> Directional mics are more sensitive to handling noises. (and wind
> noises). And super-cardioid is very directional, so yes, it is to be
> expected.
James,
That's the big snag, and a perfect omni mic would give zero handling
noise, but we need directional mics. You need a high pass (bass cut)
filter to cope with this and a windshield is basically that, starting
with zero frequency. However, if the windgag is itself vibrating, even
if the mic inside is on a resilient mount, you will get handling noise
from the windgag.
Here are some basic principles of windgags/shields:
High attenuation of zero frequency. In other words, you can't blow
through the shielding fabric.
A bass roll-off HPF material which attenuates turbulence frequencies.
This turbulence is caused by the presence of the windgag itself,
mounts, stands or other close objects including shrubbery.
Combine the last two by choosing a close weave material, but see
below.
All shapes presented to the wind should have as large as practicable a
radius of curvature, in other words avoid any sharp angles which could
cause local turbulence. The ideal would be a large sphere unless this
tended to move around in the wind. If you really must have the theory
look up Wiki Reynolds number, but I'd advise just sticking with a
minimum
radius of curvature - say a radius of curvature of 50mm to 100mm.
Consider windgagging any close stand or mount if it is likely to
generate turbulence noise. Try probing around with an omni
(windgagged) mic for hisses and whistles.
One way to keep the surface smooth is to use a fluffy material or the
fake fur used by Rycote for their "Yeti" covers, with or without a
close weave inner layer.
The effective thickness of any windgag material (or combination)
should be smaller than the highest frequency recorded. The effective
thickness of fake fur is smaller than the actual thickness, but I've
also used brushed acrylan "romper suit" fabric as a windsock material.
The wavelength at 10KHz is 34mm so aim at 1/5th or 1/10th of this. Use
trial and error and you can simply hold up a piece in front of any mic
to test HF attenuation.
If you have a quiet HF background, make sure the system noise is not
boosted over this level by HF response correction.
A close weave material will not necessarily attenuate the sound, but
it will cause longitudinal polarisation. Forget the theory, but it
interferes with phase tube (gun mic) directors because it presents a
reflective surface. Try to avoid being close and parallel to the mic
which will affect the directional response..
A bass roll-off like 6dBs per octave goes a long way to preventing
bass overloading of the recorder which is ruinous, and can be
corrected afterwards with little penalty. Alternatively or in
addition, peak low. With digital you can make up 12 or even 18 dbs and
get away with it.
That's enough. :-)
David
David Brinicombe
North Devon, UK
Cogito cogito ergo cogito sum - Ambrose Bierce
|