Build Your Own Clone Message Board

Hello, everyone. A friend of mine started working on a Leeds Fuzz a few years ago but eventually put it aside when he couldn't get it to work. He passed it off to me a few weeks ago and I've been tinkering with it. I've got bypass, LED, and distinct effect. The problem is that the signal breaks up and distorts--and not in a good way. I re-wired the whole thing because some of those connections looked a little suspect and touched up the soldering in some questionable places as well. I've built about 12 pedals over the past couple years and I feel good about my own skills, but this has me stumped. Am I overlooking something obvious?

One other thing to note: I'm a bassist so I'm playing this through a bass amp. The signal distortion is most noticeable on the G string or when strumming chords. I doubt the instrument/amp matters, but thought I would mention it regardless.

Some pictures:

Check all of your values. A stray resistor could cause that (it's happened to me before).

Will do. I'll take a look at that tonight.

That trimpot setting doesn't look right--looks all the way over to the CW end of its sweep, which can give rise to some nasty, splatty-sounding gating behavior. Try adjusting that per pg. 19 of the instructions. Note this statement (bold italics mine): "Ideally you want to set this trimpot so that it produces the most noise. This may seem counterintuitive, but this will produce the longest sustain and the best sounding fuzz. This will almost always be dead center on the trimpot's rotation (noon) or just slightly off center."

I've played around with that trim pot a bit and will look at it again.

dm, I should need to turn it just a bit more, to set it at center or off-center, correct? In other words, the notch is what I should be lining up at 12:00? I want to make sure I'm looking at it correctly--this is my first experience with internal trim pots.

"12 noon" in this case is more figurative than literal. What it really means is midway (50%) in the trimpot's sweep arc. External control pots typically have a 300 degree sweep and are commonly set up with the knob marker oriented so that full off is at 7 o'clock and full on is at 5 o'clock. Thus "12 noon"/straight up is the halfway point. However in this case, the trimpot is mounted on the PCB pointing down toward the footswitch end of the enclosure. So "12 noon" will have that flat spot of the trimmer facing straight down.

No luck with the trim pot. I adjusted it but it didn't make any difference--the signal still distorts after the initial attack. I'm going to go through and double-check the resistors again; the volume doesn't seem to be as loud as it should be, which makes me think it's something on the board....

I just double-checked all the resistors and they're in the right places. Any suggestions?

Do you have a multimeter available for checking continuity, voltages, etc?

I've got one, but I'm embarrassed to say I'm not skilled with it. Can you recommend specific tests I should try? Or even point me to a tutorial?

Excellent multimeter tutorial avilable at http://www.ladyada.net/learn/multimeter/

I think the first thing I'd be looking at would be the transistor voltages. Read through the tutorial first and then we'll take it from there.

Thanks. I'll work on that tonight. I appreciate your patience and help with this.

I've taken the transistor readings, but to double-check myself I tested a 9V battery as well. With the DC Volts set at 20 on my multimeter (and following the tutorial), I get a reading of 1810.....surely that can't be right?

What brand/model of multimeter?

It's an Ideal 61-310:

Looks pretty straightforward--red probe in the right-hand jack, black in the COM (middle), and the meter dial turned to 20 on the DCV scale. If it's giving crazy numbers, the meter prolly needs a new 9V battery. The manual is available online here: http://www.idealindustries.com/media/pd ... ual_v2.pdf

That did the trick. I swapped the battery and it seems much better now. (Wow do I feel dumb)

Here's what I'm getting from the transistors:

The "top" transistor (the one closest to the jacks):
E: 2.46
C: 6.73
B: 3.06

The "middle three" (right to left, starting with the one closest to the internal trimpot)
E: 1.60
C: 1.63
B: 2.23

E: 1.61
C: 1.63
B: 2.23

E: 4.86
C: 9.21
B: 5.45

The next one, which is to the right of the 47K resistor:
E: .70
C: 5.50
B: .13

And the bottom one, to the right of the footswitch and above the 10k:
E: .55
C: 3.66
B: 1.17

Some info in case you didn't know it - most commonly transistors are biased ON. By that I mean that for NPN silicon transistors you will find the base is always about 0.6 volts positive of the emitter. For PNP silicon transistors the emitter is always about 0.6 volts positive of the base.

All the transistors in the Leeds Fuzz are 2SC828, 2SC539 or 2SC356 (one of those types - they should all be the same). These are NPN so the bases should all be positive of the emitters. Your penultimate readings are E 0.7 C 5.5 B 0.13. Which would make the base negative of the emitter. Did you mean B 1.3 perhaps.

The transistor with 9.21 on its collector must be Q2, the two with almost the same voltages are - lets call them Q4, Q5 because although the circuit diagram does not number them that is they would be is you followed the number scheme from left to right.
So the order of your list of readings is - Q3, Q4, Q5, Q2, Q6, Q1. Apart from Q6 which I think is a measurement mistake they all look about right to me.

Here's the Leeds Fuzz board with the components labeled, including the 6 trannies:



So the order you've listed the voltage readings in is Q3, Q5, Q4, Q2, Q1, Q6. The mistake on your Q1 readings is that you switched the base and emitter values that you reported. Adjusting for that, your readings all correspond very closely to my own Leeds Fuzz.

Given that these transistor voltages appear normal, your best bet at this point would probably be to trace the signal through the circuit with a signal tester and determine where it "goes bad". This will help pinpoint the portion of the circuit where the problem is occurring.

Thanks DVM and Tark for your help. DVM, you're correct on your assessment of how I measured the trannys--that's the order in which I measured them.

I'll get or build an audio probe very soon, as I don't have one at the moment. I do have access to a scope, but it's not here at my house. There's also that whole learning curve-thingy to account for as well. But I will get to this, just not as soon as I have with the measurements.

One question...and this is largely for my own education: How do you guys know this stuff? I'm not being sarcastic...I'm actually really curious. How do you know those are the right transistor readings? Is that something you pick up after doing this stuff for a long time or is it knowledge pulled from another source? I've never done troubleshooting to this depth, so I'm trying to gather all the information I can.

We know through many years of reading and experience of trying to understand, fix, design and build stuff.
You can go nuts with lots of heavy math for electronics design if you want, but there are lots of rules of thumb that work pretty well most of the time.
Fortunately bi-polar junction transistors are pretty simple three terminal current operated devices (if you ignore what happens at an atomic level) and ohms law isn't very complicated

Looking forward to the results of that probe testing. Best of luck!

I hope to have it soon. I'm starting a drive across the country today and won't be back until Monday. I hope to have the scope by that time.

The audio probe has been ordered and I think I'll have it next week. I've never used one before, so could someone give me some pointers? How exactly does one trace the circuit? Is it just a matter of following the schematic?

Yeah, that's the basic idea--follow the signal through the effect circuit and look for where it drops out or changes in a bad way. With the Leeds Fuzz, which is basically a series of transistors stages, the key places to test are the base (B) and collector (C) legs of each tranny. AAMOF, while helping another builder here a while back, I sought some guidance from Keith "Mr. BYOC" Vonderhulls about troubleshooting a Leeds Fuzz signal problem. Here is his response to me, verbatim:

"For the Leeds Fuzz, since signal goes into the base of every transistor, the easiest way is to just test at the "B" of every transistor in order. If you test at the B of Q3 and get no signal, but you had signal at the B of Q1 and Q2, then we can assume the problem is with Q2 or one of the components supporting Q2.

At the B of Q1, your signal should sound just like clean guitar. Q1 is a gain stage, so you should notice a big volume boost at the B of Q2. Q2 is just a buffer stage, so the signal should sound the same at the B of Q3. Q3 is a buffer stage as well, but it splits the signal in to inverting and non-inverting phases. The inverted phase goes to the B of Q5 and the non-inverted goes to the B of Q4 which make up the octaver part of the circuit. The inverted signal at the B of Q5 should still sound the same as the signal at the B of Q2 and Q3. The non-inverted signal at the B of Q4 should be about the same volume as before, but sound crackly.

Q6 is another boost stage and it's just there to compensate for the volume loss caused by the clipping stage and tone stack. So the signal at the B of Q6 should sound exactly like what you expect to hear coming out of your amp, only not as loud.

Keep in mind that if the signal at the B of Q6 doesn't sound right, it doesn't necessarily mean Q4 or Q5 is to blame, because there are still the clipping diodes and tone stack that could be affecting it. So if you do find a problem at the B of Q6, back track to the C of either Q4 or Q5 (They are connected so either one works just as well as the other). The signal should sound loud, crackly, and have an octave. Then test the junction where C10, C11, D1 and D2 meet. This is where the clipping occurs. The distortion should sound much more smoothed out here and loose a little volume. Then test the signal at the center terminal of the toggle switch. You should hear the midrange get scooped out when you flick the toggle switch back and forth."