Build Your Own Clone Message Board

I finally got my Leeds Fuzz that I bought from a member here working...
I set the trimpot "correctly" at the noisiest position, but everything sounds a little under biased. Note definition for chords is a little unclear and the octave effect is not nearly as prominent as in demos I've seen
I'll measure voltages, tell me what you think:
Sound demo: http://soundcloud.com/jeff-n/short-superfuzz-demo

_________________
Website    Blog

Can't hear your demo since they block that stuff here where I work, but a comment re: the octave effect--

The effect is much more evident under some playing conditions than others--this is a "nature of the beast" thing with these octave-up producing circuits in general, not just the Leeds Fuzz. The following should give you pretty much the strongest octave output: Select your neck pickup, roll most of the treble off with your tone control, and play on your top three strings (G-B-E, assuming standard tuning) up around the 9th to 15th frets.

_________________
“My favorite programming language is SOLDER” - Bob Pease (RIP)

My Website * My Musical Gear * My DIY Pedals: Pg.1 - Pg.2

I'd agree that your demo does not sound good compared to other demos of the Leeds that are out there. It is certainly not a sound I would want to hear coming out of a pedal I owned. I'm not surprised the previous owner decided to sell it.

I'd suggest you measure voltages at collector Q1, emitter Q2, collector and emitter of Q3, base Q4?, base Q5?, collectors of Q4? & Q5? and at the collector of Q6? (probable numbers are assumed for Qn? transistors not identified on the circuit diagram).

Also check your battery (or ext power supply if you are using one) is at around 9V and that the full 9V reaches + supply track on the PCB.

I'm using a 1 spot and I'm getting 9.44 v on +, R5, R10, R16, etc

I got Q placement from here http://pic20.picturetrail.com/VOL83/511 ... 268657.jpg

C of Q1: 5.57
E of Q2: 5.01
C of Q3: 6.28
E of Q3: 3.10
B of Q4: 0.85
B of Q5: 0.00 ??
C of Q4 and Q5: 7.84 and 7.84
C of Q6: 3.50

_________________
Website    Blog

The only reading that looks obviously wrong is 0v for the base of Q5. It should be the same as Q4 with a small difference introduced as you twiddle the trimmer.

Either Q5 is blown or there is a short or open circuit near that transistor.

I suggest you check that reading again and if you measured with the probe directly on the transistor base lead or the pad it is soldered to, check the voltage at the junction of R17 100K and R19 22K perhaps by measuring form the top of the PCB where the resistor wire bends down to go through the hole. The end of either resistor will do as long as it is the end that connects to the other resistor. If that also gives you 0V and you cannot see any obvious shorts or bad joints in that area I'd say try replacing Q5.

Use a bright light and a magnifying glass to hunt for shorts and bad joints.

Since getting a replacement 2SC828 might be a little tricky you can use almost any small signal silicon NPN transistor as long as it either has the same pin-out or you bend the leads to the right order. I suppose, in the interests of balance, once you fit 'a transistor' and if it shows a sensible voltage on the base you might want to also replace Q4 with the same type.

** CORRECTED **

For reference, here are all of the transistor voltages on my fully functional Leeds Fuzz:

DC source voltage = 9.66V

Q1:
B: 0.71
C: 5.57
E: 0.15

Q2:
B: 5.57
C: 9.66
E: 4.96

Q3:
B: 3.69
C: 6.55
E: 3.12

Q4:
B: 1.99
C: 1.59
E: 1.46

Q5:
B: 2.08
C: 1.59
E: 1.46

Q6:
B: 1.23
C: 3.52
E: 0.61

All your reported values look good except for the base & collector values for Q4 & Q5. Seems odd that both transistors would be similarly off. Check R21, which is common to the voltage biasing for both those transistors. Should be a 10K resistor; make sure both ends have a good solder joint. Same with R20, which is a 1K8.

_________________
“My favorite programming language is SOLDER” - Bob Pease (RIP)

My Website * My Musical Gear * My DIY Pedals: Pg.1 - Pg.2

I'm getting 0V on both R17 and R19 in addition to the base of Q5. I actually have some spare 2SC828 bc this build is replacing a faulty Ibanez Standard fuzz vero build. So I'll try replacing it and see if that works. Any recommended Hfe's?

_________________
Website    Blog

I've got 2.08V on the ends of R17 and R19 that connect to the base of Q5, as expected. And I see 1.99V on the ends of R16 and R18 that connect to the base of Q4, also as expected. Sounds like maybe you've got a short to ground in there somewhere.

Also, please check the additional stuff I added to my preceding post.

_________________
“My favorite programming language is SOLDER” - Bob Pease (RIP)

My Website * My Musical Gear * My DIY Pedals: Pg.1 - Pg.2

I've got Q5 out and I'm about to replace it with a new one. R20 on my board is 1k8. Yes...there seems to be a short to ground: Base of Q5 is grounded...

_________________
Website    Blog

ok so it was only grounded when the trimmer was turned...it's a trimmer like this. Is that at 12 noon in the picture?

_________________
Website    Blog

Oops, sorry. Yes R20 is 1K8--I mis-typed. Went back and corrected that.

_________________
“My favorite programming language is SOLDER” - Bob Pease (RIP)

My Website * My Musical Gear * My DIY Pedals: Pg.1 - Pg.2

With respect DVM - I think you may have your transistor pin outs / voltages round the wrong way. Should not Q1 for example be C 5.57, B 0.71, E 0.15. Not B, C, E as you have it? (and Q2 ... and so on)


As far as Hfe goes for Q4 & Q5, I guess you want them to be reasonably matched, although I suppose the trimmer balances them.

That trimmer should not ground the base - there should be 22K R19 in the way.

You are right, of course--I forgot that mine is built on a beta board that was originally fabbed for 2N5088's, which have a different pinout than the 2SC828's! I just wrote down the data as I took it, with the tranny legs labeled as per the PCB diagram in the instructions. I'll go back and correct those labels....

Yup, that trimmer can't ground the Q5 base unless something else is wrong. As far as the trimmer setting goes, just set it as close to the middle of its sweep as possible--that'll be fine for now. You can optimize it once the voltage issue is resolved.

_________________
“My favorite programming language is SOLDER” - Bob Pease (RIP)

My Website * My Musical Gear * My DIY Pedals: Pg.1 - Pg.2

I replaced Q5, did a few 180's with the trimpot and this is what I'm getting

Q1:
B: 0.71
C: 5.58
E: 0.15

Q2:
B: 5.60
C: 9.43
E: 5.01

Q3:
B: 3.43
C: 6.28
E: 3.10

Q4:
B: 2.29
C: 1.68
E: 1.66

Q5:
B: 2.30
C: 1.68
E: 1.66

Q6:
B: 1.20
C: 3.53
E: 0.59

lookin' good?

_________________
Website    Blog

Yup, much better! How does it sound now?

_________________
“My favorite programming language is SOLDER” - Bob Pease (RIP)

My Website * My Musical Gear * My DIY Pedals: Pg.1 - Pg.2

it sounded pretty much the same but then the LED stopped working...I'll just revisit it tomorrow

_________________
Website    Blog

Well that base voltage seems a bit healthier but I'm a little surprised that the joint collector voltage for Q4 and Q5 is so low. 1.68 volts doesn't leave much room for the signal to swing negative on those collectors. On the other hand that is pretty much the same voltage that DVM says he gets on his working pedal.

Also, out of curiosity, what are you running the pedal into to listen and to make that recording? It will of course sound very different running in to a decent guitar amp than it will into, for example, a computer sound card input. Also note that, from what I can understand from the circuit, the Leeds Fuzz like most fuzz pedals, is designed to be first in the effects chain with its input fed directly from the guitar out. With the guitar plugged directly into it the input stage will have less gain, because the guitar electronics become part of the feedback circuit around those first two transistors, and the pedal will tend to clean up as you roll down on the guitar volume.

Damn, I really thought you were gonna be all set with those new voltages you posted.

It may be time to get yourself a signal tester so that you can follow the audio signal through the LF circuit and pinpoint where it turns nasty. Or you can build one, if you have the parts--they're very simple. Directions HERE.

I think it would also be worthwhile for you to post some good close-up photos of the build, including both sides of the PCB. We might be able to pick up a problem that you've overlooked.

_________________
“My favorite programming language is SOLDER” - Bob Pease (RIP)

My Website * My Musical Gear * My DIY Pedals: Pg.1 - Pg.2

I dunno DVM I think now that the voltages are where they ought to be there is a strong possibility that the pedal is working as it should. At least the octave component should be more obvious in the signal. I think it would be interesting to get jtn191's feedback on how he recorded that sound (see my previous post). GearManDude has a pretty good demo of the pedal on Utoob, but he is running it as the only pedal in the chain straight in to a really good tube amp. Having the diode stop working is a bit odd and I do wonder about the apparent short on the base of Q5. Was there a circuit short or was Q5 dead?

BTW - I have now run a SPICE simulation of the octave portion of the Leeds Fuzz and the voltages pretty much check out, it seems the original designers intentionally hard biased that stage so it would clip asymmetrically and the stage also has a lot of gain. It is an interesting circuit as octave doublers go. It is possible to re-bias it for a less aggressive sound by changing R18, 19 22K to 12K which might be an interesting experiment.

On its own, with the transistors re-biased and with the gain reduced, it might make a very cool octave up circuit with perhaps a smoother sound than something diode based like the green ringer.

that recording and other tests were through my Blues Jr+SM57. I also did some quick late night tests through my Vox pathfinder. I know both amps well enough to know that it's not the amp. I'll do an audio test today and see if it's some other component.

_________________
Website    Blog

All I can tell you is that I have never run my Leeds Fuzz anywhere but in the buffered noise reduction loop of my Boss NS-2 Noise Suppressor, and I can assure you that it sounds nothing like that audio clip. I know what you mean about how fuzz pedals often react negatively to being anything but right after the guitar in the signal chain (my two Fuzz Face clones sound like sh*t anywhere but first), but that doesn't seem to be a problem for the Leeds. If it still sounds like that original clip even with the correct voltage profiles through the 6 trannies, I suspect something else is still wrong in the circuit.

Incidentally, I find the octave effect in the Leeds to be a bit more subtle than what you hear in a circuit like the Octave Fuzz (Octavia clone) or the Scrambled Octave (Ampeg Scrambler clone). But it's still very evident when the guitar is set up & played as I described HERE.

_________________
“My favorite programming language is SOLDER” - Bob Pease (RIP)

My Website * My Musical Gear * My DIY Pedals: Pg.1 - Pg.2

I'd certainly agree that IF it still sounds like that clip even after fixing the problem with Q5 then something is still wrong. I just wanted to be sure the problem lies in the pedal and not elsewhere.

My comments about the octave doubler were about just that section of Q3,4 & 5. I wouldn't expect the pedal as whole to produce a clean and prominent octave.
I think that in isolation that section of the circuit might be interesting to experiment with, and within the context of the pedal, re-biasing that section could provide a useful alternate tone.

The tone coming direct out of the octave section is likely to be pretty raucous, I'm wondering if the diodes are properly back to back and both working. Otherwise after the octave up section there is not a lot left.

I have now extended my SPICE simulation so it models the entire pedal and I notice that the tone switch should make quite a big difference in tone. Flipped one way the output is almost pure square waves with some doubled pulses as a result of the octave up. This should sound quite aggressive without any extra filtering / tone controls. With the switch flipped the other way the output is a triangle wave (so the square wave get quite heavily filtered) with some fast spikes riding on it. This should sound more violin like, with maybe some upper fizz.

The first stage of Q1 and Q2 before the Expander pot is a clean gain stage with a basic gain of around 7 and some mid to treble boost. Q3,4 & 5 are the octave doubler and combined with D1 and D2 they also produce a clipped somewhat asymmetric square wave. The two diodes are followed either by a level cut or by a passive filter depending on the position of the tone switch and Q6 and associated bits are a clean output stage with a gain of about 10.