Inductive vs. noninductive dummy loads?

[Axtman]

I have read that it is best to make a dummy (speaker) load out of a noninductive resistor. Then I have read that speakers are naturally inductive so I should use an inductive reistor..

So which one is it?

[Chuck H]

The non inductive stipulation in this case probably has to do with bench testing where a resistive load is used as a base measurement(?). I've used plenty of cement resistors and those aluminum finned chassis mount types for attenuator circuits (both are wire wound types AFAIK). Their inductance is truly tiny compared to what any would consider a reactive inductance load for an amp and I've never had any trouble using them in that regard. Bench tests not withstanding I don't expect there's any detriment to any amp using wire wounds for appropriate loads. Have you seen speaker plots? They're pretty extreme at certain frequencies. No resistors inductance would ever be as discordant as an actual speaker. So this really isn't about proper operating conditions for an amp. I think it's more about proper conditions for bench testing. I think you can pretty much ignore the inductance of any resistor being detrimental WRT building attenuation circuits.

[Helmholtz]

Speaker inductance is around 1mH.
With wirewound power resistors I measured a few µHs, so totally insignificant with a dummy load.

That's actually a pity as for stability tests it would be nice to have a load that's closer to a real speaker.