Simple questions to tackle audits or troubleshooting jobs
Last month's article focused on having the right tool for the job. When conducting a power-quality audit or carrying out a troubleshooting job, the right tool is just one part of the process. You next have to figure out how to go about using those tools to achieve the objective. A slightly different approach is used for either an audit or a troubleshoot.
Of the two tasks, the PQ audit (also called a benchmark or survey) is initially the easier one. It is usually scheduled and is without pressure to get the system back up and running as soon as possible, which is often what occurs during a troubleshooting job. Whichever you are handling, there are some basic steps to follow that can make your task more productive. These steps follow familiar rules from back in English classes, which were used in writing newspaper articles: what, where, when, how and why.
o WHAT-What type of disturbances or power-quality phenomena will you be monitoring? If you don't have any clues, it is best to use a PQ monitor that can do them all simultaneously.
o WHERE-Where are the disturbances (or effects of the disturbances) occurring? Is it the entire facility, one floor or one piece of equipment? That will help determine if you start monitoring at the PCC-the point of common coupling or the service entrance-or far downstream from the electric supply source (at an individual piece of equipment) and work back toward the PCC.
o WHEN-When are the disturbances present: a specific time of day, one shift versus another, seasonally or after a new piece of equipment was installed. Suppose an ASD trips offline every day at 6 a.m. This may be caused by an overcurrent sensor in the ASD reacting improperly to a transient that originates from the utility switching in a power factor capacitor each morning at 6 a.m. The PF cap is used by the utility company to compensate for the loads that will be energized as the industry in the area begins operations.
o WHY-Is it an after-the-fact, forensic-like troubleshooting problem, or a proactive, preventive-maintenance trending of the various PQ parameters, such as total harmonic distortion or unbalance?
o HOW-How to go about conducting the survey or monitoring session is the focus of the rest of this article. The basic procedure is the same for audits or troubleshooting: plan/prepare, inspect, monitor, analyze, solve and repeat from the beginning, if necessary.
Planning and preparing often means talking to the people at the facility, finding a reasonably up-to-date one-line diagram of the facility, determining the voltage and current levels with which you will be dealing, identifying the potential safety hazards, and doing a visual inspection first. Make sure personal protective equipment and monitoring equipment are appropriate for what you will be monitoring. Putting a 300A clamp-on current transformer (CT) on a 1,000A circuit may result in a molten pool of plastic or, even worse, may start an electrical fire. Often, it is necessary to monitor for at least one business cycle, which is how long it takes for the facility's electrical-usage pattern to repeat itself. Most likely, the facility will have different operations taking place on different days of the week, making a business cycle of one week. The season can make a difference, as the HVAC load is different in most locations in the summer and winter months. Weather can also make an impact as weather-related conditions (thunderstorms versus ice storms) can affect the utility's electrical distribution system.
The next step is to analyze the data to see what it all means. If you were doing a baseline survey or periodic PQ audit, it is often useful to compare against studies conducted by area utilities or data previously collected at your facility. However, be cautious when using such data to draw hasty conclusions. For example, if a different type of instrument was used to record the original data, the results may not be directly comparable. Take harmonic calculations. The “old” method was open to the manufacturer, which may have been calculated over one cycle every five seconds. The “new” method is to use the IEC 1000-4-7 defined 200msec gapless window. The results can be quite different.
If you can correlate certain types of PQ phenomena occurring with the equipment malfunctions when in the “Columbo” troubleshooting mode, then you can work on putting in mitigation equipment, such as surge supressors or ride-through devices. Or, you can remedy the problem by relocating the offending equipment to a different circuit or rewire the circuit. Sometimes, it may require going back to the equipment manufacturer and getting its performance specs to compare against your data. If most of the sags that you record are above 70 percent of nominal and last less than 10 cycles with the equipment ride-thru rated at 50 percent of 0.5 seconds, then a call to the vendor to have them fix the problem may be in order, rather than putting in a costly mitigation device.
Often, you will find that it is an iterative process-sort of like an archeological dig. You uncover one layer, fix one set of problems, then repeat the process. With the first ones remedied, the second set will be more obvious, and a solution can then be derived for the problems. Monitoring again after installing a mitigation device or changing the wiring in the facility is a good practice. Sometimes, fixing one problem may result in a different problem-the solution for one may be the cause of another. For example, changing the capacitor banks may result in a resonant condition that wasn't there before, with a harmonic current from the ASDs in the facility, as shown in the figure at left. And it is good practice to go back periodically and monitor the same locations to make sure that new problems aren't festering below the surface. Best to be proactive, not reactive. EC
BINGHAM, a contributing editor for power quality, can be reached at 732.287.3680.