When you're called in to troubleshoot a power quality complaint, what's the logical sequence for measurements, and what tools do you really need?
With utilities deregulating, your business is likely to grow. For example, many utilities won't have the personnel needed to service small- and medium-sized customers in the manner to which they have become accustomed. Who will answer power quality complaints from the customer's side of the meter? You can, if you're willing. Do you need new equipment to handle this business? Probably not. Whether you're preparing for a LAN installation or troubleshooting a computer crash, the combination of basic inspection, logical thinking, and the right test instruments (which can be inexpensive and simple) will solve about 90% of power quality problems. It's only when you want to become a power quality expert that you need to acquire more expensive, specialized tools.
Because the vast majority of power quality complaints can be traced to wiring and grounding problems, the first step you take should always be to examine this system with your own eyes. It isn't fun, it isn't glamorous, but it usually works best.
Is there anything strange about the panel serving the receptacle? Go and look at the grounding electrode. In one instance, a visual inspection showed the ground cable had been removed from the driven electrode, the electrode carefully painted by some industrious painter, and the cable then clamped to the painted surface! Check the main bonding jumper connections, etc. Examine all along the wiring system. Do you smell hot metal or insulation?
The visual inspection also works for the popular problem of daisy-chained neutral wires as well as for detecting continuity problems with metallic enclosures, conduit, raceways, panelboards, expansion joints, and telescoping raceways. Here's the best news: Visual inspections are free.
Start with simple tools
Let's say you've received a complaint regarding an office computer, and your initial inspection didn't conclusively uncover the problem. Keeping in mind that wiring and grounding are most often to blame, start with the simplest measurement possible: Plug a wiring indicator or receptacle analyzer into the outlet, and note its readings. In their simplest form, these meters are small (2 in. square) cubes that light to indicate neutral and ground reversals, bad or missing neutral wires, and the presence of voltage. Some include a test for GFI trip protection, a feature not needed for this application. Simple and small wiring indicators cost about $10 and may give you all the information you need.
If they don't give you all you need, you next want to know the voltage level at the socket, easily and inexpensively found using a two-pronged voltage indicator, available for about $35. You'll pay more for one with digital readouts showing precise voltage. The readings should be 120V on line-to-neutral, 120V on line-to-ground, and 2V or less on neutral-to-ground.
Another useful test at this point is for impedance, a measurement often incorporated into wiring indicators. High impedance can indicate poor quality connections or an improperly installed equipment grounding conductor - problems you should have caught on a visual inspection, but possibly missed. A low impedance neutral minimizes neutral-to-ground potentials and common mode noise at the load. Impedance should ideally be kept under 0.25 ohms on circuits serving electronic loads. Analog ohmmeters that measure to this scale can be purchased for as little as $15. (Testing of the grounding electrode and conductor, requiring an earth ground tester, usually would not be required for a survey of this type.)
Features versus benefits
It's at about this level of purpose that manufacturers start adding bells and whistles, resulting in a myriad of choices in test meters and instruments (see Table 1, on page 74). For example, you can combine the features of a wiring indicator, voltage indicator, and ohmmeter mentioned above, add a measurement or two (e.g. load on line), perhaps some accuracy, and pay $100 or more. Add an LED display and more capability, such as peak voltage and ground impedance, and pay $300. You can also buy a meter that combines all these measurements plus leakage current and tests for the earth grounding system for $1200. The combinations of features, benefits, and price seem endless; hence, this article.
In making your decision about what to buy, remember you have to feel comfortable using the meter. Maybe it's convenient for you to use one meter that combines the measurements you use most often. On the other hand, you'll waste time and money if you either can't find the meter or can't remember how to extract the reading when you need it.
In reviewing the choices, it would seem the wise purchaser would save money and perhaps be less confused by buying one or possibly two low-end devices, and then perhaps own one high-end meter, probably a digital multimeter that incorporates measurements needed for more in-depth troubleshooting. These purchases will get you through 90% of the problems.
One more note here: If power quality becomes an issue at least once a month, a power quality monitor ($1500 to $12,000) that can document disturbances and, therefore, prove or disprove power as the source of equipment problems can probably pay for itself rather quickly. It will never take the place of visual inspection (see Table 2, on page 76), but it can replace most of the equipment mentioned here for the purposes of troubleshooting power quality.
More in-depth troubleshooting
If the problem persists, your next step is to check for illegal neutral-to-ground bonds by using a true rms digital multimeter (DMM), typically in the $270 to $360 range, with high/low capture and a current clamp ($100 to $200). (The digital multimeter includes the features of the $35 voltage tester mentioned above. Some include [TABULAR DATA FOR TABLE 1 OMITTED] the features of the $15 ohmmeter.) Connect the instrument at the neutral-to-ground bond nearest the service entrance. If the current reads more than a few amps, you need to hunt down an illegal bond, an example of which is shown in Fig. 1, on page 78. Although true rms often seems to cost about $200 more, it's recommended these days because without it you won't get an accurate reading when harmonics are present.
Next, capture the voltage ranges at the receptacle, also possible with this DMM using high/low capture. Sags may indicate a large load on the line, such as a vending machine, refrigerator, printer, copier, etc. (The easiest solution is to move the load.) High neutral current may indicate inadequate conductor sizing, shared neutral wires, or circuit overloading, particularly when several nonlinear loads share the circuit.
One other handy and highly recommended tool is a pair of spring-loaded fused jaw-clamp voltage leads (about $50 each). These fused clips make connecting to power panels and wiring much easier and safer. They grasp to make a good connection for measurements and work well enough to pay for themselves rather quickly.
At this point in the investigation, you'll want to check for harmonic distortion; determine if the circuit is dedicated to nonlinear loads; and attempt to capture a disturbance. The most inexpensive way to determine if there are harmonics on the line is to connect both your true rms DMM and a DMM that's not true rms (e.g. rms, averaging, peak-detecting, about $100). If the two report different readings for voltage and current, harmonics are present.
You can also connect a battery-operated oscilloscope, one that has floating inputs so you can measure line and neutral currents and view waveshapes. A current waveshape, as shown in Fig. 2, on page [TABULAR DATA FOR TABLE 2 OMITTED] 78, indicates the circuit is used by nonlinear loads exclusively. A current waveshape, as shown in Fig. 3, on page 78, indicates that mixed loads share the circuit and may be interfering with each other.
Note that if you don't use an oscilloscope that's battery operated, you're only able to take measurements with reference to ground (neutral-to-ground, line-to-ground). Oscilloscopes that are AC-powered tie the instrument chassis ground to the measuring ground. It's extremely dangerous to bypass the equipment grounding conductor, so don't do it.
You'll be able to justify the expense of a handheld battery oscilloscope if you also need to diagnose adjustable-speed drives, check power electronics signals, etc.
Correlate the symptom with the problem
If you haven't already done so, it's at this point in troubleshooting that you should start collecting other data. Therefore, when the problem occurs, you can correlate the symptom with a possible power disturbance event. What other problems are occuring in the building? What other types of equipment could be interfering (arc welders, air conditioners)? And, what types of power conditioning equipment are installed?
Ask those experiencing the problem to start a log noting the time, date, and brief description of the problem. Power disturbances are usually intermittent and, therefore, can be hard to capture, but it is possible. You can try to capture it on your oscilloscope by setting it to the mode with fastest sample rate; 40 nanoseconds (ns) is typical. Every power disturbance has a signature that, when analyzed, can help lead you to the source of the disturbance. If you've come this far and have actually captured a disturbance that correlates with the problem, store it into memory. If you're not able to decipher the source of the disturbance based on what it looks like, you can call a consultant or refer to available manufacturer-produced books that show "signature" waveshapes. These books explain the physical clues and suggest probable causes and solutions.
An elusive problem
If you're not able to capture the disturbance on the spot, connect a power monitor (an instrument designed to monitor over time and sample fast enough to capture all disturbances). Don't take this lightly. At this stage, you don't want to miss information; there are several models that do not provide the subcycle information you need to see. Typical measurements include sags, swells, waveshape faults, impulses, frequency errors (valuable for monitoring auxiliary power supplies), power failures, high-frequency noise, total harmonic distortion, temperature (fast temperature changes can be mistaken for power disturbances), and current.
There are three types of power monitors: Event indicators ($600) that light LEDs to indicate if and what type of disturbance occurred, text monitors ($2000 to $3000) that time stamp the event and describe it in terms of voltage and duration, and graphic monitors ($1500 to $12,000) that fully document, time stamp, and print an analog picture of the disturbance.
Because they need to be left in position over a period of time, power monitors are usually tamper proof or tamper resistant, with either a lockable front panel or control and data retrieval using a computer. Many include environmentally rugged cases for monitoring outdoors or in hostile factory environments.
Harmonics meters and analyzers can help identify a harmonics problem, in some cases track down the major contributing source, provide data so a solution can be devised, and verify that the installed solution works. Harmonics meters ($895 to $1895) are handheld instruments meant for taking spot checks of the electrical system. They report voltage and current Total Harmonic Distortion (THD) and display the harmonic spectrum and waveforms for each harmonic up to the 50th. One model provides simultaneous 3-phase measurements, and most provide some amount of data storage and automatic data transfer to a PC for data analysis.
Because these meters are handheld and not meant to monitor and collect data over a long period of time, they can only be used to confirm the presence of harmonics, but not to rule it out as a source of problems. Harmonics are only damaging as a source of heat that builds up over time. To analyze their effect on the distribution system, you need to use a harmonics analyzer ($8000) that can record and analyze data over at least a week's time.
We've addressed the tools from which you can choose for troubleshooting a power complaint from the receptacle. We haven't addressed prevention of power disturbances or how to do a power quality survey (see sidebar on page 74), which involves a more comprehensive approach (but still uses the same methods). Remember, visual inspection (using no tools at all) still remains the most powerful, fastest, and easiest way to solve most power quality problems. After that, if you keep in mind that power quality problems, although elusive, do obey the laws of physics, a logical approach using a few inexpensive tools should see you through most of the time.
POWER QUALITY SURVEY STEPS
1) Find out who ordered the survey and why. Understand the politics of the situation and have the appropriate people start disturbance logs,
2) Walk around take notes on all aspects of the electrical system and environment, construction work, etc.
3) Develop a monitoring plan including a list of monitoring sites and appropriate thresholds, starting with the problem location.
4) Check your results and correlate disturbances with equipment logs.
5) Choose the best solution and resurvey to verify results.
Dick Piehl is Vice President of Engineering at Basic Measuring Instruments, Santa Clara, Calif.