Although, I guess it also serves the purpose of filling the gap in the class C amp. I think the Sansui amp is a conventional class ab, with the additional (class a amp) correction amp just carrying out the feedforward.
The class A amp is not the error correction amp; there is no error correction circuit in this amp..
The class A amp drives the speakers.
If the level increases, the dumpers start to take part of the load and as a result, the feedback factor increases which throttles back the gain of the class A stage.
You guys make it all so complex and needlessly hard. It's really very simple.
Jan
The class A amp drives the speakers.
If the level increases, the dumpers start to take part of the load and as a result, the feedback factor increases which throttles back the gain of the class A stage.
You guys make it all so complex and needlessly hard. It's really very simple.
Jan
Totally agree with Jan on this some thirty years ago I did some experiments that came to the same conclusion
Trev
Trev
I disagree - they are both integral parts. However, the Class A amplifier is key to eliminating crossover distortion. Were it not a class A amp, then it would have to be AB and alignment would critical to avoid simply substituting the crossover distortion; albeit severe, of the class C dumpers with that of a low level class AB amp.
The real magic is in Walker's genius in coming up with the design. I am sure there is a lot he didn't say about his design. I said in an earlier post that an external control amp could be designed to control the point at which the dumpers switch on to support the Class A section. But what Walker did, was to come up with a design which achieves it by itself. Just how effective it is, is illustrated by virtue of the fact it needs no alignment and component tolerances are less critical. I believe it was also Quads most successful design because of low production costs.
Whatever, I could not agree with you more about its exceptional THD figure.
That was the design objective. Quad were in the middle of making 100,000 303s with 5 adjustment points each, and didn’t want any more of that.
They’re still making current-dumping amplifiers 50 years later.
I am referring to the signal at the base of Tr10, whose flat portion is certainly not the ‘normal rate of audio signal’. It is the rather minimal signal required to control the dumpers. Its sides OTOH are exactly that. They are the audio signal. There is no ‘accelerated portion’: over this range the Class A part is driving the load directly. Nothing to accelerate for.
Do what II did. Put a scope probe on Tr10 base and see for yourself.
It is a very simple circuit if you read the many descriptions from the time, it will be clear.
It is only made more complex by all the speculators trying to find some hidden complexity thereby creating it.
Jan
Not at all! The key is the balancing of the bridged feedback, if that balance is good the performance will be good.
You can have the worlds' best ideal class A amp, but if the bridge is unbalanced, it's all moot and you'll have distortion..
Unless that is clear, we go round and round with speculations and finding nonsense 'explanations' .
Jan
Leave us alone Jan - some of us need to go through these exercises, otherwise we get left behind by people like you - Ha, Ha. You could have said that earlier - but there again, I for one would still want to explore the subject, just to hammer it home into this thick head.
Of course, you are absolutely right, it is not an error correction amplifier in the conventional sense of the term. However, its variable dv/dt slew rate in effect reduces the error which exist with the dumpers when they are in the dead zone, which of course is being policed by the Class A amplifier.
A valid observation. But vision is always 20/20 in hindsight. The key to solving any equation is using the right formula, or in this case someone who can explain the problem in layman's terms, rather than want to be seen as clever.
In this regard, Quad themselves and to a degree Walker were responsible for muddying the waters - and why wouldn't they? They wanted the buying public to think they were purchasing some audio equivalent of a Rolex.
And the vast majority of the buying public are not electronic engineers.
A timely observation and I'm grateful for the Mooly's plots which are illustrating much more than I could put into words. It is also nice to know there are others actively contributing.
Jeez - I would love to hear my 909 driving a pair of Quad ELS. The gossamer thin diaphragms would result in incredible accuracy in any music. I wish I hadn't said that!!!!!
I have spent the last 54 years railing agains my need to buy a pair!
Then Technics, with their class AA, went on to improve upon the design. Instead of using class C Dumper, they made the dumper an, over biased, class AB. I know Walker's design was to rid his amp of bias controls, however, I would assume that the Technics solution is technically superior to Quad's?
Always interesting...
But I cringe a little and do a face palm every time I see the word " dumper" in a thread title.
Could just simplify Sansui idea and just over bias a class A/B and call it a day. No " dumper"
Better yet, just idle at the normal ideal 120ma per device and get the same thing.
But I cringe a little and do a face palm every time I see the word " dumper" in a thread title.
Could just simplify Sansui idea and just over bias a class A/B and call it a day. No " dumper"
Better yet, just idle at the normal ideal 120ma per device and get the same thing.
Not necessarily! They may just have been trying to gain a bit of leverage in the market. Have you any figures that support the assertion? Does anyone have any information and are they still making it? As has just been pointed out, Quad are still making CD amplifiers, which speaks volumes for the design.
I would be interested to know what the Slew figure is for the Sugden A21 or any of their pure class A Amplifiers.