Handling Broken Connections

AKD2G uses a watchdog timer and TCP keepalive to detect if communications between the drive and the HMI has been lost.

The Modbus watchdog fault, F7100, is generated if the drive does not receive any Modbus messages within a time window set by MODBUS.WATCHDOG. Zero disables the watchdog. The timer does not start until the first valid Modbus message is received. If standard process is used to disconnect no fault is generated. There is no need to write to a special register; any Modbus message resets the watchdog timer. If the drive stops receiving Modbus messages faults will occur on both axes.

The Modbus/TCP specification indicates all Modbus ports must use the TCP keepalive feature. Keepalive allows the AKD2G to detect if the connection is broken even if no Modbus reads or writes are currently being executed while the socket is open. This is achieved by injecting extra packets if no other Modbus messages are being sent.  Keepalive will fault if the cable is unplugged or cut or the PLC/HMI loses power unexpectedly. The keepalive fault is number F7101. Setting MODBUS.KEEPALIVE to zero stops the keepalive fault from being generated; by default faults will occur on both axes.