Heatpipe Reliability

Prepared by: Nelson J. Gernert/ Sr. Engineer

. . . . . . . . . .

. . . . . .. . . . Thermacore, Inc.

Heat Pipe Reliability Documentation

A reference document for companies assessing the reliability of Thermacore’s heat pipe based thermal solutions.

November 10, 1999

Thermacore, Inc. 780 Eden Rd. Lancaster, PA 17601 Phone 717.569.6551 Fax 717.569.8428

2

. . . . . . .. . .

Heat Pipe Reliability Documentation A reference document for companies accessing the reliability of Thermacore’s heat pipe thermal solutions.

Table of Contents

Table of Contents.................................................................................................2

1.0 Heat Pipe Reliability Information Disclaimer ................................................3

2.0 Introduction ...................................................................................................3

3.0 Thermacore Company Information...............................................................3

4.0 Heat Pipe Operation.......................................................................................4

5.0 Heat Pipe History ...........................................................................................5

6.0 When to Consider a Heat Pipe Thermal Solution ..........................................5

7.0 Heat Pipe Heat Sink Handling and Safety Precautions .................................7

8.0 General Comments.........................................................................................7

9.0 Laboratory Based Heat Pipe Operating Life Tests........................................7

10.0 Heat Pipes in the Field................................................................................11

11.0 Heat Pipe Thermal Cycling Information....................................................11

12.0 Heat Pipe Shock and Vibration Information .............................................13

13.0 External Heat Pipe Assembly Protection ...................................................18

14.0 Heat Pipe Pinch Off Reliability ..................................................................19

15.0 Conclusions.................................................................................................19

16.0 References ..................................................................................................20

3

1.0 Heat Pipe Reliability Information Disclaimer

The heat pipe reliability information contained in this document covers the heat pipe internal cleaning, handling, fabrication and test procedures that have been developed by Thermacore over the process of thirty years of heat pipe development and manufacturing. The use of this reliability information by other companies to support the long-term compatibility and reliability of their heat pipes is not valid.

2.0 Introduction

The enormous growth of the electronics industry is due mainly to the continued evolution of the microprocessor in personal computers. These microprocessors are now being moved out of the PC and embedded into telecom/datacom equipment, automobiles, medical equipment, test equipment and consumer electronics. There is a big difference, however, when examining microprocessors used in desktop computers and microprocessors embedded in rugged industrial type equipment. The typical desktop computer has about a three-year product life and the environment that it is exposed to is relatively benign. Many of the embedded microprocessor applications, however, have ten or more year product lives and the environment can be quite demanding. Because of the extended product life and the more demanding environment of embedded applications, all components that are routinely used in computer products become questioned for their reliability. This includes the heat pipes and heat pipe assemblies that are used to cool the microprocessors.

In response to the heat pipe reliability concerns, this document presents heat pipe reliability information that supports their use in demanding applications. Reported are extensive materials compatibility information for those heat pipe types that are relevant to cooling in embedded applications. Thermacore has had more than 700 heat pipes on extended time tests for several years. In addition, Thermacore has conducted shock and vibration tests on many heat pipe assemblies to qualify them for use in various stringent military and industrial applications. Environmental tests such as thermal cycling, humidity and salt spray tests were also conducted to qualify heat pipe assemblies for outdoor use.

Ultimately, the information contained in this document will permit engineers and managers to make an informed decision regarding the applicability of heat pipes to their rugged yet cost sensitive applications.

3.0 Thermacore Company Information

In 1963, RCA in their Electron Tube Division in Lancaster, PA, plant had mounted an intensive program to develop Thermionic converters to generate electricity in space using a nuclear reactor. The Thermonic converter closely

4

emulates the electron tube used in a television set; consequently, it fit the RCA business model to develop the Thermionic Converter. In 1964, RCA began developing the heat pipe as a means to bring the heat from the reactor to the Thermionic Converters to generate the electricity.

RCA continued the development of Thermionic converters and heat pipes for several years, however, as the space program concentrated on getting man to the moon in the late 1960’s, the U.S. interest in space nuclear power declined. RCA consequently decided to stop all work on Thermonic Converters and heat pipes.

G. Yale Eastman and Richard Longsderff, who worked in the RCA heat pipe division, formed Thermacore in 1970 to pick up the former RCA program. Initial business was in the Aerospace and the military industry, and Thermacore supplemented their engineering efforts with government R&D contracts, which gave them a strong technology flow.

For many years, much of the Thermacore’s activity was focussed on heat pipe related research and development for various government agencies such as NASA, DOD, DOE and the National Labs. That research lead to a thorough understanding of heat pipes and the materials and processes that are needed to produce long-lived heat pipes.

As the electronics and personal computer revolution took hold in the 1980’s and 90’s, faster processor speeds and higher power densities made heat dissipation a growing problem, one that was outstripping the ability of conventional heat sinks made of simple aluminum extrusions. The heat pipe provided the much-needed “next step” in the evolution of cooling products to meet the demands of the growing electronics industry. Thermacore now has heat pipe manufacturing sites in the United States, United Kingdom, Mexico, Taiwan and Korea.

4.0 Heat Pipe Operation

A heat pipe, in its simplest sense, is a heat mover or spreader; it acquires heat from a source, such as the microprocessor, and moves or spreads it to a region where it can be more readily rejected. The heat pipe moves this heat with very little drop in temperature.

A typical heat pipe is a sealed and evacuated tube, a porous wick structure and a very small amount of working fluid on the inside. Figure 1 is a heat pipe illustration that defines its internal design and its operation. A 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

5

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. Additional information about heat pipes is found in References 1, 2, 3 and 4.

Figure 1. Heat Pipe Operation

5.0 Heat Pipe History

The rudiments of the modern day heat pipe can be traced to the Perkins Tube patented in the mid-1800s. The Perkins Tube, classified now as a thermosyphon heat pipe (no capillary wick structure), provided the initial building blocks for the heat pipe. In 1944, Gaugler, from General Motors, added a capillary wick structure on the inside of the heat pipe for liquid pumping. The heat pipe remained in fallow until 1964 when Grover from Los Alomos National laboratory wrote a report and coined the name “heat pipe.” Subsequently, heat pipes have been applied to numerous applications ranging from spacecraft thermal control to permafrost temperature regulation for the Alaskan pipeline. A heat pipe remained a relatively high cost/ low volume device until the late 1980’s, early 1990s when the thermal demands of the computer industry warranted high volume production. Presently, Thermacore manufactures several thousand heat pipes per day at a price suitable for OEM use.

6.0 When to Consider a Heat Pipe Thermal Solution

Thermal system designers that have the following constraints would consider a heat pipe thermal solution:

Heat In

HeatOut

VaporFlow

PorousWickStructure

VaporGeneration

VaporCondensation

6

• Limited Height Budget: In some applications, the height over the electronic device may not provide sufficient space to provide direct cooling at this location. A heat pipe in this situation is used to move the heat to a location where it can be effectively dissipated by natural or forced convection.

• Limited or Zero Electrical Power Consumption: Cooling with a fan requires electricity. In some portable applications, a fan reduces battery power and limits the useful operating time of the product. A heat pipe, in this situation, allows the developer to acquire additional surface area for heat rejection by natural convection, thus eliminating the need for a fan. If volume constraints limit the use of a natural convection cooling solution, a heat pipe to a miniature fan/sink might be more economical than a large system fan solution.

• Zero Noise or Noise Reduction: Cooling by natural convection eliminates fan noise. If volume constraints limit the use of a natural convection cooling solution, a heat pipe to a miniature fan/sink will result in less noise than a large system fan solution.

• Low Maintenance/ High Reliability: All electro-mechanical devices such as fans have finite life. A heat pipe thermal solution has no moving parts to fail; consequently product maintenance requirements are eliminated or reduced.

• Sealed Enclosure Cooling: In some applications, the device to be cooled will be in a sealed enclosure to protect it from the environment. An example is an Industrial PC located on the dirty shop floor. Heat in this situation, needs to be rejected from the outside of the sealed enclosure. The heat pipe provides a thermal path to the enclosure wall.

• Stagnation Regions: In some situations, an electronic device can be located in a region of poor air circulation inside the enclosure. A heat pipe in this situation is used to move the heat to region where there is adequate air circulation. For example, air from a system fan is directed or ducted over the main CPU, thereby, starving other electronics within the enclosure of cooling air. Heat pipes are then used to interface with the ducted air stream.

• Low Weight: Attaching a heat sink to a printed circuit board places strain on the board. If the heat sink is too heavy it can warp and damage the board. The ultimate goal is a heat sink that has a low thermal resistance and is lightweight. Extrusion type heat sinks that mount directly over top the electronics are getting quite large for the amount of heat that must be dissipated. Heat pipe thermal solutions are offering the best performance at the lowest weight.

• Combination of Constraints: Any combination of the above constraints would warrant the consideration of a heat pipe thermal solution.

The selection of the proper heat pipe cooling solution is dependent upon the developer’s specifications, design constraints and budget. Thermacore has applications engineers available to assist with optimizing these parameters.

7

7.0 Heat Pipe Heat Sink Handling and Safety Precautions

The fluid inside the heat pipe is typically potable water; it is non-toxic and non-flammable. The heat pipe is a sealed vessel that becomes pressurized at temperatures exceeding 100oC. To avoid the danger of bursting, the heat pipe should not come in contact with excessive heat or flame that raises its temperature above 180oC.

8.0 General Comments

Heat pipes, when properly designed and manufactured, have been shown to be highly reliable and are completely maintenance-free. One group of 468 Thermacore heat pipes has been in electronics cooling service aboard aircraft for more than fifteen years with no reported failures. The mean-time-to-failure has been estimated at more than 2.5 million years.

In its simplest form, a heat pipe is a two-material system: structure and working fluid. If these materials are compatible, long un-degraded life (years) is assured. Like other devices, heat pipes have operating limits. Operation outside these limits can be expected to cause degradation or outright failure.

When operated within its ratings, a heat pipe can be expected to provide highly reliable heat transfer performance for years. Although Thermacore’s standard warranty is for one year, our products and designs are based on an extensive database that supports our objective of providing our customers with years of un-degraded life.

With all standard heat pipe products, Thermacore publishes maximum operating conditions that must not be exceeded. At a minimum, these generally include total power carried, evaporator power density and gravitational angle. The operating environment is generally specified as well. Failure to observe these limits can be expected to result in damage to the heat pipe.

9.0 Laboratory Based Heat Pipe Operating Life Tests

As part of the Thermacore heat pipe quality assurance program, Thermacore routinely operates their heat pipes for extended periods to study the long-term effects of material interaction on both discrete heat pipes and heat pipe assemblies. At any instance in time, there are over 200 hundred heat pipes on extended life test. For example, Figures 2 and 3 are photographs of various heat pipes on test (Ref. 5 & 6). The heat pipes are mounted on racks in a designated room for monitoring these heat pipes. Testing includes continuous heating and cooling with water or air. The procedures used by Thermacore to setup a heat pipe on life test are documented in Ref. 7.

8

A typical life test heat pipe is shown in Figure 4. It includes a heat pipe that has a heat input block and a condenser block that removes the heat. Thermocouples are attached to the evaporator, adiabatic and condenser region of the heat pipe. Temperature and power are recorded approximately every 200 hours. The key parameter that is monitored is the temperature difference measured between the evaporator and the condenser thermocouples (Thermocouple#1 – Thermocouple #3).

Figure 4. Typical Heat Pipe Life Test Assembly. A heat pipe material system that is considered to be compatible will exhibit a similar evaporator-to-condenser temperature difference (Delta T = Thermocouple #1 – Thermocouple #3) at the beginning of life and throughout the duration of the test, provided the heat input remains constant. A heat pipe material system that is considered incompatible will exhibit an increasing evaporator–to-condenser temperature difference as the test proceeds. This increasing temperature difference is typically due to non-condensible gas (NCG) collecting in the condenser region of the heat pipe. The NCG is swept there by the flowing vapor. The region that contains the NCG is not an active part of the heat transfer mechanism; consequently, it shows up as a cold region. As time moves on and more gas collects in this region, the heat pipe, as determined by Thermocouple #3, becomes colder in this region. Table 1 is a data summary of some of the longest running heat pipes in the Thermacore test program. This data is for copper heat pipes with three working fluids: Acetone, Methanol and Water. To date, there are over 129,332 hours (14.8 years) of operation recorded on these 35 heat pipes without any signs of degradation. The methods used to fabricate these heat pipes are the same methods used by our manufacturing operation.

9

Figure 2. Heat Pipe Life Test Rack (Air-Cooled)

Figure 3. Heat Pipe Life Test Rack (Water-Cooled)

10

Table 1. Heat Pipe Life Test Summary.

Heat Pipe #

Envelope Material

Wick Material

Working Fluid

Initial Delta-T, oC

Present Delta-T, oC

Operating Temp. Range, oC

Hours in Operation

Heat Pipe Status

1 Copper None Meth. 2 2 49 to 109 133,324 Running 2 Copper None Meth. 4 5 46 to 125 133,324 Running 3 Copper None Acetone 6 4 54 to 131 113,064 Term. 4 Copper None Acetone 4 7 45 to 126 133,324 Running 5 Copper None Meth. 6 6 38 to 114 133,324 Running 6 Copper None Meth. 6 9 48 to 69 24,558 Term. 7 Copper None Meth. 8 4 29 to 89 129,334 Running 8 Copper None Acetone 10 8 31 to 87 129,332 Running 9 Copper None Acetone 6 6 35 to 100 129,332 Running 12 Copper None Water 6 6 24 to 94 129,332 Running 14 Copper None Meth. 1 3 35 to 100 129,332 Running 15 Copper None Meth. 1 3 35 to 100 129,332 Running 19 Copper None Meth. 1 3 35 to 100 129,332 Running 21 Copper None Meth. 1 3 35 to 100 129,332 Running 22 Copper None Meth. 1 3 35 to 100 129,332 Running 27 Copper None Meth. 1 3 35 to 100 129,332 Running 29 Copper None Meth. 1 3 35 to 100 129,332 Running 42 Copper None Meth. 1 3 35 to 100 129,332 Running 44 Copper None Meth. 1 3 35 to 100 129,332 Running 68 Copper None Meth. 1 3 35 to 100 129,332 Running 77 Copper None Meth. 1 3 35 to 100 129,332 Running 81 Copper None Meth. 1 3 35 to 100 129,332 Running 145 Copper None Meth. 1 3 35 to 100 129,332 Running 146 Copper None Meth. 1 3 35 to 100 129,332 Running 147 Copper None Meth. 1 3 35 to 100 129,332 Running 148 Copper None Meth. 1 3 35 to 100 129,332 Running 267 Copper P/M Water 1 1 35 to 100 129,332 Running 268 Copper P/M Water 1 1 35 to 100 129,332 Running 269 Copper P/M Water 1 1 35 to 100 129,332 Running 270 Copper P/M Water 1 1 35 to 100 129,332 Running 271 Copper P/M Water 1 1 35 to 100 129,332 Running 272 Copper P/M Water 1 1 35 to 100 129,332 Running 273 Copper P/M Water 1 1 35 to 100 129,332 Running 274 Copper P/M Water 1 1 35 to 100 129,332 Running 275 Copper P/M Water 1 1 35 to 100 129,332 Running

11

10.0 Heat Pipes in the Field

The previous section covered laboratory based heat pipe tests operated in a controlled environment. Table 2 examines heat pipes that have been operating out in the field since 1980 to the present. Thermacore has never had a heat pipe returned for thermal performance degradation due to some type of internal incompatibility. In all cases, assemblies that have been returned were for dimensional issues or shipping damage.

To further support the reliability of Thermacore heat pipes, in 1993, Thermacore asked to have returned, a sample of 54 copper/water heat pipes from Rockwell International. These units were used for cooling electronics in a military application. Over a 15-year period, these heat pipes accumulated a total of 97,276 hours of operating time. Testing at Thermacore showed that all 54 samples met the original thermal specification. Examination of the end cap joints, the fill-tube pinch-off, the inner envelope wall and the wall/wick interface showed no signs of mass transport or corrosive degradation.

11.0 Heat Pipe Thermal Cycling Information

Thermacore has conducted numerous tests to show the stability of both heat pipes and heat pipe assemblies in thermal cycling environments. In the instance of discrete heat pipes, Thermacore conducted the following tests. 1. Samples of 100 copper/sintered wick/ water heat pipes, 0.25 inch diameter by 6 inches long, were thermally and dimensionally inspected. The units were then subjected to six hundred cycles from –85oC to +100oC, in a vertical orientation. The units were then thermally and dimensionally reevaluated. The units showed no thermal or mechanical change. 2. Thermacore also routinely evaluates its copper/water heat pipes by conducting thermal cycle tests from –65oC to +127oC. This testing is a dimensional check and is intended to assure no degradation in the wick or swelling in the tube wall. Both damage to the wick and the wall can occur if proper fluid charge is not maintained. 3. Thermacore also routinely conducts thermal cycling on heat sink assemblies. These assemblies include the mounting block, heat pipe(s), and fin stack configurations. The possible material combinations include:

12

Table 2. Heat Pipe Field Reliability Data

Company Job Disc. Run Time Approx. Quantity Shipped

Anticipated Product Life

Returns Cause/ Comments

Cincinnati Electric

C5-A Radio Comm.

1968 to 1973

468 10 yr. Life 0 No Known Failures

Intermec Print Head Cooler

1985 to 1992

25,000 10 yr. Life 1 lot (200 pcs.

Machined Hole Dimensions

Xerox Copy Machine Roller

1983 to 1993

2,700 5 to 7 yr. Life

20 pcs. Total

Shipping Damage

Mega Test Semiconductor Test Equipment

1988 to Present

20,000 4 to 6 yr. Life

3 pcs. 7 pcs.

Thermal Interface Problem. Mech. Damage

Hewlett-Packard

Heat Pipe Heat Sink for Server

1996 to Present

12,300 3 to 5 yr. Life

1 box (20 pcs.) 5 pcs.

Shipping Damage Cosmetics

Hewlett-Packard

Convex High Power Mainframe

1996 to Present

2,000 5 plus years 0 --------------

ACEC SRC Cooling for Rail Equip.

1993 to Present

2,250 30 yr. Design Life 1 yr. Field Use

3 pcs. 1 pcs.

Fins Damaged in Shipping Missing Helicoil

Notebook Cooling, Dell, Compaq

All Customers Combined

1995 to Present

1,500,000 2 to 3 yr. Life

Approx. 100 pcs.

Dimension Issues

Rockwell Military 1981 to 1991

10,000 10 yr. Life Approx. 100 pcs.

54 pcs. Test Lot returned after 100k hrs. operation No failures

13

!"Aluminum or copper heat input blocks !"Epoxy or soldered interface joints to heat pipes !"Aluminum or copper fins attached to heat pipes Each of these combinations has been subjected to thermal cycling with acceptable results, however, the test is s strongly dependent upon geometry and materials. It is recommended that all new designs, that deviate significantly from the general geometry’s that have been tested should be thermal cycled to confirm acceptable performance. Typical test requirements would be 10 cycles, -65oC to +127oC, with a 30-minute dwell at each temperature extreme. This format is characteristic of the military requirement defined in Mil-Std-810C.

12.0 Heat Pipe Shock and Vibration Information

Vibration tests have been conducted on several heat pipe assemblies to qualify them for use in various applications. In addition, some units have been installed in vibration environments by several of our customers. Below is a description of the heat pipe assemblies and the vibration conditions to which they were tested. 12.1 Heat Pipe Cold Plate Figure 5 is a plate with embedded heat pipes. There are five heat pipes in each plate. The heat pipes are copper and use water as a working fluid. The heat pipes are embedded in an aluminum web with face sheets soldered top and bottom. This plate is used as a thermal plane for military electronics. Heat from the electronics is transferred to the plate and the plate transport the heat to edge connectors. Hundreds of these units are installed in F-15 LANTIRN navigational pods. The assembly was qualified to mil-std 810C. In addition, several design qualification articles were thermal cycled twenty-five times between –66oC and 85oC with a 15-minute hold time at temperature. Following the test, the articles were examined for compliance with drawing specified geometry compliance. No deformation or deviation from the required geometry was noted.

Figure 5. Heat Pipe Cold Plates

14

12.2 Flexible Heat Pipe Cold Plates The flexible heat pipe cold plate assemblies shown in Figure 6 were developed for aircraft thermal control applications. They are made from copper and used water as the working fluid. Specific information about their design is documented in References 8 and 9. The cold plate portion attaches to the surface to be cooled and the other end attaches to the heat sink. The flexible section allows for ease of installation and it accommodates any relative motion between the heat source and heat sink.

Figure 6. Flexible Heat Pipe Cold Plates Prototype flexible heat pipes were subjected to a series of performance and qualification tests to assure that they would operate in their intended environment. These tests were vibration tests, bellows cycle tests and thermal cycle tests.

The vibration test evaluated the thermal performance and structural integrity of the flexible heat pipe in a vibration environment. The vibration levels that were tested corresponded to the aileron actuator located in region 6B on an F-18. The test procedure that was followed was taken from McDonnell-Douglas report MDC A3376, entitled “F-18 Vibration, Shock, and Acoustic Noise Design Requirements and Test Procedures fir Aircraft Equipment.” Equipment that passes this test has demonstrated the ability to provide adequate operation throughout the life of the aircraft during ground and flight operations. The test sequence consisted of three steps for each axis: a resonance survey, 90 minutes of

15

2-g sinusoidal vibration with dwells at resonance points, and a period of 34-g random vibration. During testing the flexible heat pipe was clamped to a block of magnesium, which in turn was bolted to the vibration table. Heat input was achieved with a Minco brand thermofoil heater glued tot he cold plate. The heat pipe was air cooled by convective air cooled fins that were screwed to the condenser plate. Thermal performance was monitored by several thermocouples attached to the heat pipe. During all phases of the testing, the heat pipe carried 50 watts. Overall, the heat pipe passed the vibration test. The heat pipe thermal performance never changed throughout the test and the heat pipe structural was found to be mechanically sound. Thermal Cycle – The purpose of the test was to assure that no internal or external damage to the flexible cold plate would result from repeated thermal cycles. Testing was done at Thermacore in a Tenny 5 environmental chamber. Three flexible heat pipe cold plates were subjected to the following cycle conditions. • Copper/ Water FHPCP - 3 cycles of –55oC to 135oC

10 cycles of –55oC to 85oC • Copper/ Water FHPCP - 26 cycles of –66oC to 105oC • Aluminum/ Copper/ Water - 10 cycles of –55oC to 105oC All three assemblies survived their test sequence with no visual signs of external damage. Thermal performance data taken before and after thermal cycling showed no difference in performance, indicating that no internal damage occurred. 12.3 Loop Heat Pipe (LHP) A loop heat pipe is new technology that emerged out of the former Soviet Union in 1985. The LHP exceeds standard heat pipe performance in against gravity performance i.e. evaporator above condenser. Figure 7 shows a family of LHPs that were developed for aircraft thermal control applications. A prototype LHP was vibration tested while thermally operating under the same conditions described above for the FHPCP assembly. The prototype operated normally without any temperature fluctuations and it was found to be structurally sound. 12.4 Flexible Heat Pipes Additional flexible heat pipes shown in Figure 8 were developed for cooling unmanned underwater vehicle (UUV) avionics. The curved cold plate portion interfaces with the exterior UUV skin. The flat cold plate attached to the avionics box. These units were shock tested in accordance to shipboard Mk 50 shock

16

criteria. It consisted of longitudinal water entry shock load of 400g for 1msec, vertical 200g for 5 msec, port up and starboard down 200g for 5 msec, and port side up at a 45o angle and shown to operate.

Figure 7. Loop Heat Pipes for Avionics Thermal Control. 12.5 Electrically Isolated Power Cooler The electrically isolated heat pipe power cooler, shown in Figure 9 is designed for cooling power electronics while holding off significant voltage. The ceramic isolators provide the conductive hold off capacity. The internal working fluid is FC-72. Thousands of these units are presently installed on high-speed trains in Europe. They were qualified to CEI 77 paragraph 16. This paragraph requires a search for resonant frequencies from 1 to 50 Hz with amplitudes specified by the following equations: A (mm)= 25/f(Hz) for f from 1 to 10 Hz This corresponds to 0.1 g at 1 Hz and 1 g at 10 Hz A (mm) = 250/f(HZ)2 for f from 10 to 50 Hz This corresponds to 1 g over the entire frequency range. This test was performed in three axis-X, Y and Z. The X and Z-axis showed no excitation over the 1 to 50 Hz test range. The X-axis test showed roughly 4 times amplification of the 1 g input load at 50 Hz. The output signal was still

17

sinusoidal indicting that the part was approaching but not at a resonance frequency. The specification requires a two-hour vibration test at 10 Hz or at any resonance frequencies. Tests were run in all three axis. Y and Z were run at 10 Hz 2.5 mm amplitude. Since the X-axis tests showed the part was approaching a resonance at 50 Hz the two-hour vibration test was run at both 10 and 50 Hz. The final vibration test requires 1 g loading in the X-axis, 3 g in the Y-axis and 5 g in the Z-axis. These g loads were obtained by running the shaker table at 50 Hz with appropriate amplitudes. After shock and vibration testing the unit showed no sign of damage. A warm up thermal test showed the unit still functioning. 12.6 Sonar Heat Pipe One hundred copper/water heat pipes that were 0.5” diameter and 14” long have been installed in a prototype sonar system being developed by Lockheed/Sanders. The heat pipes successfully operated under severe vibration conditions such sinusoidal vibration at 235 Hz and +30 to –1g conditions for 30 million cycles.

Figure 8. Flexible Heat Pipe Assembly Figure 9. Isolated Heat Pipe

18

13.0 External Heat Pipe Assembly Protection

The previous sections addressed the issues of heat pipe life based on internal compatibility. This section examines the external compatibility of a heat pipe assembly.

The surface of the heat pipe heat sink assembly must be protected from environmental exposure, which will either degrade the material of the heat sink or the interface between the fins and the heat pipe, or between the heat pipe and the evaporator block. Most of the heat pipe assemblies that are constructed from copper heat pipes with aluminum fins and aluminum evaporator blocks. For typical electronics cooling applications where the atmosphere is very benign, such as filtered office air, no surface treatment is needed. Unfinished copper and aluminum are acceptable.

In mild environmental conditions, Thermacore will apply one or more of the following: nickel plating, paint, anodization or powder coating. In many situations, customers have selected one of these coatings just for cosmetic reasons.

In applications where the assembly is in a salt fog or a high humidity environment, a surface coating is required to protect the assembly from oxidation and galvanic corrosion. The most relevant issue is contact of dissimilar metals, such as copper tubes with aluminum fins. An unprotected assembly would result in galvanic corrosion at the contact points between these materials.

Thermacore has done extensive work to identify a surface treatment that will protect the entire assembly. Two approaches have been identified: !" Epoxy Coating – A special epoxy coating is applied to the entire heat pipe

assembly. This coating has been tested on several heat pipe assemblies in accordance to ASTM B117 for Humidity and Salt Fog. The assemblies were thermally tested and visually inspected prior to the ASTM B117 test. After the test, the units were thermally re-tested and inspected. The thermal re-test showed no degradation in thermal performance and the visual inspection found no evidence of corrosion, oxidation or cracking of the epoxy coating (Ref. 10).

!" Tin Plating – A protection approach that has been used for several years is to tin plate the entire assembly. Units were tested per ASTM B117. There was evidence of corrosion, however, the tin plating acted as a sacrificial layer and the assembly operated as expected before and after the test (Ref. 11).

The particular protection coating to be selected is dependent upon the final application. Thermacore will council the customer regarding the coating options and the expense related with each.

19

14.0 Heat Pipe Pinch Off Reliability

A mechanical pinch system, common in the HVAC industry, is used to seal the heat pipe. This mechanical pinch of the heat pipe tubing results in a “Cold Welded” seal of the heat pipe fill tube. The fill tube is used to remove air and insert the working fluid into the heat pipe. The mechanical pressure containing strength of this pinch section was measured using a lot of twenty-five heat pipes (Ref. 12). The release temperature of the pinched section is shown graphically in Figure 10.

Figure 10. Heat Pipe Pinch-Off Release Temperature The average temperature that caused the pinch-off to open was 220oC, which corresponds to an internal pressure of 336 psi. Since most electronics are destroyed if they reach this temperature, it is very safe to assume that the heat pipe assembly is still operational.

15.0 Conclusions

Based on the information contained in this document, the following conclusions are reported: !"Thermacore maintains on of the largest heat pipe life test programs with over

200 heat pipes on test at any given time. !"Currently, over 50-copper/water and copper/ methanol heat pipes are on test.

The accumulated hours is 125,000 hours (14 years). All heat pipes are still operating and show no change in performance since beginning of life.

!"There has been no situation in Thermacore’s corporate history where heat pipes have been returned for degradation of performance. Since no heat pipes have failed in the field, a discrete MTBF calculation has not been possible.

Pinch Off Temperature

0

50

100

150

200

250

300

0 5 10 15 20 25

Heat Pipe Number

Tem

pera

ture

, o C

Series1

20

!"Fifty-four copper/water heat pipe units were pulled from a military electronics system with 97,000 hours on operating clock (system was fielded for 15 years). A re-test of the units showed that all fifty-four pieces meet original specifications. Forty-two pieces showed no change from beginning of life; the remaining twelve pieces showed minor change, but in no case more than 3oC. Examination of the end seals, tube inner wall and wick showed no signs of degradation.

!"Based on the results of the Thermacore laboratory life testing, the evaluation of the heat pipes returned from the field and the fact that the copper/water heat pipes are the most widely produced heat pipes with several million units produced annually, it is a safe conclusion that the Thermacore heat pipes are reliable and they will have an operating life exceeding that of the electronics.

16.0 References

1. Dunn, P.D., Raey, D.A., 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. Peterson, G.P., An Introduction to Heat Pipes: Modeling, Testing, and

Applications, Wiley Interscience Publication, Copyright 1994, John Wiley & Sons, Inc.

4. Faghri, Amir, Heat Pipe Science and Technology, Copyright 1995, Taylor &

Francis.

5. Heat Pipe Life Test Data, Water-Cooled, Book A, Thermacore Internal Document, Quality Assurance Department.

6. Heat Pipe Life Test Data, Air-Cooled, Book A, Thermacore Internal

Document, Quality Assurance Department.

7. Wollen, Pete, Procedure for Putting a Heat Pipe on Life Test, Thermacore Internal Memorandum, October 16, 1992.

8. Gernert, N.J., Sarraf, D.B., Steinburg, M., Flexible Heat Pipe Cold Plates for

Aircraft Thermal Control, 1991 SAE Aerotech Conference, Paper No. 912105, Long Beach, CA, September 24-27, 1991.

9. Gernert, N. J., Brown, J., Development of a Flexible Loop Heat Pipe Cold

Plate, SAE Paper No. 951436, Aerospace Atlantic Conference, Dayton, Ohio, May 23-25, 1995.

21

10. Sheffield, Rob, Assessment of Heat Pipe Assembly after Salt Spray S2000 Design, Nortel Networks Report, December 1997.

11. Hartinstine, John, Evaluation of Epoxy Based Coating on Heat Pipe Heat

Exchangers in a Humidity and Salt Spray Environment, Internal Report, February 1, 1999.

12. Dussinger, Pete, Development of Residential Gas-Fired Furnaces using Heat

Pipe Heat Exchangers, Final Report, Gas Research Institute, July 1992.