Increased Accuracy and Response Time Temperature Measurement

The TSP341-N’s ability to measure the outside wall temperature of piping for flowing media was the basis for considering suited measuring media in the previous chapter. The measurement results below should confirm this precision.

Additionally, it shows that the response time during measurement with the TSP341-N is comparable, or in many instances shorter, than during a classical measurement with a thermowell.

Precision and Response Time Measurements Using TSP341-N Temperature Sensor

An experimental measuring setup that guarantees top accuracy of the reference measurement of actual outer wall temperature as well as very quick response behavior was used to perform precision and response time measurements [1].

Experimental measurement setup in a climate chamber to measure the precision and response time of the TSP341-N.

The ambient temperature was varied over a period of several hours whilst the outside wall temperature was maintained at a constant, in order to determine the precision of the TSP341-N.

In the example shown below, at a temperature of the outer wall of T_surf = 125 ^oC the ambient temperature T_amb is varied over a range of T_amb = - 40 ^oC to 80 ^oC in increments of 20 ^oC and 60 ^oC.

Surface Temperature Measurement

The surface temperature measured by the TSP341-N at the measuring point (Sensor 1, on surface) and the measured ambient temperature near the measuring point (Sensor 2, shifted back) is shown in the diagram.

The temperatures measured at the measuring point and, especially near the measuring point, are influenced strongly by the ambient temperature. However the outside wall temperature, calculated using the non-invasive temperature sensor algorithms (Non-inv model-based), does not deviate by more than 1 ^oC from the exact temperature (Ideal surface reference). This accuracy level is also present when the ambient temperature jumps by 20 ^oC and especially when it jumps by 60 ^oC:

• No oscillation behaviour sets in and the determined temperature remains stable.

The good response behaviours of the TSP341-N at rapid temperature changes of the outer wall is shown in the two diagrams below. High accuracy and short response time of this surface sensor are achieved using thermo-mechanical optimization and special algorithms.

Compared to classic measurement with thermowell, response times are comparable and often even superior.

The outer wall temperature increases in an almost linear fashion in 8 minutes from T_surf ~20 ^oC to T_surf ~190 ^oC in the following example. Therefore, the temperature rise is ~0.3 ^oC/s.

The temperatures detected by sensor 1 and sensor 2 are only able to follow the rapid increase in temperature with a delay of several minutes. However, the outside wall temperature calculated using non-invasive temperature sensor algorithms (Non-inv-model-based) reaches the exact temperature (Ideal surface reference) after only ~17 seconds. In this example, the outer wall temperature calculated in this way only deviates from the exact temperature by ~ 6 ^oC at most, even in this very dynamic phase.

As shown in the previous diagram, at a constant surface temperature before and after it is increased the accuracy of ~ 1^oC is reached immediately.

Precise and stable measurement of a constant surface temperature at ambient temperature jumps.

Reaction of the TSP341-N to rapid changes in the surface temperature.

Reaction of the TSP341-N to a temperature jump from 25 °C to 220 °C.

The response to a temperature jump (unit step) also demonstrates the short response time of the TSP341-N.

In the following example, the temperature jump from 25 ^oC to 220 ^oC is achieved using sudden contact of the surface sensor with a surface heated to this temperature.

Determination of Outer Wall Temperature Using non-invasive Temperature SensorAlgorithms

Using non-invasive temperature sensor algorithms, the outer wall temperature (Non-inv model-based) of > 200 ^o was determined after 29 seconds. This was 90% of the sudden jump in temperature (T₉₀ ~29 s).

Compared to the classic temperature measurement with thermowell, the TSP341-N surface sensor, with a response time of T₉₀ < 30 s, fared well.

Summary

The measurements discussed demonstrate that the TSP341-N non-invasive surface-mount temperature sensor makes it possible to achieve very accurate temperature measurements. Even with rapid and large changes in ambient temperatures, high measuring accuracy is maintained.

Due to its properties, the sensor can follow even faster temperature changes of the measuring medium, quickly and with good accuracy. The response time of the TSP341-N is easily within the range of classical temperature measurement with thermowell. As a result of the required robustness of thermowells in some applications, response times there can be significantly higher as well.

References

1. J. Gebhardt et al.: “Reliable measurement of surface temperature: a step to widespread non-invasive T measurement in industry”, Proceedings of: 19. ITG/GMAFachtagung “Sensoren und Messsysteme”, Nürnberg, June 2018

This information has been sourced, reviewed and adapted from materials provided by ABB Measurement & Analytics.

For more information on this source, please visit ABB Measurement & Analytics.