> 3) does anyone have a link to a really, really good & unbiased study
> on the links between self noise & other factors in terms of the
> actual performance of the mics ?
Jez,
No, so I'll have to write one. :-)
Here's a draft. Quoting noise figures is fraught as there are no
universally accepted standards fo go on. "Self noise" is often
misquoted, but should mean the added noise caused by the microphone
and also the input stages. Yes, the impedance matching of these is
also important. Note self noise is the added noise, not the resulting
total noise, one common fudge amongst others.
"Excess Noise" is the relative noise level over what you would hear
from the thermal noise of the air. "Excess Input Noise" is this noise
referred as an equivalent input sound level in excess of the basic
thermal noise of the air and this is where the problems start. The
theoretical minimum is 3dB because air molecules are vibrating against
both sides of the diaphragm. The world record claimed is +5dBs with a
B&K instrumentation mic you would not want to use for general
recording.
The biggest problem is that thermal noise rises with frequency at 3dB
per octave (aka root2 noise) so you can't even quote a thermal noise
level without specifying frequencies.
The answer is to use a weighting curve which follows the average ear's
perception of low level noise which is the ITU-R 468 curve. (It's on
my web site) This gives "uninspiring" figures which the ignorant don't
understand, so many manufactureres give the A-Weighting figure which
is rubbish. Sorry to be blunt, but A-weighting is for medium to loud
sounds and even then is not accurate to what we hear.
The only realistic noise test it to plug various mics into your own
input impedances and listen. Alternatively, if you nave a vacuum pump
and mics which will take it, you can measure the actual levels of air
thermal noise.
David
David Brinicombe
North Devon, UK
Cogito cogito ergo cogito sum - Ambrose Bierce
|