Experimental investigation of the impact of non-uniform heat flux on boiling in a horizontal circular test section

Abstract

Presented here are the results from the steady state flow boiling of R245FA in a laboratory scale horizontal stainless-steel test tube with an inner diameter of 8.5 mm and a length of 900 mm at a saturation temperature of 35 °C and 40 °C. Experiments were conducted at mass fluxes ranging between 200 and 300 kg/m²s at inlet vapour qualities from 0.2 to 0.7 under uniform, and non-uniform imposed heat flux cases that are expected to exist in horizontal parabolic trough solar collectors. Nine (9) different heat flux distributions were investigated. Local and average heat transfer coefficients (HTC’s) were determined based on wall temperature measurements taken along the length and around the circumference of the test section. Through the choice of the fluid being linked to the possible usage of DSG technology in organic Rankine cycles, the qualitative trends and observed performance variations can be used to predict the same for a working fluid such as water. It was found that the non-uniformity of the heat flux greatly alters the HTC’s of the fluid undergoing boiling but has no effect on the pressure drop characteristics of the fluid undergoing boiling. Heating only on the sides of the tube yielded HTC’s that were 46 % lower than achieved under uniform heating. Heating only from the top proved to be more effective in heat transmission to the fluid than heating only from the bottom (as is the case on PTC solar fields), by only a slight margin, and both these cases yielded HTC’s that were 30 % lower than the uniform heating case. Applying a bell curve heat flux distribution over the tube walls yielded overall HTC’s that differed from the uniform case by a maximum of 5 %, even as the peak heat flux position changes around the circumference of the tube. A further study may be done to quantify the degree to which the non-uniformity of the heat flux influences the local HTC’s, and to develop correlations that may aid in predicting these cases. An integration with flow pattern mapping may also be done to solidify the understanding of the phenomenon governing these observations.