Run away bias on my right front KT88 power tube, VTA120

Hi all, my I did my monthly bias check today and all seemed well for the 1st 3 power tubes...set all to .500v on my vta120kit, but then my right front tube was reading 24.00v!!! I couldn't bring it below 16.00v...it was running great for the past 3 months since its build, never an issue.

I changed over all KT88 tubes, but the problem still exists.  Where do I begin tracing down the issue?

Thanks!

With the amp off, check the resistance at that bias test point to ground. Should read 10 ohms. If it doesn't, that resistor is likely bad.

Bad resistor it is...thanks!

I've dealt with this scenario twice in 18 months. As explained by Bob L. and audiobill, this so-called 10 ohm "bias-set" resistor is deliberately designed as a weak link so as to provide an Early Warning System for {usually} output tube failure. When the tube becomes senile, it becomes mentally unstable, and likely to raid the fridge - i.e. pull excess bias and/or plate current, causing this "fuse" to overheat. The amp may function again when the resistor is replaced; indeed, in my case could be operated safely even with the burnt-out resistor in place BUT bias could not be monitored. Best to address the underlying cause ASAP.
And unless there is reason to believe that an isolated KT88 is bad, many of us think replacement of the quad is in order.

peterh wrote:

vermonterbpa wrote:Bad resistor it is...thanks!

The tube located at the failed resistor should be considered dead and should not
be used again.  It's the tube that caused the resistor to overheat and brake
not the other way around.

*grumble* *growl* *mumble mumble*

I guess this is the way of it.

There are some fundamentally different perspectives on tube replacement here and this is the one that I will admit right up front is entirely contradictory to my own. I guess I am spoiled in owning a well-calibrated tube-tester that will give me sufficient -quality- information on a tube such that I know whether I can trust it or not. And as I am 'used' to tubes that are reliable for many thousands of hours as a basic expectation, and can tolerate a certain amount of abuse without failure - clearly, I am living in a fool's paradise such that I would look towards installing a heavier resistor than a failed tube. As, of course, the tube would be tested prior to replacement.

A Jim McShane matched-quad of KT88s run from $224.50 per his website. And as a reliable purveyor, one may be sure of what one is getting from that source. So, under one perspective, an incident as noted above would require the replacement of four (4) tubes based on one not-certain failure.

Now, given the insane costs of reliable tube testers (At Kutztown, recently, two Hickok 539Cs sold at $1,600 and $1,400 respectively, and a 539B sold for $1,200. Sold, not asking), one might replace a number of quads and still not reach those numbers. Note that I am citing only those devices capable of matching, albeit with additional outboard instrumentation. But, a decent MC tester may be had in the $150-$200 range (purchase such a tester only hands-on, not via eBay unless you have infinite patience and know exactly what you are doing), which will be good enough as a go/no-go device. And with individual bias capability, matching is less critical (less, not 'not').

In my house, there are approximately 28 fully operational tube-based devices dating from ~1919 to ~1972. And I have been doing this for 40+ years. So, perhaps having a tube tester is not so bad a thing in my case. But the potential waste I am seeing advocated here does bother me a bit.

I should have been more explicit; what Peter H describes is what I merely implied. This is why one should always buy new tubes from a Jim McShane. Then at least you have a fighting chance, and if, in his wisdom, he determines that a faulty tube is the culprit, he'll supply a new one. Short of arming oneself with an array of tube-testing gear AND the expertise to use it wisely, there is no other sane approach.
The other variable which should be addressed is our old bugbear Socket Maintenance. Intermittent bad contact might also cause a 10 ohm resistor to fail. But, again, chances are fairly good that said bad behavior will take the tube, too. Depends on the zodiac sign the moon is traversing, probably.

Beyond my limited mechanical skills, but has anyone given thought to putting an above-board slow-blow fuse in line with the bias resistor? Not that the 10-ohm resistors are bank-account busters, but they're a PITA to get to, and you'd have a quick visual confirmation of where the problem lieth.
Also, on the subject of new tubes for old . . . IIRC, no less than Jim McShane posted here that even after all the torture testing and close-matching he does before shipping a matched pair or quad, the values used to match will change over time. Jim, correct me if I'm wrong here, but it seems a bit wasteful to toss out three perfectly functioning pentode or beam triode output tubes because the fourth got fried.
It's also a confusing thing to use the same word, "matched," which means vastly different things. In the case of output tubes, you're matching one tube against another tube. Means a whole 'nother thing with regard to the commonly used "dual triode" signal tubes applied in driver slots and pre-amps. In this latter case, a 6XX7 or 12XX7 is actually two tubes inside a single glass bottle. The "matching" in this case is within the actual tube -- i.e., are the people upstairs dancing to the same music and volume as the people downstairs? The matching is internal, not against another tube.
I think . . .

deepee99 wrote:Beyond my limited mechanical skills, but has anyone given thought to putting an above-board slow-blow fuse in line with the bias resistor? Not that the 10-ohm resistors are bank-account busters, but they're a PITA to get to, and you'd have a quick visual confirmation of where the problem lieth.

When I planned out my ST120 I was learning a lot from Roger Modjeski of Music Reference / RAM Labs. He recommended fusing the cathodes with "high breaking ceramic fuses."

"When a fuse interrupts in a DC circuit (tube fuse) the interrupting causes the element to turn to plasma which continues to conduct long after it should. This extended time causes damage to the circuit. Ceramic fuses are often filled with sand which helps prevent the plasma from forming."

While his Music Reference forum has been closed and archived at audiocircle, it's still available to read. I learned a lot just by reading Roger's posts.
http://www.audiocircle.com/index.php?action=profile;u=39004;area=showposts;start=100


When one of my 6550 tubes failed on start-up I only had to replace the tube and the fuse without disconnecting and cracking open the whole amp.

Ironically this just happened again as I was in the middle of writing this. I was curious to measure my bias resistors so I pulled the tubes and found 3 of the 4 measured correctly but the last one was open. I looked under the hood and found nothing wrong with the resistor. I found the fuse had blown which was surprising as everything had been running fine the last time I used the amp just a couple nights ago. So I replaced the fuse and the tubes and fired it up and sure enough that corresponding tube was DOA. It didn't even light up.

What worries me is the fact that both tubes I've lost so far were in the same socket. Makes wonder about the socket. As I said the bias resistor measured as it should so I know that's not the problem.

I've been thinking about getting a new set of tubes anyway.

deepee99 wrote:Jim, correct me if I'm wrong here, but it seems a bit wasteful to toss out three perfectly functioning pentode or beam triode output tubes because the fourth got fried.
It's also a confusing thing to use the same word, "matched," which means vastly different things. In the case of output tubes, you're matching one tube against another tube. Means a whole 'nother thing with regard to the commonly used "dual triode" signal tubes applied in driver slots and pre-amps. In this latter case, a 6XX7 or 12XX7 is actually two tubes inside a single glass bottle. The "matching" in this case is within the actual tube -- i.e., are the people upstairs dancing to the same music and volume as the people downstairs? The matching is internal, not against another tube.
I think . . .

I agree regarding the output tubes. As long as the three left are matched well enough to work in the amp (or the amp does not require matched tubes) then being able to pick up one to match the remaining three makes use of the life left and is a nice money saver to boot.

On the dual triode matching - the tubes can be matched internally - triode to triode - and/or they can be matched to another tube which means that all four triodes in the pair of tubes have to match.

Jim McShane -  I have a set of KT120s that I purchased from you a few years ago.  They have performed very well but lately one of the KT120s seems to have drifting bias.   I'll set it for 50mA and a few days later is either 10mA above or below this point.  The socket pins were tightened with no help.  The other three tubes seems to hold bias well.  

Is this particular tube wearing out and/or on its way to failure perhaps?

arledgsc wrote:Jim McShane -  I have a set of KT120s that I purchased from you a few years ago.  They have performed very well but lately one of the KT120s seems to have drifting bias.   I'll set it for 50mA and a few days later is either 10mA above or below this point.  The socket pins were tightened with no help.  The other three tubes seems to hold bias well.  

Is this particular tube wearing out and/or on its way to failure perhaps?

See if the problem follows the tube; swap it with one that is behaving well in that same quad, then keep an eye on their respective biases. That will tell you if something "under the hood" needs attention or, I bet, that that one tube is ready for social security. If it's the tube going south, by all means replace it before it develops a short and damages other stuff.

deepee99 wrote:See if the problem follows the tube; swap it with one that is behaving well in that same quad, then keep an eye on their respective biases. That will tell you if something "under the hood" needs attention or, I bet, that that one tube is ready for social security. If it's the tube going south, by all means replace it before it develops a short and damages other stuff.

That's excellent advice! And be sure the sockets are clean and making good contact with the tube pins.

Thanks for your help... I finally swapped the suspect drifting bias KT120 from the front right to the rear right position. After warm up bias setup held on all tubes around their 50mA settings. Could possibly be a socket/ pin issue but took awhile to show up originally.

I do have a lot of hours on this set of KT120s. And I'll keep an eye on it and check bias daily.