Copyright 2010-2022 by H. Gragger. All Rights Reserved. All information provided herein is destined for educational and D.I.Y. purposes only. Commercial re-sale, distribution or usage of artwork without explicit written permission of the author is strictly prohibited.The original units  with their associated  trade-names are subject to the copyright of the individual copyright owner. The Author is by no means affiliated with any of those companies. References to trade names are made for educational purposes only.By reading the information provided here you agree to the Terms of Use.

This is a report of the findings I had when I tried to get the swell part working.

There are several almost-clones of the FOXX tone machine around, namely the ultimate octave (just a few time constants different) and the Experience Octave Pedal.

Pot values seem to be widely different across the schematics I found. Fuzz Central suggests all 50k lin Pots, so do GGG, the Green-Fuz schematic suggests 100k log for all except tone. Noel Grassy confirmed 100k lin pots in his original unit and that worked for me too.

Since there are at least two good PCB layouts for the FTM around (I made the Fuzz Central board) it seems feasible to just make another little board for the swell (and so I did), but the recently recovered Green-Fuz layout has been cross-checked for the swell part at least  and should work too. It appears a little less crammed than other builds.

Back To Index

---

Building And Modding The FTM Section

A word ahead: this is not a beginners project, this is by far the most elaborate pedal I ever made. This device uses many controls and switches and the prototype ended up with no less than four separate PCB´s. Even if somebody made a PCB that incorporates all, there are still many wires to tame. It was a challenge to fit everything inclusive battery inside the case, even although I used a higher case. Due to the complexity I do not recommend to build this from scratch for a novice. It may nevertheless be worthwhile to incorporate one or another idea into an existing design.

• Octave diode rewiring: Mark Hammer noticed that the FTM does not really get clean when in non-octave mode, because the signal always has to pass one diode. This always gives strange ring-modulator effects. He suggested a simple mod to wire the diodes around the octave footswitch different (and in another thread), which improves the non-octave sound dramatically. This sounds much cleaner and makes the FTM very useable even as a non-octave fuzz.
  

FTM w/ swell octave switch details

• Diode lift switch: I also installed a dpdt toggle switch to lift the diode pair after the octaving section as described there. Although a spdt switch would suffice for this purpose, the second half can be used for volume compensation when switching, similar to the way I did it for the Fuzz Ride. Lifting the diodes out of the circuit will invariably produce a big volume increase, which is obnoxious. A simple resistor in series with the FTM´s output (on the hot side of the volume pot) that gets bridged when the diodes are switched in will produce the desired volume equalization. [Note that this does, contrary to a sag switch, not interfere with the overall system gain, because the diodes, when on, limit the signal strongly and we just have to compensate for that when they are off.]
  You may need to fiddle with the compensation resistor to cater for a whole range of distortion settings. You should also install a 100k resistor across the switch contact that enables the diodes to stop DC shifts which may be heard as switching pops.
  

FTM w/ diode switch details

• Ultimate octave mod: The only difference to a stock FTM (as described here) are a few caps and a switchable tonestack. I found that the existing tonestack works fine and more extreme settings are unnecessary. The mod is, amongst other threads, described in DIYstompboxes.
• Transistors:  I tried the 2n2369A as used in the YAFF. Those are low gain (60-90) and sound flatulent.
  Good for a fuzz face, bad for a FTM. I used BC546B which sounded instantly good and that was the end of swapping. Gain is around 260.

Back To Index

---

The Dreaded Swelling Experience

The early Pete Cosey dying amp sound can nicely be done with the swell, although he achieved it in a totally different way (analog synthesizer), so it is said. A crazy effect. That´s the reason I am after the effect - not the acid swell per se.

I have spent many an hour at the P.E. swell circuit and I was close to quitting, but I finally got it working. To cut it short, this is only KIND OF A SWELL effect and not worth the effort, if you are looking for a decent swell. Moreover it has allures and only works under certain circumstances, some of which can be cured.

There is rather a long rant following for those interested to debug their swell. Others skip ahead ;-)

I have tried EVERYTHING on this circuit, swapped transistors with high and low gain, tweaked resistors, nothing worked.
The circuit kept behaving more or less like a small gain amplifier without changing the tone much. I finally left the plastic can transistors of hfe=500 because it did not seem to make any difference.

R.G. Keen has encountered that the swell part only works with a fresh battery. I tried stiff voltage (local electrolytic decoupling, which is akin to a new battery) and a several kOhm sag resistor in series, higher voltage, lower voltage, which did not change the function of the modified swell circuit either way.

As others have mentioned, the swell only works if it is driven by a fairly heavy signal. I observed the same thing. I therefore sincerely doubt that fuzzes like the Ampeg Scrambler do anything with the swell (this has a gain of nearly unity...) despite its extended HF spectrum.

It was interesting, that as soon as I took the oscilloscope probe (which has 50 pF and 1Mohm input impedance) to the input of this circuit, the signal was immediately muted. Obviously this circuit  has such big impedance, that 1MOhm would  shift too much. After removing the probe again, it did the swell action with the time constant implemented. So I knew, the circuit was perfectly capable of doing its thing. But this made it kind of hard to debug.

I had the volume of the FTM full up (and the swell nearly full down) and it was kind of farty distorting, but not swelling. Due to an accident I inserted an capacitor (same size as swell input cap) where the hot side of the swell pot is connected  and lo and behold, it worked.

As mentioned, the swell is flatulent. It seems like the massive signal on the input drives the swell circuit into some overload condition that swamps the signal momentarily and with the RC time constant it recovers, delivering a swell like effect. The only snag is that the signal gets heavily compressed and distorted when in the overload condition.

This sounds indeed like the amp is on the verge of givin´ up - good effect for that, but lousy swell effect. During this  choking condition the tone changes to trebly until it recovers. This is probably one of the nicer aspects of the thing, because it does synth-like effects.

The drawbacks of this hookup are:

• high volume setting on the FTM. When you switch the swell off you have massive volume. This is fortunately easy to remedy, since you just have to take the signal into the swell directly from the FTM board output rather the volume pot center lug. There is no reason for an additional amp section in front of it as somebody suggested - the FTM has massive gain reserve.

• The interesting portion of the effect happens on the swell pot almost fully down. On 95% of the rotational angle of the swell pot happens no swell, just amplification. This is easy to remedy as well. Put a 47k resistor in series with the hot output of the swell pot connection and  use a 2k pot on the ground side. This works a treat and lets you adjust the swell volume to match the unswelled sound level and above. Since the input is flooded with massive voltage and no more level setting in between, the only setting is the guitar volume control. You will notice, that changing the guitar volume does almost nothing for loudness level due to heavy compression, but changes the behavior of the swell choking  until down to zero of course. "Choking" - a name more suitable for the  effect than "swell" ;-).
  
• At zero input volume you encounter massive noise.

On the original P.E. units the swell is only functioning in octave mode. On my breadboard I have the luxury to have everything switchable, but it is true, fore some reason it only works in octave mode, but it does indeed work with the diodes off (which I call "native" mode) to a lesser extent.

This observation would lead to the conclusion, that an extended upper spectrum was needed to get the swell working, but indeed the swell only works with the tone pot towards a dark tone (treble cut).

Adding A Tube-Like Circuit


A tube amp of vintage variety has its input usually DC coupled to the first tube´s grid. R.O.G.´s fetzer valve does more or less emulate such a tube input to an extent, so I slung one together. Using a BF245A I would even have the 2V pinch-off voltage similar to a tube stage. The drain resistor sets the gain (relative to the source resistor), so making this one zero you have a gain of unity, a.k.a. a buffer - almost, since those are normally AC coupled. The output would be derived from the source and the ground cap there removed of course, since we don´t need additional amplification.

The DC bias on the swell shifts the j-fet´s input into a non-conductive region just as expected. This valve-like stage therefore not only does the trick independent of the following device´s characteristics, but also serves as a low impedance driver into the bargain.

Note: any other buffer that is not based on an auto-biasing j-fet input stage or similar does will very likely not achieve the desired effect.

Since the added series resistor introduces quite a hefty drive impedance (by a magnitude greater than the unaltered drive resistance), the output drive capability might be impaired, so the before mentioned j-fet stage has been moved from right after the swell section towards after the "swell" switch (which I named "choke") to cater for both circuit sections. The absence of an input blocking capacitor does not present a problem, since the FTM´s output is AC-coupled and there is no DC on the volume pot.

Back To Index

---

Reducing Pickup Loading With A Front-End Buffer

I also observed that driving the FTM directly with stock single coils over a few meters of good quality cable results in a dramatic darkening of tone as soon as the guitar´s volume is taken back. This can only be caused by heavy loading of the guitar´s output.

Again, a standard buffer in front of the FTM and that is the end of that. There is little concern about current consumption, because the FTM plus swell is with a consumption of about 2.6 mA inclusive two LED´s one of the most frugal stomp boxes I know.

While the added circuitry increased complexity somewhat, the number of possibilities has increased too.
I have refrained from making another complete schematic drawing; Since there are probably thousands of FTM clones out there, I rather display a block diagram with the add-ons that allows every builder to hot-rod their existing device to taste.

Back To Index

---

Improving The Swell Feed  And Restoring Useful Control Range

As mentioned before, the swell needs a fairly hefty signal on its input. Taking the full output signal of the FTM (on the outer lug of the volume pot) worked, but experimentation showed that a slight signal reduction would improve the situation.
It also turned out that the swell works best when treble is pretty down (despite the fact that it works better with octave setting). Since the tone control is fully down, there is not much leeway left for tone adjustment in the swell mode. It looks like the swell wants for good functioning a much lower treble setting than were useful in the "fuzz" mode.

Both conditions are very unhandy, because when switching the swell off (back to "fuzz") you are left with very loud and dark settings.

Fortunately both challenges are very easy to remedy. The swell can be fed via a separate resistive divider (a frozen separate volume control so to speak) that sets a level somewhat below full signal, but independent of the volume control.
In series to this resides an additional passive low pass filter. This way the mix also contains less treble.

I also observed that the swell does not work so good on low strings, which I attribute to increased bass content thereof.
I replaced the 220n cap on the swell input with a 22nF cap and this improved the situation vastly with no apparent bass drop in the output signal.

The new feed network allows for independent volume settings for the FTM and swell part and restores the usable and useful range of the tone control for both sections.

The values were devised empirically and can be tailored to need.

FTM swell feed details

Back To Index

---

Using The Modified Swell

The overall gain and compression FTM+swell seems so high that the swell texture is mostly subject to the guitar´s volume control with a given setting of the "sustain" control. [This is another good reason why the guitar´s smooth volume control function should not be crippled by a load, read more on my notes on buffering].

The swell´s behavior can also be modified by the FTM´s tone control setting.

By the application of the above mentioned measures the swell is to an extent freed from the bondage of the octave mode.
It works in several modes to a lesser extent. In some mode it cuts the signal envelope rather than swelling, which gives a trumpet-like effect. The Listen to the sound demos.

R.G. Keen observed with the units he had that the swell part only worked with a fresh battery. I tried stiff voltage (local electrolytic decoupling, which is akin to a new battery) and a few hundred Ohms sag resistor in series, which did not change the function of the modified swell circuit either way. So it appears pretty stable.

Back To Index

---

Downloads