NI-9512 fault when idle

[bmason18]

I have a stepper motor that I am operating to rotate a wheel using an NI-9512 module.  When operating the motor, everything works fine.  However, the longer the system is left to idle, the higher the chance that the module will display a fault via its front LED.  Most often, we see this first thing in the morning after it has been idle overnight, but it will occasionally happen over shorter time frames (an few hours or so). 

I have tried to extract the fault using property nodes and Softmotion function blocks tied to the axis resource, but the fault codes always return "0".  Power cycling the cRIO does clear the fault, but obviously this is a less than ideal solution. If it helps at all, the module is connected to a Kollmorgen P70530 stepper driver and the DIP switches have been verified to match the motor series.  All individual physical components have been swapped out with spares 1-by-1 trying to find the cause.

Any ideas on how to identify the nature of the fault would be greatly appreciated.

Thanks in advance!

[RIObotics]

Hi there,

Which cRIO model are you using?  I've used the NI-9512 before, and I'll admit that sometimes I've encountered the same issue, the red fault LED will light up for no apprenent reason.  You can try formatting the cRIO and re-installing its software.  There might clean up some bad files that have been accumulating.  After I formatted my cRIO, the faults seemed to go away, but I didn't continue working on that project long enough to know for sure.  Worth a try.

BR

John

Add motion to LabVIEW in 30min or less - TENET EMotion
Finding it hard to source legacy NI hardware? Try Extend Test

[bmason18]

This is on a cRIO-9067 running LV2020 and Softmotion 2020. 

One troubleshooting attempt I did forget to mention before is that we did try a spare module, though the problem was actually worse with it.  Faults were happening on the order of once per hour and sometimes immediately upon restart of the cRIO.  In addition, after initializing, it would run a multi-step routine without a command being sent.  Swapping back to the other module removed that behavior and took us back to the longer idle time faulting.

I have no idea why the behavior is so different between the two physical modules (or why one would run a command on its own... this one really confuses me), but I suppose it all could be due to bad files.  When I have some downtime on the equipment it is running, I will try a format and fresh install to see if that solves it.

[bmason18]

I finally got a chance to perform a fresh install on our cRIO.  Unfortunately, the fault will still usually appear after long idle periods.  Since it typically happens only after being idle overnight or over a weekend and everything Softmotion has been deprecated, I may just label it an annoyance we have to deal with and instead begin investigating the alternatives we will need to migrate to eventually anyway.

-Brian

[RIObotics]

That's good to hear.  For what it's worth, my company has a replacement solution for the now deprecated NI-9512 and SoftMotion module: https://geco-lv.com/emotion.html.  We've made our LabVIEW API to be very similar to NI SoftMotion VIs, so that the migration effort would be minimal.  I've attached an info PDF for the migration specifics, feel free to reach out to learn more.

BR

John

Add motion to LabVIEW in 30min or less - TENET EMotion
Finding it hard to source legacy NI hardware? Try Extend Test