Volume matching two amplifiers

[BeerBear]

I'd like to match the levels of two amps with a multimeter, so that they both produce the same volume when using the same speakers. For my needs, testing with a simple test tone (like a 400Hz sine wave) is fine.

Do I just measure the AC voltage from the speaker terminals? With a speaker load attached or can it work also without?
Would measuring the current be useful?

And while I'm at it, I would also like to know the power that is being sent to the speakers. Can I just do voltage squared, divided by resistance (both measured at the speaker during playback)?

 

[Sokel]

Are the amps exactly the same?That can be handy,specially if we're talking about load depended amps.
You need no load,measure at speaker terminals (beware not to short them! ) with a 50-60Hz signal (or higher if your DMM is true RMS) in AC scale.

You don't need to play very loud.

 

[voodooless]

Actually, I would recommend leaving the speakers connected. Some amps actually don't like not having a load and may give unexpected results.

As for measuring power: speakers don't have just a resistance, but a complex load that varies over frequency. So you can't just make a simple calculation. You could use your DMM in series to measure current, but then you need voltage as well... You'd need a second DMM for that, or switch over. This will work for single-tone measurements, but with real music, it would change so fast, that there is no way to derive power from this. Easier to just measure power from the AC side. That's a more averaged value, and you can make a rough guestimate of the efficiency of the amp to come to the approximate power the speakers get.

 

[Sokel]

Actually, I would recommend leaving the speakers connected. Some amps actually don't like not having a load and may give unexpected results.

I meant no extra load other than the speakers that's why I wrote speaker (and not amp) terminals but it's good to make it clearer.

 

[HarmonicTHD]

Keep in mind. Cheap DMMs can’t measure 400Hz AC. So try 50 / 60Hz instead.

 

[DVDdoug]

And while I'm at it, I would also like to know the power that is being sent to the speakers. Can I just do voltage squared, divided by resistance (both measured at the speaker during playback)?

In most cases you can just assume the nominal/rated impedance. Speakers are tested & spec'd with "constant voltage" across the frequency range and the impedance is "assumed constant" for the purposes of power rating, even though it's not.

And amplifiers are also "constant voltage"... The voltage is mostly independent of load, except at the power limits the amp might clip at a lower voltage.

Be careful with high-power test tones! Speakers are rated for program peaks and they can be blown with constant full-power, and tweeters can't handle as much power as woofers.

Keep in mind. Cheap DMMs can’t measure 400Hz AC. So try 50 / 60Hz instead.

400Hz usually isn't a problem and if you don't need power it doesn't have to be accurate it just has to be consistent. At high audio frequencies the meter may be less accurate. That's easy to test because the amplifier probably has flat frequency response.

Keep in mind. Cheap DMMs can’t measure 400Hz AC. So try 50 / 60Hz instead.

 

[eddantes]

Use pink noise. It'll match close enough.

 

[Blumlein 88]

Keep in mind. Cheap DMMs can’t measure 400Hz AC. So try 50 / 60Hz instead.

Never ran across one that would not measure 400 hz. Including very very cheap ones. There might be some out there, but they aren't common. Most DMMs specify up to 400 hz because some generator systems use 400 hz rather than 50/60 hz.

 

[Blumlein 88]

Use pink noise. It'll match close enough.

Not as precise as a sine wave, and no easier to do. You want voltage matching within 1%.

 

[HarmonicTHD]

Never ran across one that would not measure 400 hz. Including very very cheap ones. There might be some out there, but they aren't common. Most DMMs specify up to 400 hz because some generator systems use 400 hz rather than 50/60 hz.

My cheap one doesn’t but my expensive one does.

Good to know that this might be an exception.

But yes voodooless is right as one is more interested to compare the values as opposed to correctly measure the absolute value, even my cheap one might work for that.

 

[Speedskater]

The meters don't need to be accurate at the frequency of the test signal, they just need to be repetitive.
I would test with the amp connected to the loudspeaker and measure at the speaker terminals..
Using a tone in the 400 to 1000 Hz range or maybe pink noise.

 

[BeerBear]

Are the amps exactly the same?

No, different amps. Anyway, the comparison ended prematurely, because one of the amps turned out to be junk...

But I did do some measurements with the good amp, both current and voltage.
It's interesting to see how much the current/power draw varies, depending on the test tone frequency. Of course, it makes sense when you look at the impedance vs frequency graph of the speakers.

For volume matching, I think a constant sine tone is a better option, because the voltage stays constant. With pink noise it jumps around more (though it's roughly in a range). But it probably makes sense to use several sine tones at different frequencies...

One disappointing thing that I noticed is that quiet (but audible) sounds aren't picked up by the multimeter. I was hoping to measure the self noise (hiss) of the amp, but I guess that's much less than a mV.