Power Tube Failure

@inagroove 

@brownred6 

"I’ve been told by tube experts most tube testers can not match tubes. They do not supply actual high DC voltage needed to the plate of the tube and thus can not measure actual plate current."

The above statement is absolutely true. I've been told by a very reputable tech that is working on a piece of equipment I have who said exactly the same thing and pointed at a tube tester under his workbench and said he may have only used it five times in the last several decades. I believe he prefers to test his tubes in circuit.

@faustuss @brownred6 @jea48 

Thanks for all of the insight...

OK, it seems that a basic good/bad tester will get me to first-base, providing % Conductance and Leakage, but a tester with "Mutual Conductance" is needed to access the quality of dual-triode tubes (I have a dozen12AX7s hanging around).

A. Did I get that correct?

B. If yes, I wonder why do so many NOS sellers stress the need for "Matched Pair" tubes (L/R) if Mutual Conductance is more relevant than matched pairs?

C. Does the need for Mutual Conductance testing go away if a Amp/Preamp is self-biasing?

Thanks in advance for any insight and clarity.

Best,

I have a modern MaxiPreamp II which is indeed very good for assessing & matching small-signal (preamp, input, driver) tube types. There is nothing to calibrate and no tubes in it to replace. You can easily match your tubes based on its transconductance (gm) and gain readings. Generally try to get all triodes / elements within 10%. Weak / worn tubes are easily spotted by low gm readings. I would note that modern Russian tubes, when new, often measure much stronger than even "tests strong" or "tests NOS" vintage tubes of the same type. Once I accidentally severe;y abused a pair of Siemens E88CC over a ~20 hour period (running them at 12V heater instead of 6.3V lol), and the gm readings went down 25% from before (still good range, but they were super strong before this) - yep that tracks lol. 

Also have their MaxiMatcher (specifically for power tubes), and it’s super for matching or re-matching sets, and spotting tubes that are getting worn down. It can identify shorted tubes (never put these in your amp). It applies close to "real world" DC plate voltages up to 400V. That said, these modern Russian or Chinese tubes always have an element of risk. Earlier I mentioned receiving a bad octet of Mullard KT88 from Viva Tubes - put them in the tester today and they actually measure great! Strong gm, mA in range, readings pretty steady, no shorts. Unfortunately it’s just one of those edge cases where they measure good but I can’t use ’em. I do think Upscale does a better job screening new tubes with their burn-in process. Still have yet to get any power tube from them that was a problem. 

I love vintage gear, but didn’t want to mess with vintage testers. I wanted something that was consistent and just worked without fuss. I have enough tubes here to justify the spend. I don’t really see the point in an emissions-only tester - you cannot use those to match tubes. 

@inagroove 

"A little while ago a small tube (in my Phono-Preamp) died.  It was a subtle affair in which the tube turned white and the glass cracked, not necessarily in that order.  This failure was interesting, but not dangerous.

Yet, this event got me wondering - what happens when a power tube dies?"

I know that the 6550 is usually used as a power tube in push pull amplifiers but in the case of your 6SE it is employed as a voltage regulator with 6H30 as stated in AR’s literature. It serves two functions, slowing the inrush of current to the heaters in the six 6H30small signal tubes in the gain stage as well as providing a constant voltage to them. Anyone participating in this thread worth his salt would have corrected you on that.

@jea48 

"The 6922 small signal tube must be biased to ARC specs."

Generally, small signal tubes in a gain stage are not biased since their cathodes are connected to ground via a resistor that is bypassed by a capacitor.

"Not only the pairs matched but more importantly each section of the two triodes of each tube must be closely, tightly, matched. Especially the driver tube. If the tubes are not closely, tightly, match you can spend hours moving the 6922 around and still not bias the tubes to specs."

@inagroove should pay special note to your statement about pair matching since AR would have followed the most stringent protocols regarding this before they ever shipped the 6SE. There is no "driver tube" in a gain stage they are used to provide the voltage swing required in a class AB power amp stage. Regarding moving tubes around on a chassis, it is general practice to label each tube’s socket location upon installation so there is no confusion about where "matched" tubes belong in the circuit which in almost all instances is critical.

Um, @mulveling 

"Once I accidentally severe;y abused a pair of Siemens E88CC over a ~20 hour period (running them at 12V heater instead of 6.3V lol), and the gm readings went down 25% from before (still good range, but they were super strong before this) - yep that tracks lol."

the 12 in 12AX7 denotes it’s maximum heater voltage though it can be successfully used in a circuit at 6.3 volts.

"tubes were noisy on startup (which is NOT normal)"

Tubes that have not been powered up for an extended period of time will often make sporadic noises since all the various elements that make up the components inside the tube have a tendency to move around in the vacuum on an atomic level and are quickly scavenged up by gettering.

"Over the years I migrated over to (much more expensive) VAC power amps. I’m impressed by their protection circuitry"

Interestingly, tube equipment has been manufactured for well over a hundred years and much of it still survives today in perfect working order with out to much advanced "protection circuitry".

@ghdprentice 

"A friend of mine had a problem with a tube amp from Rogue within the warrantee period and they would not fix it because it was tube related... so I will not buy or recommend Rogue products."

Did your friend leave out that he was constantly tube rolling? Rogue products are designed to a price point, often mounting all components including tubes sockets unsupported on a single main board as @mulveling mentioned which leaves solder pads and circuit traces susceptible to damage from constantly pulling and reinserting tubes. Not only that but cross-referencing tube types that we’re originally not intended for a given circuit topology will certainly result in component failure down the road. Generally, equipment manufactures don’t warrant against such shenanigans and since they know their products they can’t be hidden.