Characterization and numerical simulation of liquid refrigerant flow emerging from a flooded evaporator tube bundle

Abstract

The distribution of liquid droplets emerging from an evaporator tube bundle is characterized for refrigerants R-123 and R-134a with a triangular and rotated triangular tube arrangement with a pitch of 1.167. The purpose of this research was to improve understanding of the droplet ejection process to aid in design of evaporators typically used in larger chiller systems. A laser and camera system captured images of the evaporator headspace at varying conditions. Conventional shadowgraphy techniques were applied to recognize and match droplets for velocity calculations. The evaporator conditions were varied with mass fluxes from 3.5 to 40.7 kg/s-m² (2250 to 30000 lb/hr-ft²), top-rows heat fluxes from 5.3 to 31.5 kW/m² (1700 to 10000 Btu/hr-ft²), and outlet saturation temperatures of 4.4 and 12.8 °C (40 and 55 °F). Conditions ranged from flooded to dryout on the top rows. Droplet number, size distribution, velocity, and liquid volume fraction are presented in the headspace above the bundle. A method to numerically duplicate the droplet loading in the headspace using CFD with a Lagrangian discrete-phase model is also presented and verified with R-134a and a triangular arrangement, providing a powerful design tool. Liquid distribution in the headspace is found to be a strong function of all varied properties, particularly mass flux, liquid level, and saturation temperature. Vapor shear is thought to lead to the majority of droplets generated, especially in the upper headspace. The high liquid-vapor density ratio of R-123 and corresponding high velocities make it particularly difficult to separate liquid droplets before they escape the tube bundle.

Description

Keywords

Computational Fluid Dynamics, Refrigerant, Tube Bundle, Droplet, Evaporator, Numerical Simulation

Graduation Month

May

Degree

Doctor of Philosophy

Department

Department of Mechanical and Nuclear Engineering

Major Professor

Steven J. Eckels

Date

2019

Type

Dissertation

Citation