amplifier power calculation

[Colas LeGrippa]

Hi, there must be a way to calculate the power of an amplifier, by injecting a sine wave at the input and, with volumes full up and a load connected, measuring the voltage at the input and output. There is surely a  relationship between the frequency of the sine wave, the voltage at the input, output, and the load, I guess.  By applying good math formulas, one could measure most accurately his amplifier power, and the ideal load it is ''happy'' with, what do you think ?

Colas

[kagliostro]

Not an expert about that

but I've read someone that told he used a square wave signal and measured the voltage at the speaker connection

here a very simple audio wattmeter

http://electronics-diy.presstoday.net/2010/12/power-meter-schematic-for-audio.html

this one may be is a bit too much for our purpose

http://www.chambonino.com/construct/const3.html

Kagliostro

[Iannone]

Sure there is.  Measure the AC voltage at the speaker jack and calculate:

P=E2/Z

Where E2 is the AC voltage squared and Z is the impedance of the speaker load.  (don't make the measurement on an unloaded transformer).  P is the power in watts.  This, of course, will measure peak power.  To get RMS power you will need an oscilloscope and advance the volume to the point where you can first identify output waveform clipping.  Measure the output AC voltage at that point and use the same equation.

[Colas LeGrippa]

very interesting indeed

best regards

 Colas

[PRR]

> with volumes full up...  relationship between the frequency of the sine wave, the voltage at the input

You do not have to run "full up", you do not have to know the input voltage or the frequency.

What Iannone said. Measure the voltage. square it, divide by the load you are using.

It is customary to use a Sine, and for the output to be "sine-like". Tube amps always bend the wave, and for gitar it is reasonable to turn-up until the wave is clearly bent (by eye on 'scope or by by ear on a monitor speaker).

Audio amps are not normally rated on Square wave. In most audio this is very unrealistic. We never have a full power sine, and never-never have a full-power square which will sag the power supply worse than the full sine. In over-driven guitar, maybe it IS reasonable. However many meters are calibrated for sine-waves and give different numbers for other wave shapes.

You can vary the frequency. For a first-test use 400Hz, where guitar has a lot of power and where most guitar amps do best. Then check 100Hz to see if bass is strong or weak (either way is acceptable for different guitar styles). Then check 1,600 and 6KHz to be sure treble can be strong, but it is very acceptable for treble power to be half of midband power.

You can vary the load impedance, re-compute power, and see if other loading gives more power or prettier distortion.

Push-pull amps driven past flat-top sine but not to fully-square waves should be symmetrical. (Some P-P amps go assymetrical when over-over-driven, that's OK.) SE amps never make perfectly symmetric over-drive, but strong assymetry is a sign that you should change your idle current or load.

Check your meter against your signal generator. I have seen DMMs which fall-off badly above 400Hz.

> a very simple audio wattmeter

Too simple. 3db steps. You may know it is more than 25W but less than 50W; we usually want a more specific number. Anyway, we have meters.

[Colas LeGrippa]

ok I am ready to measure the voltage at the speaker jack but at the input ? what kinda signal do I have to inject in ? 2khz ? 10mV ? no importance ?

[Colas LeGrippa]

I see, PRR , thanks for taking the time of answering so completely. ( I'll think about it all,with my scientific calculator and get back to you in......one month or so hey hey ! )
   
regards

Colas

P.S: when my neighbour calls me once, it is usually a 50 watter fully open. When the cops arrive within 10 minutes, it is a 100 watter. Easy to calculate, no ?

[stratele52]

I use 35 millivolts AC at 500 t 1000 hz and mesure AC at speaker wire . Or I play guitar and mesure AC

[jjasilli]

ok I am ready to measure the voltage at the speaker jack but at the input ? what kinda signal do I have to inject in ? 2khz ? 10mV ? no importance ?

Seems the Tone Lizzard site is finally down for good.  Too, bad it had a great tutorial on this.  Anyway:
 
1000 Hz is the audio industry standard.  Output power will vary with frequency (hence the term impedance rather than resistance). However, 1000 Hz doesn't tell us enough about guitar tone.  Hotblue has recommended testing at 400 Hz.

Select a frequency on your tone generator, and run it into the amp.  Dime vol & tone controls on the amp.  Run a sine wave > amp > to dummy load.  If you have a scope, put the scope leads across the dummy load.  Increase input voltage until the waveform distorts.  This is the max power for clean tone at that frequency.  Measure voltage at input for a reference.  Note: guitar PU's usually put out about 200mV for single coil; maybe 800mV or so for humbuckers.

Measure voltage across the dummy load.  Use the power formula to compute power in Watts:  Watts = Voltage² / Speaker impedance.  You can use RMS (meter) or peak power (scope).  

If you don't have a scope, run the amp's output to a speaker.  Increase input voltage and listen to tone until it starts sounding over-driven.  Measure voltage across the speaker terminals, and use the power formula.  This is the max power for clean tone at that frequency.  

[birt]

ok I am ready to measure the voltage at the speaker jack but at the input ? what kinda signal do I have to inject in ? 2khz ? 10mV ? no importance ?

Seems the Tone Lizzard site is finally down for good.  Too, bad it had a great tutorial on this.  Anyway:

do you mean in his article about scopes? use the way back machine! (archive.org)
http://web.archive.org/web/20050209005436/http://www.tone-lizard.com/Oscilloscopes.htm

[jjasilli]