Idea for gainclone with transformers

[Shoog]

Hi all,

I have been living with the world of Valves for about the last 5yrs, with my venerable first gainclone projects relegated to the computer room. I liked what I achieved with the Gainclones, but my first foray in tubes showed that there was a naturalness which was entirely missing with the Gainclones. So I never looked back.
However it has always troubled me that my beautiful Valve amp sits there and consumes about 180W of power just doing nothing (fully differential class A from front to back).

So I have been toying with using up my stock of LM3875 chips and was thinking about what might have been the problems associated with my previous efforts. I am inclined to agree with Joe Rasmussen in his analysis that the main problem is probably slew rate related in connection with having to deal with input crud at very high frequencies - leading to early saturation and possibly even clipping. So borrowing from my valve experience I thought of the following conceptual design as a way of dealing with this. I know that output transformers add a certain magic to the mix which is all but impossible to tie down. Also I use hi-efficiency vintage drivers and have already had them fried once before by a tube buffered gainclone - so don't want to repeat the experience.
I have also had extremely good results using mains Toroidal transformers in all my valve amps. They are good out to 100K and beyond when driven with a low impedance load.

So shoot me out of the water, Good or Bad idea. Worth a try ?

Shoog.

[pacificblue]

Worth a try. Especially if you compare that amp with and without the output transformer.