Key
22117
Conference Year
2016
Keywords
Airside heat transfer, high-performance surfaces, shape, CFD, Optimization
Abstract
The major limitation of air-to-refrigerant Heat eXchangers (HX) is the air side thermal resistance which can account for more than 90% of the overall thermal resistance. The current research on heat transfer augmentation extensively focuses on the secondary heat transfer surfaces (fins). The disadvantages of fins may include reduction of heat transfer potential due to temperature gradient, increased friction resistance, fouling and additional material consumption. On the other hand, fins contribute to reducing thermal resistance by adding significant secondary surface area. The heat transfer coefficient on the primary surfaces (tubes) is not high enough to minimize thermal resistance without significantly increasing the HX size. One contributing factor is the shape of the tube itself, which is generally limited to circular, oval, or flat. Another important aspect is the tube size; the reduction of the refrigerant flow channel significantly improves performance and compactness. This characteristic grants microchannel HX’s (MCHX) a top position amongst the current state-of-the-art air-to-refrigerant HX’s. Although the airside performance of MCHX is also improved, the need for fins has not yet been eliminated. In this paper we investigate three novel surface concepts, using NURBS and ellipse arcs, focusing on the airside tube shape with small flow channels aiming at the minimization or total elimination of fins. The study constitutes designing a 1.0kW air-to-water HX, using an integrated multi-scale analysis with topology and shape optimization methodology. Typically, such an optimization is unreasonably time consuming and computationally unaffordable. To overcome these limitations we leverage automated CFD simulations and Approximation Assisted Optimization (AAO), thus, significantly reducing the computational time and resources required for the overall analysis. The resulting optimum designs exhibit capacity similar to a baseline MCHX, with same flow rates and 20% reduced approach temperature, more than 20% reduction in pumping power, more than 20% reduction in size, while still reducing entropy generation. Experimental validation for a proof-of-concept design is conducted and the predicted heat capacity agrees within 5% of the measured values, whereas the air-side pressure drop agrees within 10%.