Phase 45 clone troubleshooting

[cnspedalbuilder]

Hello, this is my first post and I'm only working on my second pedal kit so it's going to be amateur hour.

I'm building a phase 45 clone with a univibe mod (same as stock but replace 2 caps) from a GGG kit. Here's the link: http://www.generalguitargadgets.com/effects-projects/phase-shifters/mxr-phase-45/

I've built it, inspected for shorts, cleaned up some solder joints, etc. but it's not working right. I get normal clean sounds on bypass, but when I activate the pedal I get ridiculously loud oscillating feedback. It sounds like a theremin or Moog, so it's got a vibey sort of sound, but that happens even with no play on the guitar and it's so loud that you can hear it when the amp volume is at 1. When I did one long test (which scared the hell out of my dog), the battery started to get warm so I stopped. On bypass it totally goes away.

GGG has listed the voltages for pins at the ICs and trannies. I've tried testing the voltages but this is totally new to me. I tested with red lead at these points with battery in, black lead in or on the enclosure. I couldn't get any voltage readings here, although the meter perfectly picks up the 9V battery difference. If I put the black lead on the negative side of the 9v battery and I was able to get voltages.
With that, I get for both IC1 and IC2: Pins 1 through 7 =7v, Pin 8 = ~9v

For Q1 I got 0 7 7 (sorry I am not sure which lead is gate, emitter, etc.) and Q2 I believe was 0 8 8 0 but I can't be sure because I couldn't see it perfectly as I was balancing the leads.

I am thinking that there is a short or maybe a grounding problem, but I'm at a loss to find it. I think the wiring is all right, but I'll check yet again. Does anyone have any other suggestions? Also, if you have corrections on the voltage measurements, pls. let me know. Thanks in advance for your help. I hope to pay it forward as I learn more.

[thermionix]

If you can post pictures, that always helps.

[cnspedalbuilder]

OK now I feel like a real idiot, but when I hit the attachment button
I just see "img/img in brackets

[thermionix]

AFAIK you have to host the pictures at another site.  I use photobucket, then copy the "IMG" link they provide, paste in thread post.  There are probably simpler hosting sites to use.

[cnspedalbuilder]

OK, here are links to some pics:

https://drive.google.com/open?id=0BxlQwqVFZmffQk5HMHY2UzN6TjdGTWFsNXNDMm8zOGVBSy1R

https://drive.google.com/open?id=0BxlQwqVFZmffOVozSmJTa00zMnUxeEdYMDZiSnVxWFQ4UnJZ

https://drive.google.com/open?id=0BxlQwqVFZmffZnFER05UTkpMYWtiZnBxTWhpWWZDQzNFWG1j

https://drive.google.com/open?id=0BxlQwqVFZmffUHVFR3ZkWk9rMmNCRUpvdzVCX0V1OEI1NHow

[thermionix]

That way works too.  With the voltages you reported, I expected to see a bunch of bridged IC pins, but not really seeing that.  The first pic of the foot switch is pretty blurry, but (no offense!) the soldering looks a little rough.  Doubt that's related to the voltage results your getting.

I'll step aside and let the more knowledgeable folks chime in.

[cnspedalbuilder]

I realize that google makes you log in for those links. I'll get on photobucket tomorrow. And yeah I'm very new to soldering. The PCB joints aren't bad, it's more the wires that are tough, esp. with wires that tend to come apart after you put it through the lug or terminal.

[thermionix]

I didn't have to log in.

[PRR]

> I didn't have to log in.

Google login is getting pretty sticky. I haven't used it in a while, but when I hit cns' links I got the face of my dead dog in the upper corner, showing Google remembered me (my IP? my PC? my browser?).

[thermionix]

Doesn't seem to be remembering me.  I never stay signed in to anything, Firefox is set up to dump history and cookies when closed, and I don't store any passwords.  I see OP's posted pics and a "Sign In" button at the top right of the page.  YMMV, yadda yadda yadda.

[bluebunny]

I would say you have shorts or bad connections.  All your ground points should be 0V.  All your power (to ICs) should be 9V.  All other pins are likely to be pretty much right in between - so around 4.5V.  Carefully check your soldering and carefully check continuity for all those places you expect to be connected.

P.S. When you solder wires to the board, twist the ends together and solder them.  Now poke through the board: they won't unravel.  Finally solder your joint.

[cnspedalbuilder]

I have checked continuity and it was fine across all points on the same path but I might have missed some, so will check again. I haven't checked for continuity between points on different lines, but there are a lot of possible connections so that will take a while.

Offhand, is there anything in particular that would be expected to cause a ridiculously high output oscillating feedback even with minimal input signal?

Also, w/r to the voltages, I might not be doing them correctly because this is new to me. I put the black lead in one of the enclosure screw holes which I assumed would be ground. Does that sound right? If it is, then would my failure to measure voltages on the circuit mean that somehow current is going throuhg the enclosure?

apologies in advance if these are lame questions, I am really new to the world of circuits. Thanks!

[cnspedalbuilder]

OK my friend helped me figure out what I was doing wrong with voltages. Here are the voltage readings he got for the ICs and the transistors. I think they are fairly close to what they list at the GGG site:
http://www.generalguitargadgets.com/pdf/ggg_p45_instruct.pdf

Does this help to narrow down the problem?

IC1:
4.08
4.03
3.76
0
3.94
4.06
4.04
8.59 (full voltage of battery)

IC2:
3.87
3.95
3.94
0
5.49
1.70
7.97
8.58 (again, this is equal to full battery voltage.


Q1
1.6
3.94
3.94

Q2
1.59
3.94
3.94

[PRR]

> IC2: pin 7 = 7.97

I think IC2b pin 7 is stuck high. Normally it slams 1V to 8V and back, constantly, so a meter will give a wobbly reading in the 3V to 6V range.

Examine the IC2 pins 5 6 7 area, and the connections to the Speed pot.

> Q1 Q2 Gate = 1.6

These should change with R26 TRIM pot setting. I would expect better result with FET Gates above 2V, maybe 3V. However if IC2b is stuck, the FET trim is moot.

[cnspedalbuilder]

> IC2: pin 7 = 7.97
>I think IC2b pin 7 is stuck high. Normally it slams 1V to 8V and back, constantly, so a meter will give a wobbly reading in the 3V to 6V range.
>Examine the IC2 pins 5 6 7 area, and the connections to the Speed pot.

This is great advice, I will check it out! So I forgot to mention, voltage on IC2 pin 6 was wobbling, but it would settle on 1.70, so the actual voltage might be oscillating.

In terms of examining the pins, what am I looking for? should I resolder those points? I don't think there are any solder bridges, but of course, I can check again. Is there anything else I should look for in that area?

[cnspedalbuilder]

Hello, I took a break to see if I could come back with a fresh perspective. I've definitely cleaned up some bad joints, but there is still feedback. Oddly, with a DC power input and only output cable to amp, the pedal still produces feedback when switched on. This means that, somehow, the power input leads to feedback oscillations that are going to the output, right? The frequency of feedback goes higher when there is an input but either way it's loud. Moving the PCB doesn't have any effect, so I don't think it is a bad wire connection (though I will continue to look at them).

Does anyone have any ideas of what could be happening? Thanks!

[M.A.P]

Did you check the soldering on the 3PDT switch? Looks like there is a little lead shunting two poles of the middle switch.

Have you tried to swap the two op-amps? Maybe the oscillations change then. If it does one of the op-amps is damaged.

Hope you can fix your pedal soon

[cnspedalbuilder]

**I am feeling a little dumb. It does not generate feedback if there is only output. It requires something to be plugged into the input line.

>Did you check the soldering on the 3PDT switch? Looks like there is a little lead shunting two poles of the middle switch.

Yes, I just re-did the wires and I re-did any joint that looked wonky (successfully built a pedal in between so my soldering technique has improved!  ). It's much prettier now, but still generates feedback  I can post updated pictures if it's useful to troubleshoot.

>Have you tried to swap the two op-amps? Maybe the oscillations change then. If it does one of the op-amps is damaged.

NO I haven't tried that. Are you saying that it might change the frequency of the oscillations?
>Hope you can fix your pedal soon
THANKS for the support.

[cnspedalbuilder]

@M.A.P. I tried switching the ICs but there was no change in the oscillation frequency.  There was some change in the feedback frequency when I moved the PCB up and down, but I could not localize the change to any particular wire. It does look like power goes out when I move the battery cable around (probably a cold solder joint from the battery cable to the DC jack), but that seems to be separate from the feedback issue. Another piece of info I forgot to mention is that if I play guitar through the pedal, the feedback oscillation is replaced by a fizzy, fuzz guitar sound. So inputs do go through the pedal, but it's clear that there is something wrong with the circuit that is causing the feedback and distorting the input signal.

I'm going to try more diagnostics tomorrow, but if y'all have ideas, pls. let me know.

[M.A.P]

Ok this is getting tricky

I've checked the orientation of the poled caps, the FETs and the diode with your pictures and everything looks perfect. I think you really have a problem with one or more cold solder joints.

Take some time and redo all solder joints. That would be the next logic step I would take.

I once swapped all caps and transistors of a pedal that was faulty. I redid all solder joints and checked everything. But the pedal was in perfect condition! It was a low battery that cutted the signal like a low pass filter