Distortion pedal volume/noise jumping around, can't figure out why

[canman]

Hey gang,

Been a while since I've built anything...had a temporary job, then got two jobs, then lost both jobs.  It's been a weird 3 months.  So now with all this time on my hands I thought I'd redo the first Triple Wreck pedal I built for myself.  Very poorly done...after pulling it out for the pots, I'm surprised it even worked.

Anyways, I thought I'd make the PCB myself this time with our CNC machine from a previous business venture.  I used this layout below.  I had to "rebuild" it in Diptrace so I could get the G-Code for the machine, but I didn't move any components around and for all intents and purposes, it is the "same" layout.  (My apologies for the large images, I can't figure out how to scale them back...)

 

Here is the engraved PCB.  Yes, the holes are huge.  The smallest CNC bit I have is 1.4mm, so I made the holes and traces as big as I could in diptrace.  It works fine, but I'm definitely ordering a smaller bit for PCB holes.



And here is the populated board:



Now, the issue I'm experiencing is bizarre.  Everything about the pedal works as it should.  All the controls and switches work exactly like the previous build I'd done.  But after a minute or two, I start to get this weird noise issue.  Rather than explain what it sounds like, I've recorded it:

https://soundcloud.com/jacobcanady/tw-1

To me it seems like the volume keeps jumping around.  I'm not really sure what could be causing this.  The voltages all check out on the IC's (I have an addiction to building Triple Wreck pedals for friends/family so I printed out the voltages and I've confirmed they are good on this build).  So unfortunately, I have no idea where to even start looking for the issue.

What does this sound like to you guys?  Any advice on where to start looking?

[GibsonGM]

Bad solder joint....1/2 broken wire to a pot or the battery, or a jack...something mechanical, that moves.  Even if you don't THINK it's moving.  Opening a ground or something.

Try to aggravate it, poke things with a pencil see if you can narrow it down to a pot or something.  Look at your solder joints with a magnifying glass, look for cold gray looking solder (good joints are a bit shiny).

You can also audio trace it with an audio probe, but I honestly think you can find the issue pretty quickly by eye and poke. 

Nice board

[canman]

Argh, I hate when it comes down to this...I'll check every component on the board and all the pots/wires/jacks and let you know what I find.  I suspect it could be one of the pots, they sure got manhandled in the last build!   

[Ben Lyman]

lower left you got 2x 1M side by side, looks like supposed to be 1M and 1K side by side.
Looking closer I see it's the input, a 1M in series will def lower your volume 
you could solder a 1.2k parallel right on top of that if you are feeling lazy, probably what I would do if it was my own pedal 

[canman]

 


How did I miss that??

Thanks for pointing that out!  I will replace it with the proper value...you think that's what could be causing the volume to go up and down?  That's totally weird...

[thermionix]

I'm suspecting the board itself (and connections thereto).  Some of those holes have barely any copper around them.

There's also the possibility that little scraps of copper are in the routed channels and making unwanted intermittent connections, but maybe you cleaned all that out.

If I were you, I would make another board, but instead of CNC drilling 1.4mm holes, have the "router" bit put in small starter divots, and finish the holes with a hand drill (or drill press if you have one) and a smaller bit.

[GibsonGM]

1M in series sure would lower your input level quite a bit, but wouldn't account for an up/down thing like we're hearing on the clip...what Thermionix says...the board, perhaps.

If any joints aren't totally solid, you can have that sort of symptom...as well as frayed wires, crappy wipers of pots, jack terminals in the process of breaking...a bad patch cord....amp input jack....just a process of narrowing it down 

Nice eye, Ben!  I didn't even look at the components. 

[robthequiet]

Last time I had something like this it was a flaky guitar cord -- might check area around input jack if not jacks themselves.

[canman]

My initial reaction was a bad cable...so I removed the pedal from the chain and everything works perfectly.  So, I know the issue lies within this pedal somewhere.

I pulled the PCB out and poked, prodded, and pulled on everything I could.  Couldn't find anything that triggered the issue!  I did switch the 1M resistor to 1K and when the pedal is working normally, it sounds much better now haha.  Unfortunately, I can't locate the cause of the problem.  It doesn't sound like a failed component to you guys, does it?

I think a rebuild is in order...the problem is the smallest bit I have, drill and CNC, is 1.4mm.  So, time to order a few bits!  What size is standard for a PCB, and what are your favorite sizes for drilling PCBs?

Also, I did notice something weird.  When I hovered my finger over the input jack, I got a lot more buzzing.  When I did the same thing over the output jack, nothing.  I'm confident this isn't the cause of my problem, but something interesting for sure.

[karbomusic]

At first this doesn't sound like a cold solder joint or loose connection to me however.....  I seemed to notice the volume 'ramps' up, it isn't as instant as I expected so maybe that's a clue as to where a bad connection might be if there is one. In other words, we could probably rule out some parts of the circuit because we'd expect at least some DC pop and or more instant volume change when the connection dropped and so on or maybe the gain itself is suddenly changing.

I'm not sure if you have measured various voltages but if you had a scope or something sufficient to measure changes that are that sudden (peak hold) you might catch a sudden change in a voltage that gives clues; lastly maybe a component is being squirrely. Personally I wouldn't rebuild it just yet, I think you can figure it out but I could potentially be out of my newbie mind so there is that.  It's just that hearing it ramp up like that, caught my attention.

You said after a couple minutes. Does it exclusively never do it when first powered up and always does it after a couple minutes?

[karbomusic]

I'm confident this isn't the cause of my problem, but something interesting for sure.

You are likely correct in that assumption. Should be similar to placing your finger over the end of an free-hanging guitar cord tip while plugged into an amp.

[canman]

At first this doesn't sound like a cold solder joint or loose connection to me however.....  I seemed to notice the volume 'ramps' up, it isn't as instant as I expected so maybe that's a clue as to where a bad connection might be if there is one. In other words, we could probably rule out some parts of the circuit because we'd expect at least some DC pop and or more instant volume change when the connection dropped and so on or maybe the gain itself is suddenly changing.

I'm not sure if you have measured various voltages but if you had a scope or something sufficient to measure changes that are that sudden (peak hold) you might catch a sudden change in a voltage that gives clues; lastly maybe a component is being squirrely. Personally I wouldn't rebuild it just yet, I think you can figure it out but I could potentially be out of my newbie mind so there is that.  It's just that hearing it ramp up like that, caught my attention.

You said after a couple minutes. Does it exclusively never do it when first powered up and always does it after a couple minutes?

As far as I can tell, yeah it only happens after a few minutes.  When I first tested the circuit unboxed it sounded great, so I boxed it, started jamming, and after a few minutes I noticed it got weird.

So, I unboxed it, and it acted fine.  Until a few minutes later.  Same story today whilst working on it!  I don't have a scope, unfortunately.  I might be able to get some voltages while the volume/gain gets weird, and then while they're normal. 

Either way, I'm fine with a rebuild...I'd feel better about a circuit board with normal holes anyways!

[GibsonGM]

Hmm....I support either a circuit board bend, as mentioned, or a cold solder joint.  Or bad pot, something like that.   The usual is to try to narrow down a component that's causing it, but chop-sticking it or something.   Or audio probing the stages, to see if one stage is OK while this is going on. 

If you're not against it, maybe a new board IS the trick, tho, given that there is some suspicion about it? 

[canman]

I might as well audio probe the signal path and see if I can find where things are getting weird.  I was thinking last night, even though the pots were manhandled in my early pedal building days, they never gave me issues in that horrendous build, and the soldering job on the pots this time around was much faster and cleaner, so I'm hoping the pots are not to blame.

The more I think about it, the PCB seems to make the most sense.  Is it R.G. who has said on multiple occasions that it's fairly rare for a component to fail and cause this kind of problem? 

Regardless, I have time, so I'll try to probe the audio signal and see if I can locate the problem.  I'm betting if I do find it, it will be board related. 

(Though, the benefit to making a new board is watching the CNC machine work...that never gets old!)

[karbomusic]

Still seems odd to me that the volume change isn't instant but instead swells/ramps up, that it also takes some time to occur, and only occurs when boxed up. Those have to be a clue IMHO. What I mean by the ramping is that some parts of the circuit would always be an instant change with an intermittent connection but in others not so much such as a cap discharging/charging etc.

[canman]

And that's what I'm a bit more concerned about, is a bad cap.  Not a huge issue, but I'm out of the caps I'd need to replace anything, and my meter doesn't test capacitors.  It happens when it's unboxed too, by the way.  When I first tested it, it wasn't doing it.  Probably because I didn't give it enough time for something weird to happen. 

I'd think with a bad cap though, the volume swelling issue would be more consistent.  After messing with it more yesterday, I don't think it's ramping up as much as you think.  The audio clip seems to demonstrate that, but it's pretty sudden.  Really random, no rhyme or reason to it.  Sometimes it ramps, sometimes it just snaps back to normal (or back to quiet)!

[thermionix]

My money's still on an intermittent connection.  Either on the board, or possibly a pot lug.  That's pretty common if they've been overheated.

[canman]

I've got a bunch of other pots, I can swap them out and see what happens.  I ordered some 0.8mm CNC bits to drill PCBs with, so I'll likely rebuild this anyways.  I'll let y'all know what I find with the pot swap.

[karbomusic]

How long did it take for the CNC to do that by the way?

[canman]

How long did it take for the CNC to do that by the way?

About 10-15 minutes...the feed speeds need to be pretty slow for a smooth cut in the copper.  If you go too fast it's a really rough cut, and there are tons of copper slivers that would make debugging life miserable!

The longest part was putting the circuit board design together in Diptrace, and that wasn't too bad because I had an existing layout to follow.