AnsweredAssumed Answered

VRf-Timing_Problems

Question asked by VRFuser on Aug 17, 1997
send q VRf-Timing_Problems ww vrf heiko.degenhardt@rz.tu-ilmenau.de

from: Greg Goebel / HP-MXD
      gvg@lvld.hp.com / 970-679-3030 / FAX 970-679-5971
      website:  ftp://fcext3.external.hp.com/dist/mxd/index.html
to:   VRf-Heiko Degenhardt
date: Monday, 18 August 1997 1359 MDT

> Date: Mon, 18 Aug 1997 18:23:51 +0200 (MDT)
> From: Heiko Degenhardt <Heiko.Degenhardt@rz.tu-ilmenau.de>
> To: HP VEE Reflector <hpvxd_xc%hpislsup.lvld.hp.com@hplb.hpl.hp.com>
> Subject: VRf - Timing Problems (?) with Bus I/O Monitor and ID
>
> From:    Heiko Degenhardt   (heiko.degenhardt@rz.tu-ilmenau.de)
> To:      VRf   (hpvxd_xc%hpislsup.lvld.hp.com@hplb.hpl.hp.com)
> Subject: VRf - Timing Problems (?) with Bus I/O Monitor and ID
> Date:    Mon Aug 18 18:24:16 MET DST 1997
>
> Hi all!

> I have written an instrument driver (in the HP Instrument Driver
> Language) for a Keithly K2400 Source Measure Unit.
> The driver seems to work correctly.
>
> My problem is:
> For testing the driver I use a VEE program, that contains a "Bus I/O
> Monitor" object, to see what is sent over the HPIB.
> With this object I seem to get timing problems. It looks, as if the
> communication of the instrument and my driver gets corrupted some how
> and characters are lost during data transfer.
>
> As an example:
> If I send the command ":SOUR:MODE?" (and the instrument is set up to
> perform a "sweep" measurement), then the K2400 answers with "SWE".
> But sometimes the BUS I/O MONITOR shows only "SE" or "SW" as the answer
> of the instrument, so my driver dosn't understand it.
>
> The same is with data transfer to the instrument.
> For instance, if I send ":SOUR:MODE FIXED", then sometimes the Bus I/O
> Monitor shows ":SOUR:MDE FIXED" (missing an "O") or ":SOUR:MODE FXED"
> (missing an "I"), and the K2400 says "Data corrupted or stale".
> Which characters are missed seems to be accidental (and in most cases the
> data transfer works fine).
>
> I think that the Monitor object is working correctly and shows what is
> "really" sent over the Bus. But how (and why) are some charakters lost
> while sending them over the Bus? Can the monitor object cause speed
> problems between the instrument and the controller?

The monitor actually only sends what the system software thinks it has sent
over the HPIB, not the actual traffic over the bus itself.  This means that
it might not actually be sending what you think it is sending (though I
won't believe that it will send any *more* than what it thinks it's
sending).

Of course, anything returned has clearly been sent over the bus (though there
might have been more on the bus itself).

> With direct I/O (or even if I don't use a Bus I/O Monitor object)
> everything seems to work ok.
>
> I don't know how to fix this problem. Should I "slow down" my driver (may
> be with an delay after every transmission)?

The Bus I/O Monitor actually delays the data transmission.  I have never
seen a problem like this before (it's backwards to normal expectations ...
we have had problems where *using* the Monitor causes things to work).
The only thing I can speculate on is that it is causing bus timeouts.

> Has someone else seen such problems?

> My platform is:
> Computer: Pentium 90
> Memory: 32 MB
> Software: Win95, HP VEE 3.21
> HPIB card: HP 82341

> Any help is welcome.
> With kind regards.

> Heiko.

We did have some timing problems on the 82341 cards ... this doesn't sound
like the problem here, but you might as well get the patch off the website
(see header for website address) and see if that makes a difference.

If Direct I/O is working OK I don't really see an HPIB problem though.
Please let us know if the patch makes any difference.

[<>] regards -- gvg

Outcomes