Spray Evaporation on Enhanced-Surface Tube Bundles with Low-GWP Refrigerants
Type of DegreePhD Dissertation
MetadataShow full item record
With the development of new low-GWP HFO refrigerants and increasingly stringent regulation of existing HVAC&R systems, the heat transfer performance of these refrigerants and the effects of various enhancement techniques are of particular interest. Refrigerant distribution via pressurized liquid spray in a shell-and-tube heat exchanger is one such recently developed enhancement technique that has the potential to decrease refrigerant inventory in industrial systems while matching or exceeding the heat transfer performance of the much more common flooded evaporator. A limited number of studies have been performed on HFO refrigerants and spray evaporation separately, but no published studies have combined these elements with other enhancement factors such as low-finned tube outer surfaces or different tube bundle pitch configurations. A specialized test apparatus was constructed in order to perform such a study, and shell-side heat transfer coefficients of R134a, R1234ze(E), and R1233zd(E) were investigated on bundles of tubes with two different enhanced-surface types and in two different bundle geometries at various refrigerant saturation temperatures. The results showed a weak dependence of bundle heat transfer coefficients on refrigerant saturation temperature and spray nozzle type, provided that sufficient refrigerant distribution was realized. However, heat transfer performance depended strongly on refrigerant properties, tube heat flux, enhanced-surface type, bundle geometry, and refrigerant inlet subcooling. With a low-finned tube surface designed to enhance condensation heat transfer, performance was comparable to that of nucleate pool boiling; with the surface designed to enhance evaporation heat transfer, an average enhancement of 200% was realized compared to the former. Enhancement due to the triangular-pitch bundle geometry was approximately 26% compared to the square-pitch bundle due to differences in the bundle effect, and variations in inlet subcooling between 0.5 °C and 2.0 °C resulted in an enhancement of up to 74% at the lower limit. Performance at high heat fluxes was significantly affected by refrigerant dryout at the bottom of the bundle which caused significant decreases in lower row heat transfer coefficients and represents a significant challenge in the design and implementation of a spray evaporator. The performance of R1234ze(E) was comparable to that of R134a, the refrigerant it is meant to replace. The empirical results reported in this work are meant to provide a point of comparison for engineers tasked with designing similar heat transfer equipment. A semi-empirical model was also developed to correlate the data collected in the experiments. Found in a survey of the literature, the model on which it was based was used to correlate average bundle Nusselt numbers for similar spray evaporation experiments using ammonia. The model was modified in the current work to correlate individual row Nusselt numbers, and a method for calculating local Reynolds numbers throughout the bundle was developed. A factor was added to account for the enhancement effects of different surface types, bundle geometries, and refrigerant inlet subcooling. The model was found to be an adequate fit for the data that most closely represented real operating conditions and is intended to be applicable to the design of similar multi-row shell-and-tube heat exchangers.