Key takeaways
- Contact resistance is the resistance across a closed contact interface, not the bulk resistance of the metal.
- It has two parts: constriction resistance from the small real contact area, and film resistance from surface layers.
- Use a four-wire measurement with the voltage probes inside the current path, and state the current and force.
- Accept on a limit plus a trend over life, not on one reading from a new part.
What contact resistance is
Contact resistance is the resistance measured across a closed contact interface. It includes the resistance of the contact material, the resistance where two surfaces meet, and the resistance of any film between them. It does not include the bulk resistance of the carrier or the wiring, unless the measurement is set up to include them.
The distinction matters because the same milliohm figure can mean different things depending on where the probes sit. A reading taken across a rivet head and the blade includes the contact interface, the bond line and the clinch. A reading taken across two closed contacts includes only the face-to-face interface. Both are useful, but they are not interchangeable.
Contact resistance is also not insulation resistance, and it is not the same as the resistance of the contact material measured on a bar. It is a property of the joint, and it changes with force, surface condition, temperature and history.
Constriction and film resistance
Two surfaces never touch over their whole apparent area. They touch at a small number of asperities, and the current is forced through those small spots. The resulting constriction resistance is set by the resistivity of the material and the size and number of the spots. Smoother surfaces, higher force and softer materials increase the real contact area and lower the constriction resistance.
Film resistance comes from oxides, sulfides, tarnish and contamination. A film can be thin enough to conduct by tunnelling, or thick enough to act as a partial insulator. The film is why a contact that looks clean can read high, and why a small amount of sliding or wiping at make can drop the reading sharply.
In a real contact, the two add together. A high-force contact with a thin film is dominated by constriction. A low-force contact with a tarnished face is dominated by film. That is why the same material can behave well in one mechanism and poorly in another.
The constriction formula and an illustrative number
For a single circular contact spot, the constriction resistance is approximately rho divided by twice the spot radius, where rho is the resistivity of the contact material. For several spots in parallel, the resistance falls roughly in proportion to the number of spots. The formula is a model, not a prediction, because the number and size of the spots are not known in advance.
As an illustration, take copper with a resistivity of about 1.72 times ten to the minus eight ohm metres and a single spot radius of 50 micrometres. The constriction resistance is about 0.17 milliohms. If the real contact is spread over ten such spots, the model gives about 0.017 milliohms. Both figures are illustrative and ignore film resistance, which can be larger than either.
The useful part of the model is the direction, not the absolute value. Higher resistivity raises the resistance. A larger real contact area lowers it. More force usually increases the real area. Harder materials and rougher surfaces reduce it for the same force.
The four-wire method
Use a four-wire, or Kelvin, measurement. Two leads carry a known current through the contact, and two separate leads measure the voltage across it. Because the voltage leads carry almost no current, the resistance of the leads and the probe contacts does not appear in the result. A two-wire meter includes them, and on a milliohm measurement that error can be larger than the value being measured.
Place the voltage probes inside the current probes, as close to the interface as the geometry allows. On a closed contact pair, the voltage probes should sit on the two contact members, not on the terminals. On a rivet assembly, one probe on the contact face and one on the blade near the clinch is the usual arrangement.
Stabilise the reading before recording it. Contact resistance can drift for a few seconds after the current is applied, especially at low force. Take several readings, rotate the sample or the probes, and record the spread rather than the best number.
A two-wire reading is not a contact resistance
If the meter includes the leads, the probes and the terminal interfaces, the result is the resistance of the whole loop. On a contact that reads a few milliohms, the lead resistance can dominate. Use four wires and check the zero with a shorting bar.
Dry circuit or loaded current
A dry circuit measurement uses a low open-circuit voltage and a low current so that the film is not broken down by the measurement itself. It is the right test for signal contacts and for detecting a film that would cause a failure at low signal levels. Standards for connectors and relays define the exact conditions, and the test report should state them.
A loaded measurement passes the rated or a specified current through the contact and records the voltage drop. It is closer to the service condition and it reveals the effect of heating. It can also break down a thin film, which means it may read lower than a dry circuit test on the same part.
Both tests are useful, but they answer different questions. Dry circuit tells you whether the contact will work at low signal. Loaded current tells you how much heat the joint will generate. If the application switches a load, test at the load. If it carries a signal, test at dry circuit and state the limit.
| Method | What it measures | Best for | Watch |
|---|---|---|---|
| Dry circuit, four wire | Film and constriction at low energy | Signal and low-level contacts | Films that break down under load |
| Loaded current, four wire | Voltage drop at service current | Power contacts and joints | Heating changes the reading |
| Millivolt drop | Voltage across the joint at current | Assemblies and busbar joints | Probe placement and lead length |
| Micro-ohmmeter, two wire | Loop resistance including leads | Quick checks only | Lead and probe error |
State the current, the open-circuit voltage, the force and the probe positions with every result.
What changes the reading
Force is the largest single influence. More force means more real contact area and usually a lower reading, until the contact yields or the spring takes a set. Measure at the force the mechanism actually applies, not at a convenient bench force.
Surface condition is next. Oxide, tarnish, plating thickness and contamination all add film resistance. A part that has been handled without gloves, or stored in a humid area, can read higher than the same part from a sealed bag. Clean the probes and use a consistent handling procedure.
Temperature and history also matter. Resistance rises with temperature for most metals, and a contact that has been arcing carries a different surface from a new one. A reading taken after a life test is more useful than a reading taken before, because it includes the damage the application will cause.
- Contact force and spring condition
- Oxide, tarnish, plating and contamination
- Probe material, shape and position
- Test current and open-circuit voltage
- Temperature and time under load
- Previous arcing or mechanical operations
Acceptance criteria and records
Set two limits: an absolute maximum at the start of life, and a maximum drift or absolute value at the end of life. The first proves the assembly is built correctly. The second proves it will still work after the application has aged it. A single limit on a new part checks only the first.
Record the test conditions with the result. The current, the voltage, the force, the probe positions, the temperature and the instrument are all part of the measurement. Without them, the number cannot be compared with a supplier's data or with a later batch.
For a production check, a sample per lot is usually enough if the process is stable. For a new design, test across force, temperature and life. If the reading is close to the limit, treat that as a design margin problem rather than a test problem.
Five common mistakes
The first is measuring with a two-wire meter and calling it contact resistance. The second is probing the terminals instead of the contact members, which adds the terminal and wiring resistance. The third is testing at a force that the mechanism never applies.
The fourth is recording only the average. Contact resistance is variable, and a mean can hide a few high spots that will fail in service. Report the maximum and the spread as well. The fifth is testing only new parts. The damage that matters happens during life, so a new-part reading is a baseline, not a qualification.
Frequently asked questions
What is a good contact resistance value?
It depends on the contact size, material, force and application. There is no universal number. Set a limit from the allowed voltage drop and temperature rise, then confirm it with a life test at the real load.
Why is my four-wire reading different from my two-wire reading?
The two-wire reading includes the lead, probe and terminal resistance. On a milliohm contact that can be most of the value. The four-wire method removes those contributions and is the correct method for contact resistance.
Should I test with the contact closed and the load applied?
Test at the condition the application uses. A signal contact needs a dry circuit test. A power contact needs a loaded test that reproduces the current and the heating. Testing only one condition can miss the failure mode that matters.
Does contact resistance change over life?
Yes. Arcing, mechanical wear, film growth and relaxation of the spring all change it. Measure at the start and at intervals through the life test, and set an end-of-life limit as well as an initial limit.
Sources and further reading
- IEC 60512-2-1, Connectors for electronic equipment, test 2a: contact resistance
- ASTM B539, Standard test methods for measuring resistance of electrical connections
- IEC 61810-7, Electromechanical elementary relays, test and measurement procedures
Qumek Engineering Desk
Engineering team
The engineering team behind Qumek drawings, DFM notes, tolerances and inspection standards. Every guide is reviewed against the drawings and processes used on live programmes.
Setting a contact resistance limit?
Send the contact geometry, the force and the allowed temperature rise. The engineering desk will reply with a test method and a limit that matches the application.
Send the contact data