Average vs RMS

[dr.power]

Hi guys,

What is the difference between AVERAGE power and RMS power?
LM3886 is a 68W power amplifier, which I think 68W is the RMS power of it. But why in its datasheet in the first sentences it says:
""The LM3886 is a high-performance audio power amplifier
capable of delivering 68W of continuous average power to a
4Ω load and 38W into 8Ω""?

What is avarage power? what is the difference between it and RMS power?
What is the formula for that?

Thanks

 

[MrAl]

Hi,

If you have a load that draws 10 watts for 50 percent of the time that means its instantaneous power is 10 watts during the time it is on, but because it's only on for 50 percent of the time the average power is only 5 watts.

The average power is the true power delivered to a load over a period of time.

If we knew the peak voltage of a sine wave we could get the RMS value by dividing by the square root of 2. If we knew the peak current of a sine wave we could get the RMS value also by dividing by the square root of 2. Thus we might say that if we knew the peak power we could divide by the square root of 2 to get the RMS power but it's not really right. If we knew the peak power that would mean that we knew the peak voltage and peak amps and multiplying the two we would get the peak power, so to get anything useful we'd have to divide the peak power by the square root of 2 twice, which would mean we would be dividing the peak power by 2.

Lets see, if we had 100 volts rms and 5 amps rms that would be 500va, in phase that would be 500watts and that would also be the average power and that's all we would have to know. We dont need anything called Watts RMS, so they must have made that up to avoid working with any phase angle and didnt want to call it VA (Volt Amps) because it would not make as much sense to audiophiles.
So the only time "Watts RMS" *should* be different than "Watts Average" is when the phase angle is not zero degrees, but in audio i dont think the phase angle is usually a consideration when specifying power.
So there is some justification in defining this new quantity although it's not a true measurement in every sense.

There's another problem that could come up however. That's when the peak power is defined as the maximum power the amp can put out for a short period of time but not sustain that output power forever. Trying to figure out the average power from this kind of specification would not work at all because we'd only know what it can do for a short time period and not for the entire operation time. I'd be very surprised if they defined "Watts RMS" from this starting point, but then again i tend to not believe anything the audio world specifies these days.

I think the term first came into use when they started measuring the output power by measuring the rms voltage and then calculating the power into a given load resistance. Thus the "RMS Watts" would be the same as the average power which is the same as the effective power which is the same as the continuous power.

 

[Bloodninja]

Hi,

For an ac voltage source delivered to a resistive load, the RMS value is also called the effective (average) value which is the equivelent value that is delivered by DC. RMS stands for Root Mean Value, mathematically the RMS value for points m = r1, r2, r3, ... rn is

RMSm = sqrt( (r1^2 + r2^2 + r3^2...rn^2) / n ) where n is the number of points taken.

Electrically, Pave = Irms * Vrms = (Im * Vm / 2) where Im and Vm are peak values.

 

[Endolith]

For people finding this thread through Google:

What is the difference between AVERAGE power and RMS power?

"Average power" is the correct term, and "RMS power" is an incorrect term that should not be used.

With an audio signal, the power going into the speaker varies with the waveform.
For instance, when the voltage passes through 0, the power is also 0.
Likewise, when the voltage reaches maximum, the power is also maximum ("peak power").

So to get a single number for the power, you take the average over time, called "average power". For a sine wave, average power is half of the maximum power. The graph shows the voltage waveform in red and the power waveform in blue:

**broken link removed**

Note that the average voltage is 0, because it goes from positive to negative, while the average power is positive, since power is always positive with a resistive load.

Average power can be calculated from the RMS voltage times the RMS current, which is why confused people say "RMS power":

[latex]P_\mathrm{avg} = V_\mathrm{rms} \cdot I_\mathrm{rms}[/latex]

But "RMS power" actually means the root-mean-square value of the power waveform, which would actually be 22% higher than the average power. This is a meaningless measurement, though. When people say "RMS power", it means they are confused, and are actually talking about "average power". Just don't say "RMS power"; it's wrong.

 

[audioguru]

When audio power is mentioned the word "continuous" should be used because instantaneous power is higher before the power supply voltage sags. Also the frequency and amount of distortion should be stated. The load impedance should also be stated

 

[Endolith]

It could be continuous or momentary, but yes, it should be specified, preferably along with how long it's been running, etc. Agreed otherwise.

 

[nsaspook]

As MrAl said long ago 'RMS' Power is term that describes a set method of testing for continuous "Average power".
https://thecarversite.com/yetanotherforum/uploadfiles/20141120110945135.pdf
**broken link removed**
EIA/CEA-490-A

7.1 Primary Rating and Disclosure For power amplifiers, integrated amplifiers, tuner/receivers (or such section), the primary ratings shall be: a) Power Output (see 5.1) b) Frequency Response (see 5.13.1) c) Input Impedance (see 5.9.2) d) THD (see 5.4) The primary rating should be disclosed in the format indicated in the following examples:
a) Monaural Single Channel:120 Watts RMS at 8 ohms, 1 kHz, and 1% THD
b) Stereo Two channels: 100 Watts RMS per channel at 8 ohms, 1 kHz, and 1% THD
c) Multi--Channel i) Equal Power for All Channels (Example: Five channels) Five channels: 125 watts RMS per channel at 8 ohms, 1 kHz, and 1% THD ii) Unequal Power Among All Channels (Example: Five channels) Five channels: 80 Watts RMS per Front channel (L, C,R), 40 Watts RMS per Rear channel (L,R)— at 8 ohms, 1 kHz, and 1% THD In each instance, the following information should be included with the primary rating disclosure: at 8 ohms, 1 kHz, and 1% THD.

 

[Nigel Goodwin]

"Average power" is the correct term, and "RMS power" is an incorrect term that should not be used.

You are mistaken - and presumably deliberately trying to confuse the issue?.

The exact meaning of 'RMS power' for audio power amplifiers is well documented and understood, and is the ONLY accurate way of comparing amplifiers.

The term 'could' be expanded to it's full meaning, but there's no need or point as the term is well understood.

 

[schmitt trigger]

The very front page of a datasheet are what is called the "marketing" info...essentially a list of bells and whistles designed to entice the reader to continue reading the dense, technical stuff that follows. Which by the way, evaluating that dense info is the proper way to choose a component.

On a class-AB IC audio amp, the actual power that you can get out of it, is related to its internal heat generation and how you can get rid of it. Breaking it down further, it means the supply voltage, the load impedance, and the total thermal resistance.
Pages 14, bottom 17, 18, and top of 19 discuss the many caveats and design considerations to obtain the most power. There are other charts you must consider, for instance the THD vs Output Power, where you can notice that distortion is fairly low until it reaches a power level on which it just brickwalls.

 

[audioguru]

1% distortion at 1kHz is pretty bad now-a-days for a modern power amplifier. An LM3886 IC amplifier with a plus and minus 28V power supply driving a 4 ohm speaker at 1kHz produces 0.004% total harmonic distortion when it begins clipping at 68.5W.

 

[Endolith]

The exact meaning of 'RMS power' for audio power amplifiers is well documented and understood, and is the ONLY accurate way of comparing amplifiers.

Well this thread wouldn't exist if it were well understood. Using a term to mean something different from what it literally means is always going to result in confusion.

RMS voltage is not the same as average voltage. RMS power is not the same as average power.

 

[Nigel Goodwin]

Well this thread wouldn't exist if it were well understood.

The only person who doesn't understand it appears to be you? - and you seem determined to try and confuse things as much as possible.

 

[MikeMl]

The only person who doesn't understand it appears to be you? - and you seem determined to try and confuse things as much as possible.

What Endolith has been saying is (mathematically) correct. Audiophiles are the ones who are confused...

 

[bountyhunter]

When audio power is mentioned the word "continuous" should be used because instantaneous power is higher before the power supply voltage sags. Also the frequency and amount of distortion should be stated. The load impedance should also be stated

Continuous power ratings also used to be a good indicator of how well made the amp was because it had to have heatsinks and power supplies capable of delivering the power continuously. Back in the old days, you could judge an amp's design by grabbing it and lifting it..... because the power supplies were 60 Hz transformer type that took huge transformers and heat sinks. Now most use off line switchers so you have amps with huge power ratings that are small and light.

 

[Nigel Goodwin]

What Endolith has been saying is (mathematically) correct. Audiophiles are the ones who are confused...

Not at all, he's not correctly understanding what the term 'RMS power' means when applied to an audio amplifier.

It's a simple term, easy for anyone to measure and check with simple equipment.

If you want a fuller name for it, then "continuous sinewave RMS power in to a specified impedance, at a specified distortion, at a specified frequency (or better, a specified range of frequencies), all channels driven.

For examples, 50W RMS per channel in to 8 ohms, both channels driven, less than 0.05% distortion, over 20Hz to 20Khz.

 

[MikeMl]

...
If you want a fuller name for it, then "continuous sinewave RMS power in to a specified impedance,.

His point, and my point, is that "RMS" in the above statement is redundant, and actually meaningless. The descriptors "continuous" and "sinewave" have some relevance as to the conditions at the time of the measurement.

Power is always the square of the root-mean-square of the voltage divided by the load resistance. This is true in audio, and everywhere else in engineering...

If the voltage is a sinewave, then it is relatively easy to calculate the power...

 

[nsaspook]

His point, and my point, is that "RMS" in the above statement is redundant, and actually meaningless.

I think we all get the point it's technically redundant (to engineers anyway) and (mis)used in the consumer audio industry but so is the word 'Electricity'. When 'Joe Six-pack' wants a new receiver that 'RMS Watts' value is a EIA/CEA standard rating term with a standard method he can use to compare products. It's not meaningless to him.

 

[fezder]

I too for some reason confuse between average & RMS, but then again I don't need to calculate them much.

 

[MrAl]

Well this thread wouldn't exist if it were well understood. Using a term to mean something different from what it literally means is always going to result in confusion.

RMS voltage is not the same as average voltage. RMS power is not the same as average power.

Well this thread wouldn't exist if it were well understood. Using a term to mean something different from what it literally means is always going to result in confusion.

RMS voltage is not the same as average voltage. RMS power is not the same as average power.

"Average power" is the correct term, and "RMS power" is an incorrect term that should not be used.

Hello,

The way we use language can vary quite a bit, and our language is not a context free language. In a context free language you can parse a sentence without knowing who said it or more importantly what topic was being talked about at the time. Strictly speaking what this means is that we can NEVER be sure of how a phrase renders into meaning until we also know the body of discussion from which it came. And that brings us to that little part of language we call "rendering".
In a literal word by word translation every word is taken to be just what it is and it's meaning may be context free. But in rendering that translation into the meaning that would be only understood by the local people the meaning could very well change.

There are many examples, i'll give a few here...

"log cabin"
Does this mean that a cabin is made from one 'log' ?
Does this mean we are to take the mathematical 'log' of the variable 'cabin' ?

The literal translation would make anyone wonder.

What happened is that we see RMS voltage, RMS current, and then want to apply that same method of translation to the phrase "RMS Power" or "RMS Watts". The latter two terms were NOT established in the same manner as the first two, but following the way language works they are in fact valid. They might be called idioms, which means their meaning can not be deductible from the individual words, at least not in the same way as the first two are.
If we look at it closer, what "RMS voltage" means is the root of the mean of the square of the voltage, but what "RMS Power" means is that it is the power measured using RMS values for voltage and current. So it might be a little confusing and technically probably should not have been called that, but *WE* dont always get to choose what is right in language, it is the originators who decide and if it catches on, then it prevails over time. There's always the chance that it may be changed in the future, but it's so ingrained in the audio industry that there seems to be little chance of that happening.

Another example from your previous post...
In the context of which you speak there is no such thing as "RMS Power" (meaning the RMS value of the peak power waveform), so saying that it is a meaningless measurement is meaningless in itself because there is no such measurement to begin with, yet you had no problem quoting that it was a "meaningless measurement". That's how language works too, we sometimes take short cuts. We probably dont want to have to say, "RMS times RMS Power" which would be more correct mathematically for example, and it would sound funny to say, "logs cabin".

In your first post you quote that "RMS Power" is "wrong", but it's being used by the audio industry and has been used for a number of years now. It's only wrong if you want to try to force a particular usage of language, which is almost impossible even in technical literature.

Isaac Newton tried to invent a new language that was more logical, but it never caught on. Lucky he invented gravity though or we would all be floating in space right now

 

[fezder]

"log cabin"

place where people must log off their computer or where sea captain writes sea logs?