Content-Transfer-Encoding: 8bit
Message-ID: <5.1.0.14.0.20020319163742.00a84aa0@gemini.herts.ac.uk>
Date: Tue, 19 Mar 2002 17:05:46 +0000
To: rsgb_lf_group@blacksheep.org
From: "James Moritz" <j.r.moritz@herts.ac.uk>
Subject: Re: LF: Soundcard Mods
In-reply-to: <7D653C9C42F5D411A27C00508BF8803DCB9A09@pdw-mail-r1.dstl.go v.uk>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii; format=flowed
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Sender: <majordom@post.thorcom.com>

Dear Andy, LF Group,

At 09:59 19/03/2002 +0000, you wrote:
>Has anyone tried changing the clock source on a Soundcard for a different
>frequency ?  For a  job here at work we need to sample more accurately than
>the usual crystal oscillator (does this sound familiar ?) and at different
>sampling rates.  It would be ideal if we could push it to 50kHz...

I remember looking at sound card specifications a year or so ago, and 
finding that current sound card audio codecs mostly adhere to the Intel AC 
97 specification (there are a number of different versions of this 
"standard" of course...), which includes a 48kHz sample rate, so 50kHz 
should not be pushing it much. You can download it from 
http://developer.intel.com/ial/scalableplatforms/audio/ - it contains a 
load of info about pin functions etc.

I skimmed through the data sheets for a few compliant devices - although 
they offered the same basic sample rates, the way different rates were 
implemented was different in every case - some actually had synthesised 
clocks that could be tuned in small increments, others ran at a steady 
48kHz, and messed around with the data in the digital domain - the upshot 
of this was that the ratios between sample rates were not necessarily exact 
- so you might end up needing a different clock frequency for each sample 
rate you planned to use.

Cheers, Jim Moritz
73 de M0BMU