Multi-Kilowatt Heat Pipe Heat Sink Performance
With the trend in electronics packaging towards smaller and lighter assemblies, rejecting multiple kilowatts of heat from power semiconductors to ambient air can be a challenge. This paper compares cooling a grouping of six IGBTs that generate 6 kW of heat using a state-of-the-art aluminum extruded heat sink and a heat pipe heat sink assembly under the same performance conditions. Both heat sinks were modeled using FlowTHERM CFD software. The large extruded heat sink was 42 inches long by 24 inches wide by 3 inches high. Correspondingly, the heat pipe assembly was 42 inches long, 11 inches wide and 9 inches high. The inlet cooling air flow-rate available was 600 cfm at 40°C. An air-cooled heat sink solution that maintains the IGBTs within their operating temperature limits in a small, lightweight heat sink package was the target.
Heat Pipe Operation
Power semiconductors are encountering the same thermal control situations experienced by the microprocessor industry in the early 1990s. For example, as microprocessor speed increased, so did heat. Under constrained conditions, conventional cooling approaches, such as extruded heat sinks, were insufficient to meet the ever-increasing cooling needs for the microprocessor. The computer industry found a solution using heat pipes. Today, heat pipes are used extensively for cooling microprocessors in laptops, desktop computers, high performance servers and workstations.
Depending on the particular application, the heat released from microprocessors ranges from 10W to 150W. For cost savings reasons, this heat is typically rejected to ambient air. In comparison, the heat released from large power semiconductors, such as IGBTs, diodes or thyristors, can be multiple kilowatts, a factor of 10 to 30 times more heat. Is the transition from the conventional extruded heat sink to a heat pipe cooling solution the next step for the power semiconductor industry? This paper examines that question. The alternative solution is to use a liquid pumped loop system that ultimately rejects the heat to air. Most companies would like to avoid this solution due to the reliability, maintenance and cost issues.
A heat pipe, in the simplest sense, is a heat mover or spreader; it acquires heat from a source, such as power semiconductors, and moves or spreads it to a region where it can be more readily dissipated. The heat pipe moves this heat with a minimal drop in temperature. A typical heat pipe is a sealed and evacuated tube with a porous wick structure and a very small amount of working fluid on the inside. See Figure I for a diagram of a heat pipe's operation. The porous wick structure, such as sintered powder metal, lines the internal diameter of the tube. The center core of the tube is left open to permit vapor flow. The heat pipe has three sections: evaporator, adiabatic and condenser. As heat enters the evaporator section, it is absorbed by the vaporization of the working fluid. The generated vapor travels down the center of the tube through the adiabatic section to the condenser section where the vapor condenses, giving up its latent heat of vaporization. The condensed fluid is returned to the evaporator section by gravity or by capillary pumping in the porous wick structure. Heat pipe operation is completely passive and continuous. Since there are no moving parts to fail, a heat pipe is very reliable.
Figure 1. Heat Pipe Operation
Heat Sink Descriptions
The aluminum heat sink profile that was evaluated represents a state-of-the-art large profile extruded heat sink with the highest aspect ratio available in the business. Consequently, the implied advantage is improved thermal control. The heat sink profile measured 24 inches wide and 42 inches long, with a height of 3 inches. The fin pitch was 2.5 fins per inch. Each fin was 0.08 inches thick and the base thickness was 0.67 inches. The weight and volume of this heat sink is 151 pounds and 3,024 cubic inches, respectively.
The heat pipe heat sink is shown in Figure 2. Standard 0.75" diameter heat pipes are embedded in an aluminum plate under the power semiconductors and extend from the plate to a remote fin stack. Heat from the electronics is absorbed by eleven heat pipes and transported to the plate fins, which are cooled by forced convection. The aluminum mounting plate is 17 inches long, 12 inches wide and 0.98 inch thick. The fin stack is 19.3 inches long, 10.8 inches wide and 9 inches deep. The plate fins are 0.02 inches thick with a fin pitch of 10 fins/inch. This heat sink weighs 70lbs and occupies an overall volume of 2,200 cubic inches.
Figure 2. IGBT Mounting Plate
Heat Sink CFD Analysis
Six 5 inch by 5 inch IGBTs, rejecting 1 kW each, were applied to each heat sink abutting each other in a two by three array. A 0.003 inch thick layer of interface material (k=1W/m-K) was assumed between the IGBT array and the heat sink. In each case, a 40ºC ambient temperature and a volumetric flow of 600 cfm was used. The air was fully ducted through each heat sink.
In this application, the IGBTs have a maximum junction temperature of 125 ºC and a package thermal resistance of 0.04 ºC/W (4). Using this information, the case temperature, Tcase, for the IGBTs is 125ºC – 40ºC = 85ºC. If the incoming ambient air is 40ºC, the remaining temperature drop to dissipate the heat is Tcase - Tambient air = 85ºC – 40ºC = 45ºC. This information was used as input for the following CFD analysis. In all situations, the model options selected included steady turbulent flow and conduction, negligible radiation heat transfer and negligible buoyancy effects. The air properties varied with temperature.
Extrusion CFD Analysis
Domain/Boundary Conditions: An idealized "wind tunnel" was constructed around the extrusion. The upstream and downstream surfaces were open for passage of air through the domain.
A uniform 600 cfm flow of air at 1atm pressure and 40ºC was specified on the upstream surface. The remaining sides were made to coincide with the corresponding surfaces of the extrusion. These faces were given a symmetry boundary condition.
Gridding: The domain was discretized into 478 x 79 x 11 cells. Five cells were used to resolve the flow profile between each pair of fins. The extrusion base thickness was discretized into 5 cells.
Using these parameters, the maximum temperature attained under the IGBTs was 130ºC, well beyond the temperature limitations of the IGBTs. Therefore, the extruded heat sink cannot meet the required thermal performance.
Heat Pipe Assembly CFD Analysis
The heat pipe assembly was comprised of two sub-models: the fin pack and the heat input IGBT mounting plate. In addition, each heat pipe was constructed from two components: 1) a high conductivity (k = 50,000W/m-K) cuboid represented the high effective pipe conductance along the pipe's length. 2) In addition, a "thin" enclosure was specified to coincide with the boundaries of the high-k cuboid. The effective thickness and thermal conductivity of this enclosure were chosen to represent the thermal impedance associated with radial transfer of heat into or out of the pipe across the pipe/fin or plate interface, pipe wall, and liquid-saturated wick structure.
The fin sub-model was solved first; each heat pipe was assumed to be carrying an equal share of the total heat load. This sub-model yielded the operating temperature of each heat pipe. These temperatures then served as boundary conditions to the mounting plate sub-model. Below are the particular parameters.
Domain/Boundary Conditions: Again, an idealized "wind tunnel" was constructed around the fin pack. The upstream and downstream surfaces were open for the passage of air through the domain. A uniform 600CFM flow of air at 1atm pressure and 40ºC was specified on the upstream surface. The remaining sides were made to coincide with the corresponding surfaces of the fin pack. These faces were given a symmetry boundary condition.
Gridding: The fin pack domain was discretized into 46 x 43 cells in the plane of each fin. Five cells were included between each pair of fins.
IGBT Mounting Plate Sub-Model
Domain/Boundary Conditions: The domain boundaries coincided with the boundaries of the chill plate. A symmetry condition was applied to all boundary surfaces, save the one through which the heat pipes would pass. On this surface, temperature boundary conditions were applied to each heat pipe based on the results of the fin sub-model.
Gridding: The domain was discretized into 46 x 43 x 9 cells.
Overall, the results of the CFD analysis of the heat pipe assembly (representing the IGBT mounting plate surface temperatures) show very uniform temperatures over the entire mounting plate. This is expected because of the nature and performance of heat pipes. Any temperature variation is associated with the location of the heat pipe within the fin stack i.e. the air is being heated as it passes through the fin stack.
The maximum temperature under the IGBTs is 80ºC, well under the temperature limits of the IGBTs. The CFD analysis indicates that the heat pipe assembly achieves the desired cooling results.
Figure 3. Therma-Charge
The heat pipe assembly was tested and the results were compared to the CFD predicted results. It was not necessary to test the extruded heat sink because Thermacore, under numerous occasions, showed that their modeling approach for extrusions yields results within 5% accuracy.
In the heat pipe assembly test, six heater blocks simulated the high power IGBTs. Thermocouples were mounted under each heater block for measuring the temperature of the plate. Other Thermocouples were mounted in the airflow stream for calorimetric purposes. A steady state heat load of 6kW and 40ºC ambient conditions were maintained. Airflow through the fin stack was achieved with two 300 CFM, 7.87 inch diameter by 2.76 inch brushless, 48VDC fans. The acoustical noise rating associated to these fans was 57 dBA; this is under the Belcore specification for acoustical noise suppression rating of 65 dBA (5).
At these test conditions, the average measured block temperature was 76ºC. This yields a thermal resistance value for the entire heat sink of 0.006 ºC/W. In comparison, the CFD results for the heat pipe assembly agreed to within 5Vo with the measured results. Consequently, the test results verified the validity of the modeling approach.
1. The heat pipe heat sink assembly was able to effectively dissipate the 6kW of heat while maintaining the IGBTs under their rated temperature limits. In comparison, the extrusion was not able to meet the requirements.
2. The entire IGBT mounting plate for the heat pipe assembly is very uniform in temperature (approximately 10ºC variation). Any temperature variation is associated with the location of the heat pipe within the fin stack i.e. the air is being heated as it passes through the fin stack. Uniform IGBT cooling is an important parameter when considering the effect of temperature on the quality of the output waveform.
3. The heat pipe unit weighed a total of 70 lbs. It is 81 lbs. lighter than the extrusion.
4. The heat pipe assembly has an approximate volume of 2,200 cubic inches. The extrusion has a volume of 3,024 cubic inches. Overall, the heat pipe unit occupies 27% less volume.
5. Heat pipe technology allows designers to reject multiple kilowatts of heat from power semiconductors directly to ambient air. This is an important conclusion considering the potential alternative is a liquid pumped loop system that has inherent long-tem reliability (leaks), maintenance (pump failure, fluid cleanliness, filtering) and corresponding cost issues.
6. Fan or blower noise is becoming an issue with cooling of power semiconductors. The heat pipe assembly used a dual fan pack solution that achieved an acoustic level of 57 dBA. This is under the 65 dBA ratings listed in the Belcore specification for acoustical noise suppression.
1. Dunn, P.D., Reay, D.4., Heat Pipes, Fourth Edition, Pergamon Press, Copyright 1994, Elsevier Science Ltd.
2. Silverstein, Calvin, C., Design and Technology of Heat Pipes for Cooling and Heat Exchange, Hemisphere Publishing Corporation, Copyright 1992.
3. Faghri, Amir, Heat Pipe Science and Technology, Copyright 1995 Taylor & Francis.
4. Powerex Product Selector Guide, Publication No. 12A-200, HP-10K-12/99.
5. Belcore Specification, Generic Requirements for Electronic Equipment Cabinets, GR-487- CORE, Issue 1, June 1996, Section 3.29.
< Back to Articles