A different Kind of Power Tube Arcing

[apehead]

Wow, I am a posting fool today! The Quad Reverb involved in the "Four Terminal Pot" thread is having (possibly) another problem. After being fixed for about two days, the owner (my roommate) informed me that it had failed again. He says a loud pop and a bright flash occurred, after which he shut it off. After trying to energize it again, he says it made a loud buzzing noise. Taking a look today, I could see no burnt components, but a stupid wiring mistake I had made (in regards to the "Four Terminal Pot" thread ). Fixing the wiring mistake, I replaced the tubes, and plugged the amp in. Energize. I saw nothing happen in the interior of the amp, but there was certainly a loud, ugly buzz. My roommate, who was looking at the tubes yelled "Whoa!" and jumped back. After calming down, he informed me that one of the tubes was arcing within the envelope! I simply replaced that tube with another good tube, and the amp seems to be running fine (I observed it in the energized state for about 10 minutes before shutting it back off). It seems to me that this seems a bit more than "coincidental" that this amp's power tubes keep catastrophically failing! Does anyone have any suggestions as to what the cause could be? I have checked plate and screen voltages, and they are a bit high, but only by about 20v or so.

[stokes]

Check for bad solder joints in the bias circuit,you could be losing bias voltage intermittently.

[apehead]

Losing Bias would certainly be (very) bad, but would it actually damage (or destroy) the tube internally?

[Satamax]

Yep melt them, and finaly take the PT with it after a while.

You seem to have bought crappy tubes as i see it and looking at your other thread regarding the twin. Anything comming from New sensor by any chance?

[apehead]

Nope. They are JJ 6L6GC's from Tubedepot.

[Satamax]

Nope. They are JJ 6L6GC's from Tubedepot.

Yep, tubedepot is suposed to be good. Sorry to say this, but may be it's time to see a real tech and see if your electronics skills are up to par, regarding all tthe isues you've had lately. Nothing wrong with this. It's teh best way to learn i'd think. I think that's what Bruce was trying to say in another thread, but not as bluntly as i'm puting it

[apehead]

What is a real tech? I've never met one. I wouldn't trust my toaster to most "amp techs" I've run into. I don't see how this is a reflection on my skills as a technician. The amp has a strange problem, I am asking for advice on what could cause those kinds of problems. I attempt to lay out my troubleshooting train of thought as thoroughly as possible, and no one has told me "you are going in the total wrong direction here" yet. In fact I thought the bias circuit was trouble initially (refer to "Four Terminal Pot?"). And it seems after all the tube swapping, we've returned to the bias circuit as the likely culprit for all this mess. So I think it's unfair to just dismiss this problem as over my head, I just don't have much experience with guitar amps. I work on tube LORAN transmitters for a living, so if you have a problem with one of those, let me know.

[apehead]

Forgot to mention... the replacement electrolytics in the bias circuit are only 47uF, instead of the original 80uF. Could this be leading to excess ripple in the bias voltage? Could this be a problem?

[Satamax]

Well, don't take it wrong, but if you know tube tech on transmiters, you should be able to troubleshoot without any help, well, jmho. A real tech, true they're hard to come by.

On your 4 terminal pot, i think i wouldn't have bothered and made a normal bias circuit, or bias adjust and bias balance with two pots. I see that you have a tube tester, have you checked your JJ? Btw, have you read this? http://www.geofex.com/ampdbug/ampdebug.htm Still, i know this from experience, recheck every of your solders with a magnifier. And check hiwatt's way of wiring and working with an amp (the sharpie marking of every component checked) Mechanical connection checked in blue, components in red and repairs in purple.
http://mhuss.com/php/pix.php?p=DR103bGutsFS

Your 47µ should be all right, thought could may be induce a bias sag f the cap is not staying charged.

[Satamax]

Another thing, hammer and chopstick method is another good one

[Enzo]

In my opinion, bias is not the issue. If you lose your bias, yes the tubes will red plate and eventually melt. But that is what they will do, as a rule, they won't arc and spark from a lack of bias. They just run REAL hot.

[Rick Erickson]

I wouldn't trust my toaster to most "amp techs" I've run into.

While I can symphathize with your sentiment - this is not the way to win favor with the "real" amp techs out here.

I don't see how this is a reflection on my skills as a technician.

I think it was more of a reflection on your skill as an "amp technician".
I remember hearing stories when I was young about Boeing Engineers who thought they could fix their own TV's - and what a mess they were when they came into the shop.

[apehead]

I don't know how many years of accumulated knowledge is on this forum, but I assume it's a lot. I'm only 21, and think it's a good idea to ask some dudes who have been fiddling with amps for umpteen years for some advice. Ability is not the issue, experience is.

As for Enzo's statement, are you saying that you think this is just a set of bad coincidences with crappy tubes? I thought tubes would cherry plate and melt with too high (towards zero volts) a bias, but wouldn't zero bias voltage cause burning up of the screen resistors as well, due to the massive increase in screen current? If something outside of the tube was grounding the bias, or a bad solder joint was intermittently disconnecting the bias voltage, wouldn't that be detrimental immediately?

Also I did not install that bias adjust, and I purchased some 25K pots to get rid of that mickey mouse job. I've also purchased some 80uF caps, just to be sure .

[apehead]

Well if you're a real one, don't worry, because the statement wasn't directed at you, but I think we can all agree the are a few hacks out there peddling their skills as "amp techs"! Troubleshooting is troubleshooting, whether its a clock radio or an electric range, it all boils down to isolation of the problem and logical thinking. People aren't just born "amp technicians", they learn. I am in the process of learning, and it's usually easier with teachers. So here I am, asking for some advice. Also, the Boeing analogy really doesn't apply in this situation. I am not an aeronautics engineer who thinks I know electronics by default, I am an electronics technician who wants to gain some experience.

[Enzo]

Oh let's all relax. Ape, you may not be directing hack comments at anyone here, but quite a few of us are in fact commercial amp repair techs. And meant or not meant, comments about hacks and rip offs and such when directed at techs in general still can get our hackles up. We freely try to help people like yourself to - when you think about it - take work away from our businesses. It would be nice not to have to see such remarks. That's al.

And leave us not worry about the difference between ability and experience. No one is calling anyone names or dissing their mother. After all, does it matter if one has lots of experience and no ability any more than it matters if someone has lots of ability and no experience? Bottom line is that neither gets the problem solved. Yes yes yes one can learn, but that isn';t the point.

Sometimes we suggest taking a problem to a shop because it might be easier and cheaper than taking months here. Trust me, that exact thing happens. Not saying that a year later the person wouldn't be able to look back and realize he could now do the repair.

I have over 50 years of experience working with and troubleshooting tube circuits. In THIS field - guitar amps (And most pro audio) - you want to line up ten bad amps for troubleshooting, I will go head to head with anyone. I am an experienced field service tech in a couple of industries. I am not bragging, I am just confident in my ability. I have some dim knowledge of LORAN. But if you sat one on my bench, I'd have no idea what power and I/O it needed, what freqs it used, what sort of signals it used (AM, FM, FSK?) Circuits that are as commonplace to you as 6L6 power stages are to me would be completely a mystery. Hand me teh schematics, and I come closer, but still conventions you take for granted are not in my knowledge. Standards like band freqs is something that are basic to you are complete mystery to me. I THINK there are like LORAN A, B, and C or something, yes? Are those freq bands? Circuit update levels? Data protocols? Beats me, hell it might not even matter to my repair. COuld I learn it? Of course. But when facing THAT repair, I don't have the experience. And so, really, it is over my head. I am not afraid to say that. So when someone says a repair might be over someone's head, yours or otherwise, it doesn't mean anything more than that more skill/insight/experience/specific knolwdge, etc. is needed.

OK, end of speech.

Yes, I think you have bad tubes. I don't know that I even need the word coincidence. In my shop, the first sign of sparking tubes, burning screen resistors and such, those tubes are out of there right now. Take a meter to the empty sockets and make sure all the voltages make sense, repair any burnt resistors, recheck for voltages, and slap new tubes in there. Oh, and any arcing on the sockets, those sockets get replaced FIRST.

If I read you right, your screen Rs were shattering. In a loss of bias, excess screen current might flow, but not that hard. The resistors might get hot, and might even be damaged from the heat, but that takes time, more than a second or two certainly.

Let's say the screen shorts to the cathode - or heater - which is ground. Assuming 470 ohms on the screen resistor and 470 volts B+, we have a whole amp flowing through that resistor. Well not really, the power supplt can't provide an amp at that voltage, but it tries. The poor resistor tries to dissipate 470 watts!! Shatters.

Two watt 470 ohm resistors in a typical Fender to hit 2 watts would need to conduct about 65ma, if I figure right. Not so sure how fast it could hit that mark while competing with the plate. And 2 watt resistors don't shatter at 2 watts.

SO while there may be a bias problem, that is why I don't think it is the cause of your resistor explosions.

No, having insufficient filtration on the bias supply won't mess with the tube. All it MIGHT do is leave too much ripple in the bias, which results in hum in the output. Annoying maybe, but harmless.