A/D board for 3b2/300?

[larry at kitty.UUCP]

In article <698 at bgsuvax.UUCP>, duncan at bgsuvax.UUCP (Comer Duncan) writes:
> 
> We have a couple of 3b2/300s and desire to use one for
> interfacing to experiments.  I am interested to know
> about any and all A/D boards which are known to WORK
> in a 3b2/300.  Our requirements are quite standard--
> 16 bits, not super speed but reasonable.
> 
> I would like to know the properties and costs of such
> boards as well any REAL experiences anyone has had
> with utilizing them in laboratory environments.

	My organization has done an extensive amount of analog signal
processing on 3B2's for scientific instrument and process control
applications in the past 2-1/2 years.  All of this signal processing
is in "quasi" real-time.
	We perform A/D, D/A and parallel I/O interfacing using external
devices which connect to the serial ports of the 3B2.  We use external
data acquisition systems such as the Burr-Brown PCI-3000, and interface
them to the 3B2 at 9600 baud.
	We also design our own microprocessor-based interface devices which
consist of an A/D converter with mux, microprocessor, buffer memory, and
serial port.  These interface circuits are comparatively simple and
inexpensive to design.
	Unfortunately, I know of no off-the-shelf analog signal acquisition
device which plugs directly into the 3B2 data bus.  I have toyed with the
design of such a device from time to time, but have elected not to go that
route for the time being.  One reason is that the available I/O board
area and outside-world connector-space is limited.  Another reason is the
effort involved to assure the Grand High Exaulted Mystic Rulers of AT&T
that our board design won't damage customer's 3B2's and void warrantys.
	While I would like to see someone place such a board on the
market as a product, I suspect that it is unlikely due to the comparatively
few 3B2's used for this type of application.  *I* think the 3B2 has some
good points for this type of application, but, well, the rest of the world
doesn't necessarily agree with me...

<>  Larry Lippman @ Recognition Research Corp., Clarence, New York
<>  UUCP:  {allegra|ames|boulder|decvax|rocksanne|watmath}!sunybcs!kitty!larry
<>  VOICE: 716/688-1231        {hplabs|ihnp4|mtune|seismo|utzoo}!/
<>  FAX:   716/741-9635 {G1,G2,G3 modes}    "Have you hugged your cat today?"