Full report

Original page 1

NASA/TM-2003-212500 Performance Testing of Thermal Interface Filler Materials in a Bolted Aluminum Interface Under ThemaUVacuurnConditions S.D. Glasgow and K.B. Kittredge Marshall Space Flight Center, Marshall Space Flight Center,Alabama National Aeronautics and Space Administration Marshall Space Flight Center MSFC, Alabama 35812 ~~ June 2003

Original page 2

TABLE OF CONTENTS 1. INTRODUCTION........................................................................................................................... 1 2. MATERIALS TESTED .................................................................................................................. 2 3. TEST APPARATUS ....................................................................................................................... 3 4. TEST PROCEDURES..................................................................................................................... 5 5. RESULTS........................................................................................................................................ 6 6. CONCLUSIONS............................................................................................................................. 9 ... 111

Original page 3

LIST OF FIGURES 1. Test apparatus mounted to coldplate ....................................................................................... 3 2. ......................................................................................................... 4 Test apparatus with MLI 3. Vacuum chamber. data acquisition. and cooling cart.............................................................. 4 iv

Original page 4

LIST OF TABLES 1. Thermal filler materials tested................................................................................................. 2 2. CHO-THERM-like materials at 10 in-lb................................................................................. 6 3. CHO-THERM-like materials at 25 in-lb................................................................................. 6 4. All other materials at 10 in-lb ................................................................................................. 7 5. All other materials at 25 in-lb ................................................................................................. 7 6. Vel-Therm at 10. 25. and 40 in-lb........................................................................................... 8 V

Original page 5

NOMENCLATURE A contact area d bolt diameter f friction factor N number of bolts P contact pressure T bolt torque I I AT change in temperature Vi

Original page 6

TECHNICALMEMORANDUM PERFORMANCE TESTING OF THERMAL INTERFACE FILLER MATERIALS IN A BOLTED ALUMINUM INTERFACE UNDER THERMAL/VACCUMCONDITIONS 1. INTRODUCTION A thermal interface material is one of the many tools often used as part of the thermal control scheme for space-basedapplications.For example, these materials are placed between an avionics box and a coldplate in order to improve the conduction heat transfer so that proper temperatures can be maintained. Interface materials are usually compliant and act to fill the microscopic gaps on a surface so that the area of the heat transfer path is maximized. Any flat surface has hills and valleys in it that are not visible to the naked eye. If two surfaces are placed in contact with each other, only the peaks of the hills will actually contact and create a heat transfer path, thus, greatly reducing the effective amount of energy that can transfer between the two surfaces. Under atmospheric conditions, the gases present greatly aid in heat transfer. Interface materials are not usually required in this case and, in fact, can act as insulators. However, in the vacuum of space, there are no atmospheric gases to aid in heat transfer, and these interface materials are of great benefit. Historically, at Marshall Space Flight Center, CHO-THERM@1671 has primarily been used for applications where an interface material was deemed necessary. However, in recent years, numerous alternativeshave come on the market. It was decided that a number of these materials should be tested against each other to see if there were better performing alternatives. The tests were done strictly to compare the thermal performance of the materials relative to each other under repeatable conditions and do not take into consideration other design issues, such as off-gassing, electrical conduction, or isolation, etc. The purpose of this Technical Memorandum is to detail the materials tested, test apparatus, procedures, and results of these tests.

Original page 7

1. MATERIALS TESTED Twenty different materials tested are listed in table 1 with their respective test number, manufacturer, series, model, thickness, and thermal resistance (provided by the manufacturer). They can be broken down into the following categories: CHO-THERM and similar (tests 1-5), graphite (tests 6-10), foil (test 1I), sandwich (tests 13-16), phase-change material (PCM) (tests 17-20), and other (test 12). Table 1. Thermal filler materials tested. Test No. Manufacturer Series Model 0 - - - 1 Chomerics CHO-THERM 1671 2 Chomerics CHO-THERM T500 3 Thermagon T-pli 220 4 Thermagon T-pli 205 5 Bergquist SiI-pad K-10 6 Graftech eGraf 705 7 Graftech eGraf 1210 8 Graftech eGraf 1220 9 Thermagon T-gon 805 10 Thermagon T-gon 820 11 Indium Corp. Indium foil - 12 Energy Sciences Vel-Therm A 2 0 6 4 2 5 1 Laboratory Inc. 13 Bergquist Q-pad I1 14 Bergquist Q-pad 3 15 AOS Thermal Micro-faze A6 Compounds 16 AOS Thermal Micro-faze K Compounds 17 Thermagon T-pcm HP105 18 Thermagon T-mate 291OC 19 Thermagon T-mate 2920 20 Bergquist Hi-flow 625 2 Vendor-Specified Thickness Resistance (in) (“C i n W ) Bare (no filler) - - Silicone w/Boron Nitride 0.015 0.23 Similar to CHO-THERM 1671 0.01 0.19 Similar to CHO-THERM 1671 0.02 0.21 Similar to CHO-THERM 1671 0.005 0.1 1 Similar to CHO-THERM 1671 0.006 0.41 Graphite 0.005 0.03 Graphite 0.01 0.03 Graphite 0.02 0.07 Graphite 0.005 0.07 Graphite 0.02 0.17 Foil 0.015 0.007 Other 0.02 - Sandwich 0.006 0.22 Sandwich 0.005 0.35 Sandwich 0.006 0.02 Sandwich 0.006 0.03 PCM 0.005 0.015 PCM 0.01 0.09 PCM 0.02 0.27 PCM 0.005 0.71

Original page 8

1. TEST APPARATUS The test fixture consisted of three 6-in square aluminum plates bolted to a liquid-cooled coldplate mounted in a small vacuum chamber. The filler material to be tested was placed between the two plates nearest the coldplate. Each of these plates included four imbedded resistance temperature devices (Mince@ part No. S7798PD) that were connected to an Agilent Technologies@34970Adata acquisition unit for monitoring and recording temperature data. A Minco Kapton@-insulatedthermo-foil heater resided in the interface between the two outermost plates. The heater was wired to a calibratedAgilent 6675A power supply to provide the constant voltage current across the 15.842 heater. The test fixture was mounted to the coldplate with six No. 10machine screws, which also provided the contact pressure across the interface filler. The coldplate was cooled via a Neslab@CFT-150 chiller utilizing a waterethylene-glycol coolant mixture. The contactpressure imposed on the interfacematerial by this setup can be calculatedby equation (1): T x N P = f x d x A ’ where P = contact pressure (psi), T = bolt torque (in-lb), N = number of bolts,f= friction factor (0.2 for unlubricated bolts), d = bolt diameter (in), and A = contact area (in2). Based on this equation, the contact pressure for the 10, 25, and 40 in-lb cases is 44, 110, and 176psi, respectively. The setup is depicted in figure 1. Figure 1. Test apparatus mounted to coldplate. 3

Original page 9

Following initial checkout tests, interface material was placed between the test apparatus and the coldplate to improve the heat transfer to the coldplate. Thermal interface material was also placed between the two outermost aluminum plates along with the heater to help fill surface irregularities and provide more uniform contact between the heater and the plates. Once the test fixture was assembled and mounted to the coldplate, a multilayer insulation (MLI) blanket was placed over it to reduce radiation heat transfer from the test fixture to the chamber walls. Photographs of the assembled test apparatus are shown in figures 2 and 3. Figure 2. Test apparatus with MLI. -- Figure 3. Vacuum chamber, data acquisition,and cooling cart. 4

Original page 10

1. TEST PROCEDURES The approach used in testing was to measure the average temperatures of the two plates on either side of the interface material and use the AT across the interface as a comparison of performance of the materials. A constant (+1 O F ) bottom plate temperature was maintained between each test, and the input voltage applied to the heater was maintained for each test. By using this method, testing was much simpler than trying to account for all energy losses or gains within the system, and it still gave valid results for comparison purposes. Prior to any testing, the entire assembly was placed in the vacuum chamber and baked out for 2 hr at a temperature above 176 O F . After this was complete, the chamber was repressurized, and the bolts were retorqued. All testing was done at less than 1 ~ 1 0 ~torr. A baseline test-no interface material (bare)-plus a test of each material was performed at torque values of 10and 25 in-lb. A 40 in-lb test was also done on Vel-Therm@. The bottom plate temperature and heater voltage were set for each materials test from those established in the baseline test. The settings used were arbitrary, but with the goal of an =90 O F AT. The settings ended up being =80 "Ffor the bottom plate and 70 V for the heater voltage, or =300 W of power. During the early stages of testing, one of the CHO-THERM-like materials (T-pli 220) proved to perform far better than expected and only produced a AT of =6 O F . Based on this result and the fact that a number of the materials that had yet to be tested had far lower vendor-supplied resistance values, it was decided that a higher power level was needed to provide better resolution in the results. Consequently, two subsets of results were obtained. Results from the first subset consisted of the baseline (bare) test and all the CHO-THERM-like materials tested using the previously mentioned settings. The second subset of results were from retesting CHO-THERM 1671 and T-pli 220 at a higher input power and applying those settings to the remaining materials. The settings for the second subset were a bottom plate temperature of =86 O F and an input voltage of 95 V, or =570 W of power. 5

Original page 11

1. RESULTS The results for the CHO-THERM-like materials are shown in tables 2 (10in-lb) and 3 (25 in-lb) in order from least to highest AT. It can be seen from the tables that additional torque provides better results, which is expected. It also shows that none of these particular materials are more sensitive to torque; i.e., the order of the results does not change between the two tables. Table 2. CHO-THERM-like materials at 10 in-lb. Top Bottom Torque Average Average AT ' Test No. Material (in-lb) ( O F ) ( O F ) ( O F ) 3 T-pli 220 10 6.1 I 4 T-pli 205 10 12.5 5 Sil-pad K-10 10 101.4 21.8 1 CHO-THERM 1671 10 112.6 33.3 2 CHO-THERM T500 10 117.0 80.4 36.6 0 Bare I 10 87.2 Table 3. CHO-THERM-like materials at 25 in-lb. Top Bottom Torque Average Average AT Material (in-Ib) (3 ( O F ) ( O F ) T-pli 220 25 84.7 79.8 4.9 T-pli 205 25 88.4 79.6 8.8 Sit-pad K-10 25 96.3 78.7 17.6 CHO-THERM 1671 25 105.4 79.0 26.4 CHO-THERM T500 25 106.7 78.5 28.2 Bare 25 143.6 79.9 63.7 Tables 4 (10 in-lb) and 5 (25 in-lb) show the results for the rest of the materials tested at the higher power levels. The same general trends can be seen for these materials. Two pairs of materials do swap places with the higher torque value but the ATs show that they are very close together in both cases. 6

Original page 12

Table 4. All other materials at 10 in-lb. Torque rest NO. Material (in-I b) 17 T-pcm HP105 10 12 Vel-Therm 10 3-v T-pli 220 10 20 Hi-flow 625 10 19 T-mate 2920 10 8 eGraf 1220 10 13 Q-padII 10 7 eGraf 1210 10 18 T-mate 291OC 10 11 Indium 10 10 T-gon 820 10 15 Micro-fazeA6 10 9 T-gon 805 10 6 eGraf 705 10 14 Q-pad3 10 16 Micro-faze K6 10 1-v CHO-THERM 1671 10 Bottom Average Average AT ( O F ) ( O F ) 96.9 90.2 6.7 93.3 86.3 7.0 95.3 85.6 9.7 97.2 84.1 13.1 100.3 84.5 15.8 106.4 85.7 20.7 108.0 86.7 21.3 108.6 85.9 22.7 108.5 85.6 22.9 117.2 86.4 30.8 118.5 85.8 32.7 119.0 85.8 33.2 120.7 86.4 34.3 119.5 84.7 34.8 121.9 87.0 34.9 138.2 83.7 54.5 140.6 85.7 54.9 Table 5. All other materials at 25 in-lb. Torque Average AT rest No, Material (in-lb) ~ ~ ~~ 17 T-pcm HP105 25 12 Vel-Therm 25 3-v T-pli 220 25 20 Hi-flow 625 25 19 T-mate 2920 25 13 Q-pad II 25 8 eGraf 1220 25 7 eGraf 1210 25 18 T-mate 291OC 25 11 Indium 25 10 T-gon 820 25 15 Micro-fazeA6 25 9 T-gon 805 25 14 Q-pad3 25 6 eGraf 705 25 16 Micro-fazeK6 25 1-v CHO-THERM 1671 25 Bottom (3 ("F) 91.9 85.4 6.5 91.1 84.4 6.7 93.3 85.5 7.8 99.0 86.0 13.0 101.5 85.9 15.6 103.1 85.6 17.5 103.0 85.3 17.7 106.3 86.1 20.2 106.5 85.4 21.1 107.0 85.1 21.9 109.1 85.1 24.0 109.9 85.4 24.5 112.6 86.5 26.1 114.6 86.1 28.5 115.8 86.4 29.4 125.1 86.9 38.2 128.9 86.7 42.2 7

Original page 13

Table 6 shows the results for Vel-Therm for all three torque cases. It was expected that with higher torque, the Vel-Therm would not perform as well. This is because the material consists of carbon fibers, which tend to get crushed at higher torque values, and the fibers are not effective at moving energy when this happens. As can be seen from the table, it does perform slightly better at 25 in-lb, but it loses performance at the 40-in-lb level. Table 6. Vel-Therm at 10,25, and 40 in-lb. Toque Test No. Material (in-lb) 10 93.3 86.3 7.0 12 Vel-Therm 12 25 84.4 6.7 Vel-Therm 40 91.5 84.2 7.3 12 Vel-Therm 8

Original page 14

1. CONCLUSIONS The results show that there are many materials currently available that perform quite well. Cost is not a big considerationbetween any of them with the exception of Indium@and Vel-Therm, which are much more expensive than the others. There are many design considerationsthat come into play when trying to choose a suitable candidate, but these data should help with the thermal performance aspect of that decision. From a mainly thermal perspective, the following conclusions can be made: CHO-THERM 1671 is much better than a bare interface but it is one of the poorest performers in the group tested. There is little correlation between the manufacturer’s thermal resistance data and the results from these tests, indicating that there is more to interface performance than just material properties. Graphites tended to improve with thickness. This was unexpected but may be pressure related if the graphite fillers are not as compliant as the silicone-based fillers. Indium was disappointing for the price. It may need higher pressures to conform to minor surface irregularities. There was little difference in the top two performers except price: Vel-Therm, $1000 and HP105, $16. The extra $984 buys a somewhat easier removal process; also, note that since HP105 is a PCM, it may have off-gassing problems. T-pli 220 had the best combination of thermal performance, price, and ease of use. Performance is consistent with the top two, but it is a CHO-THERM 1671-likefiller. The only category where it does not outperform 1671 is in ease of reuse, which, at $38 a sheet, should not be an issue. 9

Original page 15

Form Approved REPORT DOCUMENTATION PAGE OMB NO. 0704-0188 I I Public reportingburden for this collectionof informationis eslmaled lo average 1hour per response. includingthe time for reviewing instructions.searching existingdata sources, gatheringand maintainingme data needed. and completingand reviewingthe collectionof information. Send comments regarding this burden estimate or any other aspect of this colleclbn of information,includingsuggestionsfor reducingIhii burden, lo Washington HeadquartersS e ~ t m s .Directorate for InformationOperationand Reporls. 1215Jeflerson Davis Highway. Suite 1204. Atlington.VA 22202-4302.and lo the Officeof Managementand Budget. Papemork Reduction Project (0704-0188). Washington. Dc 20503 1. AGENCY USE ONLY (Leave Blank) 2. REPORT DATE June 2003 4. TITLE AND SUBTITLE 3. REPORTTYPE AND DATESCOVERED Technical Memorandum 5. FUNDING NUMBERS Performance Testing of Thermal Interface Filler Materials in a Bolted Aluminum Interface Under ThermalNacuum Conditions 6. AUTHORS S.D. Glasgow and K.B. Kittredge 7. PERFORMINGORGAMZATIONNAMES(S) AND ADDRESS(ES) George C. Marshall Space Flight Center Marshall Space Flight Center, AL 35812 I - 8. PERFORMING ORGANIZATION REPORT NUMBER M-1075 NAME(S)AND ADDREWES) 10. SPONSORING/MONITORING 19. SWNSORINOMONITORINGAGENCY National Aeronautics and Space Administration Washington, DC 20546-0001 11. SUPPLEMENTARY NOTES AGENCY REPORT NUMBER NASA/TM-2003-2 12500 I IPrepared for Structures, Mechanics, and Thermal Department, Engineering Directorate 12a. DISTRIBUTION/AVAILABlUTYSTATEMENT Unclassified-Unlimited Subject Category 18 Nonstandard Distribution 13. ABSTRACT (Maximum 200 words) 12b. DISTRIBUTION CODE A thermal interface material is one of the many tools often used as part of the thermal control scheme for space-based applications. Historically, at Marshall Space Flight Center, CHO-THERM 1671 has primarily been used for applications where an interface material was deemed necessary. However, numerous alternatives have come on the market in recent years. It was decided that a number of these materials should be tested against each other to see if there were better performing alternatives. The tests were done strictly to compare the thermal performance of the materials relative to each other under repeatable conditions and do not take into consideration other design issues, such as off-gassing, electrical conduction, isolation, etc. The purpose of this Technical Memorandum is to detail the materials tested, test apparatus, procedures, and results of these tests. The results show that there are a number of better performing alternatives now available. thermal interface material, filler, and testing; thermal filler; 15. NUMBEROF PAGES 14. SUBJECTTERMS 16 thermal gasket; thermal grease: heat transfer; avionics cooling; bolted 16. PRICE CODE interface thermal conduction; vacuum thermal interface 17. SECURITY CLASSIFICATION 18. SECURITY CLASSIFICATION 19. SECURITY CLASSIFICATION 20. LIMITATIONOF ABSTRACl OF REPORT OF THIS PAGE I Unclassified Unclassified NSN 7540-01-280-5500 OF ABSTRACT Unclassified Unlimited Standard Form 298 (Rev.2-89) Prescribed by ANSI Std 239-18 298-102