DEVELOPMENT, DESIGN, AND TESTING OF POWERSPHERE ...

AIAA-2003-1897 DEVELOPMENT, DESIGN, AND TESTING OF POWERSPHERE MULTIFUNCTIONAL ULTRAVIOLET-RIGIDIZABLE INFLATABLE STRUCTURES Edward J. Simburger*, Thomas W. Giants*, James H. Matsumoto*, Alexander Garcia III* John K. Lin†¶, Jonathan R. Day†#, Stephen E. Scarborough†¶, Suraj Rawal‡, Alan R. Perry‡, Craig H. Marshall‡ Henry B. Curtis§, Thomas W. Kerslake§, Todd T. Peterson§ Abstract The continuing advancement of microsatellites and nanosatellites for space application has identified a need for an innovative, simple, low mass, and low packing volume solar array system for this class of satellites. As microsatellites and nanosatellites become more capable, their demand for power also increases. The result of decreasing satellite size, thus surface area, and increasing power demand is the inevitable problem of “power choke” on satellite systems with bodymounted solar cells. The use of traditional, rigid, solar arrays necessitates larger stowage volume and mass, and also requires gimbals for pointing. In previous studies, the “PowerSphere” solar array system has been identified as one of the solutions to this “power choke” problem. This power system has several advantages over the traditional system, including lower mass, lower stowage volume, and the elimination of the pointing mechanism. The development of PowerSphere multifunctional ultraviolet-rigidizable (UV-rigidizable) inflatable structures will be beneficial not only to micro- and nanosatellites; it will enable the advancement of Gossamer type structures. Given the requirements of PowerSphere to be low mass, low packing volume, and low power for deployment, ultraviolet (UV) cured composite (i.e. UV-rigidizable) materials have been identified as the optimal candidate for this application. In conjunction with UV-rigidizable material, these structures incorporate thin film flexcircuit to make them multifunctional. Two multifunctional structures in the PowerSphere system will utilize the UV-rigidizable material, namely the inflatable hinge and the inflatable isogrid center column. The PowerSphere team, which is assembled collaboratively from Aerospace, ILC Dover, Lockheed Martin, and NASA Glenn, has made significant advancements in the development of multifunctional UV rigidizable inflatable structures. *

The Aerospace Corporation, Los Angeles, CA ILC Dover, Inc., Frederica, DE ‡ Lockheed Martin Corporation, Denver, CO § NASA Glenn Research Center, Cleveland, OH ¶ Member AIAA # Member ASME †

Copyright © 2003 by The Aerospace Corporation. Published by the American Institute of Aeronautics and Astronautics, Inc., with permission

Introduction The original charter for the development of PowerSphere is to address the issue of “power choke” in small satellite systems, particularly micro- or nanosatellites.1,2 The requirements of PowerSphere to be low cost, low mass, low packing volume, and low power for deployment dictated the research and development of multifunctional UV-rigidizable inflatable structures as an enabling technology. As the technology advanced, it was quickly realized that the development of this enabling technology would allow the realization of not only PowerSphere in particular, but Gossamer type space structures in general. Although the current research effort is still PowerSphere specific, the PowerSphere team has recognized the significance of this research as it relates to other applications. In the previous phase of research, configuration trade studies such as system configuration, preliminary mass analysis, hinge configuration, and center deployment column configuration have been completed.3,4,5 The effort of the current study is to develop and mature the selected configurations from concept to proof-of-concept engineering prototype. To successfully develop a multifunctional system involves a cooperative effort across disciplines and companies. The development of a multifunctional UV-rigidizable inflatable structure requires innovations and advancements in several technologies including materials, structures, ultra-low-mass inflation systems, electrostatic discharge (ESD) protective coatings, thin film flex circuits, and associated thin film applications and fabrication processes. The advancement of each of these technologies needs to be accomplished through systematic analysis and testing. In addition, in order to bring the PowerSphere system to the engineering prototype level in the near future, the PowerSphere team has begun fabrication trials of various subcomponents. The work presented in this paper focuses on the development, design, analysis, and testing of two PowerSphere sub-components, namely the multifunctional UV-rigidizable inflatable hinge and isogrid center column. The paper details the effort from design and analysis to prototype fabrication.

1 American Institute of Aeronautics and Astronautics

System Design The baseline design of the PowerSphere spherical solar array system consists of two semi-spherical domes that connect to a central spacecraft. Each semispherical dome consists of two different sub-modules assembled together to form a geodetic sphere of hexagon and pentagon solar panels (Figure 1). One of

(or end cap) by three inflatable UV-rigidizable hinges integrated with a flex circuit blanket. In addition, this module consists of an inflatable UV-rigidizable isogrid center column with an integrated flex circuit that also mounts to the instrument deck. A single semi-spherical dome requires three “ Sub-Module A” assemblies and one “ Sub-Module B” assembly.

Figure 3. Sub-Module B Assembly Figure 1. PowerSphere Assembly the sub-modules, which is called “ Sub-Module A” (Figure 2), consists of three hexagon solar panels, one pentagon solar panel, and four inflatable UV-rigidizable hinges integrated with flex circuit blanket to form one multifunctional module. The other sub-module, which is called “ Sub-Module B” (Figure 3), consists of three pentagon solar panels mounted to the instrument deck

Figure 2. Sub-Module A Assembly

The use of UV-rigidizable inflatable structures enable the PowerSphere system to have a small packing volume and minimal power required for rigidization. The center column will be z-folded for packing. Release ties between the instrument panel and the spacecraft bus will secure the packed center column. The hinged solar panels will fold and stack on top of each other. The solar panels will also be secured to the instrument panel with release ties (Figure 4, note that release tie and soft cover are not shown). After launch, the center column release ties are triggered allowing each center column to inflate and deploy. The UV radiation from the sun or blue or green light emitting diodes (LED) will cure the tube into a rigid supporting column. Then the solar panels release ties are triggered allowing the hinges to inflate and deploy the solar panels into the geodetic shape. Likewise, the UV radiation from the sun will cure each hinge into a rigid structural component. A passive inflation method via a material’s vapor pressure is used to inflate the center column and the hinges. This is a very simple inflation method that does not require power for initiation. The integrated flex-circuits supply the power from each solar cell to the spacecraft. The flex circuits are on the backside of the solar panels. Low profile connectors will be used to make electrical connections between Sub-Module A, the instrument panel, the

2 American Institute of Aeronautics and Astronautics

center column, and the spacecraft. The interface hinge on “ Sub-Module A” is designed with a quick disconnect feature to aid the manufacturing process and to allow component replacement at the sub-module level if the module assembly failed during fabrication or testing. A flex ribbon will be integrated into the inside of the center column. It will be attached to a connector mounted on the instrument panel and a connector on the spacecraft. This concept will provide the required airtight seal of the center column. The PowerSphere Assembly will be coated with ITO MgF2 to prevent damage due to electrical arcing.

bladder, a top UV rigidizable hinge, and a bottom nonrigidizable hinge, (Figure 5). The hinge is very flexible in the uncured state because it is comprised of thin films and uncured resin. This flexibility allows a single hinge size to fold and pack multiple layers of varying stack height. After the solar panel release ties are triggered the hinge bladders will inflate causing the solar panels to unfold into the geodetic shape. Then the top hinge will cure from UV radiation to become a rigid structural component. The cured hinge stiffness must be great enough to withstand the centripetal force caused by a spinning satellite system such that the deflection of the hinge is within acceptable geodetic shape limits. Inflated Bladder

Top Hinge

Bottom Hinge

Figure 5. Hinge Cross-Section Figure 4. Packed PowerSphere Assembly Multifunctional UV-Rigidizable Inflatable Structures The PowerSphere system consists of two multifunctional UV-rigidizable inflatable structures: (1) Multifunctional UV-Rigidizable Inflatable Hinge, and (2) Multifunctional UV-rigidizable Inflatable Isogrid Center Column. These two structural/mechanical components utilize the same rigidization (i.e. composite material) system, but employ a significantly different structural and mechanical designs to accomplish their requirements. Multifunctional UV-Rigidizable Inflatable Hinge The multifunctional UV-rigidizable inflatable hinge for the PowerSphere is a critical element with several design requirements. The hinges must be able to pack the solar panels, unfold the solar panels to the proper angle, and maintain accuracy and rigidity after deployment to the end of the mission. Hinge Design – To achieve the requirements just mentioned, the base design consists of a tubular

Details of the hinge design are as follows. The top hinge is a laminate of UV rigidization material encased between two film layers. The UV rigidization material must be encased in film to mitigate outgassing or adhere to itself when packed. The bottom hinge is a single layer of film. The top and bottom hinge widths span the length of the cylindrical portion of the bladder and bond to the solar panel frames. The top hinge length will be longer than the bottom hinge. The difference in the run-lengths is such that the proper angle of the panels is obtained upon inflation. The diameter of the hinge is dependent on the maximum stack height. The run-length of the bottom hinge must be enough to span the stack height. The bladder is indexed to the top hinge, the frame sides, and the bottom hinge. The indexing reduces the peel force on the hinge to frame bond and helps maintain the accuracy of the hinge angle. The flex circuit, which is fully integrated with the thin film solar cell, is incorporated into the hinge lay-up between the bladder and the bottom hinge. The use of the flex circuit in conjunction with UV-rigidizable material for the hinge assembly allows the PowerSphere to be efficiently packed.

3 American Institute of Aeronautics and Astronautics

Hinge Fabrication – A number of experimental hinges were fabricated for this development effort for the purpose of defining the hinge manufacturing process and to determine how accurately they could be fabricated (Figure 6). Patterns were generated using AutoCAD for the bladder, the top hinge layer, and the bottom hinge layer. The computer-generated patterns were used to precision cut and mark the film materials on a wheel cutter. After the manufacturing trials, the patterns were adjusted as necessary and a hinge was fabricated and inflated to approximately 1 psi. Using a

maximum deflection of the hinge under load. This also indicates that the design will be frequency driven in the final analysis. The primary load that will deflect the hinge comes from the predetermined rotation of the PowerSphere. In this calculation, the energy method based on Castigliano’s theorem was applied to determine the deflection of the Sub-Module A assembly at various points (Equation 1). Figure 7 shows the free-body  3 t π ⋅ P⋅ R  h ⋅ (1 + ν ) + t h + 3 δ P , R , t h , E := ⋅  2 ⋅ E⋅ I t h  2 2 12 ⋅ R  5⋅ R 

(

Where, E I(th) P R th δ ν

Figure 6. Inflatable Hinge Fabrication Experiment template, the hinge angle was measured and determined to be 40.5 degrees. This meets the requirement of the hinge design angle of 41.8 ± 2 degrees. The manufacturing knowledge gained during these experiments will be used to fabricate the hinges that will connect the solar cells for solar modules A and B in the engineering prototype. Analysis and Testing – Structural analysis and testing for the hinge is important to optimize the design. From preliminary calculations it is found that in this application the size of the hinge is not determined or governed by strength but by deflection. In other words, the parameter that governs the design of the hinge is the

)

2

2

( )

Young’s Modulus Section Modulus (function of thickness) Vertical load applied Radius of the hinge (test specimen) Thickness of the hinge (test specimen) Deflection of the hinge Poisson’s ratio

diagram of the Sub-Module A. Based on the assumptions made in the free-body diagram, the results obtained are conservative. The results of the calculation (Figure 8) indicate that with the composite material given (thus the material properties are defined) the section modulus of the hinge and the frame are the factors determining the deflection under a given rotational speed. Therefore, depending on mission

Free-Body Diagram of a Single Hinge • V = Shear Load Induced by Angular Velocity • M = Moment Induced by Angular Velocity • Varying the Thickness of Material • Applying Energy Method Based on Castigliano’s Theorem

Sub-Module A

Eq. (1)

L1 Ls

Free-Body Diagram of the Sub-Module A

Figure 7. Hinge Deflection Calculation Assumption and Free-Body Diagram 4 American Institute of Aeronautics and Astronautics

requirement the design of the hinge and the frame should be optimized according to the rotational requirement. Currently, the rotational requirement is 4 to 60 rpm. Based on the maximum rotational speed requirement and the maximum deflection of 1.0 centimeter, the section modulus of the hinge should be greater than 7.6 x 10-10 m4 and the section modulus for the frame should be greater than 1.0 x 10-10 m4. Currently, the PowerSphere design team is optimizing the hinge structural design. 8 .10

10

A ssum ing 0.01 m m axim um deflection

Section Modulus (m^4)

10 7 .10

If Ih

the isogrid pattern. The tows must be kept under tension to maintain the isogrid pattern. Furthermore, the transition of the tows to the flange must be as smooth as possible, i.e. the bending in the tows at the interface to the flange must be minimized. The UVrigidizable tows must be encased in film such that it does not off-gas or bond to itself when packed. The flange will allow for circular bolthole attachment to the solar panel container and an o-ring seal for inflation gas retention.

6 .10

10

5 .10

10

4 .10

10

3 .10

10

2 .10

10

1 .10

10

Section Modulus of the Hinge

j

j

Section Modulus of the Frame

0

1

1.5

2

2.5

3

3.5

4

4.5

5

5.5

6

6.5

ωj Angular Velocity (Hz)

Figure 8. Section Modulus of Frame and Hinge vs. Angular Velocity Multifunctional UV-Rigidizable Inflatable Center Column Center Column Design – The rigidizable inflatable center column must pack into a small volume, deploy the solar panel assembly, and must be the structural component connecting the solar panels to the spacecraft. The selected design for the center deployment column is an inflatable UV rigidizable isogrid boom consisting of a grid work of UV curable composite tows integrated into an air-tight, UV transparent, thin film lay-up (Figure 9). The flex circuit is integrated into the internal bladder of the column wall. Likewise with the hinge, the center column is very flexible in the uncured state. This allows for compact z-fold packing. The length of the center column is sized to the specific geodetic sphere for proper alignment of top and bottom spherical domes. After inflation, UV radiation will cure the center column into a rigid structural component. The strength of the cured center column is designed to withstand the applied moment at the tip of the column due to the rotation of the PowerSphere. The center column interface design involves integrating the uncured UV rigidization material to a flange. The UV rigidization material is in the form of tows that must be accurately wound onto a mandrel in

Figure 9. Center Column Center Column Fabrication – The center column (Figure 10) is fabricated using a modified filament winding operation. Patterns were generated in order to manufacture 3-inch diameter, 12-inch long isogrid booms that will be used as test specimens and engineering models. Both the bladders and antiblocking layers were precision cut and marked on a wheel cutter. The bladders and anti-blocking layers are fabricated from 2 mil thick Kynar. The tows consist of

Figure 10. Isogrid Center Column three rovings of Owens Corning’s 449 1250 S-2 glass fiber impregnated with Adherent Technologies’ ATIP600-2 UV curable epoxy resin6. The tubes are bonded onto the flanges, inflated, and then cured using UV energy from the sun. Center Column Analysis and Testing – The analysis presented in this sub-section employs classical theory to analyze the structural performance of the isogrid

5 American Institute of Aeronautics and Astronautics

of an individual tow (or rib element), were investigated. The result of the buckling analysis showed that the local Euler buckling of the rib is the limiting case (Figure 12). 5000

Com paring the Three M odes of Buckling 4500

Euler Column Buckling 4000

A ssum ption: 3 Ro ving b = 1.5535 mm [0.06116 in]

3500

Wall Buckling 3000 Rib Buckling Load (N)

column7. The structural analysis is presented without derivation. The purpose of this calculation is to optimize the various parameters (particularly radius of the column, number of axial tows around the circumference, and the diameter of the tow) of an isogrid tube for the PowerSphere application. From the isogrid center deployment column assembly, as shown in Figure 9, assumptions were made about the loading and boundary conditions to simplify and to arrive at the free-body diagram as shown in Figure 10. The center column is assumed to be a Cantilever column with fixed-free end conditions. The applied moment at the tip of the column is due to rotation of the PowerSphere during deployment. In the calculation, a lateral acceleration of 0.1 g is also applied to the center column. The cross section of the isogrid tow (Figure 11) is assumed to be circular.

P RCR ( r , n , b ) P CL ( r , n , b , L )

2500

P E( r , n , b , L)

2000

Local Rib Buckling is the Limiting Case

1500

Local Rib Buckling 1000

500

0

6

7

8

9

10

11

12

13

14

15

16

n Number of Axial Tows

Figure 12. Three Modes of Buckling Failure

Figure 10. Isogrid Center Deployment Column and Free-Body Diagram

Comparing the predicted loading condition with the local rib buckling limit (the local buckling limit was calculated with number of axial tows and tow diameter as variables) an optimized solution was obtained (Figure 13). The configuration selected for fabrication and testing is as follows: • Diameter of the column = 7.62 cm [3.0 in] • Number of axial tows = 8 • Tow diameter = 1.55 mm [0.061 in] (This diameter is equal to three roving of S2-glassfiber) 100

Comparing Three Different Tow Diameters

90

80

70

Rib Buckling Load (N)

P RCR ( r , n , 0.03532 ⋅ in) n

3 Roving b = 1.5535 mm [0.06116 in]

60

P RCR ( r , n , 0.04994 ⋅ in) n

50

P RCR ( r , n , 0.06116 ⋅ in)

2 Roving b = 1.2685 mm [0.04994 in]

n Pcomp ( r , n , b , L)

40

PRCR ( r , 8 , 0.06116in) 8

= 24.441 N

30

PRCR ( r , 12 , 0.04994in)

20

12

0

= 24.447N

1 Roving b = 0.8971 mm [0.03532 in]

10

6

7

8

9

10

11

12

13

14

15

16

n Number of Axial Tows

Figure 11. Isogrid Wall of the Center Deployment Column To optimize the parameters, three modes of failure, namely global Euler buckling of the column, general instability of the isogrid wall, and local Euler buckling

Figure 13. Rib Buckling of Three Different Tow Diameters Mechanical properties such as mass, frequency, and deflection have also been calculated for comparison (Figures 14 - 16).

6 American Institute of Aeronautics and Astronautics

Materials Development and Testing The successful development of inflatable deployable structures relies mainly on the selection of materials to be used. A class of materials called “ rigidizable materials” is usually needed in order to meet system level requirements. Rigidizable materials can be defined as materials that are initially flexible to facilitate compact packing/stowage, inflation or deployment, and become rigid when exposed to an external influence.8 The external influence can be extreme hot or cold temperatures, radiation in the electromagnetic spectrum, or a chemical reaction caused by the inflation gas.

Mass Comparison 0.045

0.04

Mass of the Isogrid (kg)

0.035

M ( n , 0.03532 ⋅ in , L) 0.03

M ( 8 , 0.06116in , L) = 0.025 kg

M ( n , 0.04994 ⋅ in , L)

M ( 12 , 0.04994in , L) = 0.025 kg

M ( n , 0.06116 ⋅ in , L) 0.025

0.02

0.015

0.01

0.005

6

7

8

9

10

11

12

13

14

15

16

n Number of Axial Tows

Figure 14. Isogrid Center Column Mass Comparison among Three Different Tow Diameters

40

Frequency Comparison 35

Frequency (Hz)

30

fff ( r, 8 , 0.0611in, L) = 28.066Hz

fff( r , n, 0.03532in , L) fff( r , n, 0.04994in , L)

25

fff( r , n, 0.06116in , L)

20

15

10

6

7

8

9

10

11

12

13

14

15

16

n Number of Axial Tows

Figure 15. Isogrid Center Column Frequency Comparison among Three Different Tow Diameters

3 .10

4

Tip Deflection Comparison

Tube Deflection at the Free End (m)

4 2.5.10

2 .10

4

δ(r, n, 0.03532 in, L) δ(r, n, 0.04994 in, L)1.5.10 4 δ(r, n, 0.06116 in, L) 1 .10

4

5 .10

5

δ(r, 8, 0.06116 in, L) = 0.069mm

0

6

7

8

9

10

11

12

13

14

15

n Number of Axial Tows

Figure 16. Isogrid Center Column Tip Deflection Comparison among Three Different Tow Diameters

16

The PowerSphere team has identified UV cured composite materials as the optimal candidate for the PowerSphere program. This decision was based on the available spacecraft energy requirements, packing volume, expected operational temperature regimes, and mass requirements. Prior to selecting the UV material as the primary resin candidate, other rigidization and deployment mechanisms were identified. These include thermoplastic shape memory composite materials and mechanical solutions. UV-Rigidizable Material Development In order to select the reinforcement fiber for the program, an extensive search of commercially available fibers was performed. Initially, this list included only carbon fibers, but was later expanded to include glass fibers, which could be used with UV-rigidizable resins. Properties that are important to fiber selection are specific stiffness, strain to failure, environmental resistance, thermal conductivity, and coefficient of thermal expansion (CTE). CTE is important in developing a near zero CTE structure that would not suffer shape changes from variable thermal inputs on the structure. Carbon, Kevlar, PBO, and Vectran all exhibit low or negative CTE’ s. S-glass also has a fairly low CTE (1.6x10-6 ppm/oC). E-glass has a slightly higher CTE (5.4x10-6 ppm/oC). Out of these fibers, only carbon, E-glass, and S-glass have the environmental resistance required for the PowerSphere application. The other fibers have various issues with degradation due to tight folding, weaving, or UV exposure. As the design of the multifunctional inflatable structure matured, it became clear from the results of various trade studies on the different design options that an UV curable resin is the most attractive choice for the rigidization mechanism. This is because UV curable resins require little spacecraft energy to cure and may provide the simplest design. The UV-rigidizable resin chosen for further development and testing was AP6002 UV epoxy resin, recently developed by Adherent

7 American Institute of Aeronautics and Astronautics

JAN3002I: AP600-2 Cured UV Resin Tube at 80°C. 0.25

JAN3002I: AP600-2 Cured UV Resin Tube at 80°C. 0.10 0.09 0.08 0.07 0.06

QCM at 160 K QCM at 220 K QCM at 298 K

0.05

Volatile

UV-Rigidizable Material Properties Testing Outgassing Test – Outgassing performance was tested for the AP600-2 UV epoxy resin because of its importance in spacecraft application. Based on the PowerSphere system level requirements, the ASTM E1559 test was performed at 80oC for 24 hours under hard vacuum, to simulate worst case conditions for the material. The sample was fabricated as close as possible to the actual end use application (an inflatable, rigidizable, deployable, UV/Glass hinge). It was a 12.7mm diameter, 5.1cm long composite tube with a ½ mil thick Mylar bladder and restraint. It was sunlight cured for four hours and it also had two 1.27mm diameter venting holes in the restraint. Normally outgassing samples are “ baked out” at a defined time and temperature to remove volatile materials prior to test. However, UV-rigidizable resins will not be able to be baked out because the resin would cure during the bake out process. Because of this, the outgassing test sample was not baked out prior to the test. Test data from the AP600-2 UV curable epoxy resin, is summarized below in Figures 17 and 18

difference in the breaking strength between the warp and the fill, this was not considered significant and therefore the data presented here is considered an average of both the warp and the fill. The AP600-2 resin was tested in 1, 2, and 3 ply composites. Five samples of each were fabricated and tested to obtain some statistical significance. All of the samples were cured under vacuum. The UV samples were cured in sunlight for approximately five hours. Table 1 summarizes the results of the testing with representative materials for comparison.

Volatile Condensable Material (%)

Technologies. The resin has a glass transition temperature (Tg) of 211oC, well above the required operational temperature. It can be cured from UV energy from the sun in approximately 10-45 minutes depending on shadowing effects, material thickness, and the UV transparency of the material. The uncured composite fabric (prepreg) coated with the UV epoxy resin must be heated to at least -20oC prior to deployment and rigidization.

0.04 0.03 0.02 0.01 0.00 0

4

8

12

16

20

24

Time (hr)

Figure 18. UV Resin %CVCM outgassing Results Table 1. Tensile Properties Test Results Resin

Fabric

Plies

Vf (%)

AP600-2 UV Epoxy* AP600-2 UV Epoxy* AP600-2 UV Epoxy* TP406 SMP Polyurethane TP406 SMP Polyurethane

S-Glass 8HS S-Glass 8HS S-Glass 8HS S-Glass 8HS S-Glass 8HS

1 2 3 1 2

0.45 0.48 0.54 0.56 0.60

Tensile Strength (MPa) 413.69 497.80 437.82 332.19 377.83

Tensile Modulus (GPa) 17.93 14.82 13.35 15.50 13.64

ASTM D3039 D3039 D3039 D3039 D3039

0.20

Total Mass Loss (%) Total Mass

QCM at 80 K

0.15

0.10

0.05

0.00 0

4

8

12

16

20

24

Time (hr)

Figure 17. UV Resin %TML Outgassing Results Tensile Properties Test – The composite tensile modulus and strength test was performed on a 58x54 count, S-glass, 8 harness satin weave fabric impregnated with AP600-2 UV epoxy. The breaking strength of this fabric is 221 and 235 lb/in in the warp and fill, respectively. Because there is only a 6%

Thermal Cycle Testing of UV Rigidizable Material Composite samples (1, 2, and 3 ply) were also fabricated from the 8HS glass fabric and the AP600-2 UV epoxy resin for thermal cycling tests. The samples were consolidated under pressure and sunlight cured for approximately five hours. They were then thermal cycled from –138oC to +100oC for up to 1000 cycles. After thermal cycling, the samples were cut into 1” wide strips and tensile tested in accordance with ASTM D3039. From Figures 19 and 20 it is apparent that there is some slight reduction in tensile properties after the thermal cycling. One of the visible effects of the thermal cycling was the “ yellowing” of the composite. To determine the cause of this yellowing, a three-ply sample was placed in a 100oC oven for 24 hrs and then examined. It was also tensile tested to determine any strength or stiffness loss. After the elevated temperature exposure, the color of the sample was

8 American Institute of Aeronautics and Astronautics

similar to the color of the thermal cycled samples. The results show that there is slightly more reduction in properties from elevated temperature exposure alone than from thermal cycling. TP406 and UV Resins Vs. Average Tensile Peak Load

9000

8027.3

8000

*From -138C to +100C

7376.8 6885.1

Average Peak Tensile Load (N)

7000

6430.9

6000

5601.6 4934.0

5000 4000 3000

2576.5 2276.8

2318.6

2000 1000 0 1 Ply, Pristine, TP406

1 Ply, Pristine, UV

1 Ply, Thermal Cycled 980 times*, UV

2 Ply, Pristine, TP406

2 Ply, Pristine, UV

2 Ply, Thermal Cycled 1000 times*, UV

3 Ply, Pristine, UV

3 Ply, Thermal Cycled 1000 times*, UV

3 Ply, 24hrs @100C, UV

Environmental Exposure

Figure 19. Peak Tensile Load Results from Thermal Cycling Test 20 17.9

18

TP406 and UV Resins Vs. Average Tensile Modulus *From -138C to +100C

16

Tensile Modulus (GPa)

14

when packed and therefore, an outer film layer (called an anti-blocking layer) is required in both of these structural concepts. The anti-blocking layer also acts as a protective barrier to reduce excess volatile materials that could be released onto the solar cells from the uncured prepreg prior to cure. The anti-blocking layer and the bladder have dissimilar functions in the final structure and therefore also have different performance requirements. The anti-blocking layer must allow UV energy to pass through the film in order to cure the prepreg in the desired wavelengths (i.e. below 385nm). To assist in the curing process, it is preferable to have a vapor deposited aluminum (VDA) coating on the bladder, which will reflect a percentage of UV energy back into the prepreg. Both films must have the ability to be packed tightly without tearing or pin holing. They are also required to have low permeability to a variety of chemicals. In addition, the films must have low elongation so that the inflated shape is close to the desired cure shape. Another useful attribute of the films would be their ability to be thermally welded to ease the manufacturing of spacecraft components.

15.5 14.8 13.5

13.6

13.3 12.3

12.7

12.0

12 10 8 6 4 2 0 1 Ply, Pristine, 1 Ply, Pristine, 1 Ply, Thermal 2 Ply, Pristine, 2 Ply, Pristine, 2 Ply, Thermal 3 Ply, Pristine, 3 Ply, Thermal TP406 UV Cycled 980 TP406 UV Cycled 1000 UV Cycled 1000 times*, UV times*, UV times*, UV

3 Ply, 24hrs @100C, UV

Environmental Exposure

Figure 20. Tensile Modulus Results from Thermal Cycling Test UV Transparent Film Study and Testing The current design of the PowerSphere utilizes an ultra-violet (UV) rigidizable material for both the center columns and the hinges. The material in these structures consists of an UV curing epoxy resin system impregnated into fiberglass fibers. These materials are initially uncured when the spacecraft is launched. Once in orbit, the individual hinges and center columns are inflated using a chemical that sublimates (from a solid to a gas) or vaporizes (from a liquid to a gas). The selection of this inflation material is currently being finalized. Upon deployment, the material is exposed to UV energy from the sun or LED and rigidizes. The UV resin impregnated fiberglass (also known as prepreg) is not impermeable to the inflatant and therefore, an internal gas-retaining layer (or bladder) is required for deployment. Also, because the prepreg is in an uncured state when launched, it could block (stick) to itself

A number of different film materials were chosen to be included in the UV transparent film trade study. These included Mylar, TPX, THV, Barex, and Kynar. A materials properties matrix was constructed to identify the most important information to be gathered about the films. Materials data was gathered on each of these films in order to understand their properties and make selections for future use in the PowerSphere. Mylar is a polyester film from Dupont that has been successfully used in past ILC UV curable isogrid boom manufacturing and curing trials. Mylar has good UV transmission in the wavelengths of interest and it also has good mechanical properties, but it is difficult to thermally weld. TPX is a Polymethylpentene film manufactured by Mitsui Petrochemicals. TPX has very good UV transmission and can be thermally welded, but it is a very brittle film and could be damaged from folding and packing. On the other hand, the THV films from Dyneon are too flexible. Note that THV films are made from a Tetrafluorethylene, hexafluoropropylene, and vinylidene flouride ter polymer. These films can be thermally welded and also have good UV light transmission. However, the THV films have tensile elongation up to 600%, which eliminates them from consideration. The Barex film, which is an acrylonitrile-methyl acrylate copolymer, also has good UV transparency and can be thermally welded, but like TPX, it is very brittle and will be prone to pinholing and damage from tight packing. Kynar is made from PVDF (polyvinylidene flouride). This film has good

9 American Institute of Aeronautics and Astronautics

mechanical and UV transmission properties. It can also be readily thermally welded. After researching all of the different films available, the PVDF films appear to be the best suited for use as the UV transparent film. Out of the available series of Kynar films, four stood out as possibilities: Kynar 740, Kynar 2500, Kynar 2800, and Kynar 2850. The Kynar 2500, 2800, and 2850 were more flexible than the Kynar 740 because of their increasing content of HFP (i.e. they are a copolymer of PVDF and HFP). These three films have tensile break elongation between 200-500%. Kynar 740 is a homopolymer (i.e. it is pure PVDF) and has much more stable mechanical properties. It has a tensile break elongation in the range of 50-200%. Also when purchased without the UV blocking agent, it has a high degree of transparency. To determine the UV transparent properties of some of the candidate film materials, the transmission spectra of a number of different samples were gathered (Figure 17). Because Kynar 740 was the leading candidate, transmission data was also gathered for Kynar 740 films of various thicknesses. As seen from Figure 21, the 1.5 mil thick Kynar has the best transparency over the widest UV wavelength range of any of the films studied. After reviewing the properties of these materials, the Kynar 740 film was selected for use in manufacturing trials of the PowerSphere. Kynar 740 (1.5 mil) was chosen for the UV transparent film layer. This selection was made based on its overall balance of mechanical properties, UV transmittance, and heat sealing ability. Further research into Kynar revealed that it is also stable to the effects of UV, has low outgassing, and has good radiation resistance. 100 90

Transmission (%)

80 70 60 50 Kynar 740 (4 mil) Kynar 2800 (4 mil) Kynar 2500 (3 mil) Mylar LBT-2 (0.92 mil) Barex (1.25 mil) Kynar 740 (1.5 mil) Kynar 740 (2 mil)

40 30 20 10 0 220

270

320

370

420

Wavelength (nm)

Figure 21. UV Transparency of Selected Films

One mil thick VDA (vapor deposited aluminum) coated Mylar MC2 was chosen for the UV reflective layer for the manufacturing trials. This film is over coated on both sides with PVDC in order to allow it to be thermally welded. It is used frequently in the food packaging industry. Another type of film, a VDA coated Polyolefin, is still being evaluated at this time and could be used in future manufacturing trials depending on the strength of its thermal weld. Currently, both Mylar LBT-2 and Kynar 740 are being utilized in the PowerSphere manufacturing trials as the design, manufacturing processes, as their seaming techniques are finalized. Additional work must be performed with the Mylar type MC2 and the Kynar 740 to develop the proper thermal welding profiles that can be used to create strong, reliable seams. Inflation Material Trade Study Given the objective of PowerSphere to be a small, low cost, simple spacecraft, the use of a passive inflation method via a material’ s vapor pressure was researched. All materials have a specific vapor pressure at a given temperature. Relatively small amounts of a given material (in either a solid or liquid state) can be applied to the inside of a tube. Once the external pressure is less than the vapor pressure of the material, the tube will inflate via the material’ s inherent pressure. The material will continue to vaporize until equilibrium is reached. This method does not require a power source for initiation, only a specific temperature to meet the necessary pressure. Nor does it require added spacecraft room as with a compressed gas bottle for example. This can be used for the inflation of both the tubular hinge and the inflatable center column. The inflation material search focused on material compatibility and vapor pressure. The main material interface of the inflation material is with the bladder material, Kynar. The inflation material’ s compatibility with the materials of the entire PowerSphere system was also considered. The maximum allowable pressure of the bladders is limited by their seam strength. Due to the smaller radius of the hinges compared to the center columns, the hinges can withstand greater pressure. Therefore, a different inflation material will be used for the hinge vs. the center column. The minimum pressure needed in the hinge and the center column for inflation and maintaining proper shape before rigidization is 0.5 psi. The maximum pressure/temperature for the hinge is at most 22.5 psi at 80°C. The maximum pressure/temperature for the center column is at most 7.5 psi at 80°C.

10 American Institute of Aeronautics and Astronautics

The recommended choice for the hinge inflation material is the mixture of 3-methylpentane and hexane. It meets the maximum pressure limit of 22.5 psi, has good material compatibility, and is inexpensive. This mixture’ s vapor pressure is equal to the minimum pressure, 0.5 psi, at -11°C. Therefore, the low temperature deployment for the hinges is limited to 11°C. The recommended choice for the center column inflation material is heptane. It is within the maximum pressure limit of 7.5 psi, has good material compatibility, and is inexpensive. Heptane’ s vapor pressure is equal to the minimum pressure, 0.5 psi, at 16°C. Therefore, the low temperature deployment of the center column is limited to 16°C. The method of controlled deployment selected is a staged deployment approach that utilizes the passive inflation material to inflate and deploy the inflatable rigidizable center columns and hinges in two controlled stages. The inflation will start to occur as soon as outside pressure is less than the vaporization pressure of the inflation material. Therefore, both the column and the hinges must be restrained until deployment is ready. The center column will be held in the packed state by fuse wire. An electrical signal to the fuse wire will initiate the deployment of the center column. The hinges will be restrained by a fabric cover. The fabric cover will contain carpenters tape that will retract the cover once the cover tie-down is released; thus allowing the full inflation of the hinges and deployment. Summary and Conclusions Significant progress has been accomplished through the systematic development, design, analysis, and testing of various technologies that make up the multifunctional UV-rigidizable inflatable structures for the PowerSphere application. Both types of structures investigated have been proven to be feasible. A fully assembled engineering unit for the thermal vacuum chamber deployment test is planned for the next phase. Acknowledgment This work was supported by NASA contract NAS301115. References [1] A. Prater, E.J. Simburger, D.A. Smith, P.J. Carian, and J.H. Matsumoto, The Aerospace Corporation, “ Power Management and Distribution Concept for Microsatellites and Nanosatellites,” Proceeding of IECEC, August 1-5, 1999.

[2] D.G. Gilmore, E.J Simburger, M.J. Meshishnek, D.M. Scott, D.A. Smith, A. Prater, J.H. Matsumoto, and M.L. Wasz, The Aerospace Corporation, “ Thermal Design Aspects of the PowerSphere Concept,” Proceedings of Micro/Nano Technology for Space Application Conference, April 11-15, 1999. [3] E.J. Simburger, J.H. Matsumoto, D.A. Hinkley, D.G. Gilmore, T.W. Giants, and J. Ross, “ Multifunctional Structures for PowerSphere Concept,” 42nd AIAA/ASME/ASCE/ASC Structures, Structural Dynamics, and Materials Conference & Exhibit, Seattle, Washington, April 16-19, 2001. [4] E.J. Simburger, J.H. Matsumoto, J.K. Lin, C.Knoll, S. Rawal, A.R. Perry, D.M. Barnett, T.T. Peterson, T.W. Kerslake, and H. Curtis, “ Development of a Multifunctional Inflatable Structure for the PowerSphere Concept,” AIAA-2002-1707 43rd AIAA/ASME/ASCE/ASC Structures, Structural Dynamics, and Materials Conference & Exhibit, Denver, Colorado, April 22-25, 2002. [5] E.J. Simburger, J.H. Matsumoto, T.W. Giants, M. Tueling, J. Ross, J.K. Lin, C. Knoll, S. Rawal, A.R. Perry, C. Marshall, D.M. Barnett, T.T. Peterson, T.W. Kerslake, and H. Curtis, “ Development of Flex Circuit Harness for the PowerSphere Concept,” 29th Photovoltaics Specialist Conference, IEEE, New Orleans, LA, May 20-24, 2002. [6] R.E. Allred, A.E. Hoyt, P.M. McElroy, S.E. Scarborough, D.P. Cadogan, "UV Rigidizable CarbonReinforced Isogrid Inflatable Booms" AIAA-20021202, 43rd AIAA/ASME/ ASCE/AHS/ASC Structures, Structural Dynamics, and Materials Conference and Exhibit AIAA Gossamer Spacecraft Forum, Denver, CO: April 22-25, 2002. [7] J.K. Lin, G.H. Sapna, D.P. Cadogan, S.E. Scarborough, “ Inflatable Rigidizable Isogrid Boom Development,” AIAA-2002-1297, 43rd AIAA/ASME/ ASCE/AHS/ASC Structures, Structural Dynamics, and Materials Conference and Exhibit AIAA Gossamer Spacecraft Forum, Denver, CO: April 22-25, 2002. [8] D.P. Cadogan, and S. E. Scarborough. “ Rigidizable Materials for Use in Gossamer Space Inflatable Structures,” AIAA-2001-1417, 42nd AIAA/ASME/ ASCE/AHS/ASC Structures, Structural Dynamics, and Materials Conference and Exhibit AIAA Gossamer Spacecraft Forum, Seattle, WA: April 16-19, 2001.

11 American Institute of Aeronautics and Astronautics