Proceedings of the Second International Energy 2030 Conference,
November 4-5, 2008, Abu Dhabi, UAE

Mold Filling Meta Model for Polymer Composite Heat Exchanger

Avram Bar-Cohen
University of Maryland, USA

Satyandra K. Gupta
University of Maryland, USA

Peter Rodgers
The Petroleum Institute, UAE

Juan G. Cevallos
University of Maryland, USA

Mahmoud Adi
The Petroleum Institute, UAE

Abstract
In this study, we present a manufacturability analysis of polymer heat exchangers using thermallyenhanced composites. The advent of new fiber filled resins [1], which use fibers with higher thermal conductivity, has brought along new design possibilities for heat exchangers using cost-effective molding techniques. The cost of polymer heat exchangers and the energy investment in fabrication are expected to be considerably lower than their metal counterparts. The most common molding technique for thermoplastics today is injection molding. Polymer heat exchangers will involve complex geometries and thin sections. Hence we believe that injection molding will be the most suitable process for making them. The work presented here focuses on injection molding of a finned plate. This geometry was deemed to be representative of a compact heat exchanger module.

Thermally-enhanced polymers possess certain material characteristics that could pose challenges during the molding process. Specifically, these resins have high filler loadings, which increases the viscosity, and the fibers’ high thermal conductivity causes the resin to cool faster in the mold. These two factors make filling the mold difficult to accomplish for long thin parts, so for this reason, successful filling of the mold is the subject of this study.

Optimizing the heat exchanger designs requires characterization of the moldability over the design space. However, molding actual parts to explore the design space would be time consuming and costly. Hence, a more suitable approach is to explore the design space using mold flow simulation software such as Moldflow. Simulation is conducted for discrete points in the design space and a meta model is constructed based on the simulation results. Moldflow filling predictions were validated using a spiral mold test. The meta model presented here can be used to predict filled volume percentage for single-gate injection, and to identify regions in the design space not suitable for injection molding.