The drift in Pt100 is what separates two Class A sensors on paper from two sensors with very different performance after months in operation. Both leave the factory with a calibration certificate, meet IEC 60751 Class A standards, and, at the same calibration points, indicate virtually the same temperature.
At first glance, they seem equivalent. After a few months of operation in an industrial plant, however, one continues to indicate almost the same temperature as on the day of installation, while the other already shows a slight drift, but one capable of compromising a critical process.
They both met the same standard. So why do they age differently?
What does IEC 60751 Class A actually certify?
IEC 60751 defines the relationship between electrical resistance and temperature for platinum resistance thermometers, as well as the tolerances for each accuracy class: AA, A, B, and C.
When a Pt100 receives a Class A rating, it means that, in the test, the error remained within the standard limit at that moment. In other words, it's a snapshot of the initial performance.
However, the standard makes no promise regarding what happens after thousands of thermal cycles, mechanical vibration, or continuous exposure to high temperatures. That's where the construction of the sensor element starts to make a difference.
Thin film and wound wire: same initial accuracy, different behaviors.
Thin film and wirewound meet the same accuracy classes as IEC 60751. What changes, therefore, is how each technology ages with use.
The thin-film element uses a very thin layer of platinum deposited on a ceramic substrate. This construction ensures good manufacturing repeatability, high mechanical strength, and small dimensions.
Platinum, however, remains adhered to the substrate throughout the sensor's lifespan. Over thousands of heating and cooling cycles, platinum and ceramic expand and contract at different rates because they have distinct coefficients of thermal expansion. This generates small internal mechanical stresses that, over time, can alter the element's electrical characteristics—the physical origin of most of the drift in thin-film Pt100. Environment and vibration weigh as heavily in this equation as the element's technology, as shown. This guide explains how to choose the right temperature sensor for high-pressure environments..
In most industrial processes, this effect is small and acceptable. Still, it exists.
In the wound wire, the manufacturer winds a very thin platinum wire around a ceramic support designed to allow it to expand with less mechanical restriction. This freedom reduces the internal stresses accumulated during thermal cycles. As a result, the sensor tends to exhibit less drift and greater stability during long-term operation.
What is drift in Pt100?
The most common mistake is thinking that a sensor "breaks" the moment it stops measuring correctly. In practice, this rarely happens.
The typical behavior is drift: the slow and gradual alteration of the sensor reading over time, even with a constant actual temperature. This results from factors such as natural aging of the platinum, internal mechanical stresses, repeated heating and cooling cycles, element contamination, continuous vibration, and prolonged exposure to high temperatures.
Therefore, it is particularly important in processes where a small temperature difference changes the quality of the final product.
The same Class A, two different performances.
Imagine two Class A Pt100 burners installed in an oven at 420°C, running continuously.
During installation, both indicate virtually the same thing, and after thousands of hours, both continue to function.
However, one of them may have drifted by a few tenths of a degree, while the other retains its almost original calibration. Thus, they continue to measure, but no longer deliver the same metrological performance.
In typical industrial processes, this difference is rarely apparent. However, in laboratories, the pharmaceutical industry, high-precision chemical processes, semiconductor manufacturing, or instrument calibration, it can determine whether a batch is approved or rejected. Many of these processes already monitor temperature using [various methods/methods]. transmitters that compensate for part of the reading errorHowever, no transmitter corrects for the drift in Pt100 caused by the technology of the element itself.
Hysteresis, the factor that's left out of the technical specifications.
Thermal hysteresis is the difference in readings when someone heats the sensor and then cools it to the same temperature.
The lower the hysteresis, the greater the repeatability of the measurement. In this respect, wound wire elements usually perform better because they experience less mechanical stress during thermal cycles.
Is thin film inferior?
No way.
For most industrial applications, thin film is the best choice: high mechanical strength, good manufacturing repeatability, low cost, small dimensions, and fast thermal response. Therefore, these characteristics have made this technology the predominant standard in modern industrial automation.
In applications involving vibration, mechanical shock, and high-volume production, thin film typically delivers the best performance-to-cost ratio, while wound yarn remains the preferred technology when maximum metrological stability over many years is the priority.
In other words, the choice depends on the application, not on the tolerance class in isolation.
How to avoid drift in Pt100 in the specification
The decision should not stop at the initial accuracy of the data sheet.
It is also worth evaluating: continuous operating temperature, expected number of thermal cycles, presence of mechanical vibration, need for periodic recalibration, required stability over the years, process criticality, and the real cost of an undetected drift.
In many cases, paying more for a more stable element costs less than dealing with years of undetected drift. See also how an RTD works and when to use each type..
How does Alutal specify the correct Pt100 for each process?
Two Pt100 sensors can leave the factory with exactly the same IEC 60751 Class A rating and, years later, exhibit very different behaviors. This does not mean that one sensor is wrong or that the standard has failed. The accuracy class certifies the initial performance; what determines the size of the Pt100 drift over the years is the construction technology, the process conditions, and the quality of the sensor element design.
In a furnace operating continuously at 420°C, for example, a difference of a few tenths of a degree between two Class A sensors installed at the same point is enough to reject a batch in a pharmaceutical process or cause a ceramic kiln to burn out of alignment. It is this type of silent error, which does not appear in any alarm, that the correct choice of sensor element prevents.
As a point of reference, a well-designed thin-film Pt100 typically derives a few hundred mΩ per year in continuous use above 300 °C, equivalent to tenths of a degree, while a wound wire of equivalent quality, under the same conditions, tends to be a fraction of that. The difference seems small in isolation, but accumulates over years of operation and in processes with many measurement points, where each sensor outside the original curve distorts the entire thermal profile.
How Alutal specifies and manufactures Pt100
Therefore, at Alutal, the selection between thin film and wire-wound goes beyond the tolerance class in the datasheet. Each application undergoes an analysis that considers continuous operating temperature, expected number of thermal cycles, vibration, installation environment, and required metrological stability over the years. Furthermore, manufacturing follows IEC 60751 from start to finish, with a traceable calibration curve and documented accuracy class for each sensor.
According to this analysis, Alutal manufactures both technologies within the Complete line of Pt100 resistance thermometers., including Series with protective well for critical processes and harsh environments.which allows you to recalibrate or replace the sensor element without stopping the line.
If your process already involves unplanned recalibrations or inconsistent results between sensors of the same class, it's worth reviewing the specification before the next batch. Speak to our technical team. to define the right sensor element for the stability your process truly needs.
