clip led on amps - Page 4 - Home Theater Forum and Systems - HomeTheaterShack.com

Old 03-02-08, 02:30 PM
Shackster
deromax

Join Date: Nov 2007
Posts: 49
Re: clip led on amps

Quote:
thxgoon wrote: View Post
Semantically, yes.
If the semantic domain is right, then the practical domain must be right also. This is science, not black magic!
Quote:
thxgoon wrote: View Post
Think about what is happening to the drivers at those points. As the voltage is changing in a linear way the subwoofer cone is moving in a way simulating the curve and the power is dissipated as useful work - sound. I'm sure I'm preaching to the choir. Now as soon as the signal clips and you hit that flat top, the woofer is 'held' at that spot and the woofer is no longer doing useful work.
False assumption. The driver IS doing something useful : it's dutyfully reproducing the square wave it is sent!!! You listen to music and mix some, I'm sure you've already heard an electric guitar with overdrive? Well, it's more or less sound clipping. Would you say that 1) Guitar amplifier's speakers are routinely destroyed by the distorded sound they are reproducing and 2) Those speakers are not doing actual useful work, that they are silent? Or alternating between sound/no sound at a rate of several hundred times a second?

Also, if you analyse the movement of a speaker cone reproducing a sine wave, you'll see that at the moment where the sine wave crests, the cone will slow down, then stop for a minute instant, then accelerate in the opposite direction. So there *is* a very short time where the cone is standing still while reproducing a sine wave. Would you qualify this as DC too? If no, then what is the required lengh of time of stop motion to be considered DC?
Quote:
thxgoon wrote: View Post
During this short time full power is sent through the VC with no way to be dissipated other than heat.
There is more heat because a square wave contain more power than in a sine wave of equal peak level. Exactly twice as much. Hopefully they outta thought you this (or they will) in EE school?
Quote:
thxgoon wrote: View Post
This is the 'dc' I am referring too. While not fitting the definition of dc over time
If it don't fit the definition, then it must not be real. DC is a continuous phenomenon that must be sustained to be of any relevance. The flat part of a clipped signal is a few millisecond long, too short to be considered DC imho.
Quote:
thxgoon wrote: View Post
when you add up all of these little bursts you end up with a lot of excess power doing no work on the driver.
A heavily clipped signal will audibly sounds louder than a clean signal. If more sound is produced, then it follows that more work is accomplished by said driver, not less. If the driver cannot cope with the increased work it is asked to do, then failure can happen.
Quote:
thxgoon wrote: View Post
Is there some larger point I am missing here?
I'm affraid yes! That is, a square wave cary twice as much energy as a sine wave of equal peak amplitude. You often see power amp with power specs like 100 watts RMS, 200 watts max. This means that if you were to clip a 100 watts sine wave so much as to render it a square wave, the resulting power would be 200 watts RMS. No wonder a driver can be burned up by heavy cliping!

I don't see how I could make my point any better than this, so I'll keep it at it. Maybe someone else may explain this better than me?

Regards,
deromax is offline

Old 03-02-08, 08:45 PM
Elite Shackster

thxgoon

Join Date: Feb 2007
Posts: 1,506
Re: clip led on amps

Deromax, I understand your point to the T. I know exactly what you are talking about. I have known what you have been trying to say from your first post. Please try to consider the point I am trying to make even if it falls outside of your 'definition' of DC. Given a short enough time frame, my definition is just as valid. Please put that aside for just a second.

Yes, the woofer is doing useful work on a square wave when traveling from peak to peak, that is not the area I am talking about. The area I am talking about is when the woofer is 'stalled' at those peaks. There it is not doing useful work and it only takes a look at the graph and the definition of work to figure this out. Work = force*distance. Lots of force, no displacement, no work.

Lets think about the square wave example. Lets say the amp is rated at 50 watts and you are clipping it so that the average power beneath the wave form is 75 watts. I don't care to do the math but lets say that would = a 200 watt non-clipping signal. Conventional logic and professional recommendation and experience tells us that one of the fastest ways to blow speakers is to underpower and clip the signal going in to them. Ok, so now lets take a 10,000 watt amp and feed it that same signal, except now it isn't clipping. Now you have that area from the square wave in addition to the area now above it under the non-clipped wave form and you have more power than you had before. But supposedly this won't wreck the woofer. So what is the difference???
thxgoon is offline
Old 03-02-08, 11:13 PM
Shackster
deromax

Join Date: Nov 2007
Posts: 49
Re: clip led on amps

A watt rms is a watt rms is a watt rms. The waveform is irrelevant and the driver don't care about it. A 75 watts rms square wave is just that : 75 watts rms of power. There is no such thing as an "clean-equivalent" watts. Do you understand what a watt rms is?

A fridge magnet that holds a piece of paper on the fridge door is not doing any work but it is usefull anyway. When a driver cone is standing still it may not doing "work" per se, but it as least holds the air pressurized and is faithfully reproducing the waveform it is asked to. The flat horizontal part is part of the signal, how else could the driver reproduce it, if not by standing still? This is not iddle time or wasted energy.

All this discussion is irrelevant anyway since no respectable sound system should be driven to significant clipping on a regular basis, be it hifi, car audio or sound reinforcement.
deromax is offline
Old 03-03-08, 12:18 PM
Elite Shackster

thxgoon

Join Date: Feb 2007
Posts: 1,506
Re: clip led on amps

Quote:
deromax wrote: View Post
Do you understand what a watt rms is?
Yes, I referred to it in my previous post.

Quote:
When a driver cone is standing still it may not doing "work" per se, but it as least holds the air pressurized and is faithfully reproducing the waveform it is asked to. The flat horizontal part is part of the signal, how else could the driver reproduce it, if not by standing still? This is not iddle time or wasted energy.
No, it is not holding the air pressurized, the sound wave has already been sent and it is standing still. You are right, there is no other way for it to reproduce the signal other than standing still and this is the problem. It takes power to hold it there and when standing still it is doing no work (making no sound) so the power is dissipated as heat instead of sound. This is the point.
thxgoon is offline
Old 03-03-08, 12:40 PM
Elite Shackster

thxgoon

Join Date: Feb 2007
Posts: 1,506
Re: clip led on amps

Quote:
deromax wrote: View Post
There is no such thing as an "clean-equivalent" watts.
The clean equivalent is the area under the red part of this curve. This is what I don't get about the average RMS argument for clipping. Yes it exists, but usually the camp will say that a larger power amplifier is the solution to blowing speaker in this scenario. But looking at this graph you can clearly see that the same signal strength on a more powerful amp that is not clipping results in still more power.
thxgoon is offline
Old 03-03-08, 03:16 PM
Elite Shackster
brucek

Join Date: Apr 2006
Posts: 7,514
Re: clip led on amps

Quote:
Do you understand what a watt rms is?
Strictly speaking (as I'm sure deromax and Thxgoon know), RMS power is quite misused and isn't really representative of equivalent heating power (average power), but consumer electronics has chosen to incorrectly use that term anyway. A rather nice entry level discussion can be found here for those that don't have an engineering background.

Anyway, let's keep this discussion completely respectful of each other.

A clipped (or any other) signal can contain some DC component, but the flat top as deromax has pointed out is not DC. If it were, an AC coupled amplifier would not pass it. So, just as a square wave will pass through a capacitor, so will the clipped signal from an amplifier reach a woofer. (A caveat to that of course is that given the value of coupling capacitance and its inherent reactance, it can create a high pass filter and could indeed distort a low frequency square wave somewhat. But we'll ignore that).

My understanding as to why a woofer would be damaged by clipping is through excess heat generated in the voice coil that is normally cooled by the cone movement. When the clipped portion of the signal is dropped across the coil, it literally stops (as Thxgoon points out), but passes large amounts of current without the required movement that is so necessary for cooling.

Yes, the coil also stops when an unclipped signal is sent to it, but only momentarily and not enough to affect cooling. A lot of air is forced through the magnetic gap as the voice coil moves with a signal. This cools the coil and dissipates the heat that would build up if movement stopped. A clipped signal increases the time that the cone is stopped. It also increase the average power that has to be dissippated. As deromax points out, you can develop twice the effective power from a square wave as opposed to a sine wave.

brucek
brucek is offline
Old 03-03-08, 04:11 PM
Shackster
deromax

Join Date: Nov 2007
Posts: 49
Re: clip led on amps

Hello. One must be cautious considering amplifiers of different power, since it's no longer apples and oranges! What I say is that *one* amp that is producing a sine wave at maximum power will produce twice as much if the wave form is allowed to become square by heavy clipping.

Now, considere a 100 watts speaker that is fed *continuous* power.

Send a 100 watts sine wave from a 100 watts amplifier to the speaker and it will survive.
Using the same amp, clip the signal so it becomes 200 watts, the speaker will blow.
Switch to a 200 watts amp and send a 200 watts sine wave, the speaker will blow.
The wave form have nothing to do with this, it's just the doubled power that is at play.

Remember I'm talking about a *continuous* sine, a harsh condition rarely seen in real life.

Now, if you know what you are doing and have a good ear to detect audible strain in speakers, the 100 watts speaker will survive just fine with a 200 watts power amplifier. This have little to do with the shape of the waveform, it's because real music is not continuous, it have a somewhat low average power with shorts peaks 10 dB (or more) higher thn the average (crest factor). Since the peaks are short, the speaker handle them just fine. In the end, you rae way better with 200% peaks of power than peaks at 100% but clipped.
deromax is offline
Old 03-03-08, 04:14 PM
Shackster
deromax

Join Date: Nov 2007
Posts: 49
Re: clip led on amps

Quote:
brucek wrote: View Post
My understanding as to why a woofer would be damaged by clipping is through excess heat generated in the voice coil that is normally cooled by the cone movement. When the clipped portion of the signal is dropped across the coil, it literally stops (as Thxgoon points out), but passes large amounts of current without the required movement that is so necessary for cooling.
Hello Brucek! My understanding is different! If you plug in a soldering iron, it takes how much time for the tip to become hot? Maybe a minute? The top portion of a 30 Hz square wave have a duration of 16.6 millisecond. Just how much heat can be produced in such a short time? Of course the system is standing still, but it was moving just mere milliseconds before and it will be moving shortly after, as well. I'd even argue that the cone move *faster* (cools better too) reproducing the extremely fast rising edge of a square wave than the more gradual moving that a sine wave would incure.

Forgeting our pre-conceived ideas for just a minute, we see that both theories can be realistic. I'd say the truth is probably somewhere in between!

I'm done with this discussion. Cheers!
deromax is offline
Old 03-03-08, 06:13 PM
Elite Shackster

thxgoon

Join Date: Feb 2007
Posts: 1,506
Re: clip led on amps

Hey Brucek, thanks for chiming in. I was hoping a veteran EE would notice this discussion.

Quote:
deromax wrote: View Post
Send a 100 watts sine wave from a 100 watts amplifier to the speaker and it will survive.
Using the same amp, clip the signal so it becomes 200 watts, the speaker will blow.
Switch to a 200 watts amp and send a 200 watts sine wave, the speaker will blow.
The wave form have nothing to do with this, it's just the doubled power that is at play.
This is my understanding as well of the blowing due to average power argument. The trouble is that the same camp usually recommends a higher power amp to prevent damage for the same output. In other words it seems that they think the 200 watt amp would not still blow.

I think Bruceks's explanation makes the most sense. VC are very light and have lots of power running through them so it wouldn't take long to heat up. Instead of thinking of a soldering iron, think of a light bulb on the other extreme.
thxgoon is offline

 Bookmarks

 Tags amps , clip , led

Message:
Options

## Register Now

Random Question
Random Question #2

User Name:
OR

## Log-in

Human Verification

In order to verify that you are a human and not a spam bot, please enter the answer into the following box below based on the instructions contained in the graphic.