Four-wire measurement separates the job of delivering current from the job of sensing voltage, which is the only practical way to measure small resistances. An ordinary two-wire reading includes the leads, the plug contacts, the probe tips and the oxide layer where they touch the workpiece. Below about an ohm, that collection is no longer negligible.
What a two-wire reading really contains
A conventional ohmmeter pushes current out one lead, through everything in the path, and back along the other. It measures the voltage across its own terminals and reports the whole loop. That loop includes several centimeters or meters of test lead, the connection at the banana jack, the contact where the probe tip meets the metal, and any surface film in between. Together these commonly amount to a fraction of an ohm and vary every time you reconnect. Measuring a ten-kilohm resistor, that contribution is invisible. Measuring a hundred-milliohm bond, it can exceed the thing you are trying to measure. Most meters offer a relative or zero function that subtracts the leads with the probes shorted, which helps but does not address contact resistance at the workpiece, and that part changes with every placement.
How separating the functions removes the error
In a four-wire arrangement, one pair of leads carries the test current and a separate pair senses voltage. The sense pair connects to a high-impedance input, so almost no current flows in it, and almost no voltage is lost along its length or in its contacts. The instrument therefore reads the voltage that exists between the two sense contacts, divided by the current it knows it is forcing. Whatever resistance sits in the current leads and their contacts affects how much voltage the source must produce but does not appear in the result. The sense contacts must lie inside the current contacts, so that the region between them contains only the component under test and none of the current path’s own resistance.
Where the clips go matters more than the instrument
Because the measurement is defined by where the sense contacts touch, clip placement is the measurement. Place the sense clips outside the current clips and you are back to including lead resistance. Place them a centimeter further along a busbar and you include a centimeter of busbar. On a component with leads, deciding whether to sense at the body or at the far end of the lead changes the answer, and the convention that applies depends on how the specification was written. Kelvin clips, which carry both connections in one jaw with the contacts electrically separated, exist to make this repeatable. Photograph or sketch the placement when a result will be compared later, since a difference of a few millimeters can explain a disagreement between two technicians.
What four wires does not fix
The technique removes lead and contact resistance and nothing else. Thermal voltages generated where dissimilar metals meet in the sense path still add an offset, which is why good low-resistance instruments reverse the current and average the two readings to cancel them. Self-heating from the test current still changes the component’s resistance while you watch. Interference picked up on the sense pair still corrupts the reading, so the pair should be twisted and kept away from power conductors. And the technique does nothing about a component that is simply different from what you expected, or about a contact that is intermittent rather than merely resistive. Four-wire is a solution to one specific problem, and confusing it with general accuracy leads to misplaced confidence.
Deciding when two wires are enough
The practical threshold is a judgment about ratio, not an absolute number. If lead and contact resistance might be a few tenths of an ohm, then measuring a hundred-ohm resistor to one percent is unaffected, while measuring a one-ohm resistor to one percent is impossible. Work out what fraction of your expected reading the parasitic resistance represents, and if it exceeds the accuracy you need, go to four wires. Typical applications that demand it include bonding and continuity of protective conductors, shunts and current-sense elements, motor and transformer winding resistance, switch and relay contact resistance, and cable joint verification. In each of those the quantity of interest is small and the consequence of an inflated reading is a decision made on bad information.
What to check before you act
- Place sense contacts inside the current contacts, never outside.
- Twist the sense pair and route it away from power conductors.
- Use current reversal or averaging where thermal offsets are a concern.
- Record exactly where the sense contacts were placed.
Common questions
Can I fake four-wire with an ordinary multimeter?
You can force a known current from a separate supply and measure the voltage with the meter on its volts range, which is the same principle. Accuracy then depends on how well you know the current.
Does the relative or zero function replace four-wire measurement?
No. It subtracts the leads as they were at the moment you zeroed, but contact resistance at the workpiece changes every time you reconnect.
The practical takeaway
Use four wires whenever the parasitic resistance of leads and contacts is a meaningful fraction of what you are measuring, and remember that the sense clips define the result. Keep them inside the current clips, twist the pair, watch for self-heating, and record the placement so someone else can reproduce your number.
References and further reading
- OSHA: 1910.304 wiring design and protection, including grounding
- NFPA: electrical safety resources and code background
- CDC/NIOSH: electrical hazards and safe work practices
Numerical scenarios are illustrative unless identified otherwise. Follow the exact product instructions; component ratings and local installation requirements can differ.
