R/E/P > Klaus Heyne's Mic Lab

Capsule and microphone distortion

(1/2) > >>

josephson:
An interesting thread in the old forum should be resurrected soon per Klaus, but since he mentioned my name (besides making mics I'm chair of the AES standards committee working group on microphones) in regard to standardized measurements of distortion, I thought I would start it up again here.

The answer is no, all microphone measurements compliant with IEC 60268-4 are to be made of the entire microphone including the capsule. The idea of making distortion measurements with a "substitute capacitance" instead of the capsule is useful perhaps for troubleshooting or establishing clipping levels but not for characterizing a microphone, particularly not when the electronics are seldom the dominant distortion producer. Unfortunately, as with most of the 268-4 standard, few microphone makers use these measurements in their data sheets.

Measurement of microphone harmonic distortion is troublesome in the real world -- it's bad enough with an omnidirectional (pressure) microphone and nearly impossible to achieve meaningful results with a directional microphone without a lot of caveats about the measurement conditions.  And harmonic distortion is certainly not the only kind. The method of difference frequency distortion measurement described by Peus and others (and it is in the standard too) might intuitively seem to be less related to real world problems, but in fact the same nonlinearities in diaphragm movement alluded to by Klaus, Oliver and others also show up in DFD or IM distortion. There are better methods, but even a plain IM distortion measurement shows the same bad actors that produce distorted sound (by my definition this means signal components triggered by or related to the program material, that aren't in the original sound) with music. There are plenty of known distortion generators in the capsule, the electronics, and the interface between them, so it's amazing that condenser mics work as well as they do.

Ultimately microphone performance measurements are only as useful as microphone customers demand them to be. Microphone makers generally believe that the vast majority of buyers have no clue about or respect for measurements, and so many specifications given over the years have been so much bunk so it's no wonder. The 268-4 standard only specifies how measurements are to be made, there is very little about what that means when a datasheet claims certain performance. We did get a tolerance requirement into the last revision for on-axis frequency response (not that any but the highest level of companies paid any attention to it) and in the next revision which we're working on now might well have more guidance about what is printed in data sheets.

--
David Josephson

Jim Williams:
Long overdue IMO. Besides those specs, a THD vs frequency plot is also needed. I have found THD varies in many mics depending on the frequency measured, especially in the low end. There was an imformative article several years ago in the AES Journal about the effect of loading on the THD of mic capsules. What was found is capsule loading resistance directly affects the THD at low frequencies.

In order to achieve .001% at 20 hz, a loading resistor of 10 GIG ohms was required. Needless to say, most mics don't have that value, most are at 1 gig ohms. THD will be ok at 1k hz, but will degrade at 20 hz. The AKG 460 and 480 models have that 10 gig ohm load resistor and I find those mics to be rather clean in the low end.

klaus:

--- Quote from: Jim Williams on March 11, 2011, 10:47:37 AM ---I have found THD varies in many mics depending on the frequency measured, especially in the low end. ...
In order to achieve .001% at 20 hz, a loading resistor of 10 GIG ohms was required.
--- End quote ---

Jim, how did you measure such miniscule level of distortion, especially in light of David's statement:

--- Quote ---Measurement of microphone harmonic distortion is troublesome in the real world
--- End quote ---
?

Barry Hufker:
I hope this is a related question.  Why is it that when discussing microphone distortion and frequency response, phase response is never measured or discussed?

klaus:
My valued colleague Scott Hampton and I worked on a test set up for way too many hours a couple of years ago, only to finally realize that the "noise" of the equipment was too much for any reliable measurement of such an extremely subtle factor to come through and show up.

So, give your ears some credit: they sense phase aberrations when machinery fails.

Navigation

[0] Message Index

[#] Next page