Hi people...i posted this in a thread that seems to have gotten deleted so i decided to repost in a new thread after speaking to a mod...i hope the information helps somebody...
minimum power will not damage your speakers...this is a myth that was created by a tech sheet JBL published many years ago by some engineer(marketing guy posing maybe ) who it seems had no idea what he was talking about and did tests not using the scientific method....if underpowering damaged speakers then u could never turn down the stereo because the speakers would be destroyed.....just think about that for a second before hitting the reply key...
for the truth about destroying speakers please read the RANE paper on the matter.....the engineers there actually did tests to prove this is nonsense...i did my own tests....here is what is usually touted....when an amp distorts is burns up speakers because of too little power(one).. when an amp distorts is produces DC(two)...both complete nonsense....
i will deal with (one) first....
i used the dynaudio esotars for a lone time...they are STILL being used by FIL in his mains in the home audio system i setup for him....the tweeters cost $650....yes...six-hundred-and-fifty dollars when i bought em...they can only be bought in pairs so if u burn one then u have to buy two....anyhow...if u check out a popular car audio forum u will find a test i did about 8 years ago...i put my $650 tweeters on an amp which was completely saturated...had a square wave on the oscope(will get to this in (two) ) and the amp was rated at 70w per channel, which meant "underpowering" and massive distortion....i had the tweeters on the amp like that for 5 minutes...i put my money where my mouth is....i ran an impedance sweep before and after the test(EXCELLENT way of testing VC damage) and the two sweeps were identical....i used my TDS(time delay spectrometry) analyzer for this test...as stated above the tweeters are fine...so myth debunked...
here is what happens.....u have an amp rated at some X number and a speaker and tweeter system rated at some higher number Y(hence "underpowering")....as the amp is turned up it starts to distort...well our ears/brains compress sound(we become "normalized" to the higher level of sound) as the SPL is increased(easy to verify..leave the car stereo on after a long drive and after a few hours come back and sit in the car and just switch the key on and be blasted out of the vehicle) so we dont notice it but the tweeters are getting louder and the mids are not getting louder...this is why when systems distort u hear pronounced high frequency response(listen to most boomers on the street for an unhealthy does of this)...the amp is maxed out at lower freqs but at higher freqs it is not maxed out and keeps amplifying the sound to the tweeters as the volume knob is turned up....THIS is what destroys the tweeters...too much power....if underpower destroyed anything then u could never turn the stereo down...many people do not realize that amplfiiers can sometimes produce up to ~twice rated power(not necessarily what is printed on the top but tested power) when under massive amounts of distortion.....if your speakers are not rated for this amount of power they will eventually fail...heat is your enemy...and guess what folks, humans can handle TONS of distortion before we notice it *especially* as the system is turned up and our ears are becominbg normalized to the hi level of sound....as a matter of fact i would wager that the *average* system in this and every other forum experiences 10% distortion frequently and very few people notice it....
(two)...people claim amps produce DC when distorting...this is foolishness...put a voltmeter on the amp while the amp has the system pounding and see for yourself...if u measure anything more than about 0.1V DC then u have issues with that amp and it should be checked by a tech....the "DC" u see and the "square waves" are really VERY saturated(technical term for a transistor trying to produce more voltage than it can) sine waves....they are basically sine waves with the tops chopped off....turn down the system and u will see the curves getting more and more round because more of the curve can now be faithfully reproduced by the transistor....same thing with vacuum tubes BTW....and as a matter of fact speaker MAGNETS saturate as well....i wont go into this one now tho...different thread
i hope this sheds a little light on what is actually happening when u destroy speakers....
the low power myth is on the same order as science not proving how bees can fly (it's been claimed for years that they break the laws of physics to do so) and that a duck’s quack doesn’t echo…before believing the hype do your own experiments and let us know what conclusion you come to.....
(two)...people claim amps produce DC when distorting...this is foolishness...put a voltmeter on the amp while the amp has the system pounding and see for yourself...if u measure anything more than about 0.1V DC then u have issues with that amp and it should be checked by a tech...
Kevin
Thanks for clearing that up for me. I went into a BUSY local store to buy some wire and overheard the install tech telling a customer "Of course you blew the sub, you were sending it straight DC". I was pretty sure this was bunk. Household DC is prety much a 60Hz sine wave, right?
I used the Autosound 2000 Test CD to get a bit of training as to what different distortion levels sound like, and you're right, I had my gains set to a point which yielded a min. of 10% distortion. They're much lower now.
So heat is the killer. Either running a sub at max. rated power for an extended period of time or 2-3 times above max. rated for a short period. I've read that the average speaker is something like 2-3% efficient so 100 watts of input yields 97 watts of heat. How does your experience rate that?
household power is not DC (direct current) at all, it is AC (alternating current)...yup..the SAME 60Hz test signal from your amps is what powers your home ...plug a speaker into the wall and u will get about...P = (V*V)/R watts going through that driver...so if the driver is 4 ohms u will have ~ (120*120)/4 = 3600W going through that speaker at 60Hz....
glad you *tested* the distortion levels yourself...most people would balk at that until they hear it for themselves....
speakers are motors in stall mode basically..this is when they are most inneficient...think about it...the cone has to start-stop-reverse direction start-stop etc...well u need the most current to start the speaker moving from rest so by definition this is when it is most inneficient...so the 3% number for average efficiency is a reasonable one....all the excess power is turned into heat...and u guys/gals thought internal combustion engines were inneficient ....an aside, this is one reason why buildings have motor control centers and "soft starters"...the "inrush current" can sometimes be up to SIX HUNDRED TIMES the average current rating needed by the motor to operate after it is up to speed....this would blow most breakers etc. so a soft starter "ramps" up the current to the motor...this helps a BUNCH with the electricity bill for buildings and also makes the design team's work less intense for switchgear and backup systems.....
Kevin
Quote:
Originally Posted by JnEsPappa
Thanks for clearing that up for me. I went into a BUSY local store to buy some wire and overheard the install tech telling a customer "Of course you blew the sub, you were sending it straight DC". I was pretty sure this was bunk. Household DC is prety much a 60Hz sine wave, right?
I used the Autosound 2000 Test CD to get a bit of training as to what different distortion levels sound like, and you're right, I had my gains set to a point which yielded a min. of 10% distortion. They're much lower now.
So heat is the killer. Either running a sub at max. rated power for an extended period of time or 2-3 times above max. rated for a short period. I've read that the average speaker is something like 2-3% efficient so 100 watts of input yields 97 watts of heat. How does your experience rate that?
Not sure where you are going with this exactly. A square wave does have more energy than a sinusoidal(that's the area under the curve). You will more easily blow your speakers using a square wave, it contains more energy. I am not arguing that underpowering speakers is the issue, but saturate an amp. and the speakers will generate more heat. Use a more powerful amp. and do the same experiment, eventually you will blow those tweets. The point is, how much heat can they dissipate before they melt. Your point is correct though, of course underpowering speakers won't 'blow' them. But it's generally agreed that clipping is a bad thing, and not just because it sounds bad.
/Mike
__________________
2001 2WD Silver Tundra: 4x4 TRD springs, Daystar 1", Carson AAL, red/blue Bilsteins, 265/75 Revo's, IS kit(minus bumper) and rims(hand polished), RF851X amp, Alpine HU, JL 6.5"XR comps, Qlogic with JL 8W3V2 subs. 185k miles and counting......
2005 Silver Sienna LE with 12" Overhead DVD
speakers have what is called the BL factor...this is baically (oversimplifying) the ratio between the mass of the moving parts of the motor and power of the motor....why is this important? well nothing is perfect and no speaker will generate a perfect square wave....you asked where i was going with this...there was a thread here last week where it was claimed that underpowering burnt speakers....the thread was deleted because of a power outage i am told and i spoke to Mark about it and he told me to put it back up...so i did...u answered your own question...the point is that underpowering wont kill speakers, that is where i am going....clipping is a non issue to speakers...they don't care what waveform they get(the acceleration is limited by the BL factor, system damping etc.) ....if anyone can find one credible paper with a repeatable test that shows this to be wrong then please share...as u say, the square wave is more dangerous and u hinted why...to those not familiar with calculus it's because the average power is greater when reproducing a square wave as opposed to a sine wave....
Kevin
Quote:
Originally Posted by longwoodklon
Not sure where you are going with this exactly. A square wave does have more energy than a sinusoidal(that's the area under the curve). You will more easily blow your speakers using a square wave, it contains more energy. I am not arguing that underpowering speakers is the issue, but saturate an amp. and the speakers will generate more heat. Use a more powerful amp. and do the same experiment, eventually you will blow those tweets. The point is, how much heat can they dissipate before they melt. Your point is correct though, of course underpowering speakers won't 'blow' them. But it's generally agreed that clipping is a bad thing, and not just because it sounds bad.
/Mike
household power is not DC (direct current) at all, it is AC
Kevin
LOL. Thanks, Kevin. I ALWAYS get that backwards.
Longwood. This debunks part of the often touted myth that UNDERPOWERING speakers will blow them. Think about this scenario. Amp gains set to reasonable distortion level and putting out say 30 watts RMS to a low sensitivity speaker. The user turns the h/u volume up and up to the point where a square wave is being sent to the amp for....amplification. Playing that square wave won't kill the speaker.
Now, adjusting the amp gain into full square wave reproduction (increasing power output like you said, measured by area under the curve) CAN blow the speaker, but that would be a result of that amount of power overheating the voice coil, NOT because of the speaker trying to reproduce the square wave.
Longwood. This debunks part of the often touted myth that UNDERPOWERING speakers will blow them. Think about this scenario. Amp gains set to reasonable distortion level and putting out say 30 watts RMS to a low sensitivity speaker. The user turns the h/u volume up and up to the point where a square wave is being sent to the amp for....amplification. Playing that square wave won't kill the speaker.
Now, adjusting the amp gain into full square wave reproduction (increasing power output like you said, measured by area under the curve) CAN blow the speaker, but that would be a result of that amount of power overheating the voice coil, NOT because of the speaker trying to reproduce the square wave.
I am not sure I agree totally here. In your scenario, the more you turn up the H/U vol., the more ths signal will clip(ie. the more it will look like a square wave). And Kevin said this:
........clipping is a non issue to speakers...they don't care what waveform they get(the acceleration is limited by the BL factor, system damping etc.) ....if anyone can find one credible paper with a repeatable test that shows this to be wrong then please share...as u say, the square wave is more dangerous and u hinted why...to those not familiar with calculus it's because the average power is greater when reproducing a square wave as opposed to a sine wave....
Kevin
--- So a square wave DOES have more average power. That's why I don't think it's entirely accurate to say clipping is a non-issue. The more you clip, the more like a square wave the signal becomes, which means more power being sent to the speaker, which means the hotter the speaker becomes.
I agree that a speaker doesn't necessarily care what it plays. But try this:
Set up your experiment however you'd like. Now play a regular song for 20min. Measure the temp. of the VC/backplate/whatever. Now play a sine wave for 20min., I'm sure that speaker will be hotter than when you played simply music. Now play a square wave for 20min., and the speaker will be hotter still.
Sure, in most cases people aren't running their equipment so close to the edge that clipping will cause failure. But to say clipping is a non issue is totally false, imo. /Mike
__________________
2001 2WD Silver Tundra: 4x4 TRD springs, Daystar 1", Carson AAL, red/blue Bilsteins, 265/75 Revo's, IS kit(minus bumper) and rims(hand polished), RF851X amp, Alpine HU, JL 6.5"XR comps, Qlogic with JL 8W3V2 subs. 185k miles and counting......
2005 Silver Sienna LE with 12" Overhead DVD
Mike:
u have shared great info....Mark is, however, correct....and so are you mostly...u dropped the ball in your very last sentence IMO...u are correct that the square wave will cause more damage in your experiment....the question is...WHY?
we agree that the square wave has more average power....lets change your experiment somewhat...lets say we have a sine wave and then we use a square wave with the level adjusted so that the *average power is the same for both waveforms*...your experiment does not take this into account which makes it meaningless....which damages the speaker quicker? this is what we are getting at...the type of wave in and of itself means nothing....if we want to get technical EVERY waveform can be broken down into a compilation of sine waves...with your experiment it is evident that u will have power differences depending on the waveform...for this to be scientifically meaningful we must keep the average power the same across the board....
i have an experiment for u along your lines of testing..this is more what Mark was getting at...get a 50w amp and turn it up till its begging for mercy...put a nice big fan on it...then get any decent 12" sub made for car audio and tell me how long it takes for the woofer to be destroyed....now bridge a kicker zr1000 amplifier to that same woofer and turn the gain up till u are a little below clipping...so no hint of square wave...just pure sine...and tell me how long the woofer takes to self destruct....
do u see why this is meaningless? we can go both ways IF power is not held constant, and we still won't have an answer the question....holding the power constant will tell us all we need to know....
Kevin
Quote:
Originally Posted by longwoodklon
I am not sure I agree totally here. In your scenario, the more you turn up the H/U vol., the more ths signal will clip(ie. the more it will look like a square wave). And Kevin said this:
........clipping is a non issue to speakers...they don't care what waveform they get(the acceleration is limited by the BL factor, system damping etc.) ....if anyone can find one credible paper with a repeatable test that shows this to be wrong then please share...as u say, the square wave is more dangerous and u hinted why...to those not familiar with calculus it's because the average power is greater when reproducing a square wave as opposed to a sine wave....
Kevin
--- So a square wave DOES have more average power. That's why I don't think it's entirely accurate to say clipping is a non-issue. The more you clip, the more like a square wave the signal becomes, which means more power being sent to the speaker, which means the hotter the speaker becomes.
I agree that a speaker doesn't necessarily care what it plays. But try this:
Set up your experiment however you'd like. Now play a regular song for 20min. Measure the temp. of the VC/backplate/whatever. Now play a sine wave for 20min., I'm sure that speaker will be hotter than when you played simply music. Now play a square wave for 20min., and the speaker will be hotter still.
Sure, in most cases people aren't running their equipment so close to the edge that clipping will cause failure. But to say clipping is a non issue is totally false, imo. /Mike
Elementary physics will answer this pretty handily.
Remember, you need to build up potential energy to get kinetic energy release, right? Good, you remembered your elementary physics. Its just like a rollercoaster going up a ramp then being released at the very top for maximum effect and duration.
So how does the above play into the heat/dissapation equation? Remember how much surface area your tweeters/speakers have at the rear to dissapate the heat generated by motion of the magnets/coil action? If your surface area to dissapate heat increases but ventilation area does not increase, you get to the point where you have excessive heat built up without anywhere to go. This is the point where the speakers will fail, from excessive kinetic energy being released from potential energy in too short a duration to allow the speaker to cool sufficiently to generate the next sound. If your speakers rear surface area around the electrical parts are large enough and have sufficient clearance/area, they will be able to disperse the heat generated by the equipment required to power the tweeters/subwoofers/speakers. Theres a way to prove this -- blow hot air into some clasped hands and note how long that hot air hangs around while blowing more hot air in. Now try loosening those fingers up so some of that hot air gets out through them and keep blowing, note the temperature doesn't increase but rather stays close to constant until of course, your heated air is greater than what can be ventilated through those gaps left by your fingers.
Also note that even the air gets warm inside the back because of convection caused by the mechanics of the electrical equipment so thus this is why some speaker cases come equipped with a fan in the side or ventilation holes to disperse this heated air before it can continue to weaken/damage the equipment.
What you guys are all missing is that distorted or clipped output will look like a high frequency noise to the crossover network which will send that clipped or distorted output to the tweeter instead of bass/mid drivers. The tweeters in most systems can only handle about 1/10th the power of the mids/bass. So the 30 distorted watts that would never hurt a bass driver will smoke the tweeter in an instant of distortion.
If you have a 50 w/ch good clean amp, and never get to the point of clipping - those 5 watt tweeters will last forever. If you have a 25w/ch amp that is driven to distortion that 25 watts of distortion will go straight to the tweets.
There are some tweeters that can handle a lot of power, typically edge wound dome tweeters. They are more tolerant of distortion, but I have cooked my share of tweeters with low power amps, and the solution was always to get a better amp. Since I have been running the MMats amps (which produce very clean power even at high levels), i have not had a tweeter get cooked - despite the fact that I listen to the same music louder now than ever before.
An example of this was obvious when NIN SIN came out, people were smoking tweeters left and right (no pun intended) because the music on the CD contained so much distortion - all the power went to tweeters. I was working at a local audio store at the time, people were bringing tweeters in by the handfuls to be replaced, all with the same story.
In fact with higher power amplifiers you will usually be able to smell the voice coil burning in the mid-bass driver before you actually damage the voice-coil. With low power distortion, the damage to the tweeters will be instant.
__________________
2000 4x4 LTD AC Sunfire Red - Sold w/ 200K
2005 4x4 LTD AC Phantom grey w/ dark grey - Sold w/ 20K
Alpine CDA-9835, W/ KCA-420i, Focal 165KP front, Focal 165V Rear speakers. powered by 2 MMats SQ2150 amps. Alpine MRD-M501 powering a JL Stealthbox with Infinity Perfect 10D VQ
Sound deadened everything with Cascade VB2HD, and Dynamat Extreme
2005 G35 Sedan 6MT - Returned due to paint work on a 'new' car.
2006 G35 Sedan 6MT - all stock. Where do I hook up the trailer and where's the bass?
no offense JeffD but the below is mostly nonsense...it has been proven, had u guys bothered to check, that harmonics(the high frequency noise your talking about) has so much less energy than the fundamental that it is a non issue for all practical purposes...again..the JBL paper i referred to above claimed the exact reasons u stated below(i.e. harmonics) as to why "low power" kills tweeters...again...nonsense...
the reason why that CD was destroying tweeters is because it was *recorded* with more high frequency energy than most other CD's, which made this high frequency distortion, as u put it, the fundamental frequencies and as such had a lot of power....
Kevin
Quote:
Originally Posted by JeffD
What you guys are all missing is that distorted or clipped output will look like a high frequency noise to the crossover network which will send that clipped or distorted output to the tweeter instead of bass/mid drivers. The tweeters in most systems can only handle about 1/10th the power of the mids/bass. So the 30 distorted watts that would never hurt a bass driver will smoke the tweeter in an instant of distortion.
If you have a 50 w/ch good clean amp, and never get to the point of clipping - those 5 watt tweeters will last forever. If you have a 25w/ch amp that is driven to distortion that 25 watts of distortion will go straight to the tweets.
There are some tweeters that can handle a lot of power, typically edge wound dome tweeters. They are more tolerant of distortion, but I have cooked my share of tweeters with low power amps, and the solution was always to get a better amp. Since I have been running the MMats amps (which produce very clean power even at high levels), i have not had a tweeter get cooked - despite the fact that I listen to the same music louder now than ever before.
An example of this was obvious when NIN SIN came out, people were smoking tweeters left and right (no pun intended) because the music on the CD contained so much distortion - all the power went to tweeters. I was working at a local audio store at the time, people were bringing tweeters in by the handfuls to be replaced, all with the same story.
In fact with higher power amplifiers you will usually be able to smell the voice coil burning in the mid-bass driver before you actually damage the voice-coil. With low power distortion, the damage to the tweeters will be instant.
Kevin,
Well, I think we are basically agreeing here. What got my attention in your previous post was when you said 'clipping is a non-issue to speakers'.
The Rane paper refers essentially to the tweeters within a loudspeaker system. Look at their conclusion: "If we can prevent an amp. from clipping, we could better utilize our loudspeakers. Limiters play an important role in preventing clipping and the resulting compression". So in effect they are saying, minimizing clipping is good. So to me, clipping is an issue. I know what your saying though, the clipped signal in and of itself is not the problem necessarily,
That paper is very good, but it doesn't address the issue of what happens to the speaker handling the low frequency sine wave burst. That is what I think people normally take for granted, when we talk about clipping we usually mean the signal to the speaker under test is clipping.
Back to my point in the previous message. It goes without saying that if you have a speaker capable of handling 1000W, a 70W amp. will never hurt that speaker. But use a speaker rated at 70W and that 70W amp. is pushing the speaker to it's limits. Now, severely clip that amp., and you have effectively produced a square wave output to that speaker. This square wave does contain considerable energy. More energy than if you were to reduce the amp. gains to the point that the same input is at the 'just clipping' point. In other words, you have NOT held the power constant. That's why I proposed that hypothetical. It's also shown in the Rane paper, look at that sine wave inf Fig. 3.
Anyways.......
Quote:
Originally Posted by Kevin7909
Mike:
u have shared great info....Mark is, however, correct....and so are you mostly...u dropped the ball in your very last sentence IMO...u are correct that the square wave will cause more damage in your experiment....the question is...WHY?
we agree that the square wave has more average power....lets change your experiment somewhat...lets say we have a sine wave and then we use a square wave with the level adjusted so that the *average power is the same for both waveforms*...your experiment does not take this into account which makes it meaningless....which damages the speaker quicker? this is what we are getting at...the type of wave in and of itself means nothing....if we want to get technical EVERY waveform can be broken down into a compilation of sine waves...with your experiment it is evident that u will have power differences depending on the waveform...for this to be scientifically meaningful we must keep the average power the same across the board....
i have an experiment for u along your lines of testing..this is more what Mark was getting at...get a 50w amp and turn it up till its begging for mercy...put a nice big fan on it...then get any decent 12" sub made for car audio and tell me how long it takes for the woofer to be destroyed....now bridge a kicker zr1000 amplifier to that same woofer and turn the gain up till u are a little below clipping...so no hint of square wave...just pure sine...and tell me how long the woofer takes to self destruct....
do u see why this is meaningless? we can go both ways IF power is not held constant, and we still won't have an answer the question....holding the power constant will tell us all we need to know....
Kevin
__________________
2001 2WD Silver Tundra: 4x4 TRD springs, Daystar 1", Carson AAL, red/blue Bilsteins, 265/75 Revo's, IS kit(minus bumper) and rims(hand polished), RF851X amp, Alpine HU, JL 6.5"XR comps, Qlogic with JL 8W3V2 subs. 185k miles and counting......
2005 Silver Sienna LE with 12" Overhead DVD
i still think that we are talking about different things on ths clipping issue (sure of it as a matter of fact) because if we are to test a factor we must keep all other factors constant...
i do, however, concur that we agree on most points, and i would say at this point all the important ones considering that most of the myths have been mentioned and dealt with IMO....i think i gave u credit for good info
Kevin
Quote:
Originally Posted by longwoodklon
Kevin,
Well, I think we are basically agreeing here. What got my attention in your previous post was when you said 'clipping is a non-issue to speakers'.
The Rane paper refers essentially to the tweeters within a loudspeaker system. Look at their conclusion: "If we can prevent an amp. from clipping, we could better utilize our loudspeakers. Limiters play an important role in preventing clipping and the resulting compression". So in effect they are saying, minimizing clipping is good. So to me, clipping is an issue. I know what your saying though, the clipped signal in and of itself is not the problem necessarily,
That paper is very good, but it doesn't address the issue of what happens to the speaker handling the low frequency sine wave burst. That is what I think people normally take for granted, when we talk about clipping we usually mean the signal to the speaker under test is clipping.
Back to my point in the previous message. It goes without saying that if you have a speaker capable of handling 1000W, a 70W amp. will never hurt that speaker. But use a speaker rated at 70W and that 70W amp. is pushing the speaker to it's limits. Now, severely clip that amp., and you have effectively produced a square wave output to that speaker. This square wave does contain considerable energy. More energy than if you were to reduce the amp. gains to the point that the same input is at the 'just clipping' point. In other words, you have NOT held the power constant. That's why I proposed that hypothetical. It's also shown in the Rane paper, look at that sine wave inf Fig. 3.
Anyways.......
i can only off you this....many people have read this thread and at least one person (our mod) has learned something...that is enough of a reason for me to keep posting....if you do not want to learn anything about this subject then please do not read this thread.....no pissing contest here....
Kevin
Quote:
Originally Posted by nicoj122
you guys should really stick to the pm's. i cant believe i even read this pissing contest thread.
Amplifier Burning Speakers
) who it seems had no idea what he was talking about and did tests not using the scientific method....if underpowering damaged speakers then u could never turn down the stereo because the speakers would be destroyed.....just think about that for a second before hitting the reply key...
Linear Mode
