ProSoundWeb Community

Sound Reinforcement - Forums for Live Sound Professionals - Your Displayed Name Must Be Your Real Full Name To Post In The Live Sound Forums => LAB: The Classic Live Audio Board => Topic started by: David Sturzenbecher on July 23, 2014, 09:58:33 am

Title: Peak to RMS Ratio
Post by: David Sturzenbecher on July 23, 2014, 09:58:33 am
I have been creating a spreadsheet of subwoofers I commonly use in designs, and I came to notice that many models (almost all) have a 6dB peak to RMS SPL rating, while having a 12dB peak to RMS power rating.  Can anyone give insight on why there is a difference?
Title: Re: Peak to RMS Ratio
Post by: Corey Scogin on July 23, 2014, 11:26:07 am
I have been creating a spreadsheet of subwoofers I commonly use in designs, and I came to notice that many models (almost all) have a 6dB peak to RMS SPL rating, while having a 12dB peak to RMS power rating.  Can anyone give insight on why there is a difference?

How are you calculating the power ratio?  See the attached image for the correct formula and this page for more details: http://www.sengpielaudio.com/calculator-db.htm
Title: Re: Peak to RMS Ratio
Post by: Hayden J. Nebus on July 23, 2014, 11:30:17 am
I have been creating a spreadsheet of subwoofers I commonly use in designs, and I came to notice that many models (almost all) have a 6dB peak to RMS SPL rating, while having a 12dB peak to RMS power rating.  Can anyone give insight on why there is a difference?

Might you be confusing field quantities, which you should square/ or multiply the log by 20, with energy quantities, for which you should multiply the log by 10?

I've made that mistake myself. 
Title: Re: Peak to RMS Ratio
Post by: Geoff Doane on July 23, 2014, 01:16:40 pm
I have been creating a spreadsheet of subwoofers I commonly use in designs, and I came to notice that many models (almost all) have a 6dB peak to RMS SPL rating, while having a 12dB peak to RMS power rating.  Can anyone give insight on why there is a difference?

Power compression? Although 6 dB seems like a lot.
Title: Re: Peak to RMS Ratio
Post by: David Sturzenbecher on July 23, 2014, 01:23:26 pm
Might you be confusing field quantities, which you should square/ or multiply the log by 20, with energy quantities, for which you should multiply the log by 10?

I've made that mistake myself.

Yup that was it right there, I was using 20 log instead of 10 log. Thanks for the refresher.
Title: Re: Peak to RMS Ratio
Post by: Don Boomer on July 24, 2014, 02:29:18 pm
The 6 dB difference between peak and average would be in line if all you were putting into the speaker was sine waves.  But of course you are likely never doing that unless all you play back is test tones.   

The newest AES speaker power handling spec now calls for 12 dB as live music as an input almost always has more than 6 dB peak to average ratio.
Title: Re: Peak to RMS Ratio
Post by: Hayden J. Nebus on July 24, 2014, 03:59:40 pm
The 6 dB difference between peak and average would be in line if all you were putting into the speaker was sine waves.  But of course you are likely never doing that unless all you play back is test tones.   

The newest AES speaker power handling spec now calls for 12 dB as live music as an input almost always has more than 6 dB peak to average ratio.

A pure sine has a crest factor of 3dB. AES2-1984, which called for pink noise with crest factor 2 (6dB) has been superceded by AES2-2012, which calls for pink noise with crest factor 4 (12dB).

 Some manufacturers may not rate their products to this standard, or to the previous standard. Buyer be informed!
Title: Re: Peak to RMS Ratio
Post by: Art Welter on July 25, 2014, 02:21:33 pm
The newest AES speaker power handling spec now calls for 12 dB as live music as an input almost always has more than 6 dB peak to average ratio.
Don,

Thanks for the heads up, had not noticed the introduction of  AES2-2012.

Seems odd that the 6 dB dynamic range AES2-1984 has been superceded by the 12 dB dynamic range AES2-2012 when the general trend since that time has been progressively more, rather than less, dynamic compression in pop music.

Some EDM even has less than 3 dB dynamic range in the sub 100 Hz range, the  AES2-2012 would grossly inflate the potential of a sub handling that type of music.

Art
Title: Re: Peak to RMS Ratio
Post by: Hayden J. Nebus on July 25, 2014, 03:35:31 pm
Don,

Thanks for the heads up, had not noticed the introduction of  AES2-2012.

Seems odd that the 6 dB dynamic range AES2-1984 has been superceded by the 12 dB dynamic range AES2-2012 when the general trend since that time has been progressively more, rather than less, dynamic compression in pop music.

Some EDM even has less than 3 dB dynamic range in the sub 100 Hz range, the  AES2-2012 would grossly inflate the potential of a sub handling that type of music.

Art
Art,
It's actually a more conservative rating, all other things being equal. I'd assume the intent was to make the noise signal more closely simulate the peak-to-average power found in live signals.

If you're trying to drive those EDM signal peaks all the way up to the AES2-2012 loudspeaker's peak power rating, sure, you're going to have a bad day. You've exceeded rated continuous power handling. But if you drive the RMS level of that same 3dB crest signal up to the loudspeaker's continuous power rating, you actually have more headroom than a transducer rated at the same continuous power under the prior '-1984 rating.

If you drive a AES2-1984 rated transducer and an AES2-2012 rated transducer together with 6dB crest pink noise up to their respective continuous power handling ratings (using RMS metering), the AES2-2012 rated transducer will still have 6dB headroom, which the other transducer will not. If you drive them with 12dB crest pink at that same RMS drive level, the  '-2012 rated transducer will run, while the '-1984 rated transducer will probably crap out sooner than later, from having been driven 6dB beyond its peak power handling. 

Here's to more conservative ratings.

EDIT: revised power handling language  per Dave's correction

 
Title: Re: Peak to RMS Ratio
Post by: Dave Gunnell on July 25, 2014, 05:26:20 pm
...while having a 12dB peak to RMS power rating...
...You've exceeded rated RMS power handling...up to the loudspeaker's RMS power rating...more headroom than a transducer rated at the same RMS power...

...up to their respective RMS power handling ratings (using RMS metering)

There's actually no such thing as 'RMS Power.'  You probably mean 'long term' or 'continuous power', but in a strict sense the term 'RMS Power' has no real meaning.

Bob Lee at QSC authored a white paper that touches on the subject of this misnomer... :).
Title: Re: Peak to RMS Ratio
Post by: Hayden J. Nebus on July 25, 2014, 07:14:51 pm
There's actually no such thing as 'RMS Power.'  You probably mean 'long term' or 'continuous power', but in a strict sense the term 'RMS Power' has no real meaning.

Bob Lee at QSC authored a white paper that touches on the subject of this misnomer... :).

You're absolutely right, thanks!  I meant to refer to the signal's RMS level and the loudspeaker's continuous power rating, respectively .

Title: Re: Peak to RMS Ratio
Post by: Peter Morris on July 25, 2014, 10:22:49 pm
I have been creating a spreadsheet of subwoofers I commonly use in designs, and I came to notice that many models (almost all) have a 6dB peak to RMS SPL rating, while having a 12dB peak to RMS power rating.  Can anyone give insight on why there is a difference?

I very simple terms, the continuous or RMS power or whatever you want to call it, is more or less the thermal capacity of the driver. The peak power is representative of the mechanical strength of the driver. The program power is roughly the recommend maximum RMS input you should use. It allows for the fact that music is not continuous so you don’t have thermal problems.

For example, my subs use 2 of these - http://www.eighteensound.it/PRODUCTS/Products/ProdID/58/CatID/8#.U9MCSmkiPq4
i.e. 3,600 watts continuous, 7,200 watts program, 20,000 watts peak. … or roughly 20 horse power for a pair subs a side!

They can sustain their program rating provided they not pushed below 30 - 32 Hz, at which point they exceed Xmax. Once you exceed Xmax the impedance drops, the cooling provided by the voice coils proximity to the magnet structure drops, and it all goes pear shaped – the smoke that powers the speaker can then escape  :-\.
 
The change from 6 dB to 12 dB crest factor was to reflect real world conditions better; music’s dynamic range is more like 10 – 20 dB,  it’s very rare (not impossible) to find peak to average ratios less than 6dB.
Title: Re: Peak to RMS Ratio
Post by: Ivan Beaver on July 25, 2014, 10:54:43 pm

 
The change from 6 dB to 12 dB crest factor was to reflect real world conditions better; music’s dynamic range is more like 10 – 20 dB,  it’s very rare (not impossible) to find peak to average ratios less than 6dB.
It all depends on what you call "music".

A lot of current popular music (EDM-RAP etc) has a very low crest factor-and at times is nothing but basically a sine wave.

It can be VERY hard on loudspeakers (especially subs).

In my opinion-it would be better to rate speakers using the 6dB rule-because it gives the "average" user a more conservative power rating.

Sure-less "bragging" rights-but what is the number there for anyway?  To me it should be to give the user an idea of what they could reasonably use in terms of a power amp to power them.

Yes many will disagree-but that is my feeling.  I try to take the conservative side when it comes to specs-allowing the average user to actually be able to achieve what is on the spec sheets.
Title: Re: Peak to RMS Ratio
Post by: David Sturzenbecher on July 25, 2014, 11:26:07 pm
Ivan, is there something special about the DBH-218? From my math, which could be wrong, it only shows a 3dB ratio on the power side, but a 6dB ratio on the SPL side


Sent from my iPhone using Tapatalk
Title: Re: Peak to RMS Ratio
Post by: Ivan Beaver on July 25, 2014, 11:40:20 pm
Ivan, is there something special about the DBH-218? From my math, which could be wrong, it only shows a 3dB ratio on the power side, but a 6dB ratio on the SPL side


Sent from my iPhone using Tapatalk
Actually the numbers are correct-but maybe a bit misleading.

The power is stated in continuous and program (NOT peak), while the SPL numbers are given in continuous and peak.

But the truth is this is actually a "typo" (kinda) left over from the days in which we did not state a "peak" power-just continuous and program.

We did not want to play the "peak" game-but since everybody else was giving peak numbers-we decided to add that also.

I will get it changed next week.

Thanks for the heads up.
Title: Re: Peak to RMS Ratio
Post by: Art Welter on July 26, 2014, 02:21:42 pm
For example, my subs use 2 of these - http://www.eighteensound.it/PRODUCTS/Products/ProdID/58/CatID/8#.U9MCSmkiPq4
i.e. 3,600 watts continuous, 7,200 watts program, 20,000 watts peak. … or roughly 20 horse power for a pair subs a side!
The peak rating is only for a duration of 10ms, one cycle at 100 Hz, only 1/2 cycle at 50 Hz ::).
That would be a short set...
Title: Re: Peak to RMS Ratio
Post by: Ivan Beaver on July 26, 2014, 06:38:31 pm
The peak rating is only for a duration of 10ms, one cycle at 100 Hz, only 1/2 cycle at 50 Hz ::).
That would be a short set...
A good example of what happens when you look at the "simple numbers"-without an understanding of where they come from or what they actually mean.

I know one manufacturer who gives power ratings on how much it would take to physically break the driver in a cycle or two.

YEAH-that is how I want to power my rig-NOT.

I believe what the power ratings should REALLY mean would be to give the user an idea of what sort of size amplifier they could reasonably use to drive the speaker without damage.

Of course there are many other "unknowns" that could quickly skew the real world-such as highly compressed music-long extended shows (like 72 hours nonstop-no breaks-no set changes etc)/

Things like that can be REAL abusive on loudspeakers as the heat just continues to build up.

So you HAVE to take the numbers-then apply a bit of "common sense" to what YOU are doing at the moment and go from there.

But of course that takes an understanding and a little bit of knowledge-things that are sorely lacking in our industry ;(
Title: Re: Peak to RMS Ratio
Post by: Peter Morris on July 26, 2014, 09:32:35 pm
The peak rating is only for a duration of 10ms, one cycle at 100 Hz, only 1/2 cycle at 50 Hz ::).
That would be a short set...

Yes but that's only to give an indication of the mechanical strength of the driver - 18sound provides a complete set of specifications - AES power, program power, peak power, thermal compression @ full power, -3db & -10dB, Xmax ... and specify how each are measured .... what else do you want ?????
Title: Re: Peak to RMS Ratio
Post by: Ivan Beaver on July 26, 2014, 09:48:02 pm
Yes but that enough to give an indication of the mechanical strength of the driver - the thermal capacity is the AES continuous, and neither of these consider Xmax ...

18sound provides a complete set of specifications, AES, program, peak, thermal compression @ full power, -3db & -10dB, Xmax and specify how each are measured .... what else do you want ?????
The "average consumer" just wants to know how many watts will it handle.

Don't confuse them with specs they don't understand.  They want a simple WRONG answer-but yet want to "believe" that the simple number tells them everything they want it to.

NOT!!
Title: Re: Peak to RMS Ratio
Post by: Peter Morris on July 26, 2014, 10:54:47 pm
The "average consumer" just wants to know how many watts will it handle.

Don't confuse them with specs they don't understand.  They want a simple WRONG answer-but yet want to "believe" that the simple number tells them everything they want it to.

NOT!!

The average consumer is not buying these drivers from 18sound ;) .. they are buying completed boxes with every bit of spin the marketing department can get away with, and it seems to be getting worse.

The person that is buying these drivers would want all this information, and in some cases more ... which 18sound, B&C etc. can all provide.
Title: Re: Peak to RMS Ratio
Post by: Ivan Beaver on July 27, 2014, 11:10:50 am
The average consumer is not buying these drivers from 18sound ;) .. they are buying completed boxes with every bit of spin the marketing department can get away with, and it seems to be getting worse.


Especially with some new products that don't have any specs-you are just supposed to "believe" that they will do the job.

At least that way the manufacturers don't have to be held to anything.

Here-drink this --------------------
Title: Re: Peak to RMS Ratio
Post by: John Roberts {JR} on July 27, 2014, 11:56:10 am
Especially with some new products that don't have any specs-you are just supposed to "believe" that they will do the job.

At least that way the manufacturers don't have to be held to anything.

Here-drink this --------------------

I have been following this exact issue for decades and it is not getting worse (probably), while there are always new wet behind the ears customers coming along, and new sales people who make faulty claims as much out of their own ignorance as bad intentions. In my decades of experience where I have even had to correct my own salespeople, I never found bad intent, just good intentions and a failure to understand the underlying physics. 

The unfortunate reality is that trying to apply (round) continuous sine wave specs to (square) varying music behavior is like the proverbial square pegs in round holes that don't fit neatly. Since marketers are forced to put those square pegs into round holes there will always be confusion and imperfect correlation to real life from using simple published data specs.

I do not see evil intentions from the majority, with only a minority of industry white knights who speak truth, but many businesses trying to survive by selling product to consumers who are never going to do the homework to fully understand complicated specs. These ignorant consumers reward products with specs presented in ways they "think" they understand.

This difficult hard to win situation is why forums like this are so valuable for the few customers who do want to understand and try to learn from the experts here. Over the decades I have found it impossible to educate every single customers before they make some bad purchase decisions out of ignorance. That is the unfortunate reality and harder for value brands than lower volume premium brands that begin with better educated customers and have more opportunity to explain details to them.

JR   
Title: Re: Peak to RMS Ratio
Post by: Peter Morris on July 27, 2014, 08:14:35 pm
I have been following this exact issue for decades and it is not getting worse (probably), while there are always new wet behind the ears customers coming along, and new sales people who make faulty claims as much out of their own ignorance as bad intentions. In my decades of experience where I have even had to correct my own salespeople, I never found bad intent, just good intentions and a failure to understand the underlying physics. 

The unfortunate reality is that trying to apply (round) continuous sine wave specs to (square) varying music behavior is like the proverbial square pegs in round holes that don't fit neatly. Since marketers are forced to put those square pegs into round holes there will always be confusion and imperfect correlation to real life from using simple published data specs.

I do not see evil intentions from the majority, with only a minority of industry white knights who speak truth, but many businesses trying to survive by selling product to consumers who are never going to do the homework to fully understand complicated specs. These ignorant consumers reward products with specs presented in ways they "think" they understand.

This difficult hard to win situation is why forums like this are so valuable for the few customers who do want to understand and try to learn from the experts here. Over the decades I have found it impossible to educate every single customers before they make some bad purchase decisions out of ignorance. That is the unfortunate reality and harder for value brands than lower volume premium brands that begin with better educated customers and have more opportunity to explain details to them.

JR

“2500 Watt 2 Way 8" Powered Loudspeaker with KLARK TEKNIK DSP Technology, Speaker Modelling and ULTRANET Networking” This is how Turbosound describes what I think is a rebadged Behringer MI speaker, and then there is EV- “Yes 2000 watts is a lot of power” -  part of their description of a 10" two-way powered MI speaker.

I probably don't have to explain…. but if you read the fine print those specification are for the peak output of the amplifier which I suspect is correct, but limited according to which speaker they use it in … as you said, this is probably not from bad intent and I suspect both of these speakers for what they are, are excellent products, but I think what some Marketing departments are doing in 2014 is a little misleading.