Prediction and control of frost formation in an air to air heat exchanger
Bantle, Myles Richard Leo
Air-to-air heat exchangers can be used to preheat ventilating air and hence increase the winter ventilation rate in livestock barns; however, frost accumulation is a major problem in this application. Currently available frost control systems operate based on some combination of time, core pressure drop, or exhaust air temperature. These systems do not result in an optimal rate of heat transfer, independent of barn temperature and relative humidity. In this project, a frost control strategy based on the measured instantaneous rate of heat transfer was studied. The control strategy involved measuring the temperature rise of the cold air stream and controlling the rate of heat transfer by positioning a damper to regulate the mass flow rate of the cold air stream. As an aid to the design of the controller, a simulation model was developed. The model was based on an existing steady-state model of a condensing heat exchanger. The model was enhanced and changed in order that it could predict the thermal performance of a heat exchanger over time as frost formed in the heat exchanger. Experiments were conducted with a 472 L/s plate-type commercial heat exchanger. The experiments were used to calibrate the heat exchanger simulation, to validate the simulation model, and to test the proposed frost control strategy. The simulation model was useful in developing the control strategy and in establishing the control parameters for the prototype controller. Also, the simulation showed that it was not possible to continuously maintain a constant rate of heat transfer which approached the maximum possible heat transfer rate available from the heat exchanger. The simulation did show that a time average rate of heat transfer approaching the maximum passible heat transfer rate was possible. The calibrated heat transfer model did satisfactorily predict the general trends of the controlled heat exchanger operation. However, there were enough differences between the experimental results and simulation results that significant redevelopments to the simulation heat and mass transfer model will be necessary to obtain good agreement. In the prototype tests, the prototype controller was confirmed to operate satisfactorily under four widely differing input conditions. Three control parameters were identified as being critical to the design of a heat transfer optimizing controller; the amount of heat transfer degradation permitted before a defrost is initiated, the maximum cold air stream mass flow rate through the heat exchanger permitted just following a defrost; and the rate at which the supply air flow rate is changed. The proposed control strategy directly measures the instantaneous rate of heat transfer. This enables the optimal average heat recovery to be obtained over a wide range of input conditions. Further development is necessary to establish the optimal control parameters and to complete development of a marketable heat exchanger frost controller.