Class A or AB ?????

Tenson said:
Wasn't there an amp topology that ran multiple voltage rails, with class A/B running on the low rails and class B on the higher ones, and then switching to the higher rails when power demands increased? It sounds quite similar.
Click to expand...

Hitachi had such a circuit in some of the Dynaharmony range and so do NAD today.
 
Sometimes called Class G, I believe. Uses diodes to switch-in a higher voltage rail or two as required. Think Doug self has done some extended work on these and how to avoid glitches from the commutation.

Biob Carver's 'magnetic field' amplifer and its later derivatives probably deserves a mention, too: a small (AB) power amplifer linked to a switching supply that bumps-up the rails to allow massive peak outputs on a very short-term basis, though not 'sliding-bias'. The original mag field amplier reputedly presents an unbelievably horrible load to the mains - 60A peaks and the like - because it worked in a resonant mode depending on some fairly crude/brutal chopping though the transformer only had something like a 50w continuous rating IIRC! I've never heard one or played with one, not sure I want to... Later derivatives are allegedly better behaved, as in the Carver 'Sunfire' woofer - 2Kw peak output and basically no heatsink. (Heard one of those, sounded horrible however it was set-up I must say. Might have worked for mindless blow-em-up movies)

Anyway - back to one of the points behind the OPs post: how much power do we really need? I've done the scope-while-listening thing, and with fairly sensitive speakers I get stupidly low average power figures - about 20-50mW. Even allowing 30dB headroom, that means a nice class A amp of only a handful of watts... really ought to get built this year ;)
 
Last edited by a moderator:
hello jez, while working at M.F. did you ever get your mits on a B200, ? what was your opinioon if you did over the A100, i always loved both the B200 and the A1000.
 
felix said:
Anyway - back to one of the points behind the OPs post: how much power do we really need? I've done the scope-while-listening thing, and with fairly sensitive speakers I get stupidly low average power figures - about 20-50mW. Even allowing 30dB headroom, that means a nice class A amp of only a handful of watts... really ought to get built this year ;)
Click to expand...

I did something similar recently and even running something like the Rogers at 86dBw sensitivity a 10 amplifier would seem enough. I do listen in the near field though.

Very loud replay using the sort of stuff designed to stress systems will push up the requirement considerably.
 
On the other hand, connecting a 10watt amp like the little tri-path does result in a squashed sound.
 
Tenson said:
On the other hand, connecting a 10watt amp like the little tri-path does result in a squashed sound.
Click to expand...

True, but those are an unusual case and the ones I've heard run on tiny little power supplies.

I''ve a 12wpc Trio amplifier here that would run rings round those things.
 
nando said:
hello jez, while working at M.F. did you ever get your mits on a B200, ? what was your opinioon if you did over the A100, i always loved both the B200 and the A1000.
Click to expand...

Yep worked on loads of them. Never got my mits on the A1000 though as it was after my time there. The A570 monoblocks and Chronos valve amps were the very top of the range at the time and the 570 was only made to order.
Always had a soft spot for the B200. It combined loads of power for such a tiny amp with good clarity and all in the same box as the A1. Has it's limitations in the under rated capacitors and transformer though. The difference in max power with one channel driven compared to both is about the largest I've ever seen! I've modified these to give 105W per channel. They are in effect a P140 power amp and the pre amp of the A1.
The A100 could be a nice amp as well and I preferred it to the A1 but the B200 would easily beat it in terms of speed, wallop and transparency. On the other hand, the A100 would win in terms of sound staging, smoothness etc. I always thought the A1 and A100 were a bit too warm and smooth sounding.

There is class H as well as class G which strives to achieve the same thing. The nomenclature of them can vary according to which side of the Atlantic you are on!
Class H works by modulating the supply rails to keep them just a few volts above that needed by the signal demands and will often use switch mode PSU's to achieve this. In effect it could be said that the voltage rails are being generated by a class D amp which then powers a class A/B amp. The principle has been used to make more efficient and higher powered class A amps but I don't like the idea of the PSU always having to play catch up with the signal!
 
Never heard of the MF A570 - that must have required an army squadron to get the thing past the front door in light of Noel Keywood's reports of getting the A470 home.

Very simple power amp circuit in the B200 and P140.
I had a pair of P140s for a while (you could mono them easily) and liked them.
Used with a P172 pre - one of the few to use tone controls.
 
The A570 was a 470 with both channels paralleled, much like the mono use of the P140 that you mention but internally wired like that.
The P270, A370, A470, A570 and several others are all actually the same design as the P140/B200 but with more output devices and bigger power supplies, heatsinks etc. To the best of my knowledge, the idea originally came from a Motorola application note using bipolar transistors which TdP modified to use MOSFET's.
Glen (Garfield) at Alchemist Products built a prototype amp that took the idea back to bipolars whilst I was working there. It sounded very nice actually but never went into production. I designed a zero feedback class A amplifier for Alchemist at around the same time which likewise sounded great but never made it past the prototype stage.
 
The A570 was a 370 only mono blocked to make the a570, huge amps, only one pair were made to a special order to a cutomer, and they got stolen from the customer, shame.
 
Arkless Repairs said:
The A570 was a 470 with both channels paralleled, much like the mono use of the P140 that you mention but internally wired like that.
The P270, A370, A470, A570 and several others are all actually the same design as the P140/B200 but with more output devices and bigger power supplies, heatsinks etc. To the best of my knowledge, the idea originally came from a Motorola application note using bipolar transistors which TdP modified to use MOSFET's.
Glen (Garfield) at Alchemist Products built a prototype amp that took the idea back to bipolars whilst I was working there. It sounded very nice actually but never went into production. I designed a zero feedback class A amplifier for Alchemist at around the same time which likewise sounded great but never made it past the prototype stage.
Click to expand...

I worked on a P170 recently which is a P140 using a dual mono PSU in a regular proportioned case.
Unmarked transformers but I'd say 250--300va per side and a pair of Elna 22000uf caps per channel. Still only one pair of mosfets per side in the P170.

Lovely simple circuit and sounds great.

mf_p-amp_170_scaled.jpg


Note the incredibly crude power supply feed to the op amp. Just dropper resistors, diode voltage clamp and a decoupling cap.
Didn't stop it performing though.
 
I know for a fact what the A570 was as I personally worked on them! Definitely A470's made into monoblocks. I can't be certain as to how many were made, certainly not many and as I said, only made to order. It was a long time ago and maybe my mind is playing tricks on me but I seem to remember working on more than just the one pair.... but no more than 3 pairs. I well remember working on a pair that had gold plated plaque's on them stating that they were specially made for this customer (a Greek shipping magnate I believe).
 
that is the pair jez that were specially made, how can someone still those monsters? you know how heavy they were,
 
Rob. The circuit you have shown there is very slightly different to the ones we have been discussing. It's a pedantic point really :)
All the variants we have been rattling on about have 2 transistor constant current sources feeding the zener diodes in place of the wirewound resistors in your diagram (which I believe I am inadvertently the source of! I lent all the MF diagrams to a "friend" who had some in for repair and who it seems scanned them all and was selling them on cd's over ebay until I found out and stopped him... probably most of the diagrams in the public domain originate from this!? or maybe not?!) which at least make the supply to the op amps a bit better than this diagram suggests. Also a small difference between the B200/P140 etc and the bigger amps was that a pair of emitter followers was used after the common base voltage amplifiers in order to drive the output mosfets from a lower source resistance (and a pair of diode connected transistors as part of the biasing but that's of no import).
The P150 and latter B200's plus the P270.2, A370.2 had a discrete long tailed pair input stage feeding into pins 1 and 5 of the LM318. Pins 2 and 3 were connected to pin 4 to shut down the op amps own input stage. The only real effect of this was to make them rather less noisy.
I believe the diagram you have pertains to some PA amplifiers made by MF and I think sold under the Bose brand name... not many were made. Also with two pairs of mosfets I think it was the circuit of the MKII Studio T but not absolutely certain.
 
Arkless Repairs said:
Rob. The circuit you have shown there is very slightly different to the ones we have been discussing. It's a pedantic point really :)
All the variants we have been rattling on about have 2 transistor constant current sources feeding the zener diodes in place of the wirewound resistors in your diagram (which I believe I am inadvertently the source of! I lent all the MF diagrams to a "friend" who had some in for repair and who it seems scanned them all and was selling them on cd's over ebay until I found out and stopped him... probably most of the diagrams in the public domain originate from this!? or maybe not?!) which at least make the supply to the op amps a bit better than this diagram suggests. Also a small difference between the B200/P140 etc and the bigger amps was that a pair of emitter followers was used after the common base voltage amplifiers in order to drive the output mosfets from a lower source resistance (and a pair of diode connected transistors as part of the biasing but that's of no import).
The P150 and latter B200's plus the P270.2, A370.2 had a discrete long tailed pair input stage feeding into pins 1 and 5 of the LM318. Pins 2 and 3 were connected to pin 4 to shut down the op amps own input stage. The only real effect of this was to make them rather less noisy.
I believe the diagram you have pertains to some PA amplifiers made by MF and I think sold under the Bose brand name... not many were made. Also with two pairs of mosfets I think it was the circuit of the MKII Studio T but not absolutely certain.
Click to expand...

Hi Jez,

That diagram mirrors the P170 I worked on recently, but it was the early Mk1 (with red panel trim) so perhaps the circuit you describe was used in the Mk2 and later amplifiers.

Here is the PCB shot of the P170 before any work was done:


DSC00582.jpg
 
You are certainly correct about the P170 Rob. It even says P170 on the diagram. The diagram I have is identical other than that yours has had "270 370 470" written on it. It is not the correct diagram for those models. I had missed out the P170 from my comments because it was before my time at MF and I never worked on one! I stand by everything I said pertaining to all those other amps :)
 
Arkless Repairs said:
You are certainly correct about the P170 Rob. It even says P170 on the diagram. The diagram I have is identical other than that yours has had "270 370 470" written on it. It is not the correct diagram for those models. I had missed out the P170 from my comments because it was before my time at MF and I never worked on one! I stand by everything I said pertaining to all those other amps :)
Click to expand...

The changes you mentioned would certainly be welcome.
I do like the simplicity of that circuit though.

Did MF have a fetish for big yellow poly caps in those days! :)
They turn up in lots of MF kit from that period.
 

Latest posts

Back
Top