Amp blew fuse, now hummmmms. Bad caps?

[jeff]

I'm working on this amp and had it up and running. The fuse blew. This amp has two fuses. It wasn't the mains fuse that blew but a fuse between the first series filter caps after rectifer and the next filter cap to OT, screens and so on..(where the standby switch would be if this amp had one) so I inspected the amp then replaced the fuse. Once replaced, the amp started to make a loud hum. I am also noticing a drop in B+. I suspect the caps need to be replaced. I tried an experiment, to see if I had blown a tube. First I removed the two outer power tubes, no hum, seemed to be working, rattled reverb tank and got sound. Then I replaced the two outer tubes I just removed and removed the two inner tubes, still no hum and heard reverb. But... with all four tubes in, the hum returns. I'm thinking maybe one of the filter caps may be going bad and with all four tubes in it's not doing it's job, but by removing half the load it puts the caps under less stress and it can handle it??? This amp is from the '80s and hasn't been used in a while.


Does this sound like it's the case? Just thought I'd ask before ordering new caps.


Interesting.....Conditions right before fuse blew:


 The amp was running fine for hour here hour there. I never fully cranked it, but at moderate volumes there was no problems. I got the idea to bypass the preamp by plugging my guitar directly into the efx loop return. The sound was lower but still sounded good, no problems. Then I got the idea to see what it sounded like with no preamp, but with a distortion pedal as the preamp. So I unplugged my cable from my guitar and plugged it into the output of my pedal. As soon as I plugged a cord into the pedal's input the fuse blew.  Does this suggest anything? Coincidence maybe? Jolt from battery turning on as input was connected? Any ideas beside bad cap?


Thanks
Jeff

[HotBluePlates]

It could be bad filter caps, but I'd want to verify that with resistance checks of the caps (a good cap should show a brief low resistance, then climb to ∞Ω).

I'd be much more cautious of snug-fitting socket contacts for the output tubes, and verifying the tubes always have bias without runaway.  I'd more likely suspect bad output tubes, or loss of bias, especially if this is not a 40 year old amp.

[jeff]

OK I'll check


As far as bias this is the amp we were talking about in another thread


New tubes
Plate 34.5mA
Screen 4.7mA
Voltage 438V
 
Would bias or socket connection cause hum?

[jeff]

Reading slow climb to 135K, probally the two 56K bleeder resistors. Should I take caps out of the circuit for testing?

[HotBluePlates]

Reading slow climb to 135K, probally the two 56K bleeder resistors. Should I take caps out of the circuit for testing?

Yes, you might have to disconnect the cap from the external circuit when bleeders are involved.  Usually unsoldering one side of the cap is sufficient (unless the bleeders are physically attached to the cap being tested).

[HotBluePlates]

Shortcut:  Pull the output tubes from the amp, power up.  Still humming & blowing fuses?  If not, probably output tube related (loss of bias, loose socket, or defective tube).

[jeff]

ok thanks


Tubes out, no hum.


 I may not have given tubes enough time to heat all the way up when doing swap test. Just tried again. Two tubes seem ok but when I tried the other two it was quiet for a while then hum slowly got louder and it looked like one of the tubes was glowing bright Blue. I shut down immediately, plate didn't look red but I shut it off as soon as I saw one tube bright blue and the other was normal


What does that blue glow on one of the two tubes indicate to you?
 

[HotBluePlates]

What does that blue glow on one of the two tubes indicate to you?

If the glow is outside the plate structure, it's probably nothing to worry about.

If the glow is inside the plate, between it and the cathode, the tube is gassy and should be thrown out.  This type of glow could be blue, pink or lavender colored.

The tubes in this thread have harmless glow outside the plate (actually from the glass), are not "gassy" and shouldn't cause any problems.

You still haven't mentioned if the sockets are tight, or if the tubes sit loosely in them.

[jeff]

Sorry I didn't get right back to you, had to go
sockets feel tight, fair amount of wiggling getting tubes in and out.


I'm just so bummed, man. I don't get it. Everything seemed to be working just fine til I plugged my pedal into the efx return then plugged my guitar cord into the pedal. Can't figure out why that would have caused a problem. That can't be coincidence, right?
 


Feeling defeated and depressed 

[John]

I can't imagine this is it, but take a good look at your EF jacks... grounding, signal wires,everything. Maybe when you plugged it it wiggled a dodgy solder joint/broke a wire/loosened a ground?

[jeff]

Thanks, will re-inspect sockets and solder joints


 Interesting... hooked a volt meter to one end of a guitar cord plugged into the output of a pedal and when I plugged a cord into the input meter jumped up to .2V. I tried two more commercially made pedals and got 2V!!! Never thought about this before but I get why. So maybe with my pedal plugged into the return jack, by plugging a cord into the input, I sent a voltage to the amp.

[shooter]

got 2V

AC or DC?

[jeff]

D.C., Pedal runs off a 9V battery

[sluckey]

The amp was running fine for hour here hour there. I never fully cranked it, but at moderate volumes there was no problems. I got the idea to bypass the preamp by plugging my guitar directly into the efx loop return. The sound was lower but still sounded good, no problems. Then I got the idea to see what it sounded like with no preamp, but with a distortion pedal as the preamp. So I unplugged my cable from my guitar and plugged it into the output of my pedal. As soon as I plugged a cord into the pedal's input the fuse blew.  Does this suggest anything? Coincidence maybe? Jolt from battery turning on as input was connected? Any ideas beside bad cap?

This is a red herring! You're chasing the rabbit down the wrong rabbit hole. This did not cause your fuse to blow.

[jeff]

Ok, just seems like too much of a coincidence that a brand new tube went bad at the exact second I plugged my pedal in, just making sure the two events are not connected. I don't want to buy a new set and have this happen again because every time one tube blows I need to buy four new ones. But the tube is bad, or as least no longer matched.


Here's what I did:
Label tubes 1,2,3, and 4
only use two inner sockets
Tube 1 and 2, no prob
Tube 2 and 3, no prob
Tube 2 and 4, Problem with tube 4!
Swapped tubes 2 and 4 in inner sockets L to R, R to L, problem with tube 4!


So by only using the inner two sockets and swapping tubes I feel I've isolated the problem to a bad tube. Or at least, somehow, this tube is no longer matched to the other three. In other words at the same bias voltage this tube is misbehaving.


Could this be the case?

Does this happen often? Buy a new set of tubes and one just 'goes bad' with under 5 hours of play time.

[sluckey]

So by only using the inner two sockets and swapping tubes I feel I've isolated the problem to a bad tube. Or at least, somehow, this tube is no longer matched to the other three. In other words at the same bias voltage this tube is misbehaving.

Could this be the case?

Does this happen often? Buy a new set of tubes and one just 'goes bad' with under 5 hours of play time.

Maybe it's just a bad tube. But who knows what else may be wrong with that amp. I'm guessing this is the same amp that had the charred pcb? And the bias problems? If so, let us know rather that continue starting new threads every time you discover a different problem. They could all very well be related. We need full disclosure.

[Sonny ReVerb]

I don't want to buy a new set and have this happen again because every time one tube blows I need to buy four new ones. But the tube is bad, or as least no longer matched.

Usually, if you bought a matched set, the vendor can send you a replacement to match the plate current of the set. Worst case, I would buy a matched pair and replace the bad one and its inner or outer socket mate.

So by only using the inner two sockets and swapping tubes I feel I've isolated the problem to a bad tube. Or at least, somehow, this tube is no longer matched to the other three. In other words at the same bias voltage this tube is misbehaving.


Could this be the case?

Does this happen often? Buy a new set of tubes and one just 'goes bad' with under 5 hours of play time.

I think tubes are like most electronic devices - under normal use, they either fail in the first few hours or work fine for decades. (Screw those extended warranties!)

[Matching can of worms ahead...] AFAIK matched tubes are only matched at a specific operating point at a specific point in time. They may not be matched at other operating points or after a period of wear.

[jeff]

NEW TEST:


When I did my previous tests I did not adjust the bias, left it where it was and looked for blue/hum.


This time I turned bias max negative installed bad tube on one side known good on other. When I turned it on, no blue, no hum. I slowly adjusted bias to pull 34mA, still no blue, no hum. I had to adjuts bias alot more negative than where it was first set. Played through it(strummed strings 2 or 3 times) and got sound. Is it possible that when they tested these tubes they were a matched set and after about 5 total hours playing one of the tubes 'drifted'. In other words -34V bias made all 4 tubes draw about 34mA but now this tube 'drifted' and needs more negative bias to maintain 34mA idle current. I've heard of a 'burn in' period were you check bias then recheck later to see if it settles in and needs readjustment, is it possible all four tubes were happy at -34V at first and just this one tube slowly drifted away from being matched? It seems to be behaving with more -bias on it than the other tubes need. By that I mean no hum/bright blue


Just trying to totally understand what happened before buying new tubes and putting them in.

[sluckey]

Everything you just said is possible. Ask your supplier for details on how the tubes were tested and matched rather than ask us to guess.

[HotBluePlates]

... I've heard of a 'burn in' period were you check bias then recheck later to see if it settles in and needs readjustment, is it possible ... this one tube slowly drifted away from being matched? ...

Yes, it's possible.

Tubes drift.  They drift most when they're new, usually the drift (eventually) settles down to less-drift.  But they will keep drifting all their useful life.

This is a big part of why I don't worry too much about biasing, other than to see the tubes aren't melting.  I built an amp with external bias test jacks, and 2 locking pots I can adjust from outside the amp.  But I haven't even bothered to check the idle current for that amp for the past ~4 years.  The amp plays fine and sounds good, I don't notice tubes melting down, so I'm happy.

[jeff]

Everything you just said is possible. Ask your supplier for details on how the tubes were tested and matched rather than ask us to guess.

I think I may have taken this post the wrong way at first.


I have some experience working on amps where everything just goes right. I don't have a lot of experince with things going wrong, and something has gone wrong. So my questions are aimed towards you guys who do have a lot more experience than I do. I was kind of asking more along the lines of 'Has anyone else got a quad where one tube drifted?' and not 'Hey guys what's wrong with MY amp and MY tube?'. Sorry


To answer your question(EDIT:ooops that wasn't a question), I called company to see if I could get one replacement tube. Box was marked 18/4.3 but the guy said that was their old matching system so they can't send me a tube to match the old numbers. He doesn't even know what those numbers mean. He says they've switched to the APEX system of matching, a better matching system, he says, but couldn't explain the testing process or what was being measured for.