The instrument consists of two platinum resistance (50 ohm) wires in a heated housing. Both elements are housed in a single external housing. The temperature-dependent resistance of each platinum wire is measured and used to determine its temperature. The wire is partially shielded from hydrometeor impacts, but the sensing wire can still become wet in some clouds. This may lead to erroneously low measurements, as discussed in the State Parameter discussion in the Documentation section. The sensor is also heated to minimize the accumulation of ice, but can occasionally become blocked by ice, especially in cases of very high concentrations of small ice crystals. (See the references at the end of this instrument description.)

The measurement of temperature involves not only this sensor but also the measurements of dynamic and static pressure, which are required to correct for dynamic heating. The "recovery factor" for this sensor is estimated to be about 0.98 on the basis of flight maneuvers during which the airspeed varied while the temperature of the air was assumed constant. The absolute calibration has been checked by integration of the hydrostatic equation between various levels where pressure and geometric height are measured accurately; these tests lead to an overall uncertainty estimate of about 0.5 C in the measured temperature. Additional errors occur in cloud because the sensor becomes wet and experiences evaporative cooling in the dynamically heated airflow entering the sensor.

The following image shows the appearance of the sensor that is attached to the outside of the aircraft fuselage: