What is your purpose in gathering this information? A DVM can get you an answer that is useful for some things, such as the DC resistance typically (not exactly) being 3 ohms for a 4 ohm speaker while it's 6 ohms for an 8 ohm speaker.

But do you need to know the actual impedance at some particular frequency? That is WAY different from distinguishing between a nominal 8 ohm speaker and a nominal 4 ohm speaker. (These are called "nominal," which means "it's not exactly that, but that's what we call it, because we either don't need better information or we don't know that we need better information.")


When I worked at Marantz around 1980, my boss had come up with a pretty good method for this. We ran plots of impedance versus frequency with the intent of knowing whether the crossover/speaker combinations presented well-controlled signals to the individual drivers. For instance, we had one midrange speaker for which the response curve in its useable spectrum was not a (more or less) level line, but instead rose across its spectrum. By manipulating the impedance of the crossover he created a crossover with a complementary non-flat impedance curve, which resulted in a flatter audio response. He wanted a skewed voltage curve to present to the driver in order to counteract the driver's audio response curve.

His method was this:
Set the output of a power amp at ten volts.
Put a 1000 ohm resistor in series with the speaker.
Measure the voltage across the speaker.

This voltage divider had two portions: a small voltage across the speaker and a large voltage across the resistor. The voltage across an 8 ohm load was almost exactly 8* mV. If the load were 10 Ohms, the voltage would be almost exactly 10* mV. This voltage was fed into the Bruel & Kjaer plotter and scaled to give us a chart showing impedance versus frequency.

This was quite accurate, especially since we were aware it was almost exactly right, at impedances under, say, 30 ohms. The higher the impedance, the less exact we needed it to be, which was nice, since it got less accurate at higher load impedances.

In addition to its being accurate, it did not lend itself to field use. You'd have to drag around an amp that could output ten volts and you'd have to spend the money on a meter that would accurately measure AC down to less than ten mV. Look at the various meters you've got -- accurate measurement of very low voltages costs money!

Was there other information you needed about this?


*or 80 mV. Somebody check my arithmetic, please.