Most desirable vacuum tubes for SET amps

 What do you think about OTL SET amps/integrateds, and the most desirable triode tube in that instance?  (I have a Vinnie Rossi L2iSE).

When you eliminate the output transformer, you eliminate with it the reason to have an SET. Its the way the SET makes distortion that makes them interesting to listen to; at low power their distortion is so low as to be unmeasurable- and this is the source of their 'inner detail'.

One argument that has been made for the last 30 years is that SETs sound better since there is no 'phase splitter'. That argument is hard to prove out in practice since there are so many variables involved! How many people have heard a push pull amplifier using exactly the same tube complement as their SET?? One thing I can state here is that they certainly won't be able to make that comparison if the amp is transformerless! DHTs have far too high a plate resistance value to be successfully used in an OTL, unless a number are used. We made a 300b-based OTL about 25 years ago that used four 6300b graphite-plate power tubes to make about 7 watts. Four 6AS7Gs (which are indirectly heated triodes) did a better job.

Recently and just for fun I've been designing a push-pull ultralinear tube amp that makes 5 watts/channel (ostensibly for my bedroom system, although it would make a good desktop or headphone amp too). Often when SETs are compared to push-pull amps the P-P amp makes a lot more power. I am interested to see what happens when the power levels are the same.

Telefunken AD1

Definitely the "Thoriated Tungsten Filament" family as 211, 811A, 845, GM70.
If you are after impact, clarity, and soundstage more than after a lush sound and coloration there is no escape from these.

Jan

Interesting comment, maybe why so many like the sound of the Elrog 300B which feature the tungsten filament.

+1 @atmasphere Thanks for your insightful summary. It does seem that the lowly 1.5 watt 45 tube is becoming increasingly popular.

I do have a question for you, though. When you talk about the distortion that SETs have, you also state that if they are listened to below their distortion threshold (not exactly how you put it) they are very linear. Am I correct to understand from your comments that a SET listened to at very low volumes is not distorting? That would explain my experience, which I appreciate is limited... I listen typically in the near field well below 80dB. My speakers are 94dB efficient, so I *think* I'm typically using a fraction of a watt. I know that drum hits and crescendos can require the amp to produce instantaneous peaks far above that... But I don't seem to hear much or any distortion in my system.