Com error - PLC M340 + SCADA VijeoCitect

[ndelucca]

Hello all,

I got a call yesterday from a worker at a factory where i installed an automation sistem with a PLC M340, operated via SCADA Vijeo Citect. He told me that everithing was saying "#BAD" instead of the usual. Next i spoke with a sistems operator at the factory and checked the following steps:
1_ No hardware errors on PLC
2_ All Ethernet cables + switches are not damaged, properly connected (the communication between PC and PLC is via Ethernet, using the OPC driver)
3_ Reset PLC, PC (scada), switch.
4_ Ping between PC and PLC (worked), according to UnityPro status comunication is OK
5_ Just in case, we programmed the PLC with a backup program (the exact version it had installed previously)
6_ Exported the DB and FB variables from unitypro to *.XVM file and checked that the OPC driver was set-up correctly.

I'm running out of ideas, the problem persist, and everithing looks properly configured, no idea why it suddenly stopped working (its been running for almos a year now, with no problems at all)

Any thoughts on how can i find the source of the problem? (you cant fix something before you find out what it is you have to fix)

Thank you all, i await your reply

 

[robertkjonesjr]

Is OPC working? Are you using OFS or a 3rd party OPC package? When you are on a page, requesting OPC data, that should drive the OPC server to generate requests as a client to the PLC. I would want to know if those requests are going out to the PLC or not, through log files and/or traffic analysis.

 

[ndelucca]

hi robert, i'm using OFS. yesterday i asked the operator to take a look at the OPC OFS screen, showed "Running", Total Coneccions: 1, and according to the data transfer it was communicating, or at least trying to communicate... I actually forgot about the network log, didn't ask what it was saying. I'll have to ask for that information.
Let's say for now that the driver is working, what would you do next?
Thanks!

 

[robertkjonesjr]

>> what would you do next?

Wireshark. It would cut the problem in half - valid requests are either on the wire, or they are not. Whether they exists or not would then guide to who is not doing their part in the communications.

The M340, assuming this is all Ethernet based, can handle multiple simultaneous requests, so one connection is low - I would consider increasing max pending and max channels.