Uni-Fuzz Repair

[FredB]

I have a Uni-Fuzz in for reapir. Basically it is a Shin Ei Super Fuzz with an AC power supply.

There are two problems with it. First this version uses a relay to switch between by pass and effect, The contacts on the relay for by pass have corroded. I tried deoxIT alhought the contacts seem too far gone for even DoxIT to halp. So a replacement is will be used.

The second problem, is that I noticed a some 120Hz buzzing in the pedal. Since I have built an ESR meter I checked the main filter caps after the rectifiers. They test at about 0.25 ohms. I believe they are 220u caps.

When I had the board out to try and clean the relay contacts, I tested the electrolytics on the main board. All of the 10u caps on the board tested between 7 and 12 ohms. In contrast some new Nichicon 10u caps I have test at 0.5 ohms.

So, the question is, should I recommend to the owner a recap job. The pedal sounds OK for a Super Fuzz, better then some I've heard, although there is the 120Hz buzzing. In the Super Fuzz circuit there is only one power rail filter cap. I could replace just that one and see if it helps get rid of the buzz.

The pedal runs off a two conductor cable. Maybe upgrading to a three conduction cable could help. When I hook up my scope to the pedal, and put the ground clip of the probe on the signal ground there is an absence of any 120Hz noise on the power rail. If I put the probe on the metal chassis, then some 120Hz noise shows up. So, I am wondering if the noise is a grounding issue rather then a filtering issue.

Anybody else have any experience with these Uni-Fuzz pedals?

[Enzo]

Never seen one, but circuits are circuits, they all work the same.

120Hz hum is power supply ripple. It MAY be from lack of filtration, but many many times it is signal grounds sharing power supply grounds. the return current for the supply gets into the signal circuits.

Before you wholesale replace a lot of caps, just tack a largish one in parallel with the existing filter caps. If that makes an improvement, then your caps are either worn out or were just not large enough to begin with. (very possible)

How is your signal ground connected to the chassis ground? It seems they are not bonded together listening to your scope experience.

How about a schematic of what you built? How does the negative side - assuming that is the grounded side - of your powr supply get to ground?

[FredB]

Never seen one, but circuits are circuits, they all work the same.

120Hz hum is power supply ripple. It MAY be from lack of filtration, but many many times it is signal grounds sharing power supply grounds. the return current for the supply gets into the signal circuits.

Before you wholesale replace a lot of caps, just tack a largish one in parallel with the existing filter caps. If that makes an improvement, then your caps are either worn out or were just not large enough to begin with. (very possible)

That exactly what I was going to try first. Although I checked the filter caps and as far as I can tell they are good.

How is your signal ground connected to the chassis ground? It seems they are not bonded together listening to your scope experience.

How about a schematic of what you built? How does the negative side - assuming that is the grounded side - of your powr supply get to ground?

Here is a schematic of the pedal:



I think there is an err in the schematic about the wiring of the primary, although that is out of concern since what ever wiring the pedal has on the primary it works fine.

As the schematic indicate power is supplied by a two conductor cord. Secondary center tap is both power ground and signal ground. The signal ground is connected to the chassis at the input and output jacks.

I have to get it back on the bench and try adding some more caps to the filters and see what happens. From studying the schematic, I doubt replacing the coupling caps will halp much with the buzz. I've recapped Super Fuzzes before and it results in only a slight change in tone.

[Enzo]

Yes, the primary is mis-drawn. If it works, it works.

The power supply board shows ground connections for the parts, but no idea where those connect to the rest of the system.

[FredB]

Yes, the primary is mis-drawn. If it works, it works.

The power supply board shows ground connections for the parts, but no idea where those connect to the rest of the system.

I did some more checking last night. Here are some pictures for a reference:







In the image of the solder side the large trace in the middle is the ground rail. The center tap of the transformer is connected on one side, the left, and the right side, is connected to the ground bus wire that grounds all jacks and pots on the top of the chassis.

The noise I have determined is mostly 60Hz with some higher frequency ripple,and faint switching spikes from the rectifier diodes. It show up on the output of the pedal. I still had yet to check how far back into the signal path it goes.

The thin trace on the bottom of the solder side image is the +9V power rail. Between the power rail and the ground rail, there is an absence of any noise. There is significantly less then 1mV ripple there.

So it seems somehow the noise is getting into the signal path and getting amplified by one or more stages. I had the bottom plate off the chassis, so the ground connection was open to it, when I noticed noise on that with the scope before. I think if it has a connection to the other half of the chassis the noise would disappear there. I'll have to check that too.

[52 Bill]

I'd suggest that you try and clean up the ground wiring at the inputs and pots. Univibes are wired with a ground buss wire just like what you have and it creates a horrible ground loop buzz.

I would try single point grounding back to the pc board.

[FredB]

I sent the Uni-Fuzz back to its owner without any modifications to the the wiring, because he requested that I keep the unit as stock as possible. Since then he has returned the unit to me to reduce the buzz. Apparently he thought the noise was the result of the bad relay or a bad foot switch. The foot switch is good.

So, I will try star grounding the front panel controls and jacks. I was also wondering what you thought of perhaps using insulated jacks. The original ones are non insulated. I see what kind of improvement the star grounding makes first.

[52 Bill]

So, I will try star grounding the front panel controls and jacks. I was also wondering what you thought of perhaps using insulated jacks. The original ones are non insulated. I see what kind of improvement the star grounding makes first.

I don't think that you'll need insulated jacks, just the rewiring.

[J M Fahey]

And 200uF is way too low.
Maybe price was an issue in the late 60's, but not today.
I'd use at least 1000uF x 16V ; 2200uF even better.
Size probably will not be a problem, elcos have shrunk a lot in 40 years.

[FredB]

As far as I can tell the after the second filter cap there is an absence of any noticable 120Hz ripple. The drain is only about 10ma though the circuit. I will try larger caps.

As far as I can tell the noise is coming from either or both of two sources other than power supply ripple. They are as you mentioned ground loop and rfi from the recifier diodes.

The noise rather than on the power rails seems to be getting into the input of some of the amplifier stages and then gets amplified to the output of those stages.

I tried star grounding will little reduction in the noise. Next will be larger filter caps.

[J M Fahey]

1) what happens if you set the drive/distortion control to 0 ?
2) what if you set it to a reasonable distortion value, but insert a shorted plug at the input?
3) I'm amazed at how the copper tracks on a 40 y.o. phenolic PCB, no solder masks or tinning of any kind, are still shiny and solderable.
Solder also looks shiny clean.
These Japanese must have used one h*ck of clear coat protective varnish or something. Cool.

[FredB]

The unit has sat in a box for those forty years. It's unused. That is probably so, because it buzzed so much to begin with. After I replaced the relay, I mentioned to the owner that the buzz was quite noticable. He requested that I send it back to him without addressing the buzz. When he got it, he immediately returned it to me again, saying the the buzz made the unit unusable. So, I'm thinking that's why it never go used since it was built.

There is a kind of varnish over the copper traces. I messed with star grounding a few different ways, although the buzz is still quite noticable. Also those old copper traces are thin and tend to separate easily from the phenolic. So, I want to minimize the resoldering, I've already messed with it more than I would have liked.

The fuzz control basically has zero effect on the buzz. The buzz remains the same regardless of the position of that control. The output level control, just makes the buzz louder or quieter. The buzz is also basically the same regardless if the input is shorted or a cord is plugged in either with or without a guitar.

I just realized there is a simple solution to this problem, just convert the unit to use a standard AC adapter. Even the old ground buss wire could be left as is in that case. Although in this case, I have already removed it.

[J M Fahey]

Love those inconstant customers
Rewiring for external PSU is the best bet.
But remember the relay coil probably "eats" more current than the actual circuit, and it would benefit adding an anti parallel diode across its coil, to lower pop.

[J M Fahey]

Love those inconstant customers
Rewiring for external PSU is the best bet.
But remember the relay coil probably "eats" more current than the actual circuit, and it would benefit adding an anti parallel diode across its coil, to lower pop.

[J M Fahey]

Love those inconstant customers
Rewiring for external PSU is the best bet.
But remember the relay coil probably "eats" more current than the actual circuit, and it would benefit adding an anti parallel diode across its coil, to lower pop.

[FredB]

Yes, the coils draws about 5x as much current as the fuzz circuit. There is an interesting phenomenon that occurs when the coil is deactivated. Since the coil draws so much current it lowers the supply voltage down to about 6-7 volts when the coil is energized. When the coil is de-energized, due to the rather high output impedance of the tiny transformer, the filtering caps and the dropping resistor, there is a lag time of more than a second where the fuzz circuit drops out of normal operation before the supply voltage finally rises back up to the nominal 9V.

The whole power supply and the rest of the circuit layout effect has a rather poor design. The input stages are closest to the power supply, making the output stages the farthest down the power rails away from the power supply. The whole thing just seems backwards and poorly thought out.

BTW The original circuit has a 100n capacitor across the coil.

[FredB]

Well, this repair is finally done, and I am quite pleased with the results. Here are some videos I made about it:

Final Tone Demo: https://www.youtube.com/watch?v=mcb6...jPDUQQ&index=1

Repair Part 1: https://www.youtube.com/watch?v=7MDx...jPDUQQ&index=4

Repair Part 2: https://www.youtube.com/watch?v=LrqT...jPDUQQ&index=3

[52 Bill]

Glad to see you got it figured out.