Thats how I remember it too - and I have vague recollections of a similar Elec Eng laboratory module !
1 Like
Your lecturer wasn’t Dr Comley at TCU by any chance ? 
No - a Mr CD McEwan at UCL for that module. Very good electronics lecturer. Nothing to do with the lager, although we always wondered.
2 Likes
Just read that REL use a class D amplifier in the S/510.
Mixing and matching doesn’t seem to be a problem - should I be surprised?
REL use Class D amps in all but their T/X series subs.
1 Like
No, it never has been a problem!
1 Like
I don’t see how it could be a problem using a class AB amp and a class D sub. What difference could that possibly make? In my case the high-level output of the amp (AB) goes into the high-level input to the sub (D, via the speaker terminals), and at that point the amp’s class is inconsequential. The subs class has to be even more inconsequential.
1 Like
It’s funny you mention that. Not knowing what was in the mixing studio means I’ll never really know what is “correct” sounding. And for that reason, I have come to realise a person can like very different presentations without having to decide which they prefer if they are willing to have multiple systems. I have a (as you say) silky smooth class A tube amp system and the more visceral AB Naim. And they serve different moods. Moment’s where I appreciate a period of silence back when I had one main system, have often become moments where I just want a period with a different presentation.
Oddly enough I have another system that is class D driving the same model speakers as the Naim does but it’s closer in presentation to the class A tubes.
1 Like
Interesting consideration. I’ve always thought in terms of a single system, making it as good as it can [affordably] be. One system that works with all music, and all moods, and only one room occupied with and focussed on music. Multiple systems could indeed be another way, though additional cost unless only cast-offs with minimal resale value are used, and more rooms requiring space and layout/acoustic considerations, but if it works for the individual why not?
Well I do lay some ground rules to keep costs under control. Only one room is really dedicated to music. The other rooms, the hifi has to accommodate the room, not vice versa. For the second “main” system, I set a budget of source+amp+speakesr must be <= cost of one Supercap DR (or 1/7th the cost of the big system). Then other systems must cost half as much again. And indeed I have unused amps and speakers here and there too, I could easily build a couple other systems.
- Ultimately, I do 95% of the cooking for the family so I like to have music in the kitchen. Solution: a kitchen system.
- I work at home chained to a desk a lot. Solution a desk system.
- I like to take a private break during the day or get a different presentation, or simply (often) can’t get any living room time because the kids are hogging it. Solution an office system.
- And then of course the main system in the custom built living room that gets the lion’s share of the funds.
I think if I had a less chaotic life and more leisure time and access to the living room, I could scale back to two systems. I didn’t plan on this. It sort of evolved. One day I built the second “main” system as a stop gap and had an epiphany, This sounds nothing like my Naim system but I really love this presentation just as much.
While you could have them all be a part of a seamless Naim multiroom ecosystem, it’s not as much fun as interacting with very different gear that makes you feel very different just to look at and touch, let alone listen too.
Believe it or not, I do enjoy quiet time though. The bedroom is a sanctuary of calm. The most you can do there is sleep or read a book.
3 Likes
Class AB is actually well overdue for a redefinition. Strictly speaking the dividing line between Class B and Class AB is whether the transistors switch off (and sometimes leave a dead patch on either side of 0V). What happened in the early days was that they ran some quiescent current through the output stage so that it would be in Class A, at least up to a point. That was the original Class AB, and how much current you could have waiting on hand depended on the thermals of the design. But there is an optimal amount of quiescent current which allows the transistors to hand over to each other approximately seamlessly and without what is known as gm doubling (the transconductance of both output transistors working at the same time, which has associated distortions and some non-optimal attributes like sharp discontinuties in output impedance). That ideal quiescent current creates what is really an optimised Class B - a perfect handover - and for a period I also called it Class B, but everyone looked at me puzzled and as though I was talking about something horrific. Fortunately Bob Cordell came out with his book on Power Amplifiers and cemented it as being in the AB domain, covering this aspect quite extensively.
Anyway, AB can now mean both the optimal quiescent current and handover (which incidentally also means not switching off the transistors entirely) and as much quiescent as your amp can manage or you would like, though very few people (I think) do it that way any more. (Everyone has distortion analysers these days and no designer likes worse graphs than they need have, even when distortion is not the main criterion.) But you are not entirely wrong, because 50 years ago - approximately when these distinctions were first coined - this optimal handover would have been described as Class B.
For completeness, a lot of what are sold as class A amps today only manage to be in Class A for a portion of their voltage swing (especially if they are driving low impedance loudspeakers) so they are, strictly speaking, Class AB amps, but of the old style variety. I don’t know how Naim amps do their Class AB because I have never asked and have never checked. I suspect it’s optimised handover, not least because they have a few other things in their armoury at that end of the amplifier, and even more so with the new NC amps which, to me at least, look like a very exciting development. In the 18 months or so before their announcement I was looking into distortion in output stages into heavy loads, so I know the potential gains that can be made and I’m very much looking forward to hearing how they translate into improvements.
3 Likes
Interesting.
1 Like
This is an interesting perspective and one which I may agree with. For a couple of years I had an MF A370 here, in fact the review sample which Hi Fi Answers had called the best amp in the world at the time. Not quite monoblocks but built as such with 2 x 700VA(?) transformers and 120,000uF of reservoir capacitors!) I should first of all say that it was nearly impossible to find a flaw in it and it was a cracking amplifier. I was so impressed with it - and that was without really using its immense power because it was driving quite small speakers. I really couldn’t (and still can’t from memory) put my finger on anything missing but I would have enjoyed the zip and drive one can get from Naim amps just a bit more over that period. I also missed my 32/Snaps (which was taking just CD and digital inputs) and its immense musicality and sweetness. Funnily enough, back when phono was the only input, I had an MF A1 Class A amp which I hated (though it is being re-released now that Tim de Paravicini is sadly dead). Its only redeeming feature was that it wasn’t as scratchy, clashy and trashy as most budget amps at the time but I found it as dull as ditchwater. Yes, there’s lots of Class A goodness in the 32 but it’s not the answer to everything - though as a supplementary heater in winter, the A370 was shockingly great. It was the metropolitan version of having an Aga.
1 Like
Good summary, but not sure you are right here… the difference between Class B and Class AB was always (since the 1980s as far as I aware ) been a choice between efficiency and distortion, in regard to push pull conduction bias or small signal offset . I am away from my office right now where my well thumbed copy of Horowitz and Hill is on a shelf… when I am back I will see what it says about class B and class AB… as we know that has been considered one of the pre-eminent electronic engineering design texts for the last 40 years or so … and I see it’s in its third edition now.
1 Like
It still is! Like most, mine’s going nowhere for as long as I’m here!
1 Like
Indeed, the bias is typically set so that you move the conduction into the start of the more linear part of the conduction curve… that is a pay off between efficiency, gain and negative feedback.
gm (transconductance) doubling distortion is about imbalances in the positive and negative driver transconductance bias such that they are not working in a totally complimentary manner in the push pull arrangement with respect to the small signal, that is both push pull drivers could be amplifying at the same time. This is a key consideration in class AB design.
Thanks for confirming the traditional Naim amps are Class AB… I couldn’t see them being anything other than that.
MF power amps !
My main system uses a modified nuvista 300 power amp. From low to flat out ( i.e too loud) it appears unburstable and sounds great driving my pmc 20.26 speakers. I cant think what id replace it with naim wise. But i know if i could, it would cost me a whole lot more.
1 Like
This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.