Implementation of Innovative Measurement Technique for Precise Experimental Characterization of Frost Growth in a Heat Pump Evaporator

peer reviewedRefrigeration systems such as heat pumps may be subject to frost formation due to ambient air humidity. This phe- nomenon plays a major role in the HP overall performance drop due to the presence of an additional thermal resistance and an increase of the air-side pressure drop. The aim of this paper is to present a new innovative experimental technique to measure the amount of frost (or con- densates) within the evaporator and therefore the air latent load. As the refrigerant distribution may vary with the frost apparition at the evaporator level, the exchanger cannot be weighted on its own to measure the frost mass over time. One of the main interesting features of the test rig consists in monitoring the weight of the heat pump as a whole. As the relative mass of the frost compared to the test rig is rather negligible, a counterweight has been designed to de- crease the total load of the force sensor and therefore to increase the measurement accuracy dedicated to the frost mass measurement. The system is more elaborated and complex to set-up, but it brings much more confidant measurements on the mass, compared to a enthalpy balance based on humidity sensors. Mainly because it relies on only one sensor and there is no error propagation throughout the test. The first phase of the experimental campaign specifically focuses on characterizing the evaporator behavior (without coating) under frosting conditions. As superhydrophobic coatings might represent an efficient solution to delay the frost formation, a second testing phase (not presented in this paper) will take place in order to characterize such coated evaporators and compare them to the bare ones. The testing of different evaporators (coated and uncoated) will mark a fundamental contribution to the progress of understanding coated heat exchangers’ behavior under frosting conditions. Lastly, the creation of the experimental database will feed the calibration of numerical models