On Wed, 07 Mar 2007 09:45:08 +0000 (GMT), Jim Lesurf
wrote:

In article , Don Pearce
wrote:


You won't see nulls caused by standing waves in a noise test. You need
the protracted phase coherence of an actual musical note to set the wave
up - it takes a few cycles to get going.

Not quite. You *will* see nulls in the frequency response with noise as a
test signal. But to do so requires the *measurement* averaging time to be
long eough - thus allowing the frequency resolution to be good enough. You
can impose the 'coherence' by measurement.


OK, this is true, and of course I did identify a frequency for him to
try exactly this way from his noise test. But that test was never
going to show the 23dB dip he was able to find with the sine tone
test.

As you say, coherence is where you impose it, but for this purpose the
easiest place to do that is in the generation of the signal.

d

--
Pearce Consulting
http://www.pearce.uk.com