..realize this is an older thread, but
My solution: Trying to keep it cheap but high quality.
the *Line Audio OM1 but I'll use the Cross Spectrum "calibrated" UMIK-1 for checking/"calibrating" the OM1 once I get my new test baffle complete.
*(..though I just received the OM1)
WHY the OM1 (instead of just using the UMIK-1)?
1. Dual Channel measurements (or semi-dual channel measurements) - because it is NOT a USB microphone. (and used in conjunction with the UMC204HD which allows a return signal.) This allows for more precise timing/phase measurements AND the ability to effectively remove the amplifier and 204HD noise from the measurement (for non-linear distortion testing).
2. ..also for non-linear distortion testing, the OM1 is much lower in noise and particularly lower in noise at higher spls. This makes a difference for **extreme nearfield distortion testing (where the closer you get to the driver under test, the LOUDER the signal). PLUS if you want to do extreme nearfield non-linear distortion testing at different sp-levels to see how distortion increases it just makes it that much louder for the microphone (adding to distortion).
**you test with the microphone up very close to the driver for lower freq.s to lower the overall influence of the room.
note: I doubt the OM1 really needs a shock mount (at least indoors on a boom), and it will almost certainly interfere a bit with measurements.
The UMIK-1 also does "double duty" as an SPL reference (basically a sound-meter) with its standard calibration at 1 kHz with REW. Only pulling it out of its case for this (and then putting it back in immediately after use), with the rare exception of checking the OM1's calibration over time.
Both will need custom "booms" from carbon tube along with a custom XLR for the OM1 integrated into the "boom" for it, and I've yet to do that either.
That along with my UMC204HD should make for a cheap but hopefully very useful result.
If I needed ***very low noise I'd probably pull-apart a Rode NT1-A, create a rear absorber panel for it along with a more appropriate "case", and then "calibrate" it with the UMIK-1.
***much lower than the OM1, but it's a caridoid and not a small omni - as a result it's not suitable for extreme nearfield testing. What I might use it for though is with a DIY Anechoic room (plus cepstral editing for correcting the low-freq. behavior of the room) and 1 meter non-linear testing at lower freq.s (..but I'm not anywhere near that project yet.)
The Rode mic. would also be good for very low-noise vocal capture.