Re: Re: Re: How important Clipping?
*Yawn* I’ll reiterate (please note bold text):
These days, with modern digital audio sources, it’s easier than ever to set up an amplifier so that it simply cannot be driven into clipping.
I’ve got an old 10W per channel Rotel in my study connected to my computers CD player. The line level input is set so the max p-p output of the CD player produces the amplifiers rated output swing – i.e. maximum volume = 0 clipping.
Connected is a pair of reasonably efficient speakers, and the combo makes plenty of noise. Even for “HiFi” listening in a reasonably large living room, the setup would be perfectly adequate.
I idea that we must all have >400W amplifiers for sufficient sound pressure levels and freedom from clipping is just plain silly.
Get it now?
For fun (and some not too efficient speakers)
PMA said:
*Yawn* I’ll reiterate (please note bold text):
These days, with modern digital audio sources, it’s easier than ever to set up an amplifier so that it simply cannot be driven into clipping.
I’ve got an old 10W per channel Rotel in my study connected to my computers CD player. The line level input is set so the max p-p output of the CD player produces the amplifiers rated output swing – i.e. maximum volume = 0 clipping.
Connected is a pair of reasonably efficient speakers, and the combo makes plenty of noise. Even for “HiFi” listening in a reasonably large living room, the setup would be perfectly adequate.
I idea that we must all have >400W amplifiers for sufficient sound pressure levels and freedom from clipping is just plain silly.
Get it now?
estuart said:
Hi Glen,
If that's so, why you are designing/building a 1kW amplifier?
For fun (and some not too efficient speakers)
Nelson Pass said:
This set of curves is also discussed by Cheever, who also went to
the trouble of matching the mathematical basis to a real world
test circuit.
Nelson,
Is Cheever that guy that tried to match the distortion spectrum to his ears? That thesis is quite controversial to say the least (some are of the opinion that he should have never been awarded his degree), because he obviously 'forgot' that if you want to recreate recorded performances with electronic means, your ear characteristics are irrelevant.
Jan Didden
GRollins said:
Grey,
The idea is that with high enough feedback, you eventually get the higher harmonics down to what they were before feedback. The lower harmonics at that point are much lower that before feedback anyway, so you win on all fronts.
The fact that some higher harmonics are missing from part of the graph doesn;t mean they were not there, only that they were below the scale.
Jan Didden
Re: Re: Re: Re: Re: How important Clipping?
Actually, even with inefficient speakers, 1kW rms per channel is probably still way over the top
Anyway, getting back to my previous post, Johan said with which I agree and think is worth some more attention:
“I am there to listen to and enjoy music within an amplifier's (system's) capabilities, not too much of the time outside it.”
The maximum output amplitude of my CD player is governed by the full-scale output of the DAC. I have a test CD with some full-scale continuous sine wave test tones.
By simply setting the gain of my amplifier so that it delivers its rated output swing (with the volume setting at maximum, of course) when amplifying these test tones, I ensure that my system is set up such that the amplifier cannot be forced to operate beyond it’s limits. It’s as simple as that.
It’s a bit like setting up an RF transmitter for freedom from over modulation. Set up in this manner, my 10W per channel amp can still make a surprising amount of noise with my venerable old Pye speakers.
Not as much noise as could be had with a 400W per channel amplifier, for sure, but by no means inadequate for most sensible personal listening.
Cheers,
Glen
estuart said:
Actually, even with inefficient speakers, 1kW rms per channel is probably still way over the top
Anyway, getting back to my previous post, Johan said with which I agree and think is worth some more attention:
“I am there to listen to and enjoy music within an amplifier's (system's) capabilities, not too much of the time outside it.”
The maximum output amplitude of my CD player is governed by the full-scale output of the DAC. I have a test CD with some full-scale continuous sine wave test tones.
By simply setting the gain of my amplifier so that it delivers its rated output swing (with the volume setting at maximum, of course) when amplifying these test tones, I ensure that my system is set up such that the amplifier cannot be forced to operate beyond it’s limits. It’s as simple as that.
It’s a bit like setting up an RF transmitter for freedom from over modulation. Set up in this manner, my 10W per channel amp can still make a surprising amount of noise with my venerable old Pye speakers.
Not as much noise as could be had with a 400W per channel amplifier, for sure, but by no means inadequate for most sensible personal listening.
Cheers,
Glen
Re: Re: How important Clipping?
Your point doesn't catch all significant real life situations, in fact maybe only one out of ten.
Regards, Klaus
Think about amps with dynamic headroom (sagging supplies), users who dare to equalize the signal a little, who still use analog sources, who use preamps? Varying mains, Intersample overs? (*fill in what I forgot*)?G.Kleinschmidt said:
Your point doesn't catch all significant real life situations, in fact maybe only one out of ten.
Regards, Klaus
I think his point is that if you set your level such that the max Vout of your source does not overdrive your amp, clipping will not occur.
Example: if your CD player gives out 1VRMS, your amp has a gain of 30 and Voutmax of 30V (a little above 100W in 8 ohms) the system will not clip, whetever you feed the CD player and however you set the level control.
If your level control/preamp has additional gain you can compensate for that by limiting the max rotation of your level control.
Then, if the system still is not loud enough, get a power amp with more output volts. But the basic idea is that you CAN avoid clipping if you want.
Jan Didden
Example: if your CD player gives out 1VRMS, your amp has a gain of 30 and Voutmax of 30V (a little above 100W in 8 ohms) the system will not clip, whetever you feed the CD player and however you set the level control.
If your level control/preamp has additional gain you can compensate for that by limiting the max rotation of your level control.
Then, if the system still is not loud enough, get a power amp with more output volts. But the basic idea is that you CAN avoid clipping if you want.
Jan Didden
G.Kleinschmidt said:
Hi Glen,
You are exactly right about being able to set up an amplifier to explicitly prevent clipping from a known digital source with a well-defined maximum output. However, what is interesting, is that the recordings that are made with little compression and high crest factor are usually recorded at a substantially lower level. So in a typical home environment, when the volume is cranked on these to the point where the average SPL is satisfyingly realistic, the opportunity for big peaks to clip an amplifier really exists.
The fact that a doubling of power corresponds to a mere 3 dB tends to be at the root of clipping and dynamic range. If you have a recording with 6 dB greater crest factor than average, and speakers that are 6 dB lower in efficiency, you now need 12 dB more power at the same average listening level, which corresponds to four doublings of power, or a factor of 16. This can get ugly fast.
Take a look at the RMAF and HE2007 workshop writeups on my web page at www.cordellaudio.com, where we demonstrated this using a custom-built Peak-Average power meter that could accurately capture peak power (calibrated to 8 ohms) on transients as small as 10 us, and read it out as power on a digital dispaly that would hold the peak reading for one second so that it could easily be read.
With regard to speaker efficiency, I would bet that more than 50% of the people who buy John's JC-1 are playing it with speakers whose efficiencies are in the range of 83 to 87 dB SPL, 2.83V at 1 meter.
Cheers,
Bob
janneman said:
Works marvelously on the bench...not so well in the listening room. All you have to do is compare the sound to live music to see what's wrong. And, no, it doesn't take soi disant Golden Ears, just attention to detail.
This was all done to death in the '70s. If gazillions of dB of negative feedback were the answer, audio would have stopped cold by 1980. No need to do anything more. Pure, perfect sound, forever.
Oh, wait...that slogan's already been used.
How about: Thin, lifeless sound with exaggerated 'detail.'
There, that should do it.
A lot of people fell for it at the time. Hell, I know I did. Professor Ludwig had taught me that feedback was a gift from the gods. I believed him. I believed the spec sheets. I bought expensive gear that sounded worse and worse and worse as the feedback increased. Eventually I realized I'd taken a wrong turn and dumped all the crap. Bought tube gear (about the only way to get low feedback in those days). The music began to sound like music again.
The funny thing is, my reassessment started about the same time I started going to classical concerts, where there's no PA system to screw up the sound.
Hmmm...
Grey
Bob Cordell said:
Hi Bob.
This I do not dispute, but the point I was trying to get across is that my old Pye speakers are efficient enough to give a perfectly adequate SPL for most domestic listening coupled to 10W-per-channel amp never driven into clipping.
With efficient speakers, one can get away with surprisingly little power. Personally, I’d prefer a lower SPL to clipped and distorted peaks.
Cheers,
Glen
Grey,
Surely one could agree on this when lots feedback is used as a "cure" for rather nonlinear open-loop behaviour. But with today's top-notch op-amps (say LM4562 or TPA6120), for small signal use, do you think that "the feedback sound" is still thin and lifeless? I'd say no, it's just as neutral as can be.
Power amps are definitely more challenging, though I'm sure we'll see heavy feedback designs in near future (given the experts assembled here) that will break the "micro-detail distortion barrier" successfully, being perfectly neutral with even the most complicated loads.
Whether these will deliver a sound that is perceived as pleasant is a different question that is more related to psychoacoustics than to engineering art, IMHO. And maybe above all, a matter of "the optimum" amp for a given speaker. A generic approach, universal amps and universal speakers, doesn't look too promising to me, that's why I favor active speaker designs where an amp and a chassis can work together as one single topological unit and the designer has all freedom to choose/optimize any parameter with respect to the combined performance (like to extend the feedback to the speaker's actual acoustic response as it is done in some highly regarded active speakers: B&M/KS-Digital, Silbersand, ...)
Regards, Klaus (active musician, both acoustic and electric instruments)
Surely one could agree on this when lots feedback is used as a "cure" for rather nonlinear open-loop behaviour. But with today's top-notch op-amps (say LM4562 or TPA6120), for small signal use, do you think that "the feedback sound" is still thin and lifeless? I'd say no, it's just as neutral as can be.
Power amps are definitely more challenging, though I'm sure we'll see heavy feedback designs in near future (given the experts assembled here) that will break the "micro-detail distortion barrier" successfully, being perfectly neutral with even the most complicated loads.
Whether these will deliver a sound that is perceived as pleasant is a different question that is more related to psychoacoustics than to engineering art, IMHO. And maybe above all, a matter of "the optimum" amp for a given speaker. A generic approach, universal amps and universal speakers, doesn't look too promising to me, that's why I favor active speaker designs where an amp and a chassis can work together as one single topological unit and the designer has all freedom to choose/optimize any parameter with respect to the combined performance (like to extend the feedback to the speaker's actual acoustic response as it is done in some highly regarded active speakers: B&M/KS-Digital, Silbersand, ...)
Regards, Klaus (active musician, both acoustic and electric instruments)
janneman said:
Well Nelson,
Not to want to sidetrack, but you might explain - at least to me - why you call this entry by Jan controversial!
I see this as a purely scientific exercise (unless we have semantics - English is only my second language). To recreate recorded performances, means to me exact, such as rewriting a piece of text exactly as the original. How does one's senses come into that? It does not matter whether one likes the text or whether one's eyesight is good, to put it like that... thus whether the original (in this case the recorded performance) was perfect, bad or whatever in between. That is what I took Jan's statement to mean; he said recorded performance, not original.
I am saying this not to be funny or challenging, but often this (at least partly) lies at the root of a misunderstanding of the purpose of (in this case) high fidelity amplification. Not directed at you specifically; when folks talk about amplifiers that did not sound nice - who said that the original was nice? One was seldom there to be able to judge! It is a little more simple to exactly reproduce something than to have it sound "nice" (according to whose taste anyway?).
john curl said:Clipping? The last time I tried to drive my amps into audible clipping, the police came and closed me down!
Also, everyone had left the listening room due to the racket. They would have probably cited me for disturbing the peace, if I was not holding an SPL meter.
Most clipping in early solid state power amps was due to V-I protection circuitry, and it came in early with REAL loudspeaker loads. Tubes can't drive very much, so they have to clip gently.
Now let's use an example: My JC-1 and the WATT 1 loudspeaker (my personal speaker)
First the 1W sensitivity is 91dB, without a subwoofer. So at 10W, I should have 101 dB spl, and at 100W maybe 111 spl, and at 400W 117 spl.
Darn, a lease breaker if there ever was one! And that is for one channel only, without a subwoofer.
The lesson is: 'Real man' amps don't easily clip. That includes Nelson and Charles in this design group. It is only the wimpy solid state power amps of 50W or less that are going to have any significant problems, and I recommmend making a bigger amp, instead of putting a 'band-aid' on a small one.
By the way, nice work John, I like it:
http://www.stereophile.com/amplificationreviews/774/index6.html
Anyway, certainly with 91 dB/W speakers you are in much better shape than most. Still, I would not doubt if systems (not yours) are clipping much more often than people think. I was a kid, the first time I saw it on a scope when I was curious about what was making a system sound not quite right at high levels. It was about 80W/ch at the time. I now have over 1000W total biamped into my reasonably sensitive 2 ohm speakers. No I don't need all the power but it is nice to have the reserve.
Obviously, speaker efficiency has to be taken into consideration as far as how much power is required. However, I think that many people associate loud with distorted, knowingly or not, and so when they want to hear it loud they tend to keep turning it up until something distorts, usually the amplifier. I'm speaking of the average listener not necessarily experts.
On another subject, do you know anything about what happened to the Spica TC-50s after Parasound took over?
Pete B.
In similar vein as my last post I must make a clean breast yet again ....
I have a problem with these qualitive statements of "high NFB sounds horrible", "no NFB sounds best", "NFB is BAD" .... as if hearing is an indisputable judging faculty. Yes, this has been repeated often - but as repeatedly the above reappear as dogma, but with no proof better than the commentator's judgement.
I respect what a person says he can hear - after all, one cannot go about poking with a probe into his brain to verify his conclusions. As such then I owe him the plain well-mannered respect that all deserve, and I gladly do so.
But if I am going to be choked blue with labels because I cannot regard (subjective) observations as proof or at least dogma, then so be it. It was said that some decades ago the "NFB is bad" business was proved to death or such. Well, it has been proved (and reported) equally well to what degree hearing could be variable, notably by tests conducted at Scandenavian centres (the name of Prof. Bengt Sorensen comes up among many others).
"NFB is bad" - how much, in what circuits, with what loudspeakers, how designed, with what stability factors, what did measurements of the same show??? Perhaps someone can respectfully explain to me why I should not have a problem with the above utterances (and since this is a learning site)?
I have a problem with these qualitive statements of "high NFB sounds horrible", "no NFB sounds best", "NFB is BAD" .... as if hearing is an indisputable judging faculty. Yes, this has been repeated often - but as repeatedly the above reappear as dogma, but with no proof better than the commentator's judgement.
I respect what a person says he can hear - after all, one cannot go about poking with a probe into his brain to verify his conclusions. As such then I owe him the plain well-mannered respect that all deserve, and I gladly do so.
But if I am going to be choked blue with labels because I cannot regard (subjective) observations as proof or at least dogma, then so be it. It was said that some decades ago the "NFB is bad" business was proved to death or such. Well, it has been proved (and reported) equally well to what degree hearing could be variable, notably by tests conducted at Scandenavian centres (the name of Prof. Bengt Sorensen comes up among many others).
"NFB is bad" - how much, in what circuits, with what loudspeakers, how designed, with what stability factors, what did measurements of the same show??? Perhaps someone can respectfully explain to me why I should not have a problem with the above utterances (and since this is a learning site)?