> I find it truly ironic that this post has the heading =B3MiniDisk data
transfer
> and bat recordings=B2
>
> It=B9s been reported that bats make sound well over 120khz. And since you
need
> twice the sample
> rate to capture a frequency....then, oh damn, 192k sample rate only gets
us up
> to 96k. I guess we=B9ll
> have to wait until the next generation of records to truly capture those
> =B3nature sounds=B2. I guess 192 will do
> for now.
Hi Andy,
yes even a sample rate of 192 kHz might be too low for direct recording bat
calls with frequencies up to 96kHz. The problem is, that depending on the A=
/D
converter, the properties of the integrated anti aliasing ailter (AAF) will
prevent to use the theoretically possible frequency range up to half the
sample rate. In practise, the maximum frequency range that can be recorded
without potential aliasing effects would be about 80 kHz in 192 kHz recorde=
rs. If
you need higher rates, then have a look at specific devices for ultrasound
recording:
http://www.ultrasoundgate.com
However, this thread 'MiniDisk data transfer and bat recordings' aimed at a
different approach to record bat sounds. There are various bat detectors th=
at
allow to convert the ultrasonic bat calls into audible low-frequency signal=
s
that can be recorded using common recording equipment. One commonly used
technique is the time expansion principle. These devices can record short b=
at
call sequences (about 3 seconds) at high enough sample rates (e.g. 300 or 4=
00
kHz) onto a solid state memory. These digitally recorded sounds can then be
replayed at a slower speed (e.g. at a playback rate of 30 kHz or even only =
15
kHz). The resulting signals can then be recorded onto a MD recorder. Due to
the low dynamic range of these time expansion bat detectors (8 bit AD, DA
conversion only), the limitations of the ATRAC compression should not becom=
e
visible. Another issue that prevents the occurrence of unacceptable ATRAC
artifacts is the fact that the slowed down signals will not have much energ=
y at
frequencies higher than 10 kHz. Most bat calls have their maximum energy at=
around
30 to 70 kHz (3 to 7 kHz in the time expanded signal). Additionally, the
electret microphones used in the cheaper bat detectors have a decreasing
sensitivity at frequencies over 100 kHz.
As you see, there are no simple answers. It always depends on what you are
doing with our equipment and what are your goals in a specific application.=
We humans usually tend to religious explanations as long as we do not fully
unterstand a thing...
Regards,
Raimund
--
+++ GMX - die erste Adresse f=FCr Mail, Message, More +++
Bis 31.1.: TopMail + Digicam f=FCr nur 29 EUR http://www.gmx.net/topmail
________________________________________________________________________
________________________________________________________________________
|