I HAVE A DIGI ONE CONNECTED TO AN ALLEN BRADLEY COMPACT LOGIX PLC. THE DATA IS TRANSFERRED FROM THE PLC TO DIGI ONE THROUGH ETHERNET AND THE DIGI CONVERTS THE ETHERNET TO MODBUS RTU. ON THE OTHER END I HAVE A MOXA MODBUS GATEWAY WHICH RECIEVES THE SERIAL DATA.
THE COMMUNICATION WORKS ABSOLUTELY FINE FOR UPTO 3 TO 4 HOURS AND THEN THE DIGI ONE FREEZES AND NO DATA IS TRANSFERRED. THE COMMUNICATION IS RESTORED BY RESTARTING DIGI ONE BY TAKING POWER OFF AND BACK ON.
PLEASE ADVISE WHAT COULD BE THE POSSIBLE CAUSE OF THIS PROBLEM AND WHAT COULD BE DONE TO RESOLVE THE PROBLEM.
This can happen if you do NOT have IGMP-snooping enabled switches. The AB/CompactLogix (& other AB/EIP devices) can sustain a Class D multicast rate which eventually over-whelms the DOIAP’s ability to handle the discarding of packets.
Another cause is for teh timeouts in the DOIAP to be almost exactly the same as the actual poll rate. For example, if the CLgix sends a new request every 2 seconds and the DOIAP is idle-time outing in 2 second, you can have a race condition which causes TCP/IP sync problems (meaning the DOIAP is tearing down the EIP socket when the CLogix sends the next request.)
Is the CLogix polling flat-out? meaning, the MSG block enable rung is tied high? Try using a timer to trigger the MSG block at a predictable rate. Or if the MSG block already ahs a timer, try speeding up the CLogix poll by 500msec, or slowing it down by 500msec.