Best Way To Put An Electrolytic On A Switch?

[Ginsly]

Long story short, I have a modified Superfuzz pcb and want to incorporate a switch that can cut the connection between Q3's emitter and C8, a 10uf electrolytic cap. Since it's polarized I want to make sure I'm setting this up correctly... How would you do this if you were me? Seems easy enough but I think I'm complicating it... not even sure which kind of switch I'd need! Yeesh...

 

[jcpst]

Have you seen the tutorial on switches here? I thought it was great and it helped me out. Not a direct answer to your question, I know. But it’s not a terribly long read either.

TUTORIAL - Which Switch is Which? - Switches Explained

This is the Test Kitchen, so let's talk about switches! In the pedal building world we really only come across 4 different kinds of switches: toggle, rotary, push button (stomp usually), and DIP. The different kinds of switches merely indicate how you 'activate' the switch. I'll explain the...

forum.pedalpcb.com

 

[Chuck D. Bones]

I'd lift one end of R14, connect a wire from the flying R14 lead and another wire to the pad on the board where R14 used to go. Run those wires to a SPDT ON/ON toggle switch. One wire goes to the center terminal, the other wire goes to one of the end terminals.

It will POP when you toggle the switch. If that's a problem, then connect a 1M resistor across the switch terminals that have wires on them.

 

[PedalBuilder]

Are you trying to set up an octave on/off switch? There are better ways to do that. Try wiring the switch to disconnect the collector of Q5 instead. With the Q5 collector disconnected, there’s no octave. Or you can place a 5k pot wired as a variable resistor between Q5’s collector and the rest of the circuit. (A C-taper works best, but a B taper will do in a pinch.) It might not seem like much resistance, but it’s enough to make the octave disappear. Connect lug 1 to the rest of the circuit, lug 2 to lug 1, and lug 3 to Q5’s collector. And voila! Now you have an octave knob that goes from no noticeable octave at 0 to full, unadulterated octave at 10.

 

[Ginsly]

@jcpst That’s a VERY helpful article- thank you!

@Chuck D. Bones this is kinda what I envisioned, and the 1M resistor seems quite important- great tip.

@PedalBuilder Bingo- that’s exactly what I’m looking to do, and your pot idea is excellent! Kinda what Orange did with their version of the Foxx Tone Machine. Removing the octave makes for a noisier pedal, though (same with Foxx TM)… I’d love to incorporate a transistor bias as a kind of noise gate, but it sounds like that isn’t so easy on the FY-6…

 

[PedalBuilder]

@PedalBuilder Bingo- that’s exactly what I’m looking to do, and your pot idea is excellent! Kinda what Orange did with their version of the Foxx Tone Machine. Removing the octave makes for a noisier pedal, though (same with Foxx TM)… I’d love to incorporate a transistor bias as a kind of noise gate, but it sounds like that isn’t so easy on the FY-6…

No idea what Orange did, but I got the octave potentiometer idea from Wattson EFY-6. I have it on my Super Fuzz clone, and it does not increase the noise at all. It's a relatively low noise circuit all things considered.

Here's the schematic, in case you're curious. I also changed the bias resistors on Q6 to increase the output level above unity, replaced the tone switch with a 50k linear potentiometer, and added a limiting resistor to the Expander control. It's not too difficult to implement all of these mods on the Uber Fuzz PCB, if that's what you're using.

 

[Ginsly]

No idea what Orange did, but I got the octave potentiometer idea from Wattson EFY-6. I have it on my Super Fuzz clone, and it does not increase the noise at all. It's a relatively low noise circuit all things considered.

Here's the schematic, in case you're curious. I also changed the bias resistors on Q6 to increase the output level above unity, replaced the tone switch with a 50k linear potentiometer, and added a limiting resistor to the Expander control. It's not too difficult to implement all of these mods on the Uber Fuzz PCB, if that's what you're using.

View attachment 77446

Man, was Wattson good stuff… even the basic FY-6 was just perfect.

The changes you made are excellent, I’ll absolutely adopt those mods.

Noise-cancelling aside, I just kinda like mis-biased fuzz! I assume there’s a reason people don’t do that with the FY-6… it must either be very difficult or sound BAD bad as opposed to GOOD bad, as underbiased fuzz can be. Some circuits just don’t want that, seems like. For instance, @bean gave me some hypothetically possibles methods of underbiasing a Fig Buster/Clusterfuzz, but none of them really worked (which he very much warned me about). To that end, I kinda want to add that 8-bit control to every single circuit possible! Sooo, so cool. If only I understood it…

 

[Ginsly]

I really thought I had gotten somewhere with my breadboarding skills... evidently not. I've tried the below modded Superfuzz schematic many times, and signal doesn't seem to make it to the other end of the 47k resistor near the beginning of the circuit (lower left, connecting to a 10n cap to ground). I'm using this schematic simply because I have a pcb of it, but if I can't figure this out I'm jumping ship and breadboarding yours, @PedalBuilder! Ha... I successfully probed a circuit for the first time, at least... My guess is no one will be able to make anything from this mess of spaghetti, but if you have any idea why signal flows on one side of the 47k and not on the other, I've hit a wall here. The red circle is where signal stops. 9v is on the top red rail, and both ground rails are linked.

 

[Chuck D. Bones]

Clearly, none of the transistors are 2N5089 on your breadboard. Make sure that no component leads come in contact with any of the transistor cases. In most metal can transistors, the case connects to the collector internally. With some, the case connects to the base. In any event, Murphy tells us that whatever touches the transistor case, it won't be good.

It appears that R1 & R2 do not connect to Q1-B. If that's true, then the DC voltage on Q1-C will be at or near +9V.

 

[Ginsly]

Clearly, none of the transistors are 2N5089 on your breadboard. Make sure that no component leads come in contact with any of the transistor cases. In most metal can transistors, the case connects to the collector internally. With some, the case connects to the base. In any event, Murphy tells us that whatever touches the transistor case, it won't be good.

It appears that R1 & R2 do not connect to Q1-B. If that's true, then the DC voltage on Q1-C will be at or near +9V.

Thanks, @Chuck D. Bones - I really appreciate you taking a look. I was able to troubleshoot and properly set up a Foxx Tone Machine, but this one has me beat. I'm certainly doing something fundamentally wrong, and I'd really, really love to figure out what that is - not just for this schematic, but for the future.

I used some metal can ECG123A transistors I had handy, as their average hfe (140-170) seemed close enough to what the Superfuzz seems to like. I'm really not sure why this layout uses transistors that are so much higher than that, and I didn't really consider this when I grabbed whatever was close by; the BOM says "pretty much any NPN Si will do", but maybe I need to try something higher. I checked the metal cans, and nothing conductive seems to be touching them.

I had used that top-left orange jumper to connect Q1-B to the second breadboard row, where C1+ and R1 meet. I just tried scooting it over to the third row where R1 and R2 meet - no difference. I also have the feeling that shifting parts of R2, R6, R8, and C4 to the bottom breadboard row has also messed something up...? I can actually hear a tiny bit of the test signal on the other side of R8, but it's very, very quiet.

This is still pretty new to me, so there are some schematic connections that I'm certainly misunderstanding. I guess I'm not always sure when components have to link directly to each other via breadboard pads or when they should/can link through another component. I've gone through some breadboarding walkthroughs here and elsewhere, but I'm still encountering situations where I'm stumped.

 

[Ginsly]

Started from scratch using @PedalBuilder's SF schematic - I actually did it right... I think? Sounds great!

I don't really hear an upper octave, and I made sure to match Q4/Q5, use BAT41s, and leave the Octave knob full up. I also played with the Octave trimmer, and although it doesn't really give more or less octave I can tell it's changing the sound (more/less noise, mostly). Using the Octave on/off pot definitely alters the fuzz still, even without a super-present octave. Very cool.

I'll likely still use the pcb I already have, but I'll certainly add the Octave on/off blend pot. Hopefully I can get the octave kickin a little more!

 

[Brett]

My guess is no one will be able to make anything from this mess of spaghetti

If you have an image editing program, it may help to annotate the images with labels so it’s easier to tell which wire is which. I started to look and immediately went cross eyed

 

[Ginsly]

If you have an image editing program, it may help to annotate the images with labels so it’s easier to tell which wire is which. I started to look and immediately went cross eyed

I hear ya, man! It's one thing to post a pcb, but I'm not sure what I was thinking posting this mess of jumpers and jumbled components...

It may very well be that the first schematic had a typo or something, because when I started over with a new layout it worked!