Are you certain that your AC voltmeter is safe to measure an incoming 480V power feed? Sure, the manufacturer may rate the meter as good for “600VAC,” but is that sufficient? Let’s look closer at voltmeter safety voltage ratings.
One often-neglected parameter of measuring AC power voltages is transients. While an input power panel may have only 480V, that’s a steady-state voltage. Consider transients. If the power feed into your building is shared with large motors or long-distance power lines, there’s a high likelihood that transients of 1000V or more may be present within the power panel. If you’re measuring that power voltage with your 15-year-old voltmeter when that transient hits, could you be in danger? Glad you asked.
A frequently misunderstood potential hazard of power measurements concerns meter input impedance. All voltmeters have a high-input impedance; the goal is to limit current drawn by the meter. With power circuits, it’s important that the meter not draw excessive current for safety reasons.
Voltage measurements are made by placing the voltmeter in parallel with the voltage source. The higher the parallel resistance (i.e., the input impedance of the meter), the less the current the meter draws. If the meter has a typical voltage input impedance of 10 megohms and it’s placed across 480V, it will draw only 48 nanoamps. Let’s see what may happen when an improper meter is used for a simple voltage measurement.
This real-world story concerns an electrician that was measuring the voltage inside a 480V breaker panel. After his first set of checks, he realized that something was wrong with the inside voltages. As he attempeted to make a second set of measurements, a fireball erupted from the panel. Both he and a nearby fire department official were killed. A third person was severely burned. What happened?
Remember, the panel had only 480VAC. An investigation showed that the voltmeter used by the electrician meter was not rated for 600V testing. When the electrician placed the probes onto the power bus the second time, the meter short circuited. When the electrician heard the noise he tried to withdraw the probes. But when he did, there were already thousands of amps of fault current running through the test leads. The test probes were immediately vaporized. The probe’s arc then followed the particle trail of the vaporized tips, creating a phase-to-phase or phase-to-earth fault. The instantaneous result was an arc blast and fireball, hotter than the temperature of an oxyacetylene cutting torch. Two people died, and a third was severely injured.
Most broadcast engineers know that a 10megohm meter should draw only 48 nanoamps across a 480V circuit. You'd think with such high-input impedance, there is nothing to worry about. But, the culprits are transient over voltages ("spikes"), which may appear on the power circuits.
You can't avoid these transients. Most are not going to do any damage, but the combination of high-voltage spikes and high current can be dangerous. Where do such spikes come from? Events like lightning strikes or motor switching can cause them. And as shown above, spikes can be severe enough to create a short circuit inside the meter.
Electrical consultant Edward Shen recommends engineers consider carefully the type of voltmeter used to measure primary voltages. Here is a recommended checklist. First determine what type of voltages you may encounter. The international standard, IEC 61010 covers voltage measurements of 1000V or less. Underwriters Lab bases its certification section UL3111 on the same standard, so either apply.
These standards specify four "Overvoltage Installation Categories." The higher the category number (in this case CAT II or CAT III), the more protection the meter offers from high voltage transients. The categories are based more on potential fault current than voltage. Within each category, there are different voltage ratings of 1000V, 600V, 300V, etc., all of which apply to both AC and DC voltages.
Rating a meter by both category and voltage can be confusing, so here's an explanation. If you ever work on power circuits then you need at least a CAT II or CAT III rated meter. Use only a meter with the proper category and an associated voltage rating, such as CAT III-600V or CAT III-1000V, marked on the front of the meter.
Old meters were rated with steady-state voltage withstand levels. So, in that sense, a 1000V-rated meter would be safer than a 750V-rated meter. But in today’s rating system that thinking only applies to meters in the same category. This means that a CAT III-1000V meter has better protection than a CAT III-600V meter. However, a CAT II-1000V meter is NOT "safer" than a CAT III-600V meter. Here’s why.
Under UL specifications, a CAT III-600V meter must pass a test against a transient with much more energy than the CAT II-1000V meter. A CAT III-600V will be tested with a transient, having the same peak voltage, but one that has one-sixth the test source impedance and therefore six times the current of the CAT II-1000V meter! Be sure you understand this key difference when selecting voltmeters.
A final meter purchase caveat. In addition to selecting voltmeters by category and voltage, look for certification by an independent testing lab, such as UL or CSA (Canadian Standards Association). Such certified meters have actually passed the mandated tests. Don’t settle for a meter that just says "designed to meet" or "conforms to IEC 61010." Those statements are no substitute for independent testing. The adage, “It looks good on paper” is no substitute for real-world testing. Don’t let yourself become that one statistic where the meter "test" failed.
For more information on power and electrical safety, see EC&M magazine.
