Jump to content


Photo
- - - - -

GPIB controlled Fluke timeout errors


  • Please log in to reply
2 replies to this topic

#1 longzoo

longzoo

    Active

  • Members
  • Pip
  • 15 posts
  • Location:Johnstown, PA
  • Version:LabVIEW 2011
  • Since:2004

Posted 23 May 2011 - 11:17 AM

i believe i had this posted in the wrong category last week so i am reposting here.
i am troubleshooting a test set that has, among many other instruments, three Fluke 8846A multimeters being controlled by GPIB. I am running LabVIEW 2009SP1. The canned vis from Fluke threw many errors so i decided to go as simple as possible and send the following commands from a subvi:

:TRIG:DEL:AUTO ON;:TRIG:SOUR IMM; :TRIG:COUN 1; INIT; FETC?

These commands set auto trigger delay on, triggered immediate, set trigger count to 1, sample count to 1, initiate, then got the reading.
i then read 64 bytes back and converted it to a double.

i call this subvi a few hundred times in my main routine. many times i get completely through the main program without issues but occasionally i will get the attached errors (see jpg). this needs to be an automated test set that runs overnight wihtout human intervention. does anyone know a way i can auto-acknowledge these errors to allow the program to run? or better yet, anyone have ideas on why i randomly get this error to begin with??

i have attached my subvi as well.

Attached Thumbnails

  • flukeError.JPG

Attached Files



#2 EricLarsen

EricLarsen

    Very Active

  • Members
  • PipPipPip
  • 71 posts
  • Version:LabVIEW 2012
  • Since:1992

Posted 23 May 2011 - 01:57 PM

The first thing to try is disable automatic error handling in your sub-vi. Wire the error cluster out and check it in your calling vi. If the instrument reports an error, try calling this vi again until you get no error.

But like you said, your best bet is to eliminate the source of the error. I've found with instruments like this it's best to limit the number of commands sent to it. It just seems like you can overwhelm it and eventually scramble its brains. I'm not familair with this instruement, but do you need to send the set up commands every time? The trigger and sample count commands may only need to be sent once to set up the instrument, then just send the fetch command every time you need a reading. Also, putting a slight delay (50ms?) between commands can help, too. It gives the unit time to play catch up. It may take some trial and error to make it preform like you want.



#3 longzoo

longzoo

    Active

  • Members
  • Pip
  • 15 posts
  • Location:Johnstown, PA
  • Version:LabVIEW 2011
  • Since:2004

Posted 08 December 2011 - 02:37 PM

Eric, I haven't been on LAVA in awhile but I posted elsewhere today and when browsing my attachments I saw that you had responded MANY months ago. That project was "mothballed" right after i posted that but may be resurrected soon. I will certainly attempt some of your suggestions and post results. Sorry i took so long to reply!