Fuzz Face w/ switchable pickup sim and input volume pot... what am I hearing?

[aion]

I'm working on an update to the Solaris and ran into an interesting issue when using a switchable pickup simulator. I was wondering if anyone could explain what's going on.

Here is the schematic. (Power and switching not shown, but it uses a TC1044 to get -9V, while the JFET is powered by the positive voltage.)



When the pickup simulator is switched on, the circuit cuts out and it takes about 2-3 seconds for it to come back to life, like a capacitor charging. This is not something that's going to happen in a live setting so I'm not terribly bothered by it, it's more of a set-and-forget switch based on the pedal's position in the chain.

The bigger issue is that when the pickup simulator is active, any time the "Input" control is turned, it does the same thing: the sound cuts out and then fades back in after a bit. The bigger the change in knob position, the longer it takes to come back, but typically 0.5 to 2 seconds. If the simulator is off, it works as expected with no cutting out.

Once the sound comes back on, everything sounds correct and the input control still does what it's supposed to. And the issue seems to be fully contained to that one control - all of the other knobs behave normally. In addition, when I remove the Input, Body and Contour controls (i.e. revert it to a stock Fuzz Face with the pickup sim) everything works perfectly, other than the slight cutout when engaging the pickup sim (although the cutout is much shorter, about 0.5 seconds).

If this is the nature of the beast then I can live with it - I just want to better understand what's going on and know whether there is a design issue someplace.

[Ben Lyman]

probably C1 charging up. I would move 2M2 (or add another) to the other side of the switch

[aion]

probably C1 charging up. I would move 2M2 (or add another) to the other side of the switch

I don't think that would solve the input control issue - the knob is already acting as a 250k resistor to ground, so the 2M2 resistor would just serve to reduce its value slightly.

[Ben Lyman]

So, the knob is only connected to the input jack when you throw the toggle switch, but it leaves one end of C1 "hanging" out there with no ground connection

[aion]

So, the knob is only connected to the input jack when you throw the toggle switch, but it leaves one end of C1 "hanging" out there with no ground connection

Right, but again, the issue isn't what happens when the pickup sim is toggled in and out - that's pretty minor and I only provided it as info because it may be related. The bigger issue is what happens when the pickup sim is enabled and the input knob is rotated. Here is a reduced test case for clarification -



In the above circuit, any time the input knob is rotated, the signal cuts out for 0.5 to 2 seconds depending on how far the knob was turned. I'm wondering if there's some sort of interaction between the transformer and the input knob that's causing this to happen when the resistance after the transformer is altered.

If there's no sensible explanation then it's always possible I have a bad part someplace, but as I mentioned before, it sounds perfectly fine once the sound comes back in, and all other controls work as expected.

[lcv]

Across the input pot there is about 4.5 V DC.  When  you operate it, the wiper is dragging up the voltage on one end of C4.
The other end of C4 is momentarily dragged up too, saturating the first Ac128. Then C4 charges up and everything is back to steady state.
Just my take

[Ben Lyman]

ahh, yes. ok I see what you're saying now. So, it's really the input pot that concerns you, that's a mystery to me. Could be the transformer reacting to the change, I don't know. Maybe look at C4, turn it around or put 2x 22uF caps facing each other instead, to negate the polarity. Sorry, not much help. I still think you could move the 2M2 though, it's not doing anything for C1 unless the sim is engaged; and like you said, when the sim is bypassed, it's just parallel to the input pot and not really doing anything anyway

[antonis]

Could be the transformer reacting to the change,

Maybe a speed-up cap across winding edges 1 - 3 should settle things up..
Or a DC blocking cap right somewhere between Q3 Source and INPUT wiper, in case of C4 issue, as said by Icv..

[aion]

Yep - thanks antonis and lcv, that was exactly it. I put a 100n cap between the JFET and the transformer and that solved it entirely.

The Eruptor gets away with it because the pickup sim isn't switchable and it goes straight into the FF input cap, but my two changes (switchable pickup sim and input volume pot) meant that the cap was necessary.

[MikeA]

You didn't ask about this, but just a random thought: the DC resistance of this transformer will approximate a pickup, but the inductance value (not listed on the data sheet) will probably be one magnitude lower measured @ 1 kHz.  I've measured this one ~ 300 mH vs humbuckers  ~ 2 to 8 H.   May or may not matter, depending on how you set the other component values, and higher L values are hard to find while this trans is readily available.  Mike