Newby Q: difference btwn Step and Impulse response

[brucemck2]

I'm just beginning to get the hang of taking individual driver measurements (using the measurement functionality of a DEQX).

A "simple" question: what is the best way to think about what "step response" is showing, and, how is that related to (or different than) the impulse response?

This came up when I was attempting to time align drivers on the left and right channels .... it's not obvious whether I should attemp to align the initial rise of the impulse responses, or align the initial peaks of the step response charts.

 

[djarchow]

I wasn't sure what you meant my time align drivers on the left and right channel. If you mean a complete loudspeaker, and if the loudspeakers are the same, then you should be able to align either peak of the impulse or the peak of the step response. You could also use an energy time curve (ETC) and again align on the first peak.

the initial rise of the impulse response is usually ambiguous enough that the peak is easier to read.

On the other hand if you are trying to time align drivers such as a woofer and tweeter, then you still won't get very close, as with any measurement system you can only determine the peak of the impulse (or any point for that matter) to within one sample width in time. For a standard 48kz soundcard, that means +- 7mm.

Regards,

Dennis

 

[MJK]

If you take the derivative of the step response, with respect to time, you should get the impulse response. They are very closely related. The conclusions drawn from either should be the same.

 

[brucemck2]

Thanks ... both the deriviative explanation and the sample rate caveat were exactly what I was looking for