I was on a start-up for a customer last week when one of the other people from the district showed up and was talking about a problem at another site. It would seem that they had two small drives running some dosing pumps and whenever the pumps would run, the signals from their metering would "wig-out". I just injected a single thought into the conversation and told them to check their grounding.
Earlier this week, that customer call me and asked if I would come to the site where they were having this problem. I arrived at the site and they proceeded to show me that they had installed a filter on the drives. They showed me that the power and signal cables had previously been ran in the same wire gutter but that they had added another small gutter above the existing one and had moved the signal cables into it. They also showed me that they had shielded wire going to the motors from the drives. Well, here are my thoughts and observations:
1. The filter was a TCI K-series which is a RFI filter that was installed on the common input feeder to the drive cabinet. Humm, great but pretty much not needed. It is however a good idea to install a simple Line Reactor at this point for drive protection and some harmonic mitigation back onto the systems service and other connect load(s). Not to mention it would have been cheaper and more effective.
2. It was great that an effort had been made to get the signal cables out of the same gutter as the motor leads and other power conductors for other equipment in the building. The problem was though, they ran the new gutter right on top of the existing one and it ran parallel to it for about 25 feet. Anyone see the problem here? Can you say inductance? This was a valiant effort but really a big waste of time. If they would have added some separation space, this would have been worth the effort. Right together, not so much.
3. The shielded cable....ah, another great idea but done incorrectly. See, the shields where tied together in the VFD cabinet. Then the other end of the shield was attached to the ground lug in the motor which also is where the ground conductor for the motor was tied and this was at best a bad connection because there was some corrosion on the lug to motor attachment point.
Looking at this, I had the customer clip off the shields in the motor conduit box and re-make the grounds for the motor conductors. I then grounded the shields in the VFD cabinet and what the heck do you know.....problem solved! Due to the fact that the shields where tied together in the VFC cabinet and were poorly attached in the motor connection boxes, it was creating a huge loop from the output of the drives to the motors and back to the feeder panel where the ground wires were connected to ground. Yes, connect to ground there as the grounds came into the VFD cabinet but only to a terminal strip and then continued to the feeder panel.
I can't tell you how many problems I have solved just starting with just a check of the basics of good wiring practices...mainly grounding. It is worth the time to do this right the first time. And please, I ALWAYS ground all my shields in the drive cabinet and cut in the field.
Fly the field and ground the garage! My instrumentation teacher gave me that years ago and they are words to live by. Albeit stupid, but easy to remember.