Mic Calibration file problem

[Achim Wulff]

Hello,
i use REW for Car Hifi with 90° calibration. Its a ECM E40WZ with pre amp.
I received my new calibration file from Hifi Selbstbau.
Problem: The peaks during measuring are to high. If i have a peak at 1Khz with 6db, and reduce in my dsp about 2 db, it reduced measured -8db. So the scale from Mic to measured is wrong.
Did i have to use C or Z weighted Mic cal?
From Hifi selbstbau, they send me an email, that i have to insert a line in the cal. file:
Sens Factor =x.xxxxdB, SERNO)
Can i use the dB at 1Khz from the calibration for the sensfactor, or must it be somehow calculated?
Regards
Achim

 

[John Mulcahy]

The cal file has no effect on scaling, it is simply subtracted from the measurement when producing SPL plots. Measurements in a car are very sensitive to position, so the differences are more likely to be due to measuring at a slightly different position.

C weighting compensation would only be used if you were measuring with a C-weighted SPL meter.

Sens Factor is only for the USB mics REW recognises, for analog mics you can calibrate the SPL reading manually. It is possible to add sensitivity information for analog mics, but there are some limitations. From the help:

If you switch between a number of analog microphones or interfaces there is a method to include sensitivity information in the mic cal file. Add a line at the start of the file of the form:

Sensitivity -12.34 dBFS

where "-12.34" is replaced by the REW dBFS input level when the microphone is driven by 94 dB SPL. When that cal file is used REW will use the sensitivity data to calculate the SPL calibration offset. Note, however, that this calibration will only be valid when using the same mic interface gain and input volume that were used when it was measured.

 

[Achim Wulff]

Ok, so i have to select Z weighted?
This problem with the difference is only with REW. If i use Arta, its normal. I use Pink Noise for all my measurments.

 

[John Mulcahy]

Yes, Z (i.e. no) weighting. If you suspect some issue with the cal file, just clear it and see what difference there is measuring without one.