Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Jasen Chung .]

Happy Holidays Everyone, 

So I'm a hobbyist being asked to check a sound system for church and please correct me if I'm wrong, but as a rule of thumb shouldn't the amp be double the wattage as the speaker for efficiency? Then I saw the features of the amp and got a bit confused with the options

- I have 2 EAS S250Z that run at  1050W (4 ohm) and 525W (8 Ohms)
- The amp being used is a QSC PL325 that pushes 500W @8 Ohms/ 850W @4 Ohms/ 1250W @2 Ohms.

1) should we be getting a higher powered amp?
2) if budget doesn't permit, what is the best setting for max bass output?

Thanks in advance experts!

[Ivan Beaver]

Happy Holidays Everyone, 

So I'm a hobbyist being asked to check a sound system for church and please correct me if I'm wrong, but as a rule of thumb shouldn't the amp be double the wattage as the speaker for efficiency? Then I saw the features of the amp and got a bit confused with the options

- I have 2 EAS S250Z that run at  1050W (4 ohm) and 525W (8 Ohms)
- The amp being used is a QSC PL325 that pushes 500W @8 Ohms/ 850W @4 Ohms/ 1250W @2 Ohms.

1) should we be getting a higher powered amp?
2) if budget doesn't permit, what is the best setting for max bass output?

Thanks in advance experts!

Efficiency has nothing to do with it.

There is no "standard" rule for amp/speaker power-but the more "common" (whatever that is) standard is having an amp twice the size of the continuous rating of the loudspeaker.

HOWEVER this depends greatly on the actual usage-the dynamic range of the source material and so forth.

In some "heavy" cases (think long term dance club), an amp half the rating or equal to the continuous rating is a good choice-due to the long term heating.

In other cases-it is recommended to have an amp 4 times the continuous rating (so it can pass 6dB peaks).

Of course the REAL answer is to have an amp of the proper size to provide the needed SPL for a particular application.  In some cases this might be only 10% of the rated power.

Believe it or not-but a sound system DOES NOT ALWAYS need to run at max output. 

[duane massey]

You need an amp that provides enough power to deliver enough spl for your needs. That could be 10 watts, or it could be 2,000 watts. Since you are using one amp, I assume you are driving each speaker with one channel, so you are driving a 4ohm load. If you are not clipping the amp, and it is loud enough, you are good. If you are clipping the amp and it is not loud enough, you need more power. Ivan said it better in technical terms, but my response is "It depends"....

[Luke Geis]

There is a general rule of 2-4 times the continuous/rms rating of the speaker. If you think of it like this, you will see why it makes sense. A speaker who's rms ( or continuous )  power is 500 watts, can in essence be powered by an amp that will provide 500 watts. However you don't want to power it with an amp who's peak power is 500 watts. That means that the amp would be working pretty hard all the time...... Another general rule for amp power is to halve it's peak power to arrive at it's rms/continuous power rating. That means an amp that is rated for 1000 watts peak should be able to push 500 watts pretty easily all day every day. But wait there is more..........

 Lets suppose we have pretty decent idea of what the MAX SPL the speaker can produce? Lets say for giggles that it's 129db calculated ( just a rough round number for a 500 watt rms  speaker ). We take into account that the speaker sensitivity is 96db at 1W/1M. That means to acquire the theoretical 129db SPL we would need an amp capable of pushing 2,000 watts!!!!! That is right at the probable peak power of that speaker. We don't want to be working a speaker that hard. But we probably don't need 129db SPL to acquire our needed SPL level at listening position. Lets say that we only need 105 at 50' from the Speakers. My calculations show that an amp capable of providing 1,200 watts should do the job. But we need some lee way in the amp still. In theory we should be able to get away with an amp that is pushing out roughly 1,600 watts to get the desired SPL at FOH. The amp is still fairly high in power for the speaker and quite likely will only just do the job without destroying the speakers. Here is why.......

Diminishing returns...... Power compression and the cables natural tendency to absorb power from the amp will reduce the actual performance of the speaker. At some point you can throw all the power at the speaker you want and it won't increase in volume, but rather it will melt down and quickly quit. Most of the performance numbers of a speaker are theory ( although some are fairly truthful ). I always assume that the speaker will provide close to the stated performance, being realistic to the true needs. A speaker with a 136db MAX SPL will probably only actually produce around 133db if powered with an amp half it's peak rated power. I don't feel like pushing my amp into clip so another 3db loss in volume from there is expected. So a realistic SPL for a speaker rated at 136db peak is roughly 130db in actuality!!!!! That means I need to be able to acquire the volume I want with that speaker provided I don't go over 130db at 1m from the speaker. In the scenario from the above paragraph the speaker had a calculated max SPL of 129db. It will probably only actually produce 123db at 1m if you run it in the safe zone. In theory you should be able to get about 100db at 50' from the speakers. With an amp pushing 1,600 watts you should never be in worry of running out of headroom and should give you enough power to reach the 105db at FOH if desired. Keep in mind this is working the speaker pretty hard as you will be running the speaker well into it's likely " PROGRAM " power rating which would be right around 1,000 watts.

But you get a little more back when you have more than 1!!!!!!! so all these numbers were factoring only 1 speaker. Another cool thing about audio is that when you double the power you in theory gain 3db in perceived volume! When a speaker is in a free space you loose volume at 6db per doubling of distance ( the classic inverse square law ). Indoors that changes a little since the sound has less space to disperse to. That means you will likely have more level at FOH than the calculated numbers, how much more is the question? So now you have just gained back some of that 6db that you lost from the calculated max SPL. and running the amp in the safe zone should yield you the desired SPL level without fear of running out of headroom!

To put things in perspective for your specific situations and speaker choice, we don't really have enough info. Is the rating peak power, or rms? What do the specs say for Peak SPL? Are they separated or placed together? Do you mean that each speaker is rated at 8ohm's and that the 4 ohm rating is with the two paired together? My guess is that the speakers are really 4ohm boxes and that the number on the back is the peak power at that ohm rating? I would plan on providing that 850 watts as a safe bet. The real question is if you are getting the SPL needed without clipping the amp? If the answer is yes then this amp should be fine. If not then you have some research to do. Hopefully this information helps.

I factor speakers and amps like this. I plan on only running up 75% of the amps available power and the rest is safety cushion. If my speaker is rated for 1,600 watts continuous and 6,400 watts peak, I plan on powering it with 1.5 times the continuous as the usable range and the last .5 as being worst case scenario. So I want to have 2,400 watts going to that speaker at any given point in time with the last 800 watts as spare energy. An amp with a 3200 watt peak running at 75% should produce 2,400 watts. I'm lucky that I have several options around that power range!!!! This is of course the low down and dirty way to conclude an amp to speaker power ratio. It goes much deeper than that. And this is only the accepted low end of determining power needs. If you factor that an amp will only actually produce half it's peak power, you would want an amp that is 4 times the continuous rating of a speaker. This should yield a true ( or closer to it ) 3200 watts of available power.

This link described what I will generalize: http://www.prosoundweb.com/article/print/power_lines_getting_the_proper_flow

The 10db crest factor is a pretty good practice. It basically prescribes that most amps are rated for a sine wave as being the peak power creator. What this means is that an amp rated for 1000 watts will actually only produce about 100 watts of power factoring the 10db crest scenario. So in order to power a speaker with 1,600 watts of power using the 10db crest factor you would need and amp capable of producing power that exceeds the speakers peak power rating! Diminishing returns much....... It's not uncommon for Larger sound companies with deeper pockets to power their speakers with amps that are higher than the peak rating of one speaker. This power is of course often distributed into several speakers which can handle all of that power. There is a higher level of responsibility that comes with that too. You can't fool around with that kind of expense......

Another good link to assessing power needs: http://www.crownaudio.com/how_much_power.htm

[g'bye, Dick Rees]

 Lets suppose we have pretty decent idea of what the MAX SPL the speaker can produce? Lets say for giggles that it's 129db calculated ( just a rough round number for a 500 watt rms  speaker ). We take into account that the speaker sensitivity is 96db at 1W/1M. That means to acquire the theoretical 129db SPL we would need an amp capable of pushing 2,000 watts!!!!! That is right at the probable peak power of that speaker. We don't want to be working a speaker that hard. But we probably don't need 129db SPL to acquire our needed SPL level at listening position. Lets say that we only need 105 at 50' from the Speakers. My calculations show that an amp capable of providing 1,200 watts should do the job. But we need some lee way in the amp still. In theory we should be able to get away with an amp that is pushing out roughly 1,600 watts to get the desired SPL at FOH. The amp is still fairly high in power for the speaker and quite likely will only just do the job without destroying the speakers. Here is why.......

Oh boy are you in for it now....... 

[Ivan Beaver]

Let's look a little bit closer at where the loudspeaker "ratings" come from.

The wattage numbers are those that a manufacturer "assigns" to a loudspeaker.  It comes from a noise source that has a 6dB crest factor-and is basically the rating that the loudspeaker can "handle" without damage.

NOT how good it sounds-just that it is not damaged.  So the USABLE/LISTENABLE wattage may be a bit lower.  And may not be the same for all freq-as the noise has a shaped response in an attempt to simulate "music"-whatever that might be.  Techno is different than jazz which is different than rock etc.  So which one is "right"? Hence the reason for so many different "standards".

The nice thing about standards is that there are so many to choose from.

6dB (for those that still don't know is a power rating of 4 times the average or RMS level.  YES YES I understand that average and RMS are NOT the same thing when talking about the points on a since wave signal-but I am simply trying to keep it "simple" for the discussion.

NOW just to dig a little bit deeper.  The wattage numbers presented ARE NOT the actual wattage being dissapated by the loudspeaker.  They are a result of the VOLTAGE applied and a CALCULATED wattage.

Once the loudspeaker starts to go into power compression (typically 1/4 to 1/2 the rated "continuous or RMS (wrongly termed) wattage". 

Since the impedance increases during power compression-the actual "wattage" is less than what is expected.

The "wattage" ratings are NOT wattage ratings, but rather ideas of the size of amplifier that might can be used.

Yes there are guidlines-and there are exceptions.  It is important to understand all the "little things" that go into proper amp size choice.

As usual-there is not simple answer.

[Ivan Beaver]

Lets suppose we have pretty decent idea of what the MAX SPL the speaker can produce? Lets say for giggles that it's 129db calculated ( just a rough round number for a 500 watt rms  speaker ). We take into account that the speaker sensitivity is 96db at 1W/1M. That means to acquire the theoretical 129db SPL we would need an amp capable of pushing 2,000 watts!!!!! That is right at the probable peak power of that speaker. We don't want to be working a speaker that hard. But we probably don't need 129db SPL to acquire our needed SPL level at listening position. Lets say that we only need 105 at 50' from the Speakers. My calculations show that an amp capable of providing 1,200 watts should do the job. But we need some lee way in the amp still. In theory we should be able to get away with an amp that is pushing out roughly 1,600 watts to get the desired SPL at FOH. The amp is still fairly high in power for the speaker and quite likely will only just do the job without destroying the speakers. Here is why.......

Oh boy are you in for it now.......

So I will 'take the bait"--------------------------------
105dB.  WHAT EXACTLY DOES THAT MEAN??

A level as measured by an SPL meter?  What weighting?  What time response?  If that is A weighted-then the C weighted response could easily be 15-120Db louder than that.  So now we have to do our calculations based on 120dB or more.  That starts to change EVERYTHING.

If that is an "average or slow" response-then we need to add ANOTHER 6-10dB-so now we could be at 130dB needed from the subs.

So the calculations arrived at for 105dB would be VERY LACKING if the actual need SPL is closer to 130dB.

As usual-it depends-and before stating a "simple number" it is VERY IMPORTANT to understand that that simple number actually means.

[Nicolas Poisson]

The most commonly found wattage for speakers now is “AES”, which is determined using pink noise with a 6 dB crest factor. A good thing is that this standard, at least, tends to become standard. Max SPL is not.

Basically a 500W AES speaker powered with a 500W RMS amplifier will be able to be used at its maximum (if required) for any signal with 6dB crest factor. Both speaker and amp reach their limits at the same point.

What if the crest factor is lower? The amplifier will be able to put out more mean power than the speaker can stand. If the amp is ran right below its limit, there is a strong risk of coil burn.

What if the crest factor is higher? Then the amplifier will start clipping before the speaker has reached its maximum capacity. This is not a problem in itself, as long as you do not run into clip. This is not a problem at all if the level produced is sufficient for the application. Indeed, this is a common practice for home stereo. However if you need more level, the amp cannot deliver while the speaker could stand it.

If you want the speaker and the amp to match, i.e. to reach their limit at the same point, you need to know the crest factor.
For typical live sound the crest factor is around 12dB (it can be much greater if you do not compress anything, up to 20-24dB). Let’s take a 500W AES speaker powered with a 1000W amp:
- the AES power already includes a 6dB crest factor
- the amp can deliver 3dB more wattage, which means 6dB more in level
This means the system is perfectly selected to reproduce the 6+6 = 12dB crests of live music. If you want to match speakers and amp considering an 18dB crest factor, you need four times the speaker AES power.

As a simplified rule, one often consider:
- Dance music, DJ, etc: amp power = speaker AES (low crest factor)
- Live music: amp power = 2x speaker AES (12dB crest factor)

Please note that nothing in this rule tells you that you will have “enough rig for the gig”. It just tells both amp and speaker will reach their limit at the same time, not that this limit is adequate for the application.

[Ivan Beaver]

The most commonly found wattage for speakers now is “AES”, which is determined using pink noise with a 6 dB crest factor. A good thing is that this standard, at least, tends to become standard. Max SPL is not.

Basically a 500W AES speaker powered with a 500W RMS amplifier will be able to be used at its maximum (if required) for any signal with 6dB crest factor. Both speaker and amp reach their limits at the same point.

Just for some clarification.

The AES signal is NOT a normal pink noise signal.  It is shaped pink noise-with both the HF and LF tapered off.

The statements about a 500 watt speaker and a 500 watt amp is NOT correct.  Because of the 6dB crest factor- when the continuous (or misused RMS) is 500 watts, the amp MUST be able to pass 2000 watt peaks-or else the signal would be compressed/clipped so the full swing of the applied signal would not be able to be applied to the loudspeaker.

If you use a 500 watt amp and put 6dB pink noise (shaped or not), the the highest continuous (AES or other) signal would be be 125 watts.

[Ivan Beaver]

- the AES power already includes a 6dB crest factor
- the amp can deliver 3dB more wattage, which means 6dB more in level

Just to address these two points.

To be honest-I have not read up on all the details of the AES test.  HOWEVER-if it already included the 6dB crest factor in it-then how come so many loudspeaker manufacturers who state the AES "wattage" ALSO inlcude a 6dB peak wattage-ABOVE the AES rating?  That would indicate 12dB above the continuous rating.

I am pretty sure the AES rating is arrived at by using the "average" voltage applied-squared-then divided by the rated impedance of the loudspeaker.  This would put the peaks 6dB higher than the continuous rating.

How do you figure that an amp can deliver 3dB more in wattage than its rating?  Yes some amps can provide a little bit more on peaks-but 3dB?  That is a whole bunch of extra "wattage" that the marketing department does not know about.  I bet they would find a way to put that into the spec sheets if that was the case.

Do you have a specific example of an amp that can do this?  Actually deliver 3dB more on peaks than its RATED output.  I am not aware of any-but I have don't know every amp out there.

I am really confused by how a 3dB increase in wattage is a 6dB in level.  Would you care to explain that a little bit more?  I have never heard of this.  A dB difference is a dB difference-whether it is in wattage-or voltage or VU units.  How they are arrived at is different-but the result is the same.

Doubling wattage gives 3dB increase.  Doubling voltage gives a 6dB increase.

[Nicolas Poisson]

Oh... I guess I should make sure I am awaken before posting anything.

Concerning the 3dB more wattage, this is an error. Of course with 3dB more watts you get 3dB more voltage, because the "10" factor for power vs. "20" for amplitudes is made on purpose. Sorry.

Concerning AES, I always thought the power indicated was the "peak" power of the signal, i.e. peak voltage multiplied by impedance. So I was thinking a "500W AES" speaker was powered by an average 125W during testing. And I used to consider roughly 500W AES = 250W sine = 125W square. I even never thought it could be otherwise indeed... I'll sleep smarter tonight.

[Luke Geis]

Oh boy....... Conversation : )

I created a hypothetical situation for no other reason than to provide a way of thinking for the OP. The numbers are farce and the calculations are general. This is why I presented two links in which both contain good material about the subject.

As far as the power ratings of the speakers, it doesn't really matter. All we can go by is what we are told. How that number was derived should be by a measured standard. If it's true or not is the big question. But we make are decisions based on that info. I myself am in the belief that the numbers are untruthful, but close enough to make a decision. I take my NAG calculations and build my system around that factoring a 25-30% safety cushion.

As for what SPL really is? A, or C weighting? Most is done in A weighting, but C weighting is more suited for determining what the system is putting out volume wise over a larger frequency range. If the Rider for your act says 120db to front of the venue, does it say A, or C weighting?  My guess is most don't specify, the ones I have seen didn't say either way, only that they wanted X volume at X position in the venue. If I need 105db at FOH position I use C weighting. When the volume police come around the measurement is done in A weighting. It's up to the designer/operator to determine what the numbers mean and what can be achieved. I think I gave good safe information that should direct someone to making an educated decision. At least I hope so.........

[Luke Geis]

here is another link that has more information on the topic: http://www.prosoundweb.com/article/understanding_target_spl_and_required_amplifier_size/

[Brad Weber]

If you think of it like this, you will see why it makes sense. A speaker who's rms ( or continuous )  power is 500 watts, can in essence be powered by an amp that will provide 500 watts. However you don't want to power it with an amp who's peak power is 500 watts. That means that the amp would be working pretty hard all the time......

Most music is not a constant signal, the rated power of an amplifier does not indicate that it always outputs that much power (actually voltage and current) and the rated power of a speaker does not define a required power.  A 500W Continuous rated speaker may work just fine with an amp rated at 5,000W, 1,000W, 500W, 100W, 1W or any other value, it all depends on the signal, the application and the operation.
 
Where the issue seems to come up with many is not having a defined use and/or having to serve a range of uses, musical genres, etc.  In that case there are some 'rules of thumb' that have been developed over the years to represent what is typically a reasonable mix of maximum output, reliability and headroom.
 

So I'm a hobbyist being asked to check a sound system for church and please correct me if I'm wrong, but as a rule of thumb shouldn't the amp be double the wattage as the speaker for efficiency? Then I saw the features of the amp and got a bit confused with the options

- I have 2 EAS S250Z that run at  1050W (4 ohm) and 525W (8 Ohms)
- The amp being used is a QSC PL325 that pushes 500W @8 Ohms/ 850W @4 Ohms/ 1250W @2 Ohms.

1) should we be getting a higher powered amp?
2) if budget doesn't permit, what is the best setting for max bass output?

As others have already noted, what really matters in that type of situation is not how the amp and speaker ratings compare but whether they amp and speaker combination can provide the desired results without damaging either one.  That pretty much answers your first question, if it currently gets loud enough without clipping the amp then you probably don't need more power.
 
I can't find any information on that speaker but based on the specification you noted I would guess that it is a dual driver model that can be run as two separate drivers (two 8 Ohm drivers rated at 525W each) or with both drivers in parallel (one 4 Ohm, 1050W rated box).  If that is true then assuming you would be running the box or both drivers off one amp channel it makes no difference.
 
What we don't know is whether the 525W and 1050W ratings represent the Continuous/RMS, Program or Peak rating.  The safest assumption is to assume that is the peak rating, thus continuous ratings of about 130W and 260W respectively.  In that case your amp which is rated at 500W into 8 Ohms and 850W into 4 Ohms at 1kHz is already probably about as much power as you would want to apply to those speakers.
 
Even if the 525W and 1050W were the continuous ratings for the speaker then you probably wouldn't want to go beyond an amp that might increase the maximum output (not the output with the same input signal but rather how loud it can get) by about 3dB.
 
Bottom line, unless you can verify that the speakers could handle the additional power and a 3dB or so increase in maximum potential output justifies the cost, I would not change amps.

[Steve Anderson]

I can't find any information on that speaker [...]

I'm guessing he's referring to an EAW SB250z

[Don Boomer]

You have to understand the difference between the potential voltage a power amp could deliver and the the amount you actually "deliver" to the speaker.  Theoretically as long as you don't exceed the peak power or the continuous rating you should have a reasonable chance of not destroying the driver. But this leaves out the possibility especially on a sub of exceeding the mechanical limits of the  box by sending a frequency that is too low.  Again theoretically if you apply a frequency an octave below box cutoff at 1/4 of the rated power you could be in trouble.

In my experience many more drivers are trashed by exceeding the mechanical limit than being burned by too much power by an order of magnitude or two.  So setting the proper HP filter is much more important in my mind than worrying about too much average power.

Back to thermal issues... As others have pointed out this is heavily influenced by the dynamic range of your music. If you have 10 dB of dynamic range then an amp rated to deliver 1000w will be putting out 100w at the time the limit LEDs fire (or there about).  This is why manufacturers recommend "program power" as a good bet. Music almost never drives a power amp to deliver its full average continuous potential.  Even if you had a highly compressed baseline such as in electronic dance music with only 3dB of dynamic range the amp would likely only be outputting less than 500w on a continuous basis.

The other thing you have to be aware of is the method used to get the power rating of a speaker.  AES and EIA are the typical standards used.  Both methods used shaped pink noise.  The AES method however stresses the lower frequencies much more than the EIA method which concentrates most of the energy in the curve around 300Hz.  The EIA method however does it for a longer time period ... 8 hours as opposed to the 2 hour AES method.  Opinions with people who know much more about it than I do go both ways.  To me it would seem that AES would be a better way especially when considering subs.

[ProSoundWeb Community]