Pop, fizz, stink, and a new hum on a homegrown tweed tremolux. Looking for ideas.

[wcrankshaw]

The other day while I was playing my tweed tremolux there was a drop on volume, some fizzy noise, and then the distinct smell of burning marshmallow. This was followed by normal volume + a new low frequency hum.

I should also note that for a long while there have been occasional pops and fizzy sounds, but nothing like this.

I've opened up everything, and don't see any visually obvious problems. I understand that an open filter cap could be the cause of the new low freq hum (makes sense to me), and I know that isn't always a visible problem.

Anyone have any good ideas on where/how to start diagnosing without just blindly replacing components? Any help would be awesome.

[The Dude]

The fizz and smell you describe sounds like an e-cap let loose. Check the ends of any axials and tops of any radials for splits and/or oozing of electrolyte.

[wcrankshaw]

The fizz and smell you describe sounds like an e-cap let loose. Check the ends of any axials and tops of any radials for splits and/or oozing of electrolyte.

I'm not seeing any issues like that. From what I've read, the electrolytic a definitely seem to be the main suspect though. There's really no simple way to test the integrity of a capacitor is there?

[Enzo]

Did you pull the "doghouse" cover off the caps up top?


And I wouldn't overlook the possibility of a failed power tube.

[The Dude]

I'm not seeing any issues like that. From what I've read, the electrolytic a definitely seem to be the main suspect though. There's really no simple way to test the integrity of a capacitor is there?

If you don't have a scope or a cap tester, probably the easiest way in your situation is to get a known good cap of similar value, a set of clip leads, and clip the cap across existing ones until the hum goes away. Then, replace the suspect cap. Observe polarity and be careful! There is extremely high voltage on those caps.

[R.G.]

If there is no obvious sign of burning, yes, pull the doghouse and look at the caps. If there is no obvious issue, put the doghouse back.

Now start swapping tubes. Pretty much the first step is to verify that a tube hasn't died. There is a reason they're in sockets.

If swapping with your spare set of known good tubes doesn't help, think for a moment. The next step requires both a high-voltage capable meter and you poking around inside a live amp, with 120Vac running through it and upwards of 400Vdc at places. It's like trying to remove eggs from a rattlesnake nest. If you don't already know that you can do this without killing yourself, or worse yet damaging the amp further, take it to a tech. It ain't worth dying for.

If you have thought your way through this, and are certain you DO know how to do it safely, make sure your estate is in order, and get out your voltmeter. Pound on the internet and find a schematic, preferably with typical voltages. Open up the amp, make sure you won't be disturbed by children, cats, garden hoses and so on, and then proceed to measure the:

- DC voltage at the first filter cap
- **AC** voltage at the first filter cap (this tells you an estimate of ripple)
- DC voltage on each tube plate in the amp.

Write these down and post them. It is very likely that the problem will be obvious from one of these steps.

Again, think about it first. It really, really isn't worth dying for. Buying a whole new amp is cheaper than a funeral.

[Dave H]

If you don't already know that you can do this without killing yourself, or worse yet damaging the amp further, take it to a tech.

I see you have your priorities right

I'd try a known good set of power tubes first. If that doesn't fix it check the power tube screen resistors for open circuit. An open screen resistor would prevent the new tubes fixing the problem when a failed tube really was the original fault.

[überfuzz]

Did you actually de-solder and lifted one lead of each filter capacitor or just looked at these filter caps..? I have had filter caps looking all fine and dandy but when I lifted one leg it turned out one of the leg-cap body connections were broken.

[wcrankshaw]

Did you actually de-solder and lifted one lead of each filter capacitor or just looked at these filter caps..? I have had filter caps looking all fine and dandy but when I lifted one leg it turned out one of the leg-cap body connections were broken.

I have not desoldered any of the leads yet, but I have visually inspected the capacitors with a finer toothed comb. The 25-25 electrolytic cap (I think) in the tremolo section appears to have blown. Obviously, this needs to be replaced, but does that sound like the culprit in general?

[wcrankshaw]

This guy:

Should I be concerned that there is another problem which caused that cap to pop, or look at it as an isolated incident?

[überfuzz]

I have not desoldered any of the leads yet, but I have visually inspected the capacitors with a finer toothed comb. The 25-25 electrolytic cap (I think) in the tremolo section appears to have blown. Obviously, this needs to be replaced, but does that sound like the culprit in general?

First guess is, no, the tremolo circuit is kind of isolated from the rest of the circuit. Hence, that cap shouldn't be able to cause the popping and what have you. However, a dip in voltage can have originated in the tremolo circuit if something vent south there. If luck smiles upon you it's just that cap. Chances are it just blew because of something you have yet to discover. Check the voltages at the filter to see if you're in the right ballpark. It's a good start if you wanna debug.

[wcrankshaw]

If there is no obvious sign of burning, yes, pull the doghouse and look at the caps. If there is no obvious issue, put the doghouse back.

Now start swapping tubes. Pretty much the first step is to verify that a tube hasn't died. There is a reason they're in sockets.

If swapping with your spare set of known good tubes doesn't help, think for a moment. The next step requires both a high-voltage capable meter and you poking around inside a live amp, with 120Vac running through it and upwards of 400Vdc at places. It's like trying to remove eggs from a rattlesnake nest. If you don't already know that you can do this without killing yourself, or worse yet damaging the amp further, take it to a tech. It ain't worth dying for.

If you have thought your way through this, and are certain you DO know how to do it safely, make sure your estate is in order, and get out your voltmeter. Pound on the internet and find a schematic, preferably with typical voltages. Open up the amp, make sure you won't be disturbed by children, cats, garden hoses and so on, and then proceed to measure the:

- DC voltage at the first filter cap
- **AC** voltage at the first filter cap (this tells you an estimate of ripple)
- DC voltage on each tube plate in the amp.

Write these down and post them. It is very likely that the problem will be obvious from one of these steps.

Again, think about it first. It really, really isn't worth dying for. Buying a whole new amp is cheaper than a funeral.

370 VDC at first filter cap
850 VAC at first filter cap
360 VDC at both power tube plates
Low 200s at first preamp tube (swinging because of tremolo).

[überfuzz]

370 VDC at first filter cap
850 VAC at first filter cap
360 VDC at both power tube plates
Low 200s at first preamp tube (swinging because of tremolo).

850 Vac at some cap in the filter..? That is not good! Check your filter in detail.

[wcrankshaw]

850 Vac at some cap in the filter..? That is not good! Check your filter in detail.

I was thinking that might be suspect. What sort of number should I be looking for there?

[R.G.]

850 VAC at first filter cap

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
8.5Vac at the first filter cap is GREAT! 85V at first filter cap is ugly, problem with ripple voltage. 850Vac at first filter cap is either a problem with your meter not doing this well or a disaster with rectifiers.

I think that your meter doesn't measure this properly, as 850Vac there would mean that the rectifiers were completely shorted, and there would be no DC at all.