Developments in C/C heat exchangers were conducted under the Air Force research laboratory thermal component development program, which began in 1996. Stevenson and Vrable studied the development of a carbon-carbon composite heat exchanger for aircraft applications. The type of heat exchanger selected was a plate-fin, cross-flow design resulting in a 40% reduction in weight compared to conventional alloys. Although metal core heat exchangers used offset strip-fin designs, manufacturing capabilities limited C/C to the plain plate-fin design.
Kearns, Anderson, and Watts examined brazing carbon-carbon plates and fins to form the heat exchanger core. The braze material consisted of 70% nickel, 18% silicon, and 12% chromium with an alloy melting point of 1473K. SEM analysis showed that the silicon and chromium reacted to form carbides, while the nickel remained in the bulk of the braze material. The study concluded that fin heights varied enough to prevent uniform contact between the corrugations and parting sheets. This reduced the amount of area available for bonding and produced structures with very little strength. It was condluded that instead of brazing fins to the plate, a preferred alternative was to manufacture an integral design that co-processes the fins and plates as one piece.
Barrett assessed closed Brayton cycle recuperators for space power applications. Barrett defined a series of CBC state point cases and conceptually designed several recuperators using temperature, pressure, and fluid flow information. Results showed that mass and pressure-loss goals were difficult to achieve using conventional HX technology. Traditional recuperator designs exceeded the mass target by 15% with a total pressure-loss of 33%. Barrett suggested options to reduce system development risk by modifying cycle state points, using enhanced heat transfer techniques, or incorporating advanced materials such as carbon-carbon composites.
Barrett conducted conceptual development of the C/C recuperator core at NASA Glenn Research Center. Studies compared the mass and volume characteristics of eight HX designs, including six metal designs with different plate-fin geometries and two C/C designs with the same plate-fin geometry, but different fiber-based materials. The recuperator designs were generated using a NASA design code HXCALC, which uses thermal requirements, state point information, and fluid properties to roughly size the heat exchanger. Calculations proceed in an iterative loop, adjusting geometries, until thermal load and pressure-drop requirements are satisfied. Then, the final geometry is used to estimate the recuperator’s mass.
Barrett predicted C/C recuperator cores to have lower masses than metallic heat exchangers in all cases, although most C/C designs showed an increase in volume compared to the metallic heat exchangers. The increase in volume is due to the choice of C/C plain-fin geometry, which is limited by manufacturing capabilities. The C/C plain-fin geometry requires more surface area than the offset strip-fin design due to a lower average convective heat transfer coefficient. At lower thermal loads, the C/C mass advantage diminishes; therefore, C/C plain-fin recuperators are only advantageous at thermal loads of 300 kWt or greater, providing a 60% mass savings and 20% volume penalty.