Highly accurate and stable temperature measurement is possible with a PT100 temperature sensor between -200 and +850°C, which makes it a popular choice in many industrial applications. In particular, the high degree of accuracy across the -50 to 150ºC range means that it is the preferred choice for temperature measurement in Pharmaceutical applications, such as in sterilisation processes.
The following guidelines will help users to specify the correct configuration.
They are written with Pharmaceutical and Healthcare applications in mind although the concepts apply across all industry sectors.
What is a PT100 Temperature Sensor?
The PT100 sensor is a popular example of a resistance temperature detector (RTD). It is based on the principle observed in metals whereby a change in temperature will cause a change in the resistance of the material. Platinum exibits a positive temperature coefficient I.e it shows an increasing resistance with increasing temperature, and so is the most common material used to construct RTD’s.
The PT100 sensor is designed to have a resistance of exactly 100Ω at 0.0°C, and a resistance increase of 0.385Ω per 1°C increase between 0 and 100°C according to ISO 60751: 2008
PT100 sensors are commonly constructed using two methods:
Wire wound sensors
Platinum wire wound sensors consist of a thin platinum wire loosely wrapped around, or threaded within, a ceramic core.
Wire wound sensors can be used over a wide range of temperatures, however they can be susceptible to mechanical shock, which induces measurement drift.
Thin Film Sensors
Thin film sensors are based upon a ceramic substrate with a deposition of high purity platinum, laser etched to give 100Ω at 0.0°C. This is then sealed within a glass adhesive.
These sensors are cheaper than wire wound detectors, and are less sensitive to impact damage. However they operate within a smaller temperature range than wire wound sensors.
PT100 sensors are commercially available to several different tolerance levels, according to BS EN 60751:2008, as class B, A and AA in order of increasing precision. In addition 1/10 DIN sensors are available which are picked to ensure a tolerance band 1/10th that of a Class B detector.
Selecting the Right PT100 Detector
This is a balance between competing factors. For example, selecting a PT100 sensor for use in a pharmaceutical autoclave chamber, it is a balance between the optimum tolerance of the temperature measurement and the resilience of the sensor to frequent handling by process operators.
So a class A thin film sensor would be most appropriate for pharmaceutical applications, whereas a class B will have sufficient tolerance for most chemical or manufacturing applications. Maximum service temperature may also dictate a ceramic detector in preference to a thin film sensor.
Impact of Probe Lead Length on PT100 Measurement Accuracy
PT100 sensors are connected to the measuring instrumentation using any one of three different connection principles:
• Two wire, with no lead wire compensation
• Three wire with partial compensation
• Four wire with full compensation.
The PT100 sensor will be located at the desired measurement point which may be some distance from where the measuring instruments are located. The length of the connection wire can influence the accuracy of the reading as described below.
Two wire connection is the simplest wiring method. Since the variable output of a PT100 sensor is resistance, it follows that the resistance of the connecting leads between sensor and instrument will also have an impact on the final measurement, and therefore the temperature inferred at the sensor.
Where the lead lengths are short, and where they are exposed to the same temperature as the sensor, then in theory this can be accounted for.
However in, for example, a Pharmaceutical autoclave chamber, where the lead length between the sensor and the instrument can be very long, a two wire connection would lead to significant measurement error.
A high proportion of the leads will be inside the chamber during the sterilising cycle, and therefore at the same elevated temperature as the sensor. The remaining lead length will be outside the chamber and at ambient temperature. These temperature differences will cause resistance change in the lead wire conductors contributing to poor reading accuracy.
The 3 wire compensation method was therefore adopted using a modified Wheatstone bridge. In this configuration the sensor lead wires are connected to the opposite sides of the bridge, effectively compensating for each other, with the third wire supplying power to the bridge.
However this will produce a thermal gradient along the leads themselves. Although the effect of Joule heating is reduced, it cannot be eliminated altogether because the heat transfer conditions at the sensor will be different from those of the matching 100Ω resistor in the bridge circuit.
To enable more accurate measurements to be made, particularly when the connecting lead wires are relatively long and passing through varying ambient temperatures, a 4 wire connection system was developed.
One pair of lead wires takes the constant current power source to the sensor and the other pair is used to measure the actual voltage drop across the sensor. Therefore by using a constant current source and being able to simply measure the change in voltage across the PT100 sensor rather than a change in resistance (Ohms Law), any fixed or varying lead wire resistance is totally eliminated.
It is worth noting that when calibrating either a 3 or 4 wire PT100 sensor in an oil bath or hot block, the difference in readings between these two connection types may appear to largely disappear. This is due to there being no temperature effect on the leads of the 3 wire connection.
The realistic choice for a PT100 sensor is therefore between a 3 and 4 wire system, and the user may be constrained by the limitations of existing instrumentation that may already be in place.
Where possible, a 4 wire system would be preferable to ensure the most accurate reading.
The use of a 3 or 4 wire PT100 temperature transmitter is a further option to reduce the overall lead length and convert the probe reading to a 4-20mA signal which will be easily integrated into process instrumentation.