View Single Post
  #5   Report Post  
Posted to rec.audio.high-end
Howard Stone Howard Stone is offline
external usenet poster
 
Posts: 45
Default More than 30W per chanel Class A transistor amps

On Thursday, February 7, 2019 at 12:43:01 PM UTC, Trevor Wilson wrote:
On 6/02/2019 3:39 am, Howard Stone wrote:
On Tuesday, 5 February 2019 10:53:05 UTC, Trevor Wilson wrote:
On 5/02/2019 8:27 am, Howard Stone wrote:
Any recommendations? Reliability and ease of service important.


**Krell KSA50. 50 Watts Class A, 75 Watts Class A/B. Very simple
topology, easy to work on, built-in fan cooling, reliable.
Krell KSA100. 100 Watts Class A. Uncertain about Class A/B power, but
likely in excess of 150 Watts. Similar construction to the KSA50, so the
same comments apply.

If unmolested, both will almost certainly require new electros by now.
The KSA100 was a particularly impressive sounding thing.

Dunno why you want Class A. A decent, high bias Class A/B amp, up to a
couple of Watts will sound just fine.

--
Trevor Wilson
www.rageaudio.com.au

---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus


Ah yes I checked your old posts and saw you have spoken highly of that Krell in the past. Noted.

Why class A? Well, why not?

I have a little second system with JR 149s, right now it's just got a Quad 303/34. I love those speakers! And I feel that with really a really top amp they could be quite fabulous.

But I don't want the trouble of valves . . . so Class A seems a good line of investigation.


**I have always regarded Class A as the lazy designer's way of achieving
good sonic performance. Thing is this:

The idea of Class A is to eliminate crossover distortion, which MAY
exist in a poorly thought out Class A/B amp. Crossover distortion (where
is exists) is caused by the non-linearity of the output devices (valves,
BJTs MOSFETs, etc) which occurs at low(ish) currents. In the case of
BJTs (Bipolar Junction Transistors - or just transistors), bias current
needs to be set at around 20mA (PER OUTPUT DEVICE PAIR) to eliminate the
effects of the kink (or knee) in the amplification curve. For MOSFETs,
the figure is somewhere around an Amp or so (1,000mA). Many BJT amp
designers set the bias figure higher than 20mA, thus giving rise to
allegedly superior performance. A bias figure that equates to a few
Watts Class A (as opposed to the more usual, 20 ~ 100 milliwatts) will
provide all the benefits of Class A, but with far more modest power
consumption.

Either way, old Krells are plentiful, well built, easy to work on,
properly cooled (using fans) and provide genuine Class A performance.
This is as opposed to many of Krell's competitors, who CLAIM Class A
performance, but don't actually deliver on that promise. Musical
Fidelity is one of the more notorious companies here.

--
Trevor Wilson
www.rageaudio.com.au


Which amps have been designed to have the crossover benefits of class A without the consumption.

I've decided to take a Electrocompaniet ECI-2 on a trial, it's not pure class A but I'll give it a go. The JR149s are by no means big speakers, but they have a reputation of performing much better if driven by a powerful amp. If it doesn't work out, then I can see a Krell KSA 100 with the offer of a trial -- though I'm a bit concerned about preamps (I only have the Quad 34 and a passive) I also noticed a cheap Sugden A21i for sale, but it's a small amp in terms of output -- may work though.