I have a production site that's mainly Beckhoff PLCs and one of the requirements they have is a heartbeat tag that both MES and the PLC will write to (PLC writes a 1, then I respond with a 0). However, every once in a while the heartbeat fails or times out, and causes the machine to fault. I never see any issues in the OPC server, and our data handshakes are always good. Attached is how I am writing back to this heartbeat tag, is there a better way to handle this? They have very limited OT support so checking the switches and network for issues has been a pain. I don't have this issue with the AB PLCs.
This is a very bad design. Writing tags from two directions almost always creates odd corner cases.
Push back on that specification. Suggest two heartbeat tags (short integers or timestamp integers). One incremented/updated by each side (with rollover) at a steady pace. Each side monitors the other for a stale value.
Thanks, I will see if they can change this on their end. Its a very simple logic change in the PLC, but because they are inetgrartors, I hope they don't push for a re-quote.
