Magnetron Delay Problem

[Jovi Bon Kenobi]

Here's the problem. When I test the board out of the enclosure I get two possible outcomes:
a.) I only get one repeat turning the full range of feedback knob with self oscillation at the very end, the dry signal is quieter than the repeats, there is faint RF, and my overall guitar signal is quieter when the pedal is engaged than when it is bypassed.
Or b.) It works perfectly with none of the above issues so I put it back into the enclosure and the problems come back.

Upon wiggling connections and stuff I can't seem to force replicate a vs b. I thought I may have solved it when I noticed that the OPA2134 was slightly lifted from it's socket and it cleared when I pushed it back in. However, once back in the enclosure the problems came back.

Link to schematic here: Magnetron Delay

I built the PCB with a couple mods added:
1. Momentary feedback footswitch
2. Internal gain trimpot moved to external control

I drew this up to make it easier to understand:


Photos incoming as soon as my housemate is out of the shared music room/my work area. ?

 

[Jovi Bon Kenobi]

Ok, here's some photos. Let me know if a clear shot is needed somewhere specific. Also, I omitted the LFO indicator LED because I didn't like it in the first one I built.

 

[Jovi Bon Kenobi]

I just noticed that the first relay bypass modules from last year aren't on the site anymore and that's what this is. Not sure that's an issue but I imagine it would be easier to troubleshoot without it. I should have tested before adding it.

 

[Jovi Bon Kenobi]

*edit* The following did not work.
I think I solved it! Something was telling me that it was a grounding issue on the gain pot as it seemed the most noisy variable upon turning. Pinching it between my fingers would produce all kinds of grounding noises. Also, I was getting continuity between lugs 1 and 3. Pretty sure that's wrong. Put a new pot in and everything works! Just gotta box it up one last time (I hope) ?

 

[moonlightpedalbuilds]

Are you using stranded or solid wire? Gauge?

 

[Jovi Bon Kenobi]

Dang, nope its back again. it's gotta be a grounding issue somewhere. it's so close to working.

Are you using stranded or solid wire? Gauge?

24 awg stranded. not pre bonded

 

[piapium]

Maybe when you cut the legs of the pots, you broke them. It happened to me once. They stopped working when I cut the lugs.

 

[moonlightpedalbuilds]

From previous experience when dealing with stranded wire, I single strand of wire may touch other parts. Especially when try to insert stranded wires into the holes in the pcb, potentiometers and switches. What I do is to make sure they at tinned before inserting them into holes.

Another thing to look at are the through-holes. Make sure your solder bleeds to the other side of the board.

Do you lose the sound when you tighten the potentiometers?

 

[Robert]

With the board out of the enclosure connect a wire (or clip lead) to a known ground point on the board.

Now touch the metal body and shaft of each pot and see if the problem reappears.

 

[Jovi Bon Kenobi]

Maybe when you cut the lugs of the pots, you broke them. It happened to me once. They stopped working when I cut the lugs.

Seems to read ok still, but note taken thanks!

From previous experience when dealing with stranded wire, I single strand of wire may touch other parts. Especially when try to insert stranded wires into the holes in the pcb, potentiometers and switches. What I do is to make sure they at tinned before inserting them into holes.

Another thing to look at are the through-holes. Make sure your solder bleeds to the other side of the board.

Do you lose the sound when you tighten the potentiometers?

Visual inspection with lighted magnifying shows most pads with wire attached are clean joints and through holes filled but some are obscured by the insulation on one side. No stray wires. Will inspect those and redo if needed, thanks!

With the board out of the enclosure connect a wire (or clip lead) to a known ground point on the board.

Now touch the metal body and shaft of each pot and see if the problem reappears.

On it! Much appreciated. Will report.

 

[Robert]

I missed the part about it sometimes misbehaving out of the enclosure, my idea might not help.

 

[Jovi Bon Kenobi]

With the board out of the enclosure connect a wire (or clip lead) to a known ground point on the board.

Now touch the metal body and shaft of each pot and see if the problem reappears.

touching a jumper between the ground point on the board and to all pots shafts and housing produces only one noticeable difference and it's to the gain pot. it produces what seems like RF.

 

[Jovi Bon Kenobi]

I missed the part about it sometimes misbehaving out of the enclosure, my idea might not help.

Ok. Yeah, it's weird. It either works just like my first mag delay out of the enclosure or it has those problems. when I handle the PCB while testing it makes it more noticable

 

[Jovi Bon Kenobi]

I have another Magnetron board. If at first you don't succeed...
Round 2...FIGHT!

 

[zgrav]

1. sometimes it works when it is not in the board
2. sometimes it does notw work when it is not in the board (only one repeat)
3. it never works in the enclosure

Is that right?

I wonder if you have a cracked trace on the board creating an intermittent connection that is part of the input to the feedback.

you can try flexing the pcb when it is out of the enclosure to see if it affects whether it works. if you can find a wonky spot you can bypass it.
also suggests the board is under some pressure in the enclosure. slightly enlarging the holes for the pots might fix that issue.

 

[Jovi Bon Kenobi]

1. sometimes it works when it is not in the board
2. sometimes it does notw work when it is not in the board (only one repeat)
3. it never works in the enclosure

Yes to all 3 if you meant enclosure instead of board for 1 and 2

I wonder if you have a cracked trace on the board creating an intermittent connection that is part of the input to the feedback.

Interesting. I mean, I use a lighted optivisor and visually inspect and throughly clean PCB's before use. This one looked and still looks fine but it's worth a shot.

 

[Jovi Bon Kenobi]

Also, with all my builds I allow a hefty gap between the pots and the board. In this case, the nubs of the pot lugs are sticking out only about 1mm above the pads before soldering. I like doing that for extra wiggle room if needed.

 

[zgrav]

I would try reflowing the connections going from the first amp stage to the PT2399 input. Something is intermittent. A bad trace is a possibility, but it could be something odd like a bad lead on a resistor.

 

[Jovi Bon Kenobi]

I would try reflowing the connections going from the first amp stage to the PT2399 input. Something is intermittent. A bad trace is a possibility, but it could be something odd like a bad lead on a resistor.

Sadly, that is beyond my scope of fully understanding but I have the schematic printed out. I'm also populating another magnetron board.
I think I went about 30+ PCB's without any problems and I hit this and my confidence gets shook. I really want to learn to read a schematic but don't know where to start.

 

[twebb6778]

I only know the basics but being able to read schematics, even just a bit, is a huge help when troubleshooting.

Take a look here if you get a chance https://learn.sparkfun.com/tutorials/how-to-read-a-schematic/all