Answer
8Sounds
like an interesting project! However, using
a sound card
to do the D/A and A/D is probably not going to work out
well, mainly because of the anti-aliasing/anti-imaging
filters found in the audio converters and the lack of
DC response. For an SEM, you want to be able to set the
beam position quickly, precisely, and repeatedly to a
particular spot and then read the target current and move
on. The huge delays of the low-pass filters (several tens
of sample periods, typically) are going to make it difficult
to correlate a beam position with a particular video pixel
value. If you slow down enough to make that work, then
the lack of DC response is going to kill your repeatability.
As
an experiment, just try hooking your sound card outputs
up to a scope configured for X-Y display and see if you
can generate the kinds of rasters you want. You’re probably
not going to be happy with what you see. You’ll want to
build or buy yourself a good laboratory type D/A-A/D board
that has high absolute precision and very little if any
filtering.
Contributor: David
Tweed
Published
September 2003