I was originally hesistant to initiate this issue because I felt grounding was not directly related to PLCs. From the sheer number of responses with suggestions and response from tomneth indicate to me that this is a huge problem for a lot of folks. I really appreciate the input.
Our older machines are much like the ones tomneth describes- no universal application and adherence to an equipment grounding philosophy. We suffer the consequences daily. They recently tried adding a ground connection for a set of stepper motors and guess what? The yield rate on the product went up measureably! Did they apply this thought to the other machines? Of course not. The up side of that is improved job security for me, but it will likely occur one machine at a time.
I floated the idea of getting a specialist and it may get rolled into a near term plan to improve the manufacturing floor. Because of the "may" aspect, I am taking previously described piece-meal path and putting in the #4 wire just for the sake of safety. (Right way to go, eh okiebob?)
One of my previous jobs was at a company that manufactured large semiconductor testers. Testers have many "buckets" loaded with customized circuit boards that draw 20 amps or more each. Each motherboard was always connected to a solitary common ground bus bar in just outside the AC breaker box with an individual #12 wire. The load was always grounded, not the supply output. When your dealing with 12V, 5V, 3.3V and 2V logic, any noise on the DC line is absolute death to the machine.
The prevelance of 24VDC in our pouching machines here makes them somewhat immune to small level noise, but it is still causing me grief.
Thanks harryg for the NEC reference.
-Grover
Grover