GPIB Write halts

[Elsa Fung]

Hi!  I have two sets of system, both are Dell computers running on Windows XP Pro with an IOTech IEEE card (PCI 488).  One of them is my development system which has all of the NI Developer Suite 2006 installed.  The other is a test station, where the deployment of the project will eventually run on.  The test station only has LabVIEW 8.0 RunTime Engine installed.

I tested a few IEEE functions on both system using a program called WinTest (directly writes to and reads from a specific addressed device) from IOTech, and both of them work properly.

However, when I try to use LabVIEW's GPIB Write and Read on the test station, the LabVIEW application gets "stuck" on the GPIB Write (the application runs totally fine on the development system).  I found the application in the LabVIEW example database, so I don't think there's anything wrong with it (all it does is: GPIB Write -> Delay -> GPIB Read).  I added messages in the application to see where the application was stuck on, but that should not have any effect on the GPIB Write.

What I mean by "stuck on" is:

The application sends an instruction to the device with the GPIB Write, the device received the instruction (I used a temperature chamber to test this: send and instruction to change the setpoint of the chamber, then check the LCD on the chamber to see if the setpoint was set, and it was), but application does not seems to proceed to the next step (Delay).

I attached the project I used.  Please help.

Thank you,

See-Ming Elsa

[JasonSmith]

The example which ships with LabVIEW is intended for use with NI GPIB hardware with the NI-488.2 Driver.  As you say you are using IOTech 488.2 adapters, you will need to make sure that you have the IOTech driver installed on both computers.  My suspicion is that the driver call (GPIB Write) is never actually returning.  In order to determine this, try turning on highlight execution in LabVIEW, and watching your application run from the block diagram.  You should be able to see the program flow stopping at some point, which I suspect will be on the GPIB Write command.

If this is the case, that the programming is hanging on the GPIB Write, you will need to contact IOTech for support, as you are interfacing with their driver and their hardware.  If the hang is ocurring elsewhere, please reply back with more information.  Also, indicate whether the abort button works when the program hangs, or if you have to actually kill LabVIEW in order to stop the application.

Jason S.
Applications Engineer
National Instruments

[Elsa Fung]

it is waiting for GPIB Write to return, so I guess I have to call IOTech for support...

Thanks!!

See-Ming

[Elsa Fung]

Just one more question, why would it be IOTech's problem if I'm using the same driver for both computers?

I uninstalled the IEEE driver from my development system, and reinstalled it using the disk I used for the test system, and my development system still works...

Also, what exactly does it mean when the GPIB Write doesn't return?

See-Ming

Message Edited by Elsa Fung on 03-07-2006 09:51 AM

[JasonSmith]

The NI-488.2 driver is only designed and licensed for use with National Instruments GPIB controllers.  Although the driver appears to be the same from within LabVIEW, IOTech must be replacing the driver in order to make it work with their GPIB interface.  What is happening is that GPIB Write calls the lower level driver (In this case he IOTech driver).  Something in that IOTech Driver appears to be locking up and failing to return control back to LabVIEW.  In this case it is the IOTech driver which appears to be the cause of the problem, so it will have to be supported by IOTech.

Jason S.
Applications Engineer
National Instruments

[Elsa Fung]

I see..

The IOTech manual does have instructions to replace the gpib-32.dll in system32 to make the card compatible with LabVIEW.  But I used the same gpib-32.dll for both computers too, and it still doesn't work.  Oh well, I contacted IOTech already (they have a funny support dept, you either leave a message on the phone for them to get back to you, or you can email them), just all these questions pop up in my mind when I'm waiting for their reply.

Thank you!  I'll post up the solution if they reply me one, so other people having similar problems can see ^_^

See-Ming

[Elsa Fung]

Apparently, IOTech cards can only run I/O Synchronously.  So if you have the same problem, right click on the GPIB Write/GPIB Read, select "Do I/O Synchronously".  If you see "Do I/O Asynchronously", that means you are already running I/O Synchronously.

See-Ming