Please see embedded comments below.
Dick
----- Original Message -----
From: Tony Duell <ard_at_p850ug1.demon.co.uk>
To: <classiccmp_at_classiccmp.org>
Sent: Tuesday, April 11, 2000 2:24 PM
Subject: Re: HPIB to HPIL adapter
> >
> > Well, I guess I'll have to look at the NI web site. I've looked for a
means
> > of deteriming basic requirements for LABVIEW drivers for instruments
(like
> > mine) which they don't support themselves. I have a TEK 'scope and
logic
> > analyzer, both GPIB capable, but for which there have apparently not
been
>
> Forget about Labview and its drivers...
>
That's my inclination as well. It's just that numerous local companies use
it in their testing and are willing to pay almost excessive (if it's what
they pay me it's not excessive) rates for people to write scripts and
drivers for Labview to drive their instruments. Many of them are already
supported, but the new drivers are worth something.
>
> I'm suggesting you go one level lower and talk to the instruments
> directly. That is to say that you send them a command (as in bytes over
> the GPIB interface ) and read data back. You're going to have to figure
> out what commands to send and what you'll get back.
>
Yes, that's described in the manuals. I expect there'll be plenty of
surprises, though.
>
> At one time this information was clearly presented in the user manual for
> the instrument. Nowadays you probably have to look in a 'programmers
> manual' or 'technical manual', but it should be available, especially
> for instruments from Tektronix or HP.
>
Well I have the doc's for my TEK instruments. The HP stuff may not be on
the web any more, but I imagine I can find it. The MSC HDC will need a
little work, but it is pretty simple.
>
> > drivers written. I've also got a GPIB HDC and one of those common HP
dual
> > 3-1/2" floppy drive boxes, either of which would probably serve well as
> > listener/loggers. It would be nice to be able to use that setup to
monitor
>
> You do realise that you're going to need a controller (== computer most
> of the time) on the GPIB. Linking the 'scope to the HP drive box without
> also linking in a controller won't do anything. I know for a fact that HP
> disk drives never operate as controllers and I doubt very much in 'scopes
> do either.
Oh, yes. The PC will do that, but at some point in the future, I might want
to make up a little controller board with the necessary firmware to make it
able to initialize and poll, etc, and then do some specific task. I still
have a couple of those 9114's from T.I.
>
> And if you have a computer on the GPIB, you might as well log the data to
> its disk drive.
>
It's not as portable, but if portability isn't a factor, that would be my
choice.
>
> > bursty events via the LA or 'scope.
> >
> > Is there anything worthwhile that doesn't involve National Instruments?
>
> Yes. Write the library yourself ;-). I've done it (not on a PC, though)--
> it's not that hard if you've got data on the GPIB card. All you really
> need (at least for simple stuff) is the equivalent of putc() and getc()
> for the GPIB bus and the ability to send controller commands (which for
> simple stuff translates to putc_with_ATN_asserted(). Maybe the
> ability to assert REN. Those 4 functions will let you talk to most GPIB
> instruments, at least for simple applications.
>
Somehow I think folks will be more interested in having this done via
Labview or some similar suite. I'm quite sure there are GPIB card vendors
for the PC beyond N.I. Surely there's something besides LabView that's just
as useful. If not, it will be necessary for me to look at whatever I can
find to enable me to prepare drivers for various instruments or, the
"generic" instrument, by using stubs for the appropriate device-specific
stuff. Writing test procedures in low-level/command line constructs is a
bit too primitive for the typical manager, and they're the one who sign the
check.
>
> -tony
>
Received on Wed Apr 12 2000 - 04:56:51 BST