What Impedance Adds That Resistance Does Not

Why this matters

A tech puts an ohmmeter on a 24 V contactor coil, reads 3.4 ohms, and calls the coil good. The reading is correct and the conclusion is unsupported, because in service that coil opposes current with roughly 267 ohms and only about 1 percent of that opposition is the quantity the ohmmeter measured. This is not a bad meter or a bad tech. It is a category error: the instrument answers a question about a de-energized path, and the fault lives in a quantity that only exists while current is changing. Knowing exactly what your ohmmeter cannot see is worth more than another decimal place on what it can.

What the instrument is really doing

An ohmmeter is a small DC source and an ammeter in one box. It pushes a known low voltage through the path, measures the resulting current, and reports the ratio. Two consequences follow and both matter in the field. First, the test is DC, so nothing in the path is changing, so any opposition that depends on change does not appear. Second, the test voltage is a few volts at most, so the reading describes how that path behaves at a few volts, not at 24, 240 or 480.

That second point is why a separate insulation-resistance test exists at all. If you use one, disconnect solid-state boards and electronic controls first because the applied test voltage will damage them, and let the instrument's own discharge function bring the winding back to zero before you touch the leads, since the winding stores the charge the test put into it. De-energize, lock and tag under 29 CFR 1910.333(b)(2) (29 CFR 1926.417 is the construction-side counterpart for lockout and tagging of circuits) and prove dead per NFPA 70E-2021, 120.5 before any of this.

Three kinds of opposition, and why they do not add up

• Resistance converts electrical energy to heat. It is present whether the current is DC or AC, steady or changing. It is what the ohmmeter reports.
• Inductive reactance appears in any coil, and it exists only because the current is changing. A changing current builds a changing magnetic field, the field pushes back against the change, and the result is opposition that rises with frequency and with inductance. At zero frequency it is zero, which is precisely why the ohmmeter never sees it.
• Capacitive reactance is the mirror. It falls as frequency rises and is effectively infinite at DC, which is why an ohmmeter across a good capacitor climbs toward open.

Impedance is the total, and it is not the sum. Resistance and reactance act a quarter cycle apart, so they combine as the two legs of a right triangle: impedance equals the square root of (resistance squared plus reactance squared). That geometry is why a small resistance sitting next to a large reactance disappears entirely from the total, and why adding those two numbers arithmetically always overstates the opposition.

Take the coil above. Sealed and holding on 24 V it draws 0.09 A, so its impedance is 24 divided by 0.09, about 267 ohms. Its resistance is 3.4 ohms. The reactance is the square root of (267 squared minus 3.4 squared), which rounds back to 267 ohms. The resistive leg contributes 3.4 out of 267, about 1.3 percent of the opposition, and it is the only leg the ohmmeter reported.

The blind spots, named

This is the part worth memorizing, because each of these is a real fault that leaves a normal ohmmeter reading behind:

• A few shorted turns in a many-turn winding. Forty turns lost out of four thousand is a 1 percent change in resistance, inside the tolerance of most field meters. But those turns form a closed loop inside a changing field, which is a shorted secondary, and it circulates current and makes heat. The DC reading barely moves while the coil cooks.
• Insulation that holds at 5 V and breaks down at 240 V. Resistance measured at the ohmmeter's test voltage is not resistance at line voltage. This is the whole reason insulation-resistance testing uses a raised test voltage.
• A joint that is continuous cold and opens hot. The ohmmeter tests it at a few milliamps and room temperature, neither of which is the condition that makes it fail.
• The difference between a coil's pull-in demand and its holding demand. Covered in the case below, and it is the one that produces chatter.
• Anything about current magnitude. The ohmmeter has no idea how much current the circuit will actually pass, so it cannot tell you whether a resistance it just reported is trivial or fatal to the circuit. Eleven ohms is nothing in series with a 267 ohm load and everything in series with a 44 ohm one, and the same instrument reports the same 11 ohms in both cases.

A contactor that chatters, and the number that hid it

A 24 V coil on a control circuit fed from a small control transformer, secondary reading 26.1 V with nothing calling. The contactor buzzes and rattles rather than pulling in and staying in.

The coil bench-checks at 3.4 ohms, which is normal for the type. The control run from the transformer to the coil measures 11.1 ohms round trip, which reads as continuity and got waved through on the first visit.

Here is what the ohmmeter could not tell anyone. A solenoid coil has an air gap in its magnetic circuit while the armature is open, and an air gap wrecks inductance. Less inductance means less reactance, and less reactance means the coil demands far more current during pull-in than it does once the armature closes the gap. On this coil, pull-in draw is about 0.55 A against a sealed draw of 0.09 A, so the opposition during pull-in is 24 divided by 0.55, about 44 ohms, against 267 ohms sealed. The coil's opposition drops to roughly a sixth of its sealed value for the fraction of a second it needs to close, and every bit of that swing is reactance. The 3.4 ohms of resistance never moved.

Now put the 11.1 ohm control run back in series and follow both states with the same numbers:

• Pulling in. 11.1 ohms at 0.55 A drops 6.1 V. Add roughly 1.0 V of sag inside the transformer itself at that draw. Coil terminal voltage lands near 19.0 V, which is 79 percent of the coil's 24 V rating. Many AC coils are specified to pull in reliably down to about 85 percent of rated voltage, but read the specification for the device in front of you rather than assuming it, because the figure varies by construction. At 79 percent this coil cannot close.
• Sealed, if it briefly closes. Current falls to 0.09 A, the same 11.1 ohms now drops 1.0 V, the transformer sag falls to roughly 0.2 V, and coil voltage recovers to about 24.9 V, comfortably above rating. It holds.

That pair is the chatter. Voltage collapses when the coil tries to close, recovers the instant it does, collapses again on the next drop-out. The mechanism is a coil whose demand changes by a factor of six between two states, in series with a resistance that only becomes significant in one of them.

The 11.1 ohms came from a corroded splice in a junction box, which by itself accounted for roughly 8 of it, with the rest being an undersized run over distance. Replacing the splice brought the round trip to about 3 ohms, pull-in drop to roughly 1.7 V, and the contactor closed on the first call. The chased-first suspect on the previous visit had been the coil, which was never faulty, and the second suspect had been the transformer, which was doing exactly what a transformer does under a momentary load.

Note the discipline point in that walkthrough: the current changed between the two states, so the drop had to be computed twice. A resistance figure alone is not a voltage drop. It becomes one only when you multiply it by the current that will actually flow at the moment you care about, and for a coil that is two different currents.

Where resistance genuinely is the whole story

Excluding cases honestly is as useful as including them. On a purely resistive load - a resistance heating element, an incandescent lamp filament, a resistance-based sensor - there is no meaningful reactance at power frequency, so resistance and impedance are the same number and an ohmmeter reading maps directly onto the current the load will draw. That is why a heating element is one of the few loads you can genuinely condemn from an ohmmeter, comparing the reading against the element's rating with the applied voltage. Even there, the resistance of a filament or element rises substantially with temperature, so a cold reading predicts cold current, not running current.

On conductors at power frequency in ordinary building-size circuits, resistance dominates and reactance is small enough that a drop calculated from resistance alone is close. That approximation weakens on long runs, on large conductors, and in steel raceway, where reactance stops being negligible.

How to verify you got this right

After a repair on a control circuit, measure the coil's terminal voltage twice: once at the instant of the call, and once sealed. Both readings need to be written down, because a sealed reading alone will pass a circuit that cannot pull in. Taking that reading means metering an energized control circuit, which sits inside the troubleshooting exception at 29 CFR 1910.333(a)(1) - permitted where the test cannot be performed de-energized - with a meter, leads and probes rated for the highest voltage present in the enclosure under 29 CFR 1910.334(c)(2), not for the 24 V you happen to be probing. A control panel is not a low-voltage panel just because one circuit in it is, and the arc-flash risk assessment under NFPA 70E-2021, 130.5 is done before the door opens.

If the pull-in reading now sits above the device's specified minimum with margin, and the sealed reading is near source voltage, the series fault is gone. If pull-in improved but is still under the specification, you removed one resistance and left another.

References

• 29 CFR 1910.333(a)(1) and (b)(2) - de-energizing and safe work practices for electrical work; 29 CFR 1926.417 for the construction counterpart
• 29 CFR 1910.334(c)(2) - test instruments and leads rated for the circuits to which they are connected
• NFPA 70E-2021, 120.5 (establishing an electrically safe work condition) and 130.5 (arc flash risk assessment)
• Manufacturer documentation for coil rating, minimum pull-in voltage and insulation-test limits
• See related: Why Voltage Sags Under Load; Continuity vs Voltage vs Resistance: Which Test to Run; How a Capacitor Behaves in a Motor Circuit