Development of Composite and Polymer Material CubeSat...

PROJECT REPORT – 9TH. SEMESTER FALL 2012

TMM 4501 Specialization project

Development of Composite and Polymer Material CubeSat Structure

with focus on Materials

Kim Sandvik

NUTS CubeSat in orbit

Norwegian University of Science and Technology Department of Engineering Design and Materials

NUTS – NTNU Test Satellite - Mechanical I - Foreword

II | P a g e


III | P a g e

PRE PROJECT

Collaboration partner:

NUTS - NTNU Student satellite project

Summary:

The project is a study about development of a composite and polymer material CubeSat

structure. The project is derived by NUTS, NTNU Student Satellite Project, and is held by

project manager Roger Birkeland. The project is a resuming of previous work by Kai Inge

Rokstad, and is conducted in collaboration with Christian Nomme. The main goal of the NUTS

project is to launch a complete double CubeSat by 2014.

The framework of the project was to follow the requirements given by the CDS, CubeSat

Design Specification, and to interact and to compromise with the internal project requirements

both given by the main goal and the other participants.

The report covers the development of the secondary structure that assembles all the internal

parts of the satellite.

For the satellite antennas there are carried out FEA-Analyses using Abaqus CAE in order to

dimension the manufacturing, buckling, and the stiffness of the antenna tape springs.

Different types of carbon fiber PrePreg for the frame and carbon fiber and polymer material for

the secondary structure is identified.

Different types of relevant coatings and coating techniques are investigated to deal with frame

wear and low conductivity.

Høgskoleringen 1

7034 Trondheim

Phone: 73 59 50 00

Internet: ntnu.no

Scope of project: Date:

Composite Satellite

Development

20.12. 2012

Pages preface:

Pages report:

012

032

Attachments: 044

Student: Supervisor:

Kim Sandvik

Nils Petter Vedvik


IV | P a g e


V | P a g e

I. Foreword

The project is a study about different aspects of development of a composite and polymer

material CubeSat structure. The project is derived by NUTS, NTNU Student Satellite Project,

and is held by project manager Roger Birkeland. The project is a resuming of previous work

by Kai Inge Rokstad, and is conducted in collaboration with Christian Nomme.

The project is a part of the final year of a 2-years master’s degree at Department of

Engineering Design and Material. The work counts 15 of 30 study points during 9th

semester,

fall 2012. The project can either act as a pilot study for the thesis, or as an independent

project. In parallel with the project, two specialization subjects of 3,75 point each was done.

The subjects were “Composite structures”, and “Product simulation”.

The framework of the project was to follow the requirements given by the CDS, CubeSat

Design Specification, and to interact and to compromise with the internal project requirements

both given by the main goal and the other participants.

Trondheim 20 /12 2012

Kim Sandvik


VI | P a g e

We would like to extend our utmost appreciation to:

Associates Professor Nils Petter Vedvik

- Extended guidance throughout the project

NUTS Project Manager Roger Birkeland

- Supplier of technical information and guidance throughout the project

Tore Landsem

- Contact person for antenna tape spring material, and prototyping

NUTS – NTNU Test Satellite - Mechanical II - Task

VII | P a g e

II. Task

NORGES TEKNISK-

NATURVITENSKAPELIGE UNIVERSITET

INSTITUTT FOR PRODUKTUTVIKLING

OG MATERIALER

PROSJEKTOPPGAVE HØST 2012

FOR

STUD.TECHN. KIM SANDVIK

NTNU Test Satellite

NUTS (NTNU Test Satellite) er ett tverrfaglig prosjekt-samarbeid ved NTNU, og skal etter

planen være klar til oppskytning i 2014. Ved IPM er det identifisert flere fokusområder

knyttet til mekaniske system på satellitten, deriblant utvikling av komposittrammen,

mekanismer for solcellepanel og antenner, dynamisk simulering, fullskala dynamisk testing,

tribologi, slitasje og nedbrytning, kvalifikasjon i henold til eksistrende regelverk samt

overordnet PLM for prosjektet.

Som en del av oppgaven skal det utarbeides en plan hvor det fokuseres på ett eller flere av de

relevante tema som er oppgitt. Dette gjøres i samarbeid med faglærer og prosjektleder i løpet

av de to første ukene av prosjektet.

Ved bedømmelsen legges det vekt på at problemstillingen presenteres klart, at besvarelsen er

skikkelig gjennomarbeidet og at kandidaten gir en selvstendig framstilling av stoffet med

egne vurderinger.

NUTS – NTNU Test Satellite - Mechanical II - Task

VIII | P a g e

Besvarelsen skal ha med oppgavetekst og skal forsynes med innholdsfortegnelse. I forordet

skal det stå hvilke fordypningsemner kandidaten tar. Rapporten innledes med en klar

formulering av problemstillinger bearbeidet i prosjektet, et sammendrag av viktige resultater,

og konklusjoner. Rapporten skal være på maksimum 30 sider, inklusive skisser innarbeidet i

tekst. Eventuelle tabeller, tegninger, detaljerte skisser, fotografier, med videre, kan medtas i et

bilag som regnes i tillegg til de 30 sider. I besvarelsen henvises til de respektive steder i

vedleggene, men besvarelsen skal skrives slik at den kan leses uten vedlegg.

Figurer og tabeller skal inneholde alle nødvendige påskrifter. Litteraturhenvisninger skal være

fullstendige med angivelse av forfatter, bok (artikkel), tittel, forlag, årstall og sidenummer.

Henvisninger foretas ved nummer i teksten og dette refererer til en nummerert litteraturliste

bak i rapporten.

Tre (3) uker etter utlevering av prosjektoppgaven leverer kandidaten et A3-ark med tekst og

bilder som beskriver hva oppgaven går ut på (en papirversjon og et elektronisk eksemplar i

pdf-format). Mal for arket finnes på instituttets hjemmeside på siden for ”prosjekt og

fordypningsemner” (http://www.ntnu.no/ipm/prosjekt).

Senest 1 uke før innlevering av prosjektoppgaven skal kandidaten levere et A3-ark som

illustrerer resultatet av arbeidet (en papirversjon og et elektronisk eksemplar).

Prosjektarbeidene presenteres som muntlige foredrag 19. oktober. Det er obligatorisk

frammøte for alle prosjektkandidater under foredragene.

Innleveringsfrist for prosjektbesvarelsen er 20. desember. Besvarelsen leveres i to

papirversjoner og elektronisk på CD eller DVD.

Kontaktpersoner:

Ved instituttet: Nils Petter Vedvik

Prosjektleder: Roger Birkeland, IET

Nils Petter Vedvik

Faglærer

NUTS – NTNU Test Satellite - Mechanical III - Table of contents

IX | P a g e

III. TABLE OF CONTENTS

I. FOREWORD V

II. TASK VII

III. TABLE OF CONTENTS IX

IV. DIAGRAM LIST XII

1 INTRODUCTION 1

2 BACKGROUND INFORMATION 2

2.1 PART DESCRIPTION 2

2.1.1 DOUBLE CUBE 2

2.1.2 P-POD 2

2.1.3 IR-CAMERA FOR SCIENTIFIC PAYLOAD 2

2.1.4 ADCS SYSTEM 2

2.1.5 EPS -POWER SUPPLY 2

2.1.6 BATTERY PACK 3

2.1.7 BACKPLANE 3

2.1.8 FRAME 3

2.1.9 SECONDARY STRUCTURE 3

2.1.10 ANTENNAS 3

2.1.11 TRANSCEIVERS 3

3 METHOD 4

3.1 PROCESS OF WORKING 4

4 SECONDARY STRUCTURE 5

4.1 DESCRIPTION 5

4.2 DEMANDS 5

4.3 CONCEPT 5


X | P a g e

4.4 MONTAGE OF CAMERA/BATTERY-PACK 6

4.5 FASTENINGS/SCREWS 7

4.5.1 SCREWS 7

4.5.2 SPACERS 7

4.5.3 INSERTS 7

4.5.4 LOCKTITE 7

4.5.5 GLUE 7

5 ANTENNA SYSTEM 8

5.1 DESCRIPTION 8

5.2 DEMANDS 8

5.3 CONCEPT 8

5.3.1 DESIGN 8

5.3.2 MONTAGE 9

5.3.3 DEPLOYMENT 9

5.4 FEM-ANALYSIS 10

5.4.1 TAPE SPRING FORMING ANALYSIS -MANUFACTURING 10

5.4.2 TAPE SPRING DEPLOYMENT ANALYSIS 14

5.4.3 TAPE SPRING STIFFNESS 16

5.5 PRODUCTION OF TAPE SPRING ANTENNA 17

5.6 PHYSICAL BENDING TEST 17

6 MATERIAL 18

6.1 FRAME 18

6.1.1 DEMANDS 18

6.1.2 COMPOSITES 19

6.1.3 SURFACE TREATMENT 21

6.1.4 MATERIAL TESTING 25

6.2 SECONDARY STRUCTURE 27

6.2.1 DEMANDS 27

6.2.2 COMPOSITE 27

6.2.3 POLYMER - SOLID AND FOAM 27

6.2.4 COMPARISON OVERVIEW BETWEEN RELEVANT MATERIALS 28

6.3 ANTENNA TAPE SPRING 29

7 SATELLITE PROTOTYPE 31


XI | P a g e

8 CONCLUSIONS 32

9 LITERATURE REFERENCES 33

10 ATTACHMENTS 34

10.1 STRUCTURE OF WORK 35

10.2 DIVIDING OF SUBJECTS 36

10.4 PART DRAWINGS 37

10.4.1 SATELLITE ASSEMBLY 37

10.4.2 ANTENNA ASSEMBLY 38

10.4.3 ANTENNA PCB CARDS 39

10.4.4 SECONDARY STRUCTURE – POLYMER 40

10.4.5 SECONDARY STRUCTURE – CARBON FIBER 41

10.4.6 CAMERA MODULE, MAXIMUM DIMENSIONS 42


10.5 TAPE SPRING MANUFACTURING TOOL DRAWING, HE 44

10.6 TAPE SPRING MANUFACTURING TOOL DRAWING, SHE 45

10.7 ANTENNA ASSEMBLY ILLUSTRATIONS 46

10.8 SECONDARY STRUCTURE ILLUSTRATIONS 51

10.9 APPENDIX 1 – BERYLLIUM COPPER DATASHEET – HARALD PIHL AB 52

10.10 PREPREG MATERIAL DATA 53

10.11 TAPE SPRING MANUFACTURING – RESULTS 55

10.12 BUCKLING AND DEPLOYING OF ANTENNA TAPE SPRING – RESULTS 63

10.13 STRENGTH OF ANTENNA TAPE SPRING – RESULTS 67

10.14 FE ANALYSIS OF ANTENNA TAPE SPRING IN ABAQUS 68

10.14.1 MANUFACTURING OF ANTENNA TAPE SPRING 68

10.14.2 BUCKLING AND DEPLOYING OF ANTENNA TAPE SPRING 74

10.14.3 STRENGTH OF ANTENNA TAPE SPRING 78

NUTS – NTNU Test Satellite - Mechanical IV - Diagram list

XII | P a g e

IV. DIAGRAM LIST

FIGURE 1 - SECONDARY STRUCTURE ............................................................................................. 5

FIGURE 2 - SECONDARY STRUCTURE – MILLED GROOVE ................................................................ 6

FIGURE 3 - SECONDARY STRUCTURE - SLOTS ................................................................................. 6

FIGURE 4 – BATTERY PACK WITH SCREWS ..................................................................................... 6

FIGURE 5 - INSERTS ....................................................................................................................... 7

FIGURE 6 - ANTENNA ASSEMBLY ................................................................................................... 8

FIGURE 7 - TAPE SPRING FOLDING PIECES...................................................................................... 8

FIGURE 8 - ANTENNA ASSEMBLY EXPLODED VIEW ......................................................................... 9

FIGURE 9 – TAPE SPRING TOOL ................................................................................................... 10

FIGURE 10 - TAPE SPRING TOOLS MOUNTED TOGETHER .............................................................. 11

FIGURE 11 - TAPE SPRING MODEL ............................................................................................... 11

FIGURE 12 – MANUFACTURE, INITIAL STEP .................................................................................. 12

FIGURE 13 - MANUFACTURE, STEP 1, VON MISES ........................................................................ 12

FIGURE 14 - MANUFACTURE, STEP 2, VON MISES ........................................................................ 12

FIGURE 15 - TAPE SPRING ROLLING STRESSES .............................................................................. 13

FIGURE 16 - TAPE SPRING MANUFACTURE, VON MISES CROSS SECTION ...................................... 13

FIGURE 17 - TAPE SPRING MANUFACTURE, PEEQ CROSS SECTION ................................................ 14

FIGURE 18 – MANUFACTURED TAPE SPRING AFTER SPRING BACK................................................ 14

FIGURE 19 - TAPE SPRING BUCKLING ANALYSIS ASSEMBLY .......................................................... 14

FIGURE 20 - TAPE SPRING BUCKLING, VON MISES ........................................................................ 15

FIGURE 21 - TAPE SPRING BUCKLING, PEEQ ................................................................................. 16

FIGURE 22 - TAPE SPRING STIFFNESS ON EARTH .......................................................................... 16

FIGURE 23 - PREPREG PRODUCTION PROCESS - OWN DRAWN MODEL ......................................... 20

FIGURE 24 - PREPREG LAYUP - [90; +45; 0; 0; -45; 90] - OWN DRAWN MODEL .............................. 21

FIGURE 25 - ELECTROLYTIC PLATING - OWN DRAWN MODEL ....................................................... 23

FIGURE 26 - HTTP://WWW.SPACE.T.U-

TOKYO.AC.JP/CUBESAT/MISSION/DEV/STR/IMG/VIBRATION-S.JPG ............................................ 25

FIGURE 27 – PHASE DIAGRAM FOR BERYLLIUM COPPER, REF; [7] ................................................. 29

FIGURE 28 – ALLOY “25 1/2H” - YIELD STRENGTH/TIME/TEMPERATURE FOR AGE HARDENING, REF;

[7] .............................................................................................................................................. 30

NUTS – NTNU Test Satellite - Mechanical 1 - Introduction

1 | P a g e

1 Introduction

NUTS, NTNU Test Satellite, is a project at NTNU to develop and launch a double CubeSat.

The project is managed by Roger Birkland, who wrote a concept study of developing a

CubeSat at NTNU as his master thesis in 2007. The project participants has since then been

final year students writing both for their pre-project and their master degrees. In the later

years the project has also gained a larger group of voluntary workers. The workforce is a mix

of dedicated people from departments all over NTNU. There are students from electronics,

communications, space technology, physics, cybernetics, computer science and mechanics.

Reference [1].

The main goal is to launch a complete double CubeSat by 2014. At this time the main payload

(load of value) for the satellite is an infrared camera. It is supposed to take pictures of gravity

waves in the upper atmosphere. Gravity waves are waves driving the large scale flows in the

middle atmosphere. They are created by the varying terrain of the earth and the effect of

weather phenomena. The waves are not yet greatly understood. It is possible to observe the

waves from the earth using a telescope. The problem is that there are many atmospheric air

layers between the observatory on earth and gravity waves. In this context they will act as a

filter, and much of the information is lost. It is therefore preferred to obtain the information

with reference above the gravity waves. At this point none of the CubeSat that are sent into

orbit has had an IR-camera as payload. Reference: [1] /payload.

The Department of Engineering Design and Material has the primary responsibility for the

structural part of satellite. In 2011/2012 Kai Inge Rokstad completed his master degree

project for NUTS. His work mainly consisted of developing a concept of using carbon

composite as structural material for the satellite. In cooperation with student Christian

Nomme the project this semester has been to develop and determine a final concept to be

used.

The structure of work is divided using a task tree, se appendix 10.1. It is divided into the main

topics that are conceptualization, analysis, experimental, and production. Different tasks for

primary-, secondary-, tertiary- structure and prototyping has been divided between each other

of us. Christian has had a strong presence on the structural analysis, while I have worked

more with the aspect of material choice and behavior, see appendix10.2.

During this study I have examined different materials for the frame and the secondary

structure. Furthermore I have worked with the concept of the secondary structure and the two

satellite antennas. At last different coating possibilities for the structure is identified.

NUTS – NTNU Test Satellite - Mechanical 2 - Background information

2 | P a g e

2 Background information

2.1 Part description

Following there is a short list that describes the most important parts of the satellite that is

referred to during the report. Reference [1]

2.1.1 Double cube

The satellite is a double CubeSat. Each module is 100x100x113,5 mm. Totally the satellite is

100x100x227 mm (w/l/h).

2.1.2 P-POD

The satellite is hitchhiking with a cargo rocket containing multiple satellites. On the cargo

rocket a P-POD is mounted. Each P-POD contains totally three CubeSats. In other words, it

could contain 3x1 cubes, 1x3 cube, or 1x2 cube + 1x1 cube.

Mainly the P-POD is a rectangular box where the satellite is resting on sliding rails at the P-

POD’s four sides. On the bottom there is a launch spring, and at the top there is a lid that is

opened when the satellites is to be launched into the satellite path.

2.1.3 IR-Camera for scientific payload

The main payload at this point is the IR-camera. The camera is supposed to observe gravity

waves in the upper atmosphere. Until now there are no CubeSats made to observe these

waves. It is possible to observe the waves from the earth, but the air layers in the atmosphere

work as a filter and will degrade the information.

2.1.4 ADCS System

The ADCS, Attitude Determination and Control System, is the system determining the

position of the satellite and will position the desired orientation relatively to the earth.

To determine the orientation of the satellite there is used sun sensors. These sensors will

determine the orientation according to the sun. As long the as the position of the satellite is

known, the orientation relatively to the earth may be calculated.

The attitude control consists of coils of wire at three of the satellite sides. When the coils are

activated with power the magnetic field of the coil will interact with the magnetic field of the

earth. Then you will be able to control the orientation at the xyz-axes.

2.1.5 EPS -Power supply

The EPS consists of the solar cells, power management system, and the batteries.

There are five solar cells modules on the satellite. One consists of two cells, while four

modules consist of four cells each. Totally the satellite is equipped with 18 cells.

NUTS – NTNU Test Satellite - Mechanical 2 - Background information

3 | P a g e

The batteries are the power bank in the system, and will be consumed when the satellite is in

the shade side of the earth, or if the satellite needs excessive power consumption.

The power management system connects the solar cells and the batteries to the electrical

system, and will control the charge cycles.

2.1.6 Battery pack

The satellite is equipped with two battery packs with four batteries each.

2.1.7 Backplane

All the electrical cards in the satellite are connected to the backplane through slots. The

backplane is like the motherboard in an ordinary computer, and is linking the cards together.

Basically it is the communication path in the satellite.

2.1.8 Frame

The frame is the structure in the satellite that is designed to provide the desired stiffness of the

construction. All loads the satellite is exposed to, both external and internal, is addressed the

frame.

It is also the part that is designed to fit the P-POD, where the frame rails are going to slide

along the P-POD rails.

2.1.9 Secondary structure

The secondary structure is the part connecting the internal parts (tertiary structure) with the

frame. It will be exposed to some loads, but should not need to provide excessive stiffness to

the structure.

The secondary structure is in direct contact with the frame, antennas, backplane, PCB cards,

battery pack, camera, and ADCS.

2.1.10 Antennas

The satellite is equipped with two transceivers with separate antennas. One transceiver has an

antenna at 145MHz, and the other has a 437MHz antenna. Together the pair of transceivers

and antennas is connected to the Modem/Microcontroller and constitutes the communication

system.

2.1.11 Transceivers

Two transceivers processes and receives signals from the antennas.

NUTS – NTNU Test Satellite - Mechanical 3 - Method

4 | P a g e

3 Method

3.1 Process of working

The NUTS project is a long running project including people with different background

during a short time interval. This means that the project method changes continuously with

new people. A good solution at one stage of the project can be poor in a later stage with

completely different personnel.

Since we are in an early stage of the project I have tried to build the report flexible. Instead of

present one correct solution, I have presented several different possibilities. All have some

positive aspects and some downsides. The different options are listed with the assumptions for

the suggestion. They also have an introduction to the process, or way of thinking, to ensure

that the reader should be able to make their own judgment on the basis of the report. The

report should be able to be used actively by the new assessments later in the project without

too much background information and research.

During the development of the project we have been addressed to several of concepts directly

or indirectly related to our tasks. They are highly important to understand the contexts in the

big picture. Much of this is well presented by the report of Christian Nomme. Due to limited

space in this report, I choose to restrict reproduction and instead refer to his report for further

details.

NUTS – NTNU Test Satellite - Mechanical 4 - Secondary structure

5 | P a g e

4 Secondary structure

4.1 Description

The secondary structure is the part connecting the internal parts (tertiary structure) with the

frame. It is in direct contact with the frame, antennas, backplane, PCB cards, battery pack,

camera, and ADCS.

4.2 Demands

- Light

- Even mass distribution

- Easy assembly

- Transfer loads evenly to the frame

- Two possibilities

o The frame provides the satellite stiffness

o The secondary structure provide the axial stiffness, and the frame the

diagonally stiffness

- Must assemble the:

o Release springs and switch

o Backplane and modules

o Camera

o Battery pack

o Stiffeners

o ADCS

4.3 Concept

The secondary structure consists of four columns, one in

each corner inside of the composite frame. The one

illustrated in this section is designed to be made of

polymer material.

The end of the columns lies against the other satellite in

the P-POD. For mounting of the release springs and switch

it is drilled four holes in the ends. Both the switches and

release springs are screwed into the holes. The cable for

the switch is drawn further trough and connected to the

PCB-module.

Inside the secondary structure there are several slots. They are for supporting the PCB

modules. At two of the columns it is a long axially slot to keep the backplane in place.

The ADCS (The Attitude Determination and Control System consisting of three copper coils)

is situated between the secondary structure and the frame. It is milled a groove along the

columns to keep the ADCS in place.

Figure 1 - Secondary structure


6 | P a g e

The photovoltaic system, antenna, stiffener and the ADCS for the top of the satellite is bolted

together as a sandwich. The complete module is then bolted to the satellite. This is done

through eight ears on the stiffener. The ears are squeezed between the secondary structure and

the frame with eight screws, the same screws mounting the top and the bottom of the

photovoltaic system at the sides of the satellite.

4.4 Montage of camera/battery-pack

There are mainly two concepts for mounting of the camera and the battery pack. Which one to

choose is dependent on the final weight of the modules. The weight is not yet known, so it

would be wise to not select one of the concepts at present time.

If the modules are lightweight they may be mounted with the

same concept as the PCB-modules. The camera will have two

slots, one at the top and one at the bottom. The battery pack

could have one in the middle, or one at the top and one at the

bottom. The montage of the satellite will with this concept be

easy.

If the modules are heavyweight they should be screwed in

place directly to the frame. This is done with the same screws

mounting the satellite together. The only difference is that the

nut is mounted on the battery pack and the camera module

instead of the secondary structure. To make the mounting of

the satellite easy the secondary structure is equipped with

supporting structure for the modules.

Figure 3 - Secondary structure - slots

Module

Slots

Backplane

Slots

Figure 2 - Secondary structure – milled groove

Groove for ADCS

Groove for Stiffener

Switch/release spring

Figure 4 – Battery pack with screws


7 | P a g e

4.5 Fastenings/screws

4.5.1 Screws

The satellite is mounted together with 40 pcs screws going through the composite frame into

the secondary structure. In addition to this there are 2x8 pcs for the assembly of the antennas,

and 2x8 pcs for assembly of the photovoltaic system, antenna, stiffener and the ADCS

sandwich.

Different screws materials are described in the report by Christian Nomme.

4.5.2 Spacers

To make sure the composite material of the frame is not squeezed there are added spacers in

the frame holes before placing the photovoltaic modules.

4.5.3 Inserts

In the secondary structure there are mounted inserts to

make sure the screws will have properly fixing. In this

way ordinary machine screws can be used. This will

result in good accuracy during montage.

4.5.4 Locktite

If the threads of the screws and/or the inserts/nuts are clean it is a risk that the screws will be

shaken loose because of the vibration during launch. To lock and secure the screws Locktite

could be used. It is a liquid compound to apply the threads that will solidify after montage. It

has many different products. For example has one the possibility for disassembly and another

is permanent and cannot be loosened.

4.5.5 Glue

Glue is another possibility instead of Locktite. This will be permanent, and could only be

applied on the final satellite before launch. The advantage of using glue is it will remove any

slack that may occur in the satellite.

Figure 5 - Inserts

NUTS – NTNU Test Satellite - Mechanical 5 - Antenna system

8 | P a g e

5 Antenna system

5.1 Description

The satellite is equipped with two antennas. The antenna situated at the top of the satellite is a

VHF antenna, and the one at the bottom is a UHF antenna.

Both concepts are flat turnstile antennas (cross dipoles). It consists of four tape springs, each

laid in different direction in the plane.

The antenna is designed according to electrical requirements submitted by Jens Abraham.

5.2 Demands

- Below 3 mm thick tape springs

- Hole for camera lens at 35 mm

- 3 mm free space between PCB cards

- Holes for mounting screws lined up with the stiffener, ADCS and the solar module.

- Proper space for electrical components

- Material of tape springs: Beryllium Copper

5.3 Concept

5.3.1 Design

The antenna is designed as a sandwich structure of

PCB-cards. For detailed view, see appendix 10.7.

In the center of the antenna sandwich there is a

camera lens hole at 35 mm that limits the space

for the electrical components.

The tape spring folding pieces has multiple tasks. When the tape

springs exit the antenna module it has to be curved to preserve its

stiffness. For that reason the curvature of the piece at the exit end

has the ordinary curvature of the antenna tape springs.

The end of the tape spring antennas has to have proper electrical

contact to the PCB card at the montage end of the antenna. The

top folding piece is therefore phased out at the same time as the

curvature is straightened out. To compensate the bottom folding

piece curvature is also straightened and the thickness is increased.

When the antennas are folded around the satellite a kink will cause a radius for the tape

spring. It is important that there will not be any mechanical damage to the tape spring that

prevents the deployment to work properly. At that reason the bottom tape spring folding piece

Figure 7 - Tape spring folding pieces

Figure 6 - Antenna assembly


9 | P a g e

has a radius at the edge. This will assist in keeping the buckling radius of the tape spring

properly during montage and launch.

The screw holes for the tape springs are situated near the center of the antenna module to get

clear from the stiffener plate and the ADCS. To make sure to keep proper tension the

curvature of the folding pieces is changed first below the first mounting screw.

There are spacers between the PCB cards for the mounting holes for the assembly to keep the

distance constant and the structure under tension.

There are four antenna contacts each placed in each corner of the bottom PCB card. It is

possible, and would have been preferred, to have one contact and lead a copper path to each

of the tape spring antennas. The reason it is split into four is the risk of magnetic field

interference between the copper path on the PCB and the different tape spring antennas. This

could make the antennas inoperable.

5.3.2 Montage

The antenna is assembled in the following order:

1) The screws for the tape springs are fitted in

the tapered holes of the top PCB plate

2) The top PCB plate is placed upside down at

the table. The four tape spring top folding

pieces are mounted

3) The four tape spring antennas are mounted

4) The four tape spring bottom folding pieces are mounted

5) The spacers for the mounting holes at the bottom PCB plate is ether soldered or glued

to the plate.

6) The bottom PCB plate is threaded on the antenna screws, and the nuts are tightened.

The top assembly consists of the solar cell modules, the antenna, the stiffener, and the ADCS.

There are eight montage screws that hold these parts together. The module is then montaged

at the top of the secondary structure with eight screws holding the frame stiffener.

5.3.3 Deployment

The tape spring antennas are wrapped around the satellite in the following order:

1) Each of the bottom antenna springs are folded along the length and the width of the

satellite.

2) The first of the top antenna spring are folded around four corners of the satellite

3) Repeat 2) sequentially with the tree other top tape springs.

The deployment of the tape springs is executed automatically after the launch of the satellite.

The tape springs are held around the satellite with thin nylon thread. When the satellite is in

orbit, the satellite will apply current to a NiCr wire that will burn over the nylon tread. The

nylon thread loses its tension, and the antennas will deploy.

Figure 8 - Antenna assembly exploded view


10 | P a g e

5.4 FEM-Analysis

The tape springs are made of Beryllium Copper. The tape springs cannot be purchased

finished, so they have to be produced at the NTNU.

There are three aspects to evaluate at this stage of designing the tape springs. It has to be

possible to form the curvature of the tape springs plastically. Secondly it has to be able to

recover its original shape when deployed. At last it should have the required stiffness for the

application.

Abaqus CAE is used for the FEM-Analyses. I have never used, nor had any experience with

Abaqus in advance. At that reason I will concentrate on developing the method for the

analyses. The methods are carefully described step by step in the appendix 10.14 (10.14.1,

10.14.2, and 10.14.3). The idea is that if parameters like length, geometry or material of the

antenna is changed, new analyses should be made based on the manual from the appendix.

5.4.1 Tape spring forming analysis -Manufacturing

To form the tape spring curvature the material has to be formed plastically. The best way to

get the best stress distribution for the tape spring will be pressing the metal plate in a form

made for the whole length of the antenna. Since this will be a small scale production this will

probably be uneconomical. This analysis will therefore be based on that the metal plate is fed

between two roller wheels that form the curvature. To make the analysis as realistic as

possible, it was desirable to model the exact manufacturing process with these tools.

The tools

The process has two forming tools, see Figure 9 – Tape Spring tool. In the analysis only ¼ is

modeled. The radius of the curvature of the “She” press is 4,15 mm, while the radius of the

“He” press is 4,00 mm. This means there will be a gap of 0.15 mm between the tools when

mounted together. The tape spring material is 0,1 mm thick, so there will be an excessive gap

at 0,05 mm. This is illustrated as Figure 10 - Tape Spring tools mounted together. The gap is

to ensure there is no squeezing of the metal plate (antenna) when rolling. The tools have a flat

edge at both side of the curvature to ensure proper distance. It is not included in the model,

but there should also be an edge at the “she” press that provides proper alignment relative to

each other.

Figure 9 – Tape Spring tool


11 | P a g e

Figure 10 - Tape Spring tools mounted together

It is assumed that the tool material is much stiffer than the antenna material. At that reason the

tools is modelled as solid 3D discreate rigid. They have to be meshed, but Abaqus wont

calculate the stresses in the parts, since they are not important in this context. Before meshing

they have to be converted to a shell element. This has to be done because Abaqus don’t

support 3D discreate rigid elements yet. As a ruule of thoumb, the mesh should be smaller

than the material formed.

Tape Spring

The tape spring material is modeled as solid 3D deformable. It is in this analysis 100x8x0,1

mm (length/width/thickness). It is meshed as C3D8R, an 8-node element. The tape spring has

16 element along the width, and 4 along the thickness, see Figure 11 - Tape spring model.

Figure 11 - Tape spring model

The analysis method

The analysis is performed in explicit mode. It has been many attempts in standard, but there

are many challenges to overcome to succeed. One of them is micro slip, where the friction

between the plate and the tool will alternately slip during step 1. Another is that there are

many points of contact during the analysis. During the press stage 1 the tape spring will be in

contact with the tools in tree points. When using explicit you can define all the surfaces in the

model as contact surfaces. At that way every surface will seek contact without any issues.


12 | P a g e

Step1

The first step could be a simplification of the manufacturing. If possible it would in reality be

preferred to feed the tape spring plate into the rolling tools. In the analysis the tape spring

plate is laid into the grove of the “She” tool. Step 1 is the step where the two tool pieces is

brought together enclosing the spring plate.

Figure 12 – Manufacture, Initial step

Figure 13 - Manufacture, Step 1, Von Mises

Step2

Step 2 is the step where the tape spring is rolled. The tools are both rotated 90 degrees (Pi/2

rad). The friction applied between the tools and the metal plate pulls the plate through.

Figure 14 - Manufacture, Step 2, Von Mises


13 | P a g e

Results

The analysis only analyzes a part of the manufacture process of the tape spring. The Figure 15

- Tape spring rolling stresses illustrates the stresses developing in the tape spring during the

plastically deformations. To check the real cross section of importance there is taken out a

section of the plate in the evenly balanced area.

Figure 15 - Tape spring rolling stresses

As illustrated in Figure 16 - Tape spring manufacture, Von Mises cross section the residual

stresses in the cross section is not above 250 MPa. More important the stresses at the edge is

only 6,3 – 20 MPa. The edge is the most vulnerable area during the buckling test, and should

be as low as possible. There are low stresses at the edge because of the excessive gap between

the tape spring and the tool, and then off course because there is no material left to pull the

edge.

Figure 16 - Tape spring manufacture, Von Mises cross section

To check the plastic deformation an PEEQ plot (Plastic Elongation EQuivalent) is illustrated

in Figure 17 - Tape spring manufacture, PEEQ cross section. In the midle there is no plastic

elongation, where the gradient increases towards the edge. This means that the plate is

subjected to bending moment. This is a good verification of the manufacture prosess, where

we only want to bend it and not expose the plate to unnecessary forces or stresses.


14 | P a g e

Figure 17 - Tape spring manufacture, PEEQ cross section

At this refinement the analyses take 49 hours to complete. It was therefore not time for

perfecting the geometry further this semester.

The geometry of the manufactured plate after spring back (geometry in equilibrium) is

illustrated in Figure 18 – Manufactured Tape spring after spring back.

Figure 18 – Manufactured Tape spring after spring back

5.4.2 Tape spring deployment analysis

If the tape spring shall perfectly recover its original shape after deployment, the stresses have

to be below the material yield strength.

The Figure 18 – Manufactured Tape spring after spring back is used as basis for the

deployment analysis. The model consists of two solid 3D parts, illustrated as Figure 19 - Tape

Spring buckling analysis assembly.

Figure 19 - Tape Spring buckling analysis assembly


15 | P a g e

The tape spring is meshed as a C3D8R, 8-node element. It has 15 elements along the

curvature, and 4 elements along the thickness. The tape spring is 100 mm long with the

geometry cross section as manufactured in Chapter 5.4.1, Tape spring forming analysis -

Manufacturing.

The second part, the corner at the satellite, is modeled as discrete rigid. It has a radius at 2 mm

and is 20 mm long.

The analysis is divided into 4 bending steps, and 4 unloading steps. The corner is moved from

its initial position with a radius of 50 mm around the origin to a degree of 45, and back again.

The end of the corner located in origin is fixed against movement in x, y, and z direction, and

rotation around y and z axis. The opposite end is fixed for rotation around y and z axis.

As illustrated in Figure 20 - Tape spring buckling, Von Mises the stresses during buckling to

90 degrees do not exceed the yield strength. The Von Mises stresses are 1075 MPa at its

maximum, where the yield strength for the material is 1169-1173. It has to me mentioned that

it is a little close, especially with respect to possible residual stresses after manufacturing.

Figure 20 - Tape spring buckling, Von Mises


16 | P a g e

When looking at Figure 21 - Tape spring buckling, PEEQ it is no plastic elongation at all

during buckling. That indicates the tape spring will resume its shape after it is wrapped

around the satellite.

Figure 21 - Tape spring buckling, PEEQ

5.4.3 Tape spring stiffness

The reason for making a curvature for the antenna is to straighten it out after launch. Its

ability to deploy will depend on its geometrical stiffness.

Initially the gravity is almost the same at the earth as the orbit. The difference is that there are

noting that stops the satellite for falling, it is only “missing” the earth. Practically, this means

that there are no gravity forces acting on the tape spring antennas that can make it buckle.

To check its stiffness the gravity effect at earth is

checked anyway. It could also be an easy way to check

the validity of the analyses in a physical test. The antenna

is modeled exactly as the Buckling and deploying of

Antenna Tape Spring analysis, only in full length. The

longest antenna is used as a basis, and it is 700 mm long.

The tape spring is only fixed in the end situated in origin.

The only force introduced in the analysis is the gravity in

–y direction. As illustrated in Figure 22 - Tape Spring

stiffness on earth the displacement in the free end is 174

mm in –y direction. It is a relative high displacement (25% of its length), but the analysis also

imply that it actually will hold without buckling when interacting with the gravity at earth.

The engineering evaluation involves how stiff the antenna actually has to be to function

properly.

What remains to be analyzed is how the antenna responds after deployment. It is important to

know that when deployed it will get an oscillation frequency. How this frequency is damped

after deployment has to be verified to ensure that the antenna reaches equilibrium without

break.

Figure 22 - Tape Spring stiffness on earth


17 | P a g e

5.5 Production of tape spring antenna

The tape spring has to be made at NTNU. The material is shipped as a plate with the

dimensions 211x800x0,1 mm.

The first step is to cut the plate in desired length and width. According to the tape spring

deployment analysis the width has to be 8 mm. The length is respectively 4x700 and 4x300

mm. The plate is cut with a plate scissor.

The second step is to form the plate with the tool designed for this application. The tool

consists of six wheels. The main forming wheels have the curvature found necessary in the

tape spring forming analysis, and the manufacturing drawings are attached as 10.5 and 10.6.

The upper wheel has a positive curvature, while the lower wheel has a negative curvature. The

two wheels in front and behind the main forming wheels is steering wheels. They are mounted

to prevent the tape spring to slide sideways.

To lower the forces acting at the tape spring folding piece the spring forming should stop 10

mm from the edge of the tape spring. In this way the tape spring will lay naturally flat on the

PCB surface after mounting.

The tape springs has to be heat treated after the forming to regain their original properties. For

further description refer to chapter 6.3.

5.6 Physical bending test

The finished tape spring should be tested physically to verify the analyzed parameters. It is

important to control the bending radius to be sure that it is designed correct to the antenna

module concept. Another test is to find the actual stiffness of the tape spring when properly

deployed. Finally it could be useful to check how vulnerable it is against buckling without

permanent (plastically) deformation.

NUTS – NTNU Test Satellite - Mechanical 6 - Material

18 | P a g e

6 Material

6.1 Frame

6.1.1 Demands

There are both structural and environmental properties of importance when selecting a proper

material for the frame. Some are to satisfy the launch party, and some are internal project

properties.

CubeSat design specification, The CubeSat Program, Cal Poly SLO:

- 2.1.7.1 Total Mass Loss (TML) shall be ≤1.0%

- 2.1.7.2 Collected Volatile Condensable Material (CVCM) shall be ≤0.1%

- 2.2.9 The rail shall not have a surface roughness greater than 1.6 µm

- 2.2.19 Aluminum 7075 or 6061 shall be used for both the main CubeSat

structure and the rails. If other materials are used the developer shall

submit a DAR and adhere to the waiver process

- 2.2.20 The CubeSat rails and standoff, which contact the P-POD rails and

adjacent CubeSat standoffs, shall be hard anodized aluminum to prevent

any cold welding within the P-POD

NUTS project specifications of importance:

- Sufficient stiffness

- Wear against P-POD equal or less relative to aluminum against aluminum.

- Impact strength (Play between P-POD and Satellite)

- Avoid potential accumulation induced from electromagnetic radiation

6.1.1.1 Avoid potential accumulation

In orbit the satellite is exposed for electromagnetic radiation. There is a fear the radiation will

induce potential accumulation in electrical components in the satellite. To avoid this challenge

the frame should be electrical conductive. At this point the danger with this problem is not

identified. It could be a problem, or it could be marginal. The maximum resistance in the

frame is therefore sat to 10 ohm for now.

There is a possibility the frame has low enough electrical resistance in itself. If the frame has

too high resistance it could be solved with a metal coating (chapter 6.1.3.2), or an epoxy

coating with metal particles (chapter 6.1.3.1)

6.1.1.2 Wear of rails

To get the satellite approved for launch it must satisfy the specifications of low wear of the

satellite against the P-POD rails. There is different ways to limit the wear. Either the epoxy

coating need sufficient hardness (Chapter 6.1.3.1), it could be used the approved aluminum

coating (Chapter 6.1.3.2), or it could be used a type of solid lubricant (Chapter 6.1.3.3).


19 | P a g e

6.1.2 Composites

One of the main goals for the frame it low weight. When it’s desirable to combine it with high

stiffness, composites is a natural option.

Composites are a generic term for a group of materials. A composite is a material composed

of two or more materials with different properties. The idea is to produce a material which is

a combination of several properties which are desired in a particular case.

6.1.2.1 Reinforcement and Matrix – Carbon and Epoxy

There are different types of fibers to choose between. There is for example high modulus

carbon, high strength carbon, E-glass, R-glass, aramid, polyethylene etc. For this application

the high modulus carbon fiber reinforcement is chosen. The tensile strength is at the same

region as the other possibilities, but the tensile modulus is almost the double of high strength

carbon. The downside of this fiber is that the cost is almost 4 times higher than any of the

other fibers.

There are mainly three different matrixes to choose between. They are Epoxy, Phenolic, and

Bismaleimide.

Table 1 - Matrices – Hexcel datasheet

Epoxy Phenolic Bismaleimide

Advantages - Excellent

mechanical

performance

- Good environmental

resistance and high

toughness

- Easy processing

- Excellent fire

resistance

- Good temperature

resistance

- Low smoke and

toxic emission

- Rapid cure

- Economic

processing

- Excellent resistance to

high temperatures

- Service temp up to

260 C

- Good mechanical

characteristics

- Good resistance to

chemical agents, fire

and radiation

For this application epoxy will be the best choice for matrix, mainly for its good mechanical

properties. Reference [2].

6.1.2.2 PrePreg

To make the satellite of carbon fiber and epoxy the PrePreg concept is used. PrePreg is a mat

of fiber reinforcement and matrix. It can either be delivered as unidirectional fibers (UD) or

weaved fibers. The PrePreg mat is uncured, and has to be stored cold until it is used.

There are mainly two methods of applying PrePreg on a mold that is appropriate for us. It is

the vacuum bag process, and the autoclave process.

Vacuum bag – Processing method

It is the vacuum bag process we have the ability to use at NTNU. The process is illustrated as

Figure 23 - PrePreg production process - Own drawn model . First the mold is applied

release wax. Then the PrePreg is placed around the mold. Following a release film, breather,


20 | P a g e

and vacuum bag is placed neatly on top. The vacuum bag is sealed against the mold, while a

vacuum pump hose is inserted under the vacuum bag. The piece is then placed into an oven

and set under vacuum.

Figure 23 - PrePreg production process - Own drawn model

Autoclave – Processing method

The autoclave process is based on the vacuum bag process. The difference is that after the bag

is set under vacuum the vacuum bag process only has 1 atmospheric pressure pressing on the

assembly. For the autoclave process the whole assembly is placed into a pressure oven that

during heating is pressurized between 1-10 bars.

Material data

The complete material data table is attached as chapter 10.10.

There is a practical problem to be dealt with during the selection of the PrePreg system. If an

aerospace certified material is to be used we have to produce it by the autoclave processing

method. At this point NTNU do not have access to this equipment themselves. One possibility

could be to let Kongsberg produce the final frame.

If a non-certified system is used there will be a little more uncertainty related to the choice of

materials. We will also be required to prove that the material is good enough ourselves,

instead of the possibility to refer to the supplier.

The material data provided by Hexcel is limited. There are only some systems that have data

such as strength and stiffness. None of them has engineering constants like poisons numbers

and shear values. For the final material we need to test the material ourselves to ensure correct

material data. As an estimate for the engineering constraints two reference materials are used.

They are T300-5208 and AS4_3501. The first problem is that the relationship will be linear if

you only have two reference materials. The other problem is that there is too large difference

between the elasticity modulus to the reference materials and the Hexcel materials. I have not

at this point been able to find a reference material with as high modulus. This means that it is

no good way to compare the materials to achieve a relative good poison and shear values. The

estimated values in orange in appendix 10.10 are therefore way off, and should not be used

until better reference materials are found.

There are different characteristics that distinguish themselves among the various types of

materials systems. The HexPly system with resin M75 and fiber K139 has the best tensile

modulus, but the lowest tensile strength. The resin M76 has the best resistance against impact,

and lowest outgassing. Whether one consider the Hexply system with resin 8552 and fiber


21 | P a g e

IM7 you got the best strength. Finally the resin M18 has lowest moisture uptake and achieves

the best surface finish.

6.1.2.3 PrePreg Layup

The layup which may be suitably for the satellite is the following:

[90; +45; 0; 0; -45; 90]

Figure 24 - PrePreg Layup - [90; +45; 0; 0; -45; 90] - Own drawn model

Different load scenarios:

The frame is mostly vulnerable to diagonal displacement. This load case cause bending

moment in the walls. The tension will be at its maximum at the surface. It is therefore most

convenient to have the outer composite layer in the direction of the tensile stresses. As long as

the reference direction is axially relative to the satellite this layer will be 90°. At the opposite

side of the tensile side the composite will be in compression. The strength in compression is

smaller. In this case it is important that the quality is good to prevent buckling and/or

delamination of the layer.

The satellite could also be exposed to torsion. At this reason there are two layers of 45°.

The core layers are at 0°. This is to take care of the axial forces that will occur.

The dilemma with this kind of layup is it is not symmetric, nor balanced. This may give

undesirable effects for e.g. during temperature changes.

6.1.3 Surface treatment

There are two challenges to overcome with a proper surface treatment. The first is wear, the

second is electrical conductivity.

6.1.3.1 Epoxy coating

A possibility is an electrically conductive epoxy coating. There is different epoxy on the

market that has achieved good conductivity by blending in silver particles.

An aspect to investigate before choosing such a solution is the wear rate of the optional

coating. Since silver in practice is a solid lubricant it may be positive gain by using a silver


22 | P a g e

epoxy as a top coat sliding against the P-POD rail. At the same time the particles may

increase the softness of the coating, and thereby be more vulnerable against two and three

body wear.

6.1.3.2 Metal coating

A metal coating could handle both wear and have high electrical conductivity. If for example

a hard anodized aluminum coating is used, we will get within the CubeSat design

specification, The CubeSat Program, Cal Poly SLO. There are different ways to coat the

satellite, all with advantages and drawbacks.

Since the satellite is non-metallic, I want to point out two methods of metal coating. The first

is plating on non-conducting materials, and the second is thermal sprayed coatings.

Plating on non-conducting materials

Plating is originally only possible on conductive metals. Due to this non-conductive materials

will need preprocessing to get conductive before the plating process is done. To do this an

electroless process is used to apply a conductive metallic layer of nickel or copper. The same

process is used when the PCB (Printed Circuit Board) is plated with for example chrome.

There are some steps involved before the plating is done:

1) The surface has to be chemical etched to produce pores in the material for the metal to

be anchored. This could be Sulphuric acid/chrome acid mixture.

2) The surface is then activated by colloidial palladium.

3) The palladium needs to be activated. A chemical treatment leaves the palladium atoms

in the pores in the material.

4) The three first steps prepare the surface of the material for metal deposition. In this

step one conduct an autocatalytic (electroless) deposition of nickel or copper. A thin

and continuous metal layer will be deposited in some minutes.

When the surface is prepared the part is ready for the electrolytic plating.


23 | P a g e

Plating is a process where the part is added a metal coating by electrolysis in a bath.

Figure 25 - Electrolytic plating - Own drawn model

Current is applied the plating metal. The plating metal is going through an anodic reaction

where metal dissolves as metal ions in the electrolyte (corrosion process). The metal ions are

then deposited on the part through a cathodic process.

Before the plating is done there are some procedures to be followed:

1) The part has to be thoroughly cleaned to remove contaminations.

2) The surface oxide film has to be removed by electrolytic cleaning in H2SO4 bath.

3) Washing to remove detergents.

There are multiple different metals that can be plated. In this case it would be practical to

plate aluminum since it is already verified by the launch team.

Anodizing is a surface treatment that uses an electrochemical process to transform the

aluminum surface. There are especially two properties which we in this case are looking for.

First: The surface gets increased wear resistance. Second: The electrical resistance is

increased. The first is important since the satellite will be exposed to vibrations during launch.

A plated aluminum layer is thin. It has its limit in both wear and possibility for delamination

from the substrate. The last is important to limit the electron transfer between the P-POD and

the satellite. If an electron transfer between the surfaces is allowed, this can lead to cold

welding. This is a phenomenon that can lead to severe damage to the coating, and will

increase the wear rate significantly.

During the electrochemical process aluminum oxide is transformed from aluminum on the

anode: 2Al(s) + 3H2O Al2O3(s) + 3H2(g). The hydrogen gas is formed at the cathode. The

oxide on the surface is formed on the interface between the aluminum and the oxide. This

means that the electrolyte needs to etch on top of the oxide layer to keep the pores open to

keep the resistance sufficiently low. When the thickness increases, and thereby the resistance,

the oxide formation growth gradually decreases. When the etching on the top of the oxide

layer and the formation at the bottom gets equal, the maximum thickness is reached.


24 | P a g e

The other coating technique is thermal spraying. There are a variety of different methods,

each for their particular applications. In common to them all is that they are a cold spray

technique. That means it’s a kinetic spray process where powder particles are accelerated to

ultra-high velocities by a jet of compressed gas. The velocity varies between 300-1200 m/s. In

this way the particles don’t need too high temperature to get a solid state plastic deformation

on impact with the substrate.

To choose the proper technique it is important to examine what kind of requirements one has

for the coating, what kind of effect it has on the substrate, economic aspect, and the practical

implementation. In this case I wish to keep the temperature of the substrate as low as possible.

Even though thermal spraying is a cold technique the particles will be partially molten, and at

the end increase the temperature of the substrate.

Flame powder spray could be an option. The technique is both cheap and easy to handle. The

particles have low temperature, but also low velocities. The issue with this is that it gives

some porosity and low density with low bond strength. For a better coating quality there are

some models with pressurized inert gas from a remote feeder. In this way higher velocity are

achieved, at the same time as it cools the substrate. The result is a less porous coating.

HVOF (High Velocity Oxy Fuel) is the same type of technique as Flame powder spray, but is

built to achieve an extremely high spray velocity. The coating will get one of the highest

density and strength compared with the other techniques, but the expense and usability will

naturally increase.

As mentioned, the coatings of thermal spray are often associated with porosity. This is

especially critical in situations when combining corrosion, stresses and wear. Since we are not

too concerned of corrosion and stresses, it is more important to look at the combination

between wear and delamination. To improve the bond between the coating and the substrate it

is important to grit blast to proper roughness.

What kind of technique to choose depends on the situation at the time of the decision. At this

point the effect of electromagnetic radiation is unknown. The plating method gives a

smoother and thinner surface coating. The downside is the complexity of the process. It is also

possible to coat it on the inside of the structure. On the other hand, thermal spraying is easier

to only coat selected parts of the satellite. The downside is it’s not that suitable for coating the

inside of the satellite.

For the metal coating chapter I refer to reference [2].

6.1.3.3 Solid lubricant

If the resistance for the carbon fiber frame is low enough it may not be necessary with a

coating except the issue with wear.

As a little unconventional method, I would like to mention solid lubricant as a possibility.

Solid lubricant is a solid material that separate two surfaces and acts like a low friction slip

film. One lubricant that has shown its good properties in vacuum is molybdenum disulfide.


25 | P a g e

While other lubricants are unstable, and may evaporate in vacuum, molybdenum disulfide

will keep its properties. It has a lamella structure that will slide freely over each other when

exposed to two surfaces sliding. It is the low shear forces that occur between the sulfur

molecules that is providing this process. In this way the wear between the P-POD and satellite

rails may be reduced.

Referring to references: [2], and [3].

6.1.4 Material testing

6.1.4.1 Wear

To get the satellite verified to be launched it has to overcome the wear specifications. The

rails in the P-POD, the launch container, are made of anodized aluminum. To secure metallic

compatibility they have specified that the satellite also should be in aluminum. To ensure no

electron transfer between the contact surfaces that can cause cold welding the rails should also

be hard anodized.

If we change the material for the rails to a non-metallic material we will avoid the challenge

with cold welding. What we have to convince is that the wear and the possibility of

interlocking of the component is at an accepted level.

Full scale wear test

The most accurate test is to test the satellite frame with

the proper weight inserted in an authentic P-POD. The

P-POD will be clamped to a vibration disk on the

vibration testing machine. It will then be tested for a

sinus vibration that emulates the actual launch.

With such a test one can obtain proper data for wear

along the rails, but more important, the edges. There is,

and should be, a little clearance between the satellite and

the P-POD. In that way the satellite will be able to twist.

The end of the rails will get a point load from both the

weight and the kinetic energy of the satellite. If we also

add sliding between the surfaces in contact we will get

significantly higher wear.

NUTS do have a P-POD available that may be used for this test. It is certainly thus possible,

but it depend on if it is adequately enough for the launch team with a more simple test for

wear.

Figure 26 - http://www.space.t.u-

tokyo.ac.jp/cubesat/mission/dev/str/img/vibratio

n-s.jpg


26 | P a g e

Pin-on-disk wear test

Although the full scale test will give the most authentic results, it may be things like

economy, time consumption, equipment availability and actual needs that are in favor for

more simple wear tests.

One of this wear tests is pin-on-disk. This test consists of the test material exposed to an

aluminum ball that is sliding over the surface. The aluminum ball is mounted to a motorized

arm. During the test the friction between the aluminum ball and test material is logged. After

the test is performed the wear area can be analyzed for wear.

Like every other simplified wear test it is important to know it is not a direct connection

between the test results and the real wear inside the P-POD. It is only to be used as a

comparison between other relevant materials exposed to the same test. It is thus possible to

see if the material is better or worse than the reference material.

In our case it is natural to select the reference material as anodized aluminum. The aim for the

test is then to ether get the same or better wear resistance between the aluminum sphere and

the relevant material than aluminum against aluminum.

6.1.4.2 Spectroscopy

Before launch the satellite has to be tested for outgassing. The satellite is first outgassed in a

vacuum own for a certain time. Afterwards the satellite is tested in a spectroscopy analyzer.

The machine set the satellite under vacuum and heat and analyses the output gases for

different atoms. The main goal for this test is to ensure that there are no unwanted elements

that are released during launch. It is not expected many surprises during this test from the

carbon fiber or the polymer we are going to use, except for the adhesives. It is thus much

more important for the electrical components as they are made of many different types of

exotic elements

6.1.4.3 Electrical resistance

An important aspect is to test the resistance in the satellite frame from end to end. It should

not exceed 10 ohm. This is relevant if the satellite is not coated with a metallic coating.


27 | P a g e

6.2 Secondary structure

6.2.1 Demands

The secondary structure is not supposed to contribute with additional stiffness, but rather be a

link between all the parts inside and outside the satellite.

CubeSat design specification, The CubeSat Program, Cal Poly SLO:

- 2.1.7.1 Total Mass Loss (TML) shall be ≤1.0%

- 2.1.7.2 Collected Volatile Condensable Material (CVCM) shall be ≤0.1%

NUTS project specifications of importance:

- Sufficient strength to carry…

o …loads from satellites the neighbor satellite inside the P-POD

o …loads from internal components

o …receive and transmit bolt tension

- As low weight as possible

- Should have good damping properties

- Withstand vacuum without excessive deformation

6.2.2 Composite

Originally the satellite stiffness is gained by the frame alone. The secondary structure is a one

single piece in the axial direction, at the same time as it is the piece laying against the next

satellite in the P-POD. This means it is perfect to provide the axial stiffness in the satellite.

If it gets desired to design the satellite with a stiffness contribution from both the frame and

the secondary structure a composite material may be used. The positive aspects with carbon

composite material is a more similar thermal expansion, it has high strength and high

elasticity modulus. The downside is the extra weight it will get.

To take account for the extra strength and stiffness there is designed a lightened version of the

secondary structure. The volume is 72,4 cm3, instead for the original 92.6 cm

3. The drawings

are attached as 10.4.4 and 10.4.5.

6.2.3 Polymer - Solid and foam

The secondary structure is originally meant to only take care of the mounting of the satellite

parts. We can therefore accept a relative low stiffness and strength. An option is to use a

polymer, ether as pure solid or as foam.

A solid polymer material is significantly stronger (~x20) and stiffer than the same material in

foam form. However, if the stiffness and strength is not needed, the foam material will be

around 20 times lighter. Another aspect is that a polymer in solid form is more ductile, while

the foam is more brittle. The vacuum properties are also a more difficult subject when dealing

with foam. One positive aspect with both solid and foam polymer is its ability to damping the


28 | P a g e

vibration of the components inside the satellite. It will also be significantly more ductile than

composite material, meaning it will address displacement in the satellite without introducing

unnecessary stresses to the inner satellite parts.

6.2.4 Comparison overview between relevant materials

A comparison between the relevant materials for the secondary structure:

Type Material Density

[kg/m2]

Secondary

Structure

Weight

[g]

Strength

[MPa]

E

Module

[MPa]

Max

Temp

[C°]

Outgassing

% TML % CVCM

CF CF-Compound

(PPS-HM)

1800 130,32 200 49000 ~0,15 ~0,01

CF-Compound

(PPA-HM)

1800 130,32 280 41000 ~0,15 ~0,01

Polymer

Polystyrene

(PS)

1050 97,23 46

-60

3000

-3600

240 0,09 0,00

Polyethylene

(HDPE)

940

-965

89,35 20

-32

600

-1400

120

-130

0,58 0,19

Polypropylene

(PP)

902

-907

83,98 25

-30

800

-1300

160

-165

0,37 0,17

Polyether ether

ketone

(PEEK)

1320 122,3 99

-100

3600 343 0,20 0,00

Foam

Polystyrene

28-45 4,16 0,8

Polyethylene 25 2,31 0,3

-1

0,79 0,25

Polypropylene 28

-33

3,06 0,25 -70/

+70

% TML: Total Mass Loss, % CVCM: Collected Volatile Condensable Material.

Material data is obtained from references [4], [5], and [6]

The volume for the secondary structure when using CF is 72,4 cm3, and when using polymer

is 92,6 cm3.

As the table indicates CF-Compound, Polystyrene (PS), or Polyether ether ketone (PEEK) is

the best choice according to low outgassing. If a CF-Compound material should be used the

strength and the stiffness should be taken into account during the structural analysis. They are

both stiff and strong, but have high weight. According to weight the foam material is far

superior. It is important to notice one should get hold on foam material with higher density. It

will then gain more weight, but it will still be highly competitive after 5-10 times the weight

in the table.

Which material to use cannot be decided until the satellite concept is selected.


29 | P a g e

6.3 Antenna tape spring

The material used for the tape spring is Beryllium Copper. For the datasheet, refer to appendix

10.9.

At arrival the tape spring has following properties:

- EN CW101C R 580

- Alloy 25

- Temper 1/2H

- Yield Strength Rp 0,2 585 N/mm^2

- Thickness t 0,10 +/-0,004 mm

- With w 211 mm

- Length l 800 mm

The ASTM designation is C17200.

The temper 1/2H stands for “Half Hard”. That means the material has been hardened (H) by

cold working. If the material has been heat treated by the standard heat treatment the temper

would be “1/2HT”.

Figure 27 – Phase Diagram for Beryllium Copper, ref; [7]

According to Figure 27 – Phase Diagram for Beryllium Copper, ref;, when heating above

1300 F (704 C°) the beryllium dissolve in the alpha phase. If it is quickly cooled down to

room temperature the beryllium maintains in solid solution. In this way the material gets soft


30 | P a g e

and ductile. The process is called annealing. Annealing should be done to prepare the material

for age hardening, but may be unnecessary because it’s often a part of the production process

in the manufacture factory.

The gamma phase is below 1100 F (593 C°). Here there is limited solubility of beryllium, and

this characteristic is what causes the hardening. After annealing the material has

supersaturated beryllium solid. When heat treated in this phase the solid cause precipitation,

and the material is hardened.

Figure 28 – Alloy “25 1/2H” - Yield strength/time/temperature for age hardening, ref; [7]

Figure 28 – Alloy “25 1/2H” - Yield strength/time/temperature for age hardening, ref;

describes how the yield strength of the material “25 1/2H” varies over time. 600 F (315 C°) is

the age hardening temperature that has the best effect of increasing the yield strength.

After the forming of the tape springs is performed according to chapter 5.5, the tape spring

has to be heat treated at 315 degree C° in 2 hours.

After heat treatment the following properties is gained

- Yield Strength Rp 0,2 1169-1173 N/mm^2

The elasticity module is not given by the supplier, but is assumed to be 19E6 psi (131 GPa)

according to ref; [7].

NUTS – NTNU Test Satellite - Mechanical 7 - Satellite prototype

31 | P a g e

7 Satellite prototype

To make a prototype we had to start with machining a mold of aluminum for the stiffeners:

The original satellite mold from last semester had to be machined because of high surface

roughness and wrong radius at the edges, and a bolt failure:

A jig to press the PrePreg into the corners were constructed:

PrePreg was cut in sections. It was then laid up in the satellite mold and on a glass plate for

coil plates and cellar panels.

The results after curing in own at 175 C for 2 hours:

NUTS – NTNU Test Satellite - Mechanical 8 - Conclusions

32 | P a g e

8 Conclusions

The secondary structure, that assembles all the internal parts of the satellite, is derived as four

columns, one in each corner. Each column is equipped with slots for the PCB modules. To

mount the battery pack and the camera two solutions is derived. The first involves using

regular slots, while the other is to use the main assembly screws. Which to choose depends on

the final weight of the modules. Both for the stiffener and the ADCS system there are milled

grooves to fit the module fastening ears between the frame and the secondary structure.

There are two flat turnstile antennas (cross dipoles) mounted at the satellite. The antenna tape

springs, made of Beryllium Copper, are wrapped around the satellite during launch. When the

satellite is in orbit they are deployed by a NiCr wire that burns over the nylon thread that

keeps the antennas in place.

The tape spring curvature has to be manufactured at NTNU. A FEM-Analysis if the

manufacturing of the tape springs is carried out. After rolling the antennas the curvature will

experience a spring back, i.e. the curvature will flatten. The curvature results are then used as

input for a buckling analysis. When a purely elastic buckling behavior is verified, the

curvature is examined for stiffness.

According to the results the antenna tape springs after manufacturing with tools at 4 mm

radius will both withstand buckling with purely elastic behavior and be stiff enough to deploy.

What should be investigated are the actual loads and fluctuations that will occur for the

antennas in orbit. This will determine if the antenna geometry is good enough or if they

should be modified.

The material for the frame is PrePreg of carbon fiber with epoxy matrix. If an aerospace

material is used the manufacturing has to be carried out of someone who have access to the

autoclave process. If we want to use an ordinary material we have to investigate the behavior

and the approval of the material.

To overcome the challenges with wear and conductivity there are investigated different types

of surface treatment. The best could be an anodized aluminum coating using a non-conducting

plating method or thermal spraying. This method will have high cost, and it must be

considered whether we get approved other options like e.g. epoxy coating.

For the secondary structure the materials CF-Compound, Polystyrene (PS), and Polyether

ether ketone (PEEK) is the best choice according to low outgassing and high strength.

Polystyrene has the lowest weight, but is also the weakest. Before a material is selected it

must be considered whether or not the secondary structure shall be used to relive stiffness of

the satellite frame.

A prototype of the satellite is carried out in PrePreg of carbon fiber. The stiffeners and the

plates for the cellar panels and ADCS system are also made. The next step is to cut out the

windows of the satellite frame and cut out the right shape of the panels.

NUTS – NTNU Test Satellite - Mechanical 9 - Literature references

33 | P a g e

9 Literature references

[1] "NUTS - NTNU Test Satellite, a Norwegian CubeSat Project," [Online]. Available:

http://nuts.cubesat.no/.

[2] Hexcel, "Prepreg Technology," [Online]. Available: www.hexcel.com.

[3] R. Johnsen, O. Ø. Knudsen and N. Espallargas, Lecture notes from TMM4205 Tribology.

[4] "Solid Film Lubricants," [Online]. Available:

http://www.machinerylubrication.com/Read/861/solid-film-lubricants .

[5] "MatBase," [Online]. Available:

http://www.matbase.com/material/polymers/commodity/hdpe/properties .

[6] "Nasa Outgassing Database," [Online]. Available:

http://outgassing.nasa.gov/cgi/uncgi/search/search_ad.sh.

[7] "Lati High Performance CF," [Online]. Available:

http://www.lati.com/en/news/2009/high_modulus_carbon_fibres_in_super-

structural_compounds.html .

[8] B. Wellman, Guide to Beryllium Copper.

NUTS – NTNU Test Satellite - Mechanical 10 - Attachments

34 | P a g e

10 Attachments

10 ATTACHMENTS 34

10.1 STRUCTURE OF WORK 35

10.2 DIVIDING OF SUBJECTS 36

10.4 PART DRAWINGS 37

10.4.1 SATELLITE ASSEMBLY 37

10.4.2 ANTENNA ASSEMBLY 38

10.4.3 ANTENNA PCB CARDS 39

10.4.4 SECONDARY STRUCTURE – POLYMER 40

10.4.5 SECONDARY STRUCTURE – CARBON FIBER 41

10.4.6 CAMERA MODULE, MAXIMUM DIMENSIONS 42


10.5 TAPE SPRING MANUFACTURING TOOL DRAWING, HE 44

10.6 TAPE SPRING MANUFACTURING TOOL DRAWING, SHE 45

10.7 ANTENNA ASSEMBLY ILLUSTRATIONS 46

10.8 SECONDARY STRUCTURE ILLUSTRATIONS 51

10.9 APPENDIX 1 – BERYLLIUM COPPER DATASHEET – HARALD PIHL AB 52

10.10 PREPREG MATERIAL DATA 53

10.11 TAPE SPRING MANUFACTURING – RESULTS 55

10.12 BUCKLING AND DEPLOYING OF ANTENNA TAPE SPRING – RESULTS 63

10.13 STRENGTH OF ANTENNA TAPE SPRING – RESULTS 67

10.14 FE ANALYSIS OF ANTENNA TAPE SPRING IN ABAQUS 68

10.14.1 MANUFACTURING OF ANTENNA TAPE SPRING 68

10.14.2 BUCKLING AND DEPLOYING OF ANTENNA TAPE SPRING 74

10.14.3 STRENGTH OF ANTENNA TAPE SPRING 78


35 | P a g e

10.1 Structure of work


36 | P a g e

10.2 Dividing of subjects

Christian Nomme Kim Sandvik

Primary structure Evaluation of concept

CAD of frame

CAD of stiffeners

Basis for analysis

Static frame analysis

Modal frame analysis

Material selection of frame

Coating?

Material orientation of frame

Implementation of material-

properties in analysis.

Material testing

Secondary structure Static analyses of structure Evaluation of concept

CAD of secondary structure

Material selection of secondary

structure

Prototype Main structure

Stiffeners

Auxiliary equipment

Prototype testing

Main structure

Stiffeners

Secondary structure

Prototype testing

Tertiary structure Photovoltaic system concept Antenna system concept


37 | P a g e

10.4 Part drawings

10.4.1 Satellite assembly


38 | P a g e

10.4.2 Antenna assembly


39 | P a g e

10.4.3 Antenna PCB cards


40 | P a g e

10.4.4 Secondary structure – Polymer


41 | P a g e

10.4.5 Secondary structure – Carbon fiber


42 | P a g e

10.4.6 Camera module, maximum dimensions


43 | P a g e

10.4.7 Battery pack


44 | P a g e

10.5 Tape spring manufacturing tool drawing, He


45 | P a g e

10.6 Tape spring manufacturing tool drawing, She


46 | P a g e

10.7 Antenna assembly illustrations

Assembly overview:

Assembly top close-up:

Assembly bottom close-up:


47 | P a g e

Disassembly of top PCB card:

Exploded view 1:


48 | P a g e

Exploded view 2:

Tape spring folding pieces 1:


49 | P a g e




50 | P a g e

Assembly exploded view, 2D:


51 | P a g e

10.8 Secondary structure illustrations

Secondary structure for polymer material:


52 | P a g e

10.9 Appendix 1 – Beryllium Copper datasheet – Harald Pihl AB


53 | P a g e

10.10 PrePreg Material Data


54 | P a g e

PrePreg Material Data (Continues)


55 | P a g e

10.11 Tape Spring Manufacturing – Results

Step1

Step 1.1 – Mises:

Step 1.2 – Mises:


56 | P a g e

Step 1.3 – Mises:

Step2

Step 2.1 – Mises:


57 | P a g e

Step 2.2 – Mises:

Step 2.3 – Mises:


58 | P a g e

Step 2.4 – Mises:

Step 2.5 – Mises:


59 | P a g e

Tape spring after manufacturing –Mises:

Tape spring after manufacturing, Cross section –Mises:


60 | P a g e

Tape spring after manufacturing, Cross section, Close-up –Mises:

Tape spring after manufacturing, Cross section –PEEQ:


61 | P a g e

Tape spring after manufacturing, Cross section –PEEQ:

Tape spring after manufacturing, Cross section, at edges –PEEQ:


62 | P a g e

Tape spring after manufacturing, Cross section, Center –PEEQ:


63 | P a g e

10.12 Buckling and deploying of Antenna Tape Spring – Results

Model:

Step 1.1 – Mises:


64 | P a g e

Step 1.2 – Mises:

Step 1.3 – Mises:


65 | P a g e

Step 2.1 – Mises:


66 | P a g e

Step 1 – PEEQ:

Step 2 – PEEQ:


67 | P a g e

10.13 Strength of Antenna Tape Spring – Results

Von Mises stress:

Displacement, y-direction:


68 | P a g e

10.14 FE Analysis of Antenna Tape Spring in Abaqus

10.14.1 Manufacturing of Antenna Tape Spring

10.14.1.1 Drawing the tools


69 | P a g e

10.14.1.1 Drawing the tape spring

10.14.1.2 Inertias of the tools

Insert these values for both she and he press, for an RP defined in each part.

Also convert the she and he press to shell element to be able to mesh them.


70 | P a g e

10.14.1.3 Meshing of parts

Insert these values. Remember explicit, linear, and second-order accuracy.

Mesh the tape spring with at least 4 elements along the thickness, and the tools with finer

mesh than the tape spring.

10.14.1.4 Tape spring material

Assign following material to the tape spring:


71 | P a g e

10.14.1.5 Step 1, Press


72 | P a g e

10.14.1.6 Step 2, Rolling

Step 2 is the same as step 1, only with time period as 1.

10.14.1.7 Interaction Properties

10.14.1.8 Interactions


73 | P a g e

10.14.1.9 BC HePress

Make a “smooth step” for the displacements amplitude.

10.14.1.10 BC ShePress

10.14.1.11 Job

Make a new job, if multiple processors edit “use multiple processors” below “parallelization”:


74 | P a g e

10.14.2 Buckling and deploying of Antenna Tape Spring

10.14.2.1 Part drawing

Model the corner as discrete rigid, and extrude it 20 mm.

Model the Tape Spring as deformable solid 3D, and extrude it 100 mm.


Assign the following material to the Tape Spring.


75 | P a g e

10.14.2.3 Interaction Properties

10.14.2.4 Interactions


76 | P a g e

10.14.2.5 BC - Tape Spring

10.14.2.6 BC – Corner


77 | P a g e


78 | P a g e

10.14.3 Strength of Antenna Tape Spring

10.14.3.1 Part drawing

Model the Tape Spring as deformable solid 3D, and extrude it 700 mm.


Assign the density of the material and the following material data to the Tape Spring.

10.14.3.3 Loads

Apply gravity load in –y direction in step 1.

10.14.3.4 BS

Fix the Tape Spring end located in origin.

Development of Composite and Polymer Material CubeSat...

Documents

Transcript of Development of Composite and Polymer Material CubeSat...