Self-heating during measurement is one of the quieter reasons a reading refuses to settle. Any instrument that measures resistance passes current through the component, that current dissipates power, and the resulting temperature rise changes the very property being measured. The display drifts, and the drift is caused by the measurement itself.
Why the number moves while you watch
Resistance measurement works by forcing a known current and measuring the voltage, or by applying a known voltage and measuring current. Either way, power equal to the product of the two is dissipated in the component. Most materials change resistance with temperature, so as the component warms, the reading changes. Metals and most wirewound components increase in resistance as they heat. Semiconductor-based sensors and many carbon compositions move the other way. The rate of drift depends on the component’s thermal mass and how well it is heat-sunk, which is why a small surface-mount part drifts within a second while a large wirewound resistor may take a minute. A reading that slides steadily in one direction after connection is the classic signature.
Components where this dominates the measurement
Low-value resistors and shunts are the obvious case, because measuring a small resistance accurately usually means forcing a substantial current, and a small resistance dissipating that current gets warm quickly. Thermistors are designed to change resistance with temperature, so the effect is dramatic: measuring one with too much test current produces a reading that reflects the meter as much as the environment, and manufacturers state a dissipation constant precisely so users can calculate the self-heating error. Filament lamps are a different extreme, since their cold resistance is a small fraction of their hot resistance, and any measurement that heats the filament at all begins climbing immediately. Motor windings, transformer windings and cable runs all show the effect on a slower timescale.
Recognizing it rather than chasing it
Self-heating has a recognizable behavior. The reading changes monotonically after connection and then stabilizes at a new value. Disconnect, let the part cool, and reconnect, and the drift repeats from the same starting point. Changing the instrument’s range often changes the test current and therefore changes how fast and how far the drift goes, which is a strong confirmation. Contrast that with a poor contact, which produces erratic jumps rather than a smooth slide, or with a thermal voltage in the leads, which drifts slowly in either direction and does not reset when you disconnect. Distinguishing these three is worth doing before deciding a component has changed value, because all three look like instability on a display.
Taking the reading before the heat arrives
The standard remedy is to reduce the energy you inject or the time you inject it for. Instruments designed for low-resistance measurement often offer a pulsed mode that applies current briefly, takes the reading, and stops, which keeps average dissipation low. Where that is unavailable, choosing a range with lower test current helps, at the cost of resolution. For thermistors and other sensitive sensors, the correct approach is to use the lowest test current that gives an adequate signal and to check the manufacturer’s dissipation constant to estimate the remaining error. Winding resistance measurements have their own convention: take the reading quickly after energizing, or record the reading together with the winding temperature so the value can be corrected to a reference temperature later.
When the heating is the measurement
Sometimes self-heating is the point rather than the problem. Measuring a winding’s resistance immediately after a machine has been running is how its average temperature rise is determined, because the resistance change of copper with temperature is well characterized. Hot-wire and thermal sensors rely on self-heating deliberately. In those cases the discipline reverses: you must control and record the current, the timing and the ambient temperature, because they are inputs to the result rather than nuisances. The general lesson holds either way. Test current is energy entering the component, and a measurement is only reproducible if the amount of energy and the moment of reading are part of the recorded method rather than left to chance.
What to check before you act
- Watch whether the drift is smooth and one-directional before suspecting a fault.
- Change range to alter test current and see whether the drift changes with it.
- Use a pulsed or low-current mode for low-value and sensitive components.
- Record ambient temperature and timing when the value must be comparable later.
Common questions
Why does my low-value resistor read higher the longer I hold the probes on?
Test current is heating it and its resistance rises with temperature. Take the reading promptly, or use an instrument with a pulsed low-resistance mode.
How do I measure a thermistor without heating it?
Use the lowest practical test current and check the manufacturer’s dissipation constant to estimate the residual self-heating error at that current.
The practical takeaway
When a resistance reading slides steadily after you connect, suspect your own test current before suspecting the component. Change range to confirm, take the reading early, use pulsed modes where they exist, and record temperature and timing whenever the number needs to be compared with another one.
References and further reading
- CDC/NIOSH: electrical hazards and safe work practices
- NFPA: electrical safety resources and code background
- CPSC: home electrical safety guidance
Numerical scenarios are illustrative unless identified otherwise. Follow the exact product instructions; component ratings and local installation requirements can differ.
