testing and measuring a microphone?

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
I guess there are many mic specs that could be the target of 'testing' a mic.

I was going to try and cross-check my two reference mics for starters.  They are ok at the calibrator frequency for both expected signal level of the fundamental, and show the same harmonic structure of the calibrator, so that's a start.  It may be easiest to just close mic a woofer, then a tweeter, and compare responses, and hopefully they are both the same (otherwise I won't know which mic is NQR).  Hopefully I can get a valid measurement over a fair audible bandwidth, as the mic specs go well beyond what I can set up - even if I can get a good comparison over say 200Hz to 5kHz then that would give me confidence that the response would  be to spec.
 
trobbins said:
I guess there are many mic specs that could be the target of 'testing' a mic.
Yes; I forgot to mention that evaluating the mic's own noise requires a superior S/N ratio of the measuring set-up.

I was going to try and cross-check my two reference mics for starters.  They are ok at the calibrator frequency for both expected signal level of the fundamental, and show the same harmonic structure of the calibrator, so that's a start.  It may be easiest to just close mic a woofer, then a tweeter, and compare responses, and hopefully they are both the same (otherwise I won't know which mic is NQR).  Hopefully I can get a valid measurement over a fair audible bandwidth, as the mic specs go well beyond what I can set up - even if I can get a good comparison over say 200Hz to 5kHz then that would give me confidence that the response would  be to spec.
Comparative method is OK for evaluating similar mics, i.e. omni vs. omni, card vs. card, fig8 vs. fig 8, but comparing a directive mic vs. a typically omni reference mic exposes to errors particularly at LF and HF.
 
We measure our production mics using the Farina exponential sin sweep deconvolved to gated impulse response. A special
low diffraction speaker on a pole is used. We use B&K lab mics as reference.

I should mention the difficulty in measuring low frequencies in directional mics. As mentioned gated impulse response
loses resolution at low frequencies...in a typical room you can only go down to about 200 Hz.
Near field testing will work for LF with omni and figure 8 (compensating for proximity effect) But this will not work
for cardioid because pattern and therefore proximity is a process variable so you can't easily calculate it.

In general you need a plane wave tube.
They are huge though. I think our PWT at Shure was about 20 meters long.

So a PWT is impractical even for most manufacturers.
But I have modeled an active termination PWT that can be very short, only a meter or less. The electroacoustic analog
shows a perfect plane wave down to 20 Hz. I haven't built it yet...I hope the model is correct...I'll see soon.

Les
 
"So a PWT is impractical even for most manufacturers.
But I have modeled an active termination PWT that can be very short, only a meter or less. The electroacoustic analog
shows a perfect plane wave down to 20 Hz. I haven't built it yet...I hope the model is correct...I'll see soon.
Les"

Schoeps' tube is 30m underground and subject to "environmental concerns".
Active PWT has been on many drawing boards for years.  I'm curious about your success on this.
 
Back
Top