Mysterious static in a preamp channel

[midwayfair]

I've got a modified Hamptone circuit that I have several channels of, and one of them is giving me some constant static in certain gain settings. The static is VERY loud -- around around -35dB at its loudest, so the channel's unusable.

The schematic below is modular, but it's the version with a volume pot (the 10K "Gain" pot hanging off the tail of Q2), transistors are all as shown (2N4401 in Q2) and it's running on 48V.

https://www.dropbox.com/s/lh6g08ayc4ygpxu/FET mic preamp.png
I've got two channels built identically to this one, and one that's identical except with a silicon transistor in Q3 instead of a second FET, but this is the only one with static. They all crackle when I turn the volume (completely expected) of course, because the wiper's changing DC.

The circuit does NOT exhibit static when the gain knob is maxed out, nor when the gain knob is backed off a couple detents from max. The static does get quieter when I turn the gain knob down, but it's actually loudest when the knob is around the center of its travel

If I take a big capacitor and silence various parts of the circuit, the first place the static goes away is the wiper of the gain pot. (NOT the emitter of Q2!)

If I pull Q2, the static goes away, so it's not coming from Q3 or later. I've replaced Q1 and Q2 with no changes. I tried a few different transistors in Q2 with no change.

I thought it might be C3 leaking, so I replaced it, and there was no change. I pulled and checked the filter caps relevant to Q2 and they're all still good.

I suppose, since this is the only channel doing this, that I should just build a new board and replace this one, but I'd really like to get some idea of what could cause this in case it happens again.

 

[midwayfair]

I left it on overnight and it went away as suddenly as it appeared (it didn't used to be staticy, in fact I recorded some tracks with it a week ago). While this is good, it adds to my apprehension.

1. Good: It works and I know why
2. Bad: It doesn't work and I know why
3. Worse: It doesn't work and I don't know why
4. Hell: It works and I don't know why <----- we are here

 

[Ike Zimbel]

I would look for hairline cracks in the circuit board traces. These can be incredibly difficult to spot...magnification helps. Flexing the board helps. Cold spray might help. Where I have found these most often is at the transition point between traces and pads. If the circuit boards have solder mask, that adds another layer of complexity as the visual transition between the solder mask and the tinned pad can easily cover up the fault. The fact that it went away after being left on overnight also supports this since the copper would expand as it warms up, forcing the edges of a hairline crack together.

 

[midwayfair]

That's a really good theory! We leave the heat off during the day most of the time so it's pretty chilly, I can fire it up a few times during the day and see if I can get it to go staticy again.

I think this board HAS been pulled at least once, so it's definitely possible that it has a hairline fracture from handling, though I did reflow solder joints so it would have to be a trace itself.

 

[JohnRoberts]

look for bad (intermittent) solder connection... maybe tap around the circuit board with a pencil eraser looking for a response to percussive troubleshooting. [edit- another troubleshooting tool was a bottle of spray cold to force thermal cycles /edit]

Changing with gain pot symptom sometimes suggest circuit instability (oscillation). Check compensation cap, PS decoupling, etc.

JR

 

[kr2p]

Did you manage to solve the problem? I have an extremely similar sounding situation with a Neve preamp that I'm trying to diagnose. The static noise is very loud, it goes away when the gain is maxed out or backed off a couple of detents, and is loudest at a medium gain setting. I can't see any solder issues (or at least, I've fixed the questionable ones I found and it made no difference) but I haven't looked for hairline cracks.

 

[rock soderstrom]

Changing with gain pot symptom sometimes suggest circuit instability (oscillation). Check compensation cap, PS decoupling, etc

Hi kr2p, your situation sounds like oscillation. Check your wiring, keep outputs away from inputs and use shielded cables. Check compensation caps and PS decoupling as JR said.

 

[midwayfair]

Did you manage to solve the problem? I have an extremely similar sounding situation with a Neve preamp that I'm trying to diagnose. The static noise is very loud, it goes away when the gain is maxed out or backed off a couple of detents, and is loudest at a medium gain setting. I can't see any solder issues (or at least, I've fixed the questionable ones I found and it made no difference) but I haven't looked for hairline cracks.

It never reappeared for me, even when firing up the circuit cold (and we leave the house at 62 a lot of days, so it's definitely been cooler than it would be even in October). If it ever reappears I'll probably replace the PCB because I wasn't able to find a bad part, misrouted wire, or microphonic stuff.

If you aren't already using shielded cable, I can say that has helped me immensely in the past and I've used it everywhere for the last few years.

 

[kr2p]

Sorry I should have specified, this isn't a diy project but a repair I'm trying to make to a 1073lb preamp! So issues such as shielded wire and parts positioning, and even power supply problems, won't apply in this case. I just came across this thread in my googling and was surprised at the similarities between your problem and mine, and I know that these diy forums are filled with people who really know their stuff!

That's interesting that it never came back, but as you said before having something work and not know why can be worrying depending on how critical the item is. I haven't been able to find any bad parts or solder joints either, and tapping around doesn't make anything noticeably better or worse.

I believe that the issue must be to do with the second gain stage. My reasoning is that the noise varies with the gain, so the problem must be before the gain stages, and the noise is there at lowest gain levels where the first gain stage is disengaged, so the problem must be in the second gain stage. Finally, the crackling noise drops considerably at the very highest (70-80db) gains, so it can't be an issue "upstream" of the first gain stage otherwise it would get louder and louder.

I'm having no luck at all seeing any solder issues or cracks though, so I've ordered some freeze spray to see if it flags any of the components as broken. I'll also have access to another one of these units later this week so I can try swapping out the preamp pcb to confirm that the problem is there. If none of these things confirm my suspicions though I'll have to go back to the drawing board.