SolarProbe ME PDF

Download as pdf or txt
Download as pdf or txt
You are on page 1of 146

Sol

arProbe+

2
008Sol
Mis
sionEngi
neer
ing

arPr
obe+Mi
St
udyReport

ssi
Mar
ch1
0,2
008

Mar
ch1
0,2
008 onEngi
neer
ingSt
udyRepor
t

07-
05537-
1
March 10, 2008

Solar Probe+ Mission Engineering Study Report


Prepared for NASA’s Heliophysics Division
By The Johns Hopkins University Applied Physics Laboratory (Under Contract NNN06AA01C)

Approved By:

_________________________________ _________________________________
Andrew A. Dantzler Robert D. Strain
Study Lead, The Johns Hopkins University Space Department Head, The Johns
Applied Physics Laboratory Hopkins University Applied Physics
Laboratory

_________________________________
J. Walter Faulconer
Civilian Space Business Area Executive,
The Johns Hopkins University Applied
Physics Laboratory

DISCLAIMER: Note that all colors used on mechanical drawings contained in this report are for
clarification of image content only and do not necessarily reflect the actual colors of the items depicted.
ACKNOWLEDGMENTS

STUDY TEAM AND SUPPORT

David Artis, JHU/APL Rainee Simons, NASA/Glenn Research Center


Scott Benson, NASA/Glenn Research Center Tom Sutliff, NASA/Glenn Research Center
Stewart S. Bushman, JHU/APL Michael D. Trela, JHU/APL
Andrew Dantzler, JHU/APL Brian R. Tull, JHU/APL
Christopher Deboy, JHU/APL Steve Vernon, JHU/APL
Michelle M. Donegan, JHU/APL Joe Wagner, MCR Technologies, LLC
David G. Drewry, JHU/APL David R. Weir, JHU/APL
Peter Eisenreich, JHU/APL Melissa Wirzburger, JHU/APL
Jacob Elbaz, JHU/APL Larry Wolfarth, JHU/APL
Martin Fraeman, JHU/APL
Daryl George, JHU/APL SCIENCE AND TECHNOLOGY
Robert E. Gold, JHU/APL
DEFINITION TEAM
Yanping Guo, JHU/APL
Meagan Hahn, JHU/APL Loren Acton, Montana State University
Roshanak Hakimzadeh, NASA/Glenn Research Marianne Balat, Centre National de la Recherche
Center Scientifique (CNRS)
Ray Harvey, JHU/APL Volker Bothmer, University of Goettingen
John Hickman, NASA/Glenn Research Center Ray Dirling, Science Applications International
Angela Hughes, JHU/APL Corporation (SAIC)
Jack Hunt, JHU/APL Bill Feldman, Los Alamos National Laboratory
Kelly Judge, JHU/APL George Gloeckler, University of Michigan
Anne King, JHU/APL Shadia Habbal, University of Hawaii
James D. Kinnison, JHU/APL Don Hassler, Southwest Research Institute (SwRI)
Geoffrey Landis, NASA/Glenn Research Center Ingrid Mann, Westfalische Wilhelms-Universitat
Valerie Lyons, NASA/Glenn Research Center Muenster
Douglas Mehoke, JHU/APL Bill Matthaeus, University of Delaware/Bartol
Chris Monaco, JHU/APL Research Institute
Steve Parr, JHU/APL Dave McComas (Chair), SwRI
David Persons, JHU/APL Ralph McNutt, JHU/APL
Sandra Reehorst, NASA/Glenn Research Center Dick Mewaldt, California Institute of Technology
John Riehl, NASA/Glenn Research Center Neil Murphy, Jet Propulsion Laboratory (JPL)
Elliot Rodberg, JHU/APL Leon Ofman, Catholic University of America
Lew Roufberg, JHU/APL Ed Sittler, NASA/Goddard Space Flight Center
Paul Schmitz, NASA/Glenn Research Center Chuck Smith, University of New Hampshire
Richard K. Shaltens, NASA/Glenn Research Center Marco Velli, Florence/JPL
Hongxing Shapiro, JHU/APL Thomas Zurbuchen, University of Michigan

ii

Mission Overview Mission Summary
Solar Probe+ will be a historic mission, flying into one of Launch Vehicle Atlas V 551

the last unexplored regions of the solar system, the Sun’s Third Stage STAR-48BV
atmosphere or corona, for the first time. Approaching as Launch Window May 21 through June 9, 2015
close as 8.5 solar radii above the Sun’s surface, Solar Probe+ First Perihelion August 2015
will employ a combination of in situ measurements and First Min. Perihelion at 9.5 RS October 2021
imaging to achieve the mission’s primary scientific goal: Cost $739.5 million (FY07$)
to understand how the Sun’s corona is heated and how the
solar wind is accelerated. Solar Probe+ will revolutionize our
knowledge of the physics of the origin and evolution of the
solar wind.

Science Objectives
• Determine the structure and dynamics of the magnet-
ic fields at the sources of the fast and slow solar wind
• Trace the flow of energy that heats the corona and
accelerates the solar wind
• Determine what mechanisms accelerate and transport
energetic particles
• Explore dusty plasma phenomena in the near-Sun
environment and their influence on the solar wind and
energetic particle formation

Solar Probe+ shown with primary solar array panel in the


deployed position.

Spacecraft Summary
Wet Mass 481 kg (lift mass: 610 kg for 30% margin)
Power • 482 W for high-power mode; includes
34% margin
• Dual solar array system
Configuration Three-axis-stabilized
Propulsion Hydrazine monopropellant 190 m/s ΔV
Attitude control Reaction wheels and thrusters
Thermal Control Passive thermal control using blankets
and active cooling for secondary solar
arrays
Telecommunications • Dual-frequency X-band and Ka-band
through articulated HGA
• Dual 128-Gbit solid-state recorders
Solar Probe+ will launch on an Atlas V551 with a STAR-48BV third for redundant data storage prior to
stage to achieve the required launch energy. downlink

07-05537
Solar Intensity

Power Generation
Solar Probe+ uses two sets of solar cell arrays, each optimized to
work over a different range of Sun–probe distances.

Instruments will be retracted inside the umbra of the


Thermal Protection System to limit solar exposure.


Mission Enablers
Challenge Solution
Solar Intensity Thermal Protection System

Temperature Control
Power Generation Secondary Solar Arrays
Temperature Control Support Structure
Mission Design Multiple Venus Gravity-Assist Trajectories
The transition structure assembly provides the
mechanical support structure that couples the
bus to the Thermal Protection System.

Mission Design

The baseline Solar Probe+ trajectory uses Venus flybys and no deep-
space maneuvers to reach a minimum perihelion of 9.5 RS in 6.4 years.
07-05537
TABLE OF CONTENTS

TABLE OF CONTENTS
1.0. EXECUTIVE SUMMARY ..................................................................................... 1-1
1.1. Solar Probe+ Spacecraft and Mission Design Summary ......................................................1-1
1.2. Solar Probe+ Summary..........................................................................................................1-3

2.0. SCIENCE OBJECTIVES AND INSTRUMENTATION ......................................... 2-1

3.0. MISSION IMPLEMENTATION ............................................................................. 3-1


3.1 Baseline Mission Design .........................................................................................................3-2
3.1.1 Mission Design Overview.........................................................................................3-2
3.1.2 Launch ............................................................................................................3-3
3.1.3 Baseline Mission Trajectory .....................................................................................3-3
3.1.4 Solar Encounters......................................................................................................3-5
3.1.5 Comparison with 2005 Solar Probe .........................................................................3-6
3.1.6 ΔV Budget and Navigation .......................................................................................3-7
3.1.7 Launch Vehicle and Third Stage ..............................................................................3-8
3.1.8 Launch Opportunity..................................................................................................3-9
3.1.9 Mission Design Trade Study ..................................................................................3-10
3.1.9.1 Trajectory Study ..............................................................................................3-11
3.1.9.2 Rationale for Selection ....................................................................................3-13
3.2 Mission Concept of Operations .............................................................................................3-14
3.2.1 Overview ..........................................................................................................3-14
3.2.2 Launch and Early Operations.................................................................................3-14
3.2.3 Mission Events .......................................................................................................3-15
3.2.4 Operations ..........................................................................................................3-16
3.3 Mission Environment .............................................................................................................3-17
3.3.1 Solar Flux ..........................................................................................................3-17
3.3.2 Radiation ..........................................................................................................3-17
3.3.3 Coronal Lighting .....................................................................................................3-18
3.3.4 Solar Scintillation....................................................................................................3-18
3.3.5 Spacecraft Charging ..............................................................................................3-18
3.3.6 Micrometeoroid and Dust .......................................................................................3-19
3.3.7 Electromagnetic Interference, Electromagnetic Compatibility, and Magnetic
Cleanliness ..........................................................................................................3-20
3.4 Spacecraft Overview .............................................................................................................3-20
3.4.1 Spacecraft Description ...........................................................................................3-20
3.4.2 Spacecraft and Mission Reliability .........................................................................3-22
3.4.3 Mass and Power Budget Summaries.....................................................................3-24
3.4.3.1 Mass Budget ...................................................................................................3-24
3.4.3.2 Power Budget..................................................................................................3-24
3.5 Mechanical Systems .............................................................................................................3-24
3.5.1 Bus Configuration and Structure Design................................................................3-25
3.5.2 Transition Structure Assembly Design Overview ..................................................3-28
3.5.3 Bus and Transition Structure Assembly System Mechanisms..............................3-30
3.5.3.1 Primary Solar Array Design Overview.............................................................3-30
3.5.3.2 Secondary Solar Array and Transition Structure Assembly Mechanism Design
Overview .........................................................................................................3-31
3.5.3.3 Science Instrument Payload Mechanical System Accommodations...............3-33
3.5.3.3.1 Aft-Mounted Science Boom....................................................................3-33

v
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

3.5.3.3.2 Mechanical System Interfaces and Science Instrument Accommodations . 3-33


3.5.3.4 High-Gain Antenna Mechanical Mechanism Overview ...................................3-33
3.5.4 Spacecraft System Structural Analysis and Launch Environments .......................3-35
3.5.5 Spacecraft System Mechanical Thermal Test and Qualification Plan....................3-36
3.5.5.1 Rationale and Test Flow .................................................................................3-36
3.5.5.2 Bus Structure Qualification Plan .....................................................................3-37
3.5.5.3 Transition Structure Assembly Qualification Plan ...........................................3-37
3.5.5.4 Thermal Shield Development ..........................................................................3-38
3.5.5.5 Spacecraft System-Level Mechanical Test Campaign....................................3-38
3.6 Thermal Protection System...................................................................................................3-39
3.6.1 System Design Requirements................................................................................3-39
3.6.2 Design Approach....................................................................................................3-40
3.6.3 Thermal Protection System Development Program...............................................3-44
3.6.3.1 Coupon and Analog Testing............................................................................3-44
3.6.3.1.1 Material Test Program............................................................................3-44
3.6.3.1.2 Analog Element Testing .........................................................................3-45
3.6.3.2 Prototype Development...................................................................................3-45
3.6.3.3 Spare and Flight Shield Fabrication, Assembly, and Testing..........................3-46
3.7 Thermal Control System........................................................................................................3-47
3.7.1 Spacecraft Bus.......................................................................................................3-47
3.7.1.1 Requirements..................................................................................................3-47
3.7.1.2 Thermal Protection System Interface ..............................................................3-47
3.7.1.3 Block Diagram.................................................................................................3-47
3.7.1.4 Thermal Analysis.............................................................................................3-48
3.7.1.5 Instrument Interface ........................................................................................3-49
3.7.1.6 Trades ..........................................................................................................3-49
3.7.2 Primary Solar Panels .............................................................................................3-50
3.7.3 Secondary Solar Array ...........................................................................................3-50
3.8 Power System .......................................................................................................................3-52
3.8.1 Power System Electronics .....................................................................................3-53
3.8.2 Solar Arrays ..........................................................................................................3-54
3.8.2.1 Primary Solar Array .........................................................................................3-54
3.8.2.2 Secondary Solar Array ....................................................................................3-55
3.8.3 Battery ..........................................................................................................3-57
3.8.4 Power System Performance ..................................................................................3-58
3.9 Avionics System ....................................................................................................................3-58
3.9.1 Avionics Suite.........................................................................................................3-58
3.9.2 Flight Software .......................................................................................................3-59
3.10 Telecommunications ...........................................................................................................3-61
3.10.1 Trade Studies.......................................................................................................3-61
3.10.1.1 Frequency Selection .....................................................................................3-61
3.10.1.2 High-Gain Antenna Design ...........................................................................3-62
3.10.2 Subsystem Implementation..................................................................................3-63
3.10.3 Performance.........................................................................................................3-64
3.11 Data Management...............................................................................................................3-64
3.11.1 Science Data Collection .......................................................................................3-65
3.11.2 Data Return..........................................................................................................3-65
3.12 Guidance and Control System ............................................................................................3-66
3.12.1 Attitude Determination..........................................................................................3-67
3.12.2 Attitude Control ....................................................................................................3-67

vi
TABLE OF CONTENTS

3.12.3 Environmental Considerations .............................................................................3-68


3.12.4 Pointing Strategy..................................................................................................3-68
3.12.5 High-Gain Antenna Control ..................................................................................3-69
3.12.6 Guidance and Control Changes Since Previous Study........................................3-69
3.13 Propulsion System ..............................................................................................................3-69
3.14 Environmental Mitigation .....................................................................................................3-71
3.14.1 Charging ..........................................................................................................3-71
3.14.1.1 Charging Analysis Methodology....................................................................3-71
3.14.1.2 Charging Results...........................................................................................3-71
3.14.1.3 Impact of Spacecraft Charging......................................................................3-72
3.14.1.4 Mitigation Strategies......................................................................................3-73
3.14.1.5 Conclusions...................................................................................................3-73
3.14.2 Methodology.........................................................................................................3-74
3.15 Technical Challenges ..........................................................................................................3-76

4.0. COST ESTIMATE ................................................................................................ 4-1


4.1. Cost Estimate Summary ........................................................................................................4-1
4.2. Cost-Estimating Methodology and Independent Cost Estimate............................................4-2
4.2.1. Cost-Estimating Methodology .................................................................................4-2
4.2.2. Cost Estimate Validation: Independent Cost Estimate............................................4-2
4.2.2.1. Summary..........................................................................................................4-2
4.2.2.2. Methodology.....................................................................................................4-4
4.2.2.2.1. Ground Rules and Assumptions..............................................................4-4
4.2.2.2.2. Parametric Cost-Estimating Models and Data Used to Generate ICEs ....4-4
4.2.3. Results ............................................................................................................4-7
4.3. Schedule...............................................................................................................................4-10
4.4. Work Breakdown Structure and Cost Detail ........................................................................4-11
4.4.1. Work Breakdown Structure Dictionary ..................................................................4-11
4.5. Subsystem Cost Detail .........................................................................................................4-14

APPENDICES
Appendix A: Solar Probe+ Mass and Power Budgets.................................................................. A-1
Appendix B: Solar Probe+ Link Analysis ...................................................................................... B-1
Appendix C: Solar Encounter Power Generation Trade Study .................................................... C-1
Appendix D: References ............................................................................................................... D-1
Appendix E: Acronyms and Abbreviations.................................................................................... E-1

vii
1.0 EXECUTIVE SUMMARY

1.0. EXECUTIVE SUMMARY this study's compliance with the requirements


The concept of a probe to investigate the laid out in the NASA Task Order
close regions of the Sun has been studied for (NNN07AA15T of Contract NNN06AA01C)
more than 40 years. In 2005, NASA tasked that initiated the study. All requirements are
The Johns Hopkins University Applied Phys- met, with the exception of meeting the mass
ics Laboratory (APL) to work with an inde- margin of 35%. The 30.1% margin we present,
pendent Science and Technology Definition however, is within the industry standard. In
Team (STDT) to determine the optimum con- addition, methods to increase this margin exist
cept for such a Solar Probe mission. The re- and warrant further study. Mass margin is dis-
sults of that study 1 described an exciting mis- cussed further in Section 3.4.3.
sion to the Sun that would meet all of the sci- 1.1. Solar Probe+ Spacecraft and Mission
ence objectives laid out by the STDT. Unfor- Design Summary
tunately, the overall cost of the described mis- In two full Science and Technology Defini-
sion and its required use of plutonium for tion Team (STDT) meetings and several
power production made the mission unrealiz- smaller discussions, the STDT and engineer-
able within available resources. ing team members refined science and meas-
On August 27, 2007, NASA tasked APL to urement objectives from previous Solar Probe
conduct a new study, called Solar Probe Lite, studies, defined instrument resource and ac-
to determine the feasibility of a non-nuclear commodation requirements, developed orbit
mission that would retain as much of the 2005 geometries and concepts of operations, and
Solar Probe concept’s science as possible, for identified risk mitigation measures and cost
a total mission cost of less than $750M. In savings options. The STDT encouraged the
performing this study, APL worked in tandem engineering team to draw on their own exten-
with the original STDT to ensure that the new sive internal experience with space missions,
concept remained commensurate with the in- along with examining external ideas and solu-
tended science. The result is a mission called tions, to develop the most detailed and techni-
Solar Probe+, so dubbed by the STDT because cally complete engineering study possible.
of the potential gains in science of the current Key requirements that flow down from the
concept over its predecessors. study ground rules and science objectives are
This report presents the engineering concept summarized here:
and cost estimate basis for the Solar Probe+
mission. The science objectives, justification, Mission Requirements
and science implementation plan, as well as • Achieve at least three orbits with perihe-
comparisons to the objectives and implemen- lion distance less than 10 RS
tation presented in the 2005 study, for the So- • Achieve the above within 10 years
lar Probe+ mission will be described in a part- • Return all data collected from each perihe-
ner report to this one, 2 written in parallel by lion pass
the STDT. Table 1.0-1 provides a summary of Spacecraft Requirements
• Survive solar intensity during perihelion
1
Solar Probe: Report of the Science and Technology pass (with a solar intensity of ~510 Suns)
Definition Team, NASA/TM—2005–212786, National • Provide reliable power over the distance
Aeronautics and Space Administration, Goddard Space
Flight Center, Greenbelt, MD (2005). range of 0.04–1 AU
2
Solar Probe+: Report of the Science and Technology • Accommodate the STDT-provided straw-
Definition Team, Southwest Research Institute, San man payload
Antonio, TX, in press (2008).

1-1
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

Table 1.0-1. Summary of study concept performance against task order requirements.
Item from Task Order NNN07AA15T Study Report Compliance
“Total mission lifecycle cost shall not exceed $750 million Total mission cost is estimated at ~$740M in FY07$.
in FY07 dollars including the launch vehicle.”
“Use the earliest technically feasible launch readiness Mission design work has yielded May 2015 as the opti-
date, assuming a April 2009 Phase A start with no con- mum launch date. Schedule and costing assumes April
straints on funding profile.” 2009 Phase A start.
“The instrument costs shall not exceed $100 million, and The STDT has reached consensus on a strawman pay-
NASA shall specify the reduced payload content from load suite. Cost is assumed to be $100M, all-inclusive.
the Solar Probe STDT reports.”
“The launch vehicle is an Evolved Expendable Launch The launch vehicle is an Atlas V 551.
Vehicle (EELV).”
“The cost estimate shall include at least one independent, An independent cost estimate was performed by MCR
parametric estimate.” Technologies, LLC, and reconciled with independent
APL cost analysts to ensure validity.
“The cost estimate shall include 30 percent reserves on The report cost estimate includes 30% on Phases B–D, as
Phases B through D excluding the EELV.” well as 5% on Phase A and 15% on Phase E. Reserves
are on total mission excluding the launch vehicle, Deep
Space Network (DSN), and payload suite.
“The schedule reserves for mission shall include at least The report schedule includes 9.5 months of funded re-
one month per year for Phases B and C and two months serve. Reserve in each phase meets or exceeds the re-
per year for Phase D.” quirement.
“Margins for mass and power shall be at least 35 percent.” Mass margin is 30.1%. Possibilities for increase of margin,
if necessary, exist and can be studied in Phase A.
Power margin is 34.4% (worst case).

“The mission pre-concept shall not use Multi-mission Ra- Solar Probe+ uses photovoltaics.
dioactive Thermoelectric Generators”

The result of this study is a technically fea- alternative trajectory design options, including
sible, acceptably low-risk and affordable mis- a Jupiter gravity assist trajectory design.
sion that can survive in the unique thermal and The baseline mission design (Figure 1.1-1)
dust environment near the Sun and that will meets the above-listed mission requirements
fully achieve all of the Solar Probe+ science and program constraints. This design features
objectives. • Launch in May 2015, with total mission du-
Trajectory trade studies were conducted ration of ~7 years
aimed at developing a feasible mission concept • First perihelion at ~0.16 AU (35 RS) in 3
for Solar Probe+ under NASA’s new direction months after launch
and guidelines. Trajectory design is the key that • 24 orbits over 6.9 years, 19 perihelia within
defines the scope of the mission and determines 20 RS , yielding 961 hours within the 20-RS
how the mission will be implemented, what can region
be accomplished for science, and how much it • Three perihelia at 9.5 RS (heliocentric dis-
will cost. A wide range of trajectory options tance), with an orbital period of 88 days
were explored with various mission scenarios. • Frequent visits of the near-Sun region over
Representative trajectory design options were 6 years, providing extensive science meas-
presented to the STDT and discussed between urements over a half solar cycle
the APL engineering team and the STDT sci-
ence team. After discussions and comparison, a In addition to providing nearly 1000 hours of
baseline trajectory design was selected from six science collection opportunity within the near-
Sun region, the baseline mission design also

1-2
1.0 EXECUTIVE SUMMARY

distances. Within 0.25


AU, two smaller,
high-temperature-
tolerant photovoltaic
arrays provide power.
These arrays are
cooled by a pumped
liquid system and
gradually are retracted
behind the TPS as the
spacecraft approaches
the Sun, effectively
keeping the incident
solar power approxi-
mately constant. A
simple blowdown hy-
drazine monopropel-
Figure 1.1-1. Baseline Solar Probe+ trajectory.
lant propulsion sys-
tem will be used for
has several advantages in mission implementa- ΔV maneuvers and attitude control. The guid-
tion. For example ance and control system consists of three star
• Perihelion gradually decreases to 9.5 RS, trackers and a high-precision, internally redun-
allowing for “practice,” leading to succes- dant inertial measurement unit that provide atti-
sively closer perihelia tude knowledge, and attitude control is provided
• Aphelion is less than 1 AU, ensuring suffi- by four reaction wheels and 12 thrusters. The
cient solar power without the need for radio- spacecraft is equipped with three antennas: a
isotope thermoelectric generators (RTGs) high-gain antenna (HGA) mounted to a dual-
axis gimbaled mast and two hard-mounted low-
The baseline Solar Probe+ is a three-axis- gain antennas (LGAs). The HGA is the prime
stabilized spacecraft designed to survive and antenna for the Ka-band science return
operate successfully throughout its many orbits downlink, whereas X-band uplink and downlink
around the Sun. The spacecraft concept utilizes capability is provided through all antennas.
a Thermal Protection System (TPS) that features When the spacecraft is within 0.59 AU of the
a large 2.7-m diameter carbon–carbon low- Sun, communications are maintained through
conductivity, low-density shield. The TPS pro- the LGAs.
tects the spacecraft bus and instruments within Although the benefits of the Solar Probe+ are
its umbra during the solar encounter. The bus numerous, it is necessary to note that there are
consists of a hexagonal equipment module and a important issues that must be more fully ad-
cylindrical adapter. It provides an efficient me- dressed in Phase A in order to help ensure a suc-
chanical structure that accommodates the in- cessful mission. Among these issues are the
struments and spacecraft subsystems and han- multiple-flyby trajectory and the detailed design
dles the loads from the TPS and the launch and qualification of the Thermal Protection Sys-
loads. Solar Probe+ will be powered by two sets tem and the secondary solar arrays.
of photovoltaic arrays. The two primary arrays
are deployed when the spacecraft is more than
0.25 AU from the Sun and are retracted at closer

1-3
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

1.2. Solar Probe+ Summary in this study report is based on a rigorous engi-
Solar Probe+ is an exciting mission of explo- neering study performed in concert with the Sci-
ration and discovery. It will journey to one of ence and Technology Definition Team (STDT).
the last unexplored regions of the solar system The described mission is technically feasible,
and reveal how the corona is heated and the so- can be accomplished within realistic resources,
lar wind is accelerated, solving two fundamental and can fully achieve all science objectives, thus
mysteries that have been top-priority science transforming our understanding of the Sun and
goals for many decades. The mission described its sister Sun-like stars and enabling exploration.

1-4
2.0 SCIENCE OBJECTIVES AND INSTRUMENTATION

2.0. SCIENCE OBJECTIVES AND report, 2 in which mass, power, and peak data
INSTRUMENTATION rate are provided. The payload suite will be
The Solar Probe+ Science and Technology serviced by a common data processing unit
Definition Team (STDT) has written a partner (CDPU) and a low-voltage power supply
report to this one, 1 describing in detail the sci- (LVPS). A single integrated payload has been
ence objectives, justification, and science im- assumed in the engineering study and cost es-
plementation plan for the Solar Probe+ mis- timate. The strawman payload consists of the
sion. The purpose of the present engineering following instruments:
study report is to describe how these science
objectives will be met. To set the context for In Situ Instrumentation
the engineering description, a top-level sum- • Fast Ion Analyzer (FIA)
mary of the Solar Probe+ science objectives is • Two Fast Electron Analyzers (FEAs)
presented here. • Ion Composition Analyzer (ICA)
• Energetic Particle Instrument (EPI)
• Determine the structure and dynamics of • Magnetometer (MAG)
the magnetic fields at the sources of the • Plasma-Wave Instrument (PWI)
fast and slow solar wind • Neutron/Gamma-Ray Spectrometer (NGS)
• Trace the flow of energy that heats the co- • Coronal Dust Detector (CD)
rona and accelerates the solar wind
• Determine what mechanisms accelerate Remote-Sensing Instrumentation
and transport energetic particles • White-Light Hemispheric Imager (HI)
• Explore dusty plasma phenomena in the
near-Sun environment and their influence In addition, the STDT has assumed re-
on the solar wind and energetic particle sources (within the overall allocation for the
formation suite) for an additional instrument, in the event
that an unforeseen beneficial component is
The Solar Probe+ science objectives will be proposed.
addressed through a combination of in situ and The entire payload suite connects to a
remote-sensing observations made from a CDPU, which integrates the data processing
near-ecliptic heliocentric orbit at progressively and LVPS for all of the payload science in-
closer distances to the Sun, with the spacecraft struments into a fully redundant system that
achieving a minimum perihelion distance of eliminates replication, increases redundancy,
~9.5 RS in roughly 6.5 years after launch. and reduces overall payload resources.
To meet the Solar Probe+ science objec- The CDPU provides a unified interface to
tives, the STDT has recommended an inte- the payload for the spacecraft. The spacecraft
grated strawman payload comprising in situ selects which side of the CDPU will be pow-
and remote-sensing instruments. For the pur- ered, leaving the redundant side off as a cold
poses of assigning resources to the compo- spare. The payload CDPU communicates with
nents of payload suite, the STDT started with the spacecraft over a MIL-STD-1553 bus, ac-
the baseline of the 2005 Solar Probe STDT cepting commands and producing Consulta-

2
Solar Probe: Report of the Science and Technology
1
Solar Probe+: Report of the Science and Technology Definition Team, NASA/TM—2005–212786, National
Definition Team, Southwest Research Institute, San Aeronautics and Space Administration, Goddard Space
Antonio, TX, in press (2008). Flight Center, Greenbelt, MD (2005).

2-1
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

tive Committee for Space Data Systems Accommodation of the payload suite is cov-
(CCSDS) packets ready for final processing ered further in Section 3.
by the spacecraft for telemetry to the ground.

2-2
3.0 MISSION IMPLEMENTATION

3.0. MISSION IMPLEMENTATION source and accommodation requirements for


Throughout the Solar Probe+ study, an en- this new concept, developed a new orbit ge-
gineering team from The Johns Hopkins Uni- ometry and concept of operations, and identi-
versity Applied Physics Laboratory (APL) and fied risk mitigation measures and cost savings
Glenn Research Center has worked closely options for the new concept. The STDT en-
with the Science and Technology Definition couraged the engineering team to draw on
Team (STDT) to define the technical mission their own extensive internal experience with
implementation described in this section. The space missions, and to examine external ideas
ground rules for implementation defined by and solutions, in order to develop the most
NASA for this study are as follows: detailed and technically complete engineering
• Preserve Solar Probe science to the maxi- study possible.
mum extent possible Our approach has been to examine the 2005
• Power the spacecraft with a non-nuclear Solar Probe mission implementation in light of
power source the NASA ground rules and science require-
• Develop a mission with duration of less ments flow-down and only address areas of
than 10 years the previous implementation that have
• Launch in 2015 changed. For example, the previous study in-
• Maintain mass and power margins of at cluded design detail for the avionics suite.
least 35% Only minor changes to this suite are needed to
• Keep total mission cost below $750M meet the requirements for the Solar Probe+
In two full STDT meetings and several concept; the avionics description in Section
smaller discussions, the STDT and engineer- 3.9 focuses on the changes to the subsystem to
ing team members reexamined the science and respond to revised requirements rather than
measurement objectives from previous Solar including a detailed design description. Other
Probe studies 1,2,3,4,5 in light of the current areas where the changes from the 2005 Solar
study ground rules, defined instrument re- Probe study are more significant include cor-
respondingly more description.
1 Key requirements that flow down from the
Gloeckler, G., et al., Solar Probe: First Mission to the
Nearest Star, Report of the NASA Science Definition study ground rules and science objectives are
Team for the Solar Probe Mission, The Johns Hopkins summarized below.
University Applied Physics Laboratory, Laurel, MD
Mission Requirements
(1999).
2
Solar Probe: An Engineering Study, prepared by The • At least three orbits with perihelion distance
Johns Hopkins University Applied Physics Laboratory, less than 10 RS
in partnership with the Jet Propulsion Laboratory, • A Sun–spacecraft–Earth encounter geome-
under contract NAS5-01072, Laurel, MD (November try that supports simultaneous Earth-based
12, 2002). observations to support Solar Probe+ obser-
3
Solar Probe: Report of the Science and Technology
Definition Team, NASA/TM—2005–212786, National vations
Aeronautics and Space Administration, Goddard Space • Minimum perihelion orbits achieved within
Flight Center, Greenbelt, MD (2005). 10 years
4
Solar Probe Risk Mitigation Study, prepared by The • Return of full data collected in a solar en-
Johns Hopkins University Applied Physics Laboratory, counter in each orbit
2006 Mid-Year Report.
5
Solar Probe Thermal Protection System Risk Mitigation Spacecraft Requirements
Study: FY 2006 Final Report, prepared by The Johns • Survive solar intensity during perihelion
Hopkins University Applied Physics Laboratory under
(~510 Suns)
Contract NAS5-01072, Laurel, MD (November 30, 2006);
and ITAR-Restricted Annex (September 17, 2007). • Provide reliable power over the distance
range of 0.044–1 AU

3-1
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

• Protect instruments and spacecraft systems Sun portions of the orbit, achieved through ad-
from dust environment near the Sun aptation of previous Thermal Protection System
• Provide large total science data return (~128 (TPS) designs. The third challenge was to de-
Gbits per orbit) velop a lightweight spacecraft concept that fits
• Accommodate significant payload mass within the performance of current launch vehi-
(~50 kg) and power (57 W) cles. Solar Probe+ is planned for an Atlas V 551
• Provide science boom for magnetometer launch with positive mass margin, yet retains
and plasma-wave search coils compatibility with the Delta IVH for risk mitiga-
• Provide actuations of instruments and anten- tion. Each section below includes a summary of
nas for proper placement and field of view trade studies conducted in the development of
(FOV) orientations during science collection this concept as well as a description of trade
These requirements differ from previous So- studies and analyses to be conducted early in the
lar Probe studies in that the science objectives Solar Probe+ program.
are satisfied by repeated passes through the The result of this study is a technically
near-equatorial coronal regions rather than one feasible and affordable mission with accept-
to two passes through the region using a polar able risk that can survive in the unique ther-
approach, as described in Section 2. The new mal and dust environment near the Sun and
mission concept results in several changes to that will fully achieve all of the Solar Probe+
mission and spacecraft requirements that offered science objectives. The sections that follow
opportunities for simplification of aspects of the describe this baseline Solar Probe+ mission.
mission. For instance, the 2005 Solar Probe Additional supporting material is presented in
study identified real-time science data downlink the Appendices.
as a mission requirement to reduce the risk of 3.1. Baseline Mission Design
data loss given that only two opportunities for 3.1.1. Mission Design Overview
science collection exist in that mission concept. The engineering team conducted studies of
The Solar Probe+ mission concept achieves the trajectories to develop a feasible mission con-
same risk mitigation by using more data collec- cept for Solar Probe under NASA’s new direc-
tion opportunities and a softer walk-in to the tion and guidelines. In order to meet these re-
critical region that allows an opportunity to quirements, significant change is required to the
practice for the required encounters and to re- Solar Probe mission concept developed during
peat measurements interrupted by anomalies. the last several Solar Probe study efforts, the last
This change in approach allows the use of a one being the 2005 Solar Probe: Report of the
simpler store-and-downlink concept of opera- Science and Technology Definition Team.6 Tra-
tions where data collection and data downlink jectory design is the key that defines the scope
are decoupled. Each section below includes a of the mission and determines how the mission
description of these changes from the 2005 Solar will be implemented, what can be accomplished
Probe concept. for science, and how much it will cost. A wide
These requirements, and the new mission de- range of trajectory options were explored with
sign, led to three major challenges addressed by various mission scenarios. Representative trajec-
the engineering team. The first challenge was to tory design options were presented to the STDT
develop a power system concept not centered on and discussed between the APL engineering
the use of radioisotope thermoelectric generators
(RTGs), which led to an extensive trade study 6
Solar Probe: Report of the Science and Technology
on systems able to generate power near the Sun Definition Team, NASA/TM—2005–212786, National
as well as at aphelion. The second challenge was Aeronautics and Space Administration, Goddard Space
to survive the thermal environment of the near- Flight Center, Greenbelt, MD (2005).

3-2
3.0 MISSION IMPLEMENTATION

team and the STDT science team. After discus- design also has several advantages in mission
sions and comparison, a baseline trajectory de- implementation:
sign was selected from six alternative trajectory • Perihelion gradually decreases to 9.5 RS,
design options, including a Jupiter gravity assist lowering mission risk
(JGA) trajectory design. Details of this trade • Aphelion is less than 1 AU, ensuring suffi-
study are given in Section 3.1.9. cient solar power without the need of RTG
The baseline mission design meets the • Minimum perihelion of 9.5 RS reduces the
above-listed mission requirements and program solar flux and corresponding thermal envi-
constraints. The baseline trajectory design pre- ronment, allowing use of photovoltaics for
sents a new mission concept with more thor- power generation through the entire orbit
ough and repeated exploration of the near-Sun • Avoidance of JGA simplifies power system
region: and allows for multiple solar encounters
• Launch in May 2015, with total mission du- 3.1.2. Launch
ration of 6.9 years The baseline mission design calls for Solar
• First solar pass at ~0.16 AU (35 RS) in 3 Probe+ to be launched in May–June 2015 dur-
months after launch ing a 20-day launch period from May 21
• 24 solar encounters over 6.9 years, 19 en- through June 9, 2015. On each day, up to 30
counters within 20 RS, 961 hours within 20 minutes is planned for the launch window. Pre-
RS region liminary trajectory analysis shows the launch
• Minimum perihelion at 9.5 RS time for May 21, 2015, as 06:10 UTC. As with
• Baseline three perihelia at 9.5 RS, occurring other interplanetary missions, the spacecraft
every 88 days will be placed into a low-Earth parking orbit
• More orbits are possible in extended mis- after lift-off from the launch pad by the launch
sion, requiring no orbit maintenance vehicle, coasting in the parking orbit for ~34
• Frequent visits at the heliosphere region minutes, and then will be injected into the de-
over 6 years, providing extensive science sired heliocentric trajectory. Candidate launch
measurements over a half solar cycle vehicle and upper stage are described in Sec-
In addition to offering more science oppor- tion 3.1.7.
tunity as described in Section 2, the trajectory 3.1.3. Baseline Mission Trajectory
The baseline tra-
jectory uses seven
Venus flybys and no
deep-space maneu-
vers to reach a mini-
mum perihelion of
9.5 RS in 6.4 years in
October 2021, as
shown in Figure 3.1-
1. The baseline mis-
sion will end after
three orbits with at
the minimum perihe-
lion of 9.5 RS. From
launch to the end of
mission, the baseline
trajectory consists of

Figure 3.1-1. Baseline Solar Probe+ trajectory.

3-3
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

Figure 3.1-2. Solar distance profile for baseline trajectory.

24 solar orbits whose perihelia gradually de- helion orbits) was computed with a full gravity
crease, from 35 RS down to 9.5 RS. Figure 3.1- field model including the Sun and all planets.
2 plots the solar distance over the entire mis- The results of the trajectory were verified by
sion and the perihelion distances of the 24 solar different mission design software used for tra-
orbits. The first perihelion occurs only 3 jectory design, flyby targeting, and trajectory
months after launch. The Sun–Earth–probe correction maneuver (TCM) planning for the
(SEP) angle and Sun–probe–Earth (SPE) angle ongoing interplanetary missions [New Hori-
as functions of time are plotted in Figure 3.1-3. zons, MESSENGER (Mercury Surface, Space
A completely integrated trajectory of the Environment, Geochemistry, and Ranging),
baseline mission from launch through the end and STEREO (Solar Terrestrial Relations Ob-
of the nominal mission (after three 9.5-RS peri- servatory)] managed by APL for NASA.

3-4
3.0 MISSION IMPLEMENTATION

The Earth positions


at the three 9.5-RS
perihelion orbits
are shown in Figure
3.1-5 as an exam-
ple. The range be-
tween the space-
craft and Earth is
illustrated in Figure
3.1-6 for those and
all other orbits.
As indicated in
Figure 3.1-2b and
Table 3.1-1, Solar
Probe+ will spend a
significant amount
of time in the near-
Sun region, cover-
Figure 3.1-3. Sun–Earth–probe (SEP) angle and Sun–probe–Earth (SPE) angles ing more than half
for the baseline Solar Probe+ trajectory. of the 11-year solar
cycle, with three to
3.1.4. Solar Encounters
four encounters per year. In 24 orbits over the
Figure 3.1-4 shows a solar encounter trajec- 6.9-year mission duration, the perihelion dis-
tory with perihelion at 9.5 RS, viewing from tance decreases from 35 RS to 9.5 RS. In 21 of
the north pole of the Sun. The time duration the orbits, the distance from the center of the
within 20 RS is 57 hours. The perihelion ve- Sun will be less then 30 RS, and in 19 of the
locity with respect to the Sun will approach orbits, the distance from the center of the Sun
200 km/s. Trajectories of other perihelion or- will be less than 20 RS. The total accumulated
bits are similar with a
greater distance and a
slower speed.
At each solar encoun-
ter, the geometry of Earth
with respect to the en-
counter trajectory is of
great interest to science,
because of the intent to
conduct observations re-
motely from the Earth
while the onboard in situ
measurements are per-
formed. With 24 solar en-
counters, this mission de-
sign allows for various
observation geometries
from the Earth when Solar
Probe+ is passing in front
of the Sun at perihelion. Figure 3.1-4. Near-Sun trajectory for baseline mission.

3-5
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

Table 3.1-1. Times in various solar dis-


tance regions.
Solar Time within
Pass 30 RS 20 RS 15 RS 10 RS
Number (hr) (hr) (hr) (hr)
4 67
5 67
6 105 10
7 105 10
8 109 50
9 109 50
10 108 55 23
11 108 55 23
12 108 55 23
13 108 55 23
14 108 55 23
15 108 55 23
16 108 55 23
17 105 57 33
Figure 3.1-5. Location of Earth during solar encounters with 18 105 57 33
perihelion below 10 RS. 19 105 57 33
20 105 57 33
time of Solar Probe+ within the regions of less 21 105 57 33
than 30 RS, 20 RS, 15 RS, and 10 RS is summa- 22 102 57 36 10
rized in Table 3.1-1. 23 102 57 36 10
3.1.5. Comparison with 2005 Solar Probe 24 102 57 36 10
The 2005 Solar Probe mission concept used Total 2149 961 434 30
a JGA trajectory to achieve a pole-to-pole near-
Sun orbit with perihe-
lion at 4 RS. The
aphelion of the final
heliocentric orbit was
5.5 AU, about the dis-
tance of Jupiter. Two
solar encounters were
baselined and were
separated by the or-
bital period of 4.6
years, which resulted
in mission duration of
8.8 years, ending after
the second solar en-
counter. The first Sun
encounter would have
occurred 4.1 years
after launch in Octo-
ber 2014. Figure 3.1-6. Distance of the spacecraft from Earth.

3-6
3.0 MISSION IMPLEMENTATION

Figure 3.1-7. Comparison of 2008 Solar Probe+ and 2005 Solar Probe trajectories.

Constrained by solar power at aphelion, the 24 solar encounters provide various perihelion
current Solar Probe+ mission concept ex- viewing geometries from Earth, allowing for
cludes a fly out to Jupiter for gravity assist. simultaneous Earth observations on different
Instead, an inward route is taken to reach parts of the Sun, as was previously illustrated
minimum perihelion at 9.5 RS through the use in Figure 3.1-5. Reducing the perihelion
of multiple gravity assists from Venus flybys. gradually in multiple orbits results in lower
A side-to-side comparison of the near-Sun en- operational risk than the JGA trajectory ap-
counter between the current and the 2005 con- proach, which directly lowers the perihelion to
cept is shown in Figure 3.1-7. More compari- 4 RS at the first near-Sun orbit.
sons are summarized in Table 3.1-2. 3.1.6. ΔV Budget and Navigation
The first in situ measurements within the
The baseline mission trajectory requires no
heliospheric region will occur only 3 months
deep-space maneuvers from launch to final he-
after launch. Starting the mission’s science
liocentric orbit with perihelion at 9.5 RS, and all
investigation right after launch and continuing
for 6+ years is unprecedented for planetary the Venus flybys are not powered. Therefore, no
missions. With three to four solar encounters deterministic ΔV is required throughout the en-
per year, the heliosphere will be thoroughly tire mission. Currently, a total of 190 m/s is allo-
and timely probed. Although perihelion is not cated for the ΔV budget to cover the injection
as close as 4 RS and the orbit is not polar, the errors at launch, the targeting of seven Venus
total time within 20 RS is 10 times that ob- flybys, and momentum control. Launch error
tained with the 2005 concept. In addition, the correction includes significant ΔV for all launch
Table 3.1-2. Comparison of 2008 Solar Probe+ and 2005 Solar Probe mission designs.
2008 Solar Probe+ 2005 Solar Probe
Minimum Perihelion 9.5 RS 4 RS
Inclination 3.4° from ecliptic 90° from ecliptic
Number of Solar Encounters 24 2
Total Time within 20 RS ~ 961 hours ~ 96 hours
Time Between Perihelia 88–150 days 4.6 years
Time from Launch to First Perihelion 3 months 4.1 years
Mission Duration 6.9 years 8.8 years
Aphelion 1 AU 5.5 AU

3-7
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

Table 3.1-3. Preliminary ΔV/fuel budget. Table 3.1-4. Venus flybys and planned TCMs.
ΔV Event Date
Usage Event (m/s) Launch 05/21/15
Trajectory Correction Launch Error TCM 1 06/05/15
Maneuver Correction 90 TCM 2 07/05/15
Venus Flyby 1 8 Venus Flyby #1 07/19/15
Venus Flyby 2 12 TCM 3 09/20/16
Venus Flyby 3 12 Venus Flyby #2 10/11/16
Venus Flyby 4 12 TCM 4 01/25/17
Venus Flyby 5 12 TCM 5 04/16/17
Venus Flyby 6 12 Venus Flyby #3 04/26/17
Venus Flyby 7 12 TCM 6 09/10/17
Momentum TCM 7 11/27/17
Attitude Maneuver Management 6 Venus Flyby #4 12/07/17
Margin 14 TCM 8 07/11/18
Total ΔV 190 Venus Flyby #5 08/01/18
Usable Propellant (kg) 52.3 TCM 9 08/09/19
Residual (kg) 0.3 TCM 10 05/15/20
Pressurant (kg) 0.1 Venus Flyby #6 06/05/20
Total Propellant Mass (kg) 52.7 TCM 11 05/23/21
Spacecraft Wet Mass (kg) 610 TCM 12 08/12/21
Venus Flyby #7 08/22/21
days. Table 3.1-3 presents the ΔV budget for Perihelion 10/10/21
Solar Probe+. The budget for momentum man-
agement includes an estimate of the frequency hance the orbit determination (OD) accuracy.
of momentum dumping based on the expected From launch (L) to L+2 weeks, there will be
solar pressure imbalance and also impulses from continuous Deep Space Network (DSN) track-
dust impacts. In this design, the propellant tank ing, followed by five 10-hour passes per week
is not full, and extra tank capacity is not in- from L+2 to L+4 weeks. At each Venus (V)
cluded in the margin cited in Table 3.1-3. A de- flyby there will be five 10-hour passes per
tailed navigation analysis associated with ΔV week from V–5 weeks to V–1 week and one
requirement refinement will be conducted in 10-hour pass per day from V–1 to V+1 week.
Phases A, B, and C/D. 3.1.7. Launch Vehicle and Third Stage
To simplify spacecraft attitude management Because of cost restrictions, the candidate
and maintain the thermal protection attitude, launch vehicle currently under consideration is
we have designed the Solar Probe+ mission to the Atlas V 551 with a STAR-48BV third
avoid burns during solar encounters. All stage built by ATK. Solar Probe+ retains
TCMs will be performed near aphelion, at dis- compatibility with the Delta IVH as a backup
tances greater than 0.5 AU. The TCMs for vehicle and with the Boeing STAR-48B as a
Venus flyby targeting will be placed before backup third stage. Based on the lift capabil-
the Venus flyby. Table 3.1-4 presents the ity provided in the Atlas Launch Systems Mis-
schedule for Venus flybys and TCMs. sion Planner’s Guide 7 and performance for
Navigation has no special requirements and the STAR-48BV provided by ATK, the esti-
will be straightforward. Optical navigation is mated launch mass is 610 kg, assuming a 20-
not required. The radiometric Doppler range
and range rate data will be used for spacecraft
trajectory determination. Delta differential
7
one-way range (Delta-DOR) tracking data Atlas Launch Systems Mission Planner’s Guide, Rev.
10a, Lockheed Martin Corporation (January 2007).
may be used before the Venus flybys to en-

3-8
3.0 MISSION IMPLEMENTATION

launch vehicle. The first flight is


scheduled for September of 2009
as part of the TacSat-4 mission.
Planned Phase A activities in-
clude significant engagement
with the NASA launch services,
the launch vehicle provider, and
the third stage provider for final-
ization of the launch configura-
tion, refinement of the lift mass
estimate, and identification and
development of mitigations for
technical risks.
3.1.8. Launch Opportunity
The synodic period of Venus is
584 days, which results in a
launch opportunity occurring ap-
proximately every 19 months.
Figure 3.1-8. Launch vehicle performance. Besides the baseline mission
launch time of 2015 suggested by
the NASA guidelines for this
day launch period. Launch vehicle perform-
study, launch opportunities before 2015 and be-
ance is shown in Figure 3.1-8. yond in a reasonable timeframe for near-term
The ATK STAR-48BV stage is a thrust- mission planning were analyzed. Details of the
vector-controlled motor offering a simple con- trajectory design for the four launch opportuni-
trol system and higher performance than the ties from 2013 through 2018 are summarized in
spinning STAR-48B, which has been used in Table 3.1-5, where the 2015 baseline launch also
programs such as New Horizons. The STAR- is included for comparison. The C3 requirement
48BV uses a loaded motor
case from the flight-proven Table 3.1-5. Launch opportunity summary.
Launch Opportunity 2013 2015 2017 2018
STAR-48B, with nozzle
Launch Date 10/25/13 05/31/15 01/10/17 08/09/18
design qualified for the Optimum C3 170.0 155.6 177.2 157.0
Conestoga program and a (km2/s2)
newly developed thrust Trajectory Flybys 7 Venus 7 Venus 7 Venus 7 Venus
vector actuator (TVA) con- Deep Space Ma- None None None None
trol system currently being neuver (ΔV)
qualified to support vector- Max. aphelion (AU) 1 1 1 1

able nozzles across the Final orbit Perihelion (RS) 9.5 9.5 9.5 9.5
Aphelion (AU) 0.73 0.73 0.72 0.73
STAR product line. The Inclination from 3.4 3.4 3.4 3.4
nozzle and thrust vector ecliptic (°)
control system will be used Orbital period (day) 88 88 87 88
on a STAR-37FMV in Near-Sun Total no. of solar 24 24 24 24
mid-2008, and the STAR- Pass passes (<0.2 AU)
48BV upper stage currently Timeline Launch to minimum
perihelion (year)
6.4 6.4 6.4 6.4

is under contract and Mission duration 6.9 6.9 6.8 6.9


scheduled for a first flight (including three 9.5-
as part of the Minotaur IV RS passes) (year)

3-9
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

Figure 3.1-10. Trajectory Option 1: Solar distance


profile.

launch vehicles. Therefore, a fundamental re-


Figure 3.1-9. Trajectory Option 1: Trajectory design. quirement of the 2005 study to perform two
shown is for the listed launch date chosen near solar encounters cannot be met with the non-
the optimal launch time of that launch opportu- nuclear restriction. A major activity of the cur-
nity. Higher C3 will be required if a launch win- rent Solar Probe+ study has been to develop
dow, for example, a 20-day launch period, needs alternative mission designs (including a single-
to be reserved. Trajectory analyses for launch pass JGA mission) that will accomplish most of
opportunities in more extended timeframes in the science objectives from the 2005 study
the future and with other types of trajectories while meeting the constraints as described in
included are planned for Phase A study. previous sections.
3.1.9. Mission Design Trade Study
The 2005 Solar Probe concept cannot
be implemented without significant
changes to mission design under the new
guidelines for the current study. In particu-
lar, NASA has directed that this study de-
velop a concept without use of a nuclear
power source. This restriction forces the
use of solar arrays for power generation
outside the solar encounter. Arrays suffi-
cient to generate power at Jupiter, even if
the spacecraft enters a lower-power “hi-
bernation” state, are too large to protect
during the solar encounter. In this concept,
portions of the primary solar arrays must
be ejected before entering the near-Sun
regions of interest and cannot be used to
power the spacecraft for a second orbit. In
addition, energy storage is impractical to
implement and more massive than can be
launched into the JGA orbit with current Figure 3.1-11. Trajectory Option 2: Trajectory design.

3-10
3.0 MISSION IMPLEMENTATION

• How close to the Sun can


Solar Probe+ get without
the JGA under current im-
plementation ground rules
with currently available
launch vehicle capability?
• How high from the ecliptic
plane of the inclination can
Solar Probe+ reach with-
out the JGA?
• Is it possible to achieve
perihelion at 4 RS and incli-
nation at 90° for the solar
flyby with the gravity as-
sists from the inner planets?
On September 24, 2007, the
engineering team gave the first
trajectory study report at the
STDT meeting. Representa-
tive trajectories of different
Figure 3.1-12. Trajectory Option 2: Solar distance profile.
trajectory types, named Tra-
3.1.9.1. Trajectory Study jectory Option 1 (Figures 3.1-
9 and 3.1-10), Trajectory Option 2 (Figures 3.1-
The trajectory study explored various mis- 11 and 3.1-12), Trajectory Option 3 (Figures
sion scenarios and trajectory options using 3.1-13 and 3.1-14), and Trajectory Option 4
gravity assists from the inner planets: Mars, (Figures 3.1-15 and 3.1-16) were presented as
Earth, Venus, and Mercury, or their combina- potential candidate trajectories for Solar Probe+.
tions in search for the alternative Solar Probe+ These trajectory options were selected from the
mission trajectory and answers to the various trajectories analyzed since the Solar
following questions: Probe+ study was initiated. Findings from these

Figure 3.1-13. Trajectory Option 3: Trajectory design.

3-11
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

Figure 3.1-14. Trajectory Option 3: Solar distance


profile.

Figure 3.1-17. Trajectory Option 6: Trajectory de-


sign.
preliminary trajectory analyses were described
in the summary remarks of the presentation:
• To achieve a solar pass with perihelion at 4
RS and 90° inclination requires JGA flyby.
• The 2015 launch JGA trajectory (Trajectory
Option 1) can achieve the same Sun en-
counter geometry as the 2005 study’s base-
line trajectory, but engineering issues limit
spacecraft functionality to one orbit only.
• Preliminary trajectory analysis shows that
using Venus and Earth gravity assists (Op-
tions 2, 3, and 4) can get perihelion down
Figure 3.1-15. Trajectory Option 4: Trajectory de- to ~0.05 AU and inclination up to ~38°
sign. from the ecliptic plane.
After discussion, the STDT down-selected
the trajectory options to two mission concepts,
Option A and Option B, for further study of
implementation feasibility:
• Option A: A JGA trajectory with a single
flyby of perihelion at 4 RS
• Option B: A new trajectory consisting of
three orbits of perihelion less than 10 RS
and mission duration less than 10 years
Trajectory Option 3 was not pursued further
because the STDT determined that a close
perihelion outweighed higher latitudes at far-
Figure 3.1-16. Trajectory Option 4: Solar distance ther distances.
profile.

3-12
3.0 MISSION IMPLEMENTATION

Two trajectory options, Trajectory Option 5


and Trajectory Option 6 (Figures 3.1-17 and
3.1-18), were developed for mission concept
Option B. Ultimately, Option 5 was chosen as
the mission baseline. Table 3.1-6 gives a com-
parison of the trajectory options considered.
3.1.9.2. Rationale for Selection
The primary science considerations for se-
lections were (i) minimum perihelion must be
below 10 RS and (ii) minimum perihelion
Figure 3.1-18. Trajectory Option 6: Solar distance must be achieved in less than 10 years in order
profile. to fit within the cost constraints as established
for this study. In addition, the study guidelines
Table 3.1-6. Comparison of trajectory options.
Trajectory Option Option 1 Option 2 Option 3 Option 4 Option 5 Option 6
Launch Date 11/20/15 05/30/15 05/30/15 09/06/14 05/21/15 05/24/15
C3 (km2/s2) 114 156 156 108 158 128
Trajectory Flybys 1 Jupiter 5 Venus 9 Venus 3 Earth, 7 Venus 1 Earth,
7 Venus 9 Venus
Deep Space Ma- None None None 1 (232 m/s) None 1 (397 m/s)
neuver (ΔV)
Max Aphelion (AU) 5.56 1 1 2.29 1 1.19
Final Orbit Perihelion (RS) 4 11.8 39.8 9.5 9.5 9
Aphelion (AU) 5.56 0.75 0.725 0.73 0.73 0.73
Inclination from 90 3.4 37.9 2.9 3.4 3.4
Ecliptic (°)
Orbital Period 4.6 year 94 days 112 days 88 days 88 days 88 days
Timeline Launch to Min. 4.1 year 3.3 year 2.1 year 10 year 6.4 year 9.5 year
Perihelion
Mission Duration 4.5 year 3.8 year 5.8 year 10.5 year 6.9 year 9.95 year
Orbits Total no. of solar 1 12 18 16 24 21
encounters (<0.2 AU)
Pros Pole-to-pole Short mis- Short mis- Good perihe- Good perihe- Good perihe-
solar flyby sion duration; sion dura- lion distance; lion distance; lion distance;
at 4 RS multiple, tion; multiple, multiple, multiple multiple,
frequent frequent frequent frequent solar frequent
solar flybys; solar flybys; solar flybys; flybys; short solar flybys;
aphelion aphelion moderate C3 mission moderate C3
<1 AU <1 AU duration; no
deep space
maneuver;
aphelion
<1 AU
Cons Single solar Low inclina- Large Long mission Low inclina- Long mission
flyby; great tion; high C3 perihelion duration, tion; high C3 duration,
aphelion distance; requiring requiring
distance; high C3 deep space deep space
long cruise; maneuver maneuver
long orbit
period

3-13
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

require significant science return. Finally, the spacecraft operations for that event are consis-
chosen mission must support a spacecraft im- tent across orbits where flyby events occur. This
plementation that can fit in the launch capabil- mission concept lends itself well to modular op-
ity of the trajectory with acceptable margins. erations with predefined sequences that are con-
From these criteria, we were able to choose sistent throughout the mission. Solar Probe+ will
the optimal mission among those considered. take advantage of the event-driven nature of the
Options 2 and 3 were eliminated immedi- mission timeline by modularizing operations for
ately because of failure to achieve acceptable events, allowing for lower risk in operating the
minimum perihelion. Option 4 also was elimi- spacecraft and for savings in planning and op-
nated because this mission does not reach erations costs. Any differences from orbit to or-
minimum perihelion within 10 years. Option bit in modularized operations will be captured
6, although it achieves the minimum perihe- through parameters that can be set independ-
lion within 10 years, requires more planetary ently for an event in each orbit. For example,
flybys and much longer mission duration than trajectory correction maneuvers (TCMs) are
Option 5 and therefore was eliminated to re- events that will occur in each orbit, yet the spe-
duce cost and complexity. cific details of each burn change with each orbit.
Spacecraft conceptual designs were devel- Parameterizing the command sequence allows
oped for the remaining options. Trajectory the team to fully develop a tested set of com-
Option 1 includes a JGA, while Option 5 re- mands once and modify that set through pa-
mains within 1 AU for the mission. This dif- rameter updates for each orbit instead of devel-
ference in trajectory led to a major difference oping a full set of commands for each orbit. The
in power subsystem concept for the two op- STEREO mission has successfully used this
tions. Option 1 requires the use of a large solar methodology for operations.
array that cannot be completely retracted be- Solar Probe+ science instruments do not re-
hind the TPS for the solar encounter. There- quire specific pointing of the spacecraft be-
fore, the extra solar panels associated with yond ram-pointing of certain instruments, nor
travel outside 1 AU must be ejected, limiting does the payload depend on mechanisms oper-
the science opportunity to a single solar pass. ated by the spacecraft beyond single-shot de-
Option 5 can be powered throughout the mis- ployable antennas and boom. Measurement
sion without this restriction, and repeated solar sequences are self-contained and can be car-
encounters give significantly more science ried out simultaneously with little or no im-
data. Given that equivalent scientific objec- pact on spacecraft or other instruments. This
tives can be accomplished with these two mis- leads to a concept for science measurement
sion options, Trajectory Option 5 was selected planning that is decoupled from spacecraft op-
as the baseline mission. erations. The spacecraft design fully supports
3.2. Mission Concept of Operations the decoupled operations concept by providing
3.2.1. Overview sufficient resources for payload operations.
The baseline mission consists of 24 solar or- The Solar Probe+ operations concept takes
bits, all of which are very similar. Operations full advantage of this decoupling to simplify
during an orbit are distinguished by events that operations, resulting in reduced risk, more ef-
occur on the orbit timeline, each of which is a ficient use of operations staff and resources,
discrete set of actions and conditions. The opera- and reduced cost.
tions for events are similar for all orbits. For ex- 3.2.2. Launch and Early Operations
ample, seven orbits include a Venus flyby. If we Solar Probe+ will be launched from Cape
define that portion of an orbit of 40 days around Canaveral within a 20-day launch window on
a Venus encounter as a Venus flyby event, an Atlas V 551 with a STAR-48 third stage.

3-14
3.0 MISSION IMPLEMENTATION

The nominal launch date is May 21, 2015. Ex- when downlinking data, transition to science
cept for launch vehicle dispersion and naviga- for the near-Sun pass, science collection, and
tional corrections, the entire ΔV necessary for transition back to data downlink.
achieving orbit will be supplied by the launch Data Downlink. The default activity for the
vehicle. spacecraft outside of all other events is to
The spacecraft will be three-axis-stabilized downlink data collected from the previous sci-
by the launch vehicle and third stage for the ence event. Contacts using Ka-band and a DSN
entire launch ascent and third stage flyout. No 34-m disk on the HGA will be conducted for an
despin maneuver is needed after third stage average of 10 hours per day. All data from one
separation. After separation, Solar Probe+ will science event will be downlinked before the
be oriented in Sun-pointing attitude. The low- next science event occurs.
gain antenna (LGA) will be the primary an- When not in contact, the spacecraft main-
tenna for communications until the high-gain tains the default pointing with ram-pointing
antenna (HGA) is deployed and checked out. instruments in the correct attitude. When in
The first Venus flyby will occur approxi- contact, the spacecraft will maintain Sun-
mately 6 weeks after launch. The time from pointing directionality while rolling the space-
launch until first encounter will be devoted to craft to allow the HGA to point to Earth as
spacecraft systems checkout and preparation needed. No science measurements requiring
for that event. Time between the first Venus specific pointing will be collected during con-
flyby and the first solar encounter will be de- tacts; however, low rate science on some in-
struments will be conducted between contacts.
voted to science instrument checkout, encoun-
Venus Flyby/TCM. Seven Venus flybys are
ter preparations, and the first changeover from
used to lower perihelion to below 10 RS. The
primary power to secondary power. During
period extending from 30 days before to 10
the first 7 days after launch, DSN coverage
days after a Venus encounter are designated a
will be continuous. After this period, DSN
Venus flyby event. Except for early operations
coverage will consist of approximately five 8-
before the first Venus flyby, all TCMs are ex-
hour contacts per week. pected to occur within the period leading up to
3.2.3. Mission Events Venus flyby or at aphelion. Operations during
Figure 3.2-1 shows a typical orbit for Solar this period are limited to
Probe+ that includes all events to be used ex- • Downlink of science data from previous
cept for early operations, which were dis- solar encounter.
cussed in Section 3.2.2. This orbit starts at • Contacts to allow analysis, execution, and
aphelion, includes a Venus flyby, periods evaluation of a navigation burn planned be-

Figure 3.2-1. Typical Solar Probe+ orbit indicating events and operational modes.

3-15
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

fore the encounter and a burn after encoun- downlinking of science data to execute the
ter to correct any residual errors. mission science:
• Monitoring of spacecraft health and safety • Maintaining safe orientation of spacecraft at
through encounter. all times
Science. At ~0.25 AU (~5 days before peri- • Generating sufficient power for spacecraft
helion), Solar Probe+ will transition from the subsystems and instruments
standard configuration to science configuration. • Transitioning from one mode to another as
While maintaining the correct Sun-pointing, the described above
primary solar arrays are retracted inside the • Conducting Venus flybys to lower perihe-
umbra formed by the TPS, and the spacecraft is lion and trajectory corrections as needed
powered only by the secondary solar arrays, • Collecting and storing data from the pay-
which are fully extended. As the spacecraft load and downlinking data to the ground
nears perihelion, the secondary solar arrays are • Receiving and executing payload and
retracted behind the TPS shield to maintain spacecraft commands
constant power. Science data collection is car- Unlike many missions such as New Hori-
ried out throughout this event; however, the zons where planning observational sequences
instruments in use and the measurement rates requires a high degree of integration between
vary, with highest use near perihelion. All data instruments and spacecraft to deconflict re-
are stored on each solid-state recorder (SSR) sources and coordinate issues such as space-
for downlink outside the science event; how- craft attitude, the Solar Probe+ mission is de-
ever, a low-rate link is maintained through the signed for simple operations. Instrument opera-
low-gain antenna (LGA) for command uplink
tions do not affect spacecraft attitude or operat-
and housekeeping telemetry downlink except
ing mode, nor are sequences of pointing ma-
during solar conjunction. During science col-
neuvers needed to perform science measure-
lection, instruments do not require off-pointing
ments. Therefore, spacecraft and payload op-
of the spacecraft or specific pointing about the
erations are decoupled. Detailed sequencing of
Sun–probe axis with the exception of a small
number of instruments that require ram- instrument commands will be produced by the
pointing. After Solar Probe+ passes perihelion, instrument teams themselves and provided to
the secondary solar arrays are slowly extended mission operations for upload after testing. As
to maintain constant power, followed by transi- with the modular nature of spacecraft opera-
tion to the downlink configuration when the tions, this decoupling of payload planning and
spacecraft is farther than 0.25 AU from the Sun mission operations simplifies the science and
(~5 days after perihelion). mission operations interface, reduces risk to the
mission, and provides a cost-effective means of
3.2.4. Operations
accomplishing the science objectives. The de-
The mission operations concept for Solar coupled operations concept has been success-
Probe+ is based on the modular nature and fully used in missions such as STEREO.
similarity of events across orbits as described Another aspect of Solar Probe+ mission op-
above. Operations for recurring events are erations designed to reduce costs is the unat-
executed using predefined sequences that are tended mode of operations within the Mission
repeated with only small changes from one Operations Center (MOC). Outside of critical
orbit to the next. This allows significant reuse events such as TCMs, flybys, and encounters,
in operations planning and tasking, resulting the MOC is capable of supporting the downlink
in lower risk to the mission at lower cost than of science data during contacts through the
for mission with less repetitive operations. DSN without the need for staffing within the
Spacecraft operations are devoted to func- MOC. The ground system monitors spacecraft
tions needed to support the collection and

3-16
3.0 MISSION IMPLEMENTATION

health and status and supports remote notifica- 3.3.2. Radiation


tion of predefined alarm conditions to Mission The radiation environment is dominated by
Operations Team members. Although the noti- solar energetic protons that are responsible for
fication comes with an initial indication of se- total dose damage in components and galactic
verity, the ground system is designed to allow cosmic rays that are the source of most single-
team members to access additional information event effects experienced by spacecraft. In
remotely. Depending on the issue, the web- addition, protons produced in solar flares also
based planning and scheduling system may al- will contribute to the single-event effects rate
low the situation to be resolved remotely. Unat- during periods when the Solar Probe+ space-
tended operations is another concept success- craft experiences this elevated environment.
fully used on the STEREO mission. Most of the Solar Probe+ mission will occur
3.3. Mission Environment during solar minimum conditions. During this
Solar Probe+ must be able to survive and op- period of the solar cycle, the total dose is neg-
erate under extreme environmental conditions, ligible, and significant radiation damage will
which present significant challenges for the en- occur only in the later years when the Sun be-
gineering design of the spacecraft. As the space- comes active during the solar maximum pe-
craft approaches and flies past the Sun, it will be riod. The total dose requirement for Solar
exposed to intense solar flux and bombardment Probe+ is 30 krad behind 100 mils of alumi-
by particles from the circumsolar dust cloud. In num shielding, based on the 95% worst-case
addition, the effects of coronal lighting and solar Jet Propulsion Laboratory solar proton model
scintillation in the near-Sun environment must for the 2 years of maximum conditions and
be included in the design of attitude control and correcting for Sun–spacecraft distance through
telecommunications systems. the orbit as defined above. The total dose re-
quirement is the same as that of the MES-
3.3.1. Solar Flux
SENGER mission and is achievable through a
The most challenging spacecraft design driver parts screening and qualification test similar to
is the intense solar flux to which Solar Probe+ that used for MESSENGER and other recent
will be exposed. At perihelion, the flux will be programs. During the design phase, a three-
roughly 510 times that at Earth orbit. As dis- dimensional shielding analysis will be con-
cussed in detail in Section 3.6, the TPS, consist- ducted to allow mass reduction in electronics
ing of a ceramic-coated carbon–carbon (C-C) enclosures and spacecraft structure without
shield, protects the instruments and spacecraft compromising total dose survivability.
bus from direct exposure to this flux. Immedi- Outside of exposure to energetic charged
ately after launch and spacecraft separation, the particles during a solar event, the single-event
TPS will point toward the Sun, and this attitude effects environment is worst near aphelion,
will be maintained through the mission. Except where single-event effects rates are similar to
for the secondary solar arrays and the plasma- that of near-Earth missions. Solar flare expo-
wave instrument (PWI) electric field antennas, sure also is expected to be no worse than flare
instruments and spacecraft components will re- environments experienced by missions such as
side within the umbra of the TPS at all times MESSENGER and STEREO. Parts used in
during solar encounter. The only other compo- Solar Probe+ are similar to those used in pre-
nents intended to extend beyond the umbra are vious missions with equivalent or worse envi-
the primary solar arrays and high-gain antenna ronments, and the parts screening and qualifi-
(HGA), which are extended only in portions of cation program used for recent interplanetary
the orbit away from the solar encounter. missions will assure that single-event effects
requirements are met and that the system will
function as needed for the life of the mission.

3-17
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

3.3.3. Coronal Lighting Definition Team 8 and the 2006 Solar Probe
Coronal lighting near the Sun is an environ- Thermal Protection System Risk Mitigation
mental factor that can have significant conse- Study and ITAR-Restricted Annex 9 include a
quences for maintaining attitude control. Exces- cone-shaped, ceramic-coated primary heat
sive coronal lighting can increase background shield. This design produced a temperature re-
noise and degrade a star tracker’s ability to de- duction of ~200 K and was an essential enabler
tect star constellations needed to determine for the mission. The Solar Probe+ mission re-
spacecraft attitude. Although coronal lighting tains the ceramic coating but replaces the cone-
conditions can be estimated from data acquired shaped heat shield with a disc. Additionally,
by remote-sensing instruments in orbit from 1 the trajectory no longer carries the spacecraft to
AU, uncertainty about the actual lighting condi- a Jovian encounter, and it now experiences a
tions will remain until a mission near the Sun is closest approach of ~9.5 RS rather than 4 RS.
performed. Because of this uncertainty, Solar Therefore, the charging situation with respect
Probe+ uses three star trackers facing in or- to both spacecraft safety and the likelihood of
thogonal directions, a high-precision inertial disruption of data collection must be reevalu-
measurement unit (IMU), and a solar horizon ated for the new design and trajectory; prelimi-
sensor (SHS) used to detect faults in attitude nary results are given in this report. Previous
control and initiate corrective autonomous re- efforts also included an investigation of charg-
covery. The guidance and control (G&C) sub- ing mitigation and early investigations of parti-
system is described further in Section 3.12. cle tracking; such work would need to be re-
3.3.4. Solar Scintillation evaluated for Solar Probe+.
The effects of solar scintillation have been The equilibrium process of spacecraft charg-
well characterized based on mission data from ing balances the current into the spacecraft (in-
the Near Earth Asteroid Rendezvous (NEAR) cident ions and electrons) with the current out
mission as well as from the Magellan and Gali- of the spacecraft (photoemission, secondary
leo missions. During the NEAR mission, meas- electron emission, and backscattered electrons).
urable telemetry losses in the X-band downlink A potential may develop on the spacecraft rela-
were experienced around solar conjunction tive to the surrounding plasma “ground” (abso-
once the angle among the Sun, the Earth, and lute charging), or different parts of the space-
the spacecraft came within 2.3°. Although cho- craft may charge to different potentials (differ-
sen for its increased data rate, Ka-band also is ential charging). For example, in the case that
less sensitive to scintillation effects; thus, Solar part of the spacecraft experiences Sun exposure
Probe+ will use Ka-band for telecommunica- while another part is shaded, the Sun-exposed
tions, both for low-rate command and telemetry side will undergo photoemission and therefore
during the solar encounter as well as during acquire a net positive charge relative to the
high-rate science data downlink outside the so- shaded side, resulting in differential charging
lar encounter. X-band communication is base-
lined in the telecommunications subsystem de-
sign as a backup for periods when Ka-band is 8
Solar Probe: Report of the Science and Technology
unavailable because of ground station condi- Definition Team, NASA/TM—2005–212786, National
tions; however, a Phase A study will be con- Aeronautics and Space Administration, Goddard Space
ducted to determine whether a dual-frequency Flight Center, Greenbelt, MD (2005).
9
Solar Probe Thermal Protection System Risk Mitigation
system is required. Study: FY 2006 Final Report, prepared by The Johns
3.3.5. Spacecraft Charging Hopkins University Applied Physics Laboratory under
The Solar Probe spacecraft described in the Contract NAS5-01072, Laurel, MD (November 30,
2006); and ITAR-Restricted Annex (September 17, 2007).
2005 Report of the Science and Technology

3-18
3.0 MISSION IMPLEMENTATION

Figure 3.3-1. Predicted dust environment at 0.1 AU (from Mann et al.10 and Ishimoto
and Mann11).

unless the parts of the spacecraft are well flux greatly in favor of smaller particles, as illus-
grounded to each other. trated in Figure 3.3-1. However, to account for
Modeling of the charging effects was per- uncertainties in the actual circumsolar dust envi-
formed by using the charging analysis program ronment, it has been conservatively assumed
NASCAP-2K, which was created by Science that there are no collisions between particles.
Applications International Corporation (SAIC) The model employs the following assumptions:
and has the capability to model spacecraft po- • The number density of the dust in ecliptic
tentials. Details of this study are presented in orbits (within ±30° inclination) varies with
Section 3.14.1. distance as 1/r between 10 RS and 1 AU.
3.3.6. Micrometeoroid and Dust • 5% of the dust within 30° of the ecliptic is
Over the years of Solar Probe studies, a de- in retrograde orbits.
scription of the near solar dust environment has • Beyond 30° inclination, the flux is 10% that
been developed by leading solar scientists from of ecliptic orbits for particles smaller than 5
throughout the world. This description predicts μm and 5% that of ecliptic orbits for parti-
that a Solar Probe spacecraft following either a cles larger than 5 μm.
polar or ecliptic orbit would encounter dust par- • 50% of the flux at >30° inclination is in ret-
ticles ranging from submicrometer up to several rograde orbits.
hundred micrometers in diameter traveling at • All dust trajectories close to the Sun are cir-
relative speeds as high as 350 km/s. For the So- cular.
lar Probe+ trajectory, it is expected that space- • As the distance from the Sun decreases to
craft would encounter mostly small particles within 10 RS, the number density of dust
consisting of carbon and some refractory silicate particles remains constant because of dust
material with a bulk density of ~2.5 g/cm3. destruction.
Mann et al.10 estimated that significant dust- While a comprehensive study will be done as
particle collisions (see Ishimoto11) occur in the part of the Solar Probe+ effort, a preliminary
inner heliosphere that redistribute the particle comparison of the Solar Probe+ trajectory with
those from the earlier studies shows a potentially
10
Mann, I., et al., Dust near the Sun, Space Sci. Rev. larger dust exposure. For the proposed Solar
110, 269–305 (2004). Probe+ mission, the spacecraft is planned to fol-
11
Ishimoto, H., Mann, I., Modeling the number density low 24 elliptical orbits with each perihelion
distribution of interplanetary dust on the ecliptic plane
lowering from 35 RS to 9.5 RS. During one en-
within 5 AU of the Sun, Astron. Astrophys. 362, 1158–
1173 (2000). counter, the Solar Probe+ trajectory will encoun-
ter about five to six times more particles than the

3-19
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

earlier study, based primarily on the elliptical vides an overview of the baseline design, dis-
trajectory path, and the Solar Probe+ trajectory cusses the fault management approach, and
includes 10 times more orbits. Therefore, the summarizes the Solar Probe+ mass and power
total dust exposure will be about 50 to 60 times requirements. Individual subsystems are dis-
higher for the Solar Probe+ trajectory than for cussed in subsequent sections.
that defined in the 2005 STDT study. Despite 3.4.1. Spacecraft Description
this increase in the environment, the mitigation Solar Probe+ is a three-axis-stabilized
factors baselined in the 2005 STDT study are spacecraft. Its most prominent feature is the
sufficient to protect the spacecraft. A detailed Thermal Protection System (TPS), a large flat
analysis is given in Section 3.14. ceramic-coated carbon–carbon (C-C) shield
3.3.7. Electromagnetic Interference, that is 2.7 m in diameter, with associated
Electromagnetic Compatibility, structure used to attach the shield to the space-
and Magnetic Cleanliness craft. The TPS protects the bus and payload
Solar Probe+ is intended to measure electrical within its umbra during solar encounter. The
and magnetic fields near the Sun. We expect that science instruments are mounted either di-
spacecraft systems will be required to meet elec- rectly to the bus, on a stand-off bracket near
tromagnetic compatibility (EMC), electromag- the fairing attachment, or on a science boom
netic interference (EMI), and magnetic cleanli- extended from the rear of the spacecraft. The
ness requirements like those in other missions science boom also carries the solar horizon
with similar instruments such as STEREO and sensor (SHS) for backup attitude safing during
Radiation Belt Storm Probes (RBSP). Generally, the solar encounter. Three deployable C-C
these requirements necessitate the use of hard- plasma-wave antennas are mounted 120° apart
ware mitigation, e.g., synchronized DC/DC con- on the side of the bus. These antennas will
verters and filters in the power system electron- protrude beyond the umbra during encounter.
ics (PSE). In the case of Solar Probe+, the fields
to be measured are larger than for mis-
sions performing similar measurements;
preliminary analysis indicates that re-
quirements will be no worse than for pre-
vious missions such as STEREO. Detailed
requirements will be developed in Phase
A. The baseline concept includes
EMI/EMC and magnetic cleanliness miti-
gations that we anticipate using, and we
have included EMI/EMC testing and veri-
fication activities in the program cost and
schedule reported in this study.
3.4. Spacecraft Overview
The Solar Probe+ spacecraft will op-
erate in environments ranging from
0.044 to 1 AU from the Sun and will
accommodate the payload defined by
the STDT. The spacecraft concept is il-
lustrated in Figure 3.4-1, and the major
components are shown in the block dia-
gram in Figure 3.4-2. This section pro- Figure 3.4-1. Spacecraft configuration.

3-20
3.0 MISSION IMPLEMENTATION

Figure 3.4-2. Solar Probe+ block diagram

Instrument data processing is provided by a rays are folded inside the umbra formed by the
common data processing unit (DPU). The TPS. During solar encounter, power is gener-
baseline design includes instrument power dis- ated by the secondary solar arrays, two panels
tribution through the DPU. A Phase A study of high-intensity solar cells mounted on move-
will be conducted to optimize payload power able, liquid-cooled base plates. At 0.25 AU, the
distribution in light of electromagnetic inter- start of solar encounter, the panels are fully ex-
ference (EMI)/electromagnetic compatibility tended outside the TPS, and as the spacecraft
(EMC) and magnetic cleanliness require- approaches the Sun, panels are partially re-
ments. The hexagonal bus carries the space- tracted behind the TPS to maintain constant
craft subsystems and provides an efficient me- temperature and power output. A lithium-ion
chanical structure to handle launch loads and battery is included as a secondary power source
integrate with the launch vehicle. to handle transient loads and power the space-
Power is provided by two separate solar ar- craft during launch and ascent until the primary
ray systems. The primary solar arrays, used solar arrays are deployed. The power system
outside 0.25 AU, are MESSENGER-heritage electronics (PSE) box controls spacecraft
panels. Array temperature is controlled by in- power and battery recharge, and provides the
cluding optical surface reflectors (OSRs) with primary power bus voltage for the bus.
cells and rotating the arrays with respect to the The Solar Probe+ avionics suite centers on
Sun to keep the cell temperature within qualifi- redundant integrated electronic modules
cation limits. Inside 0.25 AU, the primary ar- (IEMs) that house the command and data han-

3-21
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

dling (C&DH) processor, solid-state recorder counter is stored on the SSR and downlinked
(SSR), guidance and control (G&C) instrument outside the solar encounter where the tempera-
interface, and payload interface. The avionics ture constraints on the HGA are satisfied. X-
suite also includes the power distribution unit band communication through the LGAs is
(PDU), an internally redundant box that in- provided as backup for periods when Ka-band
cludes all power switching as well as pulsed communication is not available.
loads to thrusters and single-event actuators. The propulsion subsystem consists of a
Remote input/output (RIO) devices are used to blowdown monopropellant hydrazine system
collect spacecraft telemetry and communicate consisting of a single tank, 12 0.9-N (0.2-lbf)
with the avionics suite through serial data links. and two 4.4-N (1-lbf) thrusters used for mo-
The G&C subsystem consists of three star mentum control and trajectory correction ma-
trackers and one internally redundant inertial neuvers (TCMs), and associated plumbing and
measurement unit (IMU) as the primary attitude electrical hardware. The hydrazine propellant
determination sensors, with an internally redun- tank is central to the bus.
dant SHS used as a check on the primary G&C 3.4.2. Spacecraft and Mission Reliability
sensors and as a backup safing sensor. Four re- Solar Probe+ uses both hardware and func-
action wheels are used for attitude control. tional redundancy to reduce the risk of failure
The telecommunications subsystem consists and ensure mission reliability. The fault man-
of a gimbaled high-gain antenna (HGA) agement approach adopted for the mission is
mounted on an arm used to extend the HGA based on several considerations. First, during a
beyond the umbra for Earth-pointing and two solar encounter, attitude control must be pre-
low-gain antennas (LGAs). Because the ge- cisely maintained to avoid exposing instru-
ometry of each solar encounter is different, no ments and spacecraft to direct solar flux. At
real-time science data downlink is planned perihelion, the maximum off-pointing allowed
inside 0.25 AU. In this region, command and is 2°. Inability to recover quickly from an atti-
housekeeping telemetry links will be on Ka- tude control fault could result in loss of mis-
band through the LGAs, because the HGA sion. Attitude control is rendered more diffi-
must be stowed within the umbra formed by cult by uncertainties in the solar environment
the TPS. Science data collected in a solar en- such as coronal lighting effects on star track-
Table 3.4-1. Hardware redundancy. ers or torques induced by high-
Functional Area Hardware Redundancy speed dust impacts. Second, the
Avionics 2 IEMs nature of science data collection
Internally redundant PDU requires a significant amount of
Payload Internally redundant payload DPU onboard autonomy and the ability
Attitude Determination 3 star trackers
Internally redundant IMU
to quickly switch to backup sys-
Internally redundant SHS tems. In addition, the nearly 7-
Attitude Control 4 reaction wheels year mission lifetime influences
Propulsion Redundant thrusters in each axis hardware selection and redun-
Data Bus Redundant 1553 bus dancy decisions.
Redundant serial interfaces
Data Storage 2 SSRs
Hardware redundancy is incor-
Telecommunications 2 uplink/downlink cards porated in all spacecraft compo-
2 LGAs nents and subsystems that can
2 each X-band and Ka-band TWTAs practically be made redundant
Thermal Control 2 thermistor harnesses (Table 3.4-1). In addition, Solar
2 heater harnesses
Power Internally redundant PSE Probe+ incorporates functional
Extra solar cell strings on each array redundancy in many critical areas,

3-22
3.0 MISSION IMPLEMENTATION

Table 3.4-2. Functional redundancy.


Functional Area Primary System Functional Mission Impact
Failure Redundancy
C&DH Processing C&DH software fault Second IEM operates in Software must be promoted back into
safe mode operational mode
G&C Processing G&C software fault Second IEM operates in Software must be promoted back into
safe mode operational mode
Attitude Determination Star tracker IMU: short duration Communications through LGA instead
Sun Horizon Sensor of HGA
IMU Star tracker: low rates Degraded pointing performance
Sun Horizon Sensor
Attitude Control Reaction wheels Thrusters Increased propellant usage
Increased outgassing
Telecommunications Ka-band downlink X-band downlink Loss of communications P ± 8 hours
Reduced science data volume
HGA LGA Significant reduction in data rate
Power Battery Solar arrays More difficult management of switch-
over between primary and secondary
solar arrays.

as shown in Table 3.4-2. Mechanisms represent fault response sequences will be designed into
a potential source of failure, which will be ad- the flight software with the APL heritage rule-
dressed in the detailed Solar Probe+ design. In and-macro-based autonomy system. During
particular, the primary solar arrays must be the detailed mission design, the engineering
stowed during solar encounter and then ex- team will determine the necessary operating
tended following encounter, the secondary so- modes to satisfy the safety requirements for
lar arrays must be retracted through perihelion each orbital phase while minimizing software
and then extended as the spacecraft travels and operational complexity.
away from perihelion, and the HGA must move APL has successfully demonstrated their abil-
to point to Earth and also be stowed during so- ity to operate a spacecraft in a thermally sensi-
lar encounter. One major driver during this ef-
fort has been to reduce the number of mecha- Table 3.4-3. Mass summary.
nisms needed. This effort will continue in fur- Name Current Best
ther design phases. The remaining mechanisms Estimate (kg)
Instruments 47.2
incorporate fail-safe features to guarantee mis-
Accommodation Hardware 7.4
sion reliability. Telecommunications 31.8
The Solar Probe+ fault management system G&C 30.4
is distributed throughout the spacecraft design Power 119.2
as hardware, software, autonomy, and mission Thermal Protection System 68.5
operations requirements. The primary objec- Thermal Control 15.7
tive of the fault management system is to Avionics 12.7
maintain a thermally safe attitude during the Propulsion 20.5
Structure 58.9
solar encounter. In addition, the fault man-
Harness 18.5
agement system is responsible for responding
Dry Mass 428.3
to time-critical fault scenarios, maintaining a Propellant 52.7
power-positive spacecraft configuration, and,
Wet Mass 533.3
if necessary, reconfiguring the telecommuni- Launch Mass 610
cations system for emergency data rate com- Total Mass Reserve 77
munications. Some of the fault detection and Total Margin 30.10%

3-23
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

Table 3.4-4. Mass margin risk mitigation.


Current Best
Estimate Expected
ID No. Subsystem Task Description Mass (kg)
1 Launch Vehicle Identify launch vehicle mass hold-backs, etc., based on his- 23.00
torical experience and comparison of mission-unique items
across programs
2 TPS Decrease the shield diameter via a bus repackaging 5.00
3 Power, Solar Array Decrease the substrate required area/perform a cell-string 2.00
layout
4 Power, Solar Array Perform a historical margin analysis for in-flight systems and 3.00
decrease the derating penalties
5 Thermal Investigate lightweight radiators 2.00
6 Thermal/Mechanical Perform design cycle on transition structure assembly (TSA) 3.00
fluid/mechanical system
Total Potential Savings 38.00

tive environment with the MESSENGER space- spacecraft system level. Unallocated margin is
craft flying through perihelion distances of ap- then carried at the system level for distribution
proximately 0.3 AU. Because of the similarity in from the system engineering team to the various
the mission fault management requirements, the subsystems as the need arises during the pro-
Solar Probe+ fault management system will gram. The engineering team has identified a col-
utilize numerous design components and lessons lection of Phase A trade studies, shown in Table
learned from the MESSENGER mission. 3.4-4, to increase mass margin beyond 30%
3.4.3. Mass and Power Budget Summaries within 6 months of Phase A start.
3.4.3.1. Mass Budget 3.4.3.2. Power Budget
Table 3.4-3 shows the mass summary for The power subsystem is designed to provide
Solar Probe+. Mass is summarized for instru- 482 W of load power between 0.9 AU and 0.044
ments, instrument support hardware, mecha- AU distance from the Sun. Between 0.9 AU and
nisms, and each spacecraft subsystem. Masses 1 AU, less power is available, and spacecraft
given in Table 3.4-3 are current best estimates operational requirements in this regime will be
representing the best current assessment of reduced primarily by lowering power allocated
mass for that item at launch. Launch mass is to the telecommunications subsystem. Sun–
based on performance for an Atlas V 551 probe distance is greater than 0.9 AU only im-
launch vehicle with a STAR-48BV third stage, mediately after launch and around aphelion of
the baseline for Solar Probe+. The mass sum- the first few orbits, so no impact on science re-
mary includes 30.1% margin to account for turn is expected. Table 3.4-5 summarizes the
unanticipated growth and any launch reserves. average power needed during the solar encoun-
Because specifics of the launch vehicle have ter based on current best estimates. A more de-
not yet been set, propellant mass is based on tailed breakdown of the power budget for all
the need to achieve the required ΔV after in- modes of spacecraft operation is provided in
clusion of the full margin. A more detailed Appendix A.
mass budget is provided in Appendix A. 3.5. Mechanical Systems
Mass data are developed for Solar Probe+ by Mechanical systems for the Solar Probe+ mis-
using a bottom-up methodology where lead en- sion are organized into three areas of respon-
gineers responsible for their subsystems gener- sibility: the bus structure, mechanisms sup-
ate current best estimates of mass. The maturity porting both the telecommunications and
of the system is assessed, historical data are thermal/power subsystems, and science instru-
compared with the current best estimate, and an ment accommodations (including mechanisms
appropriate allocated margin is added at the and interfaces).

3-24
3.0 MISSION IMPLEMENTATION

Table 3.4-5. Power summary.


Name Post- Maneuver Cruise Checkout Approach Science
separation
Instruments 0.0 0.0 0.0 28.6 0.0 57.2
Accommodation Hardware 0.0 0.0 0.0 0.2 0.0 0.0
Telecommunications 49.7 97.7 97.7 49.7 49.7 49.7
G&C 95.2 95.2 95.2 95.2 95.5 95.5
Power 18.8 18.8 18.8 18.8 18.8 18.8
Secondary Array Thermal 0.0 43.2 43.2 43.2 43.2 43.2
Control
Avionics 35.0 35.0 35.0 35.0 35.0 35.0
Propulsion 2.9 35.9 2.9 2.9 2.9 2.9
Heaters 0.0 27.4 60.4 51.6 80.1 22.7
Harness Loss 3.0 5.3 5.3 4.9 4.9 4.9
Total 204.6 358.5 358.5 330.1 330.1 329.9
Available Load Power 0.0 482.0 482.0 482.0 482.0 482.0
Power Reserves –204.6 123.5 123.5 151.9 151.9 152.1
Margin –100% 34.4% 34.4% 46.0% 46.0% 46.1%

3.5.1. Bus Configuration and Structure and the Thermal Protection System (TPS) de-
Design sign. Figures 3.5-1, 3.5-2, and 3.5-3 identify the
The configuration of the mechanical system is significant components and the baseline bus sys-
driven primarily by the launch vehicle and third- tem mechanical configurations. Figures 3.5-4
stage-generated environments, science payload and 3.5-5 depict the flight configurations, and
field of view (FOV), subsystem requirements, Figure 3.5-6 depicts the launch configuration.

Figure 3.5-1. Solar Probe+ solar encounter con- Figure 3.5-2. Alternate view of solar encounter
figuration. configuration.
3-25
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

Figure 3.5-3. Alternate views of solar encounter configuration.


The bus structure design chosen is a conven- in a manner very similar to the approach used
tional aluminum design that employs no tech- for several recent successful missions, including
nology development and presents a very low- STEREO and New Horizons. Clean load paths
risk approach to the mission. Aluminum hon-
eycomb panels employ both embedded-edge
members and bonded clips to attach the panels

Figure 3.5-4. Science downlink configuration. Figure 3.5-5. Alternate view of science downlink
configuration.

3-26
3.0 MISSION IMPLEMENTATION

Figure 3.5-7. Spacecraft bus layout.

Figure 3.5-6. Launch configuration shows com-


patibility with volumetric stage.

efficiently transfer the launch and flight loads


from the TPS interface trusses and from bus-
mounted components through the bus structure
corners and into the payload adaptor. The struc-
tural design approach for the bus-centric pro-
pulsion tank mount consists of a simple ma-
chined and assembled bracket support struc-
ture. Ample access to the interior of the space-
craft is provided by three removable panels.
The 37-in. diameter third-stage STAR-48 in-
terface and clamp band are industry-standard Figure 3.5-8. Interior view of spacecraft bus layout.
components and are very well understood tech-
nologies. The adaptor interface to the third Figure 3.5-7 depicts the significant bus struc-
stage incorporates all needed electrical and in- ture elements. Figure 3.5-8 illustrates the pre-
strument purge interfaces for the spacecraft. liminary layout of the interior bus components,

3-27
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

mass models will be integrated and a three-


axis sine vibration test of the bus structure will
be performed before delivery for spacecraft
system-level integration. A mass simulator of
the TPS and the engineering model support
trusses will be used for these preintegration
tests, enabling development of the bus and the
transition structure assembly (TSA) support
trusses in parallel with the carbon–carbon (C-
C) TPS. The bus structure, TSA and TPS sub-
system testing, and flow into the spacecraft
Figure 3.5-9. Solar encounter flight configuration system-level integration and testing (I&T)
showing umbra.
phase is discussed in detail in Section 3.5.5.
demonstrates adequate volumetric space for Phase A design and development activities
harness and connectors, and provides an illus- planned for the bus structure include the usual
tration of the propulsion tank mounting con- configuration, science instrument, and subsys-
cept. Figure 3.5-9 illustrates the stowed space- tem accommodation tasks. An investigation
craft configuration within the umbra. into both a hybrid composite and an all-
There are no new technology development composite structure design is planned to en-
tasks outside the usual design process for the sure that the bus structure optimization of cost,
aluminum bus structure. The bus structure mass mass, and schedule has been achieved for this
has been calculated based on the current design mission and matched to the system-level re-
presented, and the values stated for components quirements. A calculated mass rollup, detailed
have been favorably compared with experience schedule and revised cost estimate, and draft
spanning several recent missions. All mechani- bus structure requirements document for the
cal ground handling equipment (MGSE) sup- system are included in Phase A tasks.
port equipment for spacecraft handling, trans- 3.5.2. Transition Structure Assembly
fer, shipping, and testing operations is planned Design Overview
to be either currently available equipment or, in The TSA provides the mechanical structure
the case of lift fixtures and transportation dol- that couples the bus structure to the TPS shield
lies and containers, typical mission unique and also serves as a platform to support the sec-
hardware of standard design. ondary solar array system and its associated
The bus structure proposed development thermal fluid cooling subsystem. Figure 3.5-10
schedule and cost baseline support a conserva- depicts the significant components of the TSA.
tive and low-risk approach successfully used The baseline structure design consists of a
on recent previous programs. The structure set of graphite-based structural members con-
follows a proto-flight approach with a single- necting the TPS, the secondary power system
flight structure. There are no engineering thermal system, and mechanism components.
models planned for the bus structure. The The mechanisms are discussed in Section 3.5.3.
baseline test and qualification program will Thermally isolated flat-panel exterior radiators
include a static load test before shipment to provide heat rejection for the secondary solar
the propulsion vendor. After propulsion sys- array fluid circulation system. These are tied to
tem integration, a thermal balance and propul- the six carbon graphite/cyanate ester (Gr/CE)
sion system thermal vacuum test will be per- trusses by short flexures that accommodate
formed. To obtain modal information and to thermal expansion mismatch. The current base-
qualify major package interfaces, component line design and mass estimates of the radiators

3-28
3.0 MISSION IMPLEMENTATION

Table 3.5-1. Data analysis summary. MEMF, mod-


el equivalent mass fraction.
Frequency Direction MEMF Comments
18.1 Hz Lateral 1 16% X Truss bending
30% Y
19.2 Hz First Z 10% X Bending of secon-
26% Y dary array support
20% Z panels and tubes
23 Hz Lateral 2 36% X Truss bending
13% Y

employ heat pipes embedded within aluminum


honeycomb panels and face sheets. The space-
craft forward deck, trusses, radiator panels,
circulation pumps, secondary solar arrays,
slider mechanism, plumbing, and control elec-
tronics are assembled, processed, and quali-
fied as a single assembly, all included in the
TSA. The integration, testing, delivery, and
integration into the spacecraft system of this
assembly is discussed in Section 3.5.5.
Figure 3.5-10. TSA components. A preliminary structural analysis has been
performed on the TSA system. The results con-
firmed the mass values estimated for these ele-
ments. A finite element model was used to es-
tablish truss geometry and sizing needed to meet
launch vehicle requirements. Three static load
cases—15 g separately in X, Y, and Z—were
run, and 15 iterations were completed, resulting
in a completion of the baseline geometry and
member sizing tasks. The design meets all re-
quirements with positive margins. Figure 3.5-11
illustrates the primary components modeled
from a finite element model analysis perspec-
Figure 3.5-11. Both pinned and fixed-boundary tive. Figure 3.5-12 depicts several of the finite
conditions were evaluated. Fifteen iterations were
completed to maximize stiffness while minimizing
element model analysis runs. Table 3.5-1 sum-
mass. All elements show positive margins of safety. marizes the data analysis results.

Figure 3.5-12. Results from fixed-boundary conditions are shown. All elements show positive margins of
safety.
3-29
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

3.5.3. Bus and Transition Structure fidelity engineering model test program is
Assembly System Mechanisms planned and costed with a schedule that pre-
Mechanisms and mechanical systems sup- cedes the flight build. The required g-negation
porting the power and TPS subsystems are system, the dynamics analysis tasks, and de-
discussed in this section, and thermal and elec- ployment data collection system are all heritage
trical performance of these systems are dis- processes tracing their roots to the deployment
cussed in their respective subsections testing of the TIMED four-panel solar array
3.5.3.1. Primary Solar Array Design Overview system. The substrates and their associated
The primary solar array system consists of high-temperature array technology use materi-
two identical wings, each comprising two pan- als and process technology from the MES-
els. The wing design interfaces to the space- SENGER solar array and differ only in geomet-
craft via a two-axis Schaffer-Moog Type II bi- ric shape and area. In order to further minimize
axial harmonic drive stepper motor, a set of the risk to the flight build cycle, several qualifi-
interface brackets, and a pyrotechnic-activated cation and test panels are planned and costed to
Hi-Shear pin puller (SP-1214) device that pro- ensure previous fabrication and cell lay-down
vides a preload to the wing and postlaunch re- processes used on MESSENGER have been
lease for subsequent mission operations. The retained by the fabrication and cell lay-down
wing design is a retractable design, driven out vendors. Primary solar array mass has been
to the deployed position after launch release calculated based on the current system re-
and retracted back to the stowed position when quirements and uses MESSENGER measured
the spacecraft is within 0.25 AU of the Sun. masses and packing factors for the panels
The drive motor system baseline is a bi-axial where appropriate.
Primary Solar Array Subsystem Qualifi-
Schaeffer-Moog Type I harmonic drive stepper
cation. The primary array flight system
motor design with extensive flight heritage.
buildup and qualification testing duplicates the
After the full deployment position is reached,
flow and processes developed for the TIMED
the array wings then are capable of rotation by
solar array wing assemblies. The flight system
the second axis of the motor in order provide
design cycle starts with the fabrication and
the array angular relationship to the Sun as
testing of a full-scale, high-fidelity engineer-
needed for thermal and power system control.
ing model. The engineering model is used to
The retractable, rotating wing, two-panel
ensure that the dynamic behavior (deploy-
system design combines flight-heritage com-
ments and retractions) correlates with the dy-
ponents from the TIMED (Thermosphere,
namic modeling analysis, so that flight system
Ionosphere, Mesosphere Energetics and Dy-
calculated mass, inertias, and physical size are
namics) four-panel design with the additional
matched to the flight design. The engineering
Solar Probe+ requirement added for a retract-
model also functions as the development
able design. A similar design incorporating a
model for harness routing and thermal blan-
retraction and redeployment capability similar
keting. The engineering model program uses
to the Solar Probe+ requirement was proposed
the same mechanical ground handling hard-
for the TIMED mission and developed to the
ware, including the gravity negation system as
preliminary design review (PDR) level by a
the flight system. Mechanism life testing at
vendor but was not used in flight as the mission
both hot and cold temperature extremes also is
design evolved. A design trade study for the
planned for both the engineering model sys-
Solar Probe+ retractable feature that will trade
tem model and for the mechanism subassem-
independently driven hinge lines using a geared
bly level. Once the engineering model system
stepper motor against the cable synchronized
performance is correlated to the dynamics
hinge lines of the earlier TIMED study will be
analysis models, the flight build can proceed.
conducted in Phase A. A full-scale, high-
This identical engineering model development

3-30
3.0 MISSION IMPLEMENTATION

namic deployments occur as a result of vibration


testing. The system is then ready for spacecraft-
level integration where it will undergo space-
craft system-level vibration, acoustic, shock, and
pop-and-catch deployments to ensure first mo-
tion occurs.
3.5.3.2. Secondary Solar Array and Transition
Structure Assembly Mechanism De-
sign Overview
The secondary power-generation system
mechanical system design is driven by the mis-
sion power generation needs within 0.25 AU.
Under the intense solar flux during solar en-
Figure 3.5-13. TIMED flight wing and g negation counter, an active cooling loop system is re-
system. quired to cool the secondary solar arrays. As
program methodology lead to the successful the solar distance decreases, a positioning sys-
TIMED solar array in flight deployments that tem pulls the concentrator arrays inboard, shad-
correlated closely with predictions. owing more strings behind two knife edges in
The flight build and test program starts with the TPS. This operation provides constant
the fabricated substrates delivered to the cell power generation while maintaining the heat on
vendor for cell lay-down and wiring. After re- the concentrator array to manageable levels.
ceipt and power system electrical performance The secondary power-generation system design
testing, the flight substrates then are assembled is an integrated part of the TSA, which couples
and deployment-tested with the identical off- the TPS into the conventional bus system. Fig-
loading g-negation fixture used for the engineer- ure 3.5-10 depicts the TSA assembly and iden-
ing model development program as described tifies the significant components.
above. The TIMED solar
array wing and g-negation
system (Figure 3.5-13) are
well understood and more
complex than that needed
for Solar Probe+. Data re-
covered during deployment
testing of the flight system
is compared against the en-
gineering model kinematics
analysis to ensure that the
behavior of the system is
fully understood and a base-
line performance set of
characteristics is captured
before wing environmental
testing. Proto-flight vibra-
tion testing (at levels above
predicted flight levels) is
followed by an additional
set of deployment tests to
ensure no changes in dy- Figure 3.5-14. TSA mechanism components.

3-31
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

Figure 3.5-14 identifies the major mechani- and the radiator panels. Aluminum fittings and
cal components of the TSA and secondary tubing (0.015-in. wall thickness × 0.500-in.
power subsystem. Two identical retractable diameter) are the baseline for mass calcula-
panel/cooling systems are used for the Solar tions and preliminary sizing of the system.
Probe+ spacecraft. The release of each system The use of fittings to connect tubing to com-
from the launch (stowed) configuration is ac- ponents is minimized where possible. Prelimi-
complished by the activation of a single flight- nary runs for plumbing have been developed
qualified Hi-Shear pin puller (SP-1214) device to identify first-order challenges as illustrated
used successfully on previous spacecraft. in several TSA-related figures. Development
Once the launch lock is released, the system is and optimization of the flight plumbing and
capable of flight deployment/retraction opera- harness will occur using a full-scale mockup.
tions. During operations outside 0.25 AU, the Phase A mechanical system tasks identified
secondary solar arrays are fully extended to for the TSA mechanisms development include
maintain temperature and minimize heater trade studies on selection of mechanical compo-
power needed for thermal balance. At 0.25 nents, a fabrication and testing development
AU, the secondary system is fully extended, program for several subassemblies, a design
and the primary solar array system (both trade study on radiator design, and evaluation of
wings) is then stowed. The temperature of the alternatives for plumbing components. Included
cell substrates (and resulting power conver- in Phase A tasks is a trade study to compare a
sion efficiency) is controlled as the spacecraft design that moves the knife edge with the
gets closer to the Sun by a closed-loop fluid baseline design, which moves the array.
management system. The system is designed Secondary Solar Array Subsystem Quali-
to keep the cells, mechanisms, drive motors, fication. The secondary solar array system
and other components within typical operating mechanisms are managed under a test and
temperature limits. qualification program very similar to the pri-
The slider mechanism baseline design con- mary solar array system, with additional testing
sists of linear bearings and a recirculating ball planned for the fluid components and elevated
lead screw system adapted from commercial temperatures attributable to the operating envi-
machine tool systems. Features include a sim- ronment imposed by the proximity to the TPS
ple, robust mechanism derived from numerical shield. A high-fidelity, full-scale engineering
controlled machine tools where debris toler- model fabrication and testing program is
ance, millions of problem-free cycles, preci- planned. The engineering model test plan starts
sion, and tolerance of thermally induced mis- at the component level for the critical motion
alignments are the normal operating require- components. Mockups of all flexible compo-
ments. The modifications planned for the slider nents (harness, cooling lines, etc.) are tested
are primarily made to reduce mass and ensure over temperature extremes for both physical
lubricants and selected components meet behavior (routing changes resulting from re-
spaceflight requirements. The drive motor se- peated motion) and, in the case of fluid compo-
lected for the system is a Schaeffer-Moog step- nents, fluid integrity under pressure. Once the
per motor with extensive flight heritage. The testing for performance and lifetime are com-
rated life, available torque, and positional accu- plete, a test to failure is planned for both the
racy (resolution) of the motor exceed mission flexible harness and flexible fluid components
requirements. to ensure that the limits of performance are
The copper used to hold the secondary solar well understood and compared to the expected
array cells is attached to the slider and extracts mission requirements. Torque changes of all
heat from the cells via internal passages flexible service loops resulting from lifecycle
through which fluid passes. Input and output testing also are measured again, duplicating the
flex hoses transfer the fluid between the plate solar array system testing and qualification plan

3-32
3.0 MISSION IMPLEMENTATION

methodology. After the component-level test- cycle. The bus mechanical team will lead this
ing is completed, the full-scale engineering effort. The current location of the boom on the
model secondary array system (less cells) is bus eliminates unnecessary bus structure mass
assembled and tested as a subsystem. Again, by taking advantage of significant open space
the system is exercised over predicted tempera- available in the payload adaptor fitting.
ture extremes. Vibration testing and postvibra- 3.5.3.3.2. Mechanical System Interfaces and
tion deployment testing is planned for the Science Instrument Accommodations
mechanism subassemblies.
Mechanical interfaces for the instruments
The flight secondary power generation sys-
consist primarily of flat surfaces with stan-
tem will be qualified and delivered to the
dard, threaded fasteners used to attach the
spacecraft I&T program as a complete subsys-
components to the bus structure. Where pin-
tem in the configuration illustrated in Figure
ning is required for alignment stability, it can
3.5-10. The system will be vibrated and
easily be accommodated with the bus design.
strength-tested by using a TPS mass simulator
Phase A tasks include a drafting of a prelimi-
attached to the forward plane of the Gr/CE
nary alignment budget and verification plan.
support trusses using titanium flexures. After
Baseline instrument purge requirements are
vibration and release testing, the system will
unexceptional and easily accommodated with
be thermally cycled, and a thermal balance test
existing support systems.
will be performed with solar cell simulators
Physical locations of the science payload
and a TPS heat-source simulator. During
instruments on the bus and the resulting in-
thermal testing, the mechanisms will be life-
strument requirements for clear FOV (CFOV)
tested (~50 cycles through the motion ex-
are compliant based on information provided
tremes). Subsequent to solar cell installation,
by the STDT. As actual instruments are se-
solar array testing will be conducted as with
lected for the mission, accommodation re-
the primary solar arrays.
quirements for each will be met; no significant
3.5.3.3. Science Instrument Payload Me- issues with instrument accommodation are
chanical System Accommodations expected. The science instrument CFOV are
3.5.3.3.1. Aft-Mounted Science Boom illustrated in Figures 3.5-15 through 3.5-19.
The bus system will supply an aft-mounted 3.5.3.4. High-Gain Antenna Mechanical
boom for science instrument accommodation. Mechanism Overview
The integration of the boom-mounted science High-Gain Antenna Mechanical Design
instruments into the boom system will require Overview. The mechanical motion components
significant accommodation efforts and inter- of the high-gain antenna (HGA) system are dis-
face definition early in the bus development

Figure 3.5-15. Instrument CFOV. Figure 3.5-16. Alternate view of instrument CFOV.

3-33
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

Figure 3.5-17. Alternate view of instrument CFOV.

Figure 3.5-20. HGA/bus layout.

Figure 3.5-18. Alternate view of instrument CFOV.

Figure 3.5-21. Alternate view of HGA/bus layout.

(Figures 3.5-20 and 3.5-21) have been per-


formed to establish the required HGA system
geometry needed to achieve the CFOV needed
for communications. A Schaeffer-Moog har-
monic drive actuator is the baseline motor for
Figure 3.5-19. Alternate view of instrument CFOV. both motion joints. The motors chosen have
cussed in this section, and the antenna element demonstrated lifetimes well in excess of the
and the radio frequency (RF) system perform- mission requirements. RF signals are transmit-
ance are discussed in Section 3.10. The FOV ted across the rotary joints via spaceflight
coverage requirements of the mission coupled heritage rotary joints and along the mast via
with the requirement to stow and deploy the rigid waveguide. The arm baseline design is of
antenna multiple times during the mission are a normal flight-heritage design using a light-
the driving requirements. Preliminary layouts weight carbon-composite tube with bonded
titanium end fittings at the attachment points.

3-34
3.0 MISSION IMPLEMENTATION

HGA Mechanical System Test and Quali- flight forcing functions for various dynamic
fication Plan. The methodology used to de- loading events. CLA loads succeed the pre-
sign and implement the HGA mechanical test liminary design loads and have customarily
and qualification plan follows closely the suc- proven to be lower than the Atlas Launch Sys-
cessful experiences of both the HGA systems tem Mission Planner’s Guide loads. There
used on the Midcourse Space Experiment typically are two CLA cycles during a pro-
(MSX) and the STEREO spacecraft. A high- gram, the first performed with a preliminary
fidelity engineering model will be mechani- spacecraft dynamics model to aid spacecraft
cally integrated and tested for proper motion design and the second with a test-verified
employing the existing TSA and the full-scale spacecraft dynamics model to finalize flight
wiring model. Measurements of drag torque margins. Loads analysis is performed by using
resulting from temperature effects on all mo- the requirements levied in NASA-STD-5002,
tion components, life-cycle testing, and other Load Analysis of Spacecraft and Payloads.13
testing is planned on all motion components The spacecraft also is exposed to an acoustic
and on the engineering model system. Follow- environment during takeoff and through tran-
ing the completion of the engineering model sonic flight. This external acoustic environment
testing, the flight build will start, and it also is transmitted through the vehicle fairing to im-
will follow a test and qualification plan that pinge onto the spacecraft and affects large-area,
uses hot and cold deployment and operational low-weight surfaces such as the TPS, the radia-
testing, sine vibration, and other subsystem- tor panels, and solar arrays. The acoustic sound
level mechanical testing before delivery to pressure levels are provided in the Atlas Launch
both the RF system and spacecraft I&T team. System Mission Planner’s Guide and vary ac-
The flight antenna element will be included cording to spacecraft volume, payload fairing
for all subsystem flight hardware testing be- dimensions, and acoustic blanket placement.
fore the spacecraft integration. Results from preliminary acoustic analysis of
3.5.4. Spacecraft System Structural the TPS, solar arrays, and lightweight radiator
Analysis and Launch Environments panels under effective root-mean-squared
The preliminary design loads for the space- (RMS) pressures indicate that the quasi-static
craft system are derived according to the most design loads based on a typical mass accelera-
recently available Atlas Launch System Mis- tion curve remain the predominant load cases.
sion Planner’s Guide.12 The expected flight Other analytical methods involving finite ele-
levels provided in the launch guide are speci- ment, boundary element, and statistical energy
fied as quasi-static accelerations for the space- methods will be used to more accurately deter-
craft center of gravity. Design loads for sub- mine the response of these lightweight structures
systems are initially based on a mass accelera- over the entire acoustic frequency spectrum.
tion curve, superseded later in the program by The Atlas Launch System Mission Planner’s
the coupled loads analysis (CLA). Guide requires a minimum natural frequency
Spacecraft and subsystem design loads are of the spacecraft to be 8 Hz in the two lateral
more accurately computed by the launch vehi- directions and 15 Hz in the thrust direction. To
cle manufacturer during the CLA cycles, in avoid coupling, the subsystem minimum fre-
which reduced dynamic models of the space- quencies typically are set higher. For the TPS
craft are combined at the launch vehicle inter- mounted on its flexure system, the fundamen-
face with the flight-verified vehicle dynamics tal frequency design goals are set at 15-Hz lat-
model. The coupled model is then excited with
13
Load Analyses of Spacecraft and Payloads, NASA-
12
Atlas Launch Systems Mission Planner’s Guide, Rev. STD-5002, National Aeronautics and Space Admini-
10a, Lockheed Martin Corporation (January 2007). stration (June 21, 1996).

3-35
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

Figure 3.5-22. Systems test flow.

eral and 25-Hz axial. Solar array and radiator is designed to decouple the three significant
panels will be designed for frequencies above development items of the spacecraft mechani-
25 Hz. Instruments and components shall have cal system: the bus structure, the transition
fundamental frequencies above 50 Hz. structure, and the top circular-shaped TPS. The
The 2005 study included an analysis show- rationale for this design and development ap-
ing that the 2005 design met all Atlas re- proach is offered for each element as follows:
quirements for stiffness. The current study • The bus structure element is considered to be
uses a flat-plate TPS, a truss structure, and a a well understood, simple design using tradi-
bus essentially unchanged (from a load path tional materials and fabrication techniques. Its
standpoint) from that previous study—a more development cycle will closely follow the
favorable design. The minimum frequency of flow established for the STEREO and New
the Solar Probe+ stack also will meet Atlas Horizons missions.
stiffness requirements. • The transition structure element is com-
3.5.5. Spacecraft System Mechanical posed of the secondary power-generation
Thermal Test and Qualification system and its associated thermal-cooling
Plan system as well as the structural support for
3.5.5.1. Rationale and Test Flow the thermal shield. The development of this
The Solar Probe+ testing, qualification, and system is more challenging than the devel-
integration plan for the bus structure and TSA opment of a standard solar array power sys-

3-36
3.0 MISSION IMPLEMENTATION

tem because of the addition of the liquid- performed. Testing then commences with a
cooling feature. The subsystem requires static load and vibration test using mass mod-
qualification testing different than that of a els. The structure is then cleaned and bagged
typical solar power system. and shipped to the propulsion vendor for pro-
• The thermal shield development program pulsion system I&T. After return of the struc-
has the lowest technology readiness level ture to APL, a thermal balance test is per-
(TRL) of the three mechanical elements. formed. At that point, the top deck is removed
The program leverages heavily the work and delivered to the TSA team to be used for
done for the 2006 Solar Probe Thermal their flight buildup, and the remaining bus
Protection System Risk Mitigation Study. 14 structure is delivered to the I&T team for har-
Decoupling the development cycles of the ness installation and population with the flight
three primary elements reduces the risk of de- components.
layed hardware development schedules. Plan- 3.5.5.3. Transition Structure Assembly Quali-
ning via design to decouple these elements fication Plan
provides schedule flexibility and allows each Transition Structure Assembly Mockup.
element of the program to manage its specific The TSA consists of the structural support truss
schedule independent of the system, provided elements, the thermal-cooling system (radia-
delivery dates are ultimately met in I&T. The tors, pumps, cooling hoses, etc.), and the sec-
decoupled hardware development, testing, and ondary power-generation system. A similar de-
qualification plan presented here lowers velopmental approach to that used for space-
schedule risk and provides the best structure craft bus development begins with a high-
for these particular program challenges. Figure fidelity mockup. The engineering mockup is
3.5-22 illustrates the flow of the significant populated with a mix of simulators and engi-
elements and captures the significant qualifi- neering model components in order to identify
cation tests of the systems. Note provisions harness and plumbing issues before the flight
have been made (illustrated by dashed arrows) build. The mockup will be thermal vacuum-
to decouple the most technically challenging chamber-compatible and of high enough fidel-
two items (the TPS and the cell substrates for ity to support any thermal system subassembly
the secondary array) from the spacecraft sys- and component performance testing required.
tem, thus providing for their arrival at the Secondary Power-Generation System
launch site should unforeseen development Mechanisms. The design of the secondary
problems surface. power-generation system lends itself to a
3.5.5.2. Bus Structure Qualification Plan modular approach and a conventional mecha-
The bus structure schedule ties significant nism development plan employing a high-
interface definition milestones to the other ele- fidelity engineering model. After assembly of
ments to ensure all systems interface properly the drive and release components, baseline
in form, fit, function, and delivery. Fabrication current and torque margin performance meas-
of the structure follows a traditional schedule. urements are made. The engineering model
After initial assembly of the bus structure, cell positioning mechanism (EMCPM) sys-
preliminary measurements and alignments are tems then will undergo vibration testing as
subassemblies. After postvibration current and
14
torque measurements are completed, the
Solar Probe Thermal Protection System Risk Mitiga- EMCPMs will be life-tested to two times the
tion Study: FY 2006 Final Report, prepared by The
Johns Hopkins University Applied Physics Laboratory
predicted number of cycles, under both hot
under Contract NAS5-01072, Laurel, MD (November and cold conditions, with current and torque
30, 2006). trending performed.

3-37
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

The design approach will use simulators of shield possesses no electrical interface to the
the solar cell/substrate assembly in order to bus system. The sole interface is a conventional
decouple the solar cell/substrate development mechanical bolted joint. A template will be fab-
and delivery schedule from the mechanism ricated, and a fit check will be performed early
test and qualification program. The flight to ensure successful integration of the flight
build will commence after completion of the TPS. The TPS will be delivered to the bus as a
engineering model development and testing in fully tested and qualified component. The pro-
order to incorporate any design changes un- gram mechanical integration plan has the TPS
covered in the mechanisms during testing. The arriving and assembled to the spacecraft as late
flight-build CPMs will undergo the same test- as just before mass properties operations at the
ing sequence as the engineering model units, launch site. This sequencing provides the
with the exception of life testing. When the maximum development time to the element
flight cell/substrate assemblies are delivered, with the lowest TRL.
they can be assembled to the flight mecha- 3.5.5.5. Spacecraft System-Level Mechanical
nisms without disturbing the cooling fluid Test Campaign
loop integrity. System-level testing is conducted by using
Transition Structure Assembly. Slightly NASA-STD-7002, Payload Test Require-
overlapping the transition structure mockup ments, 15 and GSFC-STD-7000, General Envi-
development and secondary power subsystem ronmental Verification Standard (GEVS). 16
testing, the flight truss elements will be fabri- Verification that the transition structure and
cated and assembled with mass models in the spacecraft bus will withstand the design
preparation for a static load test where the in- loads is achieved through low-frequency (5–
terfaces are structurally test-qualified to the 100 Hz range) swept sinusoidal testing. The
worst-case expected loads. After the static levels for the system-level sine vibration test
load test, a vibration test will be performed are given in the Atlas Launch System Mission
with mass models of the TPS and components Planner’s Guide. 17 These levels typically are
to complete the structural test program for the notched to match spacecraft responses derived
structural elements of the TSA. The modal from the final CLA cycle. A minimum natural
data obtained from the vibration test will be frequency for the spacecraft is specified in the
used to correlate the CLA finite element Atlas Launch System Mission Planner’s Guide
model. Population of the TSA with flight to ensure that design loads and displacements
components will then begin with various func- are not exceeded. Minimum natural and actual
tional and electrical tests planned to exercise frequencies are verified during the fixed-base
the motion components. The thermal-cooling sine test sequence of the spacecraft. With the
system will be installed, the fluid system will TPS qualified separately and integrated before
be loaded, and thermal cycling and thermal mass property testing, a mass simulator of the
balance testing will be performed. After suc- TPS will be used during spacecraft-level envi-
cessful completion of all functional and per- ronmental testing.
formance testing, the TSA flight system will
be delivered to I&T for system-level testing. 15
Payload Test Requirements, NASA-STD-7002, Rev
3.5.5.4. Thermal Shield Development A, National Aeronautics and Space Administration
(September 10, 2004).
The thermal shield development is discussed 16
General Environmental Verification Standards (GEVS)
in a separate section of the report, yet the plan for GSFC Flight Programs and Projects, GSFC-STD-
for integrating the shield into the bus system as 7000, Goddard Space Flight Center (April 2005).
17
well as the assumptions covering the integra- Atlas Launch Systems Mission Planner’s Guide, Rev.
tion plan will be discussed in this section. The 10a, Lockheed Martin Corporation (January 2007).

3-38
3.0 MISSION IMPLEMENTATION

Verification that the spacecraft will with- Table 3.6-1. TPS requirements.
stand the acoustic environment is achieved Requirement Value
through acoustic testing. The levels for the Environmental requirements
acoustic test are given in the Atlas Launch Structural requirement 15 Hz lateral
35 Hz thrust
System Mission Planner’s Guide. Testing is
Ionizing Radiation 44 krad Si
conducted by using NASA-STD-7001, Pay-
Solar Flux Max. 70 W/cm2
load Vibroacoustic Test Criteria. 18 Min 0.13 W/cm2
Mass properties testing is conducted at the
Mission Duration 7 years
spacecraft level, generating a full inertia matrix Configuration
of the test article. Measured in separate tests, TPS Mass 70.5 kg
the mass properties of the solar arrays, the TPS, Shield Diameter 2.72 m
and the instruments in their deployed condition Primary Shield Thickness 15 cm
are added analytically to generate the full-up Spacecraft Orientation Sun-pointing
spacecraft mass properties matrix in the various Science measurements
configurations encountered during launch and Mass Loss 2.5 mg/s
flight. Shock testing—including clamp-band Primary Shield Surface α/ε 0.6
separation and local sources of shock such as Surface Conductivity Conductive
solar array release, instrument doors, etc.—is Heat Flow into Spacecraft <50 W
performed twice at the system-level. Final
launch site operations include a launch configu- bound the expected environmental exposures
ration weighing and center of gravity meas- that will be encountered over the mission, de-
urement before third stage integration. fine the required TPS configuration, and sup-
port the scientific measurements being made.
3.6. Thermal Protection System
Compared with the TPS design presented in
3.6.1. System Design Requirements
the 2005 STDT and 2006 TPS Risk Mitigation20
The present system requirements for the
reports, the key difference is the change in the
Thermal Protection System (TPS) are shown in
Sun distance. At closest approach, 9.5 RS, the
Table 3.6-1. These requirements are very simi-
peak flux is 70 W/cm2, compared with 400
lar to those developed in the 2005 STDT 19
W/cm2 at 4 RS in the previous studies. The larg-
study. The Solar Probe+ closest approach is 9.5
est distance from the Sun also has changed. The
RS, compared with 4 RS in earlier Solar Probe
Solar Probe+ aphelion is less than 1 AU, com-
studies. At that solar distance, the TPS does not
pared with 5 AU in the earlier studies. At 1 AU,
require a large conical shield. Slightly lower
the minimum solar flux is 0.13 W/cm2, resulting
temperatures are achieved for the Solar Probe+
in a minimum sunshield temperature of ~0°C.
design, using a coated, flat-front sunshield, than
The current Solar Probe+ sunshield design is
were obtained with the conical design and the
based on the secondary shield design developed
closer solar distances of the earlier studies.
in the Risk Mitigation Study. 21 With the Solar
Therefore, a significant part of the design,
Probe+ mission peak flux, the sunshield design
analysis, and testing done as part of those
study efforts remains valid for the Solar achieves similar temperatures to the earlier So-
Probe+ design. The design requirements listed lar Probe studies without the conical primary

18 20
Payload Vibroacoustic Test Criteria, NASA-STD- Solar Probe Thermal Protection System Risk Mitiga-
7001, National Aeronautics and Space Administration tion Study: FY 2006 Final Report, prepared by The Johns
(June 21, 1996). Hopkins University Applied Physics Laboratory under
19
Solar Probe: Report of the Science and Technology Contract NAS5-01072, Laurel, MD (November 30, 2006).
21
Definition Team, NASA/TM—2005–212786, National Solar Probe Risk Mitigation Study, prepared by The
Aeronautics and Space Administration, Goddard Space Johns Hopkins University Applied Physics Laboratory,
Flight Center, Greenbelt, MD (2005). 2006 Mid-Year Report.

3-39
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

The Solar Probe+ mission trajectory reduces


the program risks by incorporating multiple
encounters at progressively closer distances to
the Sun. In the earlier studies, the first solar
encounter was high risk because the first time
the TPS was used, it was at its peak tempera-
tures. The Solar Probe+ mission duration is
seven years. During this time, the spacecraft
makes 24 orbits, gradually decreasing the clos-
est approach to the Sun from 35 RS to 9.5 RS.
This movement of perihelion toward the Sun
Figure 3.6-1. TPS temperature sensitivity. allows the TPS system to be tested incremen-
shield. The Solar Probe+ TPS design includes a tally. Should some performance degradation be
carbon–carbon (C-C) shell surrounding an in- found, the mission trajectory can be altered,
sulating carbon foam. The front surface of the keeping the spacecraft at a safe Sun distance.
shield includes an optical coating to reflect The spacecraft heat flow requirement is
some of the incident solar energy. The resulting based on the fact that the spacecraft has a range
peak temperature expected at closest approach of power dissipations that it can absorb and still
is ~1700 K. maintain temperature control. At closest ap-
TPS temperatures are a function of the shield proach, the spacecraft sunshield has a total in-
configuration and optical properties. Shield cident flux of ~4 MW. Heat is transferred to the
temperatures for different cone geometries and spacecraft both radiatively from the bottom of
Sun distances are shown in Figure 3.6-1. The the shield and conductively through the shield
key optical property is the ratio of the solar ab- supports. The shield is required to reduce the
sorptance to the IR emissivity, α/ε, set at the heat flow transmitted to the spacecraft to 50 W,
design value of 0.6 in Figure 3.6-1, which shows or 0.001% of the original flux. At this effective
that the shield temperatures are reduced as the shield requirement, it is difficult to adjust the
Sun distance increases or the cone angle is de- heat leak. Our inaccuracy in analyzing and test-
creased. It also shows that there is very little im- ing the shield performance is such that the goal
pact on temperature until the cone angle is be- of the shield design is to provide large margins
low 45°. At 9.5 RS, the shield temperature is and then verify the important parameters by
~1700 K. In the earlier Solar Probe studies with realistic testing.
the conical TPS, the shield temperature was a 3.6.2. Design Approach
little over 1800 K. Based on the complexity of Sun Shield. The Solar Probe+ sunshield is a
making a large C-C structure, the Solar Probe+ 2.7-m diameter shell, approximately 15 cm
TPS design is simplified by using a flat shield. thick. It consists of a pan and a top cover. The
Other requirements from the earlier Solar shell is filled with carbon foam insulation that
Probe studies are affected by the change in the provides the thermal protection to the space-
mission aphelion. When the Solar Probe mis- craft bus. The material selection for the sun-
sion included a Jupiter gravity assist (JGA), the shield is driven by the expected exterior tem-
spacecraft spent a long period in which the perature of 1700 K. C-C was the baseline mate-
direction of the sunshade relative to the Sun rial for the TPS structure in the 2005 STDT
was unconstrained. The Solar Probe+ mission report. It is one of the few structural materials
keeps the spacecraft inside 1 AU throughout that can survive the high temperatures, provide
the mission. At those distances, the spacecraft acceptable specific strength, and meet the out-
must always be positioned to keep the shield gassing requirements. The top cover is flat and
in front of the Sun.

3-40
3.0 MISSION IMPLEMENTATION

coated with a low α/ε coating. An additional viewing the side of the shell, simplifying the
micrometeoroid layer has been added into the radiative heat transfer between the shell and
sunshield to provide particle impact protection. spacecraft. This angle is an important part of
For a structure of this size to meet its tight the thermal control of the shield. The tempera-
mass estimate, it is required to be made of tures on the front surface are conductively tied
very thin-walled shell sections. The nominal to the side surface, which is free to radiate to
lay-up for the C-C shell is a six-layer isotropic space. This heat loss from the side wall reduces
laminate, approximately 0.076-cm thick, and the bottom edge temperatures enough to be lo-
made of T300 carbon fabric. Coupon testing cally cooler than the rest of the bottom surface.
of a variety of C-C materials has been per- Coating. The Solar Probe+ shield design
formed to define their material and mechani- includes an optical coating on the forward-
cal properties. Thicker-walled sections will be facing surface. The specific requirements for
used at fastener locations. that optical coating include the following:
In earlier Solar Probe studies, fabrication • Optical properties—ratio of solar absorptiv-
oven volume limited the size of C-C pieces that ity to IR emissivity, α/ε < 0.6
could be made. As part of the new Constella- • Resistance to radiation damage—inert to
tion Program, a larger oven is being installed at radiation damage including proton and elec-
one vendor, giving them the ability to make the tron bombardment and extreme ultraviolet
shell in one piece. From a mechanical perspec- (EUV) exposure
tive, the choice between single and multiple • Chemical stability—thermodynamically
piece construction will be defined in a Phase A stable in the mission environments
trade study. The positive benefits of one-piece • Charging—no impact to spacecraft or in-
fabrication include a simpler structure and less strument operation or disturbance of local
assembly. The negatives include a higher reli- plasma environment
ance on cobonded stiffeners and limiting the • Structural properties—able to maintain
vendor pool. For a multiple-piece shield design, structural integrity after surviving launch
the number of joints is a function of oven size vibroacoustic loads, temperature extremes
and spacecraft symmetry. In the earlier radio- (273 to 1700 K), and temperature cycling
isotope thermoelectric generator (RTG)-based • Particulate impact—able to survive solar
designs, the three-piece shield worked better wind particulate impact with minimal deg-
than a four-piece because of the cutouts needed radation to optical–thermal performance
for the six jack stands. With the elimination of Because the temperature of the shield is
the RTGs and the jack stands supporting the driven by its optical property, even using bare
cone in the previous concept, the truss structure C-C requires that these properties be defined.
design has been simplified, and a new symme- Because of the importance of coatings on sys-
try will emerge. tem temperatures, our inability to test these
The shield shell encloses the insulation foam coatings in a flight-like environment, and the
that provides the thermal protection for the uncertainty in the effect of the mission envi-
spacecraft below. The shell encapsulates the ronments on the coating, a factor of safety of 2
foam, providing contamination protection. has been used for the optical properties. Gen-
Should the launch environment generate any erally, the measured values of α/ε have been
particulates from the foam, they will be cap- in the 0.2 to 0.3 range, making the design re-
tured by the pan and lid. The side of the shell is quirement 0.6. While both the measurements
angled so that the Sun, at closest approach, and the margin will be reviewed over the
does not impinge on the side of the shell. The course of the program, the present limit of 0.6
angle also keeps the spacecraft body from

3-41
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

represents an achievable and conservative re- foam pieces coincident through all of the lay-
quirement on a critical design parameter. ers. Testing is planned to determine whether
As part of the TPS Risk Mitigation effort, the foam sheets can be bonded together or
two potential ceramic coatings were found that whether an interstitial carbon felt is required.
met the requirements of the Solar Probe+ mis- The target temperature for the bottom of the
sion. Ceramic materials that are visibly white TPS is 350°C. The thickness of the foam is
generally provide the optical characteristics driven by both the surface temperature and the
compatible with the proposed shield passive allowable heat flow through the shield.
thermal management strategy. These charac- Heat transfer through the C-C foam is a
teristics are low solar absorptivity and high IR combination of conduction through the carbon
emissivity. Thermodynamic stability and structures and radiation across the empty
chemical compatibility with C-C are addi- spaces. At low temperatures, conduction domi-
tional differentiators that further narrow the nates and the material conduction is nearly
list of candidate ceramics. At the end of the constant with temperature. At higher tempera-
study, both aluminum oxide (Al2O3), com- tures, radiation dominates and the effective ma-
monly called alumina, and pyrolytic boron terial conduction rises rapidly with tempera-
nitride (PBN) were found to notionally satisfy ture. The thermal conductivity of the carbon
these basic characteristics. foam has been measured, and a temperature-
During the Risk Mitigation Study, both varying thermal conductivity is used in the
coatings were subjected to testing that ensured analyses. The thickness of the sunshield has
their viability when exposed to environmental been kept at the value from 2006 the Risk
extremes beyond those required for the Solar Mitigation Study. Similar front surface tem-
Probe+ mission. Testing was done to demon- peratures, coupled with the uncertainties in the
strate thermodynamic stability and to ensure a shield performance, form an argument for the
strong interface with the underlying C-C sub- present design to be kept “as is” until the sys-
strate. Optical property performance was suc- tem can be tested. As part of the TPS develop-
cessfully demonstrated through coupon testing ment effort, a representative section of the
after exposure to vacuum, temperature, and shield will be built and tested. Based on these
radiation environments. Coating mass loss and results, the shield thickness will be reassessed.
chemical interactions were characterized over Because this testing is planned to be completed
temperature. The impact of the application by preliminary design review (PDR), the mass
process on adhesion also was investigated. savings resulting from a reduced shield thick-
Finally, the effectiveness of nondestructive ness will be available to be used elsewhere in
evaluation techniques in finding coating de- the Solar Probe+ system.
fects was explored. Thermal Protection System Supports.
Insulation. The main insulating material in The supports that connect the TPS to the
the shield is a 15-cm-thick layer of reticulated spacecraft need to be thermal isolators to pro-
vitreous carbon (RVC) C-C foam. RVC insu- tect the spacecraft bus from the high tempera-
lation is an open-pore foam of pure carbon. It tures of the shield. Because the temperatures
has very low density and can withstand very at the base of the sunshield are expected to be
high temperatures. The foam is manufactured below 350°C, titanium flexures can be used to
in 2.5-cm-thick sheets that are cut into the support the entire shield. These flexures trans-
various shapes needed to fill the required vol- fer the mechanical loads to a truss system sup-
ume. The foam packing configuration is de- ported off the spacecraft bus. The titanium
signed to eliminate thermal shorts; i.e., no- flexures also provide an additional thermal
where in the TPS will there be joints between resistance, so temperatures at the top of the

3-42
3.0 MISSION IMPLEMENTATION

truss supports are beneath 300°C. At these rial down-select process. Detailed truss analy-
temperatures, normal composites can be used ses will be performed to optimize cross-
to reduce the mass of truss members. section stresses and weight. The truss finite
In the Risk Mitigation Study, a trade was element model will be validated with static
made of potential truss materials. The materi- tests of a single truss under limit loads pre-
als examined were high-temperature compos- dicted from the TPS finite element model.
ites, C-C, titanium, and AlBeMet. The results A key part of the structural design is the
of that trade show that composites are the best fact that the driving load case is the launch
option. C-C has a slightly lower density and environment. Because this case occurs at am-
could be higher strength. However, the uncer- bient temperatures, the TPS design is testable.
tainty in the cobonded joints and material Thermal Analysis. The TPS Risk Mitiga-
properties make this material higher risk. Ti- tion Study thermal design and analysis has
tanium has very good strength but almost been updated for the Solar Probe+ configura-
doubles the mass. Finally, AlBeMet has good tion and shows the TPS can meet all of its
strength and mass capabilities but is disquali- thermal design requirements. Temperatures
fied by the higher heat leak because of its and spacecraft heat flows are well within their
large thermal conductivity. limits. All key thermal design drivers have
Structural Analysis. The Solar Probe+ been simplified and addressed satisfactorily.
TPS design meets all the stiffness and strength Enough options continue to be available to
requirements. The elimination of the primary adjust the design in the future, if unforeseen
shield and jack stands greatly simplifies the contingencies make that necessary.
design from the earlier Solar Probe studies. The key issue with the thermal design is
The structural analyses from those studies verifying the performance of the shield in its
with respect to the secondary shield has been operating conditions. With its very low thermal
updated for the Solar Probe+ design. A key conductivity, the effectiveness of the insulation
result from that earlier work is the requirement is particularly sensitive to variations in the ma-
that all highly loaded C-C joints are bolted. terial properties, thermal shorts, and spaces be-
Cobonded joints have been shown to be prob- tween the foam blocks. System sensitivity to
lematic but will continue to be evaluated. All perturbations in all of these items will be in-
highly stressed joints use conventional materi- cluded in the detailed thermal analyses of the
als such as titanium and carbon fiber/polyimide TPS. As noted above, the design approach en-
with metallic fasteners. sures that the heat flow through the shield is
As part of the detailed Solar Probe+ design well below the required value. That fact, cou-
effort, the analysis of the TPS finite element pled with the incremental approach to the Sun,
model under comprehensive thermal profiles ensures that the TPS performance will be well
will be made. This analysis is required to as- characterized before the Solar Probe+ closest
sess the stresses induced in the different parts approach. As part of the developmental testing
of the structure caused by coefficient of ther- program for the TPS, a prototype shield section
mal expansion (CTE) mismatch. Vibroacous- will be built up and tested.
tic analyses will be performed with the TPS As part of the work done in the Solar Probe
finite element model to verify that current de- STDT and TPS Risk Mitigation studies, much
sign loads are not exceeded. Broadband acous- of the needed materials property work has
tic loads on the sunshield are required to as- been started. The thermal stability of the foam
sess coating adherence and C-C matrix integ- has been measured up to 2000 K. Two com-
rity. A comparable level of detailed analysis panies made thermal conductivity measure-
will be performed on the truss after the mate- ments at high temperatures. It was found that

3-43
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

just measuring the conductivity of the foam the six truss elements. The number, configura-
was very complicated. One of the normal ap- tion, and material of the flexures supporting the
proaches used did not work well because of shield from the truss elements will be defined.
the inability of the test plates to remain in con- Truss Design. The truss design will be ex-
tact with the foam without damaging it. Data amined as part of the spacecraft bus design
was obtained by using an alternative approach, study. The trade study will include the heat
which will be repeated during the flight pro- leak to the bus from the TPS, the mounting of
gram. Alternative insulation materials, such as the TPS, and a function of truss material, num-
the aerogel infiltrated carbon foam (AICF) bers of elements, and their configuration.
were identified that have a lower thermal con- 3.6.3. Thermal Protection System Devel-
ductivity but a higher density. The combina- opment Program
tion may allow a thinner shield, reducing the The TPS development effort is aimed at re-
overall TPS mass. Many C-C, thin-shell prop- ducing the risk associated with the Solar
erty measurements have been made. The test Probe+ TPS by performing a design, fabrica-
methodology for making these measurements tion, and testing cycle early in the program.
has been developed and will be used as part of This section describes the development of the
the flight program material test program. Solar Probe+ TPS. There are three broad
Trades. The following are the trades that steps. The first step is the analog and coupon
are planned in Phase A of the Solar Probe+ testing necessary to provide the basic engi-
study. These trade studies will be undertaken neering information needed for the TPS de-
to develop options for saving mass and in- sign. The second step is the development of
creasing reliability in the overall system. full-size system prototypes. Finally, the last
Micrometeoroid Protection Layer. The step is to build the spare and flight units and
spacing, thickness, and protection of the micro- integrate them into the spacecraft.
meteoroid protection layer will be analyzed by 3.6.3.1. Coupon and Analog Testing
the University of Texas, El Paso. They were a
part of the earlier studies, and they will update The Solar Probe+ TPS will include the first
the dust model for the new trajectory and per- use in space of a variety of C-C and insulating
form the micrometeoroid protection analyses. materials. Although no new development of
Pan-Cover Connections. The number, any materials is required by the program, the
configuration, and attachment of the connec- existing materials do not have the large data-
tions between the shield pan and the top cover bases of material properties needed for design
will be defined. purposes. Furthermore, some of the design
Layered Insulation Materials. The Solar questions involve applications where material
Probe+ design assumes that the insulating property data are inadequate to provide the
foam is all carbon. There may be options to full design performance of the system. In these
save mass by moving to lower-temperature cases, analog testing is done on representative
materials near the bottom of the shield. structures to get specific design information.
Bottom Side Surface Emissivity. A high 3.6.3.1.1. Material Test Program
emissivity [no multi-layer insulation (MLI)] C-C Coupon Testing. The Solar Probe+
design is the baseline approach. This study will will include a material test program to produce
look at potential savings through the use of specific mechanical property measurements on
low-emissivity coatings or high-temperature the C-C structural materials being used in the
MLI on the bottom of the shield. design. The Risk Mitigation Study, along with
Flexure Support Details. The baseline de- several external databases, has collected some
sign includes three titanium flexures on each of mechanical properties for thin-shell C-C and

3-44
3.0 MISSION IMPLEMENTATION

high-modulus materials. The Solar Probe+ structural analyses of the test item, the “as de-
coupon test program will build on these efforts livered” capabilities of the joint can be deter-
and produce material coupon test data for all mined and included in the analysis of the
C-C materials used in the design. Some high- flight system.
temperature material property testing will be Alternatively, C-C bolts can be used as an
done. attachment method. Some C-C bolts have
Insulation Testing. It is difficult to make been tested already in the Risk Mitigation
thermal conductivity measurements at the high Study but have been shown to have very poor
temperature needed for the Solar Probe+ mis- performance. Some of the failures have been
sion. Some testing has been done of the base- traced to the bolt head design. Alternatives
line carbon foam. That testing effort will be such as the Starfire bolts were not tested be-
extended to carbon aerogels and AICF. Test- cause of schedule limitations. However, the
ing will include material CTE and thermal availability of a C-C bolt with a significant
conductivity over the range in temperatures strength capability would provide a valuable
from 200 K to 2000 K. option in the design of large C-C structures.
3.6.3.1.2. Analog Element Testing Carbon Foam, Interstitial Material. The
Flexure Interface. Support for the sunshield carbon foam insulation provides the primary
comes through the titanium flexures into the C- thermal block in the TPS. To function cor-
C shell. The local area of the shell will have to rectly, the foam must be packaged in a way
be reinforced to provide the local strength that limits thermal shorts and prevents foam
needed in the C-C laminate. This joint will in- abrasion during testing and launch. To accom-
clude a variety of complex failure modes, so an plish both these goals, analog testing of the
analog test is the best method to verify the de- assembled foam is planned. A vibration test of
sign in the early stages of the program. A full- the packaged foam is planned to ensure that
size section of the flexure and shell will be built the interstitial material used is optimized for
up and strength tested to verify the load path the sunshield. The issue to be defined is the
from the top of the flexure into the C-C shell. minimum mass than can provide the required
Cobonded Joints/Carbon Bolts. Some protection during vibration testing.
parts of the C-C structure pieces require at- 3.6.3.2. Prototype Development
tachment with fasteners. The use of metallic Prototype shield development is aimed at
materials at these high temperatures is compli- bringing the TPS to a technology readiness
cated by the CTE differences. Metallic fas- level (TRL) of 6 by the system PDR. To ac-
tener designs that use flexible elements will be complish this goal, two parallel development
developed during Phase A. Other options that projects are planned. The first effort is to build
do not involve the use of different materials and mechanically test a full-size TPS. The sec-
will be evaluated in Phase A. Two such op- ond is to build and thermally test a shield seg-
tions are available for C-C. ment. The dual approach is based on the limita-
The first is a cobonded joint where the two tions of the available test facilities and the fact
C-C pieces are built separately and then car- that the driving mechanical and thermal design
bonized together during processing to produce cases are different.
a functional bond. The capabilities of the re- Shell Prototype Development. The shell
sulting bond joints depend greatly on the spe- development brings the TPS to a TRL of 6 by
cific joint configuration and the particular load building and testing a full-size prototype
directions. To verify the capabilities of the shield. A complete TPS will be subjected to
joint, a full-size analog test item is built and vibration and acoustic testing at room tem-
structurally tested. By combining test data and perature. This testing will verify that the me-

3-45
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

chanical design of the TPS is adequate to sur- temperature testing has been aimed at defining
vive the launch environment. The test also will the ablation parameters and quantifying mate-
be used to correlate the associated analytical rial erosion. Testing was performed by using
models. This work will be completed by the arc jets in a high-enthalpy flow field. They
program PDR, allowing any design modifica- also used the Sandia Solar Tower for testing
tions to be incorporated into the flight units. the radiation transmission of TPS coupon ma-
Both APL and Goddard Space Flight Center terials, but they have not tested the TPS there.
(GSFC) have test facilities that are capable of The consensus was that it will be extremely
supporting this developmental testing. difficult, if not impossible, to subject the full-
Shield Prototype Development. The sec- size probe to heating on the order of 100
ond of the developmental prototypes is used to W/cm2. The best option would be to use a ra-
verify the thermal performance of the shield. diant lamp facility, such as the ones at NASA
This unit is aimed at verifying the protection JSC or NASA Dryden, as described in the
the shield will provide during the closest ap- preceding paragraph.
proach to the Sun and the resulting heat leak 3.6.3.3. Spare and Flight Shield Fabrication,
to the spacecraft. The configuration is required Assembly, and Testing
to include only a representative section of the Directly after PDR, the final design effort
shield because there are no test facilities pres- for the TPS would begin. Because of the long
ently capable of taking the entire shield front lead time for the fabrication of the C-C parts,
surface to the required temperature via an ap- the final design is required before critical de-
plied radiative heat flux. As noted above, in sign review (CDR). The TPS plan includes
flight, there will be ~4 MW incident on the spare and flight units to be designed, fabri-
shield’s front surface. cated, assembled, and tested in a sequential
There is a test facility at Johnson Space manner. By sequencing the two builds, the
Center (JSC) that is used for thermally testing first unit acts as a pathfinder for the second.
the Space Shuttle surfaces protected by insu- However, the overall schedule does not allow
lating tiles. The facility includes a “hot wall,” the two builds to be in done in series. There-
~30 in. square, inside a vacuum chamber. The fore, the lessons learned during the first build
limited size of the hot area requires a section will be more in the fabrication, assembly, and
of the shield be built as a test specimen. The testing areas. The first build will be important
test section would contain the C-C pan and in trying out procedures and facilities.
cover, including an edge section. Several foam During spacecraft-level testing, a shield
packing approaches would be included. The simulator is required to support the thermal
test unit would include truss and spacecraft vacuum testing of the spacecraft bus. As noted
bus simulators that would allow the heat flow above, it is not possible to take the entire
through the test item to be measured shield to temperature in any existing test facil-
As with the mechanical prototype unit, the ity. Therefore, a shield thermal simulator will
thermal testing would be completed and the be designed and built that will represent the
test data correlated with the analytical models temperature predicted on the bottom surface of
by program PDR. This timing of the prototype the shield. The spacecraft test will verify the
unit would allow any design updates to be in- performance of the bus thermal control system
cluded into the flight designs. in the presence of the shield environment. The
As part of the planning for the developmen- capability of the shield to produce those tem-
tal testing, a discussion was held with the peratures will be verified by the “hot-wall”
Mars Science Laboratory TPS team to review testing at JSC.
their testing efforts. The bulk of their high-

3-46
3.0 MISSION IMPLEMENTATION

3.7. Thermal Control System purpose. The cells will remain exposed to the
3.7.1. Spacecraft Bus Sun at all times when the spacecraft is outside
3.7.1.1. Requirements 0.25 AU, in order to limit their cold temperature
The spacecraft bus is required to operate be- to above 20°C. The cooling system is designed
tween 0.044 AU and 1 AU, with protection to maintain the cells within acceptable tempera-
from the Thermal Protection System (TPS). ture ranges at all times. It will consist of the heat
The spacecraft is broken up into eight thermal exchanger, mechanical pumps, tubing, radiators,
zones, which include the bus, battery, high- and heat pipes. There will be a valve in the
gain antenna (HGA), star trackers, TPS, pri- pumped loop system to keep the fluid out of the
mary solar array, secondary solar array, and radiator during the cold case, which will keep
secondary solar array cooling system. The the fluid from freezing and reduce the required
temperature ranges listed below are both op- heater power.
erational and survival: they must be main-
3.7.1.2. Thermal Protection System Interface
tained throughout the Sun distances listed
above and a mission duration of 7 years. The The spacecraft must keep the TPS pointed
orbit period during the closest approaches to toward the Sun at all times but is allowed to
the Sun is approximately 88 days. roll as needed. The TPS interface with the bus
Because the propulsion system is an integral will be at the end of the truss structure that is
part of the bus, the bus temperature limits are used to support the TPS from the bus. The TPS
20°C to 40°C. All bus components (including has a requirement to have no more then 50 W
the instrument interfaces) will be held to these of heat flow to the spacecraft bus from the
temperature limits with the exception of the combination of radiation and conduction. The
battery, HGA, and star trackers, which will be interface has been designed to minimize the
independently controlled. The battery tempera- transition structure assembly (TSA) conduc-
ture limits are 5°C to 30°C, the HGA tempera- tion from the TPS into the bus. The bus and
ture limits are –90°C to 250°C, and the star TPS also will include blanket “blockers” in or-
der to reduce the radiation interchange between
tracker temperature limits are –20°C to 60°C.
elements as much as possible. The items sup-
The primary array temperature limits are
ported by the TSA must be mechanically sup-
–90°C to 180°C. These arrays will be tilted off
ported but thermally isolated from the truss
the Sun at closer solar distances in order to limit
system. The attachment scheme will require a
the incident solar flux. At 0.25 AU, they will be
detailed analysis of the trusses to determine the
retracted inside the umbra and will stay in the
truss temperature at the mounting location.
shadow until the spacecraft reaches 0.25 AU
This temperature will have the biggest influ-
again, at which point they will be redeployed.
ence on the material chosen for the isolation
This sequence will occur during every orbit in
system. Preliminary temperature estimates in-
order to provide the necessary power to operate
dicate the industry-standard carbon fiber cy-
the bus and limit the required battery size.
anate ester materials have sufficient high-
The secondary solar cells will be on a paddle
temperature strengths at 230°C. Newer polyim-
that (starting at 0.25 AU) is retracted as the
ide matrix composites are available if design
spacecraft gets closer to the Sun, in order to
trades indicate a lower mass system can be
maintain as constant a solar flux and power out-
achieved by reducing the amount of thermal
put as possible while attempting to maintain the
isolation available in the titanium flexures.
cells below 100°C. In order to maintain them
below this temperature, a heat exchanger and 3.7.1.3. Block Diagram
mechanical pump loop will remove heat and Figure 3.7-1 shows a preliminary block dia-
transfer it to radiators solely designed for this gram of the thermal subsystem, including all
thermal hardware. All hardware depicted in

3-47
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

Figure 3.7-1. Thermal control system block diagram.

red is considered thermal hardware. This in- to maintain the bus temperature above 20°C in
cludes all temperature sensors, heaters, multi- the cold case, the overall power in the bus
layer insulation (MLI), heat pipes, radiators, cannot go below 270 W, meaning that when
heat exchangers, mechanical pumps, and tub- the bus power is below 270 W, heaters must
ing. The thermal interfaces for all components be turned on to maintain 270 W in the bus at
(conductive or isolative) and blanket require- all times. The baseline currently has a mini-
ments also are shown. The component place-
ments are not representative of the actual
Table 3.7-1. Spacecraft bulk temperature (°C) as a
placement on the spacecraft. function of heat into the bus.
3.7.1.4. Thermal Analysis Baseline
A worst-case hot and cold thermal analysis Heat Source 9.5 RS Cold Case
has been completed on the bus. This analysis Electrical Dissipation 300 170
Strut (Conduction) 4.0 0.0
determined the sizing of radiators and required
Strut (Radiation) 6.1 0.0
heater powers and is summarized in Table 3.7-1 TPS Through Bus Sides 4.0 0.0
and shown graphically in Figure 3.7-2. The TPS Through Top MLI 10.0 0.0
table and plot show that in order to maintain PWI 19.0 0.0
the bus temperature below 40°C in the hot Heater Power 0.0 100.0
case, the bus requires at least 0.64 m2 of radia- Total 343.1 270.0
tor area. With that much radiator area, in order Spacecraft Bulk Temperature 38 20

3-48
3.0 MISSION IMPLEMENTATION

(PWI) antennas. The an-


tennas will stick out of
the umbra into the
sunlight and will heat up
to approximately 1500°C
at the hot end. The anten-
nas are designed in order
to have a long distance
inside the umbra before
reaching the bus, which
allows them to cool to
close to bus temperatures
before reaching the bus
and therefore limits the
heat conducted into the
Figure 3.7-2. Spacecraft bulk temperature as a function of heat into the bus. The long distance to
bus. the bus also limits the
mum bus power of 190 W, and therefore the amount of radiation from the hot end into the
maximum heater power required is 80 W. bus radiators (which are near the bottom of the
As stated above, the required radiator area bus, close to the mounting location of the an-
is 0.64 m2. In order to reduce the heat load tennas). The current calculations show that the
from the TPS into these radiators, they will be three antennas will input ~19 W total into the
placed as low on the bus (away from the TPS) bus in the hottest case.
as possible. Once all components are placed 3.7.1.6. Trades
and the TPS final design is completed, the One possible trade study to be conducted in
heat load from the TPS and the instruments Phase A is an optimization of the secondary
will be analyzed again, and the final radiator power array temperature control through mov-
size and placement will be determined. ing a part of the TPS structure instead of mov-
The electronics box placement inside and ing the secondary solar array. This move would
outside the spacecraft will depend on box eliminate the need for flexible hoses in the
footprints, field of view (FOV), and thermal cooling system, but it also would require a
criteria. All the high-powered electronics mechanism that could work at 1500°C. Be-
should be placed as close to the radiator as cause of the complexity of any mechanism at
possible and may need either doublers or heat such an elevated temperature, the moving array
pipes to spread the heat evenly into the radia- panel was chosen as a baseline but further stud-
tor. The battery will have its own radiator and ies will be completed.
will be controlled separately from the rest of Another trade study still to be completed in-
the bus. The star trackers also are independ- volves the working fluid in the secondary solar
ently controlled from the bus and will have panel cooling system. Ammonia was baselined,
their own heater power. but other fluids will be studied to determine the
3.7.1.5. Instrument Interface best solution. The study will look at heat capac-
The instruments will be conductively tied to ity and operating temperatures among other
the bus with an interface temperature range of factors like corrosive properties and chemical
20°C to 40°C. The only instrument expected interactions with the pipe material.
to input to the bus any heat attributable to so- A third trade study involves the material for
lar exposure is the plasma-wave instrument the secondary solar panel cooling system radia-
tor panels. The current baseline is heat-pipe-

3-49
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

embedded honeycomb panels with aluminum get power off the arrays (the MESSENGER
face sheets. The trade study will look at compos- arrays have produced power up to 74° off the
ite, AlBeMet, and solid aluminum radiators. A Sun). A primary panel test bed is planned and
composite radiator was flown on the Compact will be used to test the tilt angle at which the
Reconnaissance Imaging Spectrometer for Mars arrays will no longer produce power.
(CRISM) instrument on the Mars Reconnais- 3.7.3. Secondary Solar Array
sance Orbiter (MRO) spacecraft. It was able to A secondary solar array will be used to
save a factor of over 3.5 on mass over the alu- generate the power needed for the spacecraft
minum alternative. The directionality of heat when it is in close to the Sun (from 0.25 AU to
flow sometimes presents challenges for interfac- perihelion and back to 0.25 AU). This solar
ing with composite radiators, which could to be distance was chosen because, as stated above,
problematic in the current Solar Probe+ configu- the current state-of-the-art solar array technol-
ration because of the mechanical pump loop in- ogy (the MESSENGER spacecraft) is capable
terface (although further definition of the inter- of withstanding solar fluxes up to 0.25 AU
face still is needed). AlBeMet has higher ther- (the MESSENGER arrays have already sur-
mal conductivity and specific heat than alumi- vived solar distances of 0.33 AU).
num and therefore is the thermally preferred A knife-edge extension has been added to
choice between those two. However, AlBeMet the TPS shield structure in two places. There
is heavier then aluminum and health hazards will be one secondary panel under each knife
involved in working with beryllium may make edge. Both the knife-edge and secondary array
this approach less effective than aluminum. panel radial and axial positions have been de-
3.7.2. Primary Solar Panels signed so that, at 9.5 RS, one row of exposed
The primary spacecraft solar panels are cells will see half of the solar disk, which cuts
based on the MESSENGER high-temperature down the solar flux hitting the panel by almost
solar array panels. The panels have a 33% a factor of two and allows the cooling system
packing factor in order to survive at higher to reject the heat. The secondary panel will
solar flux. This means that the substrate sur- use high-intensity solar cells, which are simi-
face contains 33% solar cells and 67% optical lar to those used for Earth-based concentrator
surface reflectors (OSRs). The arrays will be array applications and will be qualified for use
tilted off the Sun as the Sun distance de- in space with an extensive test program
creases, which will allow the temperature to planned for Phase A. The cells will be cooled
remain below the 180°C temperature limit for with a mechanical pump loop to transport the
the array. An analysis was completed to de- heat to radiators with heat pipes in them to
termine if a different packing factor could be further spread the heat. Each secondary panel
used to allow MESSENGER-like arrays to get will have its own mechanical pump and pump
closer to the Sun, but the increased mass be- loop as well as 1 m2 of radiator area split into
cause of the larger solar arrays (and the fact three separate panels.
that they could not reach 0.044 AU, meaning The solar cell stack-up will be a critical area
some form of secondary power generation still in this design because of thermal gradients built
was required) made this a nonviable option. up through the different layers and the maxi-
It will be possible to use these arrays in to mum allowable temperature of the cell, which
0.25 AU, but past that solar distance they is 100°C according to the manufacturer. A test
would get too hot and therefore have to be re- program will be put in place to determine the
tracted and stored within the umbra. At the pre- actual maximum temperature for the cell as
sent time, it is believed that the array can be well as the efficiency at that temperature. The
tilted to approximately 75° off the Sun and still baseline stack-up is a triple-junction solar cell

3-50
3.0 MISSION IMPLEMENTATION

the fluid from the mechanical pump will pass


to remove the heat. The fluid then will go
through the radiators to be cooled and back to
the pump. The current baseline for the fluid is
ammonia, although many other fluids will be
looked at in the next phase to determine the
best possible solution.
To ensure that the concept works as ex-
pected, the heat exchanger system will be built
up into a solar array substrate fluid test bed.
This test bed will be used to ensure that the
design can get the heat out of the solar cells
and into the fluid without a large temperature
difference. It also will be used to determine
the fluid flow rate required to keep the thermal
Figure 3.7-3. Baseline solar concentrator cell tem- gradients between cells in the same strings as
perature stack-up. low as possible.
The mechanical pumps that have been cho-
with a germanium substrate, Kapton, cell, and sen for the baseline are built by Pacific Design
coverglass (with adhesive in between layers). Technology (PDT) and have heritage from the
Figure 3.7-3 shows the expected Solar Probe+ Mars Exploration Rovers (on which they have
temperature difference through the baseline been working for several years). There also is
stack-up. a life test ongoing at PDT in which the pumps
A test setup will be created to test other pos- have been operating continuously for 5.5+
sible cell stack-ups to determine which one will years. A development plan would be put in
result in the lowest cell temperature and the place to develop the system required for Solar
lowest temperature difference through the Probe+, and a life test would be put into place.
stack. This test setup will be exposed to solar PDT also will be consulted in the determina-
illumination at various levels to determine tion of the fluid and the fluid line lengths to
thermal gradients in both the 1-AU case and the ensure that the pumps can handle the load.
0.0443-AU case. Also, a subcontract is planned Flexible hoses currently are a part of the
with one of the manufacturers of this technol- baseline design. The flexible hoses are re-
ogy for terrestrial applications in order to use quired because the secondary solar panel must
their expertise in the design and qualification. be retracted as the spacecraft gets closer to the
Another critical area is the cooling system, Sun. An extensive test program is planned to
which will be composed of the heat exchangers ensure that the hoses will work in all different
underneath the cells, mechanical pumps, flexi- configurations and over the required number
ble hosing, and multiple radiators with embed- of cycles.
ded heat pipes. The current baseline for the heat As stated previously, the radiators currently
exchangers is copper because of its coefficient are heat-pipe-embedded honeycomb panels
of thermal expansion (CTE) match to the cell with aluminum face sheets, but other materi-
substrate and its thermal conductivity. Other als, such as composite, AlBeMet, and solid
materials are being considered in order to re- aluminum also will be considered. Composite
duce mass. The leading candidates are cur- has the advantage of much lower mass but the
rently AlBeMet and aluminum. disadvantage that it is difficult to get the heat
The heat exchangers will have channels or into the fibers, unless you can add it to the fi-
pipes under each string of cells through which ber end, because the across-fiber conductivity

3-51
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

is ~10% or less of the in-fiber conductivity. solar array panels for power generation and a
AlBeMet has a high thermal conductivity and lithium-ion battery for energy storage. Two dif-
a high specific heat, which make it a better ferent types of solar arrays are used, each opti-
choice then aluminum, but the disadvantage is mized for a different range of distances from
that it is heavier and there are safety concerns the Sun. Power electronics are contained within
when working with the material. the power system electronics (PSE) box and the
The radiators will have embedded heat solar array junction boxes (SAJBs). A block
pipes in order to spread the heat and make the diagram of the power subsystem is shown in
radiator more efficient. This network will be Figure 3.8-1. A detailed trade study on the sec-
designed in order to get the maximum possible ondary solar array concept was performed, and
heat transfer across the surface for the lowest the results are presented in Appendix C.
possible mass. Also, the interface between the The power bus voltage is unregulated, and
pump piping and the radiator will be studied to all subsystems attached to the bus will be de-
find the most efficient way to get the heat out signed to work between 22 V and 35 V. The
of the piping and into the heat pipes. actual battery-dominated bus voltage will vary
The final test bed will be a solar concentrator less than these subsystem requirements. The
test bed. It will consist of all the pieces of the peak power tracking electronics, with flight
full system put together to make sure that eve- heritage from the MESSENGER and STEREO
rything works together as designed. It will use spacecraft, isolates the bus voltage from the
qualification units of hardware and will be built solar array voltage and maximizes solar array
up and tested in a flight-like environment. output over the mission’s widely varying oper-
3.8. Power System ating conditions.
The power subsystem is based on a peak Normally, only the primary solar array or the
power tracking architecture using photovoltaic secondary solar array is used as the main

Figure 3.8-1. Power system block diagram.

3-52
3.0 MISSION IMPLEMENTATION

source of power at any particular time, depend-


ing on the distance from the Sun. When the
primary solar array is not in use, it is retracted
and not exposed to the Sun. The two solar array
systems are diode-isolated from each other, and
both feed the common set of peak power
tracker (PPT) modules within the PSE.
The power subsystem is designed to pro-
vide 482 W of power between 0.9 AU and
0.044 AU distance from the Sun. Between 0.9
AU and 1 AU, less power is available. Space-
craft operational requirements in this regime
will be reduced primarily by lowering power
allocated to the telecommunications subsys-
tem. Sun–probe distance is greater than 0.9
AU only immediately after launch and around
aphelion of the first few orbits, so there will be Figure 3.8-2. STEREO PSE.
no impact on science return.
The power subsystem is single-fault toler- sharing among the modules. This approach
ant. Full redundancy is provided for power also allows the PSE to continue functioning,
control electronics and command/telemetry even if one of the PPT modules fails, although
paths. Subsystem sizing is such that the loss of with less total output power. The PPT modules
a battery cell or a string of solar cells would and controller design were flown successfully
not affect mission performance. The mission on MESSENGER and the two STEREO
requirements would still be met even with a spacecraft. The peak power tracking algo-
failure of one of the multiple, parallel peak rithm, proven on those missions, is contained
power tracking modules. within the IEM main processor. The STEREO
3.8.1. Power System Electronics PSE is shown in Figure 3.8-2.
The power system electronics (PSE) box The battery charge control electronics mini-
implements solar array peak power tracking mizes stress on the lithium-ion battery by re-
and battery charge control. The PSE provides ducing charge current when the battery ap-
primary power to the power distribution unit proaches a high state of charge, based on am-
(PDU) and has a serial digital com- pere-hour integration. Also, battery voltage
mand/telemetry interface with the integrated limiting causes the battery current to taper to a
electronic module (IEM). The PDU and IEM low value close to the end of charge. The bat-
are part of the avionics subsystem. Within the tery charge control parameters are command-
PSE, there are six PPT modules, two PPT con- adjustable as a contingency in case of drift in
troller slices, two command/telemetry inter- the control electronics or to help compensate
face slices, and battery interface slices. Power for battery aging. Bus overvoltage protection
bus filtering also is included within the PSE. also is provided as an additional control loop.
Each PPT module contains a pulse-width- The PSE box is constructed as a modular slice
modulated buck-topology DC/DC converter design, where each slice consists of a printed
and can support up to 90 W at its output. The circuit board housed in its own mechanical
use of current mode control within each PPT frame. The slices are mechanically stacked and
and centralized control (on primary and re- bolted together to form the box. The slices are
dundant sides) of all PPTs ensures current electrically connected by using a wiring harness
external to the box for power and signals.

3-53
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

Two SAJBs are used with the PSE. Each string isolation diode. The solar cells are tri-
SAJB receives power from two solar array ple-junction, gallium arsenide (GaAs)-based
wings. One SAJB is used for the primary solar cells with an active area of 12 cm2, and each
array and the other for the secondary solar ar- cell includes an individual bypass diode. The
ray. Each SAJB feeds power to the PSE PPT coverglass is 0.15-mm-thick cerium-doped
modules. Included in the SAJB are solar array microsheet, type CMG, with dual antireflec-
string isolation diodes and solar array current tive coating. The total active area of cells on
sensors. The SAJB design is similar to that all of the solar panels is ~1.5 m2.
flown on MESSENGER and STEREO. Optical surface reflectors (OSRs) are evenly
3.8.2. Solar Arrays distributed among the cells at a ratio of two
Solar Probe+ uses two solar cell arrays, each OSRs for each solar cell. This is the same
optimized to work over a different range of technique that is used on the MESSENGER
Sun–probe distances. The primary array is used solar array. As successfully demonstrated on
between 1 AU and 0.25 AU and is based on the MESSENGER, the use of OSRs and the tilt of
MESSENGER solar array design. The secon- the panels help to maintain the primary array
dary array uses high-intensity concentrator solar solar cell temperature below 180°C, which is
cells mounted on an actively cooled panel and is well within the temperature range to which
used to generate power within 0.25 AU. this design has been qualified. The MES-
SENGER solar array is shown in Figure 3.8-4.
3.8.2.1. Primary Solar Array
The primary solar array wings are retract-
The primary solar array consists of two de- able by command. At Sun distances less than
ployable, articulated wings. The primary solar
array wings are deployed from behind the heat
shield and oriented at an angle to the Sun as
shown in Figure 3.8-3. As the spacecraft ap-
proaches the Sun, the primary array is tilted to
maintain the array within flight-allowable
thermal limits while meeting spacecraft power
requirements.
Each solar panel contains 32 strings of solar
cells. Each string has 39 series cells and a

Figure 3.8-3. Primary solar array configuration (both


stowed and deployed). Figure 3.8-4. MESSENGER solar arrays.

3-54
3.0 MISSION IMPLEMENTATION

0.25 AU, the secondary solar array is used and signed and optimized for operation at high so-
the primary array must be retracted behind the lar flux, as described later in this section. Each
spacecraft heat shield because of the high panel is attached to its own linear positioner
temperature. actuated by a stepper motor. The panels can be
Although improved solar cells with increased completely retracted behind the spacecraft
efficiency are expected in the future, the power heat shield, extended beyond the shield, or
analysis conservatively assumes a 28% mini- positioned in between in fine increments (a
mum average efficiency (under standard test small fraction of a cell width) by command.
conditions of 28°C, 1 Sun, air mass zero, begin- As the positions of the panels are adjusted, the
ning of life), as such space-flight-quality pro- quantity of solar cell strings exposed to direct
duction-run cells are presently available and sunlight will vary. The amount of exposure is
have been qualified and flown. Although the therefore controlled as a function of Sun dis-
spacecraft performance will benefit from further tance. The quantity of strings that are illumi-
improvements in solar cell efficiency, it is rec- nated is minimized to reduce heat load but suf-
ognized that any newly developed cells will ficient to meet the power requirements. This
need to be qualified for the unique environment concept is illustrated in Figure 3.8-5.
of this mission. High-intensity, high-temperature A preliminary design of the secondary solar
testing will be performed on sample solar cells array panels contains 50 parallel strings of solar
and a non-flight-qualification solar panel. This cells with 27 cells per string and an isolation
panel also will be subjected to thermal cycling diode in series. These array design parameters
and electrical performance tests. are preliminary and will be updated with data
3.8.2.2. Secondary Solar Array collected during the cell qualification testing
described later in this section. A bypass diode is
Used at Sun distances between 0.044 AU
connected in parallel with each cell, and the di-
and 0.25 AU, the secondary solar array con-
odes are located on the back side of the panel.
sists of two small retractable panels. The pan-
The cell strings are arranged on the panel so that
els contain a planar array of concentrator
their series direction is parallel to the edge of the
photovoltaic cells on an actively cooled sub-
heat shield, thus ensuring that all cells within
strate that maintains cell junction temperatures
each string will be exposed to approximately the
below 120°C (see Sections 3.5 and 3.7 for de-
same illumination level, which becomes more
tails of the mechanical and thermal designs).
important closer to the Sun, where only one or
The cells on the secondary solar array are de-
two strings are exposed on a panel and the illu-
mination level on those
strings is varied by fine po-
sitioning of the panel. The
portion of the spacecraft
heat shield just above each
secondary solar array panel
forms a straight knife edge,
which improves the uni-
formity of illumination on
the cells within the strings
that are in the penumbra
between full exposure and
the umbra.
The solar cells used for
Figure 3.8-5. Secondary solar array concept. the secondary solar array

3-55
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

are triple-junction GaAs-based cells optimized was estimated by using a conservative combi-
for high-intensity illumination and high cur- nation of specifications for spaceflight-
rent density. These cells use the same epitaxial qualified solar cells and concentrator photo-
growth as high-efficiency cells with space- voltaic cells at a 120°C operating temperature.
flight heritage. The gridlines and contact met- Estimated efficiency included derating for
allization is the same as used for concentrator losses caused by assembly, coverglass and
photovoltaic cells, which have been used for coating, ultraviolet radiation, charged particle
terrestrial applications with optics having a radiation, and micrometeoroid impacts. The
very high concentration ratio. Although opti- power calculations for the concentrator photo-
cal concentration is not being used on this so- voltaic cells also include a loss factor to ac-
lar array, the close proximity to the Sun results count for the difference in spectral characteris-
in high flux, which the cell must accommo- tics between the terrestrial (air mass 1.5) con-
date. For this application, the illumination in- ditions for which these cells have been charac-
tensity varies between 16 and ~250 equivalent terized and the space environment (air mass
Suns. This intensity is well within the range zero). In addition, the effect of peak power
for which concentrator photovoltaic cells have tracker inaccuracy was included.
been designed. Characterization tests for con- At 0.25-AU Sun distance, both secondary
centrator photovoltaic cells have been per- solar array panels are almost fully exposed. In
formed at up to 1000 equivalent Suns. addition to the required load power margin,
Each cell has an active “aperture” area of the design includes two extra strings of cells.
0.989 cm2. The cell front-side metallization, At closest approach to the Sun of 0.044 AU,
dual bus bars, and gridlines are designed to the secondary arrays are retracted to expose a
minimize resistive losses to accommodate the total equivalent cell area of 34.04 cm2.
relatively high current. Wide electrical inter- Although the specific cell design proposed
connects with stress-relief and multiple-welded for the secondary solar panels has not been
contact points are used to conduct the relatively space-qualified, the epitaxial growth is the
high current between cells. OSRs and electrical same as next-generation triple-junction solar
insulation cover the cell-to-cell electrical inter- cells that have spaceflight heritage on an ex-
connects to minimize thermal load. perimental satellite. These solar cells have low
The coverglass, which is cerium-doped mi- series resistance and are well suited for opera-
crosheet with dual antireflective coating, is tion at high intensity. The cell manufacturer
used for radiation protection and optical filter- has stated that, during 2008, these next-
ing. Tradeoffs will be performed to optimize generation cells will undergo spaceflight-
the coverglass thickness and type of coating. qualification according to American Institute
The thermal effects of the coverglass thickness of Aeronautics and Astronautics (AIAA) stan-
and filter coating are much more dominant dard S-111. 22 This qualification testing in-
than typical for this application because of the cludes temperature/humidity exposure, top
high solar flux and will be studied as part of and bottom contact-weld integrity, electron
tradeoffs to determine specifics of the cover- and proton irradiation, and accelerated life.
glass design. The cell-interconnect-cover (CIC) assembly
Under the predicted range of operating con- qualification testing includes electrostatic dis-
ditions, the effective conversion efficiency charge sensitivity, mechanical strength, and
varies between 13% and 20% with a junction
temperature of 120°C, resulting in additional 22
Qualification and Quality Requirements for Space So-
2259 to 1897 W of thermal energy absorbed lar Cells, AIAA Standard S-111-2005, American Insti-
by the cells. Effective conversion efficiency tute of Aeronautics and Astronautics (January 2005).

3-56
3.0 MISSION IMPLEMENTATION

UV radiation effects. Solar cell characterization flown on many spacecraft, including NASA’s
tests include postradiation current–voltage (I– five THEMIS spacecraft. It also has been se-
V) over temperature, dark I–V, quantum effi- lected for NASA’s Living With a Star Solar
ciency, thermo-optical properties (absorptance Dynamics Observatory (SDO; a 12-year geo-
and emittance with cover), and capacitance. synchronous orbit mission) and the Lunar Re-
For the proposed cells, to use gridlines and connaissance Orbiter (LRO). In addition, this
contact metallization as is used for concentra- battery type has flown on many European
tor photovoltaic cells, most of the same cell- Space Agency spacecraft, including Rosetta (an
qualification tests will be performed. How- 11-year interplanetary mission), Mars Express,
ever, it is expected that abbreviated radiation and Venus Express.
testing can be performed, because the contact Each cell includes built-in safety features to
metallization should not have a major effect open-circuit the cell in case of excessive cur-
the radiation performance. Sample cells will rent or high temperature, and the cell and bat-
be radiation-tested under several conditions tery designs meet the range safety require-
for verification that the performance matches ments. The quantity of parallel strings has
the full characterization testing performed for been selected to provide fault tolerance so that
the next-generation solar cells. the battery requirements can be met even with
In addition, a comprehensive test program a failed string of cells.
has been planned to ensure that the cells and Because of the relatively large-scale pro-
related components will withstand the space duction runs for this type of lithium-ion cell,
environment unique to this mission. The test their manufacturing uniformity, and the use of
program includes the use of a close-match air- computer-aided cell selection and matching,
mass-zero-spectrum solar simulator and high- the cells within the battery are very well
concentration optics with a space simulation matched in performance. The vendor has per-
thermal vacuum chamber. The solar cells will formed lifecycle testing and demonstrated the
be tested for electrical performance after ex- uniformity and self-balancing of the cells with
posure to charged particle radiation as a func- cycling so that external cell-balancing elec-
tion of temperature and intensity. In addition tronics are not required. If a cell were to fail or
to cell-level testing, a non-flight-qualification have an anomalous voltage divergence, a
solar panel with multiple strings of cells will switch within the cell is designed to open and
be constructed and subjected to thermal vac- isolate the string from the others in the battery.
uum cycling and thermal vacuum balance Under normal conditions, the most signifi-
tests. Structural integrity, thermal gradients, cant discharge for the battery will be during
and electrical performance will be verified. At launch. The battery is conservatively sized so
higher levels of assembly, acoustic and vibra- that the depth of discharge (DOD) will not ex-
tion testing will be performed. ceed 50%, even if there was no solar array
3.8.3. Battery power until the primary solar array panels are
The battery has a nameplate capacity of 20 deployed and oriented to the Sun. The battery
Ampere-hours and contains space-flight- also supports short-term peak loads. In addi-
qualified lithium-ion cells. Advantages of lith- tion, the battery provides a low impedance
ium-ion cells are their high-energy density, source to clear a fuse in case of a load current
good cycle life, nonmagnetic materials, and fault. There are no repetitive eclipses ex-
successful spaceflight heritage. The battery pected, and there is no shadowing from ap-
consists of multiple parallel strings, with each pendages under normal conditions.
string containing eight ABSL 18650HC cells in A nonflight “work” battery is used during ini-
series. This modular, parallel string approach tial spacecraft integration and testing (I&T). The
for lithium-ion batteries has been successfully flight battery is installed before spacecraft envi-

3-57
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

primary array is folded into the umbra behind


the spacecraft heat shield. As the spacecraft ap-
proaches the Sun and illumination intensity in-
creases, the secondary array is retracted behind
the shield as needed to expose enough photo-
voltaic cell area to maintain the desired electrical
power output level. Figure 3.8-6 shows the
equivalent number of cells on the secondary
panel that are illuminated as a function of dis-
tance of the spacecraft from the Sun.
3.9. Avionics System
Figure 3.8-6. Solar cells illuminated as a function 3.9.1. Avionics Suite
of solar distance. The Solar Probe+ avionics system, as shown
in Figure 3.9-1, consists of the integrated elec-
ronmental testing. The battery conditioning/test tronic module (IEM), remote interface units
equipment has multiple protective features to (RIUs), and the power distribution unit (PDU).
meet the safety requirements, and the battery is The IEM contains the main processor and inter-
operated in accordance with the established faces to instruments and other subsystems. The
battery-handling plans and procedures. off-the-shelf RAD750 processor supports com-
3.8.4. Power System Performance manding, data handling, data storage [using the
Power analysis was performed taking into solid-state recorder (SSR)], and guidance and
account solar array optical, assembly, and wir- control (G&C). The IEM is an evolutionary de-
ing losses; temperature effects; degradation sign based on the compact peripheral compo-
due to ultraviolet radiation; and charged parti- nent interface (cPCI) backplane bus that has
cle radiation. The analysis also includes the flown on MESSENGER and STEREO and is
effects of intensity variations with Sun dis- planned for use on RBSP. The standard cPCI
tance and other power system losses, includ- bus allows great flexibility in combining appro-
ing solar array string isolation diodes, PPT priate processor, memory, and interface cards.
conversion efficiency, power subsystem wir- The Solar Probe+ IEM contains five 6U
ing, and spacecraft wiring harness. cPCI cards: a RAD750 CPU, a spacecraft in-
The power subsystem was designed to pro- terface card, a G&C interface card, a SSR
vide 482 W of load power between 0.044 AU card, and a DC/DC converter card. The IEMs
and 0.9 AU distance from the Sun. The Sun– are block-redundant.
probe distance is greater than 0.9 AU only The avionics subsystem also collects analog
immediately after launch and around aphelion and digital telemetry via RIUs, which are
of the first few orbits. During these times us- based on the RIO application-specific inte-
ing X-band rather than Ka-band will reduce grated circuits (ASICs) flown on several pre-
power used by the telecommunications sub- vious missions. These small, lightweight units
system. The amount of data transferred to the collect and digitize telemetry points and
ground also will be reduced. However, be- transmit the data to the IEM using the industry
cause the spacecraft spends relatively little standard I2C bus.
time at these greater distances, there will be no The PDU switches loads and controls
impact on science return. thrusters via command from either IEM. The
Between 0.9 AU and 0.25 AU the angle of PDU is internally redundant with two field-
the primary array is adjusted to deliver adequate effect transistors (FETs) in each solid-state
power to the spacecraft. Inside of 0.25 AU, the switch to ensure that every load can be turned

3-58
3.0 MISSION IMPLEMENTATION

Figure 3.9-1. Avionics system block diagram.

off. The slice-based design from RBSP con- RIU interface circuitry, backup oscillator, and
tains redundant slices for power, command, clock/timing circuitry for the card.
and telemetry. Slices based on relays and FET The G&C interface card contains the inter-
switches are stacked as appropriate. FET faces to the engineering instruments, reaction
switches incorporate resetable circuit breakers wheels, and gimbal drive circuitry for the solar
based on the power remote input/output (RIO) arrays. Also included is the interface to the aft
(PRIO) ASIC. horizon sensors used to detect a fault in atti-
The RAD750 CPU card is an existing design tude control and the fault protection module
with 16-MB SRAM, 4-MB EEPROM, and (FPM) that autonomously switches IEMs in
64-KB Fuse Link boot PROM. Various configu- response to the Sun being seen by the aft Sun
rations are available, including SDRAM-based sensors or other fault conditions. Each IEM
configurations that provide significantly more contains an FPM, which, on the spacecraft in-
main memory, and can be selected as needed. terface card, is powered by unswitched power.
The 128-Gbit SSR is based on stacked The FPM in the “off” IEM monitors health
SDRAM modules. BAE has proposed such a signals from the “on” IEM. When the FPM
design for several missions. An alternative detects a fault, it follows a decision tree and
approach is to use a flash-based SSR based on can command the PDU to switch IEMs. The
that of New Horizons. FPM disables itself after use with a latching
The spacecraft interface card is the only relay in the PDU. The FPMs can be enabled or
IEM card that contains the “standard” space- disabled by critical command via either IEM
craft interfaces that are not mission-specific. (powered or unpowered) at any time.
Much of the design is heritage from previous 3.9.2. Flight Software
programs. It contains the critical command The Solar Probe+ command and data han-
decoder, which executes some commands di- dling (C&DH) software has direct heritage
rectly in hardware and passes some directly to from RBSP and significant component heri-
the PDU. It also contains the PDU, downlink, tage from previous missions, including MES-

3-59
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

SENGER, New Horizons, and STEREO. The primary flight processor is responsible for atti-
software is implemented by using the C pro- tude estimation and attitude control, while the
gramming language and makes use of the core standby instance of the G&C application is a hot
Flight Executive (cFE) middleware by NASA spare. Attitude control of the spacecraft is neces-
Goddard. The flight software system uses the sarily limited to the primary flight processor,
VxWorks™ real-time operating system. This which is enforced in the 1553 bus application
software environment facilitates hosting sev- software.
eral concurrently executing software applica- The G&C attitude estimation and control al-
tions on a single processor. The Solar Probe+ gorithms are developed by using MATLAB
flight processor executes C&DH, G&C, and Simulink™ models. MATLAB Real-Time
autonomy and fault protection flight software. Workshop (RTW) is used to generate C code
G&C attitude estimation and attitude control from the Simulink™ models that is then com-
algorithms are implemented as tasks that exe- piled into the G&C flight software application.
cute concurrently within a single G&C appli- Several previous missions have successfully
cation. The C&DH flight software is com- used this model of G&C software development.
posed of several applications that manage, The primary flight processor C&DH software
among other things, Consultative Committee manages the telecommunications uplink and
for Space Data Systems (CCSDS) telecom- downlink using CCSDS protocols for data han-
mand protocol uplink, downlink, command dling. Commands are received in CCSDS tele-
execution, SSR, parameters, peak power track- command packets and, according to an opera-
ing, and the 1553 bus. tion code contained in the packet header, are
Among the two identical flight processors, the either processed by the primary flight processor
selection of the primary flight processor is made software or dispatched to the other subsystems
based on a discrete signal set by command via on the 1553 bus, including the secondary flight
the critical command decoder hardware. A sec- processor. The C&DH software supports storage
ond discrete signal selects which one of the two of command sequences, or macros, which can
software images to boot for the two flight proc- be executed by a ground command, an auton-
essors. The primary flight processor acts as the omy event, or a time-tagged command stored in
1553 bus controller and actively controls the the flight processor’s memory.
spacecraft, performing all G&C, C&DH, and Instrument housekeeping and science data
autonomy and fault protection software func- are routed through a common data processing
tions. The secondary flight processor, when unit (DPU) remote terminal on the 1553 bus to
powered, boots to a flight software configuration the flight processor. The DPU compresses and
that operates as a remote terminal on the 1553 packetizes these data prior to their being sent
bus and will record science data to the standby to the flight processor where the C&DH soft-
SSR in parallel with the primary flight processor ware manages the storage of the data packets
on the primary SSR. In critical phases of the or- on the SSR in the form of files.
bit, the secondary flight processor acts as a hot The C&DH software is configured to inter-
spare in standby mode in the event of a fault in leave CCSDS transfer frames of real-time te-
the primary flight processor. As a hot spare, the lemetry packets with frames of SSR playback
standby flight processor executes the same G&C data based on a commandable ratio. SSR play-
application software as the primary flight proc- back is managed using the CCSDS File Delivery
essor. Attitude information is received by both Protocol (CFDP) software that was successfully
processors via the 1553 bus from the three star used on MESSENGER. CFDP provides a
trackers and the inertial measurement unit mechanism to downlink files from the SSR by
(IMU). The G&C application executing on the using a “handshake” with the CFDP client in the

3-60
3.0 MISSION IMPLEMENTATION

ground system software. This protocol auto- toring of the various software subsystems and
matically manages retransmission of any file may initiate a failover to the secondary flight
fragments lost because of data dropouts without processor in the event of a critical software
requiring retransmission of the entire file. In ad- anomaly. The system also makes use of a
dition, file transmissions may be easily sus- hardware watchdog timer that triggers a reset
pended and resumed between contacts. The of the processor and failover to the secondary
CFDP protocol helps automate contacts by flight processor should the software become
ground stations, which increases reliability and unable to service the watchdog timer within a
reduces operating costs. programmed timeout.
In addition to the C&DH and G&C flight 3.10. Telecommunications
software applications, the flight processor also
The design of the Solar Probe+ telecommuni-
hosts an autonomy and fault protection applica-
cations system is driven by several primary re-
tion. Data collected from all subsystems are
quirements. First, the spacecraft remains near
stored in a memory buffer and can be referenced
the ecliptic plane throughout the mission, which
by uploaded autonomy rules to detect and re-
dictates antenna-pointing requirements and es-
spond to faults. Each rule can monitor one or
tablishes the Sun–Earth–probe (SEP) geometry
more telemetry points, perform computations, throughout the orbit. Second, the desired science
and execute a specified command if the premise data volume (128 Gbits) and the limited data-
of the rule evaluates “true” for a designated return time between subsequent perihelia
number of consecutive evaluations. Typically, (spaced ~88 days apart) set a high downlink data
the command is an instruction to execute a rate. Finally, the high-gain antenna (HGA)
stored macro that performs a corrective or main- must remain within thermal requirements during
tenance action. This design facilitates the ability use to maintain pointing and must be protected
to develop, test, and upload autonomy rules from temperatures exceeding survival limits at
without requiring software changes. Not only all times during the mission. These requirements
does this reduce the development cost of the ultimately establish the subsystem design and
flight autonomy system, but it also provides the data management strategy to return the required
flexibility to easily change the system behavior science data. Figures 3.1-3 and 3.1-6, previously
at any point during the mission, which is of par- presented in the mission design overview, give
ticular interest in the event of a component fail- the SEP angle and Earth–probe distance for the
ure or simply to allow operators to adapt as baseline mission.
more experience is gained with the operational
3.10.1. Trade Studies
spacecraft. This system increases reliability, re-
3.10.1.1. Frequency Selection
duces risk, and reduces cost for autonomy sys-
Solar Probe+ assumes use of the DSN 34-m
tem changes. This type of autonomy and fault
subnet for routine (nonemergency) operations.
protection software system has been success-
The DSN 34-m subnet possesses both X-band
fully used on several previous missions.
and Ka-band communications capabilities. With
The C&DH software supports receipt and
Solar Probe+ remaining in the ecliptic plane,
storage of code, parameter, command, macro,
roughly half the postperihelion series of contacts
and autonomy rule uploads, as well as
will be on the far side of the Sun (Earth range
downlink of these items or flight software data greater than 1 AU), and the Sun will interfere in
structures. Additionally, the flight software downlinks when the SEP angle is small (less
maintains a number of history logs, event logs,
than 3° for X-band, less than 1° for Ka-band).
and anomaly logs that may be downlinked to
Ka-band science links provide an approximately
support anomaly investigation. The C&DH
four to five times increase in achievable data
flight software system performs health moni-
rate over X-band (including weather effects) in

3-61
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

the 34-m subnet. This advantage is critical to one axis to position the antenna outside the
meeting science return requirements for post- umbra and a second axis to point to Earth.
perihelion contacts greater than 1 AU from The body-mounted antenna is always in the
Earth, with reasonable radio frequency (RF) umbra of the sunshield when the shield is
transmit power levels and HGA diameters. The pointed at the Sun. However, the range of pos-
lower minimum 1° SEP angle limit at Ka-band sible Sun–probe–Earth (SPE) angles over
adds several days of contact time to each post- which this antenna can support a link is limited
perihelion series versus X-band operation. To by the body of the spacecraft, the distance to
maximize the data return, a Ka-band system is the umbra, and obscurations caused by the
set as the primary science return link. Neverthe- shield. A smaller HGA can point across a wider
less, the advantages of also including a parallel range, but at a cost in downlink rate capability.
X-band downlink system outweigh the costs: For example, with a 0.65-m HGA diameter, the
better emergency performance, the ability to range of possible SPE angles is limited to 38° <
utilize the 70-m subnet (which does not have a SPE < 142°. Smaller antennas than 0.65 m pro-
Ka-band capability), greater insensitivity to vide greater coverage in the antisunshield di-
weather, greater technological maturity and rection but are limited by the shield in the sun-
flight heritage, and a more forgiving pointing shield direction, and they fall off in data rate
requirement for the same aperture size. The So- capability inversely proportional to the square
lar Probe+ telecommunications system uses both of the diameter.
Ka-band and X-band downlinks. The uplink is The mast-mounted HGA extends the an-
at X-band, given the lower data rate require- tenna out of the umbra and, along with con-
ments needed for commanding. trolling the roll angle of the spacecraft, per-
3.10.1.2. High-Gain Antenna Design mits pointing to Earth across all SPE angles (0
Several designs for the HGA system were to 180°). However, the HGA can only be de-
considered. The maximum size of the antenna ployed outside the umbra when the space-
is limited by the size of the umbra, mounting craft’s distance to the Sun is greater than 0.59
locations along the spacecraft structure, and AU, because of thermal limitations.
the pointing mechanism. A Ka-band phased Table 3.10-1 charts the idealized data return
array antenna was considered and but not se- from each of the HGA configurations for pri-
lected because of its complexity in the number mary perihelia with closest approach at 9.5 RS.
of elements required, its mass inefficiency, This comparison uses the same assumptions for
and performance losses over the range of re- each HGA design under consideration, for the
quired pointing angles. Two leading HGA purposes of determining the best HGA configu-
candidates were considered at length: a body- ration, but does not represent the actual data
mounted, shaped parabolic antenna with a sin- management for these perihelia. In both cases,
gle axis of rotation (the spacecraft roll would a minimum SEP angle of 1° was assumed for
provide the additional axis), and a mast- Ka-band operation, as was one 8-hour effective
mounted, circular parabolic antenna requiring pass per day with the DSN 34-m subnet. After
two axes of rotation along with spacecraft roll: the first and second principal perihelia, the
Table 3.10-1. Idealized post-perihelion data volume returned for principal perihelia.
Data Return Mast Body-Mounted
Perihelion 1 45 Gbits 0 Gbits
(worst-case SPE profile) (1° SEP sets return limit to 40 Gbits) (SPE always <15°)
Perihelion 2 109 Gbits 111 Gbits
(SPE < 90°, far side) (0.59 AU limits otherwise 110-Gbit return) (0.65-m dish diameter)
Perihelion 3 >1000 Gbits 510 Gbits
(SPE >90°, near-Earth return)

3-62
3.0 MISSION IMPLEMENTATION

spacecraft remains on the far side of the Sun band uplink and downlink capability is pro-
from the Earth. For the first perihelion, the SPE vided through all antennas. The HGA de-
is always less than 15°, and no high-speed ployment system remains stowed within the
downlink is possible from the body-mounted shadow of the umbra while the spacecraft is
antenna. For the second perihelion, the return within 0.59 AU of the Sun. When the space-
from both antenna cases is approximately the craft is farther than 0.59 AU from the Sun, the
same. For the third principal perihelion, where HGA mast is deployed, and the antenna
the spacecraft is on the same side of the Earth pointed to Earth for science playback. The
after closest approach, both antenna schemes main reflector is 0.8 m in diameter and uses a
provide downlink return in excess of 500 Gbits. dichroic subreflector to transmit a right-hand
In order to retrieve science results from all circularly polarized (RHCP) wave at Ka-band.
perihelia, the mast-mounted 0.8-m antenna is A horn behind the subreflector provides bidi-
selected. This case has a higher mass and rectional communications at X-band through
greater number of components but maximizes the HGA. The 0.8-m aperture at Ka-band does
science return. For orbits during which the set the required pointing error for the HGA
spacecraft is on the far side of the Sun after boresight axis to Earth at 0.2°.
perihelion, the number of passes in the inter- When the spacecraft is within 0.59 AU of
perihelion period must be increased beyond the Sun, communications are maintained
the one 8-hour pass per day assumption used through the X-band LGAs hard-mounted to
here in order to meet the 128-Gbit data-return the spacecraft structure, separated by 180° and
requirement. canted forward and aft, respectively. The
3.10.2. Subsystem Implementation LGAs provide some directivity, allowing for
The Solar Probe+ Telecommunications telemetry reception within the 0.59-AU limit
Subsystem architecture is shown in Figure and for emergency rate command and teleme-
3.10-1. The subsystem uses three antennas: a try communications to 1.7 AU.
HGA mounted to a dual-axis gimbaled mast The Ka-band (32-GHz) high-power trans-
and two hard-mounted low-gain antennas mitters are 40-W RF output power traveling
(LGAs). The HGA is the prime antenna for wave tube amplifiers (TWTAs). The TWTAs
the Ka-band science return downlink, and X- are ~50% efficient (DC power of 80 W) and
build on heritage from
TWTAs on the Kepler (35 W
at 32 GHz) and LRO (40 W at
26 GHz) missions. The X-
band transmitters are 13-W
TWTAs that are heritage from
the New Horizons mission
and are ~40% efficient (32 W
of primary DC power).
The two RF transponders
are based on the advanced
digital receiver flown on the
New Horizons mission and on
digital and Ka-band hardware
developed for NASA on the
CoNNeCT program. The
transponders each require only
Figure 3.10-1. Solar Probe+ telecommunications system.

3-63
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

Table 3.10-2. Maximum telemetry rates vs. Earth summarizes the estimated telemetry perform-
range. ance for the different antennas at varying dis-
Downlink Rates vs. Earth Range
Estimated Maximum
tances. Table 3.10-3 summarizes maximum
Telemetry Rate command rates. No significant solar interfer-
0.8-m 0.8-m ence and the use of a 34-m deep space mission
Earth– HGA, X- HGA, Ka- LGA, X- system (DSMS) antenna for the ground link are
Spacecraft band to 34 band to 34 band to 70
Distance m, 13- W m, 40-W m, 13-W assumed. In the case of the low-gain antenna
(AU) TWTA TWTA TWTA (LGA), a 70-m DSN antenna is assumed along
0.5 92 kb/s 932 kb/s 167 bps with a worst-case LGA orientation (random
1 23 kb/s 233 kb/s 42 bps
1.5 10 kb/s 104 kb/s 10 bps
tumble) to cover an emergency situation.
1.8 7 kb/s 72 kb/s 6 bps 3.11. Data Management
Table 3.10-3. Maximum command rates vs. Earth Even though the baseline instrument suite is
range. fundamentally the same, the Solar Probe+ data
Uplink Rates vs. Earth Range management concept is different from the
Estimated Maximum 2005 Solar Probe concept in two important
Command Rate
Earth–Spacecraft
ways. First, the nature of the Solar Probe+
34-m to 0.8-m 70-m to LGA,
Distance (AU) HGA, X-band X-band mission design allows for repeated solar en-
0.5 >10 kb/s 389 bps counters at fairly regular intervals, while the
1 >10 kb/s 97 bps 2005 concept is based on two solar encounters
1.5 >10 kb/s 43 bps spaced widely apart. Second, the 2005 concept
1.8 >10 kb/s 30 bps included real-time downlink of critical science
data for risk mitigation. The Solar Probe+
4 W of primary power in receive-only mode,
concept reduces risk through critical perihelia
8.7 W in receive/X-band transmit mode, 9.7 W
farther away from the Sun and through the use
in receive/Ka-band transmit mode, or 14.1 W
the previously mentioned repeated encounters
in receive/X- and Ka-band transmit mode.
to allow extensive preparation for critical en-
Other higher-power transponder options are
counters and recovery from unanticipated
available for consideration. A Phase A trade
problems, and no real-time science data
study will be conducted to determine the op-
downlink is used.
timal transponder choice by balancing factors
From an operational perspective, Solar
such as solar array size through the use of
Probe+ orbits are broken into aphelion seg-
lower power systems, development status of
ments and perihelion segments. Although
the options, and overall system cost.
some science data are taken throughout an or-
The output from either of the X-band
bit, the great majority of science data are taken
TWTAs may be steered to any of the antennas
during solar encounters around perihelion.
through a network of single-pole-double-throw
During this time, the payload stores data onto
(SPDT) and transfer (XFER) switches, which
the solid-state recorders (SSRs) through the
are themselves configured for redundant opera-
payload data processing unit (DPU) and inte-
tion. Similarly, the Ka-band TWTAs are
grated electronic module (IEM). The aphelion
switched to the HGA. Hybrid couplers are used
segment is defined by the time in which the
with each of the X-band and Ka-band TWTA
HGA can be used, nominally outside 0.59 AU,
pairs to increase downlink system reliability.
and is primarily when all data stored on the
3.10.3. Performance SSR during the previous encounter and during
Link performance is determined primarily by cruise since the last aphelion segment is
the antenna used and by the relative distance downlinked.
from the spacecraft to the Earth. Table 3.10-2

3-64
3.0 MISSION IMPLEMENTATION

Table 3.11-1. Instrument and housekeeping average data rates.


Instrument Raw Data Rate (bps) Data Rate to Recorder (bps)
Fast Ion Analyzer (FIA) 10,000 13,650
Fast Electron Analyzer (FEA) 20,000 27,300
Ion Composition Analyzer (ICA) 10,000 13,650
Energetic Particle Instrument (EPI) Low Energy 5000 6825
EPI High Energy 3000 4095
Neutron/Gamma-Ray Spectrometer (NGS) 500 683
Coronal Dust Detector (CD) 100 137
Magnetometer (MAG) 1100 1502
Plasma-Wave Instrument (PWI) 10,000 13,650
White-Light Hemispheric Imager (HI) 40,800 55,692
Housekeeping 1800 2457
Total 10,230 139,640

3.11.1. Science Data Collection lection outside encounter), the nominal 128-
The Solar Probe+ payload, as discussed in Gbit SSRs on Solar Probe+ will be simultane-
Section 2.0, is essentially the baseline payload ously filled. Two recorders are used to provide
from the 2005 Solar Probe study. Science data redundancy.
collection will be conducted in much the same 3.11.2. Data Return
manner as in the previous study, and instrument For each Solar Probe+ orbit, the period ex-
data rates are currently baselined to be very tending from 0.59 AU solar distance as the
similar to those presented in the previous study. spacecraft leaves the Sun through aphelion
Table 3.11-1 gives the average instrument/ back to 0.59 AU as the spacecraft approaches
housekeeping data rates, where the raw data the Sun is designated primarily for data
rates represent the actual science data and the downlink. As shown previously in Figure 3.1-
data rate to the recorder includes 30% margin 6, the Earth–probe distance varies from 0.3 AU
and a 5% overhead for packetization performed to 1.9 AU, depending on the specific orbit.
in the data processing unit (DPU). Actual data Based on 10-hour contacts and average
rates from the instruments will vary during the downlink rates for each aphelion downlink pe-
encounter, and detailed data volume allocations riod, we have developed a day-by-day
for each instrument in each orbit will be estab- downlink schedule that allows for downlinking
lished in Phase A. The resulting required data the full SSR data volume within the aphelion
rate of 139.6 kbps will be achieved by using a segment as well as supporting other operational
standard 1553 data bus to transfer data packets aspects of the mission (such as navigational
from the payload and housekeeping data from requirements to support Venus flybys). Table
the various subsystems. Over the encounter pe- 3.11-2 gives a summary of this contact plan
riod (and including any low-rate science col- and total data volume downlinked in each
Table 3.11-2. Orbit data return summary.
Orbit 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Total
Days in
168 149 139 119 112 108 101 99 99 96 95 95 96 95 95 94 92 91 91 87 87 87 87 87 2554
Orbit
Venus
Y Y Y Y Y Y Y 7
Flyby
Downlink
81 42 42 81 6 42 59 6 6 56 54 6 6 54 42 6 42 37 42 6 42 29 42 6 844
Days
Ave. Data
Rate 44 86 86 44 583 86 44 583 583 44 44 583 583 44 86 583 86 44 86 583 86 44 86 583 92
(kbps)
Data
Return 128 128 128 128 128 128 94 128 128 90 86 128 128 86 128 128 128 59 128 128 128 46 128 128 2765
(Gbits)

3-65
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

Table 3.12-1. Overall system pointing budgets.


Pointing Requirements (per axis)
Control Knowledge Jitter
Payload degrees, 3σ degrees, 3σ degrees, 3σ
Communication (HGA) ²0.2 x, z N/A N/A
In Situ Instruments <0.3 AU ²1 x, y, z ²0.3 x, y, z ²0.3 x, y, z over 10 s
Magnetometer <0.3 AU ²1 x, y, z ²1 x, y, z ²1 x, y, z over 0.05 s
Plasma Wave Sensor <0.3 AU ²1 x, y, z ²1 x, y, z N/A
Hemispheric Imager <20 RS ²1.0 y, z ²1.0 y, z ²0.03 y, z over 1 s

aphelion segment. Solar Probe+ is able to 3.12. Guidance and Control System
downlink the full dataset from each perihelion The Solar Probe+ guidance and control
except for six orbits where the SEP geometry is (G&C) subsystem is designed to maintain the
least favorable. This calculation uses the aver- spacecraft attitude required to protect the
age data rate for each orbit and assumes only a spacecraft bus from the harsh solar environ-
single 10-hour DSN contact for each day when ment, point antennas for communications with
contacts are planned. Therefore, the estimated Earth, provide desired viewing geometry for
data volume return is worst case. More detailed science instruments, and point thrusters for
modeling of the daily data rate in Phase A is trajectory correction maneuvers (TCMs).
expected to show an increase in data volume Three star trackers and a high-precision, inter-
returned for the worst-case orbits, and a trade nally redundant inertial measurement unit
study to be conducted in Phase A will optimize (IMU) provide attitude knowledge, and atti-
the return for these orbits by balancing science tude control is provided by four reaction
data cadence, DSN cost and schedule require- wheels and 12 0.9-N thrusters. The attitude
ments, high-gain antenna (HGA) thermal de- determination and accuracy requirements de-
sign to increase the window for HGA deploy- rived from these different activities are sum-
ment, telecommunications system topology, marized in Table 3.12-1. Pointing control is
and radio frequency (RF) transmit power. driven by the need to point the HGA within

Figure 3.12-1. G&C functional block diagram.

3-66
3.0 MISSION IMPLEMENTATION

0.2° when downlinking using the communica- error that reaches a designated threshold
tion system. The pointing knowledge and jitter should occur, the edge of the conical ring
budget is driven by the remote sensing instru- would become illuminated and projected onto
ment. A functional block diagram of the sys- the detector. The processed signal could be
tem is shown in Figure 3.12-1. used to provide attitude control for safing dur-
3.12.1. Attitude Determination ing the solar encounter.
Spacecraft attitude will be determined by Most currently available attitude control
three star trackers and an internally redundant hardware should meet the needs of Solar Probe+
inertial measurement unit (IMU). Using star with little or no custom modifications. Special
trackers in the near-Sun corona presents a care must be taken to select star trackers that
unique design challenge, which the baseline will perform well in the intense coronal lighting
design addresses by mounting the star trackers environment. During the engineering study, one
so that their field of view (FOV) are approxi- or more potentially suitable candidate star track-
mately orthogonal to the Sun as well as to each ers were identified in existing product lines. The
other. This configuration minimizes the chance SHS will be the only attitude determination de-
that all three units will be blinded by a local- vice that will need to be developed for Solar
ized coronal lighting event at the same time. Probe, and it is conceptually simple.
Special care must be taken in the selection of 3.12.2. Attitude Control
the star trackers to ensure that they will per- Although the overall pointing requirement
form properly with the elevated background for the spacecraft is 0.2°, the G&C system was
noise of the near-Sun environment. preliminarily budgeted to ~0.05° because
The IMU will provide the spacecraft rate much of the error budget will go to HGA mis-
and translational acceleration information nec- alignments and actuator setting errors. Conse-
essary for maintaining attitude control as well quently, reaction wheel control, which offers
as for closed-loop control during trajectory very tight pointing control and can easily
correction maneuver (TCMs). The IMU also maintain spacecraft attitude at better than the
can be used as a backup to the star trackers to budgeted 0.05°, has been baselined. Wheel
propagate attitude for a brief period during a control also interacts less with flexible modes
solar encounter if all three star trackers are and would be more likely to control them to
temporarily blinded. The baselined IMU is a meet the jitter budget. Thrusters are used to
single integrated box with internal redun- control attitude during TCMs and for dumping
dancy, although two separate units also would accumulated angular momentum from the
meet the needs of the mission. wheels when necessary.
In the event of long-duration star tracker A brief trade study was performed to deter-
blinding, system resets, or other attitude con- mine whether reaction wheels or thrusters
trol anomalies, a new sensor design, the solar should be used as the primary method of attitude
horizon sensor (SHS), is proposed for attitude control. Dead-band thruster control using small
safing when the spacecraft needs to be pro- minimum impulse bit rocket engines such as are
tected behind the Thermal Protection System used on Cassini and New Horizons was consid-
(TPS) umbra. The detector would be mounted ered because it appeared to offer a means of re-
at the end of the science boom and would con- ducing mass and average power during the en-
sist of a conical ring of carbon–carbon (C-C) counter. Thruster control might possibly reduce
material, a mirrored conical reflector, and a overall mass slightly; however, it also has some
detector array with a pinhole lens. The detec- disadvantages. First, as mentioned earlier, the
tor array resides in a small electronics box, required dead band for the G&C system is
which contains readout electronics for both the budgeted to ~0.05°. This small dead-band value
detector and a set of thermistors. If an attitude would require frequent thruster firings, thus

3-67
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

driving up the total propellant requirement and namic pressure (CP)/center of gravity (CG) re-
negating most of the mass savings of removing quirements to reduce solar pressure effects.
the wheels. Overly frequent thruster firings also Solar dust impacts also are an important at-
were a concern because of possible instrument titude control consideration, especially near
contamination, valve lifetimes, and possible perihelion. Dust particles that impact Solar
thruster-induced structural excitation that may Probe+ will impart an instantaneous momen-
exceed the jitter requirements. Because there are tum impulse that the wheels must take out. In
several components that could induce low- the event that the momentum impulse is too
frequency modes such as the TPS, science large for the wheels to handle, the thrusters
boom, and plasma-wave antennas, dead-band will be fired to maintain the sensitive space-
thruster control appears less attractive but will craft systems safely in the umbra of the heat
be retained as a trade for Phase A. shield. During Phase A, more detailed analysis
3.12.3. Environmental Considerations of the dust environment will be carried out and
Several environmental factors are drivers thruster selection optimized, as necessary.
for the Solar Probe+ G&C design and will re- 3.12.4. Pointing Strategy
quire more detailed study in Phase A. First, as During all phases of the mission, the
the spacecraft approaches perihelion, sunlight probe’s attitude nominally changes so that the
reflected off of dust particles will be seen by a TPS points toward the Sun, keeping the in-
star tracker looking away from the Sun struments and subsystems within its protective
through the solar corona. Coronal lighting re- umbra. TCMs will be planned for aphelion
duces the signal-to-noise (S/N) ratio for a when the spacecraft TPS can be off-pointed
tracker using a charge-coupled device (CCD), from the Sun.
thereby reducing the number of detectable Momentum dumping will occur much more
stars and degrading the performance of the frequently during perihelion because of the
star tracker. This restricts the choice of star intense solar radiation pressure. Each momen-
trackers and also drives the design to operate tum management maneuver will be completed
multiple star trackers at perihelion. quickly taking typically on the order of 3–5
Solar pressure will be very high and will minutes. For these short periods, the thrusters
change rapidly during the solar encounter. Be- will fire to remove angular momentum, and
cause the center of photon pressure is ahead of the control requirements for instrument point-
the center of mass, the solar pressure torque is ing may not be maintained.
destabilizing and is an important part of the Solar pressure torques are often used as a
dynamics of the spacecraft near perihelion. means of passive momentum control. This op-
The solar pressure torque, which decreases tion was considered as a possible augmenta-
with distance r from the Sun as 1/r2, will re- tion of Solar Probe+ attitude control during
quire multiple momentum dumps on the day the closest approach periods. It would require
of perihelion passage. Relatively small mis- an intentional pointing offset of the heat shield
alignments of the heat shield could induce that is adjusted automatically by the feedback
significant torque and momentum build-up, control system. An advantage of passive
potentially forcing more frequent use of dumping using solar pressure torques is that it
thrusters for momentum management, conse- could reduce the number of thruster momen-
quently the center-of-pressure/center-of-mass tum dumps needed. However, instrument
offset will have to be carefully monitored dur- pointing requirements might limit the range of
ing the design phase. The ΔV propellant budget offsets that could be used, reducing the overall
includes an estimate of momentum dumping contribution of a passive dumping mechanism.
frequency during perihelion. A Phase A study In addition, successful employment of this
will be conducted to optimize center of aerody- method also depends greatly on accurate mod-
eling of the solar pressure effects. Thus, the

3-68
3.0 MISSION IMPLEMENTATION

sizing of the baseline propulsion sys-


tem assumes that all momentum con-
trol must be done with thrusters, given
the uncertainties in the environment
models.
During TCMs, the attitude of the
spacecraft will be off-Sun pointed to
use the larger, 4.4-N thrusters to impart
the desired change in velocity. These
maneuvers will occur near aphelion
where it is permissible to have the
spacecraft outside of the TPS umbra for
the duration of the TCM. Upon comple-
tion of the maneuver, the TPS will
again be pointed sunward.
3.12.5. High-Gain Antenna Control
The HGA will be pointed by rotating
the spacecraft about the spacecraft
sunline and rotating the antenna using
the rotary actuator to keep it oriented
toward Earth. The second axis of motion
for the HGA is used to deploy the HGA Figure 3.13-1. Propulsion block diagram.
mast to a fixed position to give the HGA
a clear field of view to Earth. The G&C subsys- ness, a third star tracker was added to the de-
tem will compute the necessary positioning of sign to maximize the probability of having at
the gimbal for the HGA based on onboard least one star tracker operational at all times.
ephemeris models for the Earth, Sun, and space- The internally redundant IMU, the SHS, and
craft. In the event of loss of onboard ephemeris the reaction wheels are all carried over from the
knowledge or other fault conditions, the HGA previous design.
will be commanded to its safe stowed position. 3.13. Propulsion System
3.12.6. Guidance and Control Changes The Solar Probe+ propulsion subsystem is a
Since Previous Study blowdown monopropellant hydrazine system
There have been only minimal changes to that provides ΔV and attitude control capability
the G&C design since the previous Solar Probe for the spacecraft. The system consists of 12
study. 23 The digital solar attitude detectors of 0.9-N (0.2-lbf) thrusters, two 4.4-N (1.0-lbf)
the previous study, which were to be used as thrusters, and components required to control
safing sensors for the periods of the mission the flow of propellant and monitor system
when the spacecraft was outside of 0.8 AU, health and performance. The propellant and
have been deleted from the design because the pressurant are stored in the same tank, sepa-
majority of the Solar Probe+ mission will be rated by a diaphragm. As propellant is ex-
carried out within that distance. To replace pelled, the pressure of the pressurant decreases;
those sensors and further add system robust- thus, the thrust and specific impulse of the
thrusters decrease as the mission progresses.
23
Solar Probe: Report of the Science and Technology All valves will maintain temperatures above
Definition Team, NASA/TM—2005–212786, National 5°C to protect the soft seals. The propulsion
Aeronautics and Space Administration, Goddard Space
Flight Center, Greenbelt, MD (2005). system schematic is shown in Figure 3.13-1.

3-69
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

Figure 3.13-2. Baseline 0.2-lbf thruster.

Figure 3.13-4. Blowdown curve.

drazine type; when the thruster valve is


opened, propellant flows through the thruster
into a catalyst bed, where the hydrazine spon-
taneously decomposes into hot gases, which
then expand through a nozzle and exit the
thruster, producing thrust. The baseline atti-
Figure 3.13-3. Baseline ΔV thruster.
tude control system thrusters are all Aerojet
model MR-103G (Figure 3.13-2), which is a
The baseline propellant load for Solar member of the MR-103 family of thrusters,
Probe+ is 49.7 kg of hydrazine. For a 554-kg used on Voyager, Magellan, New Horizons,
wet mass launch, this translates to 190 m/s of Cassini, and several other missions. The base-
ΔV. Several flight-proven options exist for line ΔV thrusters are both Aerojet model MR-
each component of propulsion subsystem. A 111C (Figure 3.13-3). This thruster has heri-
representative set of heritage components has tage on the New Horizons, STEREO, and
been identified for preliminary performance MESSENGER spacecraft. The actual steady-
evaluation and demonstrates that system re- state thrust produced from both thrusters varies
quirements can be achieved. as the tank pressure decreases. Figure 3.13-4
The propellant tank is a 5555-in.3 (91.0-liter) illustrates thruster steady-state performance
titanium tank manufactured by ATK-PSI. This between beginning of life and end of life.
22.14-in. diameter, vacuum-rated spherical The remaining components used to monitor
tank (ATK P/N 80259) contains an elastomeric and control the flow of propellant—latch
diaphragm that pushes propellant out the bot- valve, filter, orifice, and pressure and tempera-
tom of the tank through the tank outlet. The ture transducers—have substantial heritage on
maximum expected operating pressure for the spacecraft, including New Horizons, STEREO,
Solar Probe+ mission is 300 psi. The tank has and MESSENGER.
flight heritage on the Defense Satellite Com- Latching valves isolate the thrusters from
munications System (DSCS) III spacecraft. the tank for safety and system reliability (i.e.,
The thrusters on the Solar Probe+ space- in case of a thruster leak). The Vacco ¼-in.
craft are of the catalytic monopropellant hy- latch valve (Figure 3.13-5), part number

3-70
3.0 MISSION IMPLEMENTATION

sis only crudely accounts for the material prop-


erties of the heat shield, which was expected to
be constructed from an uncoated carbon–carbon
(C-C) composite rather than the ceramic-coated
version now under study. Electron emission
characteristics were accounted for with a rough
analytical estimate only, and geometrical effects
were ignored. Furthermore, there was no esti-
mate of the differential charging of the space-
craft; only charging relative to the plasma
Figure 3.13-5. Latch valve.
“ground” was considered. Finally, only the clos-
est approach case was evaluated, which may not
V1E10747-01, will be used to isolate the be the region of greatest concern for a ceramic-
thrusters from the tank. These latch valves are coated heat shield.
proof-tested to 1000 psig with a burst pressure The analysis described here is a continua-
of 2400 psig. This item is fully flight-qualified tion of the work reported previously, adapted
and has heritage on both STEREO and New for the Solar Probe+ mission design. Our
Horizons as well as flight heritage on numerous analysis with NASCAP-2K provides estimates
other missions. of differential charging, taking the spacecraft
Filters ensure propellant purity. The Vacco geometry into account. We have considered a
propellant filter, part number F1D10767-01, is range of trajectory points, including closest
a 10-μm filter composed of stacked etched tita- approach and the range from 0.5 AU down to
nium discs. This filter has flight history on 0.1 AU. Furthermore, our modeling efforts
STEREO and New Horizons. Manual service make use of the material properties for the
valves are used for testing and loading the sys- coatings of interest. These properties include
tem on the ground. The Vacco V1E10701-01 elevated temperature-resistivity estimates and
¼-in. fill and drain valve will be used to load room temperature secondary electron emission
the hydrazine and pressurant on to Solar and backscattered electron emission measure-
Probe+. This service valve design has a two- ments. It should be noted that the NASCAP
seat seal and a maximum expected operating results depend on the accuracy to which these
pressure of 500 psig. It is proof-tested to 1750 material properties are known. In particular,
psi and burst-tested to 2000 psi. This valve has because resistivity is the material property that
flight heritage on STEREO and New Horizons. most drives the charging behavior, more pre-
3.14. Environmental Mitigation
cise temperature-dependent resistivity meas-
urements are desirable.
3.14.1. Charging
3.14.1.2. Charging Results
3.14.1.1. Charging Analysis Methodology
Studies of the Solar Probe charging problem The baseline coatings for this effort are
date to the 1980 Starprobe report.24 That analy- alumina (Al2O3) and pyrolytic boron nitride
(PBN). We have considered the charging
problem at the following trajectory points: 0.5,
24
Goldstein et al., Spacecraft Mass Loss and Electric 0.4, 0.3, 0.2, 0.1, and 0.0443 AU (closest ap-
Potential Requirements for the Starprobe Mission, A proach). This analysis was conducted for a
Report of the Starprobe Mass Loss Requirements Group solar absorptivity-to-infrared (IR) emissivity
Meeting of September 29–30, 1980, NASA Jet Propul-
sion Laboratory, California Institute of Technology,
ratio (α/ε) of 0.6; previous efforts also in-
Pasadena, CA (December 1980). cluded the lower α/ε of 0.2. The temperature
of the heat shield will be lower for smaller

3-71
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

Table 3.14-1. Differential potentials for Al2O3. Table 3.14-2. Differential potentials for PBN.
Trajectory Point (AU) Differential Potential (V) Trajectory Point (AU) Differential Potential (V)
0.5 7 0.5 7
0.4 7.5 0.4 7
0.3 7 0.3 6.5
0.2 7 0.2 7
0.1 7.5 0.1 4.5
0.0443 3.8 0.0443 3.4

Figure 3.14-2. Differential charging for PBN at 0.1


Figure 3.14-1. Differential charging for Al2O3 at 0.1
AU.
AU.

values of α/ε, resulting in a higher resistivity 3.14.1.3. Impact of Spacecraft Charging


and a change in the differential potential, so Typically, two major risk areas are studied
the results presented here are not worst case. in spacecraft charging efforts. If large differ-
The differential charging results derived ential potentials should build up between dif-
from NASCAP-2K are given in Tables 3.14-1 ferent sections of the spacecraft, there is the
and 3.14-2. Both materials experience differen- possibility of arcing, which could damage the
tial charging at low levels, below 10 V, for all spacecraft, including its electronics, commu-
of the trajectory points considered. The abso- nications devices, or instruments. Historical
lute surface charging relative to plasma ground trends indicate that surface charging potentials
also is below 10 V for all of these points. The should be limited to the low hundreds of volts
NASCAP-2K plot of the spacecraft potentials in order to protect the spacecraft electronics.
for Al2O3 and PBN for at the 0.1 AU trajectory Even for low differential potentials, however,
point for α/ε = 0.6 are shown in Figures 3.14-1 there is a risk of disrupting science data col-
and 3.14-2. The shaded portion of the space- lection. The buildup of a significant potential
craft tends to charge to a few volts negative, on the spacecraft relative to plasma “ground”
whereas the heat shield coating charges slightly may cause measurement contamination by dis-
positive, and the solar cells charge to a few rupting instrument function and by disturbing
volts positive. the local environment. The precise level of

3-72
3.0 MISSION IMPLEMENTATION

charging that is tolerable from the perspective the plasma “ground.” The differential charging
of science data collection is still under review situation is that the sunlit side of the solar arrays
but would best be kept to the low tens of volts. charge slightly positive, as does the heat shield,
3.14.1.4. Mitigation Strategies whereas the shaded portion of the spacecraft
The Solar Probe Thermal Protection System charges to a few volts negative. At these poten-
tial levels, arcing is not a concern, but mitigation
Risk Mitigation Study: FY 2006 Final Report
strategies may be considered in order to prevent
ITAR-Restricted Annex 25 contains a summary
contamination of the data collected by the in-
of mitigation options that were studied for the
struments. Possible mitigation approaches in-
previous Solar Probe+ design. These options
clude, for the solar array, the use of conductive
presume that the ceramic-coated heat shield is solar cells and, for the heat shield, doping ce-
the primary cause of surface charging, and they ramic coating or exposing portions of the shield
have not yet been studied in detail for Solar to be bare C-C.
Probe+. However, the charging results above The Solar Probe+ spacecraft will include
indicate the solar cells charge more positive dust and micrometeoroid protection for the
than the heat shield. In order to mitigate their expected particle environment. The near-Sun
charging, it may be necessary to ensure that the dust environment and its impact on the Solar
solar arrays are conductive on the sunlit side, Probe spacecraft were a major focus of the
for example, by the use of transparent conduc- 2005 Solar Probe STDT study. 26 Compared
tive oxides. If it also is necessary to mitigate with that effort, the Solar Probe+ spacecraft
charging by the ceramic coating, the principles will be subjected to a larger particle flux but at
involved in the earlier study should be applica- lower velocities. However, the spacecraft pro-
ble to the current heat shield design. In that tection approach remains similar to the one
case, one option to mitigate charging is to described in the earlier study. It includes dust
lower the resistivity of the heat shield coating protection for the Thermal Protection System
by adding small quantities of impurities to the (TPS), spacecraft bus, and solar arrays. The
ceramic material. Another possible solution to Solar Probe+ approach is described below, but
reduce the differential charging of the heat the planned effort includes a study both on the
shield is to expose small portions of the heat definition of the mission’s dust exposure and a
shield to be bare C-C. It should be noted that characterization of the protection approaches
the effects of both approaches—the effect of for the key areas.
dopants on ceramic optical properties and the The dust environment in the ecliptic portion
effect of C-C exposure on outgassing—must be of the trajectory was used in the STDT study to
investigated before either approach could be establish a statistical dust environmental model
implemented. as shown in Figure 3.14-3. Within 1 AU, the
3.14.1.5. Conclusions dust density is highest near the ecliptic plane
A preliminary charging analysis has been per- and falls off at higher inclinations. Parametric
formed for the Solar Probe+ spacecraft design studies of particle impacts were used to define
and for two different heat shield coating materi- the protection level provided for designs sub-
als. Initial results indicate that the spacecraft ject to different particle velocities and angles of
charges to a potential of several volts relative to attack. In this region, there will be thousands of
small particle impacts (submicrometer), but
25
Solar Probe Thermal Protection System Risk Mitiga-
26
tion Study: FY 2006 Final Report ITAR-Restricted An- Solar Probe: Report of the Science and Technology
nex, prepared by The Johns Hopkins University Ap- Definition Team, NASA/TM—2005–212786, National
plied Physics Laboratory under Contract NAS5-01072, Aeronautics and Space Administration, Goddard Space
Laurel, MD (September 17, 2007). Flight Center, Greenbelt, MD (2005).

3-73
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

experience from the


MESSENGER mission
has shown that the effect
of particle impacts on so-
lar arrays between 1.0 and
0.3 AU is minimal; the
Solar Probe+ spacecraft
will spend 88% of its
mission within this range.
Inside 0.3 AU, potential
protection options for the
solar arrays have been
identified, such as turning
the cell face inward dur-
ing storage, or oversizing
the arrays. With a 1/r
dust-particle dependency,
Figure 3.14-3. Distribution of dust impact with respect to particle size. the time spent within 0.3
AU is equivalent to a 6-
there will only be one or two particles large year orbit at 1 AU. Existing solar array design
enough to penetrate the TPS C-C shell. rules, allowing for performance degradation fac-
The Solar Probe+ study found that particle tors, can account for the damage expected dur-
protection is required for three key areas: the ing the Solar Probe+ mission.
TPS, the spacecraft bus, and the solar arrays.
3.14.2. Methodology
Compared with the earlier work, impacts on the
TPS will be significantly reduced by the change An investigation was performed during the
from a conical to flat shield. The TPS cross- 2005 Solar Probe STDT study, using an analy-
sectional area, in the ram direction, is reduced sis and prediction methodology similar to that
by a factor of almost 20. If there is a penetration used for other NASA missions (both low-Earth
of the TPS shell, the resulting damage could ex- orbit and deep space), to assess the risk of dust
tend well into the insulating foam. To protect to the spacecraft and to develop a methodology
against this secondary damage, an extra C-C for dust protection. By using conservative,
layer has been added to the TPS to shield the worst-case assumptions for particle size, veloc-
carbon foam. For the spacecraft bus, the ity, and obliquity, this study was performed by
Whipple-shield approach, where the multi-layer Dr. Cesar Carasco (at the University of Texas
insulation (MLI) is spaced away from the under- at El Paso) using state-of-the–art hydrodynamic
lying structure, will still provide the needed pro- codes [e.g., Coupled Thermodynamic 27
and Hy-
tection. The spacing between the MLI and drodynamic (CTH) hydrocode] that resolve
spacecraft will be updated based on the new par- the highly dynamic, nonlinear impact physics
ticle environment. The solar arrays are a new and include constitutive models of the materials
feature of the Solar Probe+ design; detailed dust of construction. Carasco’s findings predicted
environments will be generated for them in that Whipple shielding consisting of MLI (i.e.,
Phase A. However, there are several mitigating 18 layers of Kapton) at a stand-off distance of
options that indicate adequate protection for 27
them is available. Generally, the solar arrays are Boslough, M. B., et al., Hypervelocity testing of ad-
vanced shielding concepts for spacecraft against impacts
aligned parallel to the ram direction, reducing to 10 km/s, Int. J. Impact. Eng. 14, 96–106 (1993).
their exposure to the dust environment. Flight

3-74
3.0 MISSION IMPLEMENTATION

Figure 3.14-4. Large-particle, high-speed dust impact on Whipple shield.

10 mm would be sufficient to break up the and a flat disk-shaped secondary heat shield
largest particles and prevent penetration of the (consisting of a thin-walled C-C shell packed
metallic spacecraft bus or instrument housings with carbon foam insulation). The previous
as illustrated in Figure 3.14-4. Because of simi- secondary heat shield is similar to the heat
larities in both bus and instrument configura- shield proposed in the current Solar Probe+
tions and statistical dust environment models, it study with the exception that the new design
is believed that this approach also would be incorporates a ceramic coating (like the pri-
effective for the Solar Probe+ design. However, mary heat shield) for temperature control. In
as the Solar Probe+ design and mission pa- follow-on work performed by Dr. Carasco at
rameters mature, shielding studies need to be the University of Texas at El Paso, estimations
repeated with new trajectories (and correspond- of the structural response of the conical heat
ing dust environments) to predict whether MLI- shield to a matrix of dust-particle size, impact
based Whipple-shielding approaches remain speed, and impact angle of incidence were cal-
appropriate. culated by using high-performance hydrody-
During the 2005 Solar Probe STDT study namic codes. These predictions indicated that
and subsequent Solar Probe Risk Mitigation localized spallation footprints were greater in
Studies in 2006 and 2007, 28,29 the TPS was size than that of the impact particle. For a given
analyzed to determine the thermal-structural particle size and impact speed, the spallation
performance after dust impact and to predict footprint varied with angle of incidence; the
the resulting effects on mission performance largest footprints were developed for normal
with respect to mass loss and contamination of angle of incidence impacts as shown in Figure
sensitive instruments. The TPS evaluated in 3.14-5.
these studies consisted of a thin-walled, coni- Although visually appearing large in the
cal, ceramic-coated C-C primary heat shield highly magnified Figure 3.14-5, the resulting
spallation footprints are extremely small
28
Solar Probe Risk Mitigation Study, prepared by The (tenths of square centimeters but several or-
Johns Hopkins University Applied Physics Laboratory, ders of magnitude greater than the diameter of
2006 Mid-Year Report. the particle). Penetration through the heat
29
Solar Probe Thermal Protection System Risk Mitigation shield includes both the ceramic coating and
Study: FY 2006 Final Report, prepared by The Johns underlying C-C substrate. Predictions from
Hopkins University Applied Physics Laboratory under
Contract NAS5-01072, Laurel, MD (November 30, 2006); thermal models of penetrated zones indicate
and ITAR-Restricted Annex (September 17, 2007). localized temperature increases of several
hundred degrees Kelvin, but the effect on pre-

3-75
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

Figure 3.14-5. Predicted spallation footprints for particle impact of heat shield.

dicted equilibrium heat shield temperatures number of impacts and size of the particles
would be negligible. Predictions also indicated also will increase because of the increased
that large dust particles would be mostly con- number of orbits and the increased fluence.
sumed during the penetration event and that Further studies will need to be conducted to
the remaining dust fragments would be vapor- define the number of dust fluence and the sta-
ized within the large volume of the hot heat tistical size distribution of the particles pre-
shield—no solid dust forms would impact the dicted to impact the heat shield during the mul-
secondary heat shield. The small, submicro- tiple ecliptic orbits and to quantify the effects
meter dust particles were predicted to be va- of such impacts on thermal performance. Addi-
porized on impact and to not penetrate the tional studies will be necessary to characterize
conical heat shield. the degree of insulation damage by large-
By using analysis by similarity for the case particle impacts and the resulting increase in
of the Solar Probe+ design, it is shown that heat conduction through the heat shield. In ad-
dust will impact the flat surface of the coated dition, the effect of spallation on overall space-
heat shield directly. However, in this case, the craft mass loss rates and the interaction of spal-
exposed heat shield will not be hollow but will lation plumes with the local dust environment
be filled with foam essential to maintaining will need to be studied further.
control of the equilibrium temperature of the 3.15. Technical Challenges
bus. As before, small dust-particle impacts are Formal risk management begins in Phase A
not anticipated to be damaging; however, as- with risk identification. However, the Solar
sessment of large-particle impact damage will Probe+ mission involves significant new de-
be more complex. In this case, the heat shield velopment. We have included in this study a
will most likely experience similar spallation preliminary assessment of program risks and
footprints; however, unconsumed dust parti- identification of mitigations in order to de-
cles will penetrate into the insulating foam velop a mission implementation that balances
structure, which will most likely not affect risk with the ability to achieve performance
overall heat shield equilibrium temperatures requirements at low cost and on schedule.
but may induce localized hot spots with poten- Secondary Solar Arrays. The secondary
tial line of sight to the spacecraft bus. The solar arrays are used to generate power during

3-76
3.0 MISSION IMPLEMENTATION

solar encounters when the primary arrays can- yond the lift capability of the baselined launch
not be used. The technology for this system vehicle and third stage.
exists at technology readiness levels (TRLs) Mitigations include
5–9 (depending on the element), including the • Optimize mechanical structure mass by us-
temperature control system for the secondary ing composite elements instead of the base-
arrays. However, Solar Probe+ presents lined all-aluminum structure (Phase A)
unique environmental challenges, and thus the • Optimize power usage on the spacecraft
detailed design must accommodate this envi- and the power subsystem design to shrink
ronment. In addition, the secondary solar array both primary and secondary solar arrays
design includes solar cells and mechanisms (Phase A)
that must be qualified for the Solar Probe+ • Optimize launch vehicle tailoring and mar-
program. If the design or qualification of the gins to increase lift mass available to space-
secondary solar arrays does not adequately craft (Phase A)
address the near-Sun environment, the ability • Retain compatibility with Delta IVH launch
to generate sufficient power near perihelion vehicle with higher lift mass for this orbit
may be compromised. (Phase A/B)
Mitigations include Thermal Protection System (TPS) De-
• Increased power margin for the secondary sign and Manufacturing. The TPS is a criti-
solar arrays in the encounter operational cal element of the Solar Probe+ mission. Ex-
modes to allow for greater than anticipated tensive risk mitigation work has been per-
degradation (Phase A) formed to ensure that the TPS concept will
• Increased margin in the temperature control provide the needed protection from the solar
system in the secondary solar array subsys- environment. However, detailed design of the
tem to allow for greater than anticipated TPS has not been completed given the early
heat loads (Phase A) phase of the program. If unforeseen problems
• Detailed qualification program for mecha- occur during the design, manufacturing, or
nisms, temperature control elements, and testing of the TPS occur, the ability to protect
solar cells (pre-Phase A/Phase A) the spacecraft and payload from the near-Sun
• More detailed thermal modeling to optimize thermal environment may be compromised.
the design of the secondary solar array, in- Mitigations include
cluding the temperature control subsystem • Longer than normal development phases for
(Phase A) the Solar Probe+ program, with required
• Mission design that allows for slow walk-in schedule margin to allow for recovery
of perihelion to give opportunity for secon- should problems occur
dary solar array characterization and tailor- • Detailed thermal modeling to be completed
ing of operations concept before worst-case early in the program, with sufficient mar-
environmental exposure (incorporated into gins on thermal design maintained through
baseline concept) program (Phase A)
Mass Margin. The Solar Probe+ orbit pre- • Full TPS qualification program included in
sented in this study requires a high C3, and the cost and schedule as early in the pro-
maximum launch vehicle lift mass is con- gram as reasonably possible to allow for
strained. We have baselined the Atlas V 551 modifications that may be needed (pre-
launch vehicle with a STAR-48 third stage for Phase A/Phase A)
cost; however, the margin on mass for this • Modular design of TPS and spacecraft inter-
configuration is lower than usually required at face allows for late delivery of the shield to
this stage of a program. If significant mass spacecraft integration (incorporated into
growth occurs, the spacecraft may grow be- baseline concept)

3-77
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

Mechanism Reliability. Mechanisms in the mechanism for the Solar Probe+ environ-
space environment carry some degree of risk, ment (Phase A/B)
and failures in space of mechanisms have • Increased margins on mechanisms and de-
caused complications for missions in the past. tailed analysis of potential failure mecha-
In addition, the thermal environment of the nisms, including wear and lifetime issues
near-Sun portion of the orbit provides chal- • Trade study to eliminate mechanisms where
lenges to the use of mechanisms. If Solar possible (Phase A)
Probe+ mechanisms are inadequately designed, • Detailed qualification program for all
mechanism failure may occur, resulting in deg- mechanisms used in flight
radation or loss of mission. • Study to identify workarounds for failure
Mitigations include that may occur in flight (Phase A/B)
• Use of high-heritage mechanisms, including
detailed analysis of the suitability of each

3-78
4.0 COST ESTIMATE

4.0. COST ESTIMATE broken down by NASA Level 2 Work Break-


4.1. Cost Estimate Summary down Structure (WBS) element and provided
in FY07$ and real year dollars (RY$). This es-
A cost estimate for a complete Solar Probe+
timate is for the full mission cost and includes
mission was developed using the technical de-
the launch vehicle and launch services, third
sign described in Section 3 of this report. The
stage, all science and scientist participation,
cost estimate for Solar Probe+ is $739.5M in
and (DSN), environmental testing at the
fiscal year 2007 dollars (FY07$). The Solar
NASA/Goddard Space Flight Center (GSFC),
Probe+ cost estimate is provided in Table 4.1-1,
and 7 years of Phase E operations.

Table 4.1-1. Solar Probe+ mission cost table. Reserves have not been applied to Kennedy Space Center
(KSC) launch services, payload instruments, or DSN.
MISSION INSTITUTION SUMMARY FOR COST FOR SMD SOLAR PROBE+ ENGINEERING STUDY
FY Costs in Real Year Dollars (to the Nearest Thousand); Totals in Real Year and Fixed Year 2007 Dollars
Total
(Real Total
Cost Element Phase A Phase B Phase C/D Phase E Year) (FY 2007)
Task 01: Project Management $1,420 $2,755 $6,955 $8,587 $19,717 $15,636
Task 02: Project Systems Engineering $2,219 $5,962 $16,710 $22 $24,912 $21,301
Task 03: Safety and Mission Assurance $432 $1,378 $6,666 $1,246 $9,721 $8,015
Task 04: Science/Technology $833 $1,235 $3,135 $7,460 $12,663 $9,751
Task 05: Payload System $636 $1,800 $4,846 $58 $7,340 $6,264
Task 06: Spacecraft Bus $12,132 $57,215 $96,515 $896 $166,758 $144,013
Task 07: Mission Operations $200 $1,028 $10,956 $40,835 $53,020 $39,295
Task 09: Ground Data System $843 $2,507 $15,029 $295 $18,674 $15,694
Task 10: Systems Integration and Testing $231 $2,123 $23,746 $371 $26,471 $21,716
Task 11: Education and Public Outreach - $384 $935 $323 $1,643 $1,414
Task 12: Mission Design $496 $1,035 $2,874 $4,863 $9,268 $7,209
PI Mission Cost $19,441 $77,423 $188,367 $64,956 $350,187 $290,307
Payload Instruments $5,305 $33,437 $77,292 - $116,034 $100,000
Environmental Testing at GSFC - - $1,714 - $1,714 $1,398
Navigation $90 $210 $1,020 $1,680 $3,000 $2,315
KSC Launch Services - - $225,100 - $225,100 $184,468
Launch Vehicle Third Stage - - $7,700 - $7,700 $6,078
DSN - - - $22,940 $22,940 $16,509
External PI & Co-I Team, Phases A–D $681 $2,443 $4,876 - $8,000 $6,882
Science Ops. Preparations - - - $5,000 $5,000 $3,539
External PI & Co-I Team, Phase E - - - $50,000 $50,000 $35,394
Science Ops. Team - - - $8,000 $8,000 $5,663
Reserves $1,011 $24,023 $61,103 $19,445 $105,582 $86,933
Total Mission Cost $26,528 $137,536 $567,172 $172,022 $903,257 $739,489
Phase A: 5% Reserve
Phase B: 30% Reserve
Phase C/D: 30% Reserve
Phase E: 15% Reserve

4-1
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

4.2. Cost-Estimating Methodology and (CER) for others. The CERs are based on ac-
Independent Cost Estimate tual values for similar programs and are ad-
4.2.1. Cost-Estimating Methodology justed by decrementing the CERs for like
To determine mission cost feasibility, The elements priced discretely. Travel has been
Johns Hopkins University Applied Physics estimated based on a CER of similar pro-
Laboratory (APL) conducted a rigorous “bot- grams. For the purposes of this costing exer-
tom-up” cost estimate and an independent pa- cise, past experience included the MESSEN-
rametric costs analysis. The bottom-up meth- GER, New Horizons, and STEREO missions.
odology was used to estimate the Solar Probe+ This process resulted in a detailed bottom-up
mission cost, integrating a top-level schedule, (“grass roots”) cost estimate in which team
Work Breakdown Structure (WBS), resource- members populate spreadsheet templates with
level identification, and risk assessment (see resources against WBS elements that span the
Section 3.15). project schedule in 1-month increments. These
APL follows APL’s Cost-Estimating Man- lower-level inputs are quality-checked against
ual and APL’s Product Assurance System historical data for omissions, overlaps, or incon-
(PAS) document SDO-11386 that outlines sistencies. They also account for costs for APL
cost and schedule development and manage- payload administration, ground support equip-
ment standards for the Space Department. The ment, and emulators. The internal and external
cost-estimating process is described below, WBS data are then incorporated into the project-
with detailed cost information presented in level cost estimate using APL’s Resource Man-
Section 4.5. agement Information System (RMIS) to produce
Experienced functional supervisors and mission costs in fiscal year 2007 dollars
lead engineers are responsible for estimating (FY07$) and real year dollars (RY$).
the required labor resources and skill mix. Several detailed reviews are then conducted
The labor estimate is based on the technical among the study manager, the technical leads,
concept, engineering experience, an in-depth and their functional supervisors, and costs are
understanding of technical requirements, a revised accordingly. Near the end of this study
disciplined engineering process to identify period, an independent peer review evaluates
assumptions and cost sensitive areas, and ex- the draft study final report for technical and
perience on similar programs. Direct labor costing completeness, followed by a Space De-
requirements are estimated by labor classifi- partment management review of the final cost
cation. Direct labor hours associated with all plan. These reviews allow management and
phases of Solar Probe+ are included in the senior staff to evaluate the final report, to re-
cost estimate. APL direct (and indirect) labor view the costs and assumptions on which the
rates are based on the forward-pricing rates costs are based, and to access cost realism. Any
submitted to the designated Administrative omissions or errors are exposed and corrected
Contracting Officer (ACO) and the Defense during this process.
Contract Management Command (DCMC) The Solar Probe+ cost estimate, as with the
assigned to APL. These rates are applied to entire engineering concept, assumes no contri-
direct costs in accordance with APL prac- butions from foreign partners or non-NASA
tices. Procurement costs are based on vendor U.S. government agencies.
rough orders of magnitude (ROMs) or rele- 4.2.2. Cost Estimate Validation:
vant recent experience. Estimates for miscel- Independent Cost Estimate
laneous other direct costs (MODC) are dis- 4.2.2.1. Summary
crete for specific identified elements and An independent cost estimate (ICE) was pre-
based on a cost-estimating relationship pared for APL’s portion of the Solar Probe+

4-2
4.0 COST ESTIMATE

spacecraft to validate the comprehensive, bot- • Program management (PM; NASA WBS 01)
tom-up Program Office estimate (POE), quan- • Systems engineering, including mission
tify cost risks, and determine the sufficiency of analysis (SE; WBS 02)
cost reserves. The ICE costing methodology is • Mission and safety assurance (M&SA;
based on parametric cost-estimating models and WBS 03)
risk estimation tools used throughout the space • Payload administration and emulators,
exploration and observation community. ICE APL’s portion of payloads (WBS 05)
results at the third WBS level enable detailed • Spacecraft elements (WBS 06)
comparisons with the POE. • Attitude determination and control subsys-
The ICE results correspond to approximately tem (ADACS)
85% of the total Solar Probe+ Phases A–D. • Flight software development
They account for hardware and flight software • Structural and mechanical, including harness
costs for Phases A–D of the spacecraft and asso- • Thermal control
ciated management, engineering, integration, • Propulsion
and testing activities.They do not account for • Electrical power and distribution
costs of ground systems or the mission and sci- • Command, control, communications, and
ence operations centers, nor do they account for data handling (CC&DH)
the costs of instrument design and development. • Telecommunications
The Solar Probe+ program presents ex- • Launch and early orbital operations (30
tremely challenging and unique performance days), Phase D portion of mission opera-
requirements. Unfortunately, the most appro- tions (WBS 07)
priate technical solutions are not captured in the • Integration and testing (I&T, WBS 10)
space industry’s cost databases and models ICE results are generally consistent with the
used for generating ICEs. For instance, the so- POE, where the ICE point estimates are based
lar environment in which Solar Probe+ will on the current best estimate (CBE) subsystem
operate requires unique thermal shielding and masses provided by APL systems engineers.
solar array mechanisms. The ICE attempts to POEs for the spacecraft integrating functions of
represent the costs of these solutions in several PM, SE, and I&T are within 5% of the corre-
ways. For the Thermal Protection System sponding ICE point estimates. (The ICE point
(TPS), a complexity factor was developed from estimate bases estimated cost on the CBE
cost data on the Space Shuttle nose cone. A masses.) The spacecraft POE and point-estimate
mechanism cost equation was used to estimate ICE are within 3.1%. Estimates for smaller cost
the cost of the solar array mechanism. To esti- elements related to payload SE, mission opera-
mate the possible range of final subsystem tions support, and ground systems equipment do
costs, larger mass growth than typical for APL show larger variances, however. This is because
missions was assumed. When estimating flight of the conservativeness of our ICE assumptions
software development, claims about the avail- and to ICE cost equations whose scope are
ability and applicability of heritage software broader than some POE elements.
code were discounted, and mostly new soft- When technical and cost-estimating risks are
ware code was assumed to be needed. All of considered, ICE results indicate that properly
these assumptions and modeling decisions re- managed reserves will likely be sufficient to
sulted in conservative ICE results that capture complete Phase D without cost overruns. The
the technical and cost-estimating risks inherent spacecraft bus POE with 30% reserves falls at
in such a unique mission. the 70th percentile of the corresponding ICE S
Cost elements that were accounted for by curve. The S curve is a graphic representation of
the ICE include the following: the cumulative probability distribution (CPD) of

4-3
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

likely costs that takes into account uncertainty lished by Tecolote Research, Inc., in October
related to mass growth and cost estimating. A 2001. 1 USCM8 provides cost-estimating
70th percentile value corresponds to a 30% like- relationships (CERs) for estimating unmanned,
lihood of a cost overrun. The I&T POE with Earth-orbiting space vehicle costs. According
cost reserves falls at the 60th percentile; the to the USCM documentation:
SE/PM/M&SA POE falls at the 65th percentile. Each CER was developed through
All of this suggests that the mission is achiev- rigorous statistical analysis of hypothesized
able at proposed funding levels, although it will cost drivers. This was done by generating
require careful management of cost reserves, hypotheses relating cost to those underlying
shifting dollars to cover risks as they arise. parameters thought to be cost drivers. Each
4.2.2.2. Methodology tested hypothesis was based on a sound
4.2.2.2.1. Ground Rules and Assumptions understanding of the engineering principles
The following ground rules and assumptions that might drive cost. Selected CERs had to
guided preparation of the Solar Probe+ ICE: demonstrate favorable statistics and, from a
• Solar Probe+ costs are presented in fiscal behavior standpoint, be consistent with
year 2007 (FY07) dollars. Because the ICE engineering expectations. As a result,
was estimated in FY08 dollars, a 3.1% an- CERs that did not make sound engineering
nual deflation rate was used to adjust the sense were not selected despite having
Solar Probe+ ICE estimate from FY08 to good statistical measures.
FY07 dollars. The USCM8 subsystem cost equations used
• Spacecraft estimates include the costs by for the Solar Probe+ ICE are mass-driven.
subsystem to design, produce, integrate, and Nonrecurring equations represent a “new de-
test a single spacecraft. Also included are sign” effort and must be adjusted for heritage
estimated costs for management, SE, and assumptions; therefore a factor (fND) based on
I&T of spacecraft subsystems. percentage new design was applied to the non-
• The spacecraft estimate includes effort to recurring CER to account for anticipated heri-
design, code, and flight test and test bed tage as shown below:
software modules, software development
management, and integration of software (2.8ND + 0.2)
and hardware. f ND = ,
3
• Risk-adjusted technical and performance
inputs to the model are provided by the So- where ND (0.0 < ND < 1.0) is the new design
lar Probe+ system engineers, based on their fraction.
technical assessments and judgments. This factor adjusts the USCM8 nonrecurring
• Payload instrument costs were not esti- equation output to estimate more accurately the
mated, but their cost provided a basis for es- true costs of the nonrecurring effort experienced
timating APL SE costs in support of instru- by programs in the USCM8 database. The
ment design and development.
USCM8 equations enable analysts to estimate
4.2.2.2.2.Parametric Cost-Estimating quantitatively the contribution of design heritage
Models and Data Used to to final cost and were used for all Solar Probe+
Generate ICEs
hardware subsystems.
Spacecraft Hardware, Launch Vehicle
Adapter, and Other Program Costs. The ICE
used the Eighth edition of the Air Force’s Un- 1
Unmanned Space Vehicle Cost Model, Eighth Edition
manned Spacecraft Cost Model (USCM8) to (USCM8), Tecolote Research, Inc., Goleta, CA, www.
estimate nonrecurring and recurring costs at the uscm8.com (October 2001).
subsystem level. USCM8 Version 1.1 was pub-

4-4
4.0 COST ESTIMATE

Cost equations for such nonhardware ele- NAFCOM data sets identified only one mis-
ments as PM and SE are driven on the esti- sion, Tracking and Data Relay Satellite Sys-
mated costs of spacecraft hardware and soft- tem (TDRSS), with dual Ka/X-band capabili-
ware and I&T. The equations are derived from ties like those currently proposed for Solar
a total of 44 spacecraft missions. Eleven of the Probe+. A new telecommunications cost equa-
missions were funded by NASA, 20 were tion was generated from USCM8 cost data
funded by the U.S., and the remaining 11 were that weighted more heavily the TDRSS his-
funded commercially. torical data; that equation was used to estimate
Three cost estimates, representing the best, the Solar Probe+ telecommunications subsys-
most-likely, and worst cases, were generated tem. Although the new equation is likely a
for each spacecraft subsystem and cost ele- better estimator for complex dual-band tele-
ment. The three estimates provide the bases for communications subsystems, its statistical er-
probabilistic cost estimates, described below, ror is substantially larger than the standard
which enable the ICE to account for the effects equation, and cost results are less reliable.
of cost uncertainty. Based on a recent study by Flight Software Development. The open-
The Aerospace Corporation on spacecraft mass source COCOMO-II model developed by Barry
growth during the design and development Boehm’s University of Southern California
phases, subsystem masses provided by APL software research group was used to estimate
system engineers were adjusted for the three costs of flight software development. CO-
estimates as follows: CBE masses were pro- COMO-II bases its cost estimate on the effective
vided by the system engineer. The best-case amount of code that must be developed, adjusted
estimates are based on the CBE masses plus for heritage code that can be adapted, or used
10%; the most-likely estimates, CBE masses without change and productivity. COCOMO-II
plus 30%; and the worst-case estimates, CBE products include a time-phased profile of devel-
masses plus 50% (see Table 4.2-1). opment effort and low, most-likely, and high
The TPS is unique and more technologically cost estimates, which for Solar Probe+ are based
complex than typical aluminum structural com- on APL labor rates.
ponents. Accordingly, ICE analysts sought his- The estimated total number of source lines
torical data on large carbon–carbon structures of code to be developed is 110,000; 75,000 of
that could be used to adjust the USCM those lines of code serve guidance and control
structure/mechanical CER. A search of the functions, and the remaining 35,000 lines
NASA–Air Force Cost Model (NAFCOM) serve command and data handling functions.
data set identified the Space Shuttle nose cone ICE analysts chose not to discount the esti-
as an analog. Detailed cost data was used to de- mated effort for heritage code, including code
velop a complexity factor that adjusts the that might be available from the Radiation
USCM structure cost equation for the relatively Belt Storm Probes (RBSP) mission. Accord-
high cost per pound of carbon–carbon structures. ingly, the ICE results should be regarded as
For the electrical power subsystem, ICE very conservative estimates of flight software
analysts applied the USCM struc- development effort and costs.
ture/mechanical CER to estimate costs of the Payload Systems Engineering. Although
secondary solar array’s complex mechanical the Solar Probe+ instruments are outside the
components. The remainder of the subsys- scope of the ICE, an ICE was generated for
tem’s components was estimated with the APL’s effort for instrument SE and product
USCM electrical power subsystem CER. assurance (specifically, APL WBS 210 Payload
Estimating the cost of the telecommunica- Administration and WBS 280 Payload Emula-
tions subsystem from historical data proved tors). The SE cost factor from the NASA In-
challenging. A check of the USCM8 and

4-5
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

strument Cost Model (NICM) was applied to craft bus, for example, were modeled assuming
the total instrument cost basis to estimate new design would be 80–100%.
APL’s instrument SE costs. Because the factor Our ICE approach tries to quantify the cost
applied to all SE activities associated with in- risk inherent in the estimated costs. It recognizes
strument design, production, and I&T, the ICE that cost estimates represent predictions about
result likely overestimates APL efforts directed future costs and uses Monte Carlo simulation to
at ensuring interoperability of spacecraft bus assign probabilities to those predictions.
and payloads. Two kinds of uncertainty contribute to dis-
Accounting for Cost Uncertainty. Reuse crepancies between predicted and actual costs:
of heritage designs from previous APL mis- 1. Uncertainty in input variables such as
sions is critical to ensuring that the Solar mass and new design factor assumptions.
Probe+ mission is executable and affordable, 2. Statistical uncertainty in CERs and cost
which leads to a pressure for adjusting cost es- factors.
timates for likely heritage savings. However, The Monte Carlo simulation accounts for
recent analysis of NASA missions indicates both kinds of uncertainties as follows:
that, after initial design, there is a trend for in- • Subsystem mass uncertainty is applied to all
creases in mass and cost. Recent unpublished subsystem mass estimates without contin-
research of spacecraft mass growth finds 22– gency. The uncertainty is modeled using a tri-
25% mass growth is typical, and mass growth angular probability density function (PDF)
of 50% or more is not uncommon. The ICE based on a study prepared by The Aerospace
heritage design assumptions are conservative, Corporation, which shows mass growth the
in that we have assumed reuse will be minimal. minimum mass growth to be 10%, the most-
Most functional subsystems within the space- likely mass growth to be 30%, and the maxi-
Table 4.2-1. Subsystem masses used in the ICE to estimate Solar Probe+ subsystem costs account for
the likelihood during the design phases of mass growth. TT&C, telemetry, tracking, and control.

4-6
4.0 COST ESTIMATE

mum mass growth to be 50%. The results of The graphical representation of the CPD is
this study were published.2 sometimes called an S curve. CPDs offer three
• Uncertainty in the new design factor is advantages over adding the costs of individual
modeled by using a log-normal PDF with a elements to predict total cost:
mean of 1.0 and a standard deviation in unit 1. They avoid the bias inherent in arithmetic
space of 0.30, which accounts for the uncer- summing of point estimates, which tends
tainty in the new design factor assumption. to misstate total cost.
• The uncertainty in the nonrecurring and re- 2. They provide a sense of the confidence
curring CERs were modeled by using a log- we should associate with the proposed
normal PDF with a mean of 1.0 and a stan- cost. If, for example, the cost proposal is
dard deviation in unit space equal to the at the 70th percentile, we would, if our
percentage standard error of the particular model has faithfully represented the
CER, which ranges from 0.12 to 0.44. “known unknowns,” expect to overrun
• To represent the observation that increases in that prediction 30% of the time.
one element’s cost are reflected in increased 3. The shape of CPDs suggests the size of
costs for other cost elements—a frequently the cost overrun that might be incurred in
observed trend in the cost histories of space cases of where technical or other risks
systems—the values selected during the are not effectively controlled. All things
Monte Carlo simulation with an intercorrela- considered, an estimate with a small dif-
tion matrix with all elements are statistically ference between the 50th and 70th per-
correlated at a low level (Pearson r = 0.097). centile predictions is preferable to one
The effect is that, for example, selection of a with a large difference because fewer
high (or low) structure costs will result in cost reserves are required to assure mis-
correspondingly higher (lower) costs for sion success.
management, SE, I&T, etc. 4.2.3. Results
• A Monte Carlo model is built by using the The ICE results account for approximately
commercial Crystal Ball simulation tool. In- 85% of APL’s costs during Phases A–D. Fig-
puts to the model include the subsystem PDFs ure 4.2-1 shows the ICE S curve for Solar
and correlation matrix. The model simulates Probe+ costs. Table 4.2-2 provides a summary
the mission being performed thousands of of ICE results and comparison with the POE,
times. For each iteration, a cost prediction for by NASA WBS element.
each element is selected at random based on The element POEs generally are consistent
its PDF and correlation constraints. with the ICE point-estimate results. The esti-
• Adjusted cost predictions are summed at the mates for the spacecraft integrating functions of
completion of every iteration, after predic- PM, SE, and I&T are within 8%. The space-
tions have been drawn and adjusted for craft bus POE and ICE point estimate are
every cost element. within 9%. Estimates for smaller cost elements
• Finally, after completion of 10,000 itera- related to payload SE and mission operations
tions, the element and aggregate outputs are support show larger variances because of con-
characterized statistically. servative ICE assumptions and ICE methods
The result is a CPD that reports the prob- that include other costs in their calculations.
ability of each predicted cost’s occurrence. When technical and cost-estimating risks are
considered, the ICE results indicate that re-
2
Covert, R., Ten Common Things Wrong with Cost Risk serves will likely be sufficient. The spacecraft
Analysis, Presented at 76th SSCAG, Hampton Roads, bus POE with 30% reserves falls at the 70th
VA (October 22–23, 2002). percentile of the corresponding ICE S curve, a

4-7
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

be sufficient is slightly
less than two-thirds of
most cases for APL’s per-
formance of its manage-
ment and engineering
functions.
Spacecraft Bus
(WBS Element 06). The
spacecraft ICE point es-
timate of $156.0 million
is $12.7 million, or 9%
higher than the spacecraft
POE of $143.3 million.
Some of the difference is
attributable to the flight
Figure 4.2-1. Solar Probe+ ICE S curve. (All costs are in FY07$M.) software development
estimates. The flight
software estimate differ-
graphic representation of the CPD of likely ence of $10.1 million is a consequence of dif-
costs that takes into account uncertainty related ferent assumptions about the reuse of RBSP
to mass growth and cost estimating. The 70th and other heritage source code: The ICE does
percentile value corresponds to a 30% likeli- not discount effort for the availability of heri-
hood of a cost overrun. The I&T POE with cost tage code.
reserves falls at the 60th percentile; the The largest cost difference among the
SE/PM/M&SA POE falls at the 65th percentile. hardware elements is associated with the elec-
All of this suggests that the mission is achiev- trical power subsystem. The ICE point esti-
able at proposed funding levels although it will mate is $18.1 million, $8.5 million or 32%
require careful management of cost reserves, less than the POE of $26.2 million. The higher
shifting dollars to cover risks as they arise. POE is attributed to the complexity of the sec-
Key results from the ICE-POE comparison ondary solar array.
are highlighted below. The ICE for the avionics and telecommunica-
Program Management, Systems Engi- tions subsystems is $48.7 million, $9.1 million
neering, and Safety and Mission Assurance or 23.2% more than the POE of $39.5 million.
(NASA WBS Cost Elements 01, 02, and 03). The telecommunications ICE result is biased
USCM8 provides a single cost equation for es- higher by the reliance on TDRSS as the most
timating costs of these three cost elements. analogous spacecraft. Even with the abovemen-
Correspondence between the ICE and POE for tioned discrepancies among certain spacecraft
the sum of three elements is high. The ICE subsystems, the POE with 30% cost reserves for
point estimate, corresponding to the estimated the spacecraft bus falls at the 70th percentile of
cost using the CBE mass as input to the USCM the ICE S curve.
cost equation, is $49.5 million, only 7% (or Payload (WBS Element 05). Because
$3.4 million) more than the POE estimate of NASA has budgeted separately for instruments,
$46.1 million. The POE falls at the 47th per- APL’s payload costs are limited to Payload
centile on the corresponding ICE S curve. The Administration (APL WBS 210) and Payload
likelihood of the POE with 30% cost reserves Emulators (APL WBS 280). The NICM SE cost
rises to the 65th percentile on the ICE S curve.
In other words, the POE with cost reserves will

4-8
4.0 COST ESTIMATE

Table 4.2-2. Comparison of ICE point estimate and POE, by NASA WBS element. [All costs are in fiscal year 2007 millions of dollars (FY07$M).]
POE + Cost Re-
ICE Results POE ICE-POEΔ serve
Point Mean
NASA POE WBS Est. Est. Std. Dev. ICE Per- Percent- ICE Per-
WBS Description Elements FY07$M FY07$M FY07$M FY07$M centile FY07$M age FY07$M centile
TOTAL (Spacecraft Only—No$100M Payload) $244.8 $275.0 $69.5 $218.5 20% $9.2 4.2% $284.1 65%
01, 02, 03 PM, SE, S&MA 1xx $49.5 $56.2 $29.7 $46.1 47th $3.4 7.4% $59.9 65%
01 Project Management 11x $24.7 $28.0 $15.0 $13.9 10th $10.8 77.8% $18.1 30%
02 Systems Engineering 12x, 13x $24.7 $28.0 $15.0 $25.0 50th $(0.3) –1.2% $32.5 75%
Safety and Mission
03 Assurance 14x $7.1 N/A N/A N/A N/A
Payload 210, 280
05 (Management Only) only $12.4 $12.4 $5.0 $6.2 10th $6.2 99.4% $8.1 20%
06 Spacecraft 3xx $156.0 $173.7 $28.9 $143.3 15th $12.7 8.9% $186.3 70%
ADACS 31x $12.5 $14.1 $5.1 $11.5 30th $1.0 8.3% $15.0 65%
Flight Software 38x $24.1 $27.1 $4.5 $14.1 1st $10.1 71.5% $18.3 10%
Structures/Thermal 34x, 35x $47.3 $53.2 $18.6 $47.1 45% $0.2 0.3% $61.3 75%
Propulsion 39x $5.4 $5.9 $2.3 $4.5 30th $0.9 19.4% $5.9 60%
Electrical Power 32x, 33x $18.1 $19.6 $4.3 $26.6 90th $(8.5) –32.0% $34.5 95%
TT&C, C&DH, and
Communications 36x, 37x $48.7 $53.4 $13.6 $39.5 10th $9.1 23.2% $51.4 50%
Mission Operations
07 (Prelaunch) 5xx $5.5 $5.4 $3.6 $3.2 30% $2.3 71.2% $4.2 50%
10 Systems I&T 4xx $21.4 $28.0 $27.0 $19.7 50% $1.7 8.7% $25.6 60%

4-9
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

factor used by the ICE predicts the cost of all 4.3. Schedule
instrument-related activities will be $12.4 mil- For costing and engineering planning pur-
lion (point estimate), $6.2 million or 99% more poses, the Solar Probe+ mission used the top-
than the POE of $6.2 million. With cost reserves level milestone schedule shown in Figure 4.3-1.
included, the POE cost is $8.1 million. Given The schedule consists of a 13-month Phase A, a
the ICE cost factor estimates all instrument- 19-month Phase B, a 19-month Phase C, a 23-
related SE, the difference is not surprising. month Phase D, and a 7-year Phase E. Funded
Mission Operations (WBS Element 07). schedule reserves totaling 9.5 months [6.5
The only mission operations element estimated weeks for Phase B (1.0 month/year), 11.5 weeks
by the ICE is launch operations and orbital for Phase C (1.8 months/year), and 20 weeks for
support. The ICE point estimate, based on a Phase D (2.6 months/year)], as shown in the
cost factor, is $5.5 million, $2.3 million or 71% project master schedule, are funded at the peak
more than the POE of $3.2 million. The POE burn rate for cost-estimating purposes.
with cost reserves is $4.2 million. The presented schedule is reasonable, based
Systems Integration and Testing (WBS on recent experience. Table 4.3-1 compares
Element 10). The ICE point estimate of $21.4 the Solar Probe+ schedule with three recent,
million is within $1.7 million, or 9%, of the analogous, APL-led missions that successfully
POE of $19.7 million. Note, however, that the completed Phases A–D.
ICE does not include costs for integration of
payloads with the spacecraft.

Figure 4.3-1. Solar Probe+ summary schedule.

4-10
4.0 COST ESTIMATE

Table 4.3-1. Comparison of Solar Probe+ schedule 4.4.1. Work Breakdown Structure
with three recent missions. Dictionary
Phases A/B Phases C/D The following is a summary of the descrip-
Mission (months) (months)
MESSENGER 20 38
tions of each WBS element, along with top-
level costing assumptions, as appropriate.
New Horizons 16 41
WBS 01: Project Management. This WBS
STEREO 29 60
includes project management and business
Average 22 46
management, administrative support and
Solar Probe+ Schedule 24 50
earned-value management, schedule and re-
Per NASA task order NNN07AA15T, Phase source control, subcontract management, tech-
A is assumed to begin in 2009. Phases A and B nology transfer, and reporting of technical and
will involve the activities typical to those financial status to NASA. It also includes pro-
phases, including instrument accommodation ject scientist, export control issues, online
and development and documentation of the documentation, and design review support.
flowdown of the mission and science require- Project management staffing levels are based
ments to the subsystems, instruments, ground on actual expenditures on MESSENGER, New
systems, and mission operations. During this Horizons, and STEREO.
WBS 02: Project Systems Engineering.
phase, procurement activities for all major sub-
This WBS includes the mission system engi-
contracts will be initiated as the requirements
neer, spacecraft system engineer, reliability
for these components are determined. By the
engineering, system verification, electromag-
end of Phases A/B, subcontract selection
netic compatibility (EMC)/electromagnetic
should be complete, and all vendors should be
interference (EMI) system engineering, inter-
under contract.
face control documentation, onboard fault pro-
Phases C/D will include the initiation of the
tection, and precision onboard timekeeping.
detailed design process and will end with the
Systems engineering staffing levels are based
delivery of the fully integrated spacecraft to
on recent experience and have been aug-
the launch site. The 50-month Phases C/D al-
mented in recognition of the mission design
low for ample reserve during integration and
complexity and reliability issues that are in-
testing (I&T).
herent to this mission.
Phase E will start 30 days after launch and
WBS 03: Safety and Mission Assurance.
will continue for 7 years. Phase E will end 2
This WBS includes performance assurance,
months after the 24th solar pass.
contamination, safety, hardware quality assur-
4.4. Work Breakdown Structure and ance, spacecraft charging, reliability engineer-
Cost Detail ing (separate from that listed in WBS 02),
The Solar Probe+ mission cost estimation parts engineering, software quality assurance,
process used a level-three Work Breakdown contamination, and configuration control.
Structure (WBS), shown in Figure 4.4-1. This Staffing levels are based on past experience,
WBS is consistent with the NASA standard but augmentations have been made to accom-
WBS defined in Appendix G of NPR modate current NASA safety and mission as-
7120.5D, and it will be used to allocate re- surance practices.
sponsibility and resources during all phases of WBS 04: Science Team. This WBS in-
the Solar Probe+ project. cludes the project scientist and deputy project
scientist during Phases A–E as well as project
support to science teams and payload interface

4-11
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

Figure 4.4-1. The level-three WBS for the Solar Probe+ mission.

4-12
4.0 COST ESTIMATE

engineering. It is assumed that the mission this cost includes any instrument-specific data
provider will have a project scientist and sup- processing units and cost reserves necessary to
port staff. Staffing levels for this portion are ensure on-cost delivery of the payload.
based on recent experience with MESSEN- WBS 06: Spacecraft Bus. This WBS ele-
GER, New Horizons, and STEREO. ment is active in Phases A–D. It includes the
It is further assumed that the payload suite effort associated with definition, engineering,
and associated principal investigator (PI) and design, fabrication, procurements, and assem-
coinvestigators (Co-Is) will be competitively bly and testing of the spacecraft bus and com-
selected and therefore will not necessarily be ponents. After the subsystems are delivered to
associated with the mission provider’s institu- spacecraft-level I&T, the effort to support
tion. For this reason, costs that are directly as- spacecraft integration is captured under a
sociated with these portions of the science ef- spacecraft I&T WBS.
fort are shown separately from the rest of WBS WBS 07: Mission Operations. This WBS
04 in Table 4.1-1. These “external” costed element contains the entire engineering team
subelements are as follows: during Phase E and mission operations support
• PI and Co-I team for Phases A–D during Phases A–D. It also includes Deep
• PI and Co-I team for Phase E Space Network (DSN) costs. The mission op-
• The effort required for the science team to erations phase for Solar Probe+ will indeed be
prepare for science operations. This complex. Costs for this WBS are based on the
subelement is active in Phases C/D analogous efforts for MESSENGER (which
Science operations in Phase E has a similar numbers of planetary flybys) and
The costs for these subelements given in has been augmented based on lessons learned
Table 4.1-1 were provided by the Solar from MESSENGER and the added complexity
Probe+ Science and Technology Definition of Solar Probe+.
Team (STDT). These costs are deemed by the DSN costs were estimated by using the Jet
STDT to be sufficient to produce the head- Propulsion Laboratory DSN Aperture Fee
turning science expected of Solar Probe+. Tool.
In addition, the STDT strongly recommends WBS 08: NASA Launch Vehicle. This
that a Participating Science Program (PSP) be WBS includes the NASA-supplied launch ve-
established in order to broaden the pool of hicle and is used for reporting total mission
contributors to Solar Probe+ science beyond cost only. The mission design assumes an Atlas
those of the selected science team. The STDT V 551 launch vehicle. The cost presented for
has recommended that this PSP be funded at the launch vehicle is from an estimate received
$8M, in real year dollars (RY$), during Phase in February 2008 from NASA/Kennedy Space
E. However, because this recommendation Center (KSC).
falls outside the scope of this study, it is not In addition, Solar Probe+ will use a STAR-
included in Table 4.1-1. 48BV third stage. A cost estimate was ob-
WBS 05: Payload. This WBS includes tained from ATK for $5.6M in RY$. This es-
payload administration as well as the NASA timate has been augmented in order to cover
payload cost for the selected instruments for potential additional engineering activities as-
total mission cost reporting. This WBS also sociated with the third stage. (In addition, a
includes spacecraft emulators to be supplied to full-time-equivalent I&T engineer dedicated to
the instrument teams. The complete instru- third-stage I&T has been included in WBS
ment payload package cost is held at $100M 10). Note also that, unlike with the Atlas V
in fiscal year 2007 dollars (FY07$) per NASA 551, 30% reserve has been applied to the cost
task order NNN07AA15T. It is assumed that of the third stage. A more complete discus-

4-13
SOLAR PROBE+ MISSION STUDY REPORT

sion of the third stage is provided in Section tivities, we felt it prudent to include 5% on the
3.1.7. Solar Probe+ Phase A in recognition of the
WBS 09: Ground Data System (GDS). variability that risk reduction activities can
This WBS element is active in Phases A–D. introduce. Cost reserves of 30% are held
This effort contains the prelaunch GDS and against Phases B–D, which is commensurate
mission operations system (MOS) efforts, in- with APL practices and seems appropriate
cluding GDS hardware and software such as given past experience, coupled with the more
test simulators to support spacecraft and mis- than 9 months of funded schedule reserves in
sion operations testing. For the costing pur- the phases (that is, thereby reducing risk of
poses, we have assumed eight test beds. Based overrunning due to falling behind schedule).
on past experience, this number seems pru- In addition to these reserves, the estimated
dent. cost holds 15% in Phase E. Past experience
WBS 10: Project Integration and Testing. throughout industry indicates 5% to 15% is
This WBS element is active in Phases A–D. the norm. Given the complexity of the trajec-
This effort includes the I&T team and the sub- tory and potential for real-time science dis-
system engineers when they are supporting coveries, we felt it prudent to hold reserves at
spacecraft-level I&T. Costing of the WBS the high end of this range.
element assumes all normal I&T functions Funding reserves are held on all WBS ele-
will occur at APL with the exception of ther- ments, including all science activities, with the
mal vacuum and acoustic and shock testing. exception of the Atlas V 551 launch vehicle,
These test are assumed to take place at payload (considered to be self-contained
NASA/Goddard Space Flight Center (GSFC). within the $100M cap), and DSN.
The costs for these external activities are 4.5. Subsystem Cost Detail
shown separately in Table 4.1-1. Table 4.5-1 provides Solar Probe+ subsys-
WBS 11: Education and Public Outreach
tem cost detail. The cost element numbers re-
(E/PO). This WBS includes all E/PO costs as-
fer to the APL internal Work Breakdown
sociated with the Solar Probe+ mission. Dur-
Structure (WBS) subelements. These have
ing this study., E/PO was not investigated in
been mapped into the NASA WBS shown in
depth, but a program similar to that of STE-
Figure 4.4-1 in order to produce Table 4.1-1.
REO is envisioned. A mission to the Sun of-
fers many obvious exciting opportunities, es-
pecially with the in situ elements. For this
study, the E/PO allocation was defined as
0.5% of the Phases A–E cost (less the launch
system).
WBS 12: Mission Design. This WBS in-
cludes the APL effort and subcontractor sup-
port for mission design, navigation, and flight
dynamics design and analysis in Phases A–E.
It is based on experience with MESSENGER
and New Horizons.
WBS R: Reserves. This WBS contains the
project funding reserves.
Reserves are calculated as the percentage of
funds to go within each phase. Although re-
serves are not normally held on Phase A ac-

4-14
4.0 COST ESTIMATE

Table 4.5-1. Solar Probe+ subsystem cost detail. Reserves have not been applied to Kennedy Space
Center (KSC) launch services, payload instruments, or DSN.
MISSION INSTITUTION SUMMARY FOR COST FOR SOLAR PROBE+ ENGINEERING STUDY
FY Costs in Real Year Dollars (To The Nearest Thousand); Totals in Real Year and Fixed Year 2007 Dollars
Total
Phase (Real Total
Cost Element Phase A Phase B C/D Phase E Year) (FY 2007)
Program Management 11* $2,154 $3,990 $10,090 $16,005 $32,239 $25,261
System Engineering 12* $2,219 $5,962 $16,710 $22 $24,912 $21,301
Mission Design and Analysis 13* $496 $1,035 $2,874 $4,863 $9,268 $7,209
Performance Assurance Engineering 14* $432 $1,378 $6,666 $1,246 $9,721 $8,015
Payload 2* $636 $1,800 $4,846 $58 $7,340 $6,264
ADCS & GC 31* $722 $7,398 $5,108 - $13,228 $11,508
Power 32* $1,955 $9,988 $15,973 - $27,915 $24,246
Harness 33* $71 $604 $2,039 - $2,714 $2,310
Spacecraft Structures and Mechanisms 34* $1,078 $3,084 $14,501 $90 $18,754 $15,911
Thermal 35* $7,141 $12,650 $16,076 - $35,867 $31,305
RF Communications 36* $153 $10,351 $15,502 $156 $26,163 $22,521
C&DH 37* $483 $10,403 $8,538 - $19,425 $17,105
Flight Software 38* $394 $1,283 $15,132 $649 $17,458 $14,569
Propulsion 39* $134 $1,454 $3,646 - $5,235 $4,537
I&T 4** $329 $2,123 $19,701 $22 $22,176 $18,339
Launch Ops. & Early Orbital Ops. 5* - - $4,045 $349 $4,394 $3,470
Ground Data System 7** $1,043 $3,535 $25,973 $295 $30,846 $25,801
Mission Ops. and Data Analysis 8** - - $12 $40,877 $40,890 $29,222
Education and Public Outreach - $384 $935 $323 $1,643 $1,414
APL Mission Cost $19,441 $77,423 $188,367 $64,956 $350,187 $290,307
Payload Instruments $5,305 $33,437 $77,292 - $116,034 $100,000
Environmental Testing at GSFC - - $1,714 - $1,714 $1,398
Navigation $90 $210 $1,020 $1,680 $3,000 $2,315
KSC Launch Services - - $225,100 - $225,100 $184,468
Launch Vehicle Third Stage - - $7,700 - $7,700 $6,078
DSN - - - $22,940 $22,940 $16,509
External PI & Co-I Team, Phases A–D $681 $2,443 $4,876 - $8,000 $6,882
Science Ops. Preparations - - - $5,000 $5,000 $3,539
External PI & Co-I Team, Phase E - - - $50,000 $50,000 $35,394
Science Ops. Team - - - $8,000 $8,000 $5,663
Reserves $1,011 $24,023 $61,103 $19,445 $105,582 $86,933
Total Mission Cost $26,528 $137,536 $567,172 $172,022 $903,257 $739,489
Phase A: 5% Reserve
Phase B: 30% Reserve
Phase C/D: 30% Reserve
Phase E: 15% Reserve

4-15
APPENDIX A: SOLAR PROBE+ MASS AND POWER BUDGETS

APPENDIX A:

SOLAR PROBE+ MASS AND POWER BUDGETS

A-1
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

A-2
APPENDIX A: SOLAR PROBE+ MASS AND POWER BUDGETS

Table A-1. Solar Probe+ mass budget.


CBE Each CBE Total Contingency Total
Name Qty (kg) (kg) (%) (kg)
Instruments
Instruments Total: 47.2 51.9
Hemispheric Imager 1 1.5 1.5 10% 1.7
TBD Science Payload 1 5.0 5.0 10% 5.5
Fast Ion Analyzer 1 2.8 2.8 10% 3.1
Fast Electron Analyzer 2 2.5 5.0 10% 5.5
Ion Composition Analyzer 1 7.0 7.0 10% 7.7
Plasma Wave Electronics 1 2.0 2.0 10% 2.2
Plasma Wave Pre-amp 3 0.4 1.2 10% 1.3
Search Coils 3 0.6 1.8 10% 2.0
Magnetometer 1 2.5 2.5 10% 2.8
Energetic Particle Inst. High Energy 1 2.7 2.7 10% 3.0
Energetic Particle Inst. Low Energy 1 1.4 1.4 10% 1.5
Neutron-Gamma Spectrometer 1 2.0 2.0 10% 2.2
Coronal Dust Detector 1 1.5 1.5 10% 1.7
Common DPU 1 10.8 10.8 10% 11.9
Accommodation Hardware
Accommodation Hardware Total: 7.4 8.4
Fast Plasma Actuator Bracket 1 2.2 2.2 15% 2.5
Plasma Wave Antenna, Actuator 3 1.2 3.6 15% 4.1
Magnetometer Boom 1 1.6 1.6 5% 1.7
Spacecraft
Spacecraft Total: 373.7 420.1`
Telecommunications
Telecommunications Total: 31.8 34.8
High Gain Antenna 1 3.8 3.8 10% 4.2
HGA Actuator 2 2.0 4.0 5% 4.2

High Gain Arm, mounting bracket, rotary joints, etc. 1 2.9 2.9 15% 3.3
HGA Actuator Electronics 1 1.0 1.0 5% 1.1
K-Band TWTA/EPC 2 3.0 6.0 5% 6.3
X-Band TWTA/EPC 2 2.5 4.9 5% 5.1
Low Gain Antenna Assembly 2 0.7 1.4 15% 1.6
RF Plate 1 1.0 1.0 15% 1.2
Hybrids/Components 1 1.0 1.0 15% 1.2
Transponder 2 1.6 3.2 15% 3.7
Coax/Waveguide 1 2.6 2.6 15% 3.0
Guidance and Control
Guidance and Control Total: 30.4 32.0
Inertial Measurement Unit 1 6.6 6.6 5% 6.9

A-3
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

CBE Each CBE Total Contingency Total


Name Qty (kg) (kg) (%) (kg)
Reaction Wheel Assembly 4 4.2 16.8 5% 17.6
Star Tracker 3 2.0 6.0 5% 6.3
Solar Horizon Sensor 1 1.0 1.0 10% 1.1
Power
Power Total: 117.2 132.2
Solar Array System (Primary) 25.1 27.2
SA Substrate 2 2.0 4.0 10% 4.4
SA Cells, wiring, etc 2 5.2 10.4 10% 11.4
Hinge/Damper 2 1.0 2.0 10% 2.2
Solar Array Drive 2 2.8 5.6 5% 5.9
Solar Array Drive Electronics 1 2.0 2.0 5% 2.1
Fasteners, pins, etc 2 0.6 1.1 10% 1.2
Solar Array System, Secondary 66.8 76.7
Radiator Panel, Al HC 6 1.5 9.0 15% 10.4
Radiator Panel, Heat Pipes 1 3.5 3.5 15% 4.0

Straps, Clamps, Inserts, Thermal Adhesive, etc.


1 3.0 3.0 15% 3.5
Cooling Tubes, fittings, flex lines, Fasteners, Fill/Drain
Valves, etc. 1 8.6 8.6 15% 9.9
Pump 2 12.0 24.0 15% 27.6
Fluid 1 0.2 0.2 15% 0.2
Substrate Cooling/Mounting Deck 2 2.3 4.6 15% 5.3
Motor, Schaeffer-MOOG, Type 1 2 0.5 1.0 5% 1.0

Brackets, drive components, pins, fasteners, etc.


2 1.6 3.2 15% 3.7
Motor Drive Electronics 1 2.2 2.2 15% 2.5
Slider/Positioning System 2 0.91 1.8 15% 2.1
Substrate 2 1.0 2.0 15% 2.3
Cells/etc. 2 0.5 1.0 15% 1.2
Harness 2 0.8 1.5 15% 1.7
Center Deck Stiffener 1 1.2 1.2 15% 1.4
Power System Electronics 25.3 28.3
Power System Electronics 1 7.3 7.3 15% 8.4
Battery 1 8.0 8.0 5% 8.4
Power Distribution Unit 1 10.0 10.0 15% 11.5
Thermal Protection System
Thermal Protection System Total: 68.5 78.8
Shield 1 52.0 52.0 15% 59.8
Support Structure 1 10.0 10.0 15% 11.5
Dust Protection 1 6.5 6.5 15% 7.5
Thermal Control

A-4
APPENDIX A: SOLAR PROBE+ MASS AND POWER BUDGETS

CBE Each CBE Total Contingency Total


Name Qty (kg) (kg) (%) (kg)
Thermal Control Total: 15.7 18.1
Multi-Layer Insulation 1 10.0 10.0 15% 11.5
Radiators 1 0.3 0.3 15% 0.3
Heater/Thermistor Harness 1 2.5 2.5 15% 2.9
Diode Heat Pipe 1 0.9 0.9 15% 1.0
Doublers, Gaskets, etc 1 2.0 2.0 15% 2.3
Avionics
Avionics Total: 12.7 14.0
Integrated Electronics Module 2 6.0 12.0 10% 13.2
TRIO Units 14 0.1 0.7 10% 0.8
Propulsion
Propulsion Total: 20.5 21.9
Hydrazine Tank 1 6.4 6.4 5% 6.7
5-lbf Thruster 2 0.7 1.5 5% 1.5
0.2-lbf Thruster 12 0.3 3.6 5% 3.8
Latch Valve 2 0.3 0.6 5% 0.6
Filter 1 0.0 0.0 5% 0.0
Fill/Service Valve 2 0.2 0.3 5% 0.3
Pressure Transducers 2 0.2 0.4 5% 0.4
Tubing/Fasteners/Clamps 1 3.0 3.0 10% 3.3
Electrical Connectors 1 0.3 0.3 10% 0.3
Cabling 1 3.0 3.0 10% 3.3
Orifice 1 0.0 0.0 10% 0.0
Propulsion Diode Box 1 1.5 1.5 10% 1.7
Mechanical
Mechanical Total: 58.9 67.7
Primary Structure = 53.4 61.5
Top Deck 1 6.5 6.5 15% 7.5
Aft Deck 1 14.7 14.7 15% 16.9
Side Panels 6 3.6 21.6 15% 24.8
Fasteners 1 1.8 1.8 15% 2.1

Payload Adaptor, Umbilical Brackets, etc.) 1


8.1 8.1 15% 9.3
Tank Mounting Structure 1 0.8 0.8 15% 0.9
Secondary Structure = 5.5 6.3
Reaction Wheel Brackets 4 0.3 1.2 15% 1.4
Low Gain Antenna Bracket 2 0.2 0.4 15% 0.5
Medium Gain Antenna Bracket 1 0.3 0.3 15% 0.3
Star Tracker Bracket 3 0.5 1.5 15% 1.7
Solar Array Tie Down Brackets (All) 8 0.2 1.6 15% 1.8

A-5
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

CBE Each CBE Total Contingency Total


Name Qty (kg) (kg) (%) (kg)
Balance Mass 0 8.0 0.0 15% 0.0
Purge Components 1 0.5 0.5 15% 0.5
Harness
Harness Total: 18.0 20.6
Main Harness 1 16.5 16.5 15% 19.0
Plugs 1 1.0 1.0 10% 1.1
Grounding Straps 1 0.5 0.5 10% 0.6
Observatory
Observatory Dry Mass 428.3 480.3
Propellant
Propellant Total: 52.7
Useable 52.3
Residual 0.3
Pressurant 0.1
Observatory Wet Mass 533.0
Launch Mass 610.0
Unallocated Margin 77.0
Total Mass Reserves 30.1%

A-6
APPENDIX A: SOLAR PROBE+ MASS AND POWER BUDGETS

Table A-2. Solar Probe+ power budget.


Science
Momentum
Post-Separation Maneuver Cruise Checkout/Calibration Approach Science Dump
CBE CBE Duty Duty Duty Duty Duty Duty
Each Total Cycle Total Cycle Total Cycle Total Duty Cycle Total Cycle Total Cycle Total
Name Qty (W) (W) (%) (W) (%) (W) (%) (W) Cycle (%) Total (W) (%) (W) (%) (W) (%) (W)
Instruments
Instruments Total: 0.0 0.0 0.0 28.6 0.0 57.2 57.2
Hemispheric Imager 1 4.0 4.0 0% 0.0 0% 0.0 0% 0.0 50% 2.0 0% 0.0 100% 4.0 100% 4.0
Polar Source Region Imager 1 4.0 4.0 0% 0.0 0% 0.0 0% 0.0 50% 2.0 0% 0.0 100% 4.0 100% 4.0
Fast Ion Analyzer 1 3.7 3.7 0% 0.0 0% 0.0 0% 0.0 50% 1.9 0% 0.0 100% 3.7 100% 3.7
Fast Electron Analyzer 1 7.2 7.2 0% 0.0 0% 0.0 0% 0.0 50% 3.6 0% 0.0 100% 7.2 100% 7.2
Ion Composition Analyzer 1 6.0 6.0 0% 0.0 0% 0.0 0% 0.0 50% 3.0 0% 0.0 100% 6.0 100% 6.0
Plasma Wave Electronics 1 5.0 5.0 0% 0.0 0% 0.0 0% 0.0 50% 2.5 0% 0.0 100% 5.0 100% 5.0
Magnetometer 1 2.5 2.5 0% 0.0 0% 0.0 0% 0.0 50% 1.3 0% 0.0 100% 2.5 100% 2.5
Energetic Particle Inst. High Energy 1 2.3 2.3 0% 0.0 0% 0.0 0% 0.0 50% 1.2 0% 0.0 100% 2.3 100% 2.3
Energetic Particle Inst. Low Energy 1 1.7 1.7 0% 0.0 0% 0.0 0% 0.0 50% 0.9 0% 0.0 100% 1.7 100% 1.7
Neutron-Gamma Spectrometer 1 3.0 3.0 0% 0.0 0% 0.0 0% 0.0 50% 1.5 0% 0.0 100% 3.0 100% 3.0
Coronal Dust Detector 1 3.8 3.8 0% 0.0 0% 0.0 0% 0.0 50% 1.9 0% 0.0 100% 3.8 100% 3.8
Common DPU 1 14.0 14.0 0% 0.0 0% 0.0 0% 0.0 50% 7.0 0% 0.0 100% 14.0 100% 14.0
Accommodation Hardware
Accommodation Hardware Total: 0.0 0.0 0.0 0.3 0.0 0.3 0.3
Fast Plasma Actuator/Arm 1 0.0 0.0 0% 0.0 0% 0.0 0% 0.0 5% 0.0 0% 0.0 5% 0.0 5% 0.0
Plasma Wave Antenna, Actuator 1 5.0 5.0 0% 0.0 0% 0.0 0% 0.0 5% 0.3 0% 0.0 5% 0.3 5% 0.3
Spacecraft
Spacecraft Total: 201.6 325.8 292.8 244.8 245.1 245.1 253.3
Telecommunications
Telecommunications Total: 49.7 97.7 97.7 49.7 49.7 49.7 49.7
Ka-band TWT 2 80.0 160.0 0% 0.0 50% 80.0 50% 80.0 0% 0.0 0% 0.0 0% 0.0 0% 0.0
X-band TWTA 2 33.0 66.0 50% 33.0 0% 0.0 0% 0.0 50% 33.0 50% 33.0 50% 33.0 50% 33.0
Transponder rcv only 1 4.0 4.0 100% 4.0 100% 4.0 100% 4.0 100% 4.0 100% 4.0 100% 4.0 100% 4.0
Transponder (rcv+X only) 1 8.7 8.7 100% 8.7 0% 0.0 0% 0.0 100% 8.7 100% 8.7 100% 8.7 100% 8.7

A-7
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

Science
Momentum
Post-Separation Maneuver Cruise Checkout/Calibration Approach Science Dump
CBE CBE Duty Duty Duty Duty Duty Duty
Each Total Cycle Total Cycle Total Cycle Total Duty Cycle Total Cycle Total Cycle Total
Name Qty (W) (W) (%) (W) (%) (W) (%) (W) Cycle (%) Total (W) (%) (W) (%) (W) (%) (W)
Transponder (rcv+Ka only) 1 9.7 9.7 0% 0.0 100% 9.7 100% 9.7 0% 0.0 0% 0.0 0% 0.0 0% 0.0
Transponder (rcv+X+Ka) 1 14.1 14.1 0% 0.0 0% 0.0 0% 0.0 0% 0.0 0% 0.0 0% 0.0 0% 0.0
USO 2 2.0 4.0 100% 4.0 100% 4.0 100% 4.0 100% 4.0 100% 4.0 100% 4.0 100% 4.0
Guidance and Control
Guidance and Control Total: 95.2 95.2 95.2 95.2 95.5 95.5 95.5
Inertial Measurement Unit 1 29.5 29.5 100% 29.5 100% 29.5 100% 29.5 100% 29.5 100% 29.5 100% 29.5 100% 29.5
Reaction Wheel Assembly 4 11.0 44.0 100% 44.0 100% 44.0 100% 44.0 100% 44.0 100% 44.0 100% 44.0 100% 44.0
Star Tracker 1 21.7 21.7 100% 21.7 100% 21.7 100% 21.7 100% 21.7 100% 21.7 100% 21.7 100% 21.7
Solar Horizon Sensor 1 0.3 0.3 0% 0.0 0% 0.0 0% 0.0 0% 0.0 100% 0.3 100% 0.3 100% 0.3
Power
Power Total: 18.8 18.8 18.8 18.8 18.8 18.8 18.8
Primary Solar Array Drive 2 18 36.0 1% 0.4 1% 0.4 1% 0.4 1% 0.4 0% 0.0 0% 0.0 0% 0.0
Secondary Solar Array Drive 2 18 36.0 0% 0.0 0% 0.0 0% 0.0 0% 0.0 1% 0.4 1% 0.4 1% 0.4
Power System Electronics 1 10.0 10.0 100% 10.0 100% 10.0 100% 10.0 100% 10.0 100% 10.0 100% 10.0 100% 10.0
Power Distribution Unit 1 8.8 8.8 100% 8.8 100% 8.8 100% 8.8 100% 8.8 100% 8.8 100% 8.8 100% 8.8
Thermal Control
Thermal Control Total: 0.0 43.2 43.2 43.2 43.2 43.2 43.2
Secondary Array Pumps 2 21.6 43.2 0% 0.0 100% 43.2 100% 43.2 100% 43.2 100% 43.2 100% 43.2 100% 43.2
Heaters 1 0.0 0.0 0% 0.0 100% 0.0 100% 0.0 100% 0.0 100% 0.0 100% 0.0 100% 0.0
Avionics
Avionics Total: 35.0 35.0 35.0 35.0 35.0 35.0 35.0
Primary IEM 1 30.0 30.0 100% 30.0 100% 30.0 100% 30.0 100% 30.0 100% 30.0 100% 30.0 100% 30.0
Secondary IEM 1 5.0 5.0 100% 5.0 100% 5.0 100% 5.0 100% 5.0 100% 5.0 100% 5.0 100% 5.0
Propulsion
Propulsion Total: 2.9 35.9 2.9 2.9 2.9 2.9 11.1
Thrusters 1 33.0 33.0 0% 0.0 100% 33.0 0% 0.0 0% 0.0 0% 0.0 0% 0.0 25% 8.3
Cat Bed Heaters 1 11.5 11.5 25% 2.9 25% 2.9 25% 2.9 25% 2.9 25% 2.9 25% 2.9 25% 2.9

A-8
APPENDIX A: SOLAR PROBE+ MASS AND POWER BUDGETS

Science
Momentum
Post-Separation Maneuver Cruise Checkout/Calibration Approach Science Dump
CBE CBE Duty Duty Duty Duty Duty Duty
Each Total Cycle Total Cycle Total Cycle Total Duty Cycle Total Cycle Total Cycle Total
Name Qty (W) (W) (%) (W) (%) (W) (%) (W) Cycle (%) Total (W) (%) (W) (%) (W) (%) (W)
Pressure Transducers 1 1.8 1.8 100% 1.8 100% 1.8 100% 1.8 100% 1.8 100% 1.8 100% 1.8 100% 1.8
Observatory
Observatory Power 201.6 325.8 292.8 273.6 245.1 302.5 310.8

Radiated RF Power
TWT radiated power Ka-band 1 40 40.0 0.0 40.0 40.0 0.0 0.0 0.0 0.0
TWT radiated power X-band 1 12 12.0 12.0 0.0 0.0 12.0 12.0 12.0 12.0

Cooling pumps isolated from bus


Bus Subsystem Dissipated Power 189.6 242.6 209.6 218.4 189.9 247.3 255.6
Bus Heaters 0% 0.0 100% 27.4 100% 60.4 100% 51.6 100% 80.1 100% 22.7 100% 14.4
Harness Loss 3.0 5.3 5.3 4.9 4.9 4.9 4.9
Bus Power 192.6 275.3 275.3 274.9 274.9 274.9 274.9
Total S/C Consumed Power 204.6 358.5 358.5 330.1 330.1 330.1 330.1

Available Load Power 482.0 482.0 482.0 482.0 482.0 482.0


Power Reserves 123.5 123.5 151.9 151.9 151.9 151.9
Margin 34.45% 34.45% 46.03% 46.03% 46.03% 46.03%

A-9
APPENDIX B: SOLAR PROBE+ LINK ANALYSIS

APPENDIX B:

SOLAR PROBE+ LINK ANALYSIS

B-1
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

B-2
APPENDIX B: SOLAR PROBE+ LINK ANALYSIS

HGA 34m HEF Link Ver. 3.0 Name: DeBoy


SPACECRAFT COMMUNICATIONS LINK ANALYSIS Date: 1/6/2008

SPACECRAFT MISSION: Solar Probe Plus with Mast


LINK: X up Ka down HGA Link

Uplink Freq: 7.182 GHz Wavlgth: 0.0418 meter


Downlink Freq: 32 GHz Wavlgth: 0.0094 meter
Uplink Command Modulation: NRZ/PSK/PM (Data on a 16 kHz sine wave subcarrier )
Downlink Telemetry Modulation: NRZ/PSK/PM (Data on a 25.5 kHz square wave subcarrier )
Ranging Modulation: Off

HIGH GAIN ANTENNA UTILITY


Spacecraft HGA Diameter: 0.8 m Efficiency: 60.0 %
Spacecraft HGA Pointing Error (+/-): 0.20 Deg.

Calculated Parameter Uplink Downlink


Spacecraft HGA Gain 33.37 dBic 46.35 dBic
Spacecraft HGA 3dB Beamwidth: 3.44 Deg. 0.77 Deg.
Spacecraft HGA Pointing Loss: -0.04 dB -0.81 dB

LEO SPACECRAFT SLANT RANGE UTILITY


Parameter Value Units
Spacecraft Altitude (Enter Re or km for units): n/a km
Ground Antenna Elevation Angle: n/a Deg.
Slant Range (Assumes average Re= 6370 km): n/a km

Value Units
Spacecraft range (enter AU, Re or km for units): 1.8 Au

UPLINK MODULATION CHARACTERISTICS


Command data rate: 10 kbps
Command modulation index: 1.3 rad pk
Uplink ranging modulation index: 0.8 rad pk
RANGING CHANNEL CHARACTERISTICS
Ranging channel turn-around (elbow/regenerative): elbow
DOWNLINK MODULATION CHARACTERISTICS
Entered Parameters:
Telemetry data rate: info. Rate 72 kbps
Telemetry modulation index (enter BPSK, QPSK, or a value for PM): QPSK rad pk
Downlink ranging channel modulation index: 0.00001 rad pk
Downlink telemetry waveform (square/sine): square
Calculated Parameters:
S/C ranging channel bandwidth: Br= 1.5 MHz
Pwr in fundamental component of square ranging tone: gamma= 0.8106
Ranging SNR at output of S/C receiver filter: alphar= 0.6108
Cmd/noise ratio at output of S/C receiver filter: alphac= 1.0075
Effective downlink cmd modulation index: t1= 0.0000 rad. pk
Effective downlink ranging modulation index: t2= 0.0000 rad. pk
Effective downlink noise modulation index: t3= 0.0000 rad. rms

Figure B-RF1. X-band/Ka-band link analysis (1 of 6): High-gain antenna (HGA) link parameters.

B-3
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

UPLINK CALCULATIONS 1/6/2008

Parameter Notes Units Des.Value Adv. Tol. Fav. Tol. Mean Val. Variance PDF
Gnd Station TX Power: 20 kW Mid-band dBm 73.01 -1.00 0.00 72.68 0.06 tri
Gnd Station TX Antenna Gain20 deg, vacuum, mod103 p22 dBic 67.06 -0.20 0.20 67.06 0.01 uni
Gnd Antenna Pointing Loss: Con. Scan 30 mph wind, mod1dB -0.40 0.00 0.30 -0.25 0.01 uni
Gnd Station Passive Loss: 810-005 mod103 p8 dB -0.25 -0.05 0.05 -0.25 0.00 uni
EIRP: dBm 139.42 139.24 0.08

Uplink Path Loss: dB -278.17 0.00 0.00 -278.17 0.00 uni


S/C Antenna Gain: HGA dBic 33.37 -1.00 1.00 33.37 0.33 uni
S/C Antenna Pointing Loss: included above dB -0.04 0.00 0.04 -0.03 0.00 tri
Atmospheric Loss: 90% wthr, 20 deg, mod103 p2 dB -0.17 0.00 0.14 -0.10 0.00 uni
Polarization Mismatch Loss: Included in S/C antenna gain dB 0.00 0.00 0.00 0.00 0.00 tri
S/C Passive Loss: (Between antenna port and rcvr input) dB -0.70 0.00 0.20 -0.60 0.00 gau
Total Received Power: (At receiver input) dBm -106.29 -106.29 0.41

S/C Antenna Noise Temp. (At antenna port) RSB estimK 100.00
S/C Passive Loss Noise Temp. K 50.72
S/C Receiver Noise Figure: LNA w/ filter loss dB 2.00
System Noise Temp: (At receiver input) K 297.90
System Noise Density: (At receiver input) dBm/Hz -173.86 0.90 -1.00 -173.91 0.10 gau

Carrier/Total Power: dB -7.29 -0.25 0.25 -7.29 0.01 tri


Received Carrier Power: dBm -113.58 -113.58 0.42
Received Pc/No: dB-Hz 60.28 60.33 0.52
Tracking Loop Predetection Noise BW: 3 kHz dBHz 34.77 0.40 -0.40 34.77 0.03 tri
Tracking Loop Predetection SNR: dB 25.51 25.56 0.55
Carrier Tracking Loop Bandwidth (BL): 50 Hz dBHz 16.99 1.40 -1.50 16.96 0.35 tri
Received Carrier/Noise in Loop Bandwidth: dB 43.29 43.37 0.87
Required Carrier/Noise in Loop Bandwidth: dB 10.00 0.00 0.00 10.00 0.00 uni
Carrier Margin: dB 33.29 33.37 0.87
3sigma= dB 2.81

Command Subc/Total Power: dB -5.77 -0.25 0.25 -5.77 0.01 tri


Received Command Subc. Power: dBm -112.07 -112.07 0.42
Subcarrier Demod. Predetection Noise B 3.84 kHz dBHz 35.84 0.40 -0.40 35.84 0.03 tri
Subcarrier Demod. Predetection SNR: dB 25.95 26.00 0.55
Command Data Rate: dBHz 40.00 0.00 0.00 40.00 0.00 tri
Received Command Eb/No: dB 21.79 21.84 0.52
Required Command Eb/No: Pe=1.0E-06, PSK Yuen p205 dB 10.50 0.00 0.00 10.50 0.00 uni
Implementation Loss: Card spec. dB -1.50 -0.50 0.50 -1.50 0.04 tri
Command Margin: dB 9.79 9.84 0.57
3sigma= dB 2.26

Ranging/Total Power: dB -7.04 -0.25 0.25 -7.04 0.01 tri


Received Ranging Power: dBm -113.33 -113.33 0.42
Uplink Pr/No: dBHz 60.53 60.58 0.52
S/C Ranging Channel Bandwidth: 1.5 MHz dBHz 61.76 0.40 -0.40 61.76 0.03 tri
S/C Ranging Channel SNR: dB -1.23 -1.18 0.55

Figure B-RF1 (cont.). X-band/Ka-band link analysis (2 of 6): X-band HGA uplink to 34 m at 1.8 AU.

B-4
APPENDIX B: SOLAR PROBE+ LINK ANALYSIS

DOWNLINK CALCULATIONS 1/6/2008

Parameter Notes Units Des.Value Adv. Tol. Fav. Tol. Mean Val. Variance PDF
S/C Transmitter Power: 40 watts dBm 46.02 -0.50 0.50 46.02 0.04 tri
S/C Passive Loss: Estimate dB -2.00 0.00 0.20 -1.90 0.00 gau
S/C Antenna Gain: 0.8m HGA dBic 46.35 -0.50 0.50 46.35 0.04 tri
S/C Antenna Pointing Loss: 0.2° Pointing Error dB -0.81 0.00 0.81 -0.54 0.04 tri
EIRP: dBm 89.56 89.93 0.12

Path Loss: dB -291.15 0.00 0.00 -291.15 0.00 uni


Atmospheric Loss: 90% Weather 20 deg ElevationdB -1.00 0.00 0.14 -0.93 0.00 uni
Polarization Mismatch Loss: dB -0.10 0.00 0.00 -0.10 0.00 uni
Gnd Antenna Gain: 20 deg EL Angle dBic 77.50 -0.20 0.20 77.50 0.01 uni
Gnd Antenna Pointing Loss: 20 MPH Wind + 90% dB -1.30 0.00 0.30 -1.15 0.01 uni
Total Received Power: dBm -126.49 -125.90 0.14
G/T 57.96
Gnd Antenna Noise Temp. 90% wthr, 20 deg Elevation K 90.00
Solar/Planetary Noise: K 0.00
Gnd System Noise Temp: K 90.00
Gnd System Noise Density: dBm/Hz -179.06 0.25 -0.25 -179.06 0.01 gau

Carrier/Total Power: dB N/A 0.00 0.00 N/A N/A tri


Received Carrier Power: dBm N/A N/A N/A
Received Pc/No: dB-Hz N/A N/A N/A
Tracking Loop Predetection Noise BW: 1 kHz dBHz 30.00 0.40 -0.40 30.00 0.03 tri
Tracking Loop Predetection SNR: dB 22.57 23.16 0.18
Tracking Loop Noise Bandwidth (BL): 1 Hz dBHz 0.00 0.00 0.00 0.00 0.00 tri
Squaring/Quadrupling Loss: dB -12.25 -12.23 0.00
Received Signal/Noise in Loop Bandwidth: dB 40.32 40.93 0.15
Required Signal/Noise in Loop Bandwidth: dB 10.00 0.00 0.00 10.00 0.00 uni
Carrier Tracking Margin: dB 30.32 30.93 0.15
3sigma= dB 1.16

Tlm/Total Power: dB 0.00 0.00 0.00 0.00 0.00 tri


Received Tlm Power: dBm -126.49 -125.90 0.14
Tlm Data Rate: dBHz 48.57 0.00 0.00 48.57 0.00 uni
Received Eb/No: dB 4.00 4.58 0.15
Required Eb/No: Turbo Rate 1/2, 8920 frame baseline dB 1.00 0.20 -0.20 1.00 0.01 uni
Implementation Loss: dB -1.00 0.00 0.00 -1.00 0.00 tri
Other gain/loss: dB 0.00 -0.25 0.25 0.00 0.01 tri
Telemetry Margin: dB 2.00 2.58 0.17
3sigma= dB 1.25
SSNR -3.79
Rng/Total Power: dB n/a 0.00 0.00 n/a 0.00 tri
Received Ranging Power: dBm n/a n/a 0.14
Downlink Received Pr/No: dBHz n/a n/a 0.15
Tandem Pr/No (uplink and downlink): dBHz n/a n/a n/a
Downlink Required Pr/No: dBHz -10.00 0.00 0.00 -10.00 0.00 uni
Ranging Demodulator Loss: dB -0.50 0.00 0.00 -0.50 0.00 tri
Ranging Margin: dB n/a n/a n/a
3sigma= dB n/a

Figure B-RF1 (cont.). X-band/Ka-band link analysis (3 of 6): Ka-band HGA downlink to 34 m at 1.8 AU.

B-5
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

DOWNLINK CALCULATIONS 1/25/2008

Parameter Notes Units Des.Value Adv. Tol. Fav. Tol. Mean Val. Variance PDF
S/C Transmitter Power: 13 watts dBm 41.14 -0.50 0.50 41.14 0.04 tri
S/C Passive Loss: Estimate dB -3.00 0.00 0.20 -2.90 0.00 gau
S/C Antenna Gain: 0.8m HGA dBic 34.78 -0.50 0.50 34.78 0.04 tri
S/C Antenna Pointing Loss: 0.2° Pointing Error dB -0.06 0.00 0.06 -0.04 0.00 tri
EIRP: dBm 72.86 72.98 0.08

Path Loss: dB -279.58 0.00 0.00 -279.58 0.00 uni


Atmospheric Loss: 10 deg EL, 90% Weather dB -0.31 0.00 0.14 -0.24 0.00 uni
Polarization Mismatch Loss: dB -0.10 0.00 0.00 -0.10 0.00 uni
Gnd Antenna Gain: 10 deg EL Angle dBic 68.23 -0.20 0.20 68.23 0.01 uni
Gnd Antenna Pointing Loss: 90% weather dB -0.10 0.00 0.30 0.05 0.01 uni
Total Received Power: dBm -139.00 -138.66 0.11
G/T 52.21
Gnd Antenna Noise Temp. 90% Weather, 10 deg EL K 40.00
Solar/Planetary Noise: K 0.00
Gnd System Noise Temp: K 40.00
Gnd System Noise Density: dBm/Hz -182.58 0.25 -0.25 -182.58 0.01 gau

Carrier/Total Power: dB N/A 0.00 0.00 N/A N/A tri


Received Carrier Power: dBm N/A N/A N/A
Received Pc/No: dB-Hz N/A N/A N/A
Tracking Loop Predetection Noise BW: 1 kHz dBHz 30.00 0.40 -0.40 30.00 0.03 tri
Tracking Loop Predetection SNR: dB 13.58 13.92 0.14
Tracking Loop Noise Bandwidth (BL): 1 Hz dBHz 0.00 0.00 0.00 0.00 0.00 tri
Squaring/Quadrupling Loss: dB -13.52 -13.42 0.00
Received Signal/Noise in Loop Bandwidth: dB 30.06 30.50 0.11
Required Signal/Noise in Loop Bandwidth: dB 10.00 0.00 0.00 10.00 0.00 uni
Carrier Tracking Margin: dB 20.06 20.50 0.11
3sigma= dB 1.01

Tlm/Total Power: dB 0.00 0.00 0.00 0.00 0.00 tri


Received Tlm Power: dBm -139.00 -138.66 0.11
Tlm Data Rate: dBHz 38.45 0.00 0.00 38.45 0.00 uni
Received Eb/No: dB 5.13 5.47 0.11
Required Eb/No: Turbo Rate 1/2, 8920 frame baseline dB 1.00 0.20 -0.20 1.00 0.01 uni
Implementation Loss: dB -1.00 0.00 0.00 -1.00 0.00 tri
Other gain/loss: dB 0.00 -0.25 0.25 0.00 0.01 tri
Telemetry Margin: dB 3.13 3.47 0.14
3sigma= dB 1.11
SSNR -2.65
Rng/Total Power: dB n/a 0.00 0.00 n/a 0.00 tri
Received Ranging Power: dBm n/a n/a 0.11
Downlink Received Pr/No: dBHz n/a n/a 0.11
Tandem Pr/No (uplink and downlink): dBHz n/a n/a n/a
Downlink Required Pr/No: dBHz -10.00 0.00 0.00 -10.00 0.00 uni
Ranging Demodulator Loss: dB -0.50 0.00 0.00 -0.50 0.00 tri
Ranging Margin: dB n/a n/a n/a
3sigma= dB n/a

Figure B-RF1 (cont.). X-band/Ka-band link analysis (4 of 6): X-band HGA downlink to 34 m at 1.8 AU.

B-6
APPENDIX B: SOLAR PROBE+ LINK ANALYSIS

UPLINK CALCULATIONS 1/25/2008

Parameter Notes Units Des.Value Adv. Tol. Fav. Tol. Mean Val. Variance PDF
Gnd Station TX Power: 20 kW Mid-band dBm 73.01 -1.00 0.00 72.68 0.06 tri
Gnd Station TX Antenna Gain20 deg, vacuum, mod103 p22 dBic 72.62 -0.20 0.20 72.62 0.01 uni
Gnd Antenna Pointing Loss: Con. Scan 30 mph wind, mod1dB -0.40 0.00 0.30 -0.25 0.01 uni
Gnd Station Passive Loss: 810-005 mod103 p8 dB -0.45 -0.05 0.05 -0.45 0.00 uni
EIRP: dBm 144.78 144.60 0.08

Uplink Path Loss: dB -278.89 0.00 0.00 -278.89 0.00 uni


S/C Antenna Gain: Canted LGA dBic 0.00 -1.00 1.00 0.00 0.33 uni
S/C Antenna Pointing Loss: dB 0.00 0.00 0.00 0.00 0.00 tri
Atmospheric Loss: 90% wthr, 20 deg, mod103 p2 dB -0.17 0.00 0.14 -0.10 0.00 uni
Polarization Mismatch Loss: Included in S/C antenna gain dB 0.00 0.00 0.00 0.00 0.00 tri
S/C Passive Loss: (Between antenna port and rcvr input) dB -3.00 0.00 0.20 -2.90 0.00 gau
Total Received Power: (At receiver input) dBm -137.28 -137.29 0.41

S/C Antenna Noise Temp. (At antenna port) K 100.00


S/C Passive Loss Noise Temp. K 288.63
S/C Receiver Noise Figure: LNA w/ filter loss dB 2.00
System Noise Temp: (At receiver input) K 364.39
System Noise Density: (At receiver input) dBm/Hz -172.99 0.90 -1.00 -173.04 0.10 gau

Carrier/Total Power: dB -1.45 -0.25 0.25 -1.45 0.01 tri


Received Carrier Power: dBm -138.73 -138.74 0.42
Received Pc/No: dB-Hz 34.26 34.30 0.52
Tracking Loop Predetection Noise BW: 3 kHz dBHz 34.77 0.40 -0.40 34.77 0.03 tri
Tracking Loop Predetection SNR: dB -0.51 -0.48 0.55
Carrier Tracking Loop Bandwidth (BL): 50 Hz dBHz 16.99 1.40 -1.50 16.96 0.35 tri
Received Carrier/Noise in Loop Bandwidth: dB 17.27 17.34 0.87
Required Carrier/Noise in Loop Bandwidth: dB 10.00 0.00 0.00 10.00 0.00 uni
Carrier Margin: dB 7.27 7.34 0.87
3sigma= dB 2.81

Command Subc/Total Power: dB -5.65 -0.25 0.25 -5.65 0.01 tri


Received Command Subc. Power: dBm -142.93 -142.94 0.42
Subcarrier Demod. Predetection Noise B 3.84 kHz dBHz 35.84 0.40 -0.40 35.84 0.03 tri
Subcarrier Demod. Predetection SNR: dB -5.79 -5.75 0.55
Command Data Rate: 31 bps dBHz 14.91 0.00 0.00 14.91 0.00 tri
Received Command Eb/No: dB 15.14 15.18 0.52
Required Command Eb/No: Pe=1.0E-06, uncoded dB 10.60 0.00 0.00 10.60 0.00 uni
Implementation Loss: dB -1.50 -0.50 0.50 -1.50 0.04 tri
Command Margin: dB 3.04 3.08 0.57
3sigma= dB 2.26

Ranging/Total Power: dB -121.45 -0.25 0.25 -121.45 0.01 tri


Received Ranging Power: dBm -258.73 -258.74 0.42
Uplink Pr/No: dBHz -85.74 -85.70 0.52
S/C Ranging Channel Bandwidth: 1.5 MHz dBHz 61.76 0.40 -0.40 61.76 0.03 tri
S/C Ranging Channel SNR: dB -147.50 -147.47 0.55

Figure B-RF1 (cont.). X-band/Ka-band link analysis (5 of 6): X-band low-gain antenna (LGA) emergency
uplink from Deep Space Network (DSN) 70 m, 1.8 AU.

B-7
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

DOWNLINK CALCULATIONS 1/25/2008

Parameter Notes Units Des.Value Adv. Tol. Fav. Tol. Mean Val. Variance PDF
S/C Transmitter Power: 13 watts dBm 41.14 -0.50 0.50 41.14 0.04 tri
S/C Passive Loss: Estimated dB -2.00 0.00 0.20 -1.90 0.00 gau
S/C Antenna Gain: Canted LGA dBic 0.00 -0.50 0.50 0.00 0.04 tri
S/C Antenna Pointing Loss: dB 0.00 0.00 0.00 0.00 0.00 tri
EIRP: dBm 39.14 39.24 0.08

Path Loss: dB -278.00 0.00 0.00 -278.00 0.00 uni


Atmospheric Loss: 90% wthr, 20 deg, mod103 p2 dB -0.17 0.00 0.14 -0.10 0.00 uni
Polarization Mismatch Loss: dB -0.10 0.00 0.00 -0.10 0.00 uni
Gnd Antenna Gain: 20 deg, vacuum, mod103 p22 dBic 73.92 -0.20 0.20 73.92 0.01 uni
Gnd Antenna Pointing Loss: Con. Scan 30 mph wind, mod1dB -0.40 0.00 0.30 -0.25 0.01 uni
Total Received Power: dBm -165.61 -165.29 0.11
G/T 58.13
Gnd Antenna Noise Temp. 90% weather, 20 deg Elevatio K 37.94
Solar/Planetary Noise: K 0.00
Gnd System Noise Temp: K 37.94
Gnd System Noise Density: dBm/Hz -182.81 0.25 -0.25 -182.81 0.01 gau

Carrier/Total Power: dB -5.35 0.00 0.00 -5.35 0.00 tri


Received Carrier Power: dBm -170.96 -170.64 0.11
Received Pc/No: dB-Hz 11.85 12.17 0.11
Tracking Loop Predetection Noise BW: 1 kHz dBHz 30.00 0.40 -0.40 30.00 0.03 tri
Tracking Loop Predetection SNR: dB -18.15 -17.83 0.14
Tracking Loop Noise Bandwidth (BL): 0.1 Hz dBHz -10.00 0.00 0.00 -10.00 0.00 tri
Squaring/Quadrupling Loss: dB 0.00 0.00 0.00
Received Signal/Noise in Loop Bandwidth: dB 21.85 22.17 0.11
Required Signal/Noise in Loop Bandwidth: 810-5 dB 16.00 0.00 0.00 16.00 0.00 uni
Carrier Tracking Margin: dB 5.85 6.17 0.11
3sigma= dB 1.01

Tlm/Total Power: dB -1.50 0.00 0.00 -1.50 0.00 tri


Received Tlm Power: dBm -167.11 -166.79 0.11
Tlm Data Rate: 10 bps dBHz 10.00 0.00 0.00 10.00 0.00 uni
Received Eb/No: dB 5.70 6.02 0.11
Required Eb/No: Turbo code (R=1/6, 1784 frame) dB 1.00 0.20 -0.20 1.00 0.01 uni
Implementation Loss: dB -1.50 0.00 0.00 -1.50 0.00 tri
Other gain/loss: dB 0.00 -0.25 0.25 0.00 0.01 tri
Telemetry Margin: dB 3.20 3.52 0.14
3sigma= dB 1.11
SSNR -2.08
Rng/Total Power: dB -275.19 0.00 0.00 -275.19 0.00 tri
Received Ranging Power: dBm -440.80 -440.48 0.11
Downlink Received Pr/No: dBHz -257.99 -257.67 0.11
Tandem Pr/No (uplink and downlink): dBHz -257.99 -257.67 0.11
Downlink Required Pr/No: dBHz -10.00 0.00 0.00 -10.00 0.00 uni
Ranging Demodulator Loss: dB -0.50 0.00 0.00 -0.50 0.00 tri
Ranging Margin: dB -248.49 -248.17 0.11
3sigma= dB 1.01

Figure B-RF1 (cont.). X-band/Ka-band link analysis (6 of 6): X-band LGA emergency downlink from DSN
70 m, 1.5 AU.

B-8
APPENDIX C: SOLAR ENCOUNTER POWER GENERATION TRADE STUDY

APPENDIX C:

SOLAR ENCOUNTER POWER GENERATION TRADE STUDY

C-1
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

C-2
APPENDIX C: SOLAR ENCOUNTER POWER GENERATION TRADE STUDY

C.1 Trade Study Definition lar encounter increases as the perihelion is


raised from 4 RS to 9.5 RS and (ii) solar flux at
In the 2005 Solar Probe study, 1 the primary
perihelion is decreased for higher perihelion.
power source was a set of radioisotope thermal
The time spent in the solar encounter is
generators (RTGs) using plutonium as the heat
slightly different for the orbit options, ~8 days
source for thermoelectric power generation.
for orbits with perihelion at 4 RS and ~10 days
Given the current concerns over the supply of
for orbits with perihelion at 9.5 RS. The major
plutonium available for space missions, the So-
impact of this change is in sizing of options
lar Probe+ team was instructed to develop a
relying on energy storage. The magnitude of
mission using only non-nuclear power sources.
the maximum solar flux, however, is six times
The bulk of the Solar Probe+ orbit can be
greater for a 4-RS perihelion than for perihe-
powered with solar arrays. These arrays are
lion at 9.5 RS. The major impact of this differ-
sized to provide full power at aphelion and, as
ence is to disallow the use at 4 RS of photovol-
the spacecraft approaches the Sun, the arrays
taics and related concepts using mirrors to re-
are off-pointed to maintain temperature while
direct light onto solar arrays because of the
providing sufficient power including margin.
inability to prevent excessive thermal load on
Using MESSENGER heritage arrays, Solar
the bus from items directly exposed to the so-
Probe+ can be powered in this manner to
lar flux. In addition to these concerns, the so-
~0.25 AU. The heritage arrays cannot be used
lar flux time profile is similar for all orbits,
through the solar encounter where the solar
peaking at perihelion ±1.5 days, as shown in
distance at perihelion is 0.044 AU because of
Figure C-1. This environment determines the
hypersensitivity of power and temperature
times when heat conversion systems can be
versus off-pointing angle and solar panel edge
used to generate electrical power during the
effects at grazing incidence angles. During the
solar encounter and generally means that all
solar encounter, some other means of storing
heat conversion concepts require the use of
and/or generating power must be used. The
energy storage to power the spacecraft in at
major activity of the Solar Probe+ study has
least some portion of the solar encounter.
been to determine the optimal power genera-
The requirements for the secondary power
tion concept for the mission. In the subsequent
system are as follows:
discussion, the solar arrays used outside 0.25
• Provide sufficient power to meet science
AU are called the primary power generation
measurement and spacecraft operations re-
system, and the power generation method used
quirements through the entire solar encoun-
inside 0.25 AU is called the secondary power
ter as constrained by the solar flux time pro-
system. The segment of the orbit when the
file. At the time the trade study was con-
secondary power system is used is called the
ducted, the total power required was esti-
solar encounter.
mated to be 350 W.
The choice of secondary power generation
• Operate through repeated exposure to the
is largely independent of the orbit trade study
maximum solar flux at perihelion, for which
discussed in Section 4.1 except for the choice
3000 Suns was assumed for orbits with 4-RS
of perihelion. The choice of perihelion affects
perihelion, and 512 Suns for orbits with 9.5-
the trade in two ways: (i) the length of the so-
RS perihelion.
1
• Limit heat load into the spacecraft bus to
Solar Probe: Report of the Science and Technology
Definition Team, NASA/TM—2005–212786, National
prevent temperature rises within the bus
Aeronautics and Space Administration, Goddard Space structure itself. The target for the trade
Flight Center, Greenbelt, MD (2005).

C-3
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

generation, such as
electric power gen-
eration by collecting
from the solar elec-
tromagnetic field
with a tether extend-
ing from the space-
craft. These concepts
were eliminated from
the trade study as
impractical and of
very low develop-
ment status with de-
velopment schedules
incompatible with a
2015 launch.
C.2.1. Battery
Figure C-1. Temperature profile for the front face of the Thermal Protection Sys- Bank
tem (TPS) during a 4-RS perihelion encounter. Primary arrays
are used to charge a
study was to limit the heat load to 50 W, the
battery bank prior to solar encounter, which is
heat load of the RTGs in the 2005 study.
sized to provide sufficient energy for entire
• Limit mass to no more than the mass used by
encounter. Multiple advanced battery types
the RTGs in the 2005 study (120 kg), with a
were investigated, as well as capacitor-based
goal of reducing mass as much as feasible energy storage. Based on development status of
while meeting all other requirements. battery technology and suitability for use in the
In addition to the performance requirements Solar Probe+ environment, the engineering
above, the secondary power system should be team chose a 2000-ampere-hour bank of lith-
of development status compatible with a 2015 ium-ion batteries, assuming 80% depth of dis-
launch. Although we did not exclude low- charge, and a total encounter of 10 days with
technology readiness level (TRL) solutions, no use of solar arrays. The mass for this bank
the goal was to use a system that could be de- of batteries is ~800 kg. By using advanced bat-
signed, manufactured, and qualified in the de- tery technology and strict energy management
velopment cycle of the Solar Probe+ mission. during the encounter, the battery mass could be
All power system options included in the trade reduced to slightly more than 400 kg. Signifi-
study required development at the system cant battery development investment will be
level but were designed using major compo- required, and significant risk is incurred if this
nents with flight heritage or significant recent option is chosen.
investment in development for space flight. C.2.2. Thermocouple-Based Energy
C.2 Power System Options Conversion
Commercially available thermocouples are
In general, three types of system concepts
embedded in the Thermal Protection System
were chosen for the trade study: conversion of
(TPS) such that the temperature difference
solar heating to electric power, conversion of
between the front surface of the TPS and
light to electric power, and energy storage. Con-
rear of the thermocouple is used to generate
sideration was given to other methods of power
electric current. Thermocouples are semicon-

C-4
APPENDIX C: SOLAR ENCOUNTER POWER GENERATION TRADE STUDY

couples themselves is quite low (~10 kg), the


support structure needed to make the concept
work is significant. Assuming strict power
management when running on batteries, the
ability to recharge batteries during closest ap-
proach and radiator size sufficient to hold the
cold leg of the thermocouples to the required
temperature, the mass of a thermocouple-based
system to power Solar Probe+ is ~210 kg.
C.2.3. Stirling Engine Power Generation
Heat from the TPS front face is used to
drive a set of Stirling engines such that the
Figure C-2. Design of a typical thermocouple de- temperature difference between the hot and
vice. cold ends operates a mechanical electric gen-
erator. A Stirling engine uses the temperature
ductor devices typically used in waste heat difference between its hot end and cold end to
recovery (e.g., in jet engines), which are the establish a cycle of gas expansion and contrac-
energy conversion element used in radioiso- tion within the engine, converting thermal en-
tope thermal generators (RTGs) in spaceflight ergy into mechanical, and subsequently elec-
missions. Figure C-2 shows a typical thermo- tric, power. The Advanced Stirling Converter
couple design; this device is ~3-mm long and is a nominal 35-W generator developed for the
6-mm across and has a mass of 20 g. The typi- Advanced Stirling Radioisotope Generator
cal application for space flight is to embed the program by Glenn Research Center. Figure C-3
thermocouple such that the large collector is shows a schematic of a pair of engines
driven to hot temperature by some exterior mounted in the configuration chosen for Solar
source (the top surface of the heat shield in the Probe+. Two engines are driven by a common
case of Solar Probe+) while the cold legs are hot collector. In the case of Solar Probe+, this
held at much lower temperature through con- plate would be embedded in the TPS. The cold
nection to radiators. Conversion efficiency ends of these devices are controlled sepa-
depends on the difference in temperature and rately; the Solar Probe+ concept uses heat
on the temperature of the hot plate. At the So- pipes and radiators to maintain the cold end
lar Probe+ perihelion, efficiency is ~6%, and temperature of the engines.
decreases to match the thermal profile shown
earlier in Figure C-1. Thermocouples will
generate power only around perihelion ± 2
days for closest-approach orbits and may not
generate power at all for early orbits in some
mission design options with perihelion above
15 RS or so. Some form of energy storage
must be used in conjunction with thermoelec-
tric generation.
The thermocouple-based design concept
used ~500 devices embedded in the TPS
around the circumference with attachment to
radiators mounted on the support structure be-
low the TPS. While the mass of the thermo- Figure C-3. Stirling engine module adopted for the
Solar Probe+ concept.

C-5
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

C.2.4. Photovoltaic-Based Energy


Conversion
Solar cells designed for high-intensity
illumination mounted on temperature-
controlled solar array substrates. Direct illu-
mination of solar cells will not meet the power
system requirements given in Section C.1
without the use of some form of temperature
control. Two types of photovoltaic concepts
that perform this control were examined in
detail for this trade study: (i) reduction of il-
lumination impinging on cells and (ii) active
temperature control of solar cell base plate. In
the first case, the energy flux impinging on a
set of solar cells is reduced by diffusing the
Figure C-4. Solar Probe+ concept using Stirling solar input with slots or grids of holes in a
engine modules.
shield protecting the arrays of solar cells or by
The Stirling power generation concept con- using a reflector to illuminate cells with a re-
sist of six pairs of engine modules mounted duced flux. In the second case, the cells are
under the TPS such that the collectors are em- mounted on a base plate held at constant tem-
bedded in the TPS in the region where the tem- perature using an active cooling loop. In prac-
perature of the collector will be near 650°C, at tice, a photovoltaic-based system only works
which the Stirling engines are highly efficient. by using a combination of these two methods,
The radiator for each Stirling engine pair is and only for perihelia above 8.5–9 RS. A large
mounted on the outside of the TPS support number of concepts using photovoltaics were
structure near the engine module. Figure C-4 proposed and discarded as untenable because
shows the system configuration for Solar of the physical limitations of the materials,
Probe+, with two radiator panels hidden for overly difficult to manufacture (such as a grid
clarity. As with the thermocouples, the Stirling of micrometer-sized holes through the TPS
engine concept depends on a large temperature used to limit solar intensity on a solar array),
difference between hot and cold ends, which or overly burdensome on the spacecraft bus
limits ability of the Stirling engine system to (by adding extra heat leaks through the TPS to
provide sufficient power during the entire solar the bus, for instance).
encounter and requires the use of energy stor- An achievable photovoltaic concept
age as a supplementary system when the tem- emerged from preliminary engineering con-
perature of the TPS front face, as shown previ- siderations. This concept uses a knife edge on
ously in Figure C-1, is too low to support the TPS to partially shadow an array of solar
power generation with the Stirling engines. cells designed for high-intensity illumination.
Assuming strict power management when These cells use similar gridlines and contact
running on batteries, the ability to fully re- metallization as used for concentrator photo-
charge batteries during closest approach, and voltaic cells, which have been used for terres-
radiator size sufficient to maintain the cold trial applications with optics having a very
end of the Stirling engine to the required tem- high concentration ratio. These cells are
perature, the mass of the Stirling engine sys- mounted to a base plate such that the cells are
tem is ~182 kg. maintained at the proper temperature through
a liquid cooling system, which moves excess
heat to a set of radiators as described in Sec-

C-6
APPENDIX C: SOLAR ENCOUNTER POWER GENERATION TRADE STUDY

Table C-1. Power generation trade study summary.


Concept Energy Storage Thermocouple Stirling Engine Photovoltaic
Perihelion >4 RS >4 RS >4 RS >9 RS
Power Through Yes Requires batteries Requires batteries Yes
Encounter?
Operational Entire encounter Perihelion ± 1.5 days Perihelion ± 2 days Entire encounter
Range
Heat Load in <50 W <50 W <50 W <50 W
Spacecraft Bus
Subsystem Mass 400–800 kg 210 kg 182 kg 66 kg
Advantages • No mechanisms • Well understood • Moderate TRL • Lowest mass solution
• Simple semiconductor • Better energy • Retractable arrays
implementation devices conversion efficiency allow fine control of
• Highest TRL • No mechanisms than thermocouples, power generation and
• High TRL slightly better than thermal control
photovoltaics • No hardware
embedded in TPS
Disadvantages • Most massive concept • Difficult electrical • Complicated interface • Low system TRL,
• New battery chemistry interconnect scheme to TPS and support most major
may be required, • Low efficiency means structure components have
significant large radiator areas to • Potential for vibration flight heritage
development cost maintain cold from engines affecting
temperature end pointing
• Must be embedded • Hot collector
into TPS, interface embedded into or
difficult attached to bottom
surface of heat shield

tion 3 of this report. The photovoltaic concept creased mass and still achieve orbit, much less
will generate power throughout the solar en- maintain the mass margin required as part of
counter without the long-term need for batter- the overall Solar Probe+ study. The low mass
ies or other form of energy storage. The mass of the photovoltaic concept relative to the oth-
of this concept is estimated to be 66 kg. This ers considered is an enabler for the Solar
concept was eventually chosen as a result of Probe+ mission concept as a whole.
this trade study.
C.3 Trade Study Results
Table C-1 summarizes each concept and dis-
cusses advantages and disadvantages of each.
Ultimately, the choice of concept for Solar
Probe+ was driven by mass. Systems involv-
ing the use of energy storage were all more
massive than the radioisotope thermal genera-
tor (RTG) system baselined for the 2005
study 2 (which was 120 kg), and none of the
orbits under consideration in the mission de-
sign trade study could incorporate the in-

2
Solar Probe: Report of the Science and Technology
Definition Team, NASA/TM—2005–212786, National
Aeronautics and Space Administration, Goddard Space
Flight Center, Greenbelt, MD (2005).

C-7
APPENDIX D: REFERENCES

APPENDIX D:

REFERENCES

D-1
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

D-2
APPENDIX D: REFERENCES

Atlas Launch Systems Mission Planner’s Guide, Rev. 10a, Lockheed Martin Corporation
(January 2007).
Boslough, M. B., et al., Hypervelocity testing of advanced shielding concepts for spacecraft
against impacts to 10 km/s, Int. J. Impact. Eng. 14, 96–106 (1993).
Covert, R., Ten Common Things Wrong with Cost Risk Analysis, Presented at 76th SSCAG,
Hampton Roads, VA (October 22–23, 2002).
General Environmental Verification Standards (GEVS) for GSFC Flight Programs and
Projects, GSFC-STD-7000, Goddard Space Flight Center (April 2005).
Gloeckler, G., et al., Solar Probe: First Mission to the Nearest Star, Report of the NASA
Science Definition Team for the Solar Probe Mission, The Johns Hopkins University Applied
Physics Laboratory, Laurel, MD (1999).
Goldstein et al., Spacecraft Mass Loss and Electric Potential Requirements for the Starprobe
Mission, A Report of the Starprobe Mass Loss Requirements Group Meeting of September 29–
30, 1980, NASA Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA
(December 1980).
Ishimoto, H., Mann, I., Modeling the number density distribution of interplanetary dust on
the ecliptic plane within 5 AU of the Sun, Astron. Astrophys. 362, 1158–1173 (2000).
Load Analyses of Spacecraft and Payloads, NASA-STD-5002, National Aeronautics and
Space Administration (June 21, 1996).
Mann, I., et al., Dust near the Sun, Space Sci. Rev. 110, 269–305 (2004).
Payload Test Requirements, NASA-STD-7002, Rev A, National Aeronautics and Space
Administration (September 10, 2004).
Payload Vibroacoustic Test Criteria, NASA-STD-7001, National Aeronautics and Space
Administration (June 21, 1996).
Qualification and Quality Requirements for Space Solar Cells, AIAA Standard S-111-2005,
American Institute of Aeronautics and Astronautics (January 2005).
Solar Probe Risk Mitigation Study, prepared by The Johns Hopkins University Applied
Physics Laboratory, 2006 Mid-Year Report.
Solar Probe Thermal Protection System Risk Mitigation Study: FY 2006 Final Report,
prepared by The Johns Hopkins University Applied Physics Laboratory under Contract NAS5-
01072, Laurel, MD (November 30, 2006).
Solar Probe Thermal Protection System Risk Mitigation Study: FY 2006 Final Report ITAR-
Restricted Annex, prepared by The Johns Hopkins University Applied Physics Laboratory under
Contract NAS5-01072, Laurel, MD (September 17, 2007).
Solar Probe: An Engineering Study, prepared by The Johns Hopkins University Applied
Physics Laboratory, in partnership with the Jet Propulsion Laboratory, under contract NAS5-
01072, Laurel, MD (November 12, 2002).

D-3
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

Solar Probe: Report of the Science and Technology Definition Team, NASA/TM—2005–
212786, National Aeronautics and Space Administration, Goddard Space Flight Center,
Greenbelt, MD (2005).
Solar Probe+: Report of the Science and Technology Definition Team, Southwest Research
Institute, San Antonio, TX, in press (2008).
Unmanned Space Vehicle Cost Model, Eighth Edition (USCM8), Tecolote Research, Inc.,
Goleta, CA, www.uscm8.com (October 2001).

D-4
APPENDIX E: ACRONYMS AND ABBREVIATIONS

APPENDIX E:

ACRONYMS AND ABBREVIATIONS

E-1
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

E-2
APPENDIX E: ACRONYMS AND ABBREVIATIONS

ACO Administrative Contracting Delta-DOR Delta Differential One-Way


Officer Range
ADACS Attitude Determination and DOD Depth of Discharge
Control Subsystem DSMS Deep Space Mission System
AGE Ground Support Equipment DSN Deep Space Network
AIAA American Institute of EMC Electromagnetic Compatibility
Aeronautics and Astronautics EMCPM Engineering Model Cell
AICF Aerogel Infiltrated Carbon Positioning Mechanism
Foam EMI Electromagnetic Interference
AISC Application-Specific Integrated EPI Energetic Particle Instrument
Circuits EUV Extreme Ultraviolet
Al2O3 Aluminum Oxide, Alumina FEA Fast Electron Analyzers
AU Astronomical Unit FIA Fast Ion Analyzer
C&DH Command and Data Handling FOV Field of View
CBE Current Best Estimate FPM Fault Protection Module
CC&DH Command, Control, FSW Flight Software
Communications, and Data FY07$ Fiscal Year 2007 Dollars
Handling FY07$M Fiscal Year 2007 Millions of
CCD Charge-Coupled Device Dollars
CCSDS Consultative Committee for FY08 Fiscal Year 2008
Space Data Systems g Acceleration of Gravity
CD Coronal Dust Detector G&C Guidance and Control
CDPU Common Data Processing Unit GaAs Gallium Arsenide
CDR Critical Design Review GDS Ground Data System
CER Cost Estimating Relationship GEVS General Environmental
CFDP CCSDS File Delivery Protocol Verification Standard
cFE Core Flight Executive Gr/CE Carbon Graphite/Cyanate Ester
CFOV Clear Field of View GSFC Goddard Space Flight Center
CG Center of Gravity HGA High-Gain Antenna
CIC Cell-Interconnect-Cover HI White-Light Hemispheric
CLA Coupled Loads Analysis Imager
Co-I Coinvestigator I&T Integration and Testing
CP Center of Aerodynamic ICA Ion Composition Analyzer
Pressure ICE Independent Cost Estimate
cPCI Compact Peripheral Component IEM Integrated Electronic Module
Interface in. Inch
CPD Cumulative Probability I–V Current–Voltage
Distribution JGA Jupiter Gravity Assist
CRISM Compact Reconnaissance JSC Johnson Space Center
Imaging Spectrometer for Mars kg Kilogram
CTE Coefficient of Thermal KSC Kennedy Space Center
Expansion lb Pound
CTH Coupled Thermodynamic and lbf Pound of Force
Hydrodynamic Hydrocode LGA Low-Gain Antenna
DCMC Defense Contract Management LRO Lunar Reconnaissance Orbiter
Command LVPS Low-Voltage Power Supply

E-3
SOLAR PROBE+ MISSION ENGINEERING STUDY REPORT

M&SA Mission and Safety Assurance RS Solar Radius


MAG Magnetometer RTG Radioisotope Thermoelectric
MCR Mission Confirmation Review Generator
MDR Mission Design Review RTW Real-Time Workshop
MESSENGER MErcury Surface, Space RVC Reticulated Vitreous Carbon
ENvironment, GEochemistry, RY$ Real Year Dollars
and Ranging S/N Signal-to-Noise
MGSE Mechanical Ground Handling SAJB Solar Array Junction Box
Equipment SDO Solar Dynamics Observatory
MLI Multi-Layer Insulation SEP Sun–Earth–Probe
MODC Miscellaneous Other Direct SEP Systems Engineering
Costs SPDT Single-Pole-Double-Throw
MOS Mission Operations System SPE Sun–Probe–Earth
MRO Mars Reconnaissance Orbiter SSR Solid-State Recorder
MSX Midcourse Space Experiment STDT Science and Technology
NAFCOM NASA–Air Force Cost Model Definition Team
NGS Neutron/Gamma-Ray STEREO Solar TErrestrial RElations
Spectrometer Observatory
NICM NASA Instrument Cost Model SwRI Southwest Research Institute
OD Orbit Determination TCM Trajectory Correction Maneuver
PAS Product Assurance System TDRSS Tracking and Data Relay
PBN Pyrolytic Boron Nitride Satellite System
PDF Probability Density Function TIMED Thermosphere, Ionosphere,
PDR Preliminary Design Review Mesosphere Energetics and
PDT Pacific Design Technology Dynamics
PDU Power Distribution Unit TPS Thermal Protection System
PER Pre-Environmental Review TRL Technology Readiness Level
PI Principal Investigator TSA Transition Structure Assembly
PM Program Management TT&C Telemetry, Tracking, and
POE Program Office Estimate Control
PPT Peak Power Tracker TVA Thrust Vector Actuator
PSE Power System Electronics TWTA Traveling Wave Tube Amplifier
psig Pound-Force per Square Inch USCM8 Air Force’s Unmanned
Gauge Spacecraft Cost Model
PSP Participating Science Program UTC Coordinated Universal Time
PSR Pre-Ship Review W Watt
PWI Plasma-Wave Instrument WBS Work Breakdown Structure
RBSP Radiation Belt Storm Probes XFER Switch Transfer Switch
RF Radio Frequency α Absorptivity
RHCP Right-Hand Circularly ε Emissivity
Polarized
RIU Remote Interface Unit
RMIS Resource Management
Information System
RMS Root-Mean-Squared
ROM Rough Order of Magnitude

E-4
Sol
arProbe+

2
008Sol
Mis
sionEngi
neer
ing

arPr
obe+Mi
St
udyReport

ssi
Mar
ch1
0,2
008

Mar
ch1
0,2
008 onEngi
neer
ingSt
udyRepor
t

07-
05537-
1

You might also like