Phase 45 clone troubleshooting

[cnspedalbuilder]

EDIT-THE READINGS BELOW ARE OFF. The only relevant point from this post is "When I put the voltmeter probe on IC1 pin 3, the feedback (and all sound) stops," but I haven't re-checked this.

IGNORE THIS:
So I got together an audio probe (I cannot believe how much I've picked up since March), and this led me to focus on pin 3 of IC1. I redid some joints and re-tested, and it is still generating feedback, although the sound is a little different. The voltages also changed a bit, I think. Note that I used an analog voltmeter for this, so the readings might be off by up to .5 volts based on my eyesight.

Here are my rough readings:
IC1:
4
4
.5
0
3.5
3.5
3.5
9

IC2:
4
3.5
3.5
0
1
1
oscillating between 2.0 and 7.0
9

Q1=0 3.5 3.5; Q2= 0 3.5 3.5

One last, probably useless factoid: When I put the voltmeter probe on IC1 pin 3, the feedback (and all sound) stops.

Any help is much appreciated, I really feel like this is getting closer to a fix. Thanks!       
p.s. also I'll post updated photos.

[cnspedalbuilder]

Here are the photos:

[Kipper4]

Whats going on with pin 3 ic1?
meter loading?
Bad connection to Vbias?
Have you tried a continuaty test between R3 and pin 3?
The odd thing is even if R3 has a bad connection I'd expect that pins 2 and 3 would be similar because of the R2 no?
Maybe check for cold solder joints too.
Have you tried setting the trimmer?
Just thinking out loud.

[Cozybuilder]

Look at R29- it appears to be soldered in the wrong hole (its going to S3, not the LED anode)

Edit- maybe not, hard to tell, but check the orientation of the transistors (if they are not 2N5952 with DSG orientation)

[cnspedalbuilder]

OK I re-checked with my digital multimeter. I apologize for confusion, some of the error was probably due to the crappy voltmeter, but even across separate dugital multimeter readings I got some different results. This fits because, when I used the audio probe, a pin that gave feedback at one point would generate a fizzy hiss at other times. 

Anyway, here are updated #s:
IC1: 4.25  4.2  3.86  .6-1.0 (inconsistent)  3.9  4.15-4.2  3.7-3.8  9.49
IC2: 4.04  3.9  4.04  .6 to 3.0 (random oscillations)  .6-4.0 (osc) 3-6 (osc)  1.6-3.0 (osc)  9.49
Q1: 1.3-1.9 (osc) 4.04  3.90
Q2:1.4-1.7 (osc) 4.04  4.04

[cnspedalbuilder]

@Cozybuilder:If I understand your question then I think R29 is correctly placed. There is continuity between it and the "L" wire. The transistors are oriented with the outline on the PCB. From what I understand, they used to have the transistor layout backward, but apparently, in the new version, the PCB outline is correct (though I think he says you can experiment with switching it).

@Kipper 3:

Whats going on with pin 3 ic1?
My analog voltmeter reading was off. I dk why, but readings are now around 3.86
meter loading?
Bad connection to Vbias?
sorry I dk what you mean. Can you explain vBias and meter loading? I'm really green 
Have you tried a continuaty test between R3 and pin 3?
I checked, and there is continuity
The odd thing is even if R3 has a bad connection I'd expect that pins 2 and 3 would be similar because of the R2 no?
I'm really sorry about the earlier readings, now the are closer. From my limited understanding, the schematic would suggest that you have to cross R3 and R2 to get from 3 to 2. Is that wrong?
Maybe check for cold solder joints too.
I've redone a lot of the joints, but I will check again.
Have you tried setting the trimmer?
Yes, trimmer changes the sound a bit but it doesn't eliminate feedback

[cnspedalbuilder]

One addendum: With the audio probe, I found that if you follow the schematic all the way from input-->C1-->R1-->IC1a Pin3-->Pin1-->direct to R10-->C4-->Output, that is the line where the feedback and noise is most noticeable. There is some noise and FB in other parts, but that track is the one with the most consistent and loud feedback on the probe.

The line is relatively clean from C3->Q2->R26;C6->R15->R18-R25->R20,R27, etc.

[cnspedalbuilder]

A couple of relevant points that might have been lost in this thread:
- This build has the Univibe mod, which means that Caps C2 and C3 had a 10 to 1 value ratio: a 0.1uF and 0.01uF cap
-The feedback is incredibly loud and thick, suggesting that gain is going out of range
-With audio probe, feedback is present both at the input jack and the output jack. Also, if I just run a cable from amp to input jack, no other cables, you get the feedback. I'm guessing this isn't supposed to happen, right?  What could cause feedback to work backwards to the input jack?

Could the feedback reflect:
-a bad op-amp?
-something wrong with the op amp feedback loop like R4? (got this from a page at "op amps for dummies")
-a problem with capacitor C4?

[Kipper4]

When was this mentioned


"This build has the Univibe mod, which means that Caps C2 and C3 had a 10 to 1 value ratio: a 0.1uF and 0.01uF cap"

Where did you find this mod please?
if c2 and c3 values are 47nf how does a 10:1 ratio translate to 100nf and 10nf?
my maths aint that great.....
schematic
http://www.generalguitargadgets.com/pdf/ggg_p45_sc.pdf

Your audio path is IC1a,IC1b,IC2a.

IC2b is an low freqauncy oscillator.
and goes to the audio signal path through the transistors Q1 Q2
stick your audio probe on the no7 pin and turn the speed knob and you should hear the freqauncy shift.
This will tell you if your LFO is working

[cnspedalbuilder]

@Kipper4: Yes it is confusing. The caps that JD sent were indeed 0.1uF and 0.01uF, not 47 as in the stock schematic. The cap values are taken from the mod suggested by GGG: http://www.generalguitargadgets.com/pdf/ggg_p45_mods_instruct.pdf

I linked to the stock schematic that does not show the mod. I didn't think it would be helpful to link to the modified schematic because it shows all three mods and is much more complicated than the Mod1 circuit which just changes cap values. Obviously it would've been helpful to be more clear about this mod  Here is the schematic w/all mods if it helps: http://www.generalguitargadgets.com/pdf/ggg_p45_mods_sc.pdf

IC2b is an low freqauncy oscillator.
and goes to the audio signal path through the transistors Q1 Q2
stick your audio probe on the no7 pin and turn the speed knob and you should hear the freqauncy shift.
This will tell you if your LFO is working

This is useful to know. I know that on the voltages, IC2bPin7 oscillated between 1.6-3.0V. I will check how this translates to sound.

Q: Is there any possible reason why feedback would go through input jack as well as output? (or am I doing something totally wrong in diagnostics?)

[Kipper4]

Your voltages all seem to check out.
Someone said all the parts seem OK.

Look at R10 it's a negative feedback path from the output of all three op amps.

Can I just ask again.
When you plug your guitar in the input and amp at the output.
What does it do?
Does it sound phaser like?
Does it sound like a weak univibe?
Or just pure feedback loudness?

Sorry to appear so curt. I don't get why I wouldn't work if everything seems in order, and I really want it to work for you.

[cnspedalbuilder]

Look at R10 it's a negative feedback path from the output of all three op amps.

I checked the resistance on the multimeter and it came out right, but should I try to resolder that one just in case?

Can I just ask again. When you plug your guitar in the input and amp at the output. What does it do? Does it sound phaser like? Does it sound like a weak univibe? Or just pure feedback loudness?

I haven't tested it with a real amp in a week or so (because I couldn't take the noise!) but when I tested it on an amp, it sounded like Thor playing a Moog  There was a thick oscillating sound when no guitar was playing. When I played the guitar I could hear it but it was fizzy, like as if it was going through a fuzz face but at a very low output relative to the feedback. So there was something along the lines of a phase/vibe sound, I think.

Sorry to appear so curt. I don't get why I wouldn't work if everything seems in order, and I really want it to work for you.

Hey, I totally appreciate your taking the time to help. JD Sleep actually offered to check it out if I sent back to him, but I've put so much time in, I really want to figure this out on my own (well with your help!) 

Also, I'm sorry I have been reporting inconsistent information, but I am new to multimeters, audio probes, etc. This is only the second pedal I've built and before these builds, I had virtually no electronics experience. So I'm learning a lot through this process and from your feedback. Thanks! 

[cnspedalbuilder]

@kipper4: n.s. if this helps but when I probed the line of resistors below the R10 feedback line, I got fizzy noise from R5-R7, but past R7 it was *relatively* clean.

[Kipper4]

Is it possibly that you fried the fets with soldering.
No offence and I can't see too well but the solder looks a bit blobby on the fet legs.
Have you got spare fets and some sockets you could try?
While you have the meter out. What value is R1 on ohms setting?
Sorry to keep asking silly questions...

[cnspedalbuilder]

Is it possibly that you fried the fets with soldering.
No offence and I can't see too well but the solder looks a bit blobby on the fet legs.
Have you got spare fets and some sockets you could try?

Not offended, it's possible. The FETs are Q1 & Q2, right? Q1 to my eye looks ok but Q2 is a little blobby.  I did install the FETs and ICs after other PCB components in order to minimize heat damage, but it's possible that they got burned.  I don't have anything to test them, and unfortunately, don't have spare FETs or sockets. Would it make sense to resolder Q2?
Also, do you think the voltages are consistent with fried FETs? It looks to me like the Gate voltages might be off but the instructions say "2.2V plus or minus some".

While you have the meter out. What value is R1 on ohms setting?
Sorry to keep asking silly questions...

9.98K which seems to check out with the schematic. Is it just me or is this maddening? But please keep the questions coming!

[robthequiet]

Does it still change the sound if you physically move things around like battery wires and knobs? If you connect an output cable ( but not to anything at other end) does it change the sound at all via the audio probe? My gut on this would be a trim pot out of alignment or a short/open somewhere.

[robthequiet]

And just for kicks, is everything that is supposed to be at ground really at ground? With power off, check for continuity from case to battery negative, wiggle the wire to make sure. Then move on to check around for things that should be at ground, especially the ground pins in the ICs. Another thing is check that the IC pins have a good connection through the board. I think this may seem very basic and you probably already did most of this. But sometimes it's easy to overthink a problem. gl

[duck_arse]

my turn to be rude. what part # do you think your fets are? and what part # is marked on your fets? (it pays to play this game sometimes.)

also, as RtQ suggests - get a resistor leg cut-off or something similar, and poke into each pin of your IC sockets in turn. it sometimes happens that socket pins don't actually contact the IC legs.

[cnspedalbuilder]

Thanks guys for all the ?s, suggestions. Pls. see below:

Does it still change the sound if you physically move things around like battery wires and knobs? If you connect an output cable ( but not to anything at other end) does it change the sound at all via the audio probe? My gut on this would be a trim pot out of alignment or a short/open somewhere.

-Yes, the pitch and oscillation frequency would change when I moved wires around.
-For the audio probe test, do you want the audio probe positioned at the output jack when I pop in the output cable? Also, last I checked the pedal would not power on w/o input cable, but that was with the battery, it might switch on w/9V adapter
-When you say trim pot out of alignment, do you mean that it needs to be adjusted? I tried adjusting the trimpot across the whole range, and although sound changed, feedback was still there. Also, when I checked w/audio probe, points at trim pot were clean.

[cnspedalbuilder]

my turn to be rude. what part # do you think your fets are? and what part # is marked on your fets? (it pays to play this game sometimes.)

-Given that your handle is duck_arse, this is not at all rude
-Good idea: I believe it is 2N5485 (I ran to the garage to check w/a magnifying glass, and after returning, my memory is a little vague. Aging sucks.) The bill of materials and schematic say: 2N5952. According to the GGG instructions:

"The kit comes with a great sounding set of matched 2N5485 transistors.
The original units had 2N5952 JFET transistors. Note that the pin out of these
transistors may very. The 2N5952 will probably have to installed backwards from
what the layout shows.