PEER REVIEWED Pin Fin Array Heat Sinks by Cold Spray Additive Manufacturing: Economics of Powder Recycling J. Perry 1 • P. Richer 1 • B. Jodoin 1 • E. Matte 2 Submitted: 3 July 2018 / in revised form: 29 August 2018 Ó ASM International 2018 Abstract As a result of the rise in processing power demands of today’s personal computers, water-cooled pin fin heat sinks are increasingly being employed for the cooling of graphical processing units. Currently, these high-performance devices are manufactured through high- cost, high-waste processes. In recent years, a new solution has emerged using the cold gas dynamic spray process, in which pin fins are manufactured onto a base plate by spraying metallic powder particles through a mask allow- ing for a high degree of adaptability to different graphics processing unit shapes and sizes. One drawback of this process is reduced deposition efficiency, resulting in a fair portion of the feedstock powder being wasted as substrate sensitivity to heat and mechanical residual stresses requires the use of reduced spray parameters. This work aims to demonstrate the feasibility of using powder recycling to mitigate this issue and compares coatings sprayed with reclaimed powder to their counterparts sprayed with as- received powder. The work demonstrates that cold gas dynamic spray is a highly flexible and economically competitive process for the production of pin fin heat sinks when using powder recycling. The heat transfer properties of the resulting fins are briefly addressed and demonstrated. Keywords cold spray Á economics Á heat sink Á pin fin Á powder recycling List of symbols DT 1 Inlet temperature difference (°C) DT 2 Outlet temperature difference (°C) DT lm Log-mean temperature difference (°C) g f Fin efficiency g o Surface efficiency h Fin base angle (°) l Dynamic viscosity (Pa s) q Density (kg/m 3 ) A* Nozzle throat area (m 2 ) A fin Fin area (m 2 ) A tot Total heat transfer area (m 2 ) A un-fin Un-finned area (m 2 ) B Transverse fin base width (m) B s Side fin base width (m) C Cu Cost of copper ($/kg) C electricity Cost of electricity ($/kW-h) C labor Cost of labor ($/h) C N2 Cost of nitrogen ($/kg) Cp N2 Specific heat of nitrogen (kJ/kg K) D Linear spray distance (m) D h Hydraulic diameter (m) DE mask Mask deposition efficiency DE substrate Substrate deposition efficiency EC Total electricity cost ($) GC Total gas cost ($) H Fin height (m) h Convection coefficient (W/m 2 K) I 1 First-order modified Bessel function I 2 Second-order modified Bessel function k Specific heat ratio k Cu Conductivity of copper (W/m K) This article is an invited paper selected from presentations at the 2018 International Thermal Spray Conference, held May 7–10, 2018, in Orlando, Florida, USA, and has been expanded from the original presentation. & J. Perry jperr045@uottawa.ca 1 University of Ottawa Cold Spray Research Laboratory, Ottawa, ON, Canada 2 Ironside Engineering Inc., Ottawa, ON, Canada 123 J Therm Spray Tech https://doi.org/10.1007/s11666-018-0758-3