I have been happily using a UMIK-1 and REW for DIY speaker tuning and room EQ and it seemed like it was working OK. I am starting on a new speaker project and decided to get a little more serious and look at CAD help. When reading through the documents for VituixCAD I came across this in the "how to measure using REW section".
VituixCAD documentation said:
Note! Single channel measurement systems such as USB microphones (with latency variations by default) are not recommended for speaker engineering due to timing and phase variations and normalizations. REW should NOT be used without electrical loopback as timing reference or cal and timing reference for acoustical measurements to avoid timing manipulation by the program.
This reminds me of a similar mistake I made years ago when MSO was very new. The MSO docs at the time said that USB mics were not usable for gathering MSO measurements. This was correct back then, as REW did not yet have the acoustic timing reference feature.
A miniDSP applications guy found out about MSO and contacted me, saying that John was going to add the acoustic timing reference feature to REW. After hearing that, I remember thinking something like, "This can't possibly work. MSO uses measurements at multiple positions. With an acoustic timing reference, a different measurement position means the timing reference changes, because the distance between the reference speaker and the mic changes with mic position."
But then I realized that assumption was wrong, for a subtle reason.
Here's a summary of what is required for two timing references to be the same.
- With a loopback timing reference and XLR mic, the timing reference is always the same for all measurements
- For two measurements using an acoustic timing reference to have the same reference, these two conditions must be met:
- For both measurements, the speaker producing the acoustic reference must be the same
- For both measurements, the mic position must be the same
For complex summations of measurements to give the correct result, all the measurement quantities to be summed must have been taken with the same timing reference. So if the measuements were taken with an acoustic timing reference, they must meet the two criteria above for the complex summation to be valid (same reference speaker, same mic position).
In MSO, all complex summations are performed with data taken at the same listening position. There are no "cross-position" complex summations in MSO. That's why the REW acoustic timing reference works with MSO. The summations consist of all the contributions of a set of subs at a given listening position. At a different position, the timing reference is different, but all
those measurements have the same timing reference according to the above criteria.
Let's take an example where a USB mic with acoustic timing reference won't work.
There's an ambitious DIY project I've noticed: making a DIY version of the Klippel loudspeaker measurement system. The whole point of the Klippel system is to characterize the behavior of a single source at multiple positions. So if there is complex arithmetic involved, it's hard to see how it would
not involve cross-position calculations. That won't work for a USB mic with an acoustic timing reference at all.
If one were working on such a project and the USB mic gave bad results (which it will), it would be throwing out the baby with the bathwater to say that USB mics are unsuitable for
all phase-sensitive applications. They're only a problem when cross-position complex math needs to be done on the results.
