Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Audible symptoms of amplifier clipping (Read 52237 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Audible symptoms of amplifier clipping

Reply #75 –
Quote from: KumarK on
There are now many cheap amplifiers in the market, made in China, that have quoted THD/N of 10%. At about half the power specified for the amp, this comes down to 1% or less.
What is this distortion?


On a good day it is clean clipping, but my bench tests suggest that Class D amplifiers, particularly those of the Tripath persuasion don't clip cleanly. Their nonlinear distortion rises signficiantly just below clipping.

Quote
Is it damaging to the speakers as well as to the sound quality?


In general clipping is not damaging to speakers unless it is highly audible. In my book anybody who pushes their amps into clearly audible clipping deserves the results that he obtains.

Audible symptoms of amplifier clipping

Reply #77 –
Well, a speaker dipping down to 3 ohms (sometimes even 2 ohms) in the bass area and a 45° phase angle can mean serious trouble for some amplifiers.

Now if all evidence you've got is some "vague feeling that the amp doesn't provide enough power" or something of that sort ... nobody can tell what's going on. Could be anything from imagination/bias to slight clipping.
"I hear it when I see it."

Audible symptoms of amplifier clipping

Reply #79 –
Quote from: Arnold B. Krueger on
Quote from: KumarK on
There are now many cheap amplifiers in the market, made in China, that have quoted THD/N of 10%. At about half the power specified for the amp, this comes down to 1% or less.
What is this distortion?

On a good day it is clean clipping, but my bench tests suggest that Class D amplifiers, particularly those of the Tripath persuasion don't clip cleanly. Their nonlinear distortion rises signficiantly just below clipping.

Further question then, on these amps. Suppose there is such an amp that does 100wpc with 10% THD. Suppose that the amp does 50wpc with 1% THD.
I am trying to reconcile this with my understanding of clipping. Which is that when the speakers pull more power than the amp is able to deliver, the sine wave shape of the power delivered gets its curves clipped resulting in a square shape, causing distortion in sound, and beyond a point, damage to the speakers. I thought that the amp tries to deliver the power demanded, but all it can do is the aforesaid.
But the example above indicates that the amp goes from 50wpc to 100 wpc delivery, not an insignificant increase in power, but with more THD/clipping.
Where is my understanding wrong?

Audible symptoms of amplifier clipping

Reply #80 –
Quote from: KumarK on
Quote from: Arnold B. Krueger on
Quote from: KumarK on
There are now many cheap amplifiers in the market, made in China, that have quoted THD/N of 10%. At about half the power specified for the amp, this comes down to 1% or less.
What is this distortion?

On a good day it is clean clipping, but my bench tests suggest that Class D amplifiers, particularly those of the Tripath persuasion don't clip cleanly. Their nonlinear distortion rises signficiantly just below clipping.

Further question then, on these amps. Suppose there is such an amp that does 100wpc with 10% THD. Suppose that the amp does 50wpc with 1% THD.


IME an amp that puts out 100 watts @ 10% THD might have far less than 1% THD at 50 watts. It might be 0.01% THD @ 50 watts.

Quote
I am trying to reconcile this with my understanding of clipping. Which is that when the speakers pull more power than the amp is able to deliver,


Speakers don't pull power.  The amp rams it down their throats using a stick called "Voltage".

Quote
the sine wave shape of the power delivered gets its curves clipped resulting in a square shape,


That happens because the nut holding the volume control is drunk or otherwise mentally incapacitated.

Quote
causing distortion in sound, and beyond a point, damage to the speakers.


In this discussion it is power, not distortion that damages the speakers. Distortion's role in the story is that by tolerating distortion, the nut holding the volume control is able to turn it further clockwise, and the resulting power fries the drivers, etc.

Quote
I thought that the amp tries to deliver the power demanded, but all it can do is the aforesaid.


The amp tries to do what it is told to do, nothing more. 

By tolerating distortion it is capable of delivering more long average power to the speaker that we might expect from its ratings for undistorted power.

The numbers that started this discussion out - 100 watts @10% THD get even more impressive if you tolerate much more than 10% THD. @50% THD it might put out 300 watts.

Audible symptoms of amplifier clipping

Reply #81 –
50 to 100W is "only" a +3 dB difference.

Indeed, what kills drivers is too much power for too long.
"I hear it when I see it."

Audible symptoms of amplifier clipping

Reply #82 –
Quote from: KumarK on
But the example above indicates that the amp goes from 50wpc to 100 wpc delivery, not an insignificant increase in power, but with more THD/clipping.
Where is my understanding wrong?


Clipping is not usually totally "hard" - it is not like the voltage can go up to a set level totally linearily, and then stop like you had run into a brick wall. Clipping is usually more gradual, so increases once you get closer to the peak power limits.

There are some amps (have a look at some of the cheaper class D ones) where the distortion-as-a-function-of-output-power graph really goes vertical at some point.

Audible symptoms of amplifier clipping

Reply #83 –
Quote from: julf on
Clipping is not usually totally "hard" - it is not like the voltage can go up to a set level totally linearily, and then stop like you had run into a brick wall. Clipping is usually more gradual, so increases once you get closer to the peak power limits.


I've observed several 100 tubed and  SS power amps clip on the test bench, and in general the clipping is as close to being totally hard as one can practically get.  At 1 or even 0.2 dB below clipping, the THD can easily be as low as 0.02% or less.

The output voltage of most power amps does indeed go to the (+) and (-) rails within a volt or two more or less quite linearly and stop on the proverbial dime. 

The hard clipping is a consequence of negative feedback, which virtually  all but a few intentionally designed-to-be pathological power amps have. It even happens in well designed tubed amps.

Audible symptoms of amplifier clipping

Reply #84 –
Sorry to drag this thread up, but I have a question.

If a tweeter pops without hearing any distortion, does that mean too much power was applied, or that the tweeter was underpowered, and clipped?

Audible symptoms of amplifier clipping

Reply #85 –
Simply too much power can damage a tweeter, there need not be any clipping. Lots of compression on the recording played at high SPL can damage a tweeter without any added distortion from the amp.
"I hear it when I see it."

Audible symptoms of amplifier clipping

Reply #86 –
Quote
or that the tweeter was underpowered, and clipped?
That's mostly a myth...    For example a signal that goes to 100W is likely more dangerous to a tweeter than the same signal clipped-down to 50W.

Yes, clipping generates high frequency harmonics.    But there is a limit to the high frequency energy, even in a worst-case square wave.  A 100Hz square wave (or very-badly clipped 100Hz sine wave) contains harmonics that go to infinity (in frequency) but there's not much amplitude/energy in the higher frequencies...  A high-power 100Hz square wave is more likely to burn-out your woofer than your tweeter.

You can easily burn-out a tweeter with continuous test tones.    Depending on the tweeter you can probably burn it up before it distorts, and if the test-tone frequency is high enough, any harmonic distortion can be above your hearing range.    You might hear lower-frequency rattling or buzzing, but probably not harmonics.  And by the time you hear rattling or buzzing, the tweeter is probably already damaged.   

I've blown a few tweeters.    I don't remember hearing anything going wrong, the tweeter just went silent.    (The last time I thought I'd blown a tweeter it was a problem with the active crossover in my van's tri-amped system.)

Audible symptoms of amplifier clipping

Reply #87 –
[blockquote] .."the harmonic distortion components from a clipping amplifier can damage or totally burn out a mid-range or tweeter in the speaker system that it's driving. The harmonic products generated by the distortion shift the energy spectrum to higher frequencies. Experience has proven that a higher power amplifier in the same situation will not damage the speakers, simply because it isn't driven into clipping." [emphasis mine]

-Roger Russell,  Director of Acoustic Research at McIntosh Laboratory, Inc. for 25 years and the originator of McIntosh Loudspeakers
[/blockquote]

Audible symptoms of amplifier clipping

Reply #88 –
Quote from: DVDdoug on
You can easily burn-out a tweeter with continuous test tones.    Depending on the tweeter you can probably burn it up before it distorts, and if the test-tone frequency is high enough, any harmonic distortion can be above your hearing range.    You might hear lower-frequency rattling or buzzing, but probably not harmonics.  And by the time you hear rattling or buzzing, the tweeter is probably already damaged.
This is true. You can very easily destroy tweeters with what seem like fairly reasonable levels of high frequency test tones, without going near clipping.

Cheers,
David.

Audible symptoms of amplifier clipping

Reply #89 –
Here's an interesting read:
http://sound.westhost.com/tweeters.htm

If I understand correctly, the author makes a case that the additional harmonics caused by distortion only play a minor role. According to him, the main cause for trouble is the much increased average power.

Assuming he is correct, I conclude that an amplifier with more power would be of little benefit, which would contradict Mr. Russel.

Audible symptoms of amplifier clipping

Reply #90 –
I've read other similar accounts alluding to "It isn't the clipping, per se, which fries the drivers, usually tweeters first, it is the audio compression which is to blame." Rane (which among other things sells limiters) would fall into this camp. But what exactly is it which causes an amplifier to suddenly start compressing its output? OVER DRIVING IT INTO CLIPPING. So regardless of the exact mechanics involved, all a consumer needs to know is that clipping is indeed dangerous and substituting a stronger amp that doesn't clip at that exact same point should indeed be safer, although one may only eek out a tiny increase in output before THAT amp also has problems and you must be nearing the point where the tweeter is going to blow even with a perfectly clean signal, so the benefits may be marginal in some instances.

Audible symptoms of amplifier clipping

Reply #91 –
Quote from: mzil on
I've read other similar accounts alluding to "It isn't the clipping, per se, which fries the drivers, usually tweeters first, it is the audio compression which is to blame." Rane (which among other things sells limiters) would fall into this camp. But what exactly is it which causes an amplifier to suddenly start compressing its output? OVER DRIVING IT INTO CLIPPING. So regardless of the exact mechanics involved, all a consumer needs to know is that clipping is indeed dangerous and substituting a stronger amp that doesn't clip at that exact same point should indeed be safer, although one may only eek out a tiny increase in output before THAT amp also has problems and you must be nearing the point where the tweeter is going to blow even with a perfectly clean signal, so the benefits may be marginal in some instances.


Misses the fact that if you replace a smaller amp that does clip with a larger amp that does not clip, you increase the average power that the speaker driver receives.

It is the excess average power that destroys speaker drivers, and the results are similar whether you obtain this excess power with or without clipping.

Clippng's role is that it is a way to obtain far more average power from an amp than you might expect.

Audible symptoms of amplifier clipping

Reply #92 –
The added more powerful amp reproducing the same output level signal [but now without any clipping distortion, and without any compression, and without waveform distortion which increases the average power to an abnormal, excess value, since unlike the smaller amp it has just replaced, this amp is not being overdriven, yet] does NOT fry the tweeter as quickly, hence it can be correctly said to be effectively adding a bit of added tweeter protection.

Roger Russell the McIntosh Labs acoustic research director for 25 years, JBL's white paper, and the Rane sponsored AES paper are all correct that driving an amp into distortion can be harmful, regardless of their quibbling over the exact mechanics of why a weaker amp that distorts and fries a tweeter is dangerous, yet replacing it with a stronger amp reproducing that same attempted level, but does not distort, will not.

https://secure.aes.org/forum/pubs/conventions/?elib=5737

Audible symptoms of amplifier clipping

Reply #93 –
Quote from: mzil on
The added more powerful amp reproducing the same output level signal [but now without any clipping distortion, and without any compression, and without waveform distortion which increases the average power to an abnormal, excess value, since unlike the smaller amp it has just replaced, this amp is not being overdriven, yet] does NOT fry the tweeter as quickly, hence it can be correctly said to be effectively adding a bit of added tweeter protection.

Roger Russell the McIntosh Labs acoustic research director for 25 years, JBL's white paper, and the Rane sponsored AES paper are all correct that driving an amp into distortion can be harmful, regardless of their quibbling over the exact mechanics of why a weaker amp that distorts and fries a tweeter is dangerous, yet replacing it with a stronger amp reproducing that same attempted level, but does not distort, will not.

https://secure.aes.org/forum/pubs/conventions/?elib=5737



I have that paper right here before me, and the above interpretation of it seems to tax  my imagination:

"I found the newer and better sounding amplifiers (including amplifiers with soft clipping circuits) had an even
greater occurrence of blown tweeters than amplifiers with lower power. And, amplifiers with higher power
were having fewer incidences of blown tweeters. This appeared to reinforce the theory that tweeter blowouts
were occurring because of clipping. One thing was clear, when clipping occurred, tweeters blew.If you're
getting the idea I don't believe in the clipping/harmonic theory, you're right.

As you can see, the total amount of power left to make it through the crossover at the tweeter frequencies (say
above 2kHz) is less than two watts[ Remember, this simulates severe overdrive of a 100 watt amplifier with a
sine wave to make an ideal square wave. If you drove it harder, the harmonics would not increase.
This analysis shows that a small tweeter that only handles 5 or 10 watts is not going to blow out.
It takes a lot more than this to cause major failure. 8o what's happening?

Compression is what's happening [3].

The newer, higher quality amplifiers have a greater dynamic range and sound better when clipped with musical
transients. Therefore a listener may overdrive and clip the newer amplifiers on transients more often with less
audible distortion. This results in compression of the dynamics of the music source.

We can use a spectrum
analyzer (time averaging) to show spectral compression. For example, in a 100 watt amplifier, the low frequency
components limit at 100 watts while the high frequency components continue to increase until they approach
100 watts at the clipping point. The following graphs are scaled in volts. If the load impedance is 8 Ohms, then
the 100 watt level corresponds to 40 volts peak on the graph. Below clipping, the low frequency components
reach 100 watts (40 volts peak) but the high frequency components are oniy 5 or 10 watts (9 to 13 volts peak).

Let's assume a musical signal with !ow and high frequency components driving a 100 watt amplifier. We use a
low level/high frequency sinewave mixed with a high level/low frequency sinewave burst. (See Fig.2). The high
frequencies reproduced by the tweeter are at least 10 dB lower in level than the low frequencies. Now we turn
up the amplifier to clip the signal (3dB overdrive- See Fig'3). Notice that only the Iow frequency burst portion of
the waveform clips but the high frequency portion increases in level. The clipping, of course, produces
harmonics but not nearly as much as the square waves discussed earlier. The amplitude of the high frequencies
went up by 3dB in relation to the low frequency fundamental.
"

..and so on..

Audible symptoms of amplifier clipping

Reply #94 –
This is what is so confusing about these subjects.

Expert 1 chimes in and says that it's not the distortion, it's the excess power that does the damage. Expert 2 chimes in and says that the added distortion can result in speaker damage.

Expert 3 chimes in and says that the added distortion together with excess power does the damage, but depending on the conditions, may be a mix of the two.

Expert 4 (Jneutron) who is Arnolds biggest nightmare opponent has a completely different argument and I can't even begin to understand it, but it's a lot more technical than any of the explanations provided thus far.

Audible symptoms of amplifier clipping

Reply #95 –
That's why we provided a very simple answer for you: too much power kills speakers.

But you won't accept simple answers, because they are simplifications of what is really going on (doh). But you only seem to understand the simple answers. You do not grasp detailed, specific answers and then think the answers are wrong because they are different from the simple answer, because more detailed answers deal with more specific situations (doh x2).

I'm at a loss at how anyone could help you Rich B.
"I hear it when I see it."

Audible symptoms of amplifier clipping

Reply #96 –
Quote from: xnor on
That's why we provided a very simple answer for you: too much power kills speakers.

But you won't accept simple answers, because they are simplifications of what is really going on (doh). But you only seem to understand the simple answers. You do not grasp detailed, specific answers and then think the answers are wrong because they are different from the simple answer, because more detailed answers deal with more specific situations (doh x2).

I'm at a loss at how anyone could help you Rich B.


Yes, too much power kills speakers. But there are cases where too much power was not responsible for killing speakers. Ask Jneutron, who is probably more knowledgeable than most of the folk here combined.

Audible symptoms of amplifier clipping

Reply #97 –
Quote from: Rich B on
This is what is so confusing about these subjects.

Expert 1 chimes in and says that it's not the distortion, it's the excess power that does the damage. Expert 2 chimes in and says that the added distortion can result in speaker damage.

Expert 3 chimes in and says that the added distortion together with excess power does the damage, but depending on the conditions, may be a mix of the two.

Expert 4 (Jneutron) who is Arnolds biggest nightmare opponent has a completely different argument and I can't even begin to understand it, but it's a lot more technical than any of the explanations provided thus far.


So where did JNeutron start posting here?

Oh, the name of the game is to see how many spurious sources can be cited?  (notice the above post has no links, no cites, just the usual careless assertions.)

Anybody who says that excess power isn't what fries speakers needs to review the laws of physics.

The relevant question is: "Where does that excess power come from?'

Anybody who says that the problem isn't the nut holding the volume control needs a reality check.

The relevant question then becomes: "Why does the system operator abuse the system?"  This is highly germane because it is a general rule that speakers sound bad before they are stressed enough to fail.

The usual answers have something to do with visiting firemen or substance abuse.



Audible symptoms of amplifier clipping

Reply #98 –
Quote from: Rich B on
Yes, too much power kills speakers. But there are cases where too much power was not responsible for killing speakers.

Yes it was, at ultrasonic frequencies.

You see, you again take a simple answer ("too much power kills speakers") and then add your own wrong assumptions ("the power needed to destroy a tweeter is equally high at every frequency") and then you start insulting people:
Quote from: Rich B on
Ask Jneutron, who is probably more knowledgeable than most of the folk here combined.


And I assumed it was a given that we're not discussing exceptions here, e.g. design or manufacturing defects that would cause any (mechanical) system to fail btw..


Anyway, due to your inability and unwillingness to read and learn and be respectful you've earned a place on the ignore list (next to amirm) and I hope others will put you there as well, at least until you've changed.
"I hear it when I see it."

Audible symptoms of amplifier clipping

Reply #99 –
Quote from: xnor on
Quote from: Rich B on
Yes, too much power kills speakers. But there are cases where too much power was not responsible for killing speakers.

Yes it was, at ultrasonic frequencies.

You see, you again take a simple answer ("too much power kills speakers") and then add your own wrong assumptions ("the power needed to destroy a tweeter is identical at every frequency") and then you start insulting people:
Quote from: Rich B on
Ask Jneutron, who is probably more knowledgeable than most of the folk here combined.


And I assumed it was a given that we're not the exception here where a design or manufacturing defect would cause any (mechanical) system to fail..


Anyway, due to your inability to read and learn and be respectful you've earned a place on the ignore list and I hope others will put you there as well.


That's fine. It's people like you who piss me off the most. You're a fucking arrogant asshole who thinks his answer is the only valid answer.

*To add, I have been more than respectful in the past. You are the one who thinks I'm too dumb to comprehend anything. You and others put me down on several occasions.