TOPIC:

I'm definitely interested in buying one of those UMIK-1 microphones, but I've read a lot of concerns about their elevated noise floor (~50dB dBA, whereas other microphones in the same price range can do ~30dBA). To be fair this matches the specifications, which indicate max SPL at 133dB SPL and a noise floor at -74dBFS, which translates to an absolute noise floor of 59dB SPL. Nevertheless, I would appreciate being able to use the UMIK-1 to do THD measurements which require a clean, low noise floor.

I have the following questions:

• Last time I checked, there was an issue with the noise floor being dominated by a nasty spike at 1kHz followed by its harmonics. Is this fixed?
• What are the numbers regarding the noise floor for the recent revisions of the UMIK-1? I would greatly appreciate if people who ordered one recently could post a screenshot of what an RTA looks like in a very quiet room.
• Is the UMIK-1 well-isolated with regard to USB power noise or should I be careful what I plug this thing into?
  
• Does lowering the recording volume (i.e. in Windows) as any effect on the absolute (i.e. in dB SPL) noise floor? At first glance I would guess that the volume control is digital so it wouldn't have any effect, but I would sure like a confirmation.
  
• It appears that you can open the UMIK-1 and change the sensitivity using DIP switches . Depending on the where the noise is coming from, lowering the sensitivity might, in theory, lower the absolute noise floor in the same proportion. I don't require high sensitivity (I would be fine with just 110dB max SPL), so that wouldn't be a problem.
  • Has anyone tried this? Does this tweak have any effect on the absolute (i.e. in dB SPL) noise floor?
    
  • Is it safe to do this with regard to altering the microphone - i.e. is there a risk of causing damage, screwing up the calibration, etc.?

Thanks in advance for your answers!

edechamps wrote:

It appears that you can open the UMIK-1 and change the sensitivity using DIP switches . Depending on the where the noise is coming from, lowering the sensitivity might, in theory, lower the absolute noise floor in the same proportion. I don't require high sensitivity (I would be fine with just 110dB max SPL), so that wouldn't be a problem.

Err, sorry, I had that reversed. I meant increasing, not lowering, the sensitivity, of course.

@ edechamps,

Just realized with this second post that we missed your first one! Sorry for that and thanks for your patience.. Let's try to answer your questions in order.

- Noise floor issue: that issue happened on one batch back in February. That's quite a while ago.. This issue was fixed from the point we learned of the problem on subsequent batches (i.e. April).
- Noise floor would depend on your room. The fix that was done between the 2 rev was to remove the peaks.
- We haven't seen any issues on the USB power noise and effect of PC/Laptop of the past year that we have these..
- Volume control is digital and done at the driver level. The analog gain is done via DIP switches as you already noticed. Just open the mic and you can go up to 36dB of gain. That tweak has been done by quite a few especially on latest versions (no glue on body). You do have to be gentle when opening the ring though, it's not something you do everytime. You set it once and forget. It's not intended to be an analog gain control knob. You'll most likely damage the mic if you do it many times for each measurement. I don't see from the application we see (i.e. room measurement) why you'd have to change it often though.

Hoping this info helps.
Devteam

devteam wrote:

- Noise floor issue: that issue happened on one batch back in February. That's quite a while ago.. This issue was fixed from the point we learned of the problem on subsequent batches (i.e. April).
- Noise floor would depend on your room. The fix that was done between the 2 rev was to remove the peaks.

Ok, thanks. Do you have a new number for the noise floor? According to your specs, even without the peaks it would still be on the order of ~50-60dBA which is way above the noise floor of a quiet room, and still too high for my use case.

devteam wrote:

- Volume control is digital and done at the driver level. The analog gain is done via DIP switches as you already noticed. Just open the mic and you can go up to 36dB of gain.

Can you confirm that this modification increases sensitivity without increasing the noise floor in the same proportion?

@ Edechamps,

Our office isn't exactly quiet (high aircon ducts from building we can't turn off) so we're not able to take a quick measurement to answer you directly but we've asked our partner to perform a measurement in the anechoic chamber to update the specs. From a quick test in the room, it was hovering circa 30~40dBSPLA but a peak at LF was from the Aircon rumble. Once we get the real data back, it will be more useful to answer both questions.

Hoping this info helps.

I have a measurement room available that has <9dB(unwheighted!) SPL. We measured this with a 1" low noise Mic (B&K 4179, self noise < -2.5dB(A))
So if you are interested I would be happy to help out with such a measurement. Don't have a UMIK though

devteam wrote:

Our office isn't exactly quiet (high aircon ducts from building we can't turn off) so we're not able to take a quick measurement to answer you directly but we've asked our partner to perform a measurement in the anechoic chamber to update the specs. Once we get the real data back, it will be more useful to answer both questions.

Thanks. Looking forward to the results.

devteam wrote:

we've asked our partner to perform a measurement in the anechoic chamber to update the specs

Any news yet?

Just for your Info: I've performed some measurements in our anechoic chamber which has quite low noise (9dB(A) according to measurement with B&K 4179 1" ultra low noise microphone). The UMIK-1 was connected to a Dell Lattitude E6410ATG with SSD using REW and the supplied calibration file.

The A-wheighted noise of the UMIK-1 (700-0756) was 29.1 dB(A) and 43.1dB(unwheighted), which is both absolutely OK (a lower noise floor would limit the max. SPL). Please find attached the measured noise-spectrum.



If I will find the time i'll do some more tests including a comparison to a 1/2" measurement microphone from Microtech-Gefell and a calibration check with a calibrator (B&K Pistonphon).

Best regards, Daniel

Could anybody update me the noise of umik-2? spec says -105.3dbfs, and 24bit ADC has range 144.5dbfs, so its noise level is 144.5-105.3=39.2dbA? how to compare it with umik-1 figture?

Hi, fascinating to see a query here seven years after the last posts -- I just happened to be measuring my control room tonight and decided to take a look at this site for info on the UMIK noise performance. I used both my UMIK and a calibrated analog measurement mic while making pink noise and REW measurements. And I also checked the room noise floor and got a plot with REW very similar to the one above by Daniel (#9909) with a measured A-weighted value of about 29.5 dBA and C-weighted at 35.5 dBA (there is constant airflow in my studio which creates some low frequency noise which increases the C value). So I'd agree with Daniel on A-weighted self noise below 30 dBA, which is perfectly fine for room/speaker measurements.

Cheers . . .

The CEDIA spec for home theater background noise is 22db C weighted so neither UMIK is a good match for room measurments of that sort.