Clipping Damages Speakers - Yes or No
- Thread starter Guest105
- Start date
I do not see what is so difficult to understand since the explanation I offered in post #111 should have made my position clear.
To recap then:
Point 1: Mere presence of clipping does not necessarily result in driver damage.
Point 2: Excessive power (as in power far beyond a driver's heat dissipation capabilities) WILL result in driver damage.
Point 3: An amplifier driven to clipping greatly exacerbates the chances of driver failure (particularly tweeters) because:
a. The clipping amplifier generates a disproportionate amount of HF energy relative to LF output.
b. The disproportionate HF energy to HF drivers often goes unnoticed because they tend to be overpowered by the lower frequencies which are more apparent in human hearing.
c. The delicacy inherent in tweeter construction (in most designs) increasing their susceptibility to damage.
This is not something that can be answered with a simple yes/no and the original question as worded is subject to more than one interpretation.
I suspect we can all agree that clipping is bad, in that it degrades sound quality. The disagreement, of course, is as regards whether or not clipping can damage one's speakers (other than causing them to reproduce a distorted signal).
I had hoped not to have the various positions debated in this thread, but have given up on that pipe dream! Now I am simply trying to get clarification as to what it is that people holding various points of view believe as regards the specific mechanism causing the damage.
The damage theory that I do understand is the one based on clipping-induced harmonics, which holds that clipping causes a large increase in high-frequency energy. In turn, this would damage the tweeter (assuming that it's power handling capabilities are less than the clipping-induced spike). I am neither advocating nor opposing this theory...just explaining my understanding of it.
I also understand the DC theory to some degree; its advocates suggest that the AC signal is converted to DC. From here, there appear to be several schools of thought, including: 1. DC inherently damages speakers, and 2. DC causes the cone to stop moving, which overheats the coil. There well may be other variations. Again, I am NOT expressing a position...just describing it as I understand it.
The "distorted waveform" theory I understand least of all; in other words, I don't understand what it is about a distorted waveform that damages the speaker/driver. I get how it damages the sound quality...just not how it damages the speaker itself.
I also appreciate the position of the respondents who rely on their own experiences (my amp was clipping and my tweeters burned up), but must admit that I don't find a conclusion that clipping caused the damage to be logical...at least with the information provided.
In any case, I look forward to continued voting. On a related note....it is interesting to see the vast difference between the quantity of votes and the quantity of thread views.
I had hoped not to have the various positions debated in this thread, but have given up on that pipe dream! Now I am simply trying to get clarification as to what it is that people holding various points of view believe as regards the specific mechanism causing the damage.
The damage theory that I do understand is the one based on clipping-induced harmonics, which holds that clipping causes a large increase in high-frequency energy. In turn, this would damage the tweeter (assuming that it's power handling capabilities are less than the clipping-induced spike). I am neither advocating nor opposing this theory...just explaining my understanding of it.
I also understand the DC theory to some degree; its advocates suggest that the AC signal is converted to DC. From here, there appear to be several schools of thought, including: 1. DC inherently damages speakers, and 2. DC causes the cone to stop moving, which overheats the coil. There well may be other variations. Again, I am NOT expressing a position...just describing it as I understand it.
The "distorted waveform" theory I understand least of all; in other words, I don't understand what it is about a distorted waveform that damages the speaker/driver. I get how it damages the sound quality...just not how it damages the speaker itself.
I also appreciate the position of the respondents who rely on their own experiences (my amp was clipping and my tweeters burned up), but must admit that I don't find a conclusion that clipping caused the damage to be logical...at least with the information provided.
In any case, I look forward to continued voting. On a related note....it is interesting to see the vast difference between the quantity of votes and the quantity of thread views.
I wish you could read the passage in the McIntosh book where in the Laboratory they were blowing tweeters with a 100 watt amplifier but not with a 200 watt amplifier at full output.
The lower powered amplifier could never put out 200 watts yet that one blew the tweeters while the 200 watt amplifier did not - with 200 watts of output!
The lower powered amplifier could never put out 200 watts yet that one blew the tweeters while the 200 watt amplifier did not - with 200 watts of output!
Thanks. I still believe that your "YES" vote correctly reflects your point of view. Your position - if I understand it correctly - is that clipping generates a surge in HF energy...much higher than would otherwise be the case. While you caveat the damage question as dependent upon the power-handling capability of the tweeters, I would still consider the damage - if it occurs - to be as the direct result of clipping. I see this as the case because it is clipping that generates the aberrational spike in high-frequency energy...a spike well above what the amp would be producing were it not in clipping.
And, once again, to be clear...I am neither advocating nor opposing the "high-frequency energy spike" theory.
charles 1973
Super Member
It seems that before we answer the op's question, we have to define clipping. If we are referring simply to an audio signal's clipping, That is one thing. Some CD's are recorded that way (Loudness wars). An amplifier driven into clipping is an entirely different story. Both can kill a speaker if it causes the amp in question to deliver more RMS power (as in heat) than the speaker will handle.
If the audio signal has excessive clipping, and an amp is kept out of clipping, it may still be delivering more RMS power than the speaker is rated for. Speakers are generally rated with music power. With most un-clipped music short term RMS power is about 10 - 15% of that. When an audio signal is clipped enough, It's RMS power may be 30% of the music power, easily damaging the speaker.
If we are talking about an amp driven into clipping with an un-clipped audio signal, we get the same results as above, but with the addition of increased high frequency harmonic distortion that goes directly to the tweeter, and perhaps more nasty surprises. This is really bad.
If the audio signal has excessive clipping, and an amp is kept out of clipping, it may still be delivering more RMS power than the speaker is rated for. Speakers are generally rated with music power. With most un-clipped music short term RMS power is about 10 - 15% of that. When an audio signal is clipped enough, It's RMS power may be 30% of the music power, easily damaging the speaker.
If we are talking about an amp driven into clipping with an un-clipped audio signal, we get the same results as above, but with the addition of increased high frequency harmonic distortion that goes directly to the tweeter, and perhaps more nasty surprises. This is really bad.
Last edited:
JBL GUY
Addicted Member
Based on this phrasing of the question, the answer is no.
Here is a video that does a good job of explaining the composition of a square wave and the amplitude of the harmonics that make up a square wave (or clipped audio signal).
Note that the 3rd harmonic (the first odd order harmonic that occurs after the fundamental frequency) is almost 10 dB weaker than the fundamental frequency and as the frequency increases the harmonics continue to decrease in amplitude.
At about 10 minutes into the video he shows the measured levels of the harmonics of a square wave. There is no large increase in the high frequency energy content of the signal.
This is basic Fourier analysis of a square wave.
The loudspeaker does not care about the shape of the waveform.
If the amplifier clips cleanly, per you phrasing of the question, there is no large increase in high frequency energy.
Anecdotal evidence is one thing, but the only way to know what is actually happening is to be looking at the signal with an oscilloscope and spectrum analyzer at the time a speaker fails.
If an amplifier misbehaves when driven into clipping then any result is due the amplifier, not the shape of the waveform.
Abuse is abuse Clipping doesn't cause failure per se, but is responsible. 2 observations. First, based on an excellent article on the rane web site, is that when an amplifier is driven to clipping, the high frequencies, though not clipped, are amplified further and ususally exceed the power handling of tweeters. Example normal music is maybe 10 % twweter energy. When clipping, that percentage goes up and tweeters go dead. The second point is that normal music the average to max ratio is anywhere between 3 to 10 times. When clipping, that goes down to 1, plus the max output is going to be greater than the "RMS" power. So a 50 watt amplifier driven to clipping may put out 70 watts, and if all the time will kill tweeters, mids and sometimes woofer, plus damage xover components like resistors and inductors. My 2 cents.
I don't want to add any more confusion, but what would the phrasing of the question need to be for an answer of "yes"?
Would it be as simple as elimination of subsequent amplifier failures as an exception? As I'm sure is clear, I purposely excluded speaker damage in such scenarios as being caused by clipping because, at least for this poll, I assume such amplifier misbehavior to be the result of a design flaw in the amp...something which "should not" occur (but, which, of course, in reality occurs far too often).
Basically, the intent of my phrasing - perhaps not achieved - was to eliminate (or at least minimize) gray areas and semantic nuances.
JBL GUY
Addicted Member
Exactly.
The question can be asked in to ways.
Does the fact that the waveform is clipped mean that the signal will damage a speaker?
The answer to this question is no.
If an amplifier is driven into clipping is it possible for a speaker that is connected the amplifier to be damaged.
The answer to this question is yes.
In this instance, it is possible that the amplifier is producing enough power to damage the speaker, regardless of the shape of the waveform, but not regardless of the position of the volume control.
To be clear, the high frequencies are not amplified more because the amplifier is clipping; the amplitude of the high frequencies (or any frequencies for that matter) is what it is because of the setting of the volume control.
To be clear in simple terms for all of the over 3000 hits on this thread, it is all about the power that is sent to the speaker, not the shape of the waveform, although there is more power contained in the envelope of a square wave than the envelope of a sine wave. The amount of power that is sent to a speaker is dependent on the position of the volume control.
Continue to increase the volume control and the amplifier may send the speaker more power than it can handle, be it an unclipped or a clipped signal and the amplifier may not behave nicely if driven into clipping.
And to add, the term clipping is not a good descriptor, unless the amount of clipping is specified.
Last edited:
chicks
Lunatic Member
Thank you for your reply.
Would it be fair to expand on you comment regarding the power in a square wave to say that it is equal to approximately 2x the RMS wattage rating of the amp (assuming that the amp has been rated correctly)? And if so, does this suggest that if one's speakers can handle 2x the RMS wattage rating of one's amp, no amount of clipping on the part of that amp - from minimal clipping all the way to full clipping - will damage those speakers? Again, still excluding any "downstream" mis-behaviors of the amp.
Also, please note that I'm seeking a 99.9% of the time answer....one which also excludes such anomalies as - for example - tweeters which are already at the thinnest edge of their existence, being pushed into failure by the (apparently) very small increases attributable to harmonics.
Last edited:
Superampman
AK Member
And what, pray tell, would that be?
spicer
Lunatic Member
That introduces power factor. It will be in phase unless some reactance is at work. Across load resisters... the only way one is going to be able to do this... it will multiply or divide out. That's how they arrive at the output specs. But talking simply how much absolute power an amp will put out... drive it into a square wave, measure and multiple, or divide, depending on what you're measuring and computing with.
spicer
Lunatic Member
Well.... 99.9% of the time, take any...let's say, 100 watt per channel amp and drive it into severe clipping for one minute and it will destroy the tweeters in 99.9% of the home audio speakers of everybody reading or posting on this thread 99.9% of the time. Minimal clipping... not so much. Minimal clipping can be described as getting the most out of it...or, that's about all there is...like just over the red line on an engine... exceed it by too much and while it might remain exciting unpleasant things are going to happen.
Last edited:
JBL GUY
Addicted Member
You are welcome. We have been involved in this discussion before and I enjoyed our exchanges.
Yes, it is all about the amplitude of the signal that is sent to the speaker and the actual power handling capacity of the speaker (individual drivers and crossover components), not necessarily the manufacturers published specifications.
The same can be said for the amplifier. The actual maximum output power may be considerably greater than the manufacturer's published specification.
In simple terms it is all about the position of the volume control and the actual capability of the speakers and the amplifier.
And spicer makes a good point. But if the same amount of unclipped effective power is sent to each speaker component there will be the same amount of damage. When an amplifier is driven into full clipping it changes the peak to average power ratio and it skews the spectral energy content of the audio signal.
But if the amplifier in spicer's example were to be a 10 watt amplifier, then the result might be different, depending on the actual power handling capacity of the components in the speaker system.
So kcbluesman if you are looking for a 99 percent answer, spicer's answer is a good answer. It is an example of causing an amplifier to produce more power than the components in a speaker system can handle.
Sorry, but I am striving for as much clarity as possible. spicer's answer suggests that it is the clipping that does the damage, by increasing the power in some way or otherwise exacerbating the likelihood of damage. You seem to agree with his answer, but then also seem to be saying that the clipping is not the relevant factor. A reader could take it either way.
I asked in my proposed expansion of your earlier comments whether you would agree that the power at full clipping is roughly equal to 2x the RMS wattage rating...assuming that the RMS rating was accurate (not understated). I understand that if the RMS rating is not accurate, neither will be the peak capability.
Spicer's most recent post may well be true, in that 200 watts of continuous input is more than most people's speakers are capable of handling. This in and of itself says nothing about the effect of clipping, other than clarifying that it is not rated power but power at full clipping that one's speakers must be able to withstand to avoid damage from being overpowered (and, of course, assuming that one's hearing is bad enough not to notice the horrible noises being produced by a fully clipped signal!).
But if I understood your earlier posts correctly, the damage would be caused precisely because the speakers are incapable of handling the power being produced by the amp (regardless of whether or not it is in clipping, and regardless of the degree of that clipping).
Does the distinction that you are making - and which I am having trouble understanding - have to do with the "compression" concept? Perhaps not a technically correct use of the term, but by it I mean that the closer the average amplitude is to the peak capability of the amp (defined by the rail voltages), the greater the degree of clipping. But if this is what you mean, isn't the clipping simply an effect of running the amp so close to its limits? In other words, clipping is not the cause of the increased power going to the speakers....the increased power is instead the cause of the clipping?
Or perhaps your comment about the skewing of the spectral energy content of the audio signal is the key issue? Can you expand on what you mean by this comment? In what way is it skewed? You said earlier - and I'm paraphrasing to illustrate my understanding - that clipping does not cause a disproportionate increase in high frequency energy/power.
So, what is the change to which you are referring? More specifically, what is it, and what is its implication?
I've tried to explain my understanding of your comments in great detail, in the hope that by doing so you can more easily spot the error(s) in my interpretation. Thanks again.
I asked in my proposed expansion of your earlier comments whether you would agree that the power at full clipping is roughly equal to 2x the RMS wattage rating...assuming that the RMS rating was accurate (not understated). I understand that if the RMS rating is not accurate, neither will be the peak capability.
Spicer's most recent post may well be true, in that 200 watts of continuous input is more than most people's speakers are capable of handling. This in and of itself says nothing about the effect of clipping, other than clarifying that it is not rated power but power at full clipping that one's speakers must be able to withstand to avoid damage from being overpowered (and, of course, assuming that one's hearing is bad enough not to notice the horrible noises being produced by a fully clipped signal!).
But if I understood your earlier posts correctly, the damage would be caused precisely because the speakers are incapable of handling the power being produced by the amp (regardless of whether or not it is in clipping, and regardless of the degree of that clipping).
Does the distinction that you are making - and which I am having trouble understanding - have to do with the "compression" concept? Perhaps not a technically correct use of the term, but by it I mean that the closer the average amplitude is to the peak capability of the amp (defined by the rail voltages), the greater the degree of clipping. But if this is what you mean, isn't the clipping simply an effect of running the amp so close to its limits? In other words, clipping is not the cause of the increased power going to the speakers....the increased power is instead the cause of the clipping?
Or perhaps your comment about the skewing of the spectral energy content of the audio signal is the key issue? Can you expand on what you mean by this comment? In what way is it skewed? You said earlier - and I'm paraphrasing to illustrate my understanding - that clipping does not cause a disproportionate increase in high frequency energy/power.
To be clear, the high frequencies are not amplified more because the amplifier is clipping; the amplitude of the high frequencies (or any frequencies for that matter) is what it is because of the setting of the volume control.
So, what is the change to which you are referring? More specifically, what is it, and what is its implication?
I've tried to explain my understanding of your comments in great detail, in the hope that by doing so you can more easily spot the error(s) in my interpretation. Thanks again.
Last edited:
spicer
Lunatic Member
" When an amplifier is driven into full clipping it changes the peak to average power ratio and it skews the spectral energy content of the audio signal."
That is exactly correct. I chose a 100 watt amplifier for my example simply because an amp with that capacity in it's power supply will deliver enough energy to overheat most any tweeter in most any home audio system commonly available if it is severely overdriven. A ten watt amp most likely will not damage anything in most any hi fi quality loudspeaker no matter how it's used. I guess I just assume it is understood there has to be enough power available to do damage. I could have said 70 watts but at some point there just won't be enough power available to cause damage no matter what. The sentence quoted above from JBL guy is absolutely correct and yes, the clipping is what skews the spectral energy of the audio signal... if the signal was being compressed through voltage limiting in the drive circuit, like a peak limiter, that would complicate the explanation since the audio spectrum would no longer be skewed, would only change the peak to average power ratio. It would remain though that a particular driver can only handle so many watts before it overheats, doesn't matter what the signal looks like... could be DC from a big battery at what ever voltage multiplies out across the voice coil resistance to too many watts. I mean, when one converts AC to power/watts it is figured as a DC equivalent..... RMS.. root mean square... I know, for all the engineers, not quite but practically. I think the only practical thing to know about it is not to overdrive the amp. One learns the practical limits with any particular system of amplifier and speakers by buying a few tweeters. That's how one learns where the limits are and if your paying attention you'll notice how far into the peak to average power ratio change you were when the tweeters went out. When it seems to get a lot louder quickly you're in the zone. Some speaker manufactures used to include tweeter fuses in the back. Others... Bose to name one.. has used a low voltage light bulb, commonly for them a 12 volt one amp bulb, with its filament in series with the tweeter to act as a ballast for protection against high current spikes. The bulb gets hot and it's resistance goes up, when it gets real hot and starts lighting up the interior of the cabinet the resistance goes way up... saves on warranty claims.
Last edited: