Pressure Response Compensation for Couplers

Discussion in 'Measurement Techniques Discussion' started by Dr. S, Apr 21, 2020.

  1. Dr. S

    Dr. S Rando

    Joined:
    Aug 28, 2019
    Likes Received:
    9
    Dislikes Received:
    0
    Trophy Points:
    3
    Location:
    Germany
    Hey there,

    is anyone of you using compensation curves to address the boundary effects inside acoustic couplers like the 60318-4?

    [​IMG]
    Source: Audioexpress.com

    [​IMG]
    Source: Brüel & Kjær

    [​IMG]
    Source: Microtech Gefell

    [​IMG]
    Source: G.R.A.S.

    I recently became aware that all the reference microphones used in couplers are tuned for the pressure response which deviates from the commonly used free field (0°) and diffuse field (90°) curves as shown above.

    Since most affordable measurement microphones are designed for compensating room acoustics in the diffuse field, the mismatch would most likely not be that relevant. You can see that there is an estimated difference of 2.5 dB at max around 15 kHz. However, using a microphone calibrated for the free field response there will be a deviation of up to 8 dB at 15 kHz. This would result in a significant treble attenuation.

    Regards
    Dr. S
     
    Serious likes this.
  2. Serious

    Serious Inquisitive Frequency Response Plot

    Friend BWC MZR
    Joined:
    Sep 28, 2015
    Likes Received:
    1,872
    Dislikes Received:
    0
    Trophy Points:
    113
    Location:
    near Munich, Germany
    Considering that most cheap microphones are usually made for flat on-axis response, doesn't that mean microphones are usually basically free-field (not DF)? But then again these are usually 1/8" 1/4" microphones, not 1/4" 1/2" or even 1/2" 1", so I'd think it shouldn't need as much compensation.
    With IEM measurements upper treble is rolled off for multiple reasons compared to what we hear and this is one part.
    There are measurements here: https://www.itu.int/dms_pub/itu-r/opb/rep/R-REP-BS.2419-2018-PDF-E.pdf
    1/8" 1/4" microphones with no extra enclosure attached to them should be virtually omnidirectional to almost 20kHz.
     
    Last edited: Apr 22, 2020
  3. Psalmanazar

    Psalmanazar Most improved member; A+

    Slaytanic Cliff Clavin
    Joined:
    Sep 27, 2015
    Likes Received:
    4,284
    Dislikes Received:
    134
    Trophy Points:
    113
    Dr. S, you can measure rooms with most decent omnidirectional condensors. Just compensate for their response.

    Serious, nothing you say, think, or do will stop people from using Shures. Shures work and even sound good once transformer modded.
     
    Last edited: Apr 21, 2020
  4. Serious

    Serious Inquisitive Frequency Response Plot

    Friend BWC MZR
    Joined:
    Sep 28, 2015
    Likes Received:
    1,872
    Dislikes Received:
    0
    Trophy Points:
    113
    Location:
    near Munich, Germany
    @Psalmanazar What are you on about regarding Shure? I genuinely do not understand what you mean.
    Shures have a reputation for being rolled off, is that it? Still, I do not know how that relates to this.

    This is really just about the deviation from omnidirectional that omni mics show.

    EDIT: Was thinking about Shure in-ears. As far as Shure mics go I'm completely clueless. Mics for recording are very different than measurement microphones anyway.
     
  5. Dr. S

    Dr. S Rando

    Joined:
    Aug 28, 2019
    Likes Received:
    9
    Dislikes Received:
    0
    Trophy Points:
    3
    Location:
    Germany
    Looking at the graphs from hifi-selbstbau.de who measured more than 1000 cheap microphone samples I would say that most are not designed to compensate for the pressure build up from the front (free field). You can see that most samples do need quite a bit of compensation to measure flat in FF. And the cheaper you go the larger the differences.
    Furthermore, the compensation files you will get from the calibration provider will usually cover the microphone as a whole. This will also add the rod (which holds the mic), the following body (which holds the circuit) and the clamp to the curve. When mounting the capsule or front nozzle flush with the coupler though, all those interferences will vanish. This is why those "pressure microphones" are usually calibrated within chambers or with the tip mounted flush with a (theoretically endless) boundary.

    Looking at the demands of DIY and the system as a whole I think that the standard DF adaption is mostly good enough. But I would not trust a FF equalized mic tbh.
     
    Serious likes this.
  6. Serious

    Serious Inquisitive Frequency Response Plot

    Friend BWC MZR
    Joined:
    Sep 28, 2015
    Likes Received:
    1,872
    Dislikes Received:
    0
    Trophy Points:
    113
    Location:
    near Munich, Germany
    True, many of them have a rising response. I think only certain models of electret mics are close to linear on axis.
    The mic body, tip and grille likely influence the results more than the capsule itself. The cheap SPL meter I bought that was advertised as using a 1/2" capsule (which I hoped would give me low noise) seemingly uses a 1/4" mic inside a 1/2" tip, yet its directivity closely resembles that of a normal 1/2" mic (the worst of both worlds LOL). Not only that, but it has a peak in its response that I think is unlikely to be from the capsule and is most likely caused by the tip.

    Can you not get individual capsules (dangling off wires) calibrated? Then use that (possibly in addition to a simulated free field to pressure field compensation based on the mic's size) for the IEC coupler.
    If not, maybe just make a rough compensation against a cheap calibrated USB mic. That's what I do, although the resolution of the UMIK-1's calibration below 300Hz is questionable at best. There seem to be room mode artifacts in my calibration file. But at the price point that's to be expected, I guess.
     
    Last edited: Apr 22, 2020
  7. Dr. S

    Dr. S Rando

    Joined:
    Aug 28, 2019
    Likes Received:
    9
    Dislikes Received:
    0
    Trophy Points:
    3
    Location:
    Germany
    Excluding the housing from the measurement procedure would minimize the effects mentioned above, for sure.
    But on the other hand you would potentially loose any shielding properties and introduce a lot of interferences which could corrupt the calibration process.

    It may work with "bare" capsules being fed by unbalanced plug-in-power. But not so much with enhanced circuits using signal balancing, floating grounds etc. It depends on the specific design.
     
  8. Armaegis

    Armaegis Friend

    Friend BWC
    Joined:
    Sep 27, 2015
    Likes Received:
    4,201
    Dislikes Received:
    2
    Trophy Points:
    113
    Location:
    Winnipeg
    There are mics you record with, and there are mics that you can use to club a drunken club goer off the stage and go back to singing and still sound half decent. Although they can make decent ones for the former, Shures are generally well known for the latter.
     
  9. Dr. S

    Dr. S Rando

    Joined:
    Aug 28, 2019
    Likes Received:
    9
    Dislikes Received:
    0
    Trophy Points:
    3
    Location:
    Germany
    For the sake of completeness, here is another plot showing the responses for the NTi M2230:

    [​IMG]

    For some designs the pressure and diffuse field responses are actually quite close to each other.
     

Share This Page