It was . I found it by prodding around . The R10 symbol is set between the resistors and is obvious once tested . In real life much smaller . 390 mV does some nice things ( 1.34 V standard into 100 K ) . It suits NE5532 better for a start which was born out by some tests . Hypex use 100 R series resistance input to chip .
Nige, I may be missing the point, but I would wager that a high quality standard analog PSU would outshine soundwise the vast majority of SPSs. They may be much less effcicient, but there's no such thing as a free lunch, and efficiency has its price.
Put in another way, if you could separate the input stage/VAS power supply on those boards, and feed them from fully regulated and well filtered separate lines from those feeding the current stage, 1 will get 10 you'd have a much better sound.
11 years ago, a friend and I measured the amount of noise those switchmode power supplies as used in PCs send back into the grid - "criminal" is the only word applicable. They should be outlawed. And all it takes is one simple capacitor to cut the problem down by 90%, yet only ONE PSU had it.
I would imagine the switchmode supplies for audio are in fact better made, but in my book, they still have a long way to go - a VERY long way. And when they make it, they will cost as much as standard supplies, their only benefit being that they will be much smaller and hence need less in-case real estate.
Put in another way, if you could separate the input stage/VAS power supply on those boards, and feed them from fully regulated and well filtered separate lines from those feeding the current stage, 1 will get 10 you'd have a much better sound.
11 years ago, a friend and I measured the amount of noise those switchmode power supplies as used in PCs send back into the grid - "criminal" is the only word applicable. They should be outlawed. And all it takes is one simple capacitor to cut the problem down by 90%, yet only ONE PSU had it.
I would imagine the switchmode supplies for audio are in fact better made, but in my book, they still have a long way to go - a VERY long way. And when they make it, they will cost as much as standard supplies, their only benefit being that they will be much smaller and hence need less in-case real estate.
This is interesting. It has a chart of the power needed to match the SPL of instruments, for different speaker sensitivities. I still need to translate it from the French, to better-understand it.
(a.wayne, you might have been shooting way too low, with your lust for power. <grin> )
Capacit dynamique d'un systme audio
(a.wayne, you might have been shooting way too low, with your lust for power. <grin> )
Capacit dynamique d'un systme audio
Last edited:
There's always Google Translate which may be useful, though it doesn't translate the text embedded in images: Google Translate
Thanks for that, Tom. Some interesting points ... note that the piano, supposedly so difficult to do, only requires a 60W amp to drive speakers of typical efficiencies, 90dB, to maximum loudness. Which matches up with my own experience - a solo piano requires the volume to be such that the amp is at the point of clipping on a peak transient, essentially the max. volume setting on the HT system.
The most extreme volumes are only for people wanting to do silly things, like sitting a foot in front of the brass section in a full size orchestra - low 120dBs is a good goal, will take care of 99.9999% of situations ...
The most extreme volumes are only for people wanting to do silly things, like sitting a foot in front of the brass section in a full size orchestra - low 120dBs is a good goal, will take care of 99.9999% of situations ...
I'd steer away from SMPSUs for PCs in any audio application, they really are built down to the lowest possible price. Digital stuff has higher noise immunity so low output ripple isn't a major requirement. Didn't they have to meet any regulations for conductive noise?
Switchers built specifically for audio I've yet to encounter - my experience has been with the general purpose brick kinds (Meanwell, from Taiwan and 50% cheaper mainland Chinese clones of these with no regulatory approval). Yes they're not suitable for audio in stock form, but can be modded. The grounding (internal caps between mains and the output 0V) needs attention. Output ripple needs filtering (too much HF hash), common-mode noise needs suppressing. Provided you take steps to address these, I'd never go back to an ordinary linear supply. SMPSUs are definitely the future for audio poweramps.
No underestimation here , big speakers and lots of power for any sort of realism
Not so that estimation is for a listening distance of 1M, flea power and high sensitivity can make for real noise , not so for realism and size , big power wins out all the time ....
I've a feeling the combination of realism, size and big power you're talking about here is the impression that the bottom end of the keyboard is being fully expressed, realised ...
Talking about distances from speakers, this for me is always a key indicator: when the sound not up to scratch, the distance you are from the drivers is crucial, makes all the difference; when in good order, that distance becomes irrelevant - you can be 1m, 2m, 4m ... or just 0.5m away ... makes no difference, subjectively ...
Talking about distances from speakers, this for me is always a key indicator: when the sound not up to scratch, the distance you are from the drivers is crucial, makes all the difference; when in good order, that distance becomes irrelevant - you can be 1m, 2m, 4m ... or just 0.5m away ... makes no difference, subjectively ...
Very interesting indeed. For me especially.
It MIGHT have a caveat - those are pure numbers, "as is" in real life, but I have yet to hear a recording on any medium which did not compress the original sound at least a wee little bit.
This table clearly shows that with my speakers, and the power I am mostly concerned with, I am just fine with my own take on a 100W/8 Ohm amp. One of the reasons for this is that given the power supply voltages, my 100W amp will not clip up to about 172W/8 Ohms, an additional +2.3 dBW not everyone has.
It also VERY clerly shows that what Klipsch, JBL, Altec Lansing and me
have been saying all along - speaker efficiency is GOOD! The least efficient speakers I have ever had were my AR 5 boxes - typical AR of the day (early 70ies), superb tonal balance, but at 87 dB/1W/1m, not efficient enough. It drained my reVox A78 integrated amp (rated at 50W/8 Ohms) of the day at any higher volume level.Lastly, the table shows the absolute numbers, but let's not forget that at home, in our rooms, we do not listen in absolute numbers. As Wayne put it, 85 dB SPL in a room is already quite loud, but still within our comfort zone, and will not cause the SWAT team to come crashing in through the door. Or the missus to reach for her kitchen cutlery.
I understand a British company, Chord, uses them in their amps, however, this one is of their own design and manufacture, not likely to appear anywhere.
While I quite agree with your note on tuning a switchmode PSU, and while I have indeed dome some of this myself in my time (replacing suspect 150uF electrolytics with quality Siemens 330 uF, replacing some god-knows-which-if-even-He power transistors rated at 90 and 120 Watts with 200 Watt Motorola trannies, etc), I still have to say it - when you have just bought it, and know straight away it needs work, something is awfully wrong with the industry. My filters will make sure the junk from the grid never gets into it, but there's nothing they an do about poor, cheapskate parts used inside, with corresponding results.
And audiophools advocate miniature speakers because they "image better" ...
I'd like to see those 4,5 and 6 inch cones reach down to sub-80 Hz sounds with anything resembling reality.
I don't always agree with the American view that bigger is better, but in some cases, such as loudspeakers, they are quite right. Big sound comes only from big speakers, period.
I'd like to see those 4,5 and 6 inch cones reach down to sub-80 Hz sounds with anything resembling reality.
I don't always agree with the American view that bigger is better, but in some cases, such as loudspeakers, they are quite right. Big sound comes only from big speakers, period.
MC2 Audio use switching supplies in their E-series amps.
Not cheap but they do sound identical to their amps using linear supplies.
Of course they do or otherwise there would be something wrong with one of them.
You assume this is an automatic process, changing from classic analog to switchmode power supplies?
"Of course" imples that you seem to believe that it is a natural thing for switchmode supplies to sound as good as classic PSUs? If that were so, everybody would be using switchmode supplies as a matter of routine, because manufacturers could have them as outsourced and hence cheaper than otherwise.
Doesn't the fact that only a few manufacturers, all of which are upper market, use them, while most even higher upmarket brands do not?
Isn't it symptomatic that even class D kit manifacturers ALWAYS offer classic stabilized power supplies as their first and foremost upgrade over their basic kit? I wonder why?
BTW, let me also add this - I have no doubt whatsoever that in our quest for energy efficiency switchmode PSUs will eventually take over, just as class D amp own the future. In twerms of energy used, it makes sense to use a PSU requiring say just 300 VA where we are now using those requiring 600 VA, but at this time, switchmode PSUs are simply not up to the quality od linear PSUs - yet, with only very few exceptions.
At this time, they serve only to demonstrate that it is possible to obtain high quality from them, but is still prohibitively expensive at this time.
At this time, they serve only to demonstrate that it is possible to obtain high quality from them, but is still prohibitively expensive at this time.
Not so , its not all about the lower register, the really big difference between real and hi-fi is percussive energy , hi fi is weak and thin compared to real, the way drums, piano, horns , cymbals , et al, impact the air is where hi fi fails ....
I compared a Hypex switch-mode PSU with a 500 VA conventional PSU inc 2 x 22 000 uF . The Hypex PSU makes measuring the Hypex UCD 180 modules more difficult . It must mean something ? I have advised the person I am helping with this to make a two box amp one day and have a conventional PSU . Having said that the Hypex PSU has nice ideas . Sad it isn't convincing me as I would love to use it with a conventional amp . I have measured noise and it is is not subtle noise compared with a humble LM7812 . My guess is Hypex feel as long as the noise is above a certain frequency and stable then no sub harmonics will form ? Seeing how easily my spectrum analyzer creates them I don't see the real world being as mathematically ideal . We have been here before with DAC's I feel ? People would have laughed if stating nanosecond requirements in 1985 . Now no one questions it . Common-mode rejection seems unlikely to cope ? I would advise Hypex to consider using either OPA 2604 or even TL072 on the cheaper versions as input buffer amp ( NE5532 with gain of 5.5 ) . My tests seem to show considerable benefits using JFET when class D . I wonder if JFET's so called superior sound is just the ability to cope with RFI better ? I never measured anything to get excited about when JFET if the environment is quiet .
SMPSUs offer the potential to make much better use of the AC waveform, by means of power factor correction. Most at the moment don't have it though as it adds considerable cost, hence we get distorted, 'flat topped' mains. A standard linear supply, just like a non-PFC SMPSU has a fairly poor power factor which translates to excess losses in the wiring network.
Normal hi-fi, yes ... sorted out audio reproduction - no, a big no!
I know what you're talking about, I've been playing with this for nearly 30 years - and why I have little time for the sort of things most people jump up and down about. And the evidence is in, for me, without data,
, that this is all about the electronics ...The big Bryston, with Dynaudio speakers, did it at the Sydney show, when I cajoled the bloke into hitting the accelerator - I could hear the tell tale signs that it would be capable when it was running at normal volumes. Of course, not Politically Correct to have systems working at realistic levels, so he dropped it back to "nice" levels as soon as he could ... but while it was producing proper sound it made everything else at the show sound like wussy rubbish, or PA overkill ...
- Status
- Not open for further replies.