To measure a thermocouple, you need a multimeter set to either the resistance (ohms) or millivolt DC setting, depending on whether you’re checking for a broken sensor or verifying its output. A healthy thermocouple produces a small voltage when its tip is heated, and a simple resistance check can tell you in seconds whether the wires inside are intact.
What a Thermocouple Actually Produces
A thermocouple is two different metal wires joined at one end. When that junction is heated, the difference in how the two metals respond to temperature creates a tiny voltage, measured in millivolts. A Type K thermocouple at 100°C, for example, produces roughly 4 millivolts. Your furnace controller, oven, or industrial instrument reads that voltage and converts it to a temperature.
This means there are two useful things you can measure: the resistance across the wires (to check if they’re physically intact) and the millivolt output (to check if the thermocouple is actually generating a signal when heated).
Resistance Test: Checking for a Broken Sensor
This is the fastest way to find out if a thermocouple has failed. Set your multimeter to the ohms (Ω) setting and touch the two test leads to the two thermocouple wires. Polarity doesn’t matter for this test.
If the meter reads “OL” (open loop) or zero in a way that indicates no circuit, at least one of the thermocouple’s internal wires is broken. That thermocouple is dead and needs replacing.
A working thermocouple will show some resistance. A short assembly, just a few feet long, might read only a few ohms. Longer runs will read higher. As a reference point, a 4-foot length of 20-gauge Type J thermocouple wire measures about 1.4 ohms. Assemblies with longer leads can read 20, 30, or 50 ohms or more. The key takeaway is that you’re looking for a reading that isn’t open circuit. If you get a number in a reasonable range, the wires are intact.
Millivolt Test: Verifying the Output
A resistance test only tells you the wires aren’t broken. To confirm the thermocouple is actually generating a temperature signal, you need to measure its voltage output while heating the tip.
Set your multimeter to the DC millivolt (mV) range. Connect the leads to the thermocouple wires. At room temperature, you’ll see a very small reading, often just a fraction of a millivolt. Now apply heat to the thermocouple’s tip with a lighter, heat gun, or by dipping it into hot water. You should see the millivolt reading climb steadily as the temperature increases. If the reading doesn’t change when you heat the junction, the thermocouple has failed even though the wires may still be continuous.
For a rough accuracy check, boiling water works well as a known reference. At sea level, water boils at 100°C (212°F). A Type K thermocouple at that temperature should produce approximately 4.1 millivolts. A Type J should produce about 5.3 millivolts. These numbers come from standard reference tables published for each thermocouple type, and you can find them online by searching for your specific type’s millivolt-to-temperature chart.
Ice Bath Calibration for Accuracy
If you need to verify that your thermocouple reads accurately at a known low point, an ice bath gives you a reliable 0°C (32°F) reference. This method is simple but requires careful setup to get a true 0°C temperature.
Fill a clean container with crushed ice all the way to the top, then add cold water until the level sits about half an inch below the top of the ice. The ice should not be floating. If it floats, pour off some water and add more ice. You want a dense slush where the water fills the gaps between ice pieces.
Insert the thermocouple probe into the ice bath and stir gently for at least one minute. Keep the probe away from the walls and bottom of the container, since those surfaces can be slightly warmer than the ice-water mixture itself. After the reading stabilizes, it should show 0°C or 32°F within a degree or two. If your reading is off by more than that, the thermocouple may be degrading or your measurement instrument may need recalibration.
Why Your Multimeter Reading Differs From the Display
If you’ve ever measured a thermocouple’s millivolt output with a basic multimeter and then tried to convert that reading to temperature using a reference table, you probably noticed the number didn’t match what the thermocouple’s controller was displaying. That’s because of something called cold junction compensation.
A thermocouple generates voltage based on the temperature difference between its hot end (the measurement tip) and its cold end (where the wires connect to your instrument). If the connection point is at room temperature, say 22°C, the thermocouple is only producing voltage for the difference between the process temperature and 22°C, not the full temperature. Standard reference tables assume the cold end is at 0°C.
Dedicated thermocouple instruments and controllers have a built-in temperature sensor at their connection terminals. They measure the ambient temperature there and electronically add (or subtract) a correction voltage so the final reading reflects the true temperature. A basic multimeter doesn’t do this, so your raw millivolt reading will always be slightly low. For a quick go/no-go test this doesn’t matter much, but for precision work, you’ll either need a meter with built-in thermocouple mode or you’ll need to manually add the ambient offset using a reference table.
Using a Thermocouple-Capable Meter
Many modern multimeters have a dedicated thermocouple input, usually a small round connector or a miniature plug that matches the thermocouple’s connector type. When you plug the thermocouple directly into this input, the meter handles cold junction compensation automatically and displays the temperature in degrees rather than millivolts.
If you go this route, make sure the meter’s thermocouple setting matches the type you’re measuring. A Type K thermocouple plugged into a meter set for Type J will give you a wrong reading because each type has a different voltage-to-temperature relationship. Most meters default to Type K since it’s the most common.
Common Reasons a Thermocouple Fails the Test
An open circuit on the resistance test usually means the wires have broken, often from repeated thermal cycling, vibration, or corrosion. This is the most common failure mode and the easiest to diagnose.
A thermocouple that shows continuity but gives a weak, erratic, or flat millivolt output has likely degraded at the junction. Prolonged exposure to high temperatures gradually changes the metal composition at the tip, a process called drift. The thermocouple still technically works but reads increasingly inaccurate temperatures. Industrial thermocouples in high-temperature environments, particularly those running near the upper limits of their sheath materials (900°C for standard stainless steel sheaths, up to 1,200°C for high-nickel alloy sheaths), degrade faster and need more frequent testing.
Short circuits can also occur when the insulation between the two wires breaks down, allowing them to touch at a point other than the intended junction. This creates a second, unintended measurement point and produces erratic readings. A resistance test may show an unusually low value compared to what you’d expect for the wire length.
Identifying Your Thermocouple Type
Before you can interpret millivolt readings or select the right setting on a thermocouple meter, you need to know what type you have. The wire insulation is color-coded, but the coding system depends on whether the thermocouple follows the American (ANSI) or international (IEC) standard.
- Type K is the most widely used general-purpose thermocouple, covering roughly negative 200°C to 1,250°C. In the ANSI system, the positive wire is yellow and the negative wire is red. IEC coding uses green and white.
- Type J works well for lower temperatures, up to about 750°C. ANSI colors are white (positive) and red (negative). IEC uses black and white.
- Type T is common in food service and laboratory work, effective from negative 200°C to about 350°C. ANSI uses blue and red.
The red wire is always the negative conductor in the ANSI system, which trips people up since red usually implies positive in other electrical work. If the color coding has faded or been stripped away, check the connector. Thermocouple connectors are shaped so they only fit the matching type, and the connector body is typically color-coded to match the thermocouple type as well.