Shin Ei FY2 Fuzz; or look, I broke something new!

[moid]

Hello everyone (please don't groan; one of these days I might actually learn something and come up with an incredible schematic for you all to build... Guess what? It's not going to be today... and next year doesn't look great either).

Backstory (in which we meet our protagonist and his son and they set out on an epic quest to the lost lands of Fuzz. You can skip the next paragraph if you find the prose too much/ there's better stuff on the television. I won't cry.)

Many years ago, a strange sensation overtook my mind and possessed me with a vengeance, forcing me to learn the dark arts of solder and wires, fumes and metal. I was no longer happy with just buying effects pedals on Ebay and making bad noises with them; I had dreams, visions, compulsions, to create my own effects pedals... and make more bad sounds with them. One of my earliest creations was a Shin Ei FY2 pedal kit from a company. I built it; it worked (well the LED refuses to work, but who cares) and it sounded good but was very quiet. My son really liked it and used it with a boost after it to give it some volume. A year or more later I found a vero layout on Tagboard Effects for an FY2 with volume fixes and built that and added some mods to it that I liked, gave it to my son... and he didn't like it... and I forgot about it. A couple of weeks ago I built a ZVex SHO for another circuit, and while it was hanging around my son plugged the old original FY2 into it and it made a seriously impressive din! He immediately asked me to build another FY2 to the same spec as the first one with a SHO right afterwards to counter the volume loss in the same box. What could be easier I thought?

I have used this vero schematic from Tagboard Effects, which is what the pedal company who sold me the kit used



The only modification I've made is to use different transistors. For Q1 I'm using a 2N4401 and for Q2 I'm using a 2N3904. Both of these are socketed in backwards, as per the instructions on the pedal kit I bought years ago - they really sound quite rippingly evil and that's the sound my son wants. Now first the good news; the pedal works (mostly). Sound goes through, gets distorted, but... not enough - it's a gentle distortion and, after trying to swap the transistors around and substitute new ones / different types I've discovered that whatever is placed in Q1 does very little at all (some transistors have made the audio quieter and less fuzzy). I'm not entirely sure where to start to debug this (usually my pedals just don't work!). I've measured the DC voltage on Q1 and have the following

E 0.625V
B 0.618V
C 0V

Q1 is currently in backwards, which is why the CBE is reversed above.


Q2 Currently has an MPSA18 in it which is the loudest transistor I have (for testing), but I can replace it with the 2n3904 if it helps

C 0V
B 0.54V
E 0.67V

The MPSA18 is socketed the right way round.

At the moment the pedal sounds like a fairly tame fuzz - it's not bad, just not as powerful as my son wants it to be. Any guesses where I should probe next? I have an audio probe so I'm going to try to see if there is a place where the audio dies... but I'm not sure there will be one if the audio leaves the pedal. I have reflowed the solder for the Q1 socket but that hasn't changed anything. Thanks for your thoughts

[Quackzed]

c b e on the stripboard are marked correctly, but your readings look like the trannys are backwards, not sure about using em intentionally that way... i'd FIRST put em in right way per the schem and check voltages... emitters should be 0v not collectors.

[Mark Hammer]

I do not like stripboard....and threads like this are a big reason why.

This is a simple circuit, and a small one.  I've built it at least a half-dozen times.  Do yourself a favour and build it on perfboard , whether bare board or pad-per-hole.  You can lay it out so that it looks almost exactly like the schematic, which makes debugging MUCH easier.

And yes, you can probably use any of a variety of medium-gain NPNs.  Just be mindful that they don't all have the same pinout.

[Plexi]

Great backgound history: similar to my personal one.
All the pedal lover is always seduced to the solder side.. Hahaha

I hate veroboards too... I would say that you can use sockets (if you don't already did it) and try all the comb.
2n3904 and 4401 is almost the same in hfe terms. Not the MPS18: almost twice.
I'll take a look over the bias and gate resistors to know.

To see how far it can be, I would socket some Darlingtons like MPSA13.

Time ago I saw theres a design that uses the FY2 Companion + Big Muff Tone Stack + LPB1 at the end.
I'll check on my docs if I founf/remember

[moid]

Thanks everyone. Tonight I tried switching the transistors around to using 2n3904 and 2n4401 and now nothing works - audio goes through the entire circuit (gets louder) but stays clean. I think I'll return to the MPSA18, although it's weird that Q1 doesn't seem to do anything...

Quackzed - I agree the transistors are the wrong way around, but I've seen Death By Audio and Devi Ever fuzzes that do this and it seems to make the fuzzes extremely distorted.

Mark Hammer - maybe I'll try it on perfboard - I've got a small piece of this, big enough for this circuit anyway... maybe it's time to try it. I found this schematic:



Is this a good one to try? I noticed one difference between it and the vero layout - where (on the schematic) at the top there is a 0.047uF capacitor that goes around a 100K resistor, on the vero layout that 0.047uF cap doesn't seem to do that at all... but lots of people on Tagboard Effects have made that vero layout work? Weird.

Plexi - I don't have any MPSA13 - if they are important I'll have to wait until the end of the month when I get paid before I can buy any. Don't worry about looking for a modified version of the circuit thanks - my son only really wants the sound he has in the pedal I already built a few years ago. Hmmm if the worst comes to the worst I could hack the working old circuit out of the original pedal I made and solder it into this new one... feels like cheating though. Maybe I'll try perfboard first and then if that goes nowhere I'll cheat!

[Mark Hammer]

No need for MPSA13.  This is a pretty serious fuzz with only modest-gain transistors.

Note that the midscoop filter, formed by the 10k/15k pair with the 100nf to ground and the 1nf bypass cap, eats up signal  with passive loss.  If you leave out the scoop filter, and go directly from lug 1 of the fuzz pot to the volume control, you keep more signal and get a hotter output.  And the fuzz tone is nice and throaty; more in the Fuzz-Rite zone.  And without the added boost stage and the scoop filter, it's a simpler and easier build - something you could throw together in an hour.

[Quackzed]

as for q1 not doing much, make sure you have that jumper from q1's collector strip to the strip between +9v and input (leftmost black line on the layout)if so check the value?

[moid]

Thanks chaps. It's been a crap day for pedal building - I spent most of it either helping my son with a science experiment to measure water purity in local rivers (actually that was quite fun apart from discovering that one has completely dried up and another is looking very shallow) and the rest fixing the RAID drive on my PC (I think, finally now fixed, fingers crossed) and then fixing my wife's PC's graphics card, whose fan jammed from dust and then kept turning itself off and needed to be taken apart to be cleaned. So I had a look at the vero board, checked what Quackzed suggested (yes, I have that correct) and then thought about perf board. Then I thought, nah it's late, let's breadboard this first... and an hour or so later and that version doesn't work either - it passes clean signal happily, not a trace of fuzz noise though. I hope I'll get a chance to look at this during the day tomorrow. What a despairing day; I've built far more complex things than this that have worked first time... bugger

[moid]

Some good news - I got the breadboard version of the circuit working. There was nothing wrong with the way I'd assembled it, the damn power supply plug has died! I finally remembered to check for power on the circuit and got nothing, all the way to the plug. Switched to a battery or different plug and the circuit makes noise! I am now wondering if my vero circuit was working after all but the power going to it was fading away and that's why it started fuzzy and became cleaner and quieter over time? Time to try to rebuild it!

[Mark Hammer]

Always something, innit?

Many beginners will ask for troubleshooting assistance here, thinking that the root of their pedal's nonfunctionality will be exotic and well beyond the perimeter of their expertise.  And as often as not, it frequently turns out to be something banal, such as you describe.

A hearty welcome to the facepalm club.    It's a fun fuzz.  I bought mine in 1992 at a pawn shop for $25 or so.  At the time, a DPDT stompswitch - when you could find them - was $20 on its own.  So I wasn't particular interested in the fuzz itself, and cannibalized the chassis, switch, pots, and jacks for something else I was building.  Silly me.  I did save the board, however, and a decade or so later, asked around for some info on how to reconnect the pots.  I wired it up as original and loved the sound.  I've since modded it a dozen different ways, nearly all of them delightful.  It's a very generous circuit.

Note that the function of the 47nf cap in parallel with the 100k resistor from V+ is to lower treble.  Though the effect is not striking, one can replace that 100k fixed resistor with a 100k trimmer, using the outer lugs of the trimmer as if it was a fixed 100k resistor.  Connect the 47nf cap between the trimmer's wiper and the junction between trimmer and 47k fixed resistor to fine tune the tone.

[moid]

Thanks Mark. I'm a fully paid up member of Club Palm de Face. It's quite selective you know - they don't let just any old idiot with a soldering iron in. You have to really show you've gto what it takes to make that slap sound happen

I had no idea that stomp switches were so hard to get in the past; but it looks like all the 1980s/ 1990s pedals I have use custom switches (they're all Boss or DOD or Ibanez). I guess 2PDT or 3PDT switches are a 2000s sort of thing?

I have implemented your first suggestion about taking the audio from lug 1 of the fuzz pot to the volume out and that sounds great and I can easily add that with a switch to take the effect in and out of the circuit (or remove the tone control depending on how you look at it).

While playing with the breadboard circuit (dangerously addictive, this swapping components game) I managed to make a wondrous arpeggiating synth 8bit type fuzz by taking the lug2 of the Fuzz pot and attaching that to the circuit where the audio starts. This sounds great except for the rapid high pitch descending laser sound that is constantly playing in the background. I tried adding a voltage divider circuit between lug 2 of the fuzz and the connection to the audio in of the circuit, and this has controlled the laser sound a fair bit, but it still triggers randomly when I don't want it... The voltage divider is just a 10K resistor, then another 10K resistor to ground, and between them is a cable to connect to the audio in of the circuit. I tried a lot of different resistor types (keeping the first resistor at 10K) and I could either remove the effect but also the synth sound (and most of the audio) or do odd things like remove all the bass or the treble, or speed up or slow down the laser sound, but nothing got rid of it without losing the synth sounds. I also tried adding two diodes in anti parallel setup before the voltage divider (I read about using two diodes in the Boss HM2 as a crude noise gate, but they were germanium diodes that i don't have) but none of my diodes worked - they usually removed most of the signal, or left just the most broken noise.

What I'm trying to say is do you have any ideas if there is a way to keep the synthy sounds in the pedal without having the feedback laser sound? If not, I'll remove that section because it'll only annoy me if I build it into the final pedal and I can't shut it up. Thanks.

[moid]

Don't worry about the synth sounds - my son decided he didn't like them so they're no longer in the circuit now. We found a couple of other points on the circuit that make sonic variations to the main sound so I'll redraw a schematic and try to build from there.

[moid]

OK my latest disaster. I tried to build this circuit on perfboard (never did this before). Earlier on today the audio signal got al the way to lug 3 of the volume pot, but refused to go out of lug 2. I assumed the pot had an issue so swapped it for another 50K Lin pot. Now the audio stops at the first transistor (an MPSA18) on the base - I can hear music hear with my audio probe but nothing on the collector where I would expect to hear louder music.

I measured the voltage of the transistor in AC and got

Collector 0.454V
Base 0.74V
Emitter 0.11V

I was expecting the Collector to be a higher voltage than the Base, but maybe I've got that wrong? Once everyone here has gone to bed I'll start measuring the other transistor. I cannot believe how such a simple circuit can be so hard to build. Thanks for any thoughts / ideas you might have.

The transistors are socketed so  I can swap them around and I did try reversing them in case I'd got the pinouts wrong but that didn't help, and I'm sure I've got the pinouts right.

[moid]

Today's batch of misery looks like this... well, on the good side reflowing solder all over the circuit now means I have audio signal going through the circuit Yay!

Bad news - the circuit does not distort at all. I appear to have made an extremely transparent overdrive. The only part of the circuit that does anything is the switch 1 I added to take the original midscoop filter in and out of the circuit  - that does indeed work, oh and the volume pot does make the audio quieter if I turn it down but there's no fuzz at all. I've added an updated schematic below in case it shows anything I'm missing. If someone could look over this one and offer any ideas about what to check next I'd be very grateful. Thanks!

[Cozybuilder]

The 50K trimmer you added has effectively created a voltage divider with the Q2 collector resistor (47K to V+, trimmer R to ground), so the bias on that is way off now. Try adding a cap before the trimmer pot.

[duck_arse]

if you move that trimpot, and wire as a variable resistor (two lug wired) between your SW1 "3" lug and the volume pot "3" lug, you will effectively have an extra stopper resistor in series w/ the volume.

otherwise - which direction are the transistors facing today, what are the voltages, and where are the photos?

[Cozybuilder]

if you move that trimpot, and wire as a variable resistor (two lug wired) between your SW1 "3" lug and the volume pot "3" lug, you will effectively have an extra stopper resistor in series w/ the volume.

otherwise - which direction are the transistors facing today, what are the voltages, and where are the photos?

Or just cut the ground on the trimmer?  Duck is right

[duck_arse]

there's that ^, and I've just realised my mod would include the fuzz pot. you'd be mad to leave that out, tho, missing all that mixing madness. where did you say the pics were?

[moid]

Thanks chaps! So my circuit might actually be working if I hadn't added the trimmer? trust me to sabotage my own build!

Here are some fairly x-rated images of my build, apologies for the graphic imagery you are about to experience. I just noticed the burned yellow cable in one shot, I bet that's not a good thing There is an extra foot switch and pot which are for a booster circuit (not pictured) so ignore those. I haven't got round to drilling holes for the LEDs yet (I got the enclosure by mistake from a company that I ordered parts from years ago and it was pre drilled and I've never had a use for something with two foot switches until now - they let me keep it so I'm glad to finally have a use for it, but it doesn't have LED holes!)

The underside is probably a disaster solder wise - I've never worked with perfboard and it's not easy to play with. I've added a few notes on one image in case that helps; let me know if you need more.







Duck, I am not entirely sure what you mean when you are talking about moving that trimmer, but I'll try to get sometime to re read it tomorrow during the day (have been entertaining three boys aged 8 - 11 today for most of the day and I'm knackered... kids have way too much energy in them). I'll probably draw a schematic.

transistors are facing the correct way around - I decided to go along with general convention for a change; I'm getting to old to rebel against everything. I'll try to get some measurements for you if I can keep my eyes open!

[Mark Hammer]

If you want to be able to switch from midscoop to no midscoop, without suffering a volume mismatch, just make the no-scoop alternate path a single fixed resistor in series with the input to the volume pot.  I find that something in the vicinity of 33-47k does the trick nicely.

The series resistor on the pot input simply makes the volume pot behave as if it is a higher-value pot (e.g., 89k if 39k is added to 50k) that can only be turned up part way.