Analyze This Setup

[Roshanbhoj]

So I have a Behringer VS1520 speaker Eurolive speaker. That was a crazy deal locally that I couldn't pass up. I thought I would use it for house parties. It's never been used and been sitting in my garage for months because I have never had an amp. So now I found one locally that I would like to know how it would turn out if I got it to use it with the speaker. I know you guys really know what you're talking about. Here's the speaker specs:

High-performance 600-Watt 2-way PA speaker system (150 Watts Continuous / 600 Watts Peak Power)

Here's the Amp specs:

2x150W @ 4 ohms, bridged mono: 300W @ 8 ohms

From what I know, I would use the bridged mono. I know it's peak power is 600 watt but doesn't that not mean anything? It just matters that the speaker is 150W continuous right or what? How would the amp work out with the speaker?

[Incognito]

You don't want to bridge the amp for your given application, your speakers are 150 watts RMS each at 8 ohms while the amp you're considering only delivers 150 watts at 4 ohms. Generally you want an amp that can deliver 1.5 to 2 times the RMS of the speaker so for your speaker you would be looking at an amp that can deliver around 225 to 300 watts per channel at 8 ohms. Another thing to look out for is how the amp is rated, most amps in the price range you're considering rate their amps output a 1khz, what you want to find out is the amps output at 20hz-20khz.

Don't waste you're time paying any attention to any spec that is rated as peak (100% pure marketing in effect).

[nyquist]

Rule of thumb is to match your amps watts and ohm load with the peak power of you speakers. This is for headroom. So if each speaker is 300w peak @ 8ohms then your amp should be 300watts per channel. So the amp you have your eye on is not the right unit. If you run an amp too small for your drivers (speakers) then you will clip the amp (square it/distortion) this will destroy the voice coils in the speakers. Save your cash for a bit longer. An amp ratted to 300 watts @ 8 ohms per channel is the way to go. As mentioned above marketing is rife with speaker specs. The only numbers you need to know is the RMS watts of the speaker. Times the RMS by 2 will give you the real peak figure and the amp power you will need.

You mentioned bridging in your post. This essentially utilizes the amps 2 channels and combines them into the 1 channel. This makes the amp a mono amp only. It is not really the best method for this particular setup IMO.

[ampnation]

To clarify what Incognito was saying, don't match the wattage first, look at the impedence (ohms). He is telling you right about the amp you want, to go with those speakers, assuming he knows or looked up the specs (I didn't). The QSC GX3 would fit your needs. It is rated at 300w using a 1kHz test signal meaning it will be a little lower using a full band signal (20Hz-20kHz) but that's okay because 300w is the MOST you want your amp rated at without proper limiting in place. Frankly, when you can, you should also add a loudspeaker management system or at least something with limiting.

[ampnation]

Rule of thumb is to match your amps watts and ohm load with the peak power of you speakers. This is for headroom. So if each speaker is 300w peak @ 8ohms then your amp should be 300watts per channel. So the amp you have your eye on is not the right unit. If you run an amp too small for your drivers (speakers) then you will clip the amp (square it/distortion) this will destroy the voice coils in the speakers. Save your cash for a bit longer. An amp ratted to 300 watts @ 8 ohms per channel is the way to go. As mentioned above marketing is rife with speaker specs. The only numbers you need to know is the RMS watts of the speaker. Times the RMS by 2 will give you the real peak figure and the amp power you will need.

You mentioned bridging in your post. This essentially utilizes the amps 2 channels and combines them into the 1 channel. This makes the amp a mono amp only. It is not really the best method for this particular setup IMO.


Rule of thumb is to match your amps watts and ohm load with the peak power of you speakers. This is for headroom. So if each speaker is 300w peak @ 8ohms then your amp should be 300watts per channel. So the amp you have your eye on is not the right unit. If you run an amp too small for your drivers (speakers) then you will clip the amp (square it/distortion) this will destroy the voice coils in the speakers. Save your cash for a bit longer. An amp ratted to 300 watts @ 8 ohms per channel is the way to go. As mentioned above marketing is rife with speaker specs. The only numbers you need to know is the RMS watts of the speaker. Times the RMS by 2 will give you the real peak figure and the amp power you will need.

You mentioned bridging in your post. This essentially utilizes the amps 2 channels and combines them into the 1 channel. This makes the amp a mono amp only. It is not really the best method for this particular setup IMO.

I disagree. If he tries running 600w per channel into those speakers, he's likely to fry them unless as I indicated, he has limiters in place in which case 600w is just overkill... or headroom, depending on how you look at it. Program is 2x RMS and Peak is 4x RMS... once in a while you'll see specs with "continuous" and a peak number at 2x continuous...I always assume in those instances that they are using continuous as a synonym for program to be on the safe side, especially since these are usually bottom feeder type cabinets.

To be clear, you're talking 300w but the OP specified his speakers are 600w peak... 300w would be 2x RMS, aka program.

[Roshanbhoj]

Hmmm. Just to clarify I only have on speaker so wouldn't putting all the power into one channel by mono bridging it be okay? This still means that there's 300W on the amp and on the 150W continuous and 300W program speaker which I guess you're saying wouldn't be okay?

[ampnation]

So I have a Behringer VS1520 speaker Eurolive speaker...


Hmmm. Just to clarify I only have on speaker so wouldn't putting all the power into one channel by mono bridging it be okay? This still means that there's 300W on the amp and on the 150W continuous and 300W program speaker which I guess you're saying wouldn't be okay?

OHHHHHH... my/our bad because we have poor reading comprehension hehe... Yes. In that case, you're fine. Make sure you know how to put the amp in bridged mode. For the record, there's at least one person on this board, one who actually designs speakers, who claims you should never use bridged mode unless you are VERY knowledgeable about how it actually works, so I'll say this is probably less desirable than using one channel. Keep in mind that an amp will produce more power to one channel if only one channel is driven but it will still be less than bridge mode.

[sss18734]

What is the amp that you're using?

[Roshanbhoj]

OHHHHHH... my/our bad because we have poor reading comprehension hehe... Yes. In that case, you're fine. Make sure you know how to put the amp in bridged mode. For the record, there's at least one person on this board, one who actually designs speakers, who claims you should never use bridged mode unless you are VERY knowledgeable about how it actually works, so I'll say this is probably less desirable than using one channel. Keep in mind that an amp will produce more power to one channel if only one channel is driven but it will still be less than bridge mode.

Oh okay. I always thought it was as simple as moving the switch to mono bridge mode. I'm glad this will work out though. I can finally use the speaker!

[windspeed36]

Rule of thumb is to match your amps watts and ohm load with the peak power of you speakers. This is for headroom. So if each speaker is 300w peak @ 8ohms then your amp should be 300watts per channel.

You only really need 1.5 to 2x the continuous draw. Never in my lifetime have I heard of someone picking amplifiers based on the cabinets peak power. For example, there would be no point in finding an amp that can put out 6400w continuous for a JBL SRX728 because you are never going to drive the speaker into peak.

[Roshanbhoj]

^I understand that. Thanks for making that straight.

[nyquist]

I've been designing and specing PA systems for over 15 years 10 of which have been with a very large audio supplier. When wo put together a system the method we always use is to look at the capability of the cabinet and assume that yes it will be used to its full potential. This is of course most important for subs. I must have misread the OP's post. 1 Box rated at 150rms x 2 will give 300 peak amp power (generally) so a bridged amp in 8ohms of 300 watts is on the money.

[ampnation]

I've been designing and specing PA systems for over 15 years 10 of which have been with a very large audio supplier. When wo put together a system the method we always use is to look at the capability of the cabinet and assume that yes it will be used to its full potential. This is of course most important for subs. I must have misread the OP's post. 1 Box rated at 150rms x 2 will give 300 peak amp power (generally) so a bridged amp in 8ohms of 300 watts is on the money.

yet spec sheets typically give RMS, program & peak with program and peak being 2x and 4x RMS respectively so you should make it clear you mean 2x RMS when you say peak. Just sayin'

[jayhwk]

I actually disagree with the "old wisdom" of 2x the RMS power for beginners. The difference between 150W and 300W is 3dB. That's the threshold of volume difference that the average audience member will notice. You'll get a barely noticeable volume change, but the potential to damage you speakers will have increased significantly.

I have had a speaker designer from one of the large American loudspeaker manufacturers tell me that he personally would not put more than 4/3 the RMS rating into loudspeakers he designed since he didn't make enough money to recone the drivers all the time. I use loudspeakers he designed, so I follow his advice.

This document by EV which I've posted here a couple times is a really great explanation of power ratings:
http://www.electrovoice.com/downloadfile.php?i=2320

EDIT: I edited out a comment about peak power which I made without thinking. Peak is usually a number made up by a manufacturer and often is meaningless.

[Incognito]

I actually disagree with the "old wisdom" of 2x the RMS power or 1x the peak power for beginners. The difference between 150W and 300W is 3dB. That's the threshold of volume difference that the average audience member will notice. You'll get a barely noticeable volume change, but the potential to damage you speakers will have increased significantly.

I have had a speaker designer from one of the large American loudspeaker manufacturers tell me that he personally would not put more than 4/3 the RMS rating into loudspeakers he designed since he didn't make enough money to recone the drivers all the time. I use loudspeakers he designed, so I follow his advice.

This document by EV which I've posted here a couple times is a really great explanation of power ratings:
http://www.electrovoice.com/downloadfile.php?i=2320

The 1.5 to 2 times RMS is not for use of added SPL but for added headroom.......HOWEVER if you keep a close eye on things being sure to stay out of the red then supplying RMS is perfectly fine (you can deliver less then RMS as well as long as you're speakers are reaching your intended full potential before the amp reaches it's full potential) though a strong limiter would be a good safety net as a just in case.

[SockoPro]

Without getting too technical with all the formulas I will explain it this way.

The higher the impedance (ohms) the easier the amp can make the power more efficiently and have more headroom (gain before distortion). Either way you are not gaining any power capacity with your setup of two speakers.

When you goto say 150 watts @ 8 ohms stereo to 300 watts bridged at 4 ohms that doesn't mean each speaker is getting 300 watts. 300 watts in the coefficient sum both of the 2 speakers in parallel have available to them. To understand the rule you need to know ohms laws.

http://www.the12volt.com/ohm/ohmslawcalculators.asp

Amplifier energy output is like DC power. The formula I am using for the figures below is E = Square Root of (P x R).

Basically

150 watts @ 8 ohms = 34.64102 volts

300 watts @ 4 ohms = 34.64102 volts

Notice your speakers are getting the same amount of available voltage from the amplifier fet rails either way. If you bridge it, the 2 speakers must share the available power. If you leave it in stereo each speaker has a dedicated amount of power.

Now while I have your head spinning what are the impedance (ohms) of your speakers?

When you parallel (daisy chain the two together) you are decreasing your resistance closer to a dead short (0 ohms) and the amp has to work even harder. For example two 8 ohm speakers in parallel equals a 4 ohm load, two 4 ohm speakers in parallel equals a 2 ohm load.

Here is a good read about it. http://www.prestonelectronics.com/audio/Impedance.htm

[jayhwk]

The 1.5 to 2 times RMS is not for use of added SPL but for added headroom.......HOWEVER if you keep a close eye on things being sure to stay out of the red then supplying RMS is perfectly fine (you can deliver less then RMS as well as long as you're speakers are reaching your intended full potential before the amp reaches it's full potential) though a strong limiter would be a good safety net as a just in case.

The definition of "Headroom" is SPL - it's the ability of your system to produce a SPL higher than your target (if your target is 95 dB, and you have 10 dB of headroom, that means your system has a max output of 105 dB).

Therefore, going from a 150W amp to a 300W amp would allow for an extra 3dB of headroom.

[SockoPro]

Headroom on an amp is the unused, usable gain structure available over top of the desired playing level. Headroom gives you the ability to play the entire note without clipping off the edges. If you are clipping the input or outputs on deep notes your under powered.

[Incognito]

The definition of "Headroom" is SPL - it's the ability of your system to produce a SPL higher than your target (if your target is 95 dB, and you have 10 dB of headroom, that means your system has a max output of 105 dB).

Therefore, going from a 150W amp to a 300W amp would allow for an extra 3dB of headroom.

No, you have it all wrong... (as far as the definition of headroom goes at least) Headroom is not to be used for added SPL, it's there for the dynamics of music & a buffer zone from what your speakers require & the max output of the amp. This means you're amp will never run out of stored available power to deliver the full potential of a given speaker.

[Al Poulin]

No, you have it all wrong... (as far as the definition of headroom goes at least) Headroom is not to be used for added SPL, it's there for the dynamics of music & a buffer zone from what your speakers require & the max output of the amp. This means you're amp will never run out of stored available power to deliver the full potential of a given speaker.


I would have to agreee with this. I would also add that live music generally requires much more headroom (especially an uncompressed mic'd kick drum and even microphone for example) than playback/CD type music which is already compressed to shit in the studio (these days)...

Al

[Al Poulin]

I've been designing and specing PA systems for over 15 years 10 of which have been with a very large audio supplier. When wo put together a system the method we always use is to look at the capability of the cabinet and assume that yes it will be used to its full potential. This is of course most important for subs. I must have misread the OP's post. 1 Box rated at 150rms x 2 will give 300 peak amp power (generally) so a bridged amp in 8ohms of 300 watts is on the money.

An inexperienced user matching an amplifier's continuous output with a speaker's peak power handling is simply a recipe for disaster. Most peak ratings are there for sales/marketing reasons and should not be used to match amps to speakers at all IMO.

I would stick with matching the amp's continous output to the speaker's program power for the average user and closer to RMS for someone who's new to live sound/DJ.

If you want to match your amps to the peak rating of your speakers and have proper limiting, high pass filtering etc... that's fine but simply asking for trouble IMO. One OOPS could be very costly...

Al

[ampnation]

I saw a write up some time ago but can't remember who wrote it. It was from someone with great credentials... something like a longtime engineer for a some company like Crown or QSC. It had to do with selecting the right power amp and talked about how the numbers were derived at. As I recall, that person was only speaking of the recommendation as 2x RMS, not 1.5x to 2x and to be sure I would prefer 2x to 1.5x but getting it that precise isn't always possible and given the variation in how manufacturers spec their equipment, one 2x spec is not equivalent to all other 2x specs.

At any rate, what I got from this writeup boiled down to this... 1x RMS is what you will use directly 99% of the time and the power capacity between 1x and 2x RMS is overhead and is there for transients... that is to say, to give you headroom for the occasional spike measured in milliseconds plus making it so your amp isn't working as hard to produce what you need during that 99% of time.

That is the reader's digest condensed version and it really is more complicated than that and in fact even what I wrote isn't exactly accurate I'm pretty sure... it just gives a conceptual overview. It was quite clear to me however that the author DID know all the math behind the recommendation and that there was little reason to doubt his reasoning and thus little reason to use any other benchmark when deciding which amp to use.

[jayhwk]

No, you have it all wrong... (as far as the definition of headroom goes at least) Headroom is not to be used for added SPL, it's there for the dynamics of music & a buffer zone from what your speakers require & the max output of the amp. This means you're amp will never run out of stored available power to deliver the full potential of a given speaker.

You are saying the same thing I am except you've still got a misconception about how SPL fits into this. Read what you're saying carefully:

What is the definition of "full potential of a given speaker" in acoustical terms? What is a "buffer zone from what your speakers require". How do you define "the dynamics of music"? You'll find that all of these qualitative terms can be quantified by an increase in the sound pressure level.

If you are "realizing the full potential" of a loudspeaker with the headroom, you're allowing the speakers to produce more SPL than your average SPL (when the occasion comes along where more is needed). The peaks of music can be 6-10dB higher than the average level, so an optimal sound system has 6-10dB of headroom above the average level. Headroom is the difference between the maximum available performance of a loudspeaker and the desired performance of a loudspeaker. The higher your average level goes, the less headroom you have. If you've got 6dB of headroom with your amplifier, you could push the average level 6dB higher, but you would now be clipping off those peaks, and you would have 0dB of headroom.

[Incognito]

You are saying the same thing I am except you've still got a misconception about how SPL fits into this. Read what you're saying carefully:

What is the definition of "full potential of a given speaker" in acoustical terms? What is a "buffer zone from what your speakers require". How do you define "the dynamics of music"? You'll find that all of these qualitative terms can be quantified by an increase in the sound pressure level.

If you are "realizing the full potential" of a loudspeaker with the headroom, you're allowing the speakers to produce more SPL than your average SPL (when the occasion comes along where more is needed). The peaks of music can be 6-10dB higher than the average level, so an optimal sound system has 6-10dB of headroom above the average level. Headroom is the difference between the maximum available performance of a loudspeaker and the desired performance of a loudspeaker. The higher your average level goes, the less headroom you have. If you've got 6dB of headroom with your amplifier, you could push the average level 6dB higher, but you would now be clipping off those peaks, and you would have 0dB of headroom.



The definition of "Headroom" is SPL - it's the ability of your system to produce a SPL higher than your target (if your target is 95 dB, and you have 10 dB of headroom, that means your system has a max output of 105 dB).

Therefore, going from a 150W amp to a 300W amp would allow for an extra 3dB of headroom.

You're basing the added power that should be considered headroom as useable power for added SPL, the point of headroom is not to make your system louder, it's to make sure the requirements of your speaker doesn't exceed the full potential of your given amp. If you use this extra headroom for added SPL as you suggest then you no longer have any headroom (since you've used it to gain that added SPL). This all comes down to the point of what started this conversation (RMS rating vs Program rating vs peak rating) RMS is the requirement of the speaker while the program rating is the speakers requirement rating plus headroom.

You mention that an optimum sound system would be rated with up to 10db headroom, to achieve this would mean you would have to apply 10 times the power so for example:
the speakers in question in this thread are 150 watts RMS at 8 ohms but to achieve the 10db headroom you mention it would mean you would have to have an amp that can deliver 1,500 watts into 8 ohms. Totally impractical plus exceeds even the peak rating of the given speaker (which is a totally bogus rating anyway).

Here take a look at this chart to understand

http://img823.imageshack.us/img823/4729/image2hn.jpg

The general rule of thumb for a speakers program rating only requires 3db of headroom.

Now if you get into headroom from a mixer/signal strength point of view that is completely different from where headroom for speakers (i.e. program rating) are concerned.

[jayhwk]

You're basing the added power that should be considered headroom as useable power for added SPL, the point of headroom is not to make your system louder, it's to make sure the requirements of your speaker doesn't exceed the full potential of your given amp. If you use this extra headroom for added SPL as you suggest then you no longer have any headroom (since you've used it to gain that added SPL). This all comes down to the point of what started this conversation (RMS rating vs Program rating vs peak rating) RMS is the requirement of the speaker while the program rating is the speakers requirement rating plus headroom. .

Then we are saying the same thing. I'm trying to point out that you can't separate the concept of sound pressure level from headroom. I'm also not suggesting that you should use the headroom for added SPL (in fact, I pointed out that you would have no headroom if you did). What I'm saying is that headroom is simply defined as the difference between desired operating level and maximum possible operating level, and that SPL (not max SPL) is the units of measure for both of those quantities.



You mention that an optimum sound system would be rated with up to 10db headroom, to achieve this would mean you would have to apply 10 times the power so for example:
the speakers in question in this thread are 150 watts RMS at 8 ohms but to achieve the 10db headroom you mention it would mean you would have to have an amp that can deliver 1,500 watts into 8 ohms. Totally impractical plus exceeds even the peak rating of the given speaker (which is a totally bogus rating anyway).

The general rule of thumb for a speakers program rating only requires 3db of headroom.

Now if you get into headroom from a mixer/signal strength point of view that is completely different from where headroom for speakers (i.e. program rating) are concerned.


You're assuming that the 150W amplifier is always putting out 150W. It's not. If you can achieve the desired average level with 50W, then your 6dB peaks would require a 200W amplifier. It's too bad that most devices don't have meters that show both average and peak levels.

I agree that planning for 10dB peaks is not practical. That's not to say that music doesn't have 10 dB peaks. That figure comes from source that I am working with (the previously linked EV document).

[windspeed36]

I've been designing and specing PA systems for over 15 years 10 of which have been with a very large audio supplier. When wo put together a system the method we always use is to look at the capability of the cabinet and assume that yes it will be used to its full potential
What exactly are you classifying as 'full potential' - the way I see it is full potential is the maximum a speaker can be safely driven without clipping or distortion of its drivers. That point is certainly well below the peak power designated on a driver.

There is no way in hell I'd use your recomendation of matching peak power of a cabinet to the output power of an amplifier because you get to a point where a) the driver would simply blow up or b)they don't make 16kw amplifiers to feed JBL V25's based on your method.

[SockoPro]

Full potential should be defined as being driven the full spectrum 20hz - 20 khz (or however you have it crossed) of the musical notes with no clipping off the peaks without distortion....

[Incognito]

You're assuming that the 150W amplifier is always putting out 150W. It's not. If you can achieve the desired average level with 50W, then your 6dB peaks would require a 200W amplifier. It's too bad that most devices don't have meters that show both average and peak levels.


I am not assuming the amp will be delivering 150 watts all the time (you spec an amp to speaker based on the speaker being able to reach full potential as per it's given RMS spec then added headroom is a bonus as well as any driver protection such as limiters), that was never in question (keeping in mind that headroom plays a more significant role in tops as opposed to subs). What was called into question was your questioning providing program power (1.5 to 2 times RMS) to a speaker because of potential damage (which will only stands any chance of happening if you're not using the headroom as headroom but to actually try to gain extra SPL) in your opening statement.


I actually disagree with the "old wisdom" of 2x the RMS power or 1x the peak power for beginners. The difference between 150W and 300W is 3dB. That's the threshold of volume difference that the average audience member will notice. You'll get a barely noticeable volume change, but the potential to damage you speakers will have increased significantly.



What confuses me though is above you mention the potential dangers (if used incorrectly) of having 3db headroom (2 x RMS) but then go on to suggest a properly designed system should have 6db to 10db of headroom (between 4 times to 10 times RMS). Also peak rating is a rating that should NEVER be taken into consideration since it's 100% marketing & none applicable in a real world setting (for the beginners out there)


The peaks of music can be 6-10dB higher than the average level, so an optimal sound system has 6-10dB of headroom above the average level.

You then counter your given statement by saying this in a later post


I agree that planning for 10dB peaks is not practical.

So you can see where I am little confused as to where you're coming from on this topic.

[jayhwk]

Sorry, maybe I could have done a better job of explaining the situation. I should condense down my recommendations:

I recommend that headroom be achieved by specifying a sound system as a whole that is able to produce the targeted SPL + the desired headroom. I would do that by using loudspeakers that produce enough SPL to reach the desired SPL target + headroom powered with amplifiers that provide the RMS rating to 4/3 of the RMS rating. Headroom does not only come from amplifiers - it is a property of a sound system as a whole.

I specify at RMS or 4/3 of RMS because if there is uncontrolled feedback, a dropped mic, or an extra crunchy 1/4" jack in the system, that headroom that under normal operation is used to reproduce the near instantaneous peaks of music harmless to loudspeakers will be turned into SPL and more often than not will damage the loudspeaker.

I just designed a system for a banquet/multipurpose room that required 85 dB SPL throughout the space for voice reproduction only. We decided that 10 dB of headroom was ideal, so we specified loudspeakers that were capable of producing 95dB SPL when powered at the RMS rating.

My post about 10dB being impractical was a bit confusing, I apologize. That statement was mostly target at loud music reproduction systems. It's easy to go from 85 to 95 dB. It's not as easy to go from 107 to 117 dB. People listening to music that loud probably won't notice the difference, though...

DSPs and limiting also confuse the situation - if you use both average and peak limiters (that are properly set), you can use much larger amplifiers and be confident that the system won't be damaged.

Note that I've edited out my peak power comment - I was thinking of EV's definition of "program power", which is always 2x the RMS. Some manufacturers define peak as 4x the RMS, others just make it up.

[Subzeero]

How one powers a speaker depends on the severity of the duty cycle.

Much of today's dance music presents a signal that is very demanding due to it's dynamic range.
Jazz or acoustic ensemble recordings are much less demanding than dance music.
Powering speakers at up to twice the continuous rating is not the ultimate solution to powering a speaker in every application. It only applies to the amplification used in less demanding scenarios.
I have followed this approach for years and I have never blown a speaker. I find that it makes a great deal of sense.

Here is a quote from a JBL White Paper that deals with demanding applications.



What about loudspeaker abuse in normal
operation? Should there be any de-rating of systems to
allow for this?


Yes. JBL defines the following three categories
of loudspeaker application, each requiring an
adjustment of the system's IEC rating:

A. For carefully monitored applications where
peak transient capability must be maintained, a system
should be powered with an amplifier capable of
delivering twice its IEC rating. For example, a studio
monitor system rated at 300 watts can be safely driven
by an amplifier capable of 600 watts output.


Discussion: Careful monitoring is the key here.
High quality music production today demands high peak
factors in the recorded signal. Such peak signals are
normally of such short duration that they hardly stress
the system's components. Thus, the extra 3-dB margin
(times two) of power will result in cleaner overall
operation of the system, with less listening fatigue.


B. For routine application where high
continuous, but non-distorted, output is likely to be
encountered, a system should be powered with an
amplifier capable of delivering the IEC rating of the
system.

Discussion: This case describes the bulk of
sound reinforcement activities. Such systems can often
be inadvertently overdriven, or can go into feedback.
When powered with an amplifier equal to their IEC
rating, the user is guaranteed of safe operation.


C. For musical instrument application, where
distorted (overdriven) output may be a musical
requirement, the system should be powered with an
amplifier capable of delivering only one-half of the IEC
rating for the system.

Discussion: Much rock music is produced at full
output with the amplifier well into clipping, and this is a
matter of musical choice. When an amplifier capable of,
say, 300 watts of undistorted sinewave output is driven
well into clipping, its output power can approach 600
watts! So, Derating the system to one-half its IEC power
will result in safe operation of the loudspeaker.


Here is the complete technical paper from JBL

http://www.jblpro.com/pub/technote/spkpwfaq.pdf