Bonjour,

somewhat sceptical about the gorgeous measurement results that Serge showed with his KT88 Power-SRPP, I have breadboarded a "mini-version" using trioded PCL82, which proved the concept right on the spot (the thread is over at another board, http://geek.scorpiorising.ca/GeeK_ZonE/index.php?topic=2695.msg31291#msg31291).

I made some further experiments using tubes I expected to be suitable, notably EL86, not only in triode mode, but also using/duplicating the original Philips pentode SEPP circuit that can be found in the EL86 spec sheet. Also some other tube were tried that seemed suitable, like PCL805. But I never managed to obtain such fine measurements as with the trioded PCL82 pentode section again (or like Serge with his KT88 version).

So, I am wondering what (which parameters) makes a tube especially suited for a "Power-SRPP" stage? I have asked at other boards, but nobody seems to know - so far it is only guesswork

Maybe someone has an idea and some traceable explanation/substantiation coming along

Thank you,

Tom

Hi Tom,

Sadly, not yet

You may have noted another thread on experimenting an SEPP using that silly 6080.

Perhaps results obtained with this configuration could help

Yves.

It's just a guess:

des problèmes de capacité entre les électrodes ( les tubes qui suportent des courants plus forts, ayant des élements de plus grande taille, sont peut être moins aproprié).

je crois que les montages srpp étaient utilisés à la base dans des montages HF.

hervé.

Bonsoir Yves,

For SRPP somehow I presume comparably large possible voltage swing of the tube being used to be some sort of key to success. At least that is what my empirical experiments suggest: Go primarily for voltage swing, not current swing. This also fits in the context of what SRPP was invented and used for, I think.

SEPP (like with the 6080/6AS7 circuit in another thread) might be another story, though.

But I think that low mu tubes (like 6080) are contradicting the intentional working of the SRPP circuit, since so much power is wasted to generate the bias voltage across cathode resistors

Tom

Mmmmh, just compare triode strapped data you published for ECL82 and EL86

I can't see notable difference except for plate dissipation

Yves.

Ola

du bist auf das?




BOB

Bonsoir Yves,



The published EL86 empirical data was taken using pentode mode, which allows for larger plate swing (much closer to 0V) than triode mode at same B+, so it is not directly comparable in the sense of my "theory" (talking about triode mode straight SRPP).

Squeezing the last out of EL86, there is a Loftin-White pentode mode design out, running at 550V, claiming 8W Po, which would be about double the Po and plate efficiency than the original Philips pentode SRPP EL86 delivers on the bench (not only on paper). I have not checked yet if the claims hold true. I can post a link to the article (German language), if of interest.

Leaving the grounds of straight triode(d) SRPP completely altogether, there is the Murray amplifier, using PL36, gaining about 18W Po at considerably less than 1% THD, or something: http://www.wdehaan.demon.nl/tubeamps/murray/index.html Please note that would be in the range of double the plate efficiency of Serge´s KT88 SRPP at comparable Po, and still allows cheap toroids to be used.

But such beasts are very far and different from what I intended with my initial question about suitability of triode(d) tubes for straight SRPP action

Tom

Hi Tom,



Thanks a lot for the link to this very intersting schemo
At first, the phase spilitter looks like a bootstraped paraphase

Fixed bias and "only" 510V B+

I intuitively supposed that those sweep tubes was good candidates due to they large current capacities at lo plate voltages

Yves.

Bonjour Yves,

more on the Murray can befound here: http://www.tubecad.com/2004/blog0024.htm

Tom