occasional VISA error -1073807346 (0xBFFF000E) with VS2008 and Measurement Studio

[Penny]

Hello,

I have a piece of C# code in VS 2008 that talks to a Lakeshore temperature controller via the serial port.  The VISA communication was automatically generated by Measurement Studio.  The code is set to take a temperature reading every few minutes.  Occasionally (once every few days), it crashes with the VISA error -1073807346 (0xBFFF000E).  The rest of the time it works fine.

I looked at other posts about this error in the NI Discussion Forum, and found a few that recommend reinstalling VISA.  Is this correct?  If so, would that mean doing a complete uninstall/reinstall of our current version, which is VISA 4.6.2?

Thanks for any advice.

Penny

[Michael_W.]

Hi Penny,

I have a few clarification questions for you to help my ensure my understanding.

Do you mean that you leave this program running all the time and every few days it crashes?  Or that you run this program often and it occasionally crashes when you start it?

Which forum posts are you referring to?

Uninstalling and reinstalling VISA certainly wouldn't hurt so you may want to consider starting there.

[Penny]

Hi Michael,

Thanks for replying.  We are running our application every few days, for 1-2 days at a time.  Sometimes it does not crash at all.  Other times it crashes after having run for several hours without any problems.

There is another post in the LabView forum that I have looked at:  http://forums.ni.com/t5/LabVIEW/Error-0xBFFF000E-VISA-or-Error-1073807346/m-p/519307 .  It references this NI link: http://digital.ni.com/public.nsf/websearch/25121C491778AD2086256FFE007CC361?OpenDocument .  I also looked on the rest of the internet, which is probably where I saw the suggestion to reinstall VISA.

We are also wondering if our anti-virus software might be causing the problem somehow, so I think we are going to disable it altogether and see what happens.

Thanks,

Penny

[Michael_W.]

Hi Penny,

Checking the antivirus is a good start.  I'm not sure how much onboard feedback your instrument provides but after the error occurs is there any indication on the device itself that it is in an error state?  This may help us discover what went wrong in the communication.

[Penny]

Hi Michael,

Just coming back to update this post - we fixed the problem by adding a 50 ms delay between serial commands.  Probably should have thought of that in the first place, but just thought I would update the post.

Thanks again,

Penny