Power Of Audio Power Amps?

Thread Starter

forumer

Joined Oct 24, 2014
17
when we talk about the output power of an audio amp, say 20W, what power are we talking about?

i see 3 kinds of numbers in LTspice:-

1) by pointing the cursor at a net or device, i can get DC voltage and current readings, and therefore a power reading.

2) when pointing the finger cursor at a device, i can also get a "power dissipation" reading, is this only a DC power reading?

3) i can also get LTspice to display AC voltage and current waves of the output node, measure their peak values, and get an AC peak to peak power reading.

which of these is taken as the output power of an audio power amp per our convention, or at least the base for such a rating?

i'm not able to verify the LTspice simulated result on power, or anything, by measuring the actual built circuit, i only have a multimeter. this multimeter can measure down to 2mA, but manual says there needs to be 2V present, otherwise it couldn't measure anything. so i'm getting nothing, zero's only. how do you get 2V between two points for current measuring, i have no idea what that means. so do i have any hope here, measuring the actual output power, or current, of an audio amp circuit with just a multimeter?

thank you for all comments and helps!
 

Alec_t

Joined Sep 17, 2013
14,263
Don't believe advertised 'power levels' for audio amps. They're measured in special 'music Watts', not real Watts :).
Post the schematic and/or .asc file of your amp and perhaps we can help.
 

crutschow

Joined Mar 14, 2008
34,201
The amp power is the power delivered (dissipated) into the rated load (typically 8 ohms for an audio amplifier). For a good amplifier that is in continuous RMS watts. The total power required from the power supply(s) is always more than that.

The "power dissipation" measurement in LTspice plots the instantaneous power dissipated by the device. You can then CNTL-Left Click on the plot label for the power trace, which will give the average (RMS) power integrated over the wave-form display time. To do this power measurement accurately for a AC signal it should be done over an integral number of complete cycles.
 

MikeML

Joined Oct 2, 2009
5,444
Zapper beat my to it. Look at this simple LTSpice sim of a complex signal (like audio):

218.gif

Note the rms Voltage across R1. Power is ((Erms)^2)/R1 = 7.887^2/10 = 6.22W

Now, using the Alt-Left click thermometer cursor in LTSpice to plot instantaneous power dissipated in R1 gives this trace, and letting LTSpice integrate that plot to find average power gives:

218a.gif

Only caveat is that the integration interval should include full cycles of the input waveform...
 

rogs

Joined Aug 28, 2009
279
I found that Bertus' link redirects somewhere horrible?......anyone else found that?....

I'm sure it probably says in Rod Elliot's article anyway, but there's no such thing as an 'RMS Watt' ...or a 'peak power Watt' or indeed any other kind of 'qualified' Watt.

A Watt is a unit of work. That's it. No 'variables'.

You can have RMS Voltage... or RMS amps... but not RMS Watts
 

Thread Starter

forumer

Joined Oct 24, 2014
17
The amp power is the power delivered (dissipated) into the rated load (typically 8 ohms for an audio amplifier). For a good amplifier that is in continuous RMS watts. The total power required from the power supply(s) is always more than that.

The "power dissipation" measurement in LTspice plots the instantaneous power dissipated by the device. You can then CNTL-Left Click on the plot label for the power trace, which will give the average (RMS) power integrated over the wave-form display time. To do this power measurement accurately for a AC signal it should be done over an integral number of complete cycles.
hey, i've found the right forum for these stuff!

alright, all that makes sense to amp power rating i guess.

a few follow up questions if you don't mind crutschow,
so "CNTL-Left Click on the plot labe" give an AC power reading, the average AC power reading to be exact.

how do people usually select a bjt, for example, for a certain so called output power?
do we have to include DC power the bjt would be dissipating as well?
for the AC part, do we usually use the peak to peak value, or RMS, or average, or integrated average?
 

Thread Starter

forumer

Joined Oct 24, 2014
17
Zapper beat my to it. Look at this simple LTSpice sim of a complex signal (like audio):

View attachment 74644

Note the rms Voltage across R1. Power is ((Erms)^2)/R1 = 7.887^2/10 = 6.22W

Now, using the Alt-Left click thermometer cursor in LTSpice to plot instantaneous power dissipated in R1 gives this trace, and letting LTSpice integrate that plot to find average power gives:

View attachment 74645

Only caveat is that the integration interval should include full cycles of the input waveform...
thanks a lot Mike for your explanations and good clear illustrations as well!
i tried these readings in LTspice, they worked just as you describe, thanks, i didn't know that before!
i have a few follow up questions on which kind of power number is to be used when selecting a transistor for a certain output power. these questions are included in my reply to crutschow's post above.
 

Thread Starter

forumer

Joined Oct 24, 2014
17
I found that Bertus' link redirects somewhere horrible?......anyone else found that?....

I'm sure it probably says in Rod Elliot's article anyway, but there's no such thing as an 'RMS Watt' ...or a 'peak power Watt' or indeed any other kind of 'qualified' Watt.

A Watt is a unit of work. That's it. No 'variables'.

You can have RMS Voltage... or RMS amps... but not RMS Watts
i'm seeing 404 rogs.

wow, very rigorous thinking, didn't realize that, but it makes perfect sense!

so when we integrate power over time, we get energy?
so the unit of RMS Power is actually energy?
but then we say we did that over a period of time, i.e. per some time interval, so it becomes power again, and we use that as power rating of our amps?
but what period of time? one cycle of a 1KHZ sine wave, (for audio realted stuff)? what amplitude and what current then?

...da more questions you ask, da more confused you become...
 

Lestraveled

Joined May 19, 2014
1,946
.......A Watt is a unit of work. That's it. No 'variables'.........

You can have RMS Voltage... or RMS amps... but not RMS Watts
rogs, You are incorrect. A watt is a unit of power, not a unit of work. A watt second (or joule) is a unit of energy.

Also, you most certainly can have RMS watts, peak watts and average watts.
 

Jony130

Joined Feb 17, 2009
5,487
r
Also, you most certainly can have RMS watts, peak watts and average watts.
Term "RMS" refers only to effective value of a AC voltage and current.
And we should not use the term "RMS" to description the real (true) power.
The true power is simple average value of the instantaneous power.
And for this we don't need roots and the square.
So RMS power is inaccurate term, that have no logical justification.
 

Lestraveled

Joined May 19, 2014
1,946
Term "RMS" refers only to effective value of a AC voltage and current.
And we should not use the term "RMS" to description the real (true) power.
The true power is simple average value of the instantaneous power.
And for this we don't need roots and the square.
So RMS power is inaccurate term, that have no logical justification.
RMS is an industry standard for evaluating a waveform of changing voltage and current. It appears that the Federal Trade Commission felt that RMS was too complex of a term when applied to audio amplifiers. They wanted a simpler term to compare output power, so they decided on average power based on a single sine wave. In other words, the FTC dumbed down the specs so some amplifier salesmen wouldn't make their amplifiers appear more powerful by using RMS instead of Average power.
 

Alec_t

Joined Sep 17, 2013
14,263
In other words, the FTC dumbed down the specs so some amplifier salesmen wouldn't make their amplifiers appear more powerful by using RMS instead of Average power.
But since most amps are probably sourced from the Far East, where they may not have even heard of the FTC, RMS Power or Music Power is still the dominant measure :).
 

rogs

Joined Aug 28, 2009
279
rogs, You are incorrect. A watt is a unit of power, not a unit of work. A watt second (or joule) is a unit of energy.

Also, you most certainly can have RMS watts, peak watts and average watts.
My bad... it's a 'Watt Hour' that could be described as a unit of 'work'. As you have rightly corrected , the Watt itself is a unit of power.

There's still no such thing as an 'RMS Watt' though....or a 'peak Watt' or an 'average Watt'. All in common usage . All wrong.
 

Lestraveled

Joined May 19, 2014
1,946
.......There's still no such thing as an 'RMS Watt' though....or a 'peak Watt' or an 'average Watt'. All in common usage . All wrong......
From Wikipedia, "Continuous average power ratings are a staple of performance specifications for audio amplifiers and, sometimes, loudspeakers.", which means, average wattage.

Here is an interesting youtube video on RMS , average and peak power by AltecLansing.

rogs, if you are going to be in the business, you need to understand all of it.
 
Last edited:
Top