Lead Wire Effects on the Calibration of Resistive Sensors
Lead wires to resistive temperature sensors present a problem for calibration because they add a small but possibly significant value to the sensor resistance. This additional resistance is usually not in the temperature environment being measured and will, usually, have a different temperature coefficient than the main sensor.
Customer readout instrumentation will normally use one of several methods for handling this potential source of error.
Sensors with two lead wires offer no way to directly measure lead resistance as they are connected directly at the sensing element with no access to this junction.
In use, the additional resistance of leads is either accepted by readout circuitry as a small offset or, in some cases, ignored.
Although it cannot be measured directly, the resistance of a single lead wire is calculated as part of the calibration service and reported in the event that this value is needed for customer application.
Sensors with three lead wires (two connected to one side or the element, one to the other) are intended to compensate for lead resistance by using circuitry to measure the actual resistance of the extra wire and subsequently subtracting this from the element resistance.
As a result, after subtracting the known lead resistance, the resistance-to-temperature formula used by the application can assume no offset or addition by leads. (There is a very small remaining error due to assuming the measured lead is identical to the unmeasured lead, but this is usually ignored.)
To accommodate this usage, Thermalogic calibration result tables for three wire sensors provide element resistance versus temperature less the measured lead value. In case it is needed, the measured resistance of a single lead (on the two-wire side) is also reported as part of the calibration service.
The most accurate lead wire configuration for sensor readout is four wires, two connected to each side of the sensing element. With these leads, the actual lead resistance can be measured by application circuitry and subtracted in use from the total value.
Alternatively, and more common, one set of leads is used to feed a constant, known, current through the element and the other set of leads is used, with high impedance circuitry, to measure the voltage across the element. The “pure” element resistance is than calculated by dividing the measured voltage by the known current.
To accommodate this usage, Thermalogic calibration result tables for four wire sensors provide element resistance versus temperature using a four-wire measurement method. This provided a result for the isolated element as required by a four-wire application circuit.