quirky GPIB-USB behavior - visa session connection lost

[instrumento]

Has this been fixed yet, e.g. by a new VISA layer?   Has this been acknowledged by NI as a VISA layer bug? I've been having this problem in two stations and not in a third one, all running identical software, the difference being that the one where everything works has a PCI GPIB card and the ones where the code fails have GPIB-USB adapters.   I'm not particularly fond of replacing that part of the code to use GPIB calls, but may have to.

[Chris_R.]

Hi instrumento,

There were some issues improved/repaired in some of the later versions of NI-VISA.  Have you upgraded to the latest version?  If not, you can find it here.  What version of LabVIEW and VISA are you using when you see the problem?

[instrumento]

I'm running VISA 4.0, Laview 7.1.   Seems to be a timing issue with a certain peripheral, when calling VISA Status Byte (waiting for SRQ) in a loop.  We have a delay before the first call to VISA STB, inside a while loop, and another delay inside the while loop.  By increasing those delays we seem to be able to survive, or diminish the probability of VISA erroring out, at a great cost to execution speed later on in the process.  Seems to always happen when I have another VI open with an image preview, through IMAQ, i.e. camera is streaming and I'm getting an image and putting in the front panel of a VI, where the camera is attached to a NI PCI based frame grabber.  The other VI is moving a stage over VISA/GPIB in jog mode via pushbuttons, and eventually one of the moves causes a VISA connection lost to the stage controller.  On the (identical) system with a PCI GPIB card the problem never occurs, or is so infrequent that we never seem to hit it.   If the video is not streaming I don't recall ever having problems, but I'm not 100% sure on this.