it works out, I'll post them.

All the best wishes to you and your family this holiday season.
Cheers,
Dino