Measuring Amplifier Outputt

[trevorg]

Is there anyway to measure an amplifier's RMS output or even peak output with a meter and then convert it on a scale.

Example: Meter reads .2volt ac = 20watts
Please help

[rone]

In principle all it takes is a measure of volts, but you can't play music and read with a typical voltmeter. In practice, it is more involved because a speaker is not a resistor and music is not a pure single frequency waveform. You need to monitor the output waveform for distortion because your voltage reading will not be accurate unless the output is a pure sine wave. The best way is to use a dummy resistive load and a sine wave generator on the input and monitor the waveform with an oscilloscope or distortion analyser.

By the way, you can blow up your amp or speakers very easily with test signals - so I wouldn't try it unless you are sure you know what you are doing. The FTC requires certain things for power ratings - I believe a preconditioning at 1/3 power for a length of time is one of the prerequisites, and many amps get REALLY hot during this phase. Imagine how hot at full power.

Much music has a typical peak to average power rating of about 10dB, which means that if you play music so that the peaks are just clipping, the average power is only ~1/10 the rated power of the amp.

[curt_c]

Provided Link: http://www.ftc.gov/os/2000/12/amplifierrulefrn.pdf


The preconditioning is 1/8 rated power now. This allows the MFG's to use save money by using smaller than what might be considered optimum heatsinks, and/or allows them to provide a 2 ohm rating.

See FTC doc below:

C

[wg_ski]

> In principle all it takes is a measure of
> volts, but you can't play music and read
> with a typical voltmeter. In practice, it is
> more involved because a speaker is not a
> resistor and music is not a pure single
> frequency waveform. You need to monitor the
> output waveform for distortion because your
> voltage reading will not be accurate unless
> the output is a pure sine wave. The best way
> is to use a dummy resistive load and a sine
> wave generator on the input and monitor the
> waveform with an oscilloscope or distortion
> analyser.

> By the way, you can blow up your amp or
> speakers very easily with test signals - so
> I wouldn't try it unless you are sure you
> know what you are doing. The FTC requires
> certain things for power ratings - I believe
> a preconditioning at 1/3 power for a length
> of time is one of the prerequisites, and
> many amps get REALLY hot during this phase.
> Imagine how hot at full power.

> Much music has a typical peak to average
> power rating of about 10dB, which means that
> if you play music so that the peaks are just
> clipping, the average power is only ~1/10
> the rated power of the amp.

Most of us playing around with amps don't have expensive test gear. But voltmeters and scopes we do. Drive the amp to visible clipping with a sine wave using a dummy load resistor. If you have no scope, connect a 1k ohm, 10 watt resistor in series with a speaker to keep the volume down and put that in parallel with the dummy load to monitor when it clips.

Measure the PEAK output voltage, using a peak detector and DMM set to DC volts or a DSO if one is available. The peak detector may be a 1N4007 diode and 10uF cap. Convert to RMS volts for a sine wave by dividing by the sqaure root of two. If you omit this calculation you get "car stereo watts" :-). Then calculate watts into your load resistance. This is not rigorous, but it is conservative. It is accurate if the waveform is a single tone that is undistorted before clipping. If anything, the power measured this way will *drop* as you over drive the amp because the rail voltages continue to fall so the peak output is less. If you measure AC volts, the effective voltage will continue to rise after the amp clips so you get the usual EIA-rated over-inflated power reading - with this arrangement it will NOT. And the meaurement is taken under full load on the power supply, and the meter is probably accurate at DC. This should only take a few seconds and even a cheapo amp should hold up that long before thermalling.