USB mic vs external mic preamp measurement input

[dlr]

I purchased a miniDSP Umik-2 to have as a calibrated mic that also would be used for some distortion meaurement of drivers. I also have other measurement systems with other microphones. One is calibrated and dedicated through a supplied preamp. I also have a 25+ year old hand-built wand mic using the Panasonic WM-60AY 25+ years old. Over time it appeared that discrepencies in measurements are related to the Umik-2, not the others. This is using REW for near-field measurements set at 10mm placement primarily.

Current measurements of a woofer I've had going back to 2004 when compared to a measurement I made in 2004 of that same woofer show nearly perfect overlay of current vs 2004 results in REW after shifting curves to overlay. Except for the Umik-2. It has a deviation starting about 80Hz down to 10Hz. Right where I planned to make distortion measurements and to use for splicing to far-field. I have tried to find any reason for this other than the mic. Windows 10 and Windows 11 both don't seem to cause this unless I'm missing something there. Same for REW. I can find no reason for it. When I test, I measure with one mic, reposition another one, change the mic used in REW, and measure again. Nothing else is changed. The difference is either in the mic or due to the Umik-2 being USB vs. line input from an external mic preamp for the others.

Is there any way that the issue is due to using a USB mic in Windows/REW?

 

[John Mulcahy]

Are you using the UMIK's cal file? If you are, you could amend it to align with your other mics.

 

[dlr]

Are you using the UMIK's cal file? If you are, you could amend it to align with your other mics.

Yes, I am using the cal file. It shows about 5dB correction needed at 20Hz, but with the correction it's down another 5dB at 20Hz. I tested this. Removed the cal file. The error was 10dB down.

Edit: I labored for weeks using the Umik-2 as reference until I compared the 2004 measurement to the others. All mics other than the Umik-2 are nearly perfect overlays to the 2004 result in the typical range for a 10mm measurement.

Edit2: The Umik-2 and my other calibrated (for SPL curve, not absolute SPL) in far-field are nearly identical up to 20kHz. They show divergence in the low end, though.

 

[dlr]

Are you using the UMIK's cal file? If you are, you could amend it to align with your other mics.

I had considered that, but I bought it partly for the absolute SPL calculation, partly for the distortion capabilities. If there is an issue with the low end, it may be the mic, may be the the cal supplied. If it's the cal, then yes, I can amend it to align. If it's the mic, then amending the cal would simply show a change only in the graph. But there may be an underlying issue that affects the distortion within the measurement if it's changed since their cal. I'd rather have one is guaranteed to meet the calibration supplied with the SPL curve.