Ditto Wayne .
Did anyone try this with power amplifiers ? Related reading below .
http://www.essex.ac.uk/csee/researc...bdocs/J14 Mills-Hawksford power amplifier.pdf
Did anyone try this with power amplifiers ? Related reading below .
http://www.essex.ac.uk/csee/researc...bdocs/J14 Mills-Hawksford power amplifier.pdf
His arguments about some sort of hitherto-unappreciated granularity in low-level preamplification was also...errr...controversial
Yes. He seemed not to understand that shot/partition noise (which was essentially what he was talking about) only occurs when charge carriers are uncorrelated, which does not happen in a conductor (otherwise every circuit junction would generate noise). Basic errors like that make me cautious about everything else he says; perhaps unfair, but we all have to apply filters because none of us has time to read and understand everything written on audio.
He seems to appear to be having a very good time though, which I appreciate. However, while thoughtful and inventive, it seems he doesn't do a lot of background reading sometimes. And yet his name gets associated with circuits (e.g., the Hawksford cascode). What's in a name?
BTW a good albeit brief discussion of electrons in a conductor can be found in the late Peter Dunn's electronics book for physicists, Gateways into Electronics.
I have pieces of the schematics, mostly the inputstage. What do you mean by current gain stage, the output stage ??
There is a member that has one and is refurbishing it, promised me schematics as he reverse engineers and rebuilds the amp. The outputstage is based on diamond topology according to the literature. The inputstage design is very different indeed, I have ever seen anything even resembling it.
The Citation XX power amp was a first in relatively high power amps to attempt to do everything 'right'. It was not the only amp attempting to do it, but at least, it was made commercially, and is still known, even today.
Most of my amps contain virtually all of the additions put into the XX, today. e.g. High Current, Good Slew Rate, Symmetrical topology, etc., etc.
Most of my amps contain virtually all of the additions put into the XX, today. e.g. High Current, Good Slew Rate, Symmetrical topology, etc., etc.
It is something of an elusive phantom - everybody and their dog know about it, but few have seen it or heard it.
Perhaps it's the model's rarity that adds the extra aura ... Plus some really wild for the time claims, like the 500 Amp peak current capability.
@homemodder
I have that as well, but no predriver-driver-output stage at all. Ultimately, I'd simply like to have it whole, even if I don't plan on building one.
A propos John's comment: I just checked up and confirmed that the first next commercial project after the two Citations was H/K's power amplifier 870.
It had the honor of being the first commercial (hence simplified, but also cheaper) offshoot of the Citation technology. It was a low global NFB model, using what was to be H/K's standard topology for the remaining 80ies, all of the 99ies and well into the 2000s. Its current capability, as described by H/K's Instantaneous High Current Capability was quoted as 90 Amps (from two pairs of 150W devices per channel), its slew rate as 160 V/uS (cascoded FET), and THD as 0.08% into nominal 100 Watts into 8 Ohms.
That was fairly advanced in those days. Later development of the same topology, which includes my own 6550 integrated amp (purchased in December 1993), enabled similar performance (albeit with only 45 Amps of current) from SEPP designs based on Toshiba's 2SC3281/2SA1302 tranistors, much used by H/K for 16 or 17 years, and still used today in form of new version of the same from Toshiba, 2SC5200/2SA1943. Current top of the line integrated amp sports a power rating of 150 W/8 Ohms and uses no less than 6 pairs of the above trannies per channel.
Have you heard it by chance, John?
It had the honor of being the first commercial (hence simplified, but also cheaper) offshoot of the Citation technology. It was a low global NFB model, using what was to be H/K's standard topology for the remaining 80ies, all of the 99ies and well into the 2000s. Its current capability, as described by H/K's Instantaneous High Current Capability was quoted as 90 Amps (from two pairs of 150W devices per channel), its slew rate as 160 V/uS (cascoded FET), and THD as 0.08% into nominal 100 Watts into 8 Ohms.
That was fairly advanced in those days. Later development of the same topology, which includes my own 6550 integrated amp (purchased in December 1993), enabled similar performance (albeit with only 45 Amps of current) from SEPP designs based on Toshiba's 2SC3281/2SA1302 tranistors, much used by H/K for 16 or 17 years, and still used today in form of new version of the same from Toshiba, 2SC5200/2SA1943. Current top of the line integrated amp sports a power rating of 150 W/8 Ohms and uses no less than 6 pairs of the above trannies per channel.
Have you heard it by chance, John?
True. It did have an outrageous price for those days and was criticized for it, but then, no matter what you do, you can never please the press, can you?
It was a BIG departure from the cost of power amps, at the time, AND there was little hardware to show where the cost was. It was in the development of the amp, that took years, rather than months, like it should have, and politics in mid-management at that time. I teased Matti about it, when he first released it.
For example, he LEFT OUT the all protected, dual tracking voltage regulator that I designed for the project, as he said it was TOO EXPENSIVE to implement. Oh? Do you mean, Matti that you could only afford a couple of power caps with a rectifier bridge?
For example, he LEFT OUT the all protected, dual tracking voltage regulator that I designed for the project, as he said it was TOO EXPENSIVE to implement. Oh? Do you mean, Matti that you could only afford a couple of power caps with a rectifier bridge?
In no way I am aware of, to tell you the truth.
Ever since that spec appeared in H7K's literature, I questioned it. I intensly dislike unspecified claims, you know, I demand the nitty gritty, into what impedance, for how long, under which conditions, that sort of thing.
This is the reason why I am so critical of the audio industry, sometimes probably too critical - I read an Electrocompanied (Norway, based on Otala's circuits) ad in which the claim was that a nominally 200 WPC amp could pump out 1,500 Watts (sic!) in peaks.
Then I found an explanation in the smallest fond I have ever seen in actual press that this was for 1 millisecond only.
To my mind, 1 millisecond is a totally useless time interval. Even the rather relaxed IEC standards take 20 mS as the least meaningful value, and I would question even that, the nitpicker that I am.
In psychology, this type of thing is referred to as "guided logic". It is not a lie, but it is in fact useless, yet it's so stated as to lead you to believe what they can legally deny responsability for.
I (desperately) want to believe that the true masters of the art will never stoop that low.
John, far too many people, even in the trade, have never had the opportunity to hear exactly how a solid state amp can sound when fed from fully electronically regulated power supplies.
How regulation can make a 30W amp sound bigger, bolder and cleaner than a 100W amp. If anyone thinks I am overstating the case, get hold of a Naim 30WPC integrated amp from the early 80ies, it had full power regulation, and it did wipe the floor with typical commercial offering from Japan in the 100 WPC range.
Regulation does the power doubling into halved impedance particularly well, assuming it's well designed.
Its only failing is by association only. The weakest point of most electronically regulated power amps is their next to zero headroom, HOWEVER, this no inherent failing of the regulation as such, rather the fact that due to cost reasons, many designers try to go as low as they can with the voltages. If they went further, it would cost more, but is otherwise quite possible.
My own 680 integrated amp, made in 1999, designed by Richard Miller and his crew, is also billed as being capable of 90 Amp, again from two pairs of output devices per channel, this time rated at 130 W.
It's rated at 85/139W into 8/4 Ohms, and is a true dual mono configuratio, the only link between the channels being the transformer primary; separate windings for each channel, separate full wave bridge rectifiers, and 2 8,200 uF caps per channel.
So I did an Inquisition on it.
To cut a long story short, it managed to pump out 510 Watts into 2 Ohms for 40 mS, but went no further. But that's still "only" 22,6 Amps. Admittedly, I was using lab resistors, if those had been loudspeakers, I would have had the cops put an end to the testing on charges of public order disturbance.
pretty decent presentation - the "audio" takeaway is that VFB "wins" at "low frequency"
their idea "low frequency" is below 100 kHz or even 1 MHz
Last edited:
And beyond that, how ensuring that the power supply side of things is thoroughly clean in all senses makes all the difference. I've been laughed at a few times for "claiming" that a lowly 20W chip amp can sound very, very impressive if you do all the right things, but I certainly know what's achievable ...
Only a few months ago I did the rounds of listening to a large range of pro, powered monitors from the best names in the industry and to sum it up they were a joke: I was pushing them all hard, overload led indicators flashing continuously and they still sounded pretty gutless. A key reason for this is the lousy, conventional power supplies used ...
Frank
- Status
- Not open for further replies.