Leak had a whole design team of engineers, I think they would have spent many months finalising the sound to what it currently has, changing the circuit in the way you describe may alter it to being a far worse performer than it is unmodified.
I think you might be putting just a touch too much subjective wind in these designers sails. These guys didn't really know what they were creating, subjectively speaking in the grand scheme of things.
They were just doing the best they could with what was available. Carbon comp resistors were it, polyester caps were it.
They were actually mainly going for low measured distortion and linearity hence the Ultra Linear OP stage and plenty of global FB. The same goes for the designers of the very first solid state mixing consoles they
honestly thought they would sound much better than their valve predecessors!
History tells us it didn't quite pan out that way. A friend of mine, years ago interviewed the great Rupert Neve and pretty much same story - he wished for quite a while that all those
early SE / choke loaded SS mix consoles would go away so people would embrace his newer and what he genuinely believed were superior sounding and measuring consoles. A lot of the stuff he does now
goes back to and draws heavily from the early vintage era SE designs - because that is peoples preference.
So come on Zen, share this latest supply wonder, is it the 'electronic choke by chance?
No I have never been interested in EC's. I have just observed that most of these typical CLC power supplies ring (oscillate) when I put them in a simulator - some quite badly.
My real life experience on various circuit elements supports this. It's a matter of being aware of it and having the counter measure in your 'tool box' to try.
Years ago I did a mix console upgrade in a studio, replaced all opamps, biased them into class A etc. Had to use more local decoupling caps Result was + step in all areas but it sounded a bit edgy and etched.
The engineer said I had increased the HF response. I said no it's something else, the response is flat - thinking on my feet I proceeded to insert a 5 or 10 ohm R feeding every bypass cap to separate them with a lossy type network and bingo - edge gone. They were ringing at some VHF.
I never denigrate the hard work done by others (and I am thinking of Steve V here) when it comes to the hard yards of trying out many different combinations of components and circuits for the benefit of others; I particularly enjoyed reading the recent experience with the AN resistors, the same material but different outcomes..................
Maybe it's unintentional, however I see quite a degree of denigration - towards more modern designers / designs. A lot of these people have put a huge amount of work into the advancement of audio.
Their intentions are absolutely honorable however full understanding of objective versus subjective can take a lifetime to get a handle on - see R Neve above. Same goes for TDA1540/1, a brilliant Dutch engineers
work. He was doing same thing - just the best he could with the technology at that time. It was a big part fluke that it ended up being subjectively possibly the best sounding DAC ever made.
I suppose I also accept that its all about the circuit and the way various items react with each other and simulations done with the Duncan pwr supply designer etc will not reveal what they sound like, only measure / resonate / bounce.
Duncan's is great for quick and dirty - but it does make some interesting mistakes. Spice is the ticket and much better, but as always its GIGO. I use them as helpers to see a potential issue, to tell me where to
look, where to tweak. That's what they are good for.
For the many who took the leap of faith into the NOS world of the KDAC did we all fry our tweeters with the RF noise, well hell no. I know faith can be blind but perfectly measuring equipment sounds bland and sterile, I have many such pieces gathering dust.
I don't think 0 x OS has ever been a major LOF it's not unfiltered DSD
I'd been meaning to try it for years ever since Kusunoki / Thorsten did it.
In the end this all should be part of a timeline of knowledge gathering - not an absolute serving peoples egos. Too common in this game.
The Leak stereo 20 is not a bad amp, it's not the best Leak by a country mile, messing with the circuit is going to upset it, by all means use the best valves and components but alter the overall circuit at your peril. IMV the correct biasing of the valves will bring more reward than swapping many components, get that right before attempting any more 'tweaks'.
V
V
I think a small / moderate sized inductor in the power supply to replace the R is a pretty safe bet on an improvement with the Leak. I like the design in general, quite
simple and efficient.
Z