It's scary to think how many drives the engineer might have replaced before discovering the true culprit.
Variable-frequency drives (VFDs) and power quality problems seem to go hand in hand. If it isn't causing problems in the building by affecting other equipment, the VFD itself is the victim of a supply problem. You should expect the occasional problem, but what do you do when the customer has already replaced three units and the problem still persists?
After calling his supplier to replace a 7½-hp drive for the fourth time, the electrical engineer at a cosmetics manufacturing plant decided a new course of action might be necessary. He had returned each unit with a note that read “Keeps tripping on overload,” believing they were all defective. After listening to his problem, I made an appointment with the engineer to visit the site and investigate what was affecting these drives. I was very impressed with his knowledge of the system and with the installation itself. He had mounted eight drives that controlled the motors serving a production conveyor in a NEMA 12 freestanding cabinet with filtered air-cooling.
The engineer explained that all the units worked well, but he believed he had received a series of defective VFDs, because they all appeared to trip on overload when no overload was present. As proof of the problem, he showed me a chart tape of his current measurements taken over several days. The charts indicated the system had experienced no overloads. At the same time, however, it didn't make sense for this many units to have the same problem — especially when the rest of the drives worked fine.
Before I begin troubleshooting any power quality problem, I always ask myself, “Is this a global problem, or is this a local problem?” In this situation, all the other units powered by the same source were working fine, so it had to be a local problem. We looked at the motors, and they all were of the same design. The loading of the conveyor did not seem to be a problem either.
It wasn't until I asked the engineer to take another set of current readings at the motor that I discovered the reason for the overcurrent trips. As the engineer prepared to connect his meter to the conductors serving the motor in question, I could see the wires ran in a metal wire-way alongside the conveyor. When he pulled the leads out of the tray, I noticed they were wrapped around another larger set of conductors that served an air compressor located near the conveyor.
I asked the engineer to force the air compressor on by readjusting the pressure switch. As soon as the unit kicked on, the VFD tripped out on overcurrent. The compressor motor conductors were inducing a secondary current onto the VFD motor leads, which were inducing a false current to the protective circuits of the VFD. The fast peak current of the starting compressor motor never registered at the chart recorder, but the VFD's solid-state protection had no problem picking it up.
It's scary to think how many drives the engineer might have replaced before discovering the true culprit. While most of us know not to run signal and control wiring with large motor conductors, it's easy to forget small motor conductors. However, the rules change when it comes to solid-state overcurrent protection. Microprocessor-based motor controls have far superior protection capabilities that can and will cause nuisance trips if you're not careful. Good wiring practices and proper routing of conductors inside a control cabinet are a necessity for today's controls. Often, a group of circumstances and occurrences can conspire to give the illusion of a power quality problem.
The preceding scenario qualifies as a noise problem. Electrical noise is nothing more than random signals getting into circuits where they are not wanted. This indirect transmission of signals and power may be hard to comprehend, but one look at a radio tower will demonstrate how easily it can happen.
There are four ways unwanted noise (that is, voltage or current) can enter a circuit. It pays to have a basic understanding of how and where these conditions can exist.
Capacitive coupling occurs when two conductors are insulated from each other. In the case of a lightning discharge, air acts as the insulating material between two potentials. This electrostatic noise can cause headaches when phantom voltages appear on conductors and prevent solid-state control logic from operating correctly. These strange voltages can be very frustrating for electricians.
Inductive coupling is current-based, and it occurs when induced magnetic fields cut across nearby conductors. This is the same principle used in a transformer's primary and secondary windings. It is difficult to believe one single loop can induce substantial current into a secondary conductor. A fast rise in current (as in our story) or a fast rate of change (oscillation) can also affect the increase. The proximity of the conductors and even their size and shape are concerns, so wrapping excess wire in a coil inside the cabinet is not a good idea.
You should even consider the type of signal conductors you use when placing conductors. You should isolate input signals from output signals when they run in the same raceway.
Conducted noise does not require a traditional conductor to carry unwanted noise. Solid-state devices (like transistors and SCRs) never really open up and isolate a circuit from a potential. Instead, they limit the amount of current that passes through them. Even in the off state, there is always a small amount of leakage current that passes through the device.
Another example of this is when we create ground currents by using isolated grounding techniques. A current will flow between grounds using the earth as a conductor.
RFI noise, also known as electromagnetic interference (EMI), has become a greater concern over the past few years with the increase in transmission devices in use. We sometimes overlook the fact that any conductor or device switched at a high frequency (10 KHz and above) will act like a transmitter, with surrounding conductors becoming receivers. Shielding and isolation techniques are often necessary to prevent this unwanted noise from affecting our circuits.
In the end, all circuits can generate noise and all circuits have the potential to be affected by noise. This is the nature of working with electrical devices. Diligent installation, wiring practices, and forethought into the potential problems that may arise are the keys to keeping noise problems from affecting our operations.