Modeling the Effects of Refrigerant Charging on Air Conditioner Performance Characteristics For Three Expansion Devices

Abstract

An experimental and analytical study concerned with the off-design refrigerant charging of air conditioners is presented. A series of experiments were conducted to characterize the effects of refrigerant charge and type of expansion device on the system performance (capacity, EER, SEER, etc.) of an air conditioner. All experiments were performed according to the ASHRAE Standard [1983]. The effects of off-design refrigerant charge in the system, type of expansion device (capillary tube, TXV, and short-tube orifice), and outdoor dry-bulb temperature (82° to l00°F) on the wet and dry steady state and cyclic tests are addressed in this study. The fully charged condition was established as a base case for all the expansion device systems. A full charge was obtained by charging the unit to the superheat or subcooled condition specified by the manufacturer charging chan. Once the full charge was determined, refrigerant was then added in 5% increments from -20% of full charge to +20% of full charge to cover the full range of charging conditions for a particular expansion device being tested. The investigation of off-design charging indicated that the system performance variables (total capacity, EER, and SEER) of the unit with capillary tube were more sensitive to off-design charging than the systems with TXV and short-tube orifice. From -20% to +20% charging, the capacity and EER showed a strong dependence on the outdoor temperature, but varied little with charge for the TXV and short-tube orifice expansion systems. A -20% charging resulted in a 21% reduction in SEER while a +20% charging produced an 11% reduction in SEER for the unit with capillary tube. For the TXV, SEER dropped 2% and 8% for -20% and +20% refrigerant charging, respectively. The SEER was constant at 9.4 for all the charging condition except for -5% charging where it peaked to 9.9 for the short-tube orifice expansion system. This trend would suggest that the range of sensitivity of a short-tube orifice system is confined within a small refrigerant charge. A new heat exchanger model based on tube-by-tube simulation was developed and integrated into the ORNL heat pump model. The model was capable of simulating the steady state response of a vapor compression air-to-air heat pump and air conditioner commonly used in residential applications. The simulated results were compared with laboratory tests at two outdoor temperatures. It was found that the ORNL model estimates were within an average of 3% of the experimental results from - 10% to +10% charging conditions. The model predicted the system performance up to 8% higher than measured results at +/-20% and +/-15% charging conditions. Among the eight void fraction models studied, the Hughmark model showed the best agreement between superheat and subcooled temperatures, refrigerant flow rate and capacity and the measured results.