Smart level sensor based on thermal resistance measurement with self calibration

A new detection method of level sensor based on the thermal resistance of gas and liquids using modeling of the current-voltage-curve is presented. The model directly examines the thermal resistance (R
_th
) of the sensor exposed to a specified medium whose value extracted simultaneously with the parameters of the sensor. In compared to the in air with 348 K/W, the thermal resistance in water decreased around 82%, and/or in silicon oil 67%, in transmission oil 68% and in petroleum 71%. From these results, the sensor status between "empty" (in air) and "full“ (in fluid) are clearly distinguishable. The change of overall thermal resistance due to the dirt was measured experimentally using a variety of fluids and the results were validated with the mathematical simulation. The changing of the thermal resistance is evaluated using the mathematical model based on heat transfer concept, enable to assess if soiling on the sensor surfaces so far increased, then the sensor must be changed or cleaned.