Experimental investigation of refrigerant charge determination for heating performance in a CO2 heat pump system for electric vehicles

Given the scarcity of investigation of refrigerant charge for the heating performance of the CO2 automotive heat pump system, a test of charge determination with a 350-650 g range over a wide operating condition is comprehensively conducted. It is concluded that 450-550 g charge can be an appropriate level. The inefficient charge level is more sensitive for performance deterioration than the excessive level. Each refrigerant charge level can generate-30 kg/h mass flow rate lift, and more than 50 K discharge temperature increase in respect of the valve openings. Lifting the compressor speed can generally decrease system coefficient of performance (COP) by-18% under-20 degrees C ambient and by-40% under 0 degrees C ambient, respectively. As to the discharge pressure concerning the control strategies, for a fixed compressor speed, the cabin side air supply temperature in general has a linear relationship (with a fixed value of slope) in respect of the discharge pressure for low discharge pressure domains (less than 70 bar), no matter what the valve settings are or which charge amount level we choose or what the ambient condition is. In addition, with a fixed compressor speed, under the same discharge pressure, with-20% refrigerant charge reduction from the appropriate level, the COP generally can be dropped by more than 12%, and the drop is more pronounced under high ambient conditions. What's more, for a fixed optimal valve setting with different compressor speeds, under the same discharge pressure,-20% refrigerant charge loss can provide more than 20% COP drop. This may be helpful for fault diagnosis of system performance deterioration when charge leakage is involved. The conclusions in this study can be served as the guidelines for automotive heat pump systems.