Photocoupler

Due to the poor contact of the network connector, or due to interference, etc., the DP network that has been used for a long time may be "flashed", that is, occasionally disconnected instantaneously and quickly returned to normal. If OB86 is not downloaded, the CPU will stop when it is flashed. It is now common to download an empty OB86 to solve the problem of downtime caused by flash.
But there are risks in doing this. If it is not a flash, but a real network failure or a failure of the slave, if it does not stop, it may cause catastrophic consequences. To do this, the following measures can be taken to determine whether it is a flash or a persistent slave failure.
1) Determine which network is the fault of which slave in OB86. If it is an "entry event" (a fault occurs), the slave-specific M position is set. If the slave's fault is an "outgoing event" (the fault disappears), the slave-specific M bit is reset.
2) In OB1, start the 200ms timing with the M bit, and call the SFC 46 (STP) when the time is up to stop the CPU. If it is flashing (the time when M is 1 is less than 200ms), the timing of this timer is aborted and will not stop.
3) Use one word to record the number of times OB86 is called, and hmi to display the number of times OB86 is called.
If the flashing fault occurs frequently, the hardware must be processed, such as replacing the DP connector to solve problems such as grounding, shielding, equipotential bonding, and anti-jamming.