4-Wire vs 2-Wire Resistance Measurement: When Lead Resistance Matters

The Problem: Every Wire Has Resistance

When you connect a multimeter to a resistor and select the ohms function, the instrument passes a small test current through the circuit and measures the voltage. By Ohm's law (R = V/I), it calculates resistance. The problem is that the voltage measurement includes not just the voltage across the resistor under test (the DUT. Device Under Test), but also the voltage dropped across the test leads themselves.

A typical pair of test leads has resistance of approximately 0.05–0.2Ω per wire. Meaning the total lead resistance in series with the circuit is 0.1–0.4Ω. For measuring a 10 kΩ resistor, this is negligible (0.004% error). For measuring a 1Ω resistor, a 0.2Ω lead resistance is a 20% error. For measuring a 10 mΩ motor winding or contact resistance, the lead resistance is often 10–100× larger than the value being measured.

This is the fundamental limitation of 2-wire resistance measurement: it is only accurate when the DUT resistance is large relative to the lead resistance. The conventional rule of thumb is that 2-wire measurement is acceptable when the DUT resistance is at least 100–1000× the lead resistance. Meaning 2-wire works above approximately 100Ω with standard test leads, but should not be used below 10Ω without knowing and correcting for lead resistance.

A related but distinct source of 2-wire error worth understanding separately is contact resistance at the test point itself, as opposed to the resistance of the wire running between the instrument and that test point. Every clip, probe tip, or crocodile clamp makes contact with the DUT through a physical interface that itself has some resistance, influenced by surface oxidation, contact pressure, and cleanliness of the mating surfaces. This contact resistance behaves identically to lead resistance in a 2-wire measurement: it sits in series with the DUT and adds directly to the reading, and it can vary noticeably from one connection to the next even using the exact same test leads, simply because contact pressure and surface condition are not perfectly repeatable by hand. This variability is part of why repeated 2-wire measurements of the same low-resistance component can show scatter that has nothing to do with the component itself, and it is a variability that 4-wire measurement eliminates just as effectively as it eliminates fixed lead resistance, since the sense connection is likewise immune to the current-carrying contact's resistance.

How 4-Wire (Kelvin) Measurement Works

The 4-wire, or Kelvin, measurement method solves the lead resistance problem by using two separate pairs of leads:

• Force (current) leads: the outer pair, which carry the test current from the instrument to the DUT and back.
• Sense (voltage) leads: the inner pair, which connect directly to the DUT terminals and carry the voltage measurement back to the instrument.

The key insight is that the voltmeter (or analog-to-digital converter) measuring voltage has very high input impedance, typically 1 GΩ or greater. This means virtually no current flows through the sense leads. Since V = IR and I ≈ 0 in the sense leads, the voltage drop across the sense lead resistance is negligible regardless of how long or resistive the sense leads are.

The voltage measured is therefore V_DUT = I × R_DUT, with no contribution from lead resistance. The only sources of error remaining are: the accuracy of the reference current source (I) and the accuracy of the voltmeter (V). Both are well-controlled in precision instruments.

In practice, Kelvin connections are made using either:

• Kelvin clips: alligator-style clips with separate current and voltage contacts on each jaw
• 4-binding-post connections: used on resistance standards, decade boxes, and precision resistors
• 4-wire probe sets: used with low-resistance meters (micro-ohmmeter instruments)

Mathematical Foundation

In a 2-wire measurement:

• Total measured resistance = R_lead1 + R_DUT + R_lead2
• Error = R_lead1 + R_lead2 (both lead resistances add directly)

In a 4-wire measurement:

• Current I flows through: R_force1 + R_DUT + R_force2
• Voltage measured: V_sense = I × R_DUT (sense lead resistance not in voltage path)
• Calculated resistance: R = V_sense / I = R_DUT (exact, ignoring sense lead resistance)

The only residual error is if the sense leads have non-negligible resistance and the instrument's input impedance is not truly infinite. At 1 GΩ sense input impedance with 0.5Ω sense lead resistance: the current through sense leads is V/1GΩ ≈ 0.1V/1GΩ = 0.1 nA. Voltage error = 0.1 nA × 0.5Ω = 0.05 nV. Completely negligible for any practical measurement.

RTD Application: Why Calibration Labs Always Use 4-Wire

The most common reason engineers encounter 4-wire measurement requirements is RTD (Resistance Temperature Detector) measurement. A Pt100 RTD has a resistance of exactly 100Ω at 0°C and increases at approximately 0.385Ω per °C. This means:

• At 100°C: Pt100 resistance ≈ 138.5Ω
• At 200°C: Pt100 resistance ≈ 175.8Ω

A lead resistance error of 0.2Ω would cause a temperature measurement error of 0.2Ω ÷ 0.385Ω/°C = 0.52°C. This is often larger than the accuracy specification of the Pt100 element itself (Pt100 Class A = ±0.35°C at 100°C). In other words, the lead resistance error completely destroys the accuracy advantage of using a precision RTD.

For this reason:

• SAC-SINGLAS accredited calibration laboratories always use 4-wire connections for Pt100 calibration
• High-accuracy process instruments (Endress+Hauser, Yokogawa) use 4-wire RTD inputs for their highest-accuracy temperature transmitters
• Industrial transmitters with 3-wire inputs use the 3-wire compensation method as a practical compromise

3-Wire RTD Connection: The Industrial Compromise

The 3-wire connection adds a third wire to allow the instrument to measure and compensate for lead resistance. The instrument measures the resistance of one lead wire (via the third wire) and subtracts this from the total measurement, assuming both main lead wires have equal resistance.

This works well when:

• The three wires are the same gauge, material, and length (equal resistance per leg)
• Ambient temperature is stable (lead resistance changes with temperature)
• Accuracy of ±0.1–0.5°C is sufficient

The 3-wire method fails (and returns to near-2-wire error) when:

• Lead lengths are unequal (common in field installations)
• Temperature gradients exist along the cable run
• Better than ±0.2°C accuracy is required

For any calibration application requiring measurement uncertainty below ±0.1°C, 4-wire connection is mandatory.

Low-Resistance Testing: Milliohm and Micro-ohm Measurement

The 4-wire technique is essential for measuring resistances below 1Ω:

• Motor winding resistance: three-phase motors typically have winding resistances of 0.1–10Ω. Measurement accuracy is important for motor condition assessment and comparison between phases.
• Contact resistance: relay contacts, circuit breaker contacts, switchgear joints. All have contact resistances in the milliohm to ohm range. IEC 62271 specifies maximum contact resistance for switchgear.
• Bond wire resistance: PCB ground bonds and structural electrical bonds must meet resistance requirements (e.g. < 25 mΩ per MIL-STD-1760).
• Cable resistance testing: for continuity verification and comparison against manufacturer specifications.

Instruments for low-resistance measurement include the Cropico DO5000, Tinsley 5840D, and Keithley Model 580 micro-ohmmeter. All of which use 4-wire Kelvin connections and provide accurate measurements from 1 μΩ to 100Ω.

Offset-Compensated Ohms: A Related Technique for Semiconductor and Diode-Junction Measurements

A further refinement relevant to precision low-resistance measurement, particularly on components containing semiconductor junctions or other non-linear elements, is offset-compensated ohms measurement. Some devices under test, diodes, transistor junctions, or components with parasitic diode paths, exhibit a voltage offset even with no current flowing, which a simple resistance measurement cannot distinguish from a genuine resistive voltage drop. Offset-compensated measurement takes two readings at two different test currents, then calculates resistance from the change in voltage divided by the change in current, mathematically cancelling any fixed voltage offset present in the circuit regardless of its source.

This technique is less commonly needed in general industrial resistance testing than the current-reversal method described above, but it becomes relevant when testing components with embedded protection diodes, certain sensor types, or PCB assemblies where a resistance measurement might inadvertently forward-bias a semiconductor junction elsewhere in the circuit and produce a misleadingly low or unstable reading. Precision source-measure units and higher-end bench multimeters offer this as a selectable measurement mode specifically for these situations, and it is worth knowing the technique exists even if it is not the default measurement mode for routine calibration work.

High-Resistance Measurement: When Lead Resistance Doesn't Matter, but Other Things Do

Above approximately 1 MΩ (megohm), lead resistance is typically negligible relative to the DUT resistance. However, high-resistance measurement introduces different challenges:

• Surface leakage currents: on insulation surfaces, particularly in humid conditions, leakage currents can flow across the surface of the insulation, bypassing the bulk resistance being measured. These currents add a parallel conducting path that reduces the apparent resistance.
• Guard electrodes: a third guard electrode surrounding the high-voltage electrode collects surface leakage currents and redirects them back to the source, preventing them from entering the measurement path. This is a 3-terminal measurement, different from the 2-wire/4-wire distinction.
• Electrification time: high-resistance materials (polymers, ceramics) absorb charge slowly. Resistance readings at 1 second differ significantly from readings at 1 minute. IEC 60167 standardises electrification times for insulation resistance measurement.

Thermal EMF and Current Reversal: The Error Even 4-Wire Doesn't Automatically Fix

Eliminating lead resistance error is not the end of the precision resistance measurement story. Whenever two dissimilar metals are joined, at a test clip, a switch contact, a connector, or simply where copper meets a different alloy in a Kelvin clip, a small thermoelectric voltage (the same Seebeck effect that makes thermocouples work) is generated if there is any temperature difference across that junction. This thermal EMF is typically only a few microvolts, but for low-resistance measurements using small test currents, a few microvolts is far from negligible relative to the voltage being measured, and a 4-wire connection alone does nothing to cancel it out.

The standard technique to eliminate thermal EMF error is current reversal: the instrument takes one measurement with the test current flowing in one direction, then reverses the current and takes a second measurement, then averages the two readings. Because the thermal EMF voltage is independent of current direction (it depends only on temperature difference at the junctions) while the voltage due to the resistance under test reverses polarity along with the current, averaging the two readings cancels the thermal EMF contribution while preserving the true resistance signal. Precision micro-ohmmeters and resistance bridges used in accredited calibration laboratories perform this current-reversal technique automatically, and it is one of the less visible reasons a laboratory-grade low-resistance measurement achieves meaningfully better repeatability than a general-purpose 4-wire measurement made with a standard bench multimeter, even when both use the identical Kelvin connection principle.

Temperature Coefficient of Resistance: Why Precision Resistors Need Environmental Control

A resistor's resistance value is not fixed; it changes with the resistor's own temperature, described by its temperature coefficient of resistance (TCR), typically expressed in parts per million per degree Celsius (ppm/°C). A general-purpose resistor might have a TCR of several hundred ppm/°C, while a precision resistance standard used as a calibration laboratory reference is manufactured and selected specifically for a very low TCR, often single-digit ppm/°C, so its value remains stable across normal ambient temperature variation.

This matters directly for 4-wire measurement accuracy because the measurement current itself causes a small amount of self-heating in the resistor under test, exactly analogous to the self-heating effect discussed for RTDs, and a resistor with a high TCR will show a measurably different resistance value depending on how much test current is applied and how long it has been flowing before the reading is taken. Accredited laboratories account for this by using appropriately low test currents for the resistance value being measured, allowing adequate settling time before recording a reading, and, for reference-grade resistance standards, controlling the laboratory's ambient temperature tightly enough that the standard's own TCR does not introduce a meaningful error into the calibration being performed with it.

Calibration Implications: Resistance Standards and Decade Boxes

Resistance standards (including precision resistors, resistance decade boxes, and shunt resistors), are always calibrated and specified in 4-wire (Kelvin) configuration. This means:

• The calibration certificate value applies only when the standard is measured using 4-wire connection
• If used in a 2-wire circuit, the reading will be higher by the lead resistance of the circuit
• Connection adaptors (e.g. BNC to 4-binding-post) should be calibrated as part of the measurement system if they introduce significant resistance

When a resistance standard is calibrated at a SAC-SINGLAS accredited laboratory, the certificate will specify the terminal configuration used. Engineers using the standard in field applications must use the same terminal configuration to realise the stated calibration value.

This distinction matters in practice more often than it might appear, particularly when a resistance standard or a decade box is handed between departments or between an internal calibration function and an external contractor. A standard calibrated and certified using 4-wire terminals, then subsequently used by someone unaware of that terminal requirement in a 2-wire circuit, will produce a reading that appears to disagree with its own calibration certificate, not because the standard has drifted, but because it is being used in a configuration different from the one the certificate actually describes. Building a habit of checking a resistance standard's certificate for its stated terminal configuration before use, and matching your own measurement setup to it, prevents this entirely avoidable class of apparent calibration discrepancy.

Connection Method Comparison

When to Use 4-Wire Measurement

Use 4-wire (Kelvin) measurement when:

• Resistance is below approximately 1 kΩ and accuracy is important
• Measuring RTDs (Pt100, Pt1000) with uncertainty requirements below ±0.5°C
• Measuring motor windings, contact resistance, or bond resistance in the milliohm to ohm range
• Calibrating or verifying resistance standards and decade boxes
• The calibration application is SAC-SINGLAS or ISO/IEC 17025 accredited (4-wire is mandatory for low-resistance reference standards)

Use 2-wire measurement when:

• Resistance is above 10 kΩ (lead resistance is negligible)
• Quick continuity checking or go/no-go testing
• Lead resistance can be measured and subtracted (via null/compensation function)
• Accuracy requirements are coarse (>1% is acceptable)