This article presents the impact of accuracy of sensors, or more specifically Pt100 temperature sensors, on result analysis of experimental studies. For this purpose, an experiment was carried out consisting in measuring the temperature on the surface of a partition - a concrete wall, beneath its insulation layer. The tested surface was separated from external environment and could only be influenced by the wall structure. Therefore, the expected result of the experiment, i.e. the difference in temperature sensor readings in identical locations on both sides of the partition, should reach a value close to 0. This article also presents the values of absolute error for sensors which were determined before their installation on the surface, and on which their location depended. The obtained deviations were included in the results of the experiment, which led to a decrease in temperature differences on both sides of the partition, in some cases even reaching the expected value of 0. This analysis showed how important it is to know the measurement error and then eliminate it in result interpretation.
Read full abstract