Signal level is often not known.
For each acquired data point, I'm configuring multiple samplings of
the signal with each sample taken at one of the card's gain settings.
These samplings are taken close in time to each other by setting the
interchannel delay to 10 or 100 usec. Overall repetition period
between these multiple samplings is slow (10 Hz max). I then look at
each set of data acquired at all the gain settings to find the one
with highest gain that did not saturate. That data is the one that is
kept. So when the signal is small, high gain is used. When it's
large, low gain is used. I could just use input ranges of some
arbitrary values in the channel-scan list (say 0.1, 0.5, 1.0, 2.0,
5.0, 10.0 V) and check the acquired data to see if
it exceeds those
limits, but it'd be better to use the actual input ranges the card has
available.
On Wed, 24 Sep 2003 22:48:05 -0500 (CDT), mikeporter
wrote:
>Just a thought, but shouldn't you be setting the input limits based on
>the signal you're measuring?