Oscillation in bypass mode

[Marbles]

Hey Guys,

Unfortunately not possible to post pics at the moment, but maybe you already have an idea.. I built the OKKO Dominator, so I guess that qualifies as high gain haha.

I did not have any issues when it wasn't boxed up yet. Used it in my test rig without any issue. Ofcourse there was the usual noise, but it was the kind of noise I'm used to and the one that dissapears when boxed.

So I boxed it. Confident that everything was working (and being somewhat tight) I ofcourse hot glued the board in place on the pot covers. The offboard wiring was done, I connect it to my powersupply, hook up my guitar, and even in bypass mode: peeeeeeeeeeeeeeeep.

A high peep noise. Even when the circuit is in Bypass. That is what I have trouble understanding.

The leads from switch to jacks are pretty short. I use a 1590bb upright, so there is not that much distance to cover.

Almost all my grounds are soldered to that same spot (input jack) only the circuits ground is grounded to the output.



The input from cicuit to switch is fairly long, cause it has to go from all the way up across the circuit board down. They do cross the power lines, but in a 90 degree angle. I did press on the IC's (in sockets) a bit when glueing it a bit, but that wouldn't explain it in bypass right?


The fact that it happens in bypass mode puzzles me. Also that it was not like this before. In my testrig it was silent when bypassed. Turning it on makes it slightly louder, but not that much. Could it be as simple as a ground not being soldered well?

Should I focus solely on the offboard wiring?

[Ciaran Haslett]

Parasitic oscillation.  Can happen for a number of reasons, especially with high gain stages.  You could try a few things...

Shielded in/out cable (was never much good for me)
Up the suppression caps (the 100p and 470p)
Change the charge pump
Add more filtering to both V+ rails (e.g. 100n caps ground)
Do you use an "Input Ground" wiring scheme like the one on the main site?  That will eliminate the oscillation on bypass.

[Marbles]

Thanks so much for replying. I understand it's kind of impossible to say anything sensible with not seeing anything. It did seem I had a bit less gain btw.. But that could be the boxing.

I use this as wiring normally:




Which is basically the same if I understand correctly. The grounds from input and output I usually never connect. Because of this problem I did, but to no effect unfortunately.

So, before trying all the other things (which I will eventually, so thanks for that!) I 'Hope' it's something else. especially cause these issues were not apparent when testing.

Would the fact that the input and output are grounded (which I understand that should ground the circuit completely) and still having oscillation point me in the direction of either a problem within my switch, or a problem with the ground connections?

Or am I just being too hopeful to not having to pull any of the other stunts? :D ( I'm afraid that will mess with the tone_

[Ciaran Haslett]

Hmmm

Assuming all your wiring is correct (and it sounds like it) then the only thing that has really changed is the proximity of boards/wiring etc.  I wonder if moving the charge pump somewhere else might solve it?

Your switch wiring is input grounding so that would definitely point me towards the pump/power supply.  Also...it seems there was uncertainty RE that 100p cap.  I'd tack another 100p in parallel to it and see what happens.

So...confirm your grounds are all connected with a meter, then maybe start moving wires/pump board about and see what happens

[Marbles]

Ugh, no luck yet. I remember when I started out, I would already box them and then debug/change parts and get frustrated with the small wires etc . Guess I'm in that same spot now haha.

I resoldered the ground wires. All the ground spots give me 0v.
Moved the pumpboard as far as I could. Changed it somewhat, but didn't stop it. Well, I thought it did, but realized I was connecting the traces of pins 6 and 7 with my finger. Tried another chip, no difference.

I hooked up my test rig again. Put the alligator clips for ground on the power jack, 9v clip on the power jack, input clip to the lug of the switch where the input of circuit goes, and output to then output lug.
This still gave me the high pitch noise. I guess this is the same situation as it was before boxing, so I'm starting to think something went wrong while boxing. I am 100% sure there wasn't any noise in bypass though before. This should at least rule out the powerjack I guess.

Did I test the grounds wrong though? I don't use a multimeter a lot, so I just put it on volts, put the black pin in one of the screwholes and with the red pin I checked all the ground points.

Could I test the extra 100p cap by just holding another next to it, legs touching? Or is that not enough?

Again, thanks a lot for taking the time!

[induction]

Some (not all) brands of hot glue are conductive. If yours is, then the glue is probably creating shorts on your board.

Have you used this glue for this purpose before? Can you scrape the glue off without damaging anything and see if the problem is fixed?

Worth a try, maybe.

[Marbles]

No way! Geez.. Yeah worth a try.. I think haha. I have used it before though, but worth a shot. It looks like it has to come out now anyway, sigh. Thanks!

[Marbles]

So.. reading another thread.. I read that people have had problems with 7660's squealing cause of the quality. Could that also explain my 'problems'? Is there a substitute more reliable I could test in that spot?

[Travis]

The 7660S itself is not inhently bad to use, but many of us (myself included) have been getting bad parts from Tayda etc

If the charge pump is causing the issue, it probably won't only squeal in bypass

As time goes on, I'm realizing the only things I'm interested in buying from Tayda are the pots and resistors. They sell a lot of junky stuff!

[Marbles]

Alright, a little update. Scraped away the glue (fairly easy) and still beeeeep.
Loosened all the pots, thinking maybe a wire got trapped: Beeeeep
Took the charge pump as far away as possible: Beeeeep

Got really tired of the beeeep. Snapped the grounds, 9v, input and output and took the circuit out. Connected it to my testrig aaaand:.................... Silent

So I'm kind of lost here haha. It leaves me with 3 or 4 culprits: Jacks, 3dpt, powerinput or led.

I'm going to guess the LED is out of the question. It is connected to the same lug from the powerinput as the 9v that goes to the chargepump (and the wires were tangled together) but it seems unlikely?

The jacks: I clearcoated the enclosure. So the outside has a little layer. But the jacks are grounded on the inside of the enclosure, so I would say that's not very likely either?

The powerinput: Like mentioned above: 1 lug goes to the charge pump and led directly (with 2.7k resistor) The ground is 'grounded ' on the input just like all the other stuff. I don't think I can mess up a lot there. Which would mean it could be a faulty power supply.

The switch: Most expensive, so hope not haha. There are no lugs connected/bridged as far as I can see. If there is weird noise sent to outputs however, it would be the most likely suspect? On the other hand: everything form the switch should be grounded, right?


I know, I know: "Dude, just switch the %$& parts and it will work probably". And you guys would be right. I'm just trying to figure out how it's possible to hopefully be able to debug things like this quicker in the future and understand :)

 

[Marbles]

Ok, I switched the power supply. Still the same.
I replaced the switch: Still the same.

The noise gets considerably louder (in bypass) when I touch the bottom row of the charge pump. It also get louder when I hold the charge pump closer to the switch.

i tried to tape the charge pump in the beginning (all around it), but that didn't help. I have tried to power it with a battery, didn't help either.

The annoying part is that it's in bypass too. It rules out a lot of things. But geez. What's left? haha

[Ciaran Haslett]

Jesus you're having a nightmare lad!

Still sounds like the only real thing that has changed is proximity.  When you test, is your board wires all nice and spread out on the desk...no wires running under the board?  Then in the box...wires running under/over the board e.g.?

Have you tried shielding the in/out wires?  Instead of buying shielded cable...try ghetto fashioning some.  Wrap the in/out wires with tin foil and ground only one side.

I'm fast running out of ideas haha?  Maybe a few pics of the boxed build?

[Marbles]

Haha, yeah, tell me about it.

After changing the switch and supply, the board wasn't in the enclosure. Just floating around.

It does make me think: The sounds gets worse when the chargepump is close to the switch or/and output. Even when in bypass. That might explain the lack of problems with my testrig.

My testrig is a 1590BB with all the offboard wiring inside, and alligator clips coming out of the eclosure. In other words: The charge pump is on the outside of the enclosure and the offboard inside. Maybe it would squeel the same if I would put it inside my testrig.

I have some shielded cable somewhere I think and will try that. Do I only have to do the wires from the switch to the jacks?

I am going to build a new charge pump. It is a small thing to build and seems to be the part emitting something that is picked up by either the switch or output. It sounds weird but that's really what it looks like.
Thanks for wlkaing me through this :D

[Marbles]

Alright, I rebuilt the charge pump, thinking that was the culprit. It's not lol.
I figured it was, cause if I bring the chargepump close to the switch or input, it gets worse. Also replaced the input just in case while I'm at it. I thought it was less, so (getting a wee bit tired of it) decided: I'll place the charge pump as far away from the switches and jacks as possible. I couldn't hear it in bypass anymore. Not when it was on either (so I thought).

Not wanting to change anything, I kept all the wires and ratsnest as is. and fastened all the knobs again. Thinking maybe closing it will have Faraday helping me with even less hums and whatnot.

Played it on love volume and it was ok. Highest gain settings though; Peeeeeeeeep.
Played around with it more, and realized that turning the treble up makes it a lot worse. So Gain and Treble are the enemies right now.

Also what I thought was interesting: when the midrange pot is all the way down, and I turn it up a little bit, there is a small pop.

The switch pops too the first time. Second time it doesn't, after a long pause it does. This is a picture of the top. Again my apologies for the mess. I know how annoying it is to have non clear pics with wires everywhere and bad angles:

[Travis]

Now that you've isolated some problem areas, I bet shielded wire for the treble, gain, input, etc could do the trick

[Marbles]

Alright, let's do this haha.

Do I have to shield the switch to output and input jacks? Or board to switch?

[Ciaran Haslett]

Ideally any wire that carries audio should be shielded but I think you've found your trouble areas.  So I'd start with shielding each pot wire, one at a time, all the while testing to see if if made a difference.  If not then move on to the next wire....rinse and repeat.  Or shotgun every single one if you don't have the patience but I don't think that's necessary and will only add cost to your build.

[Marbles]

My god hahaha. I built the Darkglass Microtube to procrastinate and even making the circa 200 cuts was a relief haha.
Oh well... Going to start with the inputs and outputs as suggested first then, sigh.
Just to be sure: I'm fighting symptoms here, but theyre not symptoms of a bad build?

Don't want to miss out. Like im not reaching it's full potential cause i made an error if that makes sense.

[Ciaran Haslett]

Naw lad....you've proved already that the build is sound.  You're fighting a high gainer crammed into a little box now.

[Marbles]

Ok, just a double check.. I have done some shielded wiring, though I found out i might not be using the right stuff... The ground wire is in my cable, loose, but it's not spiralling around it. Hope that's ok...

I have a feeling it's a bit less. But, I discovered something else.. Yes, the drama continues, but I thought maybe it means something, and maybe it doesn't, but want to bring it up..

As said before, the midrange pot is behaving a bit funky. There is a audible plop in the beginning of the turn, but I found out that at a sweet spot, the squeel is exactly in that spot. I filmed a bit to explain.

https://www.youtube.com/watch?v=lfOPaA2iZDY

Maybe that points to which wires should be shielded?