[Top] [All Lists]

Re: [laptop-tapers] RE: freq response and aliasing of Edirol FA-10=

Subject: Re: [laptop-tapers] RE: freq response and aliasing of Edirol FA-10=
From: Gianni Pavan <>
Date: Thu, 12 Aug 2004 23:41:36 +0200
1 at 192 kHz

Hi all,
        recording at 192 kHz is very useful when studying animals emitting their
communicative or echolocation signal at frequencies higher than those we
can hear. Mice, rats, bats, insects do emit sounds up to 100 kHz, but some=

emits (ultra)sounds ranging up to 120 kHz. Many others emit audible sounds=

with harmonics extending to more than 20 kHz.
To record those sounds it is required to sample at very high rates with
industrial sound acquisition boards that are expensive, require specific
drivers, and often have 12 or 14 bits only.
By using a very good board with 192 kHz sampling you can spend a lot less,=

you can use standard windows software and you get at least 16 bits of
resolution. The LynxTwo board allows to work reliably up to 90-92 kHz. This=

means you can record most (but not all) of the ultrasounds emitted by
terrestrial and aerial mammals. With marine mammals you can get a
significative portion of their echolocation sounds that may extend up to
160 kHz.
Then, if you want to hear those sounds you can simply playback the files at=

reduced rates, 48 kHz for example.

I was hoping to find a portable device to setup a field recording, laptop
based workstation able to capture sounds up to 90 kHz.
Unfortunately the FA-101 is not as good as the PCI board LynxTwo. The data=

I presented about the FA-101 has two meanings:

1 - the frequency response is not flat: -23 dB at 90 kHz is a lot because
signals at that frequency are weak by their own and the transducer are not=

very sensitive and often are very noisy.
Though, the attenuation could be compensated by equalization if the
anti-alising filtering is good.

2 - unfortunately the a-a filter is very soft and signals or harmonics
extending over the Nyquist frequency, that is half of the sample rate, can=

be "reflected" below Nyquist and "contaminate" the recording with artifacts=

Now I'm willing to know how the other 192 kHz devices behave.
Testing the frequency response and the a-a filters is quite easy if you
have a signal generator ranging up to 200 kHz!
Any measure on other devices will greatly appreciated!


Centro Interdisciplinare di Bioacustica e Ricerche Ambientali
Universita' degli Studi di Pavia


<Prev in Thread] Current Thread [Next in Thread>

The University of NSW School of Computer and Engineering takes no responsibility for the contents of this archive. It is purely a compilation of material sent by many people to the naturerecordists mailing list. It has not been checked for accuracy nor its content verified in any way. If you wish to get material removed from the archive or have other queries about the archive e-mail Andrew Taylor at this address: andrewt@cse.unsw.EDU.AU