Electronics cooling solutions, technology comparison
Several trade studies were performed to demonstrate the difference in heat transfer technologies. The first two studies compare peak temperatures for conduction cooling plates, HiK™ plates, vapor chambers, and encapsulated conduction cooling plates with identical dimensions. As shown in Figure 19, the plates are 9.0 inch long by 4.0 inch wide (23 x 10 cm), and are 0.120 inch thick (3 mm). This thickness was chosen because it is the current minimum thickness for vapor chambers.
The heat load is 50 W power over a 5 cm2 area. The two most common heat sinks for electronics cooling are simulated:
- A water-cooled cold rail that is 0.5 inch (1.27) cm wide, see Figure 19. The water cooling is simulated by setting the temperature at 25°C.
- Forced air cooling over the entire back surface of the plate. The fans and heat sink are simulated with a set heat transfer coefficient of used 1000 W/m2-K, with an air temperature of 25°C.
Five case studies were examined for each heat sink:
- Conduction through a 6063 aluminum plate (baseline)
- Encapsulated conduction plate with uniform conduction of 550 W/m-K.
- HiK™ plate with a uniform conduction of 1,000 W/m-K.
- 6063 aluminum HiK™ plate, with the optimal copper/water heat pipe locations. The heat pipes locations are different for the two heat sinks, see Figure 19. The temperature profiles for this case used ACT’s in-house model for HiK™ plates, which is more accurate that a uniform conduction.
- Copper/Water Vapor Chamber.
Comparison: Cold Rail Cooling
The cold-rail-cooling simulations are shown in Figure 20. In the cold rail cooling case the aluminum plate had a maximum temperature of 210°C (it is not shown, since it would make a comparison of the other devices more difficult. The maximum temperature drops as we progress from the 550W/m-K enhanced conduction plate to a uniform 1000 W/m-K HiK™ plate, an optimized HiK™ plate, and finally a vapor chamber. As you can see there is a significant reduction in temperature using the passive two-phase cooling approach. In terms of cost, the HiK™ plate is cheapest, followed by the vapor chamber. Encapsulated Conduction Plates are the most expensive.
Comparison: Convection Cooling on Back
For the convection cooling case, we see a similar progression in thermal performance across the different cooling technologies; see Figure 21 (Note that Figure 21 has a different temperature range than Figure 20). On the optimized HiK™ plate design, you can see the thermal patterns from the heat pipes spreading the thermal load across the plate surface, improving the convection side heat transfer. The vapor chamber shows further improvements over a HiK™ plate as a result of 2D spreading. The peak temperature of the aluminum plate was 55°C peak.
Isothermalization – Convection Cooling
An additional case study was conducted to look at cooling discrete heat sources over a common plate, evaluating HiK™ plates and vapor chambers; see Figure 22. In this case the heat sink was forced air convection cooling off the backside of the plate. Power is supplied to multiple 1cm2 locations with heat loads ranging from 10 to 200 W. Note that the low heat flux components are almost isothermal. If you want to keep something as isothermal as possible, use a vapor chamber.
More information on When to Use Heat Pipes, HiK™ Plates, Vapor Chambers, and Conduction Cooling: