View Full Version : Very interesting article on negative feedback


hacker
06-05-06, 12:43 AM
Some light reading for a saturday morning:

http://milbert.com/articles/FutureWithoutFeedback/FutureWithoutFeedback.html

Makes me wonder about hacking the 321 and NAP boards for less feedback... :D

Carl

Paul L
06-05-06, 02:27 AM
So why did Vereker and others carry on with gobfuls of negative feedback anyway and ignore their own listening results? I expect our resident Euphoniphobe will be along to tell us in due course.

martin clark
06-05-06, 02:49 AM
Articles like the one linked are a bit teleological, in that with most valve designs you simply don't have enough gain in the first place to make high levels of feedback possible. The other misdirection is that they are talking of *globally* applied feedback - if your using an active device, there's inherently lots of local feedback you can't do away with - and wouldn't want to.

None of this is to say that feedback can't be applied badly though... and doing-away with ill-ythought-out f/back arrangements is probably a good thing.

markt
06-05-06, 02:55 AM
"So why did Vereker and others carry on with gobfuls of negative feedback anyway and ignore their own listening results? I expect our resident Euphoniphobe will be along to tell us in due course."


I think that's a little unfair, doubtful that they ignored their listening results, and if anything had identified this very problem at the start, Martin colloms points out that the problem isn't feedback itself but it's application to circuits that have not been optimised first i.e. making the circuit shorter to keep time domain errors as small as possible then using feedback to clear up what is left, no, Mr Colloms was probably pointing to the misuse of feedback to overcome design flaws, the more intelligent designs being faster and needing less of it, so the diss of Naim is absurd.

I've listened to the 321 stage without feedback, it does indeed sound less smeared, noisy though :)

david ellwood
06-05-06, 02:56 AM
I am reading doug selfs book on amplifier design at the moment and it is a great read. I am finding it interesting sorting the useful and substantiated theory from his vitriolic hatred of all things subjective.

His take on the low negative feedback brigade is particularly interesting and he shows a clarksonesque disregard for their ideas on its merits.

Anyone else any opinions on dougs ramblings?

martin clark
06-05-06, 03:09 AM
It's a great book, and contains lots of hard work, investigation and some thought -provoking (and very readable) commentary. The strength of the book is that it is an exhaustive investigation of one particular (though dominant) power amp topology. However this necesarily means there are some intrigung possibilites for further investigation completely missed - such as the matter of compensation. Self assumes a straightforward dominant-pole with 'miller' cap , and there's no investigation of the way this can (negatively) affect HF performance - or of the alternatives (both outlined in J L Hoods writings for instance).

Actually I think Selfs rants are pretty-well judged, even amusing ripostes to some of the claims from the other side of the fence...

martin clark
06-05-06, 03:37 AM
Coincidence? A very good page on feedback by Rod Elliott appeared today!:

http://sound.westhost.com/articles/distortion+fb.htm

markt
06-05-06, 04:47 AM
The Avondale NCC200 for instance, it's that well known circuit, but with a few extra feedbackback loops to tighten things up nicely, yet it has the good naturalness associated with valves to my ears, probably the increased parts quality which is another thing associated with high cost valve amps, good parts certainly help a good design IMO.

trancera
06-05-06, 10:16 AM
Coincidence? A very good page on feedback by Rod Elliott appeared today!:

http://sound.westhost.com/articles/distortion+fb.htm

Rods articles are excellent, and so are a lot of his projects.

Its his P09 Active crossover board I use and I'll never go passive X Over again, tremendous.

James
07-05-06, 02:18 PM
Is that why Densen amps sound so good?

James

BobMaximus
07-05-06, 04:56 PM
I seem to remember someone here saying they had tried building some of the Self power-amp circuits. I think they came to the conclusion that, although the theory looks very persuasive, they did not sound at all good. It could have been Andy Weeks, but searching on this site doesn't exactly facilitate the finding!