Maybe for women.
Their hearing usually exceeds ours.
And only had one girlfriend who complained about metal dome tweeter.
But only certain songs.
Otherwise was very clean tweeter with no fatigue.
Most the time was electronic mixes and very common for DJ to push highs to much.
Actually common for males to push highs too much.
So electronic sounds sounded harsh to her on some tracks.
This problem is noted many times though, specially live performance.
But can guarantee most the fatigue is speaker not amplifier.
.01% is more than fine for 20k
as mentioned many many Class D amps dont do much better than .1% at 20k
and in blind test with your most familiar, most favorite track ever.
Blind test likely change opinion. Since most is assumption based on numbers.
If we are testing with 3 inch " wideband" which is fine to many.
On paper high end is jagged regardless and actually makes no difference.
People dont realize how high pitch even 8k is.
20k is harmonics actually more like 12k is the tip top of harmonics in most music.
And the microphone is recorded with, even best one. is making more distortion.
Difference between metal dome and 3" wideband is clarity in top harmonics.
" Detail" like maybe magazine or review will say.
Amplifier distortion at .1 or .01%
Not really. If people make claims with wideband as reference system, no dont listen.
Maybe only metal dome will expose a recording. I only hear microphone distortion
in Jazz recordings. which I know from studio work and buying countless microphones.
Amplifier is way way below that. And most dont even know microphone distortion.
It just sound like normal music to them.
Their hearing usually exceeds ours.
And only had one girlfriend who complained about metal dome tweeter.
But only certain songs.
Otherwise was very clean tweeter with no fatigue.
Most the time was electronic mixes and very common for DJ to push highs to much.
Actually common for males to push highs too much.
So electronic sounds sounded harsh to her on some tracks.
This problem is noted many times though, specially live performance.
But can guarantee most the fatigue is speaker not amplifier.
.01% is more than fine for 20k
as mentioned many many Class D amps dont do much better than .1% at 20k
and in blind test with your most familiar, most favorite track ever.
Blind test likely change opinion. Since most is assumption based on numbers.
If we are testing with 3 inch " wideband" which is fine to many.
On paper high end is jagged regardless and actually makes no difference.
People dont realize how high pitch even 8k is.
20k is harmonics actually more like 12k is the tip top of harmonics in most music.
And the microphone is recorded with, even best one. is making more distortion.
Difference between metal dome and 3" wideband is clarity in top harmonics.
" Detail" like maybe magazine or review will say.
Amplifier distortion at .1 or .01%
Not really. If people make claims with wideband as reference system, no dont listen.
Maybe only metal dome will expose a recording. I only hear microphone distortion
in Jazz recordings. which I know from studio work and buying countless microphones.
Amplifier is way way below that. And most dont even know microphone distortion.
It just sound like normal music to them.
In my eyes it depends on how that 0,001% at 20 kHz are achieved.
Without global feedback it would be extremely impressive.
With global feedback it might be impressive. Depends on the amount of feedback used.
For every magnitude of ten in decreasing distortion at the output, the input stage increase by a magnitude of 10.
I have seen bad cases where distortion at input stage can be both 1 and 10% of distortion when output stage deliver 0,001%.
So a distortion figure say nothing by itself.
You have to investigate all stages in an amplifier to fully understand it's potential to overall good sound.
Just as a comment, I set target for distortion at listening level to a maximum of 0,05% if 2nd harmonic and 0,02% if 3rd or higher harmonics when testing with a single sine wave test tone.
The reason for that is my detection level of 0,6% 2nd harmonic and 0,25% for 3rd harmonic of distortion.
The difference in a magnitude of ten is because IMD distortion can be much higher during playback of music, so some margin is essential.
I have the same target for all stages in an amplifier design regardless of the amount of global feedback used.
Last edited:
High distortion in the input stage with clean output is a good sign. It means that almost all amplifier distortion is removed from the output.
Feedback works by - ideally - only have the distortion at the input stage, in appropriate phase.
Lets run some numbers. Assume you have an amp that has a gain before feedback of 100,000 and an output of 1V.
The input to that amp is then only 10uV.
Also assume it has 1% distortion, and we close the feedback loop for a gain of 20, and an input signal of 1V/20 = 50mV.
(But remember that signal-wise, the amp itself sees still the 10uV between its inputs, or input and ground, which is the difference between the source signal and the feedback signal!).
We then 'must have' a distorted input signal of 1% of 1V /20 = 100uV at the amp terminal.
That swamps the 10uV input signal and looks like a huge input stage distortion. Which is good.
Jan
Feedback works by - ideally - only have the distortion at the input stage, in appropriate phase.
Lets run some numbers. Assume you have an amp that has a gain before feedback of 100,000 and an output of 1V.
The input to that amp is then only 10uV.
Also assume it has 1% distortion, and we close the feedback loop for a gain of 20, and an input signal of 1V/20 = 50mV.
(But remember that signal-wise, the amp itself sees still the 10uV between its inputs, or input and ground, which is the difference between the source signal and the feedback signal!).
We then 'must have' a distorted input signal of 1% of 1V /20 = 100uV at the amp terminal.
That swamps the 10uV input signal and looks like a huge input stage distortion. Which is good.
Jan
Last edited:
@jan.didden
As always, it depends. IMHO.
If you need to limit bandwidth of the input stage into audible range it's a bad thing.
I prefer listening to the harmonics from the recorded signal rather than the harmonics from the output stage.
Too much feedback and a limited bandwidth of input stage create listening fatigue.
As always, it depends. IMHO.
If you need to limit bandwidth of the input stage into audible range it's a bad thing.
I prefer listening to the harmonics from the recorded signal rather than the harmonics from the output stage.
Too much feedback and a limited bandwidth of input stage create listening fatigue.
Recently a friend compared Hiraga class A amplifier with my amplifier (a typical blameless topology with TMC compensation). He prefer the timbre of Hiraga amp, but Hiraga amp have bad imaging stereo. I think he like the timbre of class A amplifier because of harmonic profile the distortion.
Yeah I've heard that myth a lot.
I tried to debunk it in an aX column but not many read that apparently.
Jan
@jan.didden
If an article contradicts what you see and experience it's a bitter pill to swallow.
So you have to push it down in my throat for sure 🙂
If an article contradicts what you see and experience it's a bitter pill to swallow.
So you have to push it down in my throat for sure 🙂
Not really, if you realize that it is Standard Procedure for your mind to play tricks on you.
That's what we are.
Science and the scientific method are pretty unnatural for us because it involves getting against your ego and actually think,
and be ready to admit you had it wrong. THAT's hard to swallow; some people never get to that in their entire life!
Jan
That's what we are.
Science and the scientific method are pretty unnatural for us because it involves getting against your ego and actually think,
and be ready to admit you had it wrong. THAT's hard to swallow; some people never get to that in their entire life!
Jan
Last edited:
@jan.didden
In this case I will be stuborn and hard to convince.
If I design an amplifier with no feedback at all the input stage, VAS stage usually have much less and different distortion to the output stage.
Depending on bias and how much of class A operation of course.
Once feedback is introduced from output to input stage will make the input stage produce a totally different setup of harmonics that are induced from the output stage. Input stage take a hugh leap of more distortion that wasn't there without feedback.
By increasing feedback the problem worsen for the input stage, and eventually you need to limit the frequency range of the input stage to keep the amplifier out of oscillation.
With a limited frequency range the input stage have less capacity to pick up necessary harmonics from the recording and more or less have a sonic signature similar to the output stage cross over distortion.
If limited to much it will start to sound monotonic and all acoustic instruments starts to sound anemic and without the air, ambience and bloom that a correct handling of harmonics in the recording gives.
The amplifier simply can not convey the harmonics in the correct way if bandwidth is too limited because of the amount of feedback.
I see it in measurement and can hear it, so I will be a hard fellow to convince into thinking in a different way.
Sorry, but that is the way it is. For me at least...
In this case I will be stuborn and hard to convince.
If I design an amplifier with no feedback at all the input stage, VAS stage usually have much less and different distortion to the output stage.
Depending on bias and how much of class A operation of course.
Once feedback is introduced from output to input stage will make the input stage produce a totally different setup of harmonics that are induced from the output stage. Input stage take a hugh leap of more distortion that wasn't there without feedback.
By increasing feedback the problem worsen for the input stage, and eventually you need to limit the frequency range of the input stage to keep the amplifier out of oscillation.
With a limited frequency range the input stage have less capacity to pick up necessary harmonics from the recording and more or less have a sonic signature similar to the output stage cross over distortion.
If limited to much it will start to sound monotonic and all acoustic instruments starts to sound anemic and without the air, ambience and bloom that a correct handling of harmonics in the recording gives.
The amplifier simply can not convey the harmonics in the correct way if bandwidth is too limited because of the amount of feedback.
I see it in measurement and can hear it, so I will be a hard fellow to convince into thinking in a different way.
Sorry, but that is the way it is. For me at least...
Post # 103. What you think is input stage distortion is the inverse of the output distortion that is introduced by the fb to the input to cancel it at the output.
And it is not the 'limited frequency' range that is of importance - how would you define that??
It is how many loop gain you have at the frequency of interest.
If two amplifiers have the same loop gain at 20kHz they have the same distortion at 20kHz.
The one with the narrow OL bandwidth will have huge more loop gain at lower frequencies thus less distortion at lower frequencies than the wide bandwidth thing. That's good.
Edit: if you are interested: https://audioxpress.com/article/audio-myths-why-narrow-bandwidth-may-be-better
Jan
And it is not the 'limited frequency' range that is of importance - how would you define that??
It is how many loop gain you have at the frequency of interest.
If two amplifiers have the same loop gain at 20kHz they have the same distortion at 20kHz.
The one with the narrow OL bandwidth will have huge more loop gain at lower frequencies thus less distortion at lower frequencies than the wide bandwidth thing. That's good.
Edit: if you are interested: https://audioxpress.com/article/audio-myths-why-narrow-bandwidth-may-be-better
Jan
Last edited:
@jan.didden
Lets not get into a lengthy discussion.
I know what I know and you know what you know and we will have a hard time to agree on this topic.
Difference in experiences and as a result of that difference in opinions.
No big deal...
Two amplifiers, one with BJT output and one with Mosfet output. Same open loop gain and frequency response and same loop gain with feedback.
Will those have the same amount of distortion at 20 kHz?
Will the harmonic structure of the distortion be the same?
Sorry, had to comment oon that one since it seemed very odd for a statement. Maybe I misunderstood you.
Lets not get into a lengthy discussion.
I know what I know and you know what you know and we will have a hard time to agree on this topic.
Difference in experiences and as a result of that difference in opinions.
No big deal...
Two amplifiers, one with BJT output and one with Mosfet output. Same open loop gain and frequency response and same loop gain with feedback.
Will those have the same amount of distortion at 20 kHz?
Will the harmonic structure of the distortion be the same?
Sorry, had to comment oon that one since it seemed very odd for a statement. Maybe I misunderstood you.
Last edited:
Maybe, maybe not. And that's not what I said, so I don't know why you pluck that out of the air.
Two amplifiers with the same ol distortion at 20kHz and the same loop gain have the same distortion after feedback.
That should have been clear in the context.
It's hard to cram 100 years of design understanding in a 200 word post so you rely on the understanding of your readers.
Jan
I'm working, and with this tempo in conversation I didn't hade the time to read it, only skimming. Will read it in more detail later today.
Opinion for now is that your reasoning is very much contradictory to my experience of designing audio amplifiers.
Sorry if I misunderstood you.
I just reacted to this:
"If two amplifiers have the same loop gain at 20kHz they have the same distortion at 20kHz."
There are so much that influence the amount and structure of distortion that this seemed a bit odd to me.
Change the bias point and stand by current in one of two other wise exactly alike amplifiers will give different amount and structure of distortion.
Opinion for now is that your reasoning is very much contradictory to my experience of designing audio amplifiers.
Sorry if I misunderstood you.
I just reacted to this:
"If two amplifiers have the same loop gain at 20kHz they have the same distortion at 20kHz."
There are so much that influence the amount and structure of distortion that this seemed a bit odd to me.
Change the bias point and stand by current in one of two other wise exactly alike amplifiers will give different amount and structure of distortion.
Last edited:
Well I would be interested in how you documented and measured stuff that led to your opinion.
Measurements clearly contradict you.
Jan
Measurements clearly contradict you.
Jan
Now I have read your article and it completely contradicts the experience of mine.
We will never agree, and I think you are sorely mistaken about the importance of the input stage ability to convey recorded harmonics that are so important for the music we listen too.
With too limited bandwidth and too much feedback the input stage will take on the sonic character of the output stage.
I have tried to, but never succeeded in using Op-amps just because of this reason.
They simply have a sound of their own and have very limited ability to convey the recorded signal with all harmonic structure intact.
They will be altered by the too dominant distortion created by the output stage and sound sterile and monotonic.
We will never agree, and I think you are sorely mistaken about the importance of the input stage ability to convey recorded harmonics that are so important for the music we listen too.
With too limited bandwidth and too much feedback the input stage will take on the sonic character of the output stage.
I have tried to, but never succeeded in using Op-amps just because of this reason.
They simply have a sound of their own and have very limited ability to convey the recorded signal with all harmonic structure intact.
They will be altered by the too dominant distortion created by the output stage and sound sterile and monotonic.
Last edited:
So, this is a clear sign of high linearity of the amplifier. The same effect occurs with high linearization of the speaker system.
Last edited:
So your perceptions contradict the facts. Then you have to make a choice; I don't have to spell it out for you.
The earth is the center of the universe, everyone can clearly see that.
Those astronomers must be wrong!
Nice talking to you, all the best,
Jan
Exactly! With a high linearity amp and speaker you hear what the record producer intended.
You may not like it, and you may want a bit more 'warmth' or whatever, but then you are 2nd guessing the producer.
Which is OK, taste is personal.
Some would go so far as adapting famous art for personal preference.
Jan
Attachments
In fact, it gets even worse, in a highly linear speaker system, you start to hear what the sound engineer hid behind the non-linearity of his speakers. All the mixing flaws are very easily audible, I even easily heard bad track gluing on some tracks, I could easily separate the instrument tracks from the vocal track in my head. The sound is so sterilely clean that all the mixing dirt is visible.