Backfeeder PLL noise bleed through

[Guardians of the analog]

Ok, I know this is a well documented issue. I looked up troubleshooting threads going back to 2019 about this one. Some people have built multiple boards with the same issues, any time it's been brought up the issue hasn't been resolved. @Robert is it possible it's a board/layout issue with some kind of bleed thru?

Yes, all my ICs, are from a reputable source. I have checked all solder joints, checked continuity, audio probed, triple checked values and checked all wiring. I even removed the feedback switch, rechecked continuity, hardwired, rewired the switch and it's still an issue. The pedal functions with it's intended purpose, the distortion works, even the feedback switch functions like it should. The rhythmic background noise is the issue. The video demonstrates the noise and how everything functions despite the noise.
Has anyone built one without the noise? Has anyone had one with this issue that found a solution? I have not seen many build reports and Chuck's build is the only one I've seen that does not mention it.

 

[Chuck D. Bones]

Let's see a straight down photo of your FEEDBACK switch wiring.
Sounds like the PLL is tracking and holding the way it should, but the mute circuit is not working right.
Did you sub any parts?

 

[Guardians of the analog]

Let's see a straight down photo of your FEEDBACK switch wiring.
Sounds like the PLL is tracking and holding the way it should, but the mute circuit is not working right.
Did you sub any parts?

No parts substitutions, when I tested the circuit originally I had only tested the distortion section as I did not have the momentary switch from tayda yet. I figured it would likely work without issue. I took it out of the box to troubleshoot.

What area is the mute section so I can focus there next. Also would the switch being faulty be a possibility? I'm making an order for stompboxparts and I'll get a new one just in case.

 

[Chuck D. Bones]

You used 2SC2603 for Q4, Q5 & Q8? If so, where did you get them and did you verify them in a transistor tester?
Where did you obtain the G10K pot? Those are hard to come by.

You're FB switch is wired same as mine.

The mute circuit is Q4, Q5, Q8 and all of the parts that connect to them. Vref is also part of the mute circuit, make sure that C101 has good solder joints.

With the power off, measure the resistance across R51. With the FB switch not pressed, it should read zero Ohms (or close to it).

With the power on, measure the voltage across R57 (one end is GND). With the FB switch not pressed, it should read zero Volts (or close to it).

 

[Guardians of the analog]

You used 2SC2603 for Q4, Q5 & Q8? If so, where did you get them and did you verify them in a transistor tester?
Where did you obtain the G10K pot? Those are hard to come by.

You're FB switch is wired same as mine.

The mute circuit is Q4, Q5, Q8 and all of the parts that connect to them. Vref is also part of the mute circuit, make sure that C101 has good solder joints.

With the power off, measure the resistance across R51. With the FB switch not pressed, it should read zero Ohms (or close to it).

With the power on, measure the voltage across R57 (one end is GND). With the FB switch not pressed, it should read zero Volts (or close to it).

Forgot about those transistors I used 5088s with the base and collector legs twisted. The g10k pot was a w20k with 20k parallel resistor. No other mods or subs. I'll measure the resistance and recheck all of the components around those parts when I get home from the store.

 

[Guardians of the analog]

The mute circuit is Q4, Q5, Q8 and all of the parts that connect to them. Vref is also part of the mute circuit, make sure that C101 has good solder joints.

C101 is good, vref is getting to where it needs to at the correct voltage at all spots. All parts connected to Q4,5 and 8 are all correct and I checked continuity so that checks out.

With the power off, measure the resistance across R51. With the FB switch not pressed, it should read zero Ohms (or close to it).

0.2 ohm

With the power on, measure the voltage across R57 (one end is GND). With the FB switch not pressed, it should read zero Volts (or close to it).

.5mV

I don't know what gives this is why I wonder if it's a manufacturing defect somewhere or if something is bleeding into the audio path somewhere.

 

[Paradox916]

I love the idea of this pedal. I’m interested in how this pans out.

 

[Guardians of the analog]

I love the idea of this pedal. I’m interested in how this pans out.

I really think it's a cool pedal and would like to figure it out too, it's not my first rodeo but this one has me stumped, as well as some other well-known builders

 

[Chuck D. Bones]

Your measurements say that the mute circuit should be working. Let's try forcing the mute. With the pedal running and connected to an amp, try these things, one at a time:

1. Temporarily short D7-A to gnd. Does the funky sound stop?
2. Temporarily short one end of R56 to gnd. Try both ends, one at a time. Does the funky sound stop?
3. Temporarily short IC7-2 to IC7-4. Does the funky sound stop?

 

[Guardians of the analog]

Your measurements say that the mute circuit should be working. Let's try forcing the mute. With the pedal running and connected to an amp, try these things, one at a time:

1. Temporarily short D7-A to gnd. Does the funky sound stop?
2. Temporarily short one end of R56 to gnd. Try both ends, one at a time. Does the funky sound stop?
3. Temporarily short IC7-2 to IC7-4. Does the funky sound stop?

Sorry for the late response, got the flu going thru the house. Unfortunately none of the above worked. I'm half tempted to try another board, whole new set of components and see if that does it too. I want to understand why there are various reports of this but not you.

 

[Chuck D. Bones]

Sure, you can do that if you like. Buuuuuut, if we don't know what went wrong with this board, we are ill-prepared to prevent it from happening on the next board.

I think the high density and parts count makes the likelihood of a build error exceed 50%. Mine was given to me by someone who couldn't get it to work. Took a fair amount of troubleshooting (and swearing) to get it going. It has 2N5088 for Q2 & Q7, 2SC603 for Q4, Q5 & Q8. No pin bending req'd. Mine was built with IC sockets, which I don't like because you can't see shorts and debris hidden between the socket and the board.

 

[Guardians of the analog]

Sure, you can do that if you like. Buuuuuut, if we don't know what went wrong with this board, we are ill-prepared to prevent it from happening on the next board.

I think the high density and parts count makes the likelihood of a build error exceed 50%. Mine was given to me by someone who couldn't get it to work. Took a fair amount of troubleshooting (and swearing) to get it going. It has 2N5088 for Q2 & Q7, 2SC603 for Q4, Q5 & Q8. No pin bending req'd. Mine was built with IC sockets, which I don't like because you can't see shorts and debris hidden between the socket and the board.

Yours has 2n5088 for Q2 and 7 with no pin bending? I thought all the transistors on this board have the usual japanese ECB pinout?

 

[Chuck D. Bones]

I was surprised to find out during my own troubleshooting exercise that contrary to what is implied in the Build Docs, those transistor footprints are E-B-C. A quick continuity check on the board will verify the pinout on Q2 & Q7. Since the distortion channel works, Q2 is installed correctly. We get signal from the Feedback channel, so Q7 is installed correctly.

 

[Guardians of the analog]

I was surprised to find out during my own troubleshooting exercise that contrary to what is implied in the Build Docs, those transistor footprints are E-B-C. A quick continuity check on the board will verify the pinout on Q2 & Q7. Since the distortion channel works, Q2 is installed correctly. We get signal from the Feedback channel, so Q7 is installed correctly.

I honestly don't know where to go from here. All my values are right for the components. It did it in and out the box, it's not a grounding issue. There are no shorts, all ICs are getting the voltages they should, audio probing isn't helping me figure out where it's coming from in the circuit. I removed and tested every cap, IC and transistor. Was hoping maybe @Robert had a possible idea. I guess I'll shelve it for the time being and come back to it. Bummer because I had real ideas on using the feedback switch

 

[Chuck D. Bones]

Let's try some more audio probing. These places should have no signal (or next to none) when the FB switch is not pressed.
Q4-C
Q5-C
Q7-E

 

[Robert]

I was surprised to find out during my own troubleshooting exercise that contrary to what is implied in the Build Docs,

You must have one of the first batch of PCBs, the PCB footprints were corrected back in 2019 and are all ECB.

 

[Guardians of the analog]

Let's try some more audio probing. These places should have no signal (or next to none) when the FB switch is not pressed.
Q4-C
Q5-C
Q7-E

No signal on either of these audio probing

 

[Chuck D. Bones]

That tells us that the mute circuit is working. If you depress the FB switch, then you should hear the feedback tone on all three transistors. Q4 will be the fundamental, Q5 the 2nd harmonic (octave up) and Q7 will be a blend of the two.
Is this problem intermittent? If you put the audio probe on Q2-C do you hear the PLL noise? How about IC7-7? IC8-2?

 

[Guardians of the analog]

That tells us that the mute circuit is working. If you depress the FB switch, then you should hear the feedback tone on all three transistors. Q4 will be the fundamental, Q5 the 2nd harmonic (octave up) and Q7 will be a blend of the two.
Is this problem intermittent? If you put the audio probe on Q2-C do you hear the PLL noise? How about IC7-7? IC8-2?

I get the PLL sound. The sound stops when the switch is depressed, it starts back up when you let goof the switch

 

[Chuck D. Bones]

If that's true, then the FB switch is backward. With the power off, measure continuity between pins 1 & 2. Should be open circuit. If it measures near zero Ohms, then the switch is backwards. You can either unsolder all 8 leads, rotate the switch 180° and then reconnect the leads per the Build Doc... OR you can move the wire from pin 3 to pin 1, from pin 6 to pin 4, and swap the wires that go to pins 7 & 9. The wire that's on pin 7 now probably won't reach.

---- BOARD ----