Home Theater Forum and Systems banner

1 - 8 of 8 Posts

·
Premium Member
Joined
·
134 Posts
Discussion Starter #1
Hi folks, if you'd be interested in trying a 90 degree calibration file for your mic and don't have one, here's a procedure to make one:

http://johnr.hifizine.com/2014/09/how-to-make-a-90-degree-calibration-file/

Now, it seems complicated, but if you follow the steps it should be simple. I haven't (yet) been able to find a more straightfoward method. It does seem to work tho (will try more proper measurements tomorrow).

Let me know if you try it :nerd: :)
 

·
Registered
Joined
·
261 Posts
can't you just do it using trace arithmetic alone rather than exporting to a spreadsheet? i.e.

take 2 measurements for 0 and 90 degree
create a 3rd by A-B to give you the difference in the response
invert the result (not sure if REW has this)
add that to the 0 degree
export as a frequency response & then convert that to cal file format

I did something something similar for converting a mic cal file to an FFT filter for speclab, details in http://data-bass.ipbhost.com/index.php?/topic/290-using-speclab-for-measuring-in-room/?p=4124
 

·
Banned
Joined
·
4,838 Posts
For the UMIK-1 specifically rcohen made a utility to generate a 90 degree file, see this post.
User flavio wrote out easy-to-follow instructions for that utility:

"To use that "micdelta" utility with Win7...

- The 0° calibration file of your UMIK-1 should be copied inside the micdelta folder together with the other files

- Press and hold Shift and right click on the micdelta folder to open the command prompt at that location and click on Open Command Window Here

- Enter the following string replacing umik1_serXXXXXX.txt with the name of your calibration file:
micdelta.exe umik1_avg_0.txt umik1_avg_90.txt umik1_serXXXXXX.txt umik1_serXXXXXX_90.txt

- Enter

DONE"
 

·
Premium Member
Joined
·
134 Posts
Discussion Starter #5
can't you just do it using trace arithmetic alone rather than exporting to a spreadsheet?
You would need the A-B operation to work on the magnitude only (no phase) in dB. I wasn't able to make REW do that.. :( (I could have missed something?)

Thanks JohnM and AudiocRaver, I'll try rcohen's program when I get a chance.
 

·
Premium Member
Joined
·
1,562 Posts
The method I used is similar to what HifiZine posted above (measurements and spreadsheet). I settled on a mic distance of 3" after trying 1" and 10" distances. I used my TW for the measurements as it runs from 1700 to 20k Hz. I did not smooth 1/3 octave, but did a similar thing by fitting the best fit linear line through the delta data and using its formula to calculate the 90° cal data. The SPL delta data shown below was the result of the A / B math of the 2 mic orientations. I don't think there would be any significant difference between what I did and what HifiZine did. The measured SPL delta is for my Beyerdynamic MM1.

90 Vs 0 deg.PNG

I suspect that rcohen method is very good also for the EMM6. My only concern is that it appears that some mics start the rolloff (and peaks) at a slightly different freqs than other mics of the same type (see the charts from CSL). I am guessing that this is maybe due to the mic diaphragm tension and resulting natural resonance freq of the mic. Whatever the reason, measuring the individual mic at both orientations will be a little more accurate that a generic calculation. The difference may be trivial for most EMM6 mics however. For other mic models the rcohen method may result in larger error as different mic models should average a different correction factor as it tend to be also be impacted by mic capsule diameter and barrel diameter.

Mic Phase:
I have also seen a very old Bruel & Kjaer document that indicated the mic phase is close to the minimum phase of the power response of the mic. The power response of a mic is very close to the 90° response. The power response drops off possibly 1 degree more at 20kHz as near as I have been able to glean from various charts. That is trivial in terms of impact on phase so I use the 90° SPL to calculate the phase response of the mic and use those numbers for both 90° and 0° cal files. My measurements confirmed that the phase is not impacted noticeably by mic orientation. Below is the 90° chart for my MM1 including the phase response. This is not easy to do as I had a tough time learning to use FRD properly to calculate the minimum phase. Possibly there is an easier to use program to do this however.

SPL at 90.PNG
 

·
Registered
Joined
·
261 Posts

·
Premium Member
Joined
·
1,562 Posts
3ll3d00d,
That's a great thread. I only scanned it, but agree with most all the points made there. The last post reflects my position in the matter (Post 15, Tsardoz). This is a DIY hobby for some of us. I just like the puzzle. I have always been interesting in calibration issues and assisted with lab calibration issues occasionally on my job. I agree there is no practical reason to worry about this. All the reference info I found shows the phase impact to be very modest and most experts would agree we are not sensitive to reasonable phase rotation issues anyway.

That said, I use a rePhase filter to take out the phase rotation of my system. It is nice to see normal looking IRs and step responses. Since I am working to that end it behooves me to correct the various calibration issues that impact this. There are other calibration issues as well and this is not most important issue. It is just one that I was able to address.

B&K stopped providing phase data on their mics long ago citing the same issues discussed in the thread you linked. I agree that the minimum phase assumption is probably not entirely correct. It is less correct though to assume there is no phase rotation. My MM1 has a capacitor and a transformer in the mic preamp. It is not logical that they do not impact the phase response of the mic output. I think using the minimum phase response of the mic is closer to correct than assuming phase to be zero.

A simpler way to answer your 2 questions is; "the relevance and impact is shown in the phase chart I posted". I am sure I would not be able to hear any sound quality impact. Possibly those that use special power cords, speaker wire, and can detect differences in amplifiers could? :rolleyes:

I probably should not have mentioned this in this thread. I mostly just wanted to support the method offered to create the 90° cal file.
 
1 - 8 of 8 Posts
Top