As an Amazon Associate, Akermin earns from qualifying purchases. Product links on this page are affiliate links.
Low impedance mode connects a load of a few kilohms across the meter input, which drains the tiny current that a capacitively coupled conductor can supply and collapses the reading to near zero. A conductor that is genuinely connected to a supply barely notices the extra load, so its reading stays put. The mode is therefore a test rather than a setting: switch it in, and a voltage that vanishes was never a source, while a voltage that holds is real. The reason an ordinary meter shows the ghost at all is that its ten megohm input deliberately draws almost no current, which is exactly the condition a coupled voltage needs in order to survive.
The ten megohm input that makes a dead conductor look live
Two insulated conductors running side by side down the same conduit form a capacitor. The metal of each is a plate, the insulation and the air between them is the dielectric, and the capacitance is small but entirely real, typically in the hundreds of picofarads for an ordinary domestic run and more for a long commercial one. When one of those conductors is energised at mains frequency, alternating current flows through that capacitance into the other conductor whether anyone wants it to or not.
If the second conductor is connected to anything at all at its far end, that current drains away harmlessly and no voltage builds up. Disconnect it at both ends, which is precisely what happens when a switch is open or a fitting has been removed, and the current has nowhere to go except into whatever you connect at the test point. A modern digital meter presents around ten megohms, which is enormous compared with the reactance of the coupling capacitance, so almost the whole supply voltage appears across the meter. The background piece on why ghost voltage appears covers the wiring arrangements that produce it most often.
What a low impedance input does to the capacitive divider
Treat the arrangement as a series circuit: the supply, then the coupling capacitance, then the meter input, then back to the supply reference. The coupling capacitance has a reactance of one divided by two pi f C. At fifty hertz with a coupling capacitance of 500 picofarads that works out at roughly 6.4 megohms. The meter input sits in series with it, and because one element is reactive and the other resistive the two combine as the square root of the sum of their squares rather than by simple addition.
The fraction of the supply that lands on the meter is therefore the input resistance divided by that combined figure. With ten megohms of input the resistance dominates and most of the supply voltage appears. Drop the input to a few kilohms and the reactance dominates completely, leaving a reading that is for practical purposes zero.
A conduit run worked through at three input impedances
| Meter input impedance | Fraction of the source that appears | Indicated voltage from a 230 V source | What the display suggests to an electrician |
|---|---|---|---|
| 10 megohms, an ordinary digital meter | 0.844 | About 194 V | Looks like a live conductor |
| 1 megohm | 0.155 | About 36 V | Looks like a fault or a leakage path |
| 100 kilohms | 0.016 | About 4 V | Looks like residual noise |
| 3 kilohms, a low impedance mode | 0.0005 | About 0.1 V | Reads as dead, which it is |
The middle rows are the dangerous ones, because thirty six volts on a display invites a theory. People start hunting for a degraded insulation path or a shared neutral when nothing is wrong except the impedance of the instrument they are holding. The Fluke 117 states low input impedance specifically to help prevent false readings caused by ghost voltage, alongside automatic alternating and direct voltage selection and a direct millivolt range stated at 600.0 mV with 0.1 mV resolution. The VEVOR six thousand count true RMS multimeter lists a low impedance mode among its functions together with auto ranging, a bar graph, data hold and maximum and minimum recording.
Three situations where low impedance mode is the wrong tool
The first is any low energy circuit. A fire alarm loop, a building management signal pair, a thermocouple extension, a four to twenty milliampere transmitter output, a pilot circuit sitting on an electronic input: all of these are sources with real but tiny current capability, and a few kilohms across them does not reveal a ghost, it collapses the signal and can trip or latch something downstream. The mode cannot distinguish a coupled voltage from a deliberately high impedance source.
The second is any measurement where the loading itself changes the thing you are trying to see. Measuring across a partially open contact, across a high resistance joint, or on a circuit protected by a residual current device can all go differently with a load attached. The third is any situation where you have not yet established that the conductor is safe to load at all, which is why the sequence always starts with a non-contact check rather than with the mode switch. The overview of non-contact voltage detectors covers that first step, and measuring alternating voltage correctly covers the probe technique that follows it.
What the four listings here say about stray voltage
| Instrument | Low impedance mode stated | Non-contact detection stated | Input impedance value stated |
|---|---|---|---|
| Fluke 117 multimeter | Yes, described as preventing false readings from ghost voltage | Yes | Not stated in the listing bullets |
| VEVOR true RMS multimeter | Yes, listed among the measurement functions | Not stated in the listing bullets | Not stated in the listing bullets |
| JASLEEN clamp meter | Yes, described as identifying and eliminating ghost or stray voltages | Yes, selectable under any function | Not stated in the listing bullets |
| Klein Tools NCVT1P pen | Not applicable, it makes no connection | Yes, stated across a 50 to 1000 V alternating range | Not applicable |
Three of the four name the feature and none of them publishes the impedance it presents, which is the figure that would let you predict the collapse. That is a common gap rather than a defect in any one listing, and the manual usually carries it. The Klein Tools NCVT1P belongs in the table for the opposite reason: it states a bright indicator that shows green when the tester is working and flashes red with an audible alert when voltage is detected, which makes it a good first step and a hopeless last one, because it responds to coupled fields just as readily as to connected ones.
Proving a conductor dead without fooling yourself
The order that holds up is always the same. Prove the tester works on a known source. Check the conductor with the tester. Isolate, lock off and apply your own warning. Measure with a connected instrument between every pair of conductors and between each conductor and earth, not just the pair you suspect. Switch the low impedance mode in and watch what the numbers do. Then prove the tester still works on the known source afterwards, because a tester that failed silently during the job invalidates everything you did with it.
The behaviour of the numbers under load is the actual evidence. A reading that falls from most of the supply voltage to a fraction of a volt the moment a few kilohms appear was a capacitive artefact. A reading that falls only part of the way is telling you about a real but resistive source and deserves investigation. A reading that does not move at all means the conductor is connected to something with current behind it, and the only correct next action is to stop and find out what. For the direct current side of the same discipline, measuring direct voltage properly covers why the coupling effect largely disappears there.
Frequently Asked Questions
Is a ghost voltage dangerous to touch?
The voltage is real but the source behind it can supply almost no current, so in most cases contact produces a mild sensation rather than a hazardous shock. That is not a reason to treat it casually. You usually cannot tell by looking whether the reading in front of you is a coupled artefact or a live conductor fed through a high resistance fault, and the correct response to both is the same: load the input down, repeat the measurement, and lock off before proceeding.
Why does a non-contact pen also respond to a ghost voltage?
Because a non-contact pen works by sensing exactly the same capacitive coupling. It detects the alternating electric field around a conductor through a tiny capacitance between that conductor and its own sensing electrode. A conductor sitting at a coupled potential radiates a field in the same way a connected one does, just more weakly, so the pen alerts. A pen confirms presence, never absence.
Does low impedance mode damage anything?
On a mains distribution circuit it should not, because the load it presents draws a fraction of an ampere at normal voltages and the mode is time limited in most instruments to keep the internal resistor cool. The risk lives elsewhere. Place the same load across a control circuit, a signal pair, an intrinsically safe loop or an electronic sensor output and you can trip a relay, false a protection input or injure the circuit you were diagnosing.
Can I get the same effect with a wiggy style solenoid tester?
Broadly yes, since a solenoid tester has always presented a low impedance by its nature. That is why those instruments never showed ghost voltages in the first place. What they lack is the resolution, the range of functions and the display behaviour of a modern meter, and the solenoid draws a much heavier current. Low impedance mode exists so that one instrument can behave like both when you ask it to.
Does the length of the cable run change how big the ghost reading is?
Strongly. The coupling capacitance between two conductors sharing a route grows roughly in proportion to how far they run together, so a long shared conduit produces a larger coupled voltage than a short one. Spacing, the dielectric between them and whether an earthed conductor runs in between all matter too. The same pair of circuits in two different buildings can produce very different ghost readings.
What should I do if the reading barely drops in low impedance mode?
Treat the conductor as live and stop measuring until you have found out why. A voltage that survives a few kilohms across it is being supplied by something with real current behind it. That could be a circuit you have not isolated, a shared neutral carrying current from another circuit, a backfeed through an interlock, or a generator or inverter source you did not know was present.