Huh?steveksux said:Well, I wouldn't exactly call it concise....
Clipped amp puts out more than the rated RMS power, so a 200w amp clipped can put out 400W of filthy power and burn out your unsuspecting 200W speaker.
Clipped signals have higher ratio of power in the high end. So tweeters are at more risk from clipped waveform than clean given the same power.
So given a 200W RMS speaker with an appropriately rated tweeter, it will likely survive short peaks of 400W from a 400W amp, where as the 400w clipped watts of a 200W amp will be more likely to fry the tweeter. 400w clean might split 350 w to woofer, 50 to tweeter, and 400w clipped might be more like 300w to woofer, 100w to tweeter. That's why tweeters are more vulnerable to distortion. That applies if you clip your input, ala "crunchy" sound on purpose. Guitar players with 200w speakers and 50W amps can distort all they want, the waveform isn't the issue, its avg power in the voice coil. Amp can't put out enough to burn out the speakers in that case.
The extra power generated when clipping a 200w amp can blow your 200w speaker given enough time if you don't hear the distortion and turn down. Same for a 400w amp driving your 200w speaker though. Distortion is generally a warning sign, ignore at your own risk.
Randy
Enjoy the benefits of a premium membership!
Upgrade Account
-
TalkBass has been independent since 1998. Add your voice.
Create a free account to reply to discussions, view embedded media, and browse with fewer display ads.
Join free Log in
Want zero display ads or expanded classifieds tools? Compare plans.
RMS
- Thread starter BassRocker8713
- Start date
bassix said:
No one said clipping didn't matter. One reason it matters it that it sounds bad. Another reason is that it means your amp can no longer handle dynamics or reproduce tone accurately.
What we said is that clipping isn't a problem quite the way you think it is. That is, in and of itself, it does not necessarily damage speakers. This should be an obvious point, seeing that guitar players do it all the time. Your contention was that the mere presentation of a square wave to a speaker would damage it. This would mean that you could damage an 800 W woofer by severly clipping a 1 W amp into it. Not likely. What Mark and steveksux explained was that this was not true. If clipping results in a situation in which too much power is presented to a driver, then the driver will have a problem. But if the driver can handle the amount of power being thrown at it, it basically doesn't care what the waveform looks like or whether you're clipping or not. Overexcursion is another issue, but is not directly related to clipping.
Companies make bigger amps (1) because they can, (2) because headroom makes things sound better, and (3) because that way they can be louder.
bassix said:
Round and round we go.
Clipping is bad, overpowering a woofer beyond it's physical limits is bad. Also woofers are one thing, tweeters are another. Clipping generates lots of high frequency energy. Woofers are not only more robust but because they have inherently poor high frequency response (compared to a tweeter) the clipping artifacts are less audible. With a passive crossover and a tweeter the crossover will send all that extra high frequency energy to the tweeter and distroy it in short order. My $.02 worth.Richard Lindsey said:
to your point about headroom,,,,YES, my point exactly. Things sound better because the amplifier isn't working so closely to it's limits. It has the reserve power to produce transients.
I did a Google search to see what I could find out about this whole subject of amplifier clipping. Some interesting stuff.
http://www.rocketroberts.com/techart/amp.htm
[Invalid or Expired Link Removed]
Invalid Link Removed
[Invalid or Expired Link Removed]
http://www.crownaudio.com/amp_htm/amp_info/how_much_power.htm
This is all good stuff. I learned a lot. For example the effect of the squaring off of a sine wave - apparently NOT exactly the same thing as a square wave - damages speakers at high levels, but not always in the way I thought. I love this stuff! It's always worthwhile to meet a challenge and try to learn from it. Thanks!
The JBL article in particular addresses this subject very well for what we've been talking about here.
http://www.rocketroberts.com/techart/amp.htm
[Invalid or Expired Link Removed]
Invalid Link Removed
[Invalid or Expired Link Removed]
http://www.crownaudio.com/amp_htm/amp_info/how_much_power.htm
This is all good stuff. I learned a lot. For example the effect of the squaring off of a sine wave - apparently NOT exactly the same thing as a square wave - damages speakers at high levels, but not always in the way I thought. I love this stuff! It's always worthwhile to meet a challenge and try to learn from it. Thanks!
The JBL article in particular addresses this subject very well for what we've been talking about here.
bassix said:
Owyes it is, a square wave is a totally chopped-off sine
And this is the interesting part from the jbl site:
"When an amplifier receives an input signal capable of driving it beyond its power rating, the result is clipping. This means that the negative and positive peaks of the amplifier's output signal are "clipped" off. The amplifier may also clip in an asymmetrical fashion, meaning that the positive side of the signal is clipped more than the negative (or vice versa). When subjected to an asymmetrical clipped waveform, one end of the loudspeaker's voice coil is "on average" spending more time outside of the gap (corresponding to the direction that is clipped) than the other. The end of the coil that is spending more time outside of the gap has poor heat transfer to the magnet structure. As a result, it overheats and burns. "
bassix said:
Before you patronize me, you should probably be sure that I don't know what I'm talking about....
I never said that clipping doesn't matter or that it was good. I'm simply saying that clipping in and of itself isn't necessarily a root cause of speaker damage and that damage from so-called underpowering doesn't occur in the way that people think it does. The jist of it is that an amp will put out more than its rated power under clipping, theoretically as much as twice rated power. How's that? An amp clips when the voltage of the output signal tries to exceed the power supply rail voltage. Obviously the amp can't output higher voltage than the supply rail, so the top of the waveform is clipped off at the rail voltage. If you drive the amp hard enough (say with a pure sine wave), eventually you'll get something that looks somewhat like a square wave out. Now to take the extreme examples of a pure sine and a pure square wave, it's easy to see that, at the same amplitude, there's more area under the square waveform than under the sinusoidal one. That means that the square wave has more energy or power than the sine wave. Twice as much, in fact (the math bears this out). Now, amps don't really ever produce pure square or sine waves, of course, but it's useful to use them as an example. So in this way, it's possible for an amp to put out almost twice its rated power under severe clipping. You can see, then why you could damage a 500W speaker with a 400W amp, but it's extremely unlikely you could damage the same speaker with a 100W one.
Asymmetrical clipping is pretty uncommon in modern power amps. In fact it only really occurs in class-A amplifiers. An interesting note is that asymetrical clipping is often used intentionally in distortion pedals because it favours even-order harmonics, which sound more musical than odd-order ones do.
broerevb said:Owyes it is, a square wave is a totally chopped-off sine
And this is the interesting part from the jbl site:
"When an amplifier receives an input signal capable of driving it beyond its power rating, the result is clipping. This means that the negative and positive peaks of the amplifier's output signal are "clipped" off. The amplifier may also clip in an asymmetrical fashion, meaning that the positive side of the signal is clipped more than the negative (or vice versa). When subjected to an asymmetrical clipped waveform, one end of the loudspeaker's voice coil is "on average" spending more time outside of the gap (corresponding to the direction that is clipped) than the other. The end of the coil that is spending more time outside of the gap has poor heat transfer to the magnet structure. As a result, it overheats and burns. "
A square wave is not really a totally chopped off sine wave. It's actually the combination of a fundamental sinusoidal frequency and an infinite series of sinusoidal harmonic frequencies. Every waveform can be broken down into a series of sine waves (thank you Fourier!!).
A clipped sine wave doesn't go from 0 to peak amplitude instantaneously, a square wave does. Real amplifiers can't react fast enough to truly produce a square wave.
As for the asymmetrical thing above, this would only be an issue if the clipped signal had enough amplitude to drive the coil out of the gap.
Mark Reccord said:
Well, that's the mathematical way to look at it. The fourier analysis is an approach of reality in sine functions, just to simplify the calculations. But that's theory, way beyond normal bass-life I'd say.
Here we are discussing amplifier clipping, the real thing, and those squares are incomplete sine waves, chopped off due to the lacking headroom of the amplifier's electronics... Agreed with the other two points however.
Several observations:
We can directly prove that a square wave doesn't necessarily damage a speaker....buy a 555 timer and with a couple of capacitors and resistors, it's easy to make a square wave generator. Hook this up to a speaker, the speaker will be undamaged.
Granted, this is a low power, 50% duty cycle square wave (it's 50% because it goes from 0 to +Vdc then back to zero). With a little more work we could make this a true 100% duty cycle square wave generator, going from +Vdc to -Vdc. Even that won't damage the speaker. So let's take this one step further: let's lower the frequency of our square wave generator. Down from 1000 hz, to 40 hz, down lower...At 10 hertz the speaker will be moving in then out 10 times per second....Let's go even lower--to 0 hz. Now we are giving the speaker pure DC. This is the absolute worst case of a square wave: a square wave with an infinite period or 0 Hz frequency.
Okay, real life. Take an 8 ohm speaker. Stick a 9 volt battery directly to it. This is a now a speaker with a 0 Hz square wave generator connected to it. The speaker will go all the way in or out and stay there. Since this is pure DC, the speaker will dissipate 9x9/8 watts, or 10.1 watts of power. This will all go to heat in the voice coil. If the voice coil is big enough to dissipate the heat, the speaker will sit there undamaged until the battery drains itself. (If the voice coil cannot dissipate the 10 watts of heat, it will physically start to distort and may eventually overheat and burn out). Conclusion: if the speaker can dissipate the power, a square wave doesn't matter one hill of beans.
But the REAL issue (to me at least) is the blanket statement of "always buy an amp twice the speaker rating". Not every situation is the same!! Some people carefully monitor their rig and won't ever clip their amp. But....Suppose Joe Noobie buys a speaker cab rated 200 watts RMS. He then buys an amp rated 400 watts RMS (following the "amp twice as big" rule). But unfortunately he's playing in a metal band with 2 guitarists, each driving Marshall stacks. So he cranks his amp up loud...then louder, trying to compete in the volume wars. He cannot recognize the sound of his amp clipping, and as stated above, it potentially can deliver twice it's rated power at full clipping. Now he's delivering 800 watts into a cab rated 200 watts!! Even given the transient nature of music, this has the potential of frying the cab.
The "best" approach would be to realistically evaluate how much power one will need. Then buy a rig that can deliver that power with adequate headroom--and that includes the cab...it's part of a system!!
We can directly prove that a square wave doesn't necessarily damage a speaker....buy a 555 timer and with a couple of capacitors and resistors, it's easy to make a square wave generator. Hook this up to a speaker, the speaker will be undamaged.
Granted, this is a low power, 50% duty cycle square wave (it's 50% because it goes from 0 to +Vdc then back to zero). With a little more work we could make this a true 100% duty cycle square wave generator, going from +Vdc to -Vdc. Even that won't damage the speaker. So let's take this one step further: let's lower the frequency of our square wave generator. Down from 1000 hz, to 40 hz, down lower...At 10 hertz the speaker will be moving in then out 10 times per second....Let's go even lower--to 0 hz. Now we are giving the speaker pure DC. This is the absolute worst case of a square wave: a square wave with an infinite period or 0 Hz frequency.
Okay, real life. Take an 8 ohm speaker. Stick a 9 volt battery directly to it. This is a now a speaker with a 0 Hz square wave generator connected to it. The speaker will go all the way in or out and stay there. Since this is pure DC, the speaker will dissipate 9x9/8 watts, or 10.1 watts of power. This will all go to heat in the voice coil. If the voice coil is big enough to dissipate the heat, the speaker will sit there undamaged until the battery drains itself. (If the voice coil cannot dissipate the 10 watts of heat, it will physically start to distort and may eventually overheat and burn out). Conclusion: if the speaker can dissipate the power, a square wave doesn't matter one hill of beans.
But the REAL issue (to me at least) is the blanket statement of "always buy an amp twice the speaker rating". Not every situation is the same!! Some people carefully monitor their rig and won't ever clip their amp. But....Suppose Joe Noobie buys a speaker cab rated 200 watts RMS. He then buys an amp rated 400 watts RMS (following the "amp twice as big" rule). But unfortunately he's playing in a metal band with 2 guitarists, each driving Marshall stacks. So he cranks his amp up loud...then louder, trying to compete in the volume wars. He cannot recognize the sound of his amp clipping, and as stated above, it potentially can deliver twice it's rated power at full clipping. Now he's delivering 800 watts into a cab rated 200 watts!! Even given the transient nature of music, this has the potential of frying the cab.
The "best" approach would be to realistically evaluate how much power one will need. Then buy a rig that can deliver that power with adequate headroom--and that includes the cab...it's part of a system!!
here was my lesson learned
all i know is that I almost killed an Ampeg BXT-410HL4 600watts RMS/1200 watts peak with an ampeg 350h (350 watts-solid state) at a road gig.
We had no PA support (just vocals, a little bit of kick/snare) but no subs, and I was competing with two loud guitarists. one had a tube 2x12 combo and a 2x12 extention, the other had a 2x12 combo and a 4x12 extention. we've all been there.
i basically had my old little 350h amp maxed out power amp and about 75% pre-amp just to keep up. it was fine for a while, then i heard this weird unintentional geddy lee/jack bruce tone start coming out of the speakers and then all the sudden a weird humpty hump 808 bass drum emerged everytime i hit the E string.
I dont' know what i did but, i turned down and made it though the gig. that cab and head never sounded the same after that show. i immediately sold it and got some new stuff, and i am much more careful.
all i know is that I almost killed an Ampeg BXT-410HL4 600watts RMS/1200 watts peak with an ampeg 350h (350 watts-solid state) at a road gig.
We had no PA support (just vocals, a little bit of kick/snare) but no subs, and I was competing with two loud guitarists. one had a tube 2x12 combo and a 2x12 extention, the other had a 2x12 combo and a 4x12 extention. we've all been there.
i basically had my old little 350h amp maxed out power amp and about 75% pre-amp just to keep up. it was fine for a while, then i heard this weird unintentional geddy lee/jack bruce tone start coming out of the speakers and then all the sudden a weird humpty hump 808 bass drum emerged everytime i hit the E string.
I dont' know what i did but, i turned down and made it though the gig. that cab and head never sounded the same after that show. i immediately sold it and got some new stuff, and i am much more careful.
