How many on ASR don't runs subs of any kind.? & why.....

[Mort]

Wikipedia, but it's surely a well-known problem if amps are asked to deliver more power they are than designed to do

It was an hypex nc502 amp on a philharmonic audio BMR monitor. I really doubt the amp was overdriven.

 

[NTK]

"Because the clipped waveform has more area underneath it than the smaller unclipped waveform, the amplifier produces more power than its rated (sine wave) output when it is clipping. This extra power can damage the loudspeaker. It may cause damage to the amplifier's power supply or simply blow a fuse."

This statement has no relevance nor meaning. Regardless of whether the waveform is clipped or not, a "smaller" waveform (assuming you are talking about a sine wave and smaller means lower amplitude) always have less area than a "larger" one.

 

[gwing]

Wikipedia, but it's surely a well-known problem if amps are asked to deliver more power they are than designed to do

Thanks, Wikipedia can be a great source of info and I'm actually one of their donors myself. Specifically referring to 'https://en.wikipedia.org/wiki/Clipping_(audio)' which seems to be the source of your quotation:

I do recommend you carefully re-read that page which does describe the issue well, but a conclusion that the destruction of woofers is because of amps delivering more power than they are designed for is, I think, a misreading of the article.

It is true that turning up the gain of an amplifier so that you are asking it to deliver more volts than it can provide does result in the signal being clipped at the maximum output voltage of your amplifier. It is also true that the area under the signal graph is greater for the clipped signal than an unclipped sine wave signal from the same amplifier and hence more energy will be going into the speaker however the clipping isn't the cause of this it is because we have turned up the gain to produce a greater signal. If we swapped our amplifier for a more powerful higher voltage capable amplifier and used that instead with the required gain set to the level our smaller amplifier clipped at then that larger amplifier would be able to deliver that requested higher amplitude signal without clipping and *even more* energy would be going into the woofer. i.e. the problem here is not that the clipping is damaging the woofer but that we are simply driving it beyond its capability and a more powerful, non-clipping, amplier given that same gain requirement would have even greater area under its signal curve and thus would drive the woofer even harder.

Note that the case is different for tweeters. As the Wiki article describes, clipping the signal drives its shape towards that of a square wave and a square wave (as a greatly simplified explanation) contains a lot of high frequency overtones. This introduces a lot of spurious high frequency signals into the signal, in addition to the high notes in the music itself thus increasing the energy going into the tweeter and causing said tweeter to burn out. However this problem from clipping is a problem for the tweeters and isn't going to damage your woofers. In really extreme cases I guess it might also damage a midrange driver.

 

[levimax]

Thanks, Wikipedia can be a great source of info and I'm actually one of their donors myself. Specifically referring to 'https://en.wikipedia.org/wiki/Clipping_(audio)' which seems to be the source of your quotation:

I do recommend you carefully re-read that page which does describe the issue well, but a conclusion that the destruction of woofers is because of amps delivering more power than they are designed for is, I think, a misreading of the article.

It is true that turning up the gain of an amplifier so that you are asking it to deliver more volts than it can provide does result in the signal being clipped at the maximum output voltage of your amplifier. It is also true that the area under the signal graph is greater for the clipped signal than an unclipped sine wave signal from the same amplifier and hence more energy will be going into the speaker however the clipping isn't the cause of this it is because we have turned up the gain to produce a greater signal. If we swapped our amplifier for a more powerful higher voltage capable amplifier and used that instead with the required gain set to the level our smaller amplifier clipped at then that larger amplifier would be able to deliver that requested higher amplitude signal without clipping and *even more* energy would be going into the woofer. i.e. the problem here is not that the clipping is damaging the woofer but that we are simply driving it beyond its capability and a more powerful, non-clipping, amplier given that same gain requirement would have even greater area under its signal curve and thus would drive the woofer even harder.

Note that the case is different for tweeters. As the Wiki article describes, clipping the signal drives its shape towards that of a square wave and a square wave (as a greatly simplified explanation) contains a lot of high frequency overtones. This introduces a lot of spurious high frequency signals into the signal, in addition to the high notes in the music itself thus increasing the energy going into the tweeter and causing said tweeter to burn out. However this problem from clipping is a problem for the tweeters and isn't going to damage your woofers. In really extreme cases I guess it might also damage a midrange driver.

This subject has a lot of myths surrounding it. Here is an article where they take an in depth look at it. https://sound-au.com/tweeters.htm

The conclusion of this article is that clipping has little if anything to do with speaker damage, rather it is the maximum output of the amp that damages speakers.

 

[Frank Dernie]

I'm struggling to think why distortion would be likely to blow a bass unit. When we get distortion, typically from clipping when the amp is overdriven, that produces higher harmonics of the actual signal thus driving tweeters to excess but why would that damage the bass units?

Occam's razor suggests it is indeed likely to be just the bass unit being driven to excess, which could easily happen if you have a powerful amp and eq trying to get more out of it than it can deliver.

Particularly if you attempt to fill in FR nulls in the bass with DSP an easy mistake for the non-technical to make.

 

[krabapple]

Particularly if you attempt to fill in FR nulls in the bass with DSP an easy mistake for the non-technical to make.

Automated 'room correction' DSP won't do this, so it's really a problem among do-it-yourselfers using e.g. REW.

 

[gwing]

Automated 'room correction' DSP won't do this, so it's really a problem among do-it-yourselfers using e.g. REW.

I guess there isn't one answer for all devices but what is the poor automated room correction DSP going to do if there is a null in the bass response? Sure it can implement a tone curve so that you have cuts everywhere else in the frequency response and indeed technically no boost in the bass but what happens then? It just sounds quieter so you turn up the volume and there you have your bass boost.

 

[krabapple]

I guess there isn't one answer for all devices but what is the poor automated room correction DSP going to do if there is a null in the bass response?

Almost nothing. As it should.

Sure it can implement a tone curve so that you have cuts everywhere else in the frequency response and indeed technically no boost in the bass but what happens then? It just sounds quieter so you turn up the volume and there you have your boost in the bass

But if you turn it up the post-EQ relative levels between frequencies remain the same, no?

Anyway, EQ isn't the solution to modal nulls. Better low-bass-source placement wrt MLP wrt room geometry, is.

 

[Frank Dernie]

I guess there isn't one answer for all devices but what is the poor automated room correction DSP going to do if there is a null in the bass response? Sure it can implement a tone curve so that you have cuts everywhere else in the frequency response and indeed technically no boost in the bass but what happens then? It just sounds quieter so you turn up the volume and there you have your bass boost.

You can't fill in nulls until the laws of physics are repealed so any knowledgeable engineer writing automated room compensation will ignore the nulls (or maybe, if he/she is adventurous add 3dB or so of boost).

 

[Willem]

I guess there isn't one answer for all devices but what is the poor automated room correction DSP going to do if there is a null in the bass response? Sure it can implement a tone curve so that you have cuts everywhere else in the frequency response and indeed technically no boost in the bass but what happens then? It just sounds quieter so you turn up the volume and there you have your bass boost.

Equalization will not cure dips. Only different locations and/or multiple subs will help here.

 

[Frank Dernie]

Almost nothing. As it should.


But if you turn it up the post-EQ relative levels between frequencies remain the same, no?

Anyway, EQ isn't the solution to modal nulls. Better low-bass-source placement wrt MLP wrt room geometry, is.

None are daft enough to cut all output to the level of the lowest point in the FR afaik

 

[Mort]

indeed technically no boost in the bass but what happens then? It just sounds quieter so you turn up the volume and there you have your boost in the bass

The volume knob increases evenly across the spectrum while the cuts are limited, calculated and surgical. The results are not going to be the same.

Anyway, some (older especially) DSP tried to boost room nulls. That's true. But newer versions (GLM 5.x for example) rely on cuts only, afaik. Some even include room measurements to better identify the nulls and avoid boosting.

 

[Frank Dernie]

Anyway, some (older especially) DSP tried to boost room nulls.

Did it????
I wouldn't imagine any acoustics engineer understanding little enough to think they could do the impossible

 

[levimax]

Did it????
I wouldn't imagine any acoustics engineer understanding little enough to think they could do the impossible

This year I had DIRAC DLBC (latest version) with a stereo set up with 2 subs that definitely tried to fill in a 100 Hz null with some boost. While the FR measured smoother with REW than anything I could do manually, if I turned it up I could hear the woofers on the mains bottoming out and the red clipping light coming on (Neumann KH 310's). I had never seen or heard that before. When I raised the crossover from 70 Hz to 120 Hz the issue went away. I am sure DIRAC doesn't go crazy with boost but even 3-5 dB boost in a high energy area like 100 Hz can cause problems. I am using manual DSP now and I never use any boost.... best way to get rid of sharp measured dips is Psychoacoustic smoothing

 

[bmc0]

In a crossover, the output is the convolution of the response of the electrical XO and the driver.

For each individual band. The final result is the sum of all bands.

Because the clipped waveform has more area underneath it than the smaller unclipped waveform, the amplifier produces more power than its rated (sine wave) output when it is clipping.

This statement is true by itself, but you've misinterpreted the implications. If the clipped waveform had not been clipped, the power would be even greater.
An idealized amp (ignoring any PSU sag, etc.) will produce up to twice its rated unclipped output when driven far into clipping. In other words, if you ask a 10W amp to produce 100W, you'll get approximately 20W. But if you ask a 100W amp to produce 100W, you'll get 100W.

 

[krabapple]

None are daft enough to cut all output to the level of the lowest point in the FR afaik

You could do it halfway , and get halfway there.

 

[rdenney]

Thank you for the insight into your choice of instrument to suit the venue you are playing in - very interesting. This is exactly akin to my claim that the speakers must be carefully chosen to suit the room in which they are installed. Type of speaker is the foremost consideration (horns suit my difficult room, electrostatics sadly don't), but this is not akin to using DSP.

With your musical instruments, you select the one to suit the venue, but you can't adjust its sound further to compensate for room acoustics to achieve perfectly flat response at the listener's seat. That could only be achieved by using a microphone to measure, then DSP to adjust the signal and then to transmit the adjusted sound to the audience - through loudspeakers. What an abominable thought! Let's accept the source sound reproduced by the tuba, violin, flute, or loudspeaker that has been carefully selected as suitable for the space in which it will be playing and be happy with the pleasure it gives the listener.

My tuba creates the art (or would if I was better), and if I was making a recording of solo works, a lot of effort would go into tailoring the result to reflect my intentions.

The playback system’s job is to be faithful to that art, and that’s the objective that room correction (including equalization) serves.

Why would I want my carefully crafted recorded tuba sound to be subject to someone’s ringing room modes and amateur paranoia about equipment that pros use without a second thought?

Rick “nothing to worry about—no such recording is in the works” Denney

 

[bungle]

But newer versions (GLM 5.x for example) rely on cuts only, afaik.

I think it uses boosts too but only broad ones and usually connected with a cut, but I think it tries to limit that to 6db (and manually you can do whatever you want, it will give a warning though if you exceed 6db), example:

And here the end results with REW (the error around 1400 Hz is the cross-over of two-way S360A, issues with bass are because of small room/placement/listening position/lack of treatment, the highest frequency roll-off happens I think because of distance of 3 meters, so higher frequencies start to lose their power faster, the room seems to give some 6Hz low end extension):

 

[Hear Here]

For each individual band. The final result is the sum of all bands.


This statement is true by itself, but you've misinterpreted the implications. If the clipped waveform had not been clipped, the power would be even greater.
An idealized amp (ignoring any PSU sag, etc.) will produce up to twice its rated unclipped output when driven far into clipping. In other words, if you ask a 10W amp to produce 100W, you'll get approximately 20W. But if you ask a 100W amp to produce 100W, you'll get 100W.

The point I was making was in regard to the damaged woofers that were played at high volume. It could have been that the damage was more related to amp distortion (that undoubtedly is not good for speakers - or ears for that matter) than solely high volume. However we've since established that the amp is a 500 watt Class D one, so unlikly to be distorted as a result of being over-driven. Conclusion - speakers being driven too loudly and probably the drivers are too small for such abuse!

 

[ZolaIII]

@bungle it does fine and you can adjust limits manually. First peek is fundamental and it skips it's first two harmonics and first belonging deeps. Why did it skip 700 Hz peek still in reach of IIR PEQ Q factor limit I don't know, I wouldn't leave it like that.