differnt sensitivity between two UMIK-1 - question/curiosity

[MCH]

hello,
Since recently i own two different UMIK-1 calibrated microphones. I got both used so i don't know if one is older than the other or belong to two different generations... they do look identical.
I plan to keep only one of them but of course like any other geek would do, took the chance to compare them.
Surprisingly for me, while the FR seems reasonably similar (identical 40-600 Hz, some differences outside that range), one of the mic captures consistently ca 3,5 db more SPL than the other across most of the spectrum (sorry if i use the wrong terminology).
3.5 db seems like a huge difference to a noob like me, maybe it is not... Is this normal? is higher or lower sensitivity somehow better? - thinking on which of the two mics i should keep.
Thank you for your comments.
Here is an example (taken with same equipment other than the mic, same levels, same everything except of course the calibration files):

 

[bloodshoteyed]

did you use the downloadable calibration files?

 

[MCH]

Yes, each mic used with its own calibration file

 

[Speedskater]

It's not surprising that they are a little different.
In room repeatability and reproducibility are not easy.
* * * * * *
an old Earthworks paper with lots of DIY info:
How Earthworks Measures Microphones
By Alex Khenkin, Director of Engineering, Earthworks
How to measure microphones and the implications relating to measuring loudspeakers

https://earthworksaudio.com/wp-content/uploads/2012/07/how-earthworks-measures-mics.pdf

 

[DVDdoug]

3.5 db seems like a huge difference to a noob like me, maybe it is not...

That's more than I'd expect too. I wouldn't expect that much variation from a regular mic (from Shure, etc.). Maybe they changed something in the design or maybe they just don't care since it comes with a calibration file....

With the correct calibration files they shouldn't be that far off.

I'd put some colored tape on the mics (or something like that) so you can easily identify which-is-which. A band of colored heat-shrink tubing might be "more professional".

An advantage of a USB measurement mic is that the level is calibrated. With an analog measurement mic the frequency response is calibrated but the digital level depends on the preamp/interface.



,,,I read somewhere that the "permanent charge" in electret microphones deteriorates over time and they lose sensitivity. I was doing some "experiments" recently with my antique Radio Shack SPL meter so I bought a little calibration gizmo (about $100 so about the same price as another inexpensive SPL meter). The old analog meter was "perfect" (as accurately as I could read it).

 

[MCH]

Thanks guys,
As I guess many others here, I only really care about relative FR, and this is not a big deal. But i can imagine others need to measure absolute SPL levels, so I was expecting a closer result.
I am not a REW expert and chances are that i did somthing wrong, even though i only swapped mics and cal files, so i will try to do a few more measurements and see what happens.
Cheers.

 

[Absolute]

Go to minidsp.com and check if there's more/new calibration files. I remember there were two versions of one of mine, with the gain being the only difference.

 

[MCH]

more on this (again, just for curiosity, not trying to solve or demonstrate anything):

I did another simple test. I measured the SPL with the SPL meter in REW. Mics placed one right besides the other, same rig playing the same pink noise at the same volume. Just swaped the cable connected and respective calibration files:
Umik-1 A: that generally gives higher readings (blue line in post #1): ca. 78.4 db
Umik-1 B: that generally gives lower readings (orange line in post #1): ca. 74.3 db
(all these values are the aprox average, as the reading in rew is always moving 0.3 db up and down)

Coincidentally, if i look at the two calibration files (@Absolute : there is only one file for each of the two mics in minidsp.com), the difference in "sens factor" between the two mics is 1.879 - (-1.632) = 3.511 db, that is ca. the difference i get in the readings:

(-1.632db is for the Umik A and 1.879 is for Umik B)

was REW ignoring the sens factor??

With Umik B still conected (the one that gives the lower readings), i changed the calib file to the one of Umik A, and the reading of the SPL went from 74.3 db to 77.6 db . This seems to indicate that REW indeed takes the "sens factor" into account.
Mistery not solved....

 

[ernestcarl]

I would either get a third SPL meter microphone to check, or adjust one of the microphone calibration file so that they match closer only in order to put your mind at ease. If I remember correctly, the difference between my UMIK-1 and another dedicated SPL meter is only about 1.5-2 dB max

 

[Gruesome]

more on this (again, just for curiosity, not trying to solve or demonstrate anything):

I did another simple test. I measured the SPL with the SPL meter in REW. Mics placed one right besides the other, same rig playing the same pink noise at the same volume. Just swaped the cable connected and respective calibration files:
Umik-1 A: that generally gives higher readings (blue line in post #1): ca. 78.4 db
Umik-1 B: that generally gives lower readings (orange line in post #1): ca. 74.3 db
(all these values are the aprox average, as the reading in rew is always moving 0.3 db up and down)

Coincidentally, if i look at the two calibration files (@Absolute : there is only one file for each of the two mics in minidsp.com), the difference in "sens factor" between the two mics is 1.879 - (-1.632) = 3.511 db, that is ca. the difference i get in the readings:

View attachment 226190
(-1.632db is for the Umik A and 1.879 is for Umik B)

was REW ignoring the sens factor??

With Umik B still conected (the one that gives the lower readings), i changed the calib file to the one of Umik A, and the reading of the SPL went from 74.3 db to 77.6 db . This seems to indicate that REW indeed takes the "sens factor" into account.
Mistery not solved....

I know my comment comes a bit late, but could it be as simple as a missing minus sign in the 7047079 calibration file for the Sens Factor?