KT150 Tube Matching


Ok. Educate me. I don't know enough to even be dangerous. What are the critical factors used when matching a set of KT150 tubes, say, to go in an Audio Research amp? Transconductance? Bias current draw? A combination of the two? Is there any spec. or threshold on the variations? Does Audio Research do anything special to match their replacement tube sets? 
imgoodwithtools
When shopping for power tubes (NOT NOS), I look for dealers with higher quality and larger supply of tubes for a better match. Upscale states they pay extra for tubes with higher micromhos.  ARC is 2x price of Upscale ... don't think you can go wrong with either. 


 Tubes need to be burned in a bit before matching to get a close match. Otherwise they can drift apart as they burn in leading to a poor match. I believe Upscale does this. I know Jim McShane does. The last 2 quads of KT120s I got from him were dead on. Next to no drift. I did not have to adjust the bias again after the first check. 


I would not be so worried about getting new output tubes that supposedly come closely matched. When you set the bias for the tubes in your amp, you are in a sense "matching" the output for all the tubes yourself. Autobias amps do it for you automatically. This makes the output of each tube very close to equal. I would be more worried about getting a good match on the input or driver tubes for the amp, which make more of a difference in balance. Those tubes are not usually in a circuit that is automatically or manually biased. Also, as Lostbears stated above, brand new tubes need to be burned in for a good while before you can get a stable reading on how closely matched their output is. Some of the big tube dealers mentioned above charge a LOT of extra money for "matching" of new tubes. I have personally tested quite a few pairs of "platinum matched" tubes right out of the box, on my Amplitrex AT1000, which were off by as much as 35% on their GM readings! Needless to say, they got sent back for a refund immediately. I later ordered the same tubes new from a China dealer on E-Bay for half the price, and they tested within 5% of each other and very strong when they arrived.
Audiophiles don't need to be so fussy about tubes who's readings "match" so closely. Your ears will never hear the difference between a pair of tubes that match within 3% of each other or a pair that are off by 10%, especially when they both test very strong.


Most Audio Research amps and I am sure others, run the tubes in pairs. There is one driver tube for each pair. While you can check the bias of each tube, you bias the pair with one pot. So you want closely matched pairs of tubes. You don't want to have one tube bias reading really low while the other is high.

 You need to check the bias often when you replace power tubes. The bias will drift as they burn in. You don't want a tube running really hot for a long time. They will settle down as they burn in. This is a lot less of an issue if you get power tubes that were burned in a bit before matching. They won't drift nearly as much.

 I don't know about paying extra for platinum matching and such. I just email Jim McShane. Tell him I need two quads of the power tube I want. I also tell him what amp they are for. His tubes are fairly priced and I have never been disappointed Yet. I have been buying tubes from him for nearly 20 years now.




 
I recently purchased a "matched" quad of KT150 tubes from a well known, seemingly well respected tube vendor. My goal is not to bash him, but to educate myself. One of the "matched" tubes has a transconductance 20% under the average of the other three tubes. When I called to ask about that, I was told transconductance is not a critical factor when matching these tubes for an AR amp. That bias current is the critical factor for tube matching in this particular situation. True? False?