Creating a response correction file from NIST traceable certificate

[Matthew J Poes]

Hi Jon,

I just picked up a BSWA measurement microphone with 1/2" 'Class 1' capsule. The company I purchased it from included a NIST traceable certificate. It's a piece of paper with an image of the response measured both free field and with an electrostatic actuator (pressure field). It then has an "upper" and "lower" response deviation in a table. It looks like the "upper" is the free field response deviation and the lower is the pressure field response deviation from 0. First, would that make sense to you?

Second, I assume I can use this to create a correction file. Any helpful tips for that? I can probably scan the document in and read the table into excel with OCR software I have, and then just output that as a tab delimited txt file. It may also be quicker just to manually type it in.

I'm not sure if this is just the norm for NIST or my mic, my paper document only extends down to 20hz, but extends up to 31.5khz. The capsule also only has an indicated response down to 20hz. Everything below 500hz has a correction of 0 for both pressure and free-field. When I asked the company what it's true low-end limit was, they said likely single digits, but they don't have the facilities or capability to test that. Any thoughts on what it's true low-end limit probably is? I assume if it hasn't rolled off at all below 20hz, it is flat to below 20hz, but that still could mean 18hz, 15hz, or 2hz. Based on the capsule (it appears to be a copy of a G.R.A.S. capsule as it has identical specs and an identical appearance) I suspect it extends below 5hz (That capsule was -3dB at 5hz).

 

[3ll3d00d]

vituixcad has a decent trace tool -> https://kimmosaunisto.net/Software/Software.html#SPL_Trace

 

[John Mulcahy]

Could also extract the data with something like this: https://automeris.io/WebPlotDigitizer/index.html

Data below 20 Hz is much harder to produce. As an example of LF extension for another mic, UMIK-1 tends to be -3 dB at 12 to 13 Hz per the miniDSP cal files (which go down to 10 Hz).

 

[Matthew J Poes]

Thanks everyone. The NIST file had the raw numbers, no need to trace the response. I just wanted to be sure I was using the right numbers. I hand typed them in last night, took all of ten minutes.

I was told that the “accurate” way to assess the lower limit of the capsule is based on a pressure shift at the side of the capsule?

In any case it looks like I may be getting some help with this. One of the leading manufacturers of measurement equipment had a local affiliate that has indicated an ability to calibrate my mic at low frequencies if I bring it.

The person I talked to was very skeptical of the LF accuracy of the calibration files with these usb mics. He said he would tell me why when I stopped by, but seemed to have strong opinions about it. Since I have such a mic, I will bring it by (but I know the approach he is going to use on my mic won’t work on the Dayton mic).