Taperssection.com
Gear / Technical Help => Microphones & Setup => Topic started by: moricle on June 10, 2015, 08:14:50 AM
-
I have read about people checking pairs of new microphones to ensure they have similiar sound/levels. I was curious what's a common way to do this check. I'm checking a new bought pair of Rode M5s. I assume I'd plug in the mics (in stereo at the same time or one at a time?) and then record something. Then check the wave forms and listen for any differences in volume? I appreciate any feedback!
-
Bingo, that's the way I'd do it. I'd set the Mics up as a pair in the typical way I'd use them for a show then record something from my stereo. I'd then look at the wave forms in Audacity, do a gain check individually on each channel to make sure they were as close to even as possible. There are probably better ways to do this, maybe recording a test tone but for quick & easy this is the way I'd do it.
-
The easy, practical approach is to record something and listen.
However, instead of setting them up in a typical stereo configuration I'd set them up so they are pointing the same direction with the capsules coincident. That way the sound reaching each microphone will be as similar as possible, making your comparative listen and visual level check for sensitivity across the two more fruitful. Another easy thing to listen for is differences in self-noise (level, 'texture', 'color'). With the gain cranked up the self-noise from both mics should be at about the same level and have the same smooth hiss without spittyness or popping.
If you want to get fancier, you can record something with the mics setup that way and compare their frequency curves. A frequency analyzer plugin (plenty of free VST analyzer plugins available) helps do that, and one that can be set to accumulate an average curve over time will make comparison easier.
With the mics setup that way you can also do the good old "null test". Record something. Make sure exactly the same amount of gain is applied to each channel by the preamp and/or recorder. Flip polarity on one channel, sum the two and see how well they cancel. The closer matched they are the more the common signal in both will cancel each other and the less sound you will hear. Listening to this 'difference signal' can be informative since you're hearing what the two recorded signals do not have in common with each other, so it can help identify the frequency regions where the two are not exactly the same. It's a good idea to listen this way while adjusting the gain of one channel, to see what maximum cancellation is that you can achieve. Doing that compensates for any differences in recording gain as well as differences in the sensitivity of the microphones. If the recording chain gains really were exactly identical, then the difference in gain required to achieve the best null (quietest summed signal result) indicates the difference in sensitivity of the two microphones.
-
Damn, I always come away smarter from reading a post by Gutbucket. Thanks.
-
Thanks, hope that helps.
Two clarifications-
When I mention comparing the averaged frequency curves of a recording made with the two microphones using a frequency analyzer (with the mics mounted so that their pickup would be as identical as possible), what you will see is not the on-axis anechoic frequency response curve of the microphone which one would hope to find on the manufacturer's specification sheet. The curve will not be anywhere near as flat and smooth. Instead, the curve represents the averaged frequency response of the recorded sound combined with the microphone's response curve. The important thing to look for is not the overall shape of these curves, that's irrelevant. Instead look for any differences between the curves from the two microphones. It's the differences between the two curves which indicates differences in frequency response between the two microphones. The closer the curves are to each other the better.
As for the null test, don't expect to get a complete cancellation null. Total cancellation would requires a completely error-free match. It's not going to happen even for perfectly matched top quality mics. Besides error from the mics not being perfectly coincident (which is physical impossible), there will always be minor differences in response between even superbly matched pairs. Just listen for the greatest extent of the cancellation while adjusting the gain of one channel and note the resulting gain differences. Most of what you hear after that adjustment will hopefully be rather whimpy, whispy stuff, significantly quieter than the sum of the two mics without polarity inverting one of them, or either on it's own.