TOPIC:

Hello,

I'm developing a custom application using Umik-2.
I downloaded the calibration file for my mic, but I don't understand how these values will help me get from dBFS to dbSPL.

Specifically,
1. What is Sens Factor?
2. What is AGain?
3. What are the units of the two columns? I'm assuming the first column is frequency. Is the second decibels?

Can I simply add the second column to dBFS measured at each frequency to get dBSPL?

Thanks,
Josh

I have seen this explanation  regarding the UMIK-1 a while ago:

The cal file sensitivity figure is the dBFS input reading for 100 dB SPL with the input volume set to 100%, which is 24 dB of digital gain. The UMIK internal gain (18 dB in your case) is taken into account in the cal file sensitivity figure, so your UMIK with the internal gain set to 0 dB would need a sensitivity figure of -18.673 dB.

@ Sanworks

Each OS has different interpretation of dBFS to % and it's not a trivial answer. Easiest to use REW as a "reference" for that reason to test your software. 
1. What is Sens Factor? 
ANS: It's the sensitivity of the mic. 

2. What is AGain?
ANS: It's the analog gain. Not relevant since it's always steady and can't be easily modified. 

3. What are the units of the two columns? I'm assuming the first column is frequency. Is the second decibels?
ANS: Freq and dB offset.

@ devteam

It would be good to have a proper explanation for the sensitivity number in the calibration files.

Does this statement still apply to the latest UMIK-1 cal files?
The cal file sensitivity figure is the dBFS input reading for 100 dB SPL with the input volume set to 100%, which is 24 dB of digital gain. The UMIK internal gain (18 dB in your case) is taken into account in the cal file sensitivity figure

Does this apply to the the UMIK-2 as well?

Thanks.
 

Hello Dev team,

If possible, I'd really like a proper definition of Sens Factor.
Earlier in the thread, it was given as:

1. What is Sens Factor? 
ANS: It's the sensitivity of the mic. 

How is "sensitivity" defined? What are its units?
How would I combine Sens Factor with dB Offset to get from the calibrated frequency to the raw measurement?

Thanks,
Josh

Sens  factor is a calibration that tells the software exactly how many db it is reading 

otherwise you need a mic calibrator to calibrate the spl level 

sens factor tells the software it’s offset so spl readings are accurate 


in REW if you remove that from the cal file , you’ll get a warning dialogue box that will say spl isn’t accurate 

In my calibration file, Sens Factor =-14.82dB. 

If Sens Factor is an offset that "tells the software exactly how many db it is reading", does this mean that I subtract 14.82 from all the values in the second column to get the true calibration?
Or is it used somehow else?

A precise definition of the Sens Factor would really help.

Thanks!
-Josh

I’m not sure how REW reads it because spl is relative to me and I don’t ever really need it to be accurate 

there’s one way you can find out do a test make a copy of the calibration photo and remove it and see what happens

but I’m not sure if it subtract  or adds 

OK, sounds like we need devteam or someone else at miniDSP to weigh in here. What is the precise definition of Sens Factor?

Full disclosure, I'm not trying to compete with REW. I need this information for a custom MATLAB/Python application that would use UMIK2 as a calibration tool for sounds in Neuroscience experiments. If this works and the results are consistent across microphones, a whole bunch of brain scientists around the world will buy UMIK2's from miniDSP.

There is an article here www.analog.com/en/analog-dialogue/articl...one-sensitivity.html
For digital microphones such as the UMIK-2 the sensitivity number is the digital output level relative to full scale when the SPL is 94dB.
This is the standard way of specifying it with 94dB being the reference level, I believe that the UMIK-2 conforms to this way of specifying it but the UMIK-1 does not.

www.focusfidelity.com