Performance Modeling of Supercritical Carbon Dioxide Zigzag-Channel Printed Circuit Heat Exchangers

Supercritical carbon dioxide (sCO2) is currently being studied as the working fluid in power producing Brayton cycles due to its excellent physical and thermodynamic properties, especially near the critical point. Printed circuit heat exchangers (PCHEs) are being considered for use as condensers and recuperators for this purpose due to their high strength and compact designs. Many experimental and numerical studies are being conducted to characterize and optimize sCO2 PCHE operation and develop correlations to describe their thermal-hydraulic performance. Additionally, a few experimental and numerical structural assessments of these PCHEs have been conducted, but all have been somewhat limited due to the difficulty measuring actual stresses in an operating PCHE and the computer resources needed to accurately conduct a fluidstructure interaction (FSI) examination using finite element analysis (FEA). The entire average data reduction method has traditionally been the means by which to evaluate the Nusselt number (����), Colburn factor (��), and Fanning (��) or Darcy friction factor (����) for these heat exchangers. Then the staged integral method of data reduction was proposed which greatly improved the accuracy of these analyses. Computational fluid dynamics (CFD) is used to examine the design and operating factors which influence the size of data sampling interval that must be used to accurately apply the staged integral method to zigzag-channel and straight-channel PCHEs. Then a non-uniform staged integral method is proposed and evaluated against several test cases. Data indicates that the interval size required for PCHEs data analysis is primarily driven by channel bend angle (��) for zigzag-channel designs and by mass flow rate (��̇ ) for straight-channel designs. The non-uniform staged integral method was of limited use for zigzag-channel heat exchangers operating near the critical point. The method was quite useful iv for reducing the number of required data sampling points for straight-channel PCHEs with low mass flow rates operating near the critical point. The sensitivity of the ����, ��, and ���� for zigzag-channel PCHEs to 20 geometric design factors and operating conditions were investigated. A two-level Plackett-Burman, non-geometric resolution III experimental design with no replications was used. This experimental design was then folded over to produce a resolution IV design to eliminate confounding of main factor effects. Main factor and two factor interaction effects were examined. CFD was used to simulate each tested case. Data indicates that zigzag-channel PCHE thermal-hydraulic performance parameters are most sensitive to changes in ��, bend angle radius of curvature (��), ��̇ , and channel width (��). Additionally, some two factor interactions of the more significant main factors were found to be of importance. Data from the zigzag-channel PCHE factor sensitivity study was used to develop the ���� and ���� correlations. The Gauss-Newton nonlinear regression method was applied to this data to create best fit correlations. The developed correlations for the hot and cold channel ���� and ���� were validated against experimental data and were found to model the data within ±30%. The new correlations predicted the numerical results from inlet parameters with an average error of approximately ±30%. These correlations appear to provide the best estimate, over current leading correlations, of PCHE thermal-hydraulic performance from inlet parameters prior to any simulation or experimentation. In order to attempt a complete investigation of zigzag-channel sCO2 PCHEs, a FSI study was conducted. A previous pseudo two-dimensional (2D) study of a sodium-sCO2 PCHE was replicated, comparing linear elastic model and bilinear isotropic hardening model results, then multilinear elastic hardening model results were included. Next, previously unperformed, three- v dimensional (3D) one-way coupled FSI studies of an experimental and two notional zigzagchannel, sCO2 PCHEs were conducted. All results were evaluated against the stress intensity limits set forth by the American Society of Mechanical Engineers (ASME) Boiler and Pressure Vessel Code (BPVC) Sections III and VIII. Most of the examined PCHEs meet the requirements for general use but exceed the maximum allowable stress intensities for application as nuclear components