Help with Amplifier

[mtrycrafts]

I can't see how you can conclude the amp was distorting. 90 watts to a 93db/1 meter/ 1 watt speaker far exceeds normal power requirements for a speaker of that sensitivity. If anything, he simply exceeded the power disappation ability of the tweeter with the average continous signal[afterall, the speaker system might be rated for 100 watts, but a tweeter is typically only really rated for about 10-20 watts]. With certain music, the tweeter can be far more likey to blow due to the spectral distribution being not typical[and tweeters are matched with the system on an average expected spectrum distribution, proporotionately]. The guy was just probably using these speakers at too high of a volume level[compared to what the tweeter can actually handle, which is not reflected in the power rating of the speaker].

-Chris

I have been going over a number of clipping related articles:

http://sound.westhost.com/tweeters.htm

http://www.rane.com/note128.html

http://www.st-andrews.ac.uk/~www_pa/Scots_Guide/audio/clipping/page2.html

And they all come to the same conclusion that a tweeter is over driven by power that destroys the tweeter, not the fact that the amp is clipping, period. That is the end of the story

 

[mtrycrafts]

Something in the speakers failed, and I'd wager the most likely cause is clipping the amp. It doesn't matter if you have a 105dB efficient speaker in a bathroom if you feed it with a bad signal. We don't know the exact conditions, or what the damage actually was, so it's hard for anyone to say for sure, but based on what we do know, that is my guess.

A buddy of mine fried two tweeters, in two different speakers one time. He had a Denon 4801 driving KEF Q7s and Studio 20s for surrounds. He tossed the remote down somewhere and it hit something and pushed the "+" button in, turning the volume all the way up. He *RAN* over to the receiver and turned it off and hoped for the best, but when I came over to check it out, it was immediately evident that both of the main and surround speakers on the right side had no highs coming from them. Closer inspection showed that both speaker's tweeters had gone bye-bye. We pulled the one from the Studio 20 and it definitely smelled burnt. You could see where the thin voice coil wire had heated up and melted like a light bulb element. The KEF tweeter was not as obvious, but it also smelled burnt and definitely did not work. After replacing both tweeters, both speakers play just fine - no obvious or audible damage to any of the other drivers.

I would strongly suggest you read all three of the links:

http://sound.westhost.com/tweeters.htm

http://www.rane.com/note128.html

http://www.st-andrews.ac.uk/~www_pa/Scots_Guide/audio/clipping/page2.html

This supports what WmAx is saying. The tweeters blow because of excess power the tweeter cannot handle.

If it was clipping alone, then I suggest you attach a Walkman to that speaker and clip it, see if that clipping will destroy the speakers

 

[mtrycrafts]

Of course, the tweeter was probably rated for about 10 - 15 watts. A few seconds of power greatly exceeding that, distorted or not, will overheat the voice coil.

-Chris

Yes, the speaker doesn't care if a signal is distorted as long as its power is at or below the tweeter's limit. Power destroys, not the clipping.

Perhaps a Walkman driven to clipping should then destroy all speakers on the market, right

 

[mtrycrafts]

I'm with Mac on this one. Clipping has the potential to destroy speakers in seconds. More than enough power will simply allow you to turn it up to levels that nobody would ever listen at without distortion, up to the point at which the drivers or crossover fail. There's no question in my mind that this is what happened, because I've done it myself with my first stereo, and I've seen others do it many, many times it both home and car audio.



Not true. I read an article on clipping that says that what happens is the amp's draw actually spikes up considerably, and even a low power amp can deliver far beyond it's rated power. However, when the amp reaches the point ot saturation where it can no longer deliver the required amount of current to keep the waveform correct, clipping will occur creating the distortion that you hear. That distortion results in the drivers heating up and quite possibly failing. It isn't a guarantee that it will happen in every case, but that IS how it happens. As Mule notes - if you feed a speaker with CLEAN signal, even in excess of the speaker's rating, it is not likely to have an issue. If you feed any speaker with a distorted signal for a length of time, it will eventually fail.

Mac is just plain wrong!!!

A tweeter blows due to power beyond its limit:

http://sound.westhost.com/tweeters.htm

http://www.rane.com/note128.html

http://www.st-andrews.ac.uk/~www_pa/Scots_Guide/audio/clipping/page2.html

 

[mtrycrafts]

This is turning into a BS argument too. Look if the guy wants his eardrums to bleed then fine. He needs a bigger amp with PA speakers. He abused his system and grenaded it. Simple as that. It is my opinion that the amp clipped and did it, Chris and Mrty think it was a plethora of other things. The damn thing blew, the guy needs new gear if he wants to go deaf, so buy an amp and some speakers.

Thread Over.

Of course the speaker blew, so he said and no reason not to believe him.

Why it blew is the question. Too much power, not other factors, to the driver that blew, period, end of the real story.

 

[mtrycrafts]

I'm not even sure if you guys are still talking about my SB2s, but from what ive read in this thread, it sounds like clipping was the culprit. It was the drivers that blew on the speakers, not the tweeter. I was havin a party at my place and they had been playing loud as hell for hours, like +5 db volume setting. I was usually around to stop the stupid drunk kids from turning up the volume, but i was preoccupied with a girl, but thats a bullet ill always take

No, not the clipping at all. The driver was over driven with power. The driver is not as good as it may be rated to be.

If it is clipping alone, then you should be able to bow it with a Walkman at clipping, right?

 

[WmAx]

What I am saying is, when that signal is reproduced on a speaker that is not being driven at the limit of the amp driving it, it isn't creating the same condition at the speaker that a clipped signal being fed to it by an amp that IS clipping does.

If this is true, then you should be able to explain how to identify signals (1) from (2) in the previous excercise.

-Chris

 

[mtrycrafts]

I'm not questioning whether or not the signal was clipped when creating the sound that was recorded or saying that the waveform that is recorded is different from what was originally created. What I am saying is, when that signal is reproduced on a speaker that is not being driven at the limit of the amp driving it, it isn't creating the same condition at the speaker that a clipped signal being fed to it by an amp that IS clipping does. If you took pink noise and sent it to a speaker within the limits of the amp, then what you get is pink noise. If you turn it up to the point where you clip that amp and distortion is present, then the lack of the amp's ability to deliver sufficient power for the speaker to reproduce the signal is what is actually causing the distortion in the speaker, not the signal itself.

But so what?
I think you are not seeing the forest from the trees If the amp is just at the initial unset of clipping from your pink noise experiment with full frequency band a speaker will fail if it receives power beyond its rating, period. If that amp is less powerful than the individual drivers themselves are, it cannot get more power. As you increase the clipping, the sine wave begins to look like a square wave, the average power will increase to the point of its peak power capability 1.414x, all across the bandwidth since you are using a full band pink noise. If a driver cannot take that power, it smokes. If you replace that 100 amp with a 1000w amp and using full band pink noise, as you increase power above 100 watts, well below that amps clipping point, you will start smoking the speakers as you approach their power limit, yet the amp is not clipping. What blew the speakers then??? POWER.

I hope you will read those 3 links I posted.

 

[mtrycrafts]

In that case, it is not the clipped waveform that damage the speaker but the way up rms output resulting from amp clipping that does it, right?

Yes, absolutely. If one of the drivers in the speaker is not rated for that power, it smokes
If a speaker in fact has 250 RMS watt drivers in it, including the tweeter, a 50 watt amp just cannot smoke the tweeter, period.
But, speakers are not rated with all drivers capable of that rated power for good reasons.

 

[mtrycrafts]

Dont think so. When a speaker is rated at 125W/RMS 500w/peak that means it will handle 125 W of music content.
He blew the speakers because he clipped the amp, period end of story.

Whichever speaker went blew because it received more power than it was rated for, period, not because the amp clipped.

For your case to be true, then a 1000W amp will not smoke any of the drivers in the speaker up to the limit of the amp as long as it is not clipping??? Please explain further, thanks.

 

[WmAx]

Ok Chris, you know darn well that when an amp clips its RMS output goes way up. that is the point he is trying to make, I believe.

How can RMS output be increased because of clipping? Instead, don't you mean that total power is increased, as a result of the added harmonics as a result of the clipped waveform?

-Chris

 

[MacManNM]

How can RMS output be increased because of clipping? Instead, don't you mean that total power is increased, as a result of the added harmonics as a result of the clipped waveform?

-Chris

Symantics. you know what I meant. A square wave (clipped audio amp output) has more AVERAGE power than an unclipped sinewave. So a recorded distorted signal and a clipped amp are no where near the same.

 

[MacManNM]

Mac is just plain wrong!!!

A tweeter blows due to power beyond its limit:

http://sound.westhost.com/tweeters.htm

http://www.rane.com/note128.html

http://www.st-andrews.ac.uk/~www_pa/Scots_Guide/audio/clipping/page2.html

His tweeters never blew, I believe he stated that. Tweets rarely blow due to clipping, they never see DC (the clipped waveform). The crossover prevents that. So how am I wrong?

 

[Votrax]

His tweeters never blew, I believe he stated that. Tweets rarely blow due to clipping, they never see DC (the clipped waveform). The crossover prevents that. So how am I wrong?

A highpass (tweeter) crossover just filters out lower frequencies and passes the higher frequencies. So you can still clip high frequency signals. Of course since it is a higher frequency the duration of pure DC is much shorter.

 

[PENG]

A highpass (tweeter) crossover just filters out lower frequencies and passes the higher frequencies. So you can still clip high frequency signals. Of course since it is a higher frequency the duration of pure DC is much shorter.

How could the d.c. components go through the high pass filters?

 

[WmAx]

Symantics. you know what I meant. A square wave (clipped audio amp output) has more AVERAGE power than an unclipped sinewave. So a recorded distorted signal and a clipped amp are no where near the same.

No, just trying to keep the thread accurate.

As for reproducing clipping via an unclipping amp vs. the actual clipping amp, it's the same signal on the output terminals. How would you tell the difference in the prior comparison I presented?

-Chris

 

[MacManNM]

A highpass (tweeter) crossover just filters out lower frequencies and passes the higher frequencies. So you can still clip high frequency signals. Of course since it is a higher frequency the duration of pure DC is much shorter.

Hence it's not DC. See the crossover is a filter, the transients get through, but the DC that creates thermal problems doesn't. Of course this is true until you short the cap in the crossover, and grenade the driver.

 

[mtrycrafts]

His tweeters never blew, I believe he stated that. Tweets rarely blow due to clipping, they never see DC (the clipped waveform). The crossover prevents that. So how am I wrong?

What???????

DC???? Please tell me that you made a simple mistake??? THERE IS NO DC AT CLIPPING!!! Are you insinuating then that a square wave is a DC signal????
You are way out in left field.

Read the 9th and 10th poster where they discuss just this very same silly DC issue:

http://groups.google.com/group/rec.audio.misc/browse_thread/thread/f35bbfa2dab67181/ff63c42a0db22c8c?lnk=st&q=amp+clipping&rnum=3#ff63c42a0db22c8c



Yes, I saw after my post and others where he clarified that he blew his main driver. So what. His guest just over powered the driver. IT is excess POWER over the rated power that smokes the drivers, not clipping per say.

As to your other comment on tweeters, well, they are the ones that seem to smoke when amps are clipped!!!!!! You know why???? Tweeters are way under powered compared to the other drivers!!! The Power to them from the harmonics, a result of clipping, is too powerful for the tweeters, they smoke. Read the links!!! There is no DC at CLIPPING anywhere!!!!

 

[mtrycrafts]

How could the d.c. components go through the high pass filters?

There is no DC at clipping; another audio myth. Please read the 9th and 10th posters at this link that discuss this very same issue. NO DC anywhere It has sine wave AC harmanics creating what looks like a square wave, AC

http://groups.google.com/group/rec.audio.misc/browse_thread/thread/f35bbfa2dab67181/ff63c42a0db22c8c?lnk=st&q=amp+clipping&rnum=3#ff63c42a0db22c8c

 

[MacManNM]

No, just trying to keep the thread accurate.

As for reproducing clipping via an unclipping amp vs. the actual clipping amp, it's the same signal on the output terminals. How would you tell the difference in the prior comparison I presented?

-Chris

Ok Chris, you are trying to bait me now. You know darn well that a clipped signal into an amp is going to look the same as the signal put into the amp when it's clipping. Now, if the input level is increased, then the duty cycle of the output will increase, hence increasing the average output power. Im sure that you will say that there is some technical problems with the way I describe this, but you know what I'm saying.