Krill - The Next Generation

[sonnya]

Wahab, here is one thread about Mirand, it's a design by a guy on this forum called sonnya, it's quite impressive in terms of complexity and figures.
http://www.diyaudio.com/forums/solid-state/139439-mirand-a1.html
If you search on sonnya posts/threads you will find more.


About the Krill,

I have been looking at it from the side view so far and what came to my mind from the very beginning is one can use diodes at the inputs of the driver/OP which will also prevent the base charge to be sucked out, the OP stage can only be pulled on through the diode but not off.
Actually a circuit that is supposed to work according to the non-switching off scheme should not show any significant rise in cross conduction current by rise of working frequency, eg. that is there is an intelligence designed into the circuit to keep the cross conduction fairly constant within its working frequency which in this case is the audio band and for different loads (hence higher load means more base charge), can this circuit show such a consistency say from 20 Hz to 20 kHz and, different and reactive loads??

Cheers Michael

Not to steal this thread, but the MIRAND A1 has some years on its back..

And there are room for some improvement.
The Amp's error correction and VAS stage is based on BC846/56 family. The v
was the best transistor you could get you hands on from a lot distributors but now you can easily get larger quantity through farnell, mouser and digikey of the more exotic types featuring more linear gain,higher Ft and lower parasitic capacitance.
Also the driver stage can be improved with BJT's with a higher Ft and lower cob.
The output transistors can be doubled for improvement of the stability by making this buffer stage more stiff. I know that there is a lot of talk about multiple output transistor as they do switch simultaniously. But if they where runing in class A they would rarely switch off.
Also opamp's for the servo can be upgraded.