If your facility has had a slew of equipment failures that seem to have been caused by ghosts and goblins, you probably need a Power Quality (PQ) Survey. It’s tempting to rush into getting one done in hopes it will show you where the problem is. One problem with that is power quality issues are almost never due to one problem.

If your facility has not hade such a survey before, the level of survey you need will likely require a capital request. That means people above you are going to sign off on it, and they will want to see the results. They will want analysis and a report.

You don't want half the report to be about basics you should have already addressed. That could imply you aren’t up to the job. But it’s also hard to use at a practical level.

A plant full of basic issues will generate a lot of noise that hides the signal. You need to clean up these issues before commissioning the survey so the signal isn't lost in the noise. This will not only validate the PQ Survey expenditure to the bean counters, it will also help you identify problem causes that are currently invisible.

What normally (okay, always) happens when a huge list of fixits comes back is work commences on the list and then stops getting done. Two years later, another power quality survey gets done and again it's not really acted upon due to handling all the background noise instead of hearing the signal.

One approach some plant engineers try is to limit the PQ survey to a particular area of the plant, so the cost is below the radar. Then, the thinking goes, we can apply the lessons learned to other areas. This is a bad approach. For one thing, a full PQ survey starts at the Point of Common Connection (e.g., the utility service) and works its way down through the branch circuits. You cannot extrapolate from a selection of branch circuits to go the other direction.

Limiting cost is not the goal. In fact, you don't want to limit the cost. You want the full survey done, unless you already know where to look. What you want to limit is the amount of “should have already done that” work from the results of the PQ survey. Take care of obvious problems, first. You do not need a PQ survey to:

• Identify the 73 bonding errors in your plant, a visual inspection will reveal those.
• Discover that your four largest motors have no power factor correction, a visual inspection can show you that.
• Reveal that you have a terrible voltage imbalance in the panel supplying the 50HP motor that is replaced twice a year. Looking in the panel and seeing that the lighting loads (which are, by definition, single-phase) are all supplied from the same two phases.
• Clue you in that you may have a high number of high impedance connections. Overtorquing is the most common cause of these, because the standard method of "checking connections" is to give them an additional twist. This method makes good connections go bad. If your maintenance procedures say to “Check connections”, you no doubt have a lot of high impedance connections. Test each connection using the correct instrument and replace those that fail. Insert this method in place of anywhere in a procedure that something like “check connections” is used.
• Show the plant equivalent of the “canary in the mine”. That would be junction boxes and panels with unused openings that are still open. If you see more than a couple violations of NEC Sec. 110.12(A), you can expect a staggering number of NEC violations in the plant. Some of these, especially those involving Art. 250 Part V, can play a big role in PQ problems.

You want to address the basics. Make a list of what inspections you have the capability to do, and then do them. Only then should a PQ survey be conducted. to help identify what you may have missed and what is not basic. Often, it’s worth hiring a qualified electrical services firm or electrical consultant to do such inspections. Or ask the firm that’s doing your PQ survey if they might go bigger with this effort and conduct the visual inspections a week or two before they conduct the survey. That gap should give you enough time to correct the issues that are most likely creating power quality problems prior to conducting a survey that is best used to reveal problems you can’t see with a simple visual inspection.

Remember, the point isn’t to delay conducting the PQ Survey or to limit its scope. The goal is to eliminate defects that would otherwise skew the results or bury them in noise. This is how you can most effectively use the PQ Survey to get a clear picture of your hidden power quality problems.