In search of low distortion omnidirectional microphones for DIYers

Jump to Latest
#161
@b_force

Honestly, for me, I thought ARTA/STEPS was more of a challenge to set real SPL readings...
This is how I set my SPL levels in REW:

Click SPL meter:
1724309224231.png

Select SPL, Z, Slow:
1724309274434.png


Click Calibrate:
1724309354651.png


choose "Use an external signal"
1724309344627.png


Connect your (B&K) SPL calibrator to your connected microphone and play the test tone eg. 1000Hz, 94dB. Then hit OK:

Type SPL level of your test signal in the box, then click FINISHED:
1724309420110.png



Is this what you meant by SPL calibration with REW?

Or do you mean calibration using a frequency response data files to correct for non-flat frequency responses that many microphones, amplifiers have, as explained by @JohnPM in post #158?

In practical terms, I think we all agree that actual SPL values are not important in loudspeaker design. But it is nice to be able to compare measurements, at least within a dB or so. But when taking distortion measurements it's absolutely critical to equalize and match the SPL of the fundamental...
 
#163
tktran303 said:
Type SPL level of your test signal in the box, then click FINISHED:
Click to expand...
Yep, that's exactly the procedure I did, and the values are totally off.

tktran303 said:
In practical terms, I think we all agree that actual SPL values are not important in loudspeaker design. But it is nice to be able to compare measurements, at least within a dB or so. But when taking distortion measurements it's absolutely critical to equalize and match the SPL of the fundamental...
Click to expand...
For me it's needed when I go back to data months later or when clients want to get data published.
 
#169
Odd. That means the calibration process didn’t set the SPL. I’d consider uninstalling/reinstalling REW when funny things like that happen.
Though delving deeper is probably beyond the scope of this thread.
@b_force Consider posting a query at the official REW forum if you’d like some troubleshooting help.
 
#170
b_force said:
levels from measurement don't line up with the actual SPL
Click to expand...
What kind of measurement? SPL meter? Sweep? RTA? Fractional octave RTA measurements have an "Adjust RTA levels" appearance option for easier comparison with sweep measurements, help for that is copied below. That doesn't affect the overall levels shown in the levels panel, which has unweighted, A-weighted and C-weighted figures for the range specified by the distortion LP and HP settings, if used. Sweep measurements don't have an SPL as such since the result is a transfer function, but REW draws the transfer function at the level of the stimulus as a convenience. SPL meter readings are only limited by the accuracy of the calibrator and the mic response file, if used.

Adjust RTA Levels​

The RTA plot shows the energy within each octave fraction bandwidth. As the RTA resolution increases, from 1 octave through to 1/48 octave, the octave fraction bandwidths decrease and, for broadband test signals such as pink noise, the energy in each octave fraction decreases correspondingly. Whilst the RTA is correctly showing the actual level within each octave fraction, this variation of trace level with RTA resolution can be awkward when using the RTA with a pink PN noise signal to adjust speaker positions or equaliser settings. The Adjust RTA Levels option offsets the levels shown on the RTA plot to compensate for both the bandwidth variation as resolution is changed and the difference between a sweep measurement at a given sweep level and a full range pink PN RTA measurement at the same level, allowing direct comparison between RTA and sweep plots. Whilst the levels shown are not the true SPL in each octave fraction, they are more convenient to work with. N.B. This option should only be used with broadband test signals, such as pink noise or pink PN.
 
#171
tktran303 said:
Odd. That means the calibration process didn’t set the SPL. I’d consider uninstalling/reinstalling REW when funny things like that happen.
Though delving deeper is probably beyond the scope of this thread.
@b_force Consider posting a query at the official REW forum if you’d like some troubleshooting help.
Click to expand...
Very odd, but I am curious if other people actually compared the levels from REW against other things?
 
#172
JohnPM said:
What kind of measurement? SPL meter? Sweep? RTA? Fractional octave RTA measurements have an "Adjust RTA levels" appearance option for easier comparison with sweep measurements, help for that is copied below. That doesn't affect the overall levels shown in the levels panel, which has unweighted, A-weighted and C-weighted figures for the range specified by the distortion LP and HP settings, if used. Sweep measurements don't have an SPL as such since the result is a transfer function, but REW draws the transfer function at the level of the stimulus as a convenience. SPL meter readings are only limited by the accuracy of the calibrator and the mic response file, if used.
Click to expand...
That's a lot of text and none of it really matters.
Al that is needed is to simple measure the loop sensitivity of a black box.
Or in other words, under the same setup conditions and if my output signal is the same between REW and any other programs, the SPL must be the same.

Since a mic calibrator outputs a 1kHz signal any kind of weighting isn't relevant.

Anyway, I am most certainly not gonna discuss the entire procedure here.
That is going way offtopic, I was just curious if any other people actually compared those results.
But it sounds like most assume the levels are correct, unless I am misreading things.
 
#173
I can confirm it was done and compared with other microphones.
In my microphone testing, in this economy, I had much luck to beg, borrow, buy the following:

miniDSP Umik-1 (x2, older microUSB models)
Sonarworks Xref20 (older pre-2015 model)
Merry Electronics MMA209-003 (MEMS) (2024 purchase) x4
Line Audio Omni 1 (x2, 2024 purchase)
Sennheiser MD42 (purchased 2nd hand ?year model)
Earthworks M23 (2024 model)
Brüel and Kjær 4133 (1982 model) + Brüel and Kjær 2669L preamp + Brüel and Kjær 2690 signal conditioner

All were used to measure a single driver. The SPL had to be checked and calibrated to within +/- 0.1dB using a combination of hardware and software calibration. Only in this way can the harmonic measurements be fairly compared.

I did this in REW and then in STEPS.

Whatever issue you are having can be resolved, but it is beyond the scope of this thread to determine what the issue is. Come check out this help forum here: https://www.avnirvana.com/forums/official-rew-room-eq-wizard-support-forum.10/

As usual, most of my difficulties were later found to be user error, but many helpful people over there have made the same mistakes and learnt from them and explained how to move forward. Conversely, it's a two way street, and the designers of the software make changes to the most commonly/repeated faults. This is how user interfaces (UI) and user experiences (UX) evolve.
 
Last edited:
#174
b_force said:
Very odd, but I am curious if other people actually compared the levels from REW against other things?
Click to expand...
I've compared REW SPL readings to Smaart, ARTA, Open Sound, and Crosslite+.
All of them read identical, ime. As they all adhere to same standards, I guess.

I find REW is the quickest/easiest to calibrate SPL of any of them, using a mic calibrator as per tktran's post #161.
REW also has the unique capability to read peaks to the sample, which makes it the fastest peak capturing meter I know of, maybe matched only by NTI etc..
 
  • Like
Reactions: b_force
#175
mark100 said:
I've compared REW SPL readings to Smaart, ARTA, Open Sound, and Crosslite+.
All of them read identical, ime. As they all adhere to same standards, I guess.
Click to expand...
Thanks for confirming, that's actually very useful! 🙂
I will just have a look again with a fresh install.

Those steps as before was exactly what I did as well, seemed very straightforward, so I have no clue what went wrong. 🤷‍♂️

Anyway, enough offtopic.👍
 
#176
Hörnli said:
Your displayed work with REW took me months, atop the wasted years with REW and microphone distortion from others at diyAudio.
Click to expand...

Ultimately I hope to compare and contrast how an omnidirectional mic, like the dynamic Sennheiser MD42, or electret condenser Line Audio Omni 1, comes to a 200V externally polarised scientific instrument like the Brüel & Kjær 4191/2669.

The red pill might taste bitter initially...















spoiler alert...
 
Last edited:
#177
Let’s test the PTT6.5W04 again, using the 200V externally polarised microphone:
IMG_1686.png



Some commentary is appropriate, I believe, to aid interpretation. Note the brown line which the noise floor of the environment.

Notice how the higher order harmonics are all bunched together- they all seem to around the same level.

Let’s see that again in terms of dB down, relative to the fundamental:
IMG_1688.png

It appears the harmonic distortion noise floor is about - 80 to -90dB.


How does this compare to other measurements?

Here is the gorgeous Buchardt 10th Anniversay speaker, using a “Mid / woofer 1 x 6.5” PURIFI Extended Stroke Driver with Ultra Low Distortion, Custom made with triple Voice Coil”
IMG_1747.jpeg

As measured by @bikinpunk

Reference:
https://www.erinsaudiocorner.com/loudspeakers/buchardt_a10/


2 measurements are shown; at 86dB and 96dB average. Because the bass tuning is different, we should not pay much attention to what’s happening below 40Hz. The tweeter also takes over above 2KHz, so let us focus on the region 40Hz to 2KHz.
IMG_1687.png
IMG_1665.png



Notice how harmonics seem to be about 6-10 dB higher in level, and the higher order harmonics "only" seem to be around -70dB...
 
Last edited:
#179
What about other speakers measured with the Klippel NFS. From fellow countryman Alan March:

IMG_1752.jpeg


https://marchaudio.com/product/sointuva-awg-stand-mount-speaker/



IMG_1750.png

Reference:
https://www.erinsaudiocorner.com/loudspeakers/march_audio_sointuva/

Or the one with the aluminium cone PTT6.5NAA?

IMG_0622.png
IMG_1748.png

Reference:
https://www.erinsaudiocorner.com/loudspeakers/march_sointuva_awg/


And finally one reviewed/measured at ASR, with the PTT6.5X mid woofer in a 1/4 cu foot cabinet and SB Acoustics 5x7” passive radiator:

IMG_1751.png


Reference:
https://www.audiosciencereview.com/...r-directiva-open-source-speaker-review.27094/
 
Last edited:
  • Like
Reactions: lrisbo
#180
So what is going on here?

To understand we need to go back to the microphone- that is, the capsule and the pre-amplifier.


1726038150961.png


1726038176386.png


Let us extend this the 2nd harmonic distortion line to the left:
1726045077653.png

We see that the 0.01% (-80dB) distortion point is ~116dB for the 2nd harmonic.
Adapted from:
Brüel & Kjær Microphone Handbook,
Figure 5.17, Chapter 5

​

What about the 20KHz 1/2" microphone?
Here is the type 4189, "probably the world’s most popular ½-inch free‐field microphone"
Reference: Brief history of the B&K microphone

1726045873048.png

Adapted from:
Brüel & Kjær Microphone Handbook,
Fig 3.17, Chapter 3:​

Reference:
Brüel & Kjær Microphone Handbook
https://www.bksv.com/media/doc/ba5105.pdf


TBC...
 
  • Like
Reactions: Deleted member 375592 and lrisbo
You must log in or register to reply here.
Top Bottom