Category Archives: heat sink

In Part 1 and Part 2 of our 3 part series, “Heat Sink Testing Methods and Common Oversights” we wrote about the heat transfer path and summarized different types of heat sink experimental set ups. Here in our final installment, part 3, we’ll show the differences in thermal resistance as determined by each testing method.

Example of Heat Sink Test Results

To show the differences in thermal resistances, as determined by each testing method, we can compare the resistances of a maxiFLOW^TM and a straight fin heat sink. The maxiFLOW heat sink has a base size of 42.5 x 42.5 mm, and is 17.5 mm high. However, the straight fin heat sink has a base size of 80 x 76 mm and is 20 mm high. Due to these size differences, the thermal resistance of the smaller maxiFLOW sink will be higher than that of the straight fin heat sink. The straight fin sink is 284% bigger in volume than the maxiFLOW heat sink.

As shown in Table 1, the maxiFLOW heat sink has a thermal resistance of 1.5 K/W when ducted, and 1.9 K/W when unducted at 1 m/s,. This results in a 21% difference between the testing methods. The dense straight fin heat sink was simulated using CFD. It has a ducted thermal resistance of 0.38 K/W, and resistance of 1.31 K/W when unducted as shown in Table 2.

Table 1: Ducted and Unducted Thermal Resistance of an ATS-52425P-C2-R0 at ~1 m/s (200 LFM) (42.5 x 42.5 x 17 mm) [2]

Table 2: Ducted, Unducted, and Dual Thermal Resistance of a Straight Fin Heat Sink at 1 m/s (~200 LFM)  (80 x 76 x 20mm)

This results in a difference of 71%. The large difference in thermal resistance between testing methods is due to the dense fins. Airflow goes around the heat sink in an unducted test, as shown by the particle tracks in Figure 5. All of the particle tracks go through the heat sink in the ducted simulation, as shown in Figure 4.

Figure 4: CFD Image of a straight fin heat sink in a ducted environment, dissipating 10 W at an inlet Air Flow Velocity of 1 m/s, (~200 LFM)

CFD Image of a straight fin heat sink in an Unducted Environment, Dissipating 10 W at an Inlet Air Flow Velocity of 1 m/s (~200 LFM)

Table 3: Summary of Heat Sink Testing Methods

References
1. Sergent, J. and Krum, A., Thermal Management Handbook for Electronic Assemblies, First Edition, McGraw-Hill, 1998.

2. Advanced Thermal Solutions, Inc., Data Sheet for maxiFLOW ATS-52425P-C2-R0 Heat Sink. http://media.digikey.com/pdf/Data%20Sheets/Advanced%20Thermal%20Solutions%20PDFs/ATS-52425P-C2-R0.pdf

3. Advanced Thermal Solutions, Inc., Heat Sink Design and Characterization, Tutorial.

In Part 1 of our 3 part series, “Heat Sink Testing Methods and Common Oversights (Part 1 of 3)” we wrote about the heat transfer path, here in part 2 we’ll cover different types of heat sink experimental set ups.

Heat Sink Set up for Testing

A wind tunnel, air temperature and velocity sensor, thermocouples and a power supply are needed to test heat sinks. A resistor is used to dissipate electrical power in the form of heat energy. The resistor is typically attached to the heat sink using thermally conductive double-sided tape. Such tape also attaches the resistor to a board, which is typically a low thermal conductivity printed circuit board or FR4.

As with a heat sink mounted on a component, the heat dissipated in the resistor is transferred to the environment by two paths. However, the thermal resistance from the resistor to the heat sink and to the board is only the interface resistance, Heat Transfer to the air via the heat sink will not equal the energy dissipated in the resistor. If this value is used, there will be an error in determining the thermal resistance of the heat sink because the energy dissipated, is not the heat transfer to the environment via the heat sink.

Thermal resistance from the board to the air increases with decreasing board thickness. Resistance decreases with increasing board thermal conductivity and increased air flow velocity across the board. Though, as previously stated, a low thermal conductivity printed circuit board is used to minimize the loss through the board.

Figure 2. Heat Sink Experimental Set Up

Heat Sink Experimental Set Ups

The first method of heat sink testing is set up in an unducted environment. This is similar to the flow experienced in typical applications. The airflow through the heat sink is affected by its fin density. The higher the fin density, the more airflow bypasses the heat sink. This provides realistic data for the thermal performance of the heat sink.

The test is easy to set up, but requires a higher quality testing facility. One prominent issue is that the heat loss to the board must be taken into account.

Figure 3  Heat Sink Experimental Set Ups: Unducted (a), Ducted (b)
and Dual Heat Sink Testing Methods [3].

The second method of heat sink testing is set up in a duct. This forces all of the airflow to go through the heat sink. There is no air flow around the heat sink or bypass airflow. It is moderately easy to set up. Vendor supplied thermal resistance data is commonly provided for ducted test results. However, the results are optimistic and can give misleading data when heat sinks are used in an unducted application.

The third method is dual heat sink testing, which uses two identical heat sinks with a heater sandwiched between them. The assembly is suspended on the centerline of a research quality wind tunnel. Dual heat sink testing is a good approach because there are no heat transfer losses to the air, e.g. via a board.  But this method is rarely used in industry because it is time consuming to set up and because the approach velocity is difficult to measure without using a quality testing facility.

In part 3 we’ll show the differences in thermal resistance as determined by each testing method.  CLICK HERE FOR PART 3

ATSs Thermal Characterization Lab is the perfect partner to help you put this 3 part article series into practice. And the best part is, the first half day is free!  Click here to learn more:  Thermal Characterization Lab

References
1. Sergent, J. and Krum, A., Thermal Management Handbook for Electronic Assemblies, First Edition, McGraw-Hill, 1998.

2. Advanced Thermal Solutions, Inc., Data Sheet for maxiFLOW ATS-52425P-C2-R0 Heat Sink. http://media.digikey.com/pdf/Data%20Sheets/Advanced%20Thermal%20Solutions%20PDFs/ATS-52425P-C2-R0.pdf

3. Advanced Thermal Solutions, Inc., Heat Sink Design and Characterization, Tutorial.

Today we’ll be kicking off a three part series on Heat Sink Testing Methods and Common Oversights. This important phase in developing a thermal management solution can sometimes be overlooked. While CFD is a critical first step, the actual testing of physical samples verifies your data and insures your design meets the real world application needs.

The effective use of an electrical component is limited by its maximum operational junction temperature. To achieve a desired component temperature, excess heat dissipated by the device must be transferred to the environment [1]. The most common method for transferring heat from the component to the environment is to use a heat sink.

To estimate a component’s junction temperature, a required value is the heat sinks thermal resistance. The thermal resistance of a heat sink can be determined analytically or experimentally. This article looks at three experimental methods of testing heat sinks.

First, it is necessary to understand the heat transfer path from the component to the environment, and then to understand the differences between the practical and experimental application of a heat sink.

Heat Sink Mounted On a Component: Practical Use

In a practical application, the heat transferred to the air from multiple junctions of a component follows a complex 3D heat transfer path. Simplified, the heat transferred from the junction of a component, , to the air follows two heat transfer paths, as shown in Figure 1. The first heat transfer path is from the junction to the air via the heat sink The second path is also from the junction to the air, but via the board The portion of heat transfer via the heat sink depends on the thermal resistance of the two paths.

For BGA components without a heat sink, heat transfer to a board is typically 80% of the total heat transfer rate. When a heat sink is mounted to the BGA, the thermal resistance from the case to the air is decreased. Heat transfer to the board will decrease and more heat will be transferred to the air.

Figure 1. Heat Sink Applied to a Component, Mounted on a Board.

In part 2 of our 3 part series on Heat Sink Testing, we’ll cover three examples of how to do this testing.  CLICK HERE FOR PART 2

ATS’s Thermal Characterization Lab is the perfect partner to help you put this 3 part article series into practice. And the best part is, the first half day is free!  Click here to learn more: Thermal Characterization Lab

References
1. Sergent, J. and Krum, A., Thermal Management Handbook for Electronic Assemblies, First Edition, McGraw-Hill, 1998.

2. Advanced Thermal Solutions, Inc., Data Sheet for maxiFLOW ATS-52425P-C2-R0 Heat Sink. http://media.digikey.com/pdf/Data%20Sheets/Advanced%20Thermal%20Solutions%20PDFs/ATS-52425P-C2-R0.pdf

3. Advanced Thermal Solutions, Inc., Heat Sink Design and Characterization, Tutorial.

In Part 1 of our 2 part series on Heat Sink Types, we covered the following heat sink types: Extruded, Stamped, Bonded Fin, and Folded Fin. Here in part 2, we will  look at active heat sinks, forged, swaged, single fin assembly and skived. At the end, we’ll have a concise cheat sheet for you for easy reference.

Active Heat Sinks are those that have some kind of air mover on them. This might be a fan or blower, and generally it is attached in some way to the heat sink. The air move provides local air helping to cool down a semiconductors hot spot. Many active heat sinks are often seen cooling high end graphics processors (GPUs) on graphics boards. The performance of this kind of heat sink is high. Generally speaking though it’s not a terrific solution long term. Air movers are made up of moving parts (fan, bearings, motor) and while rated as long as they can, the truth is these air movers can break, rendering the underlying heat sink less than effective. These solutions can tend to be costly as well.

Forged Heat Sinks conjure up the image of a blacksmith forming metal into a heat sink, but, in today’s modern world, forging is done by compressing a metal, in this case aluminum or copper. Forged heat sinks may be used in a wide variety of applications. They are medium performing heat sinks that are generally inexpensive to make. Their only real limitation is they are somewhat limited in design and air flow management.

Swaged Heat Sinks are made by a swaging manufacturing process, which is a cousin to the forging process. The swaging process involves forming the metal to a die. Swaged heat sinks generally are very good performers and good for high power applications, that is, power supplies, and other such equipment power equipment. Swaged heat sinks tend to be bulky with a limited ability to manage air flow.

Single Fin Assembly Heat Sinks are versatile for all applications. Their performance can scale from low to high performing. Some of the reasons why thermal engineers use them is because single fin assembly heat sinks are light weight and feature low profiles, so they be deployed in tight spaces. The main drawback is that they can be on the pricey side.

Skived Heat Sinks are those that are produced using the skiving process. The skiving process basically takes a block of metal, often copper, and then create fins by making very fine cuts from that block. Skived heat sinks can be used in many applications requiring medium to high performance. The skiving process produces heat sinks with very high fin density, creating a good deal more surface area for heat dissipation. A drawback though is the thick base often seen in skived heat sinks, higher weight and the heat sinks needing to be placed properly for air flow direction

That wraps up our 2nd part of this two part series. To get a copy of the companion cheat sheet for this information, simply click this link: “Heat Sink Types Cheat Sheet”

Need a heat sink like the ones we discussed here? Then please check out our eSHOP to purchase online, visit our heat sink section to see our complete list, or email us at [email protected] with your questions and we will respond as quick as we can!

And don’t forget our QoolPCB program, Cool an entire PCB, all heat sinks included, for $50!!