I agree it is a bad practice, but I have to wonder if it's because the current flows through the EGC or GEC. How would that damage the equipment? Consider this: a lot of electronic equipment have filtering and/or surge protection connected to its supply ground. Could the welding current find its way back through the filtering and/or surge protection to the normally-energized side of the circuitry then return to ground through its grounded supply conductor?
The most common damage mode will be when "grounded"
inputs or outputs of the PLC are suddenly at significantly different voltages. A protector from the
power supply input of the PLC to ground will not protect any other terminals.
If the cabinet ground goes one way and the sensor ground goes another way, the smoke will find a way out.
Just having sensors, one side of which is grounded for "protection", attached to different parts of the machine and coming back to the same PLC can do the job too.
Many PLCs will include protection diodes or other safeguards on their inputs and outputs, but they will not handle the level of current a welder can generate.
There may be other reasons too, but there are so many other ways that an improper welding return connection can damage even mechanical components that proper procedure is critical. (One example is welding current going through the bearings of a dovetail slide.
)
You are absolutely right that the damage mechanism need not be related specifically to EGC or GEC current. But that is one source for ground voltage differentials.