Home Theater Forum and Systems banner

Typical distortion of mics measuring subwoofers

4K views 12 replies 5 participants last post by  bjs 
#1 ·
ricci said:
Very high volume close mic LF driver measurements are going to be rough on the mic and it will exhibit high distortion if you are up over 120db, which really isn't even driving the sub very hard at all when the mic is so close to the cone. It may affect the accuracy of the measurements adversely especially any distortions. I'd limit things to under 120db indicated by the mic. (Look for a test that Herb kindly did for me on the ECM8000 capsule). Some of the better Earthworks or AO Pacific mics would probably be fine, but not these.
I pulled this quote from the "mic damage" thread as it is a good introduction into the second part of my question which is typical distortion levels of the 6mm measurement microphones when measuring subwoofers (ie 10-100Hz).

I've been able to find very little information on this subject but am trying to get a sense of typical distortions in this scenario. Especially when used in a near-field mic arrangement where the levels tend to be quite high.

Here is the results that Herb provided Ricci. Very useful information!

anechoic said:
I just ran a quick test comparing the distortion of my one of my reference mics (ACO Pacific 7052, rated to 144 dB) to an ECM8000 at 1 kHz.

At 115 dB, my ref mic gets 0.13% distortion, the ECM8000 reads 0.54%

At 120 dB, my ref mic gets 0.22% distortion, the ECM8000 reads 0.81%

At 125 dB, my ref mic gets 0.46% distortion, the ECM8000 reads 1.37%

and,

Tests at 30 Hz:

At 115 dB, my ref mic gets 1.8% distortion, the ECM8000 reads 3.1%

At 120 dB, my ref mic gets 2.95% distortion, the ECM8000 reads 5.0%

At 125 dB, my ref mic gets 4.2% distortion, the ECM8000 reads 7.4%

FWIW, this is a different ECM8000 than the one used for the 1 kHz test.
Also ARTA has an application note (http://www.fesb.hr/~mateljan/arta/AppNotes/AP5_MikroMeasChamber-Rev03Eng.pdf) which includes this chart in the appendix:

Text Line Slope Font Parallel


Note the measurements are at 300Hz which is inbetween the data points Herb measured. The WM60 is presumably the precursor to the current crop of 6mm Panasonic capsules in mics from Behringer, Dayton etc.

Since my subwoofers usually show a significant increase below 30Hz I wonder about the distortion at 10Hz and 20Hz? Does distortion typcially increase significantly below 30Hz?

And I wonder if the native 6mm capsules inherently have much lower distortion than shown by the Behringer. I have an IBF Acoustic microphone that appears to have significantly less distortion than my Behringer for example. This raises the question if the Dayton mic may be lower distortion than the Behringer?

I also have a Nakamichi CM-300 mic rated at 3% at 140dB. Presumably rated at 1kHz (they don't say). It is a 12mm microphone. I've noticed it has significantly less distortion than my Behringer but more than my IBF Acoustic mic.

So, does anyone have data providing more information on this subject? Typical distortion levels of other mics relative to SPL and Frequency? Distortion patterns at 10 and 20Hz? Other comments?
 
See less See more
1
#2 ·
I suspect that microphone distortion will increase as the frequency decreases in much the same way that a subwoofer's does. It takes more air displacement for the lowest freq's (4x per each octave decrease) so it would stand to reason that the microphone diaphragm deflection will also increase causing higher distortion levels right? :ponder:

Hopefully someone like Herb with much more knowledge on the subject will comment.
 
#5 ·
Hopefully someone like Herb with much more knowledge on the subject will comment.

Unfortunately I don't have a whole lot of knowledge when it comes to discussing mic distortion other than that I know that the mic pre-amp can contribute a lot to distortion so for a given mic it may not be related to just the capsule.

I'll have to look through some of the ANSI and IEC standards documentation that I have and see what they say about distortion.
 
#3 · (Edited)
(EDIT: Posted prior to seeing Ricci's comment. Don't want it to sound like I view his contribution as "nobody"!)

Nobody has good info or measurement data to add? Oh well, I'll start answering my own questions then.:eek:


bjs said:
Since my subwoofers usually show a significant increase below 30Hz I wonder about the distortion at 10Hz and 20Hz? Does distortion typcially increase significantly below 30Hz?
I should be able to handle this one. :)

There is no physical reason why a microphone would suddenly increase distortion readings below 30Hz unless it started to roll off there and was being used uncalibrated.

On the other hand, my subwoofers are sealed and have their fb (box resonance) around 30Hz. Since they start to roll off at that point one would expect distortion to increase at rates similar to the rolloff (ie 12dB/octave).

Since I prefer to test than speculate, here's a quick test measuring distortion in the box (with fb=30Hz) and then free air (fs=20Hz).

First, the distortion in the box. I took this previously so please ignore the multiple lines showing the harmonics and look at the top THD line. As expected, distortion begins to rise quickly below the box resonance. Levels are 2-3% down to the 30Hz resonance and then rise quickly to about 7% by 20Hz as predicted.

Oscilloscope Text Line Plot Font



Now here is the same test but with the driver in free air. This drops the resonance to 20Hz and one can see that indeed, the distortion increase shifts to below 20Hz (note the plot starts at 10Hz instead of 20Hz as in the previous). The distortion by 10Hz is 16-18%, again as would one expect given a base level around 4% in the 20-30Hz region.

Text Oscilloscope Line Slope Plot



As an aside, these plots were taken at about 116dB. How much of the 3-4% distortion indicated between 20-40Hz is the driver and how much is the mic is unknown. But given the driver is swinging around 30mm peak to peak I have a hard time believing it has less distortion than the mic. If both were contributing half, that would leave the mic with 1.5-2% at these frequencies and levels. I suspect it is actually less.
 
#4 ·
Since I had the test setup, I decided to do one more experiment.

How much will the distortion levels change if I increase the level by 10dB (to about 126dB)?

I didn't want to change the subwoofer level and introduce another variable so I just moved the mic in from 20cm to 3cm from the dustcap (lesser of two evils in terms of variables). This gained almost 10dB.

The results are below:

Text Oscilloscope Line Plot Slope



Basically, it didn't change much.

This mic (IBF Acoustic Emm8) doesn't seem too fussed by these high levels.

My Behringer mic however sits in a dusty drawer. Past experience has shown it is not a happy camper when pushed like this.
 
#10 ·
Since I had the test setup, I decided to do one more experiment.

How much will the distortion levels change if I increase the level by 10dB (to about 126dB)?

I didn't want to change the subwoofer level and introduce another variable so I just moved the mic in from 20cm to 3cm from the dustcap (lesser of two evils in terms of variables). This gained almost 10dB.
That seems like the best way to characterize LF mic distortion, especially if it's done outdoors. Set the sub's SPL to 140dB or whatever close-mic'd. Gradually move the mic away and run a series of distortion curves until the SPL was say 80dB where mic distortion should be negligible and most of the measured distortion would be the sub itself. Subtract that 80dB curve from the others and you should have mic distortion.
 
#6 ·
RICCI shamed me into digging out my ECM8000 from the dusty drawer and repeating the above test for comparison.

Same setup as the 3cm plot above although due to a brain cramp in setting the drive voltage the ECM8000 data was taken about 1dB lower level. This would favor the ECM8000 slightly because the subwoofer distortion levels would be lower.

Oscilloscope Text Line Slope Plot


The levels are very close around 100Hz, off by quite a bit at 20Hz and way off by 10Hz. This even though the freqency response curves match extremely well.

What does it all mean? Who knows really. It's not much different than measuring frequency response with an uncalibrated mic, one can analyze things to death but without proper characterization via NIST traceable methods we're just guessing.

Since my IBF and Nakamichi agree with each other all the way down to 10Hz (<1%) AND they record much lower values than the ECM8000 I consider them accurate until proven otherwise! :hissyfit:
 
#7 ·
Wow...that sucks....for me. (I've only got ECM8000's at this point). Time to get something better.

Let me get this straight, at 1db lower drive level the ECM8000 shows 8.5% THD at 20hz vs only 3.8% for the IBF, at 30hz it's the same story at 5.5% for the ECM8000 and only 3.2% for the IBF. at 10hz the ECM8000 shows 57%!!! versus 21% for the IBF. I guess this could be attributed to a much faster roll off by the ECM8000 down low but still...
 
#8 ·
Yes, you've got it.

As for the faster rolloff...calibrated they both read the same and these measurements are with calibration. However uncalibrated the IBF is dead flat while the ECM is down about 9 dB at 10Hz, perhaps whatever cause the ECM to fall-off natively also affects it's distortion.
 
#12 ·
The faster roll off will cause the fundamental at say 10hz to read lower than 20hz or 30hz, causing the harmonics to be at a higher level relative to the 10hz fundamental, raising the THD reading. In theory the calibration should even that out, but I'm not so sure and there may be other effects of the steeper roll off that cause the THD to read higher that is unaffected by the cal file.
 
#9 ·
just out of curiosity, what kind of Preamp are you guys running while doing these tests? Some aren't as clean as others, and can add their own noise...

The Behringer mics I'm only vaguely familiar with (used the ECM8K's once) but there are others that would be a lot better (and have a heftier price tag).. DPA 402x (I cant remember the model on the omni), AKG 460/480 with CK-62 capsule, Studio Projects C4 w/omni..etc.

I'm coming from a live taping background, where we're all about the gear used, which is why I inquire.. :)

-Jim
 
#13 ·
Jim: I'm using the Behringer Xenyx802 mixer as the pre-amp. What is the cheapest mic you would recommend (flat FR and ultra LOW THD)? The ones you listed or something else?

Dennis: Yes, I couldn't think of a better method. Indoors the range of SPL is too limited to be useful though so I'm waiting on good weather!

Ricci: Yes, calibration could have side-effects. Increase in mic noise floor for example (eg correcting -10dB at 10Hz on a 50dBA mic would make 1% the lower limit at that frequency due to mic noise alone). And there could be issues with the software interpolation of the mic calibration file at the boundaries (ie 10Hz). Lot's of potential issues for sure.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top