Maybe. An amp & a speaker has to be considered a system. One should not be considered without the other.
dave
There are SS amps with high output impedance, and tube amps with low output impedance. Every loudspeaker benefits from an optimum output impedance,and that is not "as low as possible"
dave
High enough to cause FR errors that are at or above well-established JND thresholds with a given loudspeaker.
The lower the better. I know you're implying some voodoo witch nonsense about the evils of feedback or indulging in some other odd audiophool fetish, but...not interested.
Yes, there are some crappy solid state amps, particularly in the "high end" and hobbyist worlds. There are probably some genuinely good light bulb amps, but one needs to see FR measurements into a nonresistive load to separate them from the usual audiophool dross. But commodity solid state amps are both sonically transparent and fairly cheap. The only reasons not to use one are an interest in superior energy efficiency or a desire for smaller form factors. Then the better (ie load invariant) class D amps (Hypex, icepower asx2, Abletec, etc) rise to the top.
Certainly my main amp (PP EL34) is better than any ofmy SS amps,someof which are quite good.
dave
Wow…didn’t realise that would start such an ‘energetic’ discussion.
Perhaps I should have given more detail about my specific case in my OP rather than making it so broad.
In my case the tweeter is driven by the power section of a NAD 701, Damping factor (ref. 8Ω, 50Hz) >50, which is sort of typical of an SS amp.
I think the speaker is nominally 6 ohm, voice coil inductance 0.05mH
So because the amp impedance is an order of magnitude or two lower than the speaker (or the compensation) so I can now happily continue to use the speaker without worrying about compensation.
Thanks everyone!
Perhaps I should have given more detail about my specific case in my OP rather than making it so broad.
In my case the tweeter is driven by the power section of a NAD 701, Damping factor (ref. 8Ω, 50Hz) >50, which is sort of typical of an SS amp.
I think the speaker is nominally 6 ohm, voice coil inductance 0.05mH
So because the amp impedance is an order of magnitude or two lower than the speaker (or the compensation) so I can now happily continue to use the speaker without worrying about compensation.
Thanks everyone!
Hi. Seems I opened up a can of worms. Not my intention.
I 100% agree that good tube amps are good amplifiers.
I simply think that the aspects that draw in "tubephiles" - people who are looking for a distinctly so-called "tube" sound - are not interested in good amplifiers.
Even if 97 in 100 tube amplifiers have no difficulty with a loudspeaker not containing zobel compensation, if the most popular three in 100 do, and they outsell the other 97, then that's my angle at it.
This is the multi-way forum and I think we're all more interested in talking loudspeakers, not getting into amp arguments.
Cheers.
I'm unaware of any, unless they are "current drive" and intentionally (more or less) do this.
Please give some examples, if you can.
JND, fwiw is a rather high threshold, imo. Especially given how it is defined. It is possible to hear substantially below JND, but not by using the typical JND method.
The other problem is that a majority of all speakers have variations far greater than JND...
If there is a fetish (poor choice of phrases) it is yours. Ok?
The issues with negative feedback, since you bring it up are pretty well documented both here and elsewhere. Suggest you might want to read Bruno Putzey's article in a recent Linear Audio mag, if you have not already.
Glad you like translucent amps.
As long as you think those class D amps are all that you need or want, then that's what you think. To think that your, or my world view, is going to be the absolute truth is only good for the one who holds that view
Stereophile measures the FR of amps into a simulated loudspeaker load. That's a good place to start for examples.
Completely out of the scope of the discussion, of course.
Actually it's quite apt. Perhaps you'd consider the other alternative, religious fanatic, less harsh.
My weltanschauung here is simply "reality." Everything I've written is supported by every credible listening test ever done on the subject, and represents the consensus of the serious audio community.
I guess ur a very serious guy.
Dunno how you can term me a fanatic, it seems to fit your views far better than mine.
Regardless, throwing these terms about does nothing to further the discussion or back up your ideas with fact. And, since ur a serious audio guy, you ought to come with some substance.
JND is not beyond the scope of this conversation, since you cited it.
Based on how you described it and what you attributed to it, somehow I doubt that you really understand it, at least in this context. If you don't think that the variations in frequency response in a typical speaker need to be considered given your claim that some amplifiers are changing the response (apparently for the worse, and significantly) then what is it that you think is being changed or what is the problem?
Please cite what you think a "credible listening test" is? Has it been replicated by any others? Were there any controls to assure that what they claim to be testing for is actually what is/was being tested?
So far the listening tests that I have seen are far from dispositive, and the better ones yield some potentially useful information, but usually the "statistically valid results" are valid only for the actual test, and don't carry over to all situations.
By this is only my opinion. Probably wrong at that.
Dunno how you can term me a fanatic, it seems to fit your views far better than mine.
Regardless, throwing these terms about does nothing to further the discussion or back up your ideas with fact. And, since ur a serious audio guy, you ought to come with some substance.
JND is not beyond the scope of this conversation, since you cited it.
Based on how you described it and what you attributed to it, somehow I doubt that you really understand it, at least in this context. If you don't think that the variations in frequency response in a typical speaker need to be considered given your claim that some amplifiers are changing the response (apparently for the worse, and significantly) then what is it that you think is being changed or what is the problem?
Please cite what you think a "credible listening test" is? Has it been replicated by any others? Were there any controls to assure that what they claim to be testing for is actually what is/was being tested?
So far the listening tests that I have seen are far from dispositive, and the better ones yield some potentially useful information, but usually the "statistically valid results" are valid only for the actual test, and don't carry over to all situations.
By this is only my opinion. Probably wrong at that.
Btw, I see ur a follower of David Rich.
You say in your blog:
"The simple fact of the matter is that any non-broken, competently designed amplifier operating within its design limits will sound exactly the same as every other non-broken, competently designed amplifier operating within its design limits. That has been proven conclusively by controlled listening tests, to the point where anyone who claims otherwise is either simply ignorant of the relevant science, or willfully ignores it. The latter is often due to a pecuniary interest in propagating voodoo and mythology."
Let's skip the obvious vendors of snake oil.
But your statement is simply not true.
What is true is that below some threshold of (let's call it) resolution it will be very difficult to tell one amplifier apart from another.
If you read about Dr. Geddes' "GedLee Metric" and before that the work of D.E.L. Shorter, you may get an awakening, and some illumination on what is actually going on here.
The other thing that could be said by way of an analogy is:
"any properly designed and built new car will be a pretty good car to drive..."
Which is about what you have said, in effect.
Or maybe: "...all tires meet DOT standards in the USA, so all tires are equally good..."
If one only considers first order, or low order effects and no higher order effects, then everything does tend to bunch together and appear the same.
'nuff said here.
You say in your blog:
"The simple fact of the matter is that any non-broken, competently designed amplifier operating within its design limits will sound exactly the same as every other non-broken, competently designed amplifier operating within its design limits. That has been proven conclusively by controlled listening tests, to the point where anyone who claims otherwise is either simply ignorant of the relevant science, or willfully ignores it. The latter is often due to a pecuniary interest in propagating voodoo and mythology."
Let's skip the obvious vendors of snake oil.
But your statement is simply not true.
What is true is that below some threshold of (let's call it) resolution it will be very difficult to tell one amplifier apart from another.
If you read about Dr. Geddes' "GedLee Metric" and before that the work of D.E.L. Shorter, you may get an awakening, and some illumination on what is actually going on here.
The other thing that could be said by way of an analogy is:
"any properly designed and built new car will be a pretty good car to drive..."
Which is about what you have said, in effect.
Or maybe: "...all tires meet DOT standards in the USA, so all tires are equally good..."
If one only considers first order, or low order effects and no higher order effects, then everything does tend to bunch together and appear the same.
'nuff said here.
AFAIC, Class D amplifiers are far from load invariant. Most, if not all of them, to maintain stability, are forced to drastically reduce their damping factor to tube amp, or lower figures above a few khz to allow their passive LC filters to reduce the massive switching noise, more than a little of which leaks out anyway. A medical electrical engineer with whom we are working has found that a Class AB amplifier works much better than a Class D type as a driver for resonant ultrasonic vibrator applications because the Class D amps he used modulate their switching frequency too much to the input signal, literally changing its fundamental frequency by up to 100 hz and throwing the sonic vibrator out of resonance intermittently due to cross modulation with the Class D switch frequency.
Pardon me if I find the idea that adding a quarter ohm resistor in series with the output of a zero ohm output voltage amplifier would 'ruin' its quality humorous since most peoples' speaker cables add more resistance than that. Har de har har!
A zero ohm output impedance is not really some holy grail of amplifier 'quality'. My current DC coupled OTL, incidentally, will maintain a DF of 100 up to 20 khz which is more than many SS amps are capable of. Even the first amps I owned at age 17, a pair of 1950's era consumer Bogen amplifiers each had a damping control that would let the user cross right through a zero ohm output impedance to negative impedance, allowing people in the light bulb amp days to compensate for their speaker cables.
Pardon me if I find the idea that adding a quarter ohm resistor in series with the output of a zero ohm output voltage amplifier would 'ruin' its quality humorous since most peoples' speaker cables add more resistance than that. Har de har har!
A zero ohm output impedance is not really some holy grail of amplifier 'quality'. My current DC coupled OTL, incidentally, will maintain a DF of 100 up to 20 khz which is more than many SS amps are capable of. Even the first amps I owned at age 17, a pair of 1950's era consumer Bogen amplifiers each had a damping control that would let the user cross right through a zero ohm output impedance to negative impedance, allowing people in the light bulb amp days to compensate for their speaker cables
.
Last edited:
sreten, the folks at KEF would like to politely disagree with you,
I expect. Brothers from the British Isles, yes?
Hi,
In what way and what contex ? I doubt they disagree with the plain facts.
rgds, sreten.
sreten, KEF built speakers that were specifically "compensated" to maintain a nearly non-reactive load looking back at the amplifier.
Undoubtedly, following your line of reasoning the engineers at KEF were wasting their time going for something they shouldn't have? It was a waste of time? Had no effect, no benefit?
Which plain facts would you like to present, sir?
Undoubtedly, following your line of reasoning the engineers at KEF were wasting their time going for something they shouldn't have? It was a waste of time? Had no effect, no benefit?
Which plain facts would you like to present, sir?
If we change the statement to read 'competently designed accurate amplifier' then it becomes a self-defining statement. So if two amplifiers sound different, then one doesn't meet the 'competently designed accurate amplifier' clause.
Hi,
Your talking utter nonsense in the context of the OP.
And FWIW that phase of KEF marketing BS where they
"compensated" (mostly) some of their speakers to be
near resistive (except the real bottom end resonance)
so poor amplifiers didn't have to worry about reactive
loads, was and is still a total waste of time and effort.
They don't do it anymore and you can't spot spin.
Proper speaker design is not made better by "compensation".
rgds, sreten.
Competent valve amplifiers simply still prefer high
impedance loads and don't care if its also reactive.
Building speakers only justifiable for poor high
impedance valve amplifiers makes no sense,
however KEF's BS was about amplifier power.
Last edited:
I was at KEF when this was developed but it was not one of my projects. It was first developed for the 104.2, with probably the most complex passive crossover ever found on a system.
Laurie Fincham speculated that some amplifiers would be happier with a totally resistive load with no phase angle at any audio frequencies. He thought that some amps would be more tolerant of 4 ohms if it was a resistive 4 Ohms. We did know that amps were sensitive to the minimum real part of their impedance curves and had seen Quad get into trouble with their 405 Series 1 and its over eager current protection.
It was all speculation and no listening trials were conducted.
Essentially, he did it because he could.
David S
What kind of speaker cable has more than 0.25 Ohm?
Mine has <4.5 Ohm per km or less than 0.045 Ohm for the length I use.
The thinnest speaker cable (1.5mm^2) would give about 0.13 Ohm for a 5m run (10m there and back).
Microphone cable has a bit more than 0.25 Ohm for a normal speaker cable length but nobody in their right mind would use that for speaker duty.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.