Trying to understand influence of 18db gain on UMIK-1 vs 12db suggested in guide

[sebna]

Hi All,

First of all I apologize if this is the wrong place to ask this kind of question. Please direct me to right one in case it is.

I bought myself UMIK to try and be more scientific about my HT setup (in non-dedicated room).

I have found this great guide linked in reading materials of this form and located in AVSforum (It seems that I cannot link it directly)

Which I am trying to follow but my UMIK is set to 18db gain when the guide calls for 12bd. I am trying to open this thing to change it to 12db but I cannot so far (I tried to soak it in Isopropyl and in nail polish acetanol based).

It normally should not matter that much to what gain setting UMIK is set but there is also gain slider in Windows 10 with mixes thing up. The guide calls to set windows 10 to 9.9db for UMIK with 12db internal gain. So when I do so with 18db UMIK I am about 10db a drift (lower reading) in compare to my random dedicated SPL meter which reads almost the same as my HTC One phone. To compansate for it in Windows I have to put gain in range of +20-24db, but I have read that windows gain is very bad as it is a digital gain.

So what will happen if I switch UMIK to 12db - will I get close to my other SPL meters reading with windows gain 9.9?

Does anybody know the right Windows gain setting for 18db UMIK in case I would fail to open this god thing!

Many thanks my friends for any advice or explanation provided.

 

[John Mulcahy]

You are making life unnecessarily complicated for yourself, Jerry's guide already says it doesn't matter what the mic internal gain is. There is absolutely no need to alter the internal gain of the UMIK unless you need to measure levels above about 115 dB. If you are using the UMIK with REW then REW will automatically take into account the effect of the Windows volume setting and show the correct SPL. Set the Windows UMIK volume to 0 dB and move on to measuring.

 

[sebna]

You are making life unnecessarily complicated for yourself, Jerry's guide already says it doesn't matter what the mic internal gain is. There is absolutely no need to alter the internal gain of the UMIK unless you need to measure levels above about 115 dB. If you are using the UMIK with REW then REW will automatically take into account the effect of the Windows volume setting and show the correct SPL. Set the Windows UMIK volume to 0 dB and move on to measuring.

Hi John,

I have re-read the guide again and indeed Jerry sates that 9.4db gain in Windows 10 is close to 0db of real gain if I understand it correctly but why then with 18db UMIK-1 and calibration file imported to REW and those settings in Win10 I am getting 10 db lower SPL reading to my dedicated SPL meter?

This will completely skew my measurements will it not? If I would like to use REW measurement to set the absolute SPL reading at MLP (so how can I do that if it is off by 10db)?

If setting of gain of 9.4db in Windows is universally good independently of internal gain of UMIK then it would mean that either my UMIK or my SPL meter or both are widely inaccurate (by 10db or anything inbetween)? But my meter matches my phone perfectly so that would suggest that UMIK is off by 10db?

EDIT: I will check tomorrow with another SPL meter. If it matches my phone and other SPL meter should I then recalibrate the UMIK in REW? If I would be doing this does it mean that I no longer need the 90 deg cal file supplied by miniDSP or does it mean that I still need it but that sensitivity parameter in Cal file will be replaced internally by value acquired by REW during re-calibration process

Thank you

 

[John Mulcahy]

It would be very unusual for a UMIK to be off by 10 dB, I've never come across that. What serial number is your UMIK?

 

[sebna]

It would be very unusual for a UMIK to be off by 10 dB, I've never come across that. What serial number is your UMIK?

Hi John,

The serial number is 705-5638

I will be checking today one more time the comparison between 4 devices.

Thanks

 

[John Mulcahy]

Make sure you have the latest cal file for the mic from the miniDSP site, may be worth downloading it again.

 

[sebna]

Ok. After setting up proper test they came very close indeed. What is interesting is that in 90' mode the wind cover makes about 5db difference to UMIK-1 when in normal mode it is more or less transparent acoustically.

I am happy to proceed with my measurements . Thank you John for your help.

The photo below is from dbA fast but the drift was about the same in dbC S/F and dbA S.

I measured at 1m using 1khz test tone.

 

[John Mulcahy]

Better using noise signals signals to check levels, such as the pink PN or pink random speaker cal signal, tones will show large variability over short distances due to reflections.

 

[sebna]

Better using noise signals signals to check levels, such as the pink PN or pink random speaker cal signal, tones will show large variability over short distances due to reflections.

I have noticed the effect of reflections. It was sever. I gave them even chances by standing exactly behind them and having them measuring virtually in the same spot (as per picture).

If I will be ever doing it again I will use pink noise. Lesson learned. For the moment I am happy with the answers my test provided.

I will start taking measurements today / tomorrow . The first goal is to find optimal location for the Sub to squeeze maximum out of it. The room is not easy and MLP is fixed and ATM Sub is not getting the job done so I hope I can find the right spot for it.

After that I will look into adjusting phase and time alligment in general and addressing waterfall problems if sever found.

If I plan to run 5.1.4 do I ever measure measure the full set at once?

 

[John Mulcahy]

Nope.

 

[Dent]

Hi John,

I have re-read the guide again and indeed Jerry sates that 9.4db gain in Windows 10 is close to 0db of real gain if I understand it correctly but why then with 18db UMIK-1 and calibration file imported to REW and those settings in Win10 I am getting 10 db lower SPL reading to my dedicated SPL meter?

This will completely skew my measurements will it not? If I would like to use REW measurement to set the absolute SPL reading at MLP (so how can I do that if it is off by 10db)?

If setting of gain of 9.4db in Windows is universally good independently of internal gain of UMIK then it would mean that either my UMIK or my SPL meter or both are widely inaccurate (by 10db or anything inbetween)? But my meter matches my phone perfectly so that would suggest that UMIK is off by 10db?

EDIT: I will check tomorrow with another SPL meter. If it matches my phone and other SPL meter should I then recalibrate the UMIK in REW? If I would be doing this does it mean that I no longer need the 90 deg cal file supplied by miniDSP or does it mean that I still need it but that sensitivity parameter in Cal file will be replaced internally by value acquired by REW during re-calibration process

Thank you

In the guide, Jerry never said that 9.4db gain in Windows 10 is close to 0db of real gain. He just said that is what he uses and it works good for him. I use Windows 10 and set the UMIK volume level to 0.0db. 0.0db is unity gain. Even John said in a reply to you to just set the UMIK volume in Windows 10 to 0db and move on. To me, setting the UMIK volume to 9.4db is 9.4db above unity gain. Now if it doesn't clip in REW does it matter even if the gain is above unity? I don't really know but I just keep mine at 0db (unity) and it works for me. 9.4db (9.4db above unity gain) works for Jerry. We all get readings either way. I have a UMIK-1 with an internal gain of 18db just like you. Our UMIKs already have an internal gain 6db above Jerry's 12db internal gain UMIK so I figure our UMIKs are already outputting a 6db stronger signal than Jerry's for a particular Windows UMIK volume level setting.