CEEPS (Customer) asked a question.
I have a P2000 plc controlling two large Nidec drives over modbus TCP/IP. It gives the run and inhibit commands as well as speed settings. Some readings are taken from the drive such as % load, health etc: This has been running for a couple of years without any issues, but recently the two drives have been failing with network errors BG/Overrun. After speaking to the drive manufacturers and sending them data from the drives I have been informed that the fault is coming from the PLC's. Network traffic has been reduced to the point of only starting the drive and setting the speed with the Automatic polling set to 1000. What I would like to know is if it is possible that the PLC is causing issues with the drive. I have recently updated the PLC software and the firmware. Maybe the faults have started since the updates have been done but I had been working on how to fix the drives. The drives have never had any updates carried out.
Are you seeing comms errors on the PLC side as well? For those drives, what does BG/Overrun mean? Depending on your network you could potentially have another device writing to the drives or you might be seeing broadcast traffic issues.
You can run Wireshark and see the traffic on the network.
https://www.wireshark.org/
Regards,
Garry
This may require a smart switch (for port mirroring/sniffing.) I use a Netgear Prosafe FS726T smart switch.
It is a rather large (26-port) switch, but... it was free.
Always beware of one manufacturer pointing fingers at another.
The blame is not always necessarily accurate.
You did not provide any information about the model of motor/drive.
Please provide links to docs if you have them.
Check that the modbus addresses that the P2 is writing to do not have a Max Write Cycle (or equivalent.) You may be damaging the MEMORY.
Interesting, I have a P2000 running two Nidec M751 Digitrax drives for a little over 2 years. Yesterday I got a call that the servo is acting "weird". After some back and forth, I had them swap the entire control system with another (I had an unused duplicate). Everything cleared up. I have another identical system running next to this one (three systems total). I am traveling to the site next week to investigate the issue with first system. I have not done any PLC firmware updates in over a year so I know it is not related to that. So how exactly did you determine it was network errors? Is this the error code on the drive? What does your system do when the drive has network errors?
I would check and see what kind of broadcast traffic is going all over the network. Managed or at least "smart" switches with IGMP Snooping should be used to ensure packets dont spam all over the network and cause issues.
The PLC itself does not appear to have any issues and the only communication is between the drive and the PLC. From the drive manuals the BG Overrun is just put down as Background task overrun. This is on the network card and it turns the drive off. It seems that to recover from the fault you have to power the drive down. The drive is a Nidec-Unidrive M700 and this is a link to the page for the fault. https://www.manualslib.com/manual/1510722/Nidec-Unidrive-M700.html?page=204#manual. It may be a coincident but the company recently suffered a ransome ware attack and has had to rebuild the complete IT system.
"and the only communication is between the drive and the PLC."
Are you sure? How did you verify this? From some of Nidec's other literature, it seems as though the problem is related to broadcast storming and the drive is shut down when it can't process traffic. If switches have been reconfigured recently and there are other devices on your control network (especially desktop/laptop computers and the like) the drive could be seeing unintended traffic. To take the physical PLC out of the loop, you could try the Modbus TCP Master tool to troubleshoot just the drive.
I have downloaded a program called wireshark which appears to be for checking the amount of data on the network. The computer it is running on has a lot of modbus data as to be expected but how do I use the software to see what the drive is receiving. Its ip is 172.20.5.105 and this was not showing up on the PC. When I ping to the drive its response is less than one ms with the odd 14ms.
I am not an expert on Wireshark, but I believe you need to have a managed switch in between the PLC and drive with a port set up to send mirrored traffic from the one going to the drive. The PC running wireshark will be connected to the mirrored port. That is how I have heard people mention it before anyway.