A Perspective on the Design and

Development of the SpaceX Dragon

Spacecraft Heatshield

by
Daniel J. Rasky, PhD

Senior Scientist, NASA Ames Research Center

Director, Space Portal, NASA Research Park
Moffett Field, CA 94035
(650) 604-1098 / [email protected]

February 28, 2012

 2

How Did SpaceX Do This?

Recovered Dragon Spacecraft!

After a “picture perfect” first flight, December 8, 2010 !
 3

Beginning Here?

SpaceX Thermal Protection Systems Laboratory, Hawthorne, CA!

“Empty Floor Space” December, 2007!
 4

Some Necessary Background:

Re-entry Physics
•  Entry Physics Elements
–  Ballistic Coefficient
–  Blunt vs sharp nose tip
–  Entry angle/heating profile
–  Precision landing reqr.
–  Ablation effects
–  Entry G’loads
»  Blunt vs Lifting shapes
–  Lifting Shapes
»  Volumetric Constraints
»  Structure
»  Roll Control
»  Landing Precision
–  Vehicle flight and turn-around
requirements

Re-entry requires specialized design and expertise for the Thermal

Protection Systems (TPS), and is critical for a successful space vehicle
 5

Reusable vs. Ablative Materials

 6
Historical Perspective on TPS:
The Beginnings

•  Discipline of TPS began during World War

II (1940’s)
–  German scientists discovered V2 rocket was
detonating early due to re-entry heating
–  Plywood heatshields improvised on the vehicle to EDL
solve the heating problem

•  X-15 Era (1950’s, 60’s)

–  Vehicle Inconel and Titanium metallic structure
protected from hypersonic heating AVCOAT
»  Spray-on silicone based ablator for acreage
»  Asbestos/silicone moldable TPS for leading
edges
–  Spray-on silicone ablator found to be inadequate
»  Unable to protect the vehicle beyond Mach 6
»  Required considerable labor to refurbish
Arc Jet Testing
 7
Historical Perspective on TPS:
Ablatives
•  Mercury/Gemini/Apollo (1960’s)
–  Needed a lighter weight system than DoD re-entry body TPS of
high density carbon or quartz phenolic
–  Developed polymer based moldable ablators with high
temperature honeycomb reinforcement to withstand re-entry
and lunar return environment: Avcoat 5026-39/HC-G for Apollo
–  Approximately 1/3 the weight of high-density carbon-phenolic EDL

•  Viking (1970’s)
–  Apollo heatshield too heavy for Mars entry
–  Silicone based moldable light-weight ablator reinforced with a
high-temperature honeycomb developed: super-lightweight
ablator - SLA-561
–  Similar to Apollo TPS but lighter weight (~1/2 the density)
–  Good insulator with a robust architecture

•  Pioneer-Venus, Galileo (1970’s, 80’s)

–  NASA did not have materials to handle severe entry conditions
for the Venus or Jupiter entries
–  DoD developments in high density carbon phenolic used to
meet mission requirements
–  NASA did not fully explore material payload impacts from use
of DoD class heatshields Arc Jet Testing
 8
Historical Perspective on TPS:
Reusables
•  Reusable materials technology investment dominated TPS development
efforts in the late 70’s through 80’s, 90’s and early 2000’s
–  Shuttle: Development of first reusable TPS
»  Reinforced Carbon-Carbon (RCC), Ceramic Tiles (LI-900), TPS Blankets (FRSI & AFRSI),
Refractory metals (Coated Niobium)
–  NASP: Investigation of advanced reusable TPS
»  Ceramic Matrix Composites (CMC’s), Metal Matrix Composites (MMC’s), Actively Cooled
Systems
–  X-vehicles (X-33, X-37, X-38, X-43): Development and investigation of more moderately
advanced TPS
»  Metallic TPS, Advanced Carbon-Carbon, CMC’s, sharp hypersonic leading edges, high-
temperature tiles for leading-edges
 9
Ablative TPS Technology Development:
Post Apollo/ Viking/ Galileo Era

•  Lightweight Ceramic Ablators research

initiated at Ames in the early 1990’s (Rasky,
Tran)
–  Goal was to produce a new generation of ablators,
making use of advancements in materials technologies
»  ceramic substrates with polymer impregnants
–  Superior capabilities fit well with the Faster, Better and
Cheaper philosophy
»  adopted for Mars Pathfinder, Mars Exploration
Rovers, Mars Sclence Laboratory, Stardust,
SpaceX Dragon
 10

A New Class of Ablators:

Light-Weight Ceramic Ablators (LCA’s)
Traditional Ablators*! Light-Weight Ceramic Ablators!
Polymer Based! Ceramic Based!
Top Surface!

Polymer! Fibrous!
Matrix! Ceramic!
Substrate**!

Polymer!
Ceramic &! Impregnant!
Organic! (Multiple &!
Fillers! Graded!
Possible)!

(*e.g. Avcoat -5026, SLA-561V, Carbon-Phenolic)! (**e.g. silica, carbon, alumina fibers)!

Disadvantages:! Advantages:
– Little strength at high temperature – Good structural integrity at high temperature,
requiring reinforcing (e.g., honeycombs)! avoids need for reinforcing honeycombs
– Restrictive design and performance – Multiple and graded polymer impregnants
characteristics (e.g., thickness limits, possible to optimize ablative and insulative
pressure limits, heavy)! performance (e.g., SPLIT)
– Labor intensive manufacturing process, – Billet fabrication process giving a low cost,
giving high fabrication costs and lot to flexible, CAM compliant material
lot variations!
 11

Light Weight Ceramic Ablator Family

•  SIRCA
–  Silicone Impregnated Refractory Ceramic Ablator
–  Uses flight certified ceramic substrates (Shuttle)
and silicone impregnants (Viking)
–  Densities: 0.20 - 0.40 gm/cc
SIRCA!
–  For heat fluxes < 300 W/sqcm
–  Patents 5,536,562 & 5,672,389

•  PICA
–  Phenolic Impregnated Carbon Ablator
–  Uses Fiberform substrates from FMI, with flight
grade phenolic impregnant
–  Densities: 0.25 - 0.60 gm/cc PICA!
–  For heat fluxes > 300 W/sqcm
–  Patents 5,536,562 & 5,672,389

•  SPLIT
–  Secondary Polymer Layered Impregnated Tile
–  Used with either SIRCA or PICA to improve ablator Phenolic!
effectiveness by augmented passive phase
change and transpiration cooling
–  Densities: 0.25 - 0.80 gm/cc
PICA/SPLIT!
–  Patents 6,955,853

PMMA!
 12

PICA Forebody for Stardust

Fastest entry ever of a spacecraft at Earth! (12.9 km/s)
January 15, 2006

Forebody design details:!

–  Single piece Fiberform carbon
substrate vacuumed formed to
rough shape by FMI!
–  Substrate impregnated with
phenolic, and then machined to
final shape by FMI!
–  0.82 m diameter heatshield then
Post-Flight Stardust Sample Return integrated and bonded to
Probe spacecraft structure by LMA!
–  Qualified for Stardust entry
environment: !
»  Heat flux = 950 W/cm2,
Pressure = 0.45 atm,
Heat load = 36 KJ/cm2!
–  Significant impact crush
capability demonstrated for hard
landing after entry!

Great re-enty video: http://www.youtube.com/watch?v=H1JxIp2B7Jc!

 13

Stardust Capsule,
including PICA
Heatshield,
on display at the
Smithsonian National
Air and Space Museum

•  Part of the “Milestones

of Flight” Display
 14

Back to SpaceX…
•  By 2007, SpaceX had selected
PICA as their material of choice
for the Dragon primary
heatshield
–  Elon very impressed with Stardust
performance and capabilities

•  Fall, 2007, Dr. Rasky

approached by Elon Musk to
help transfer PICA technology to
SpaceX
Early Dragon primary heatshield mockup - 2007!

•  Spring 2008 through 2009, Dr. Rasky works closely with SpaceX
(~1/2 time at SpaceX facilities) and other colleagues at NASA
Ames to transfer PICA, and support Dragon heatshield design
 15

Successful Tech Transfer of PICA

•  Laboratory sized samples
successfully made at
Hawthorne
–  Spring 2008
–  A number of formula variations
produced and investigated
–  Three different carbon fiber tiles
substrates used
–  PICA-X formulation established by
fall, 2008

•  Full size production billet of PICA-X undergoing inspection!

PICA-X demonstrated
–  Prototype produced in fall, 2008
–  Using a custom designed vacuum
oven with very precise thermal
control (both spatially and
temporarily)
 16

Test Validation of PICA-X

•  PICA-X successfully certified
for flight
–  Very successful arc-jet test series
conducted at NASA Ames in
December 2008
–  Three different carbon-fiber
substrate PICA-X versions tested
–  All performed above expectations

Successful certification arc-jet testing at

•  Production capability NASA Ames – December 2008!
established
–  Batch processing for PICA-X
demonstrated by fall 2009
–  Ability to produce PICA-X in excess
of that needed for Dragon
 17

PICA-X Installed on Dragon

•  PICA-X being installed on Dragon carbon-composite

carrier structure, 2010
 18

PICA-X Heatshield Installed

on Dragon, 2010
 19

Dragon Integrated to Falcon-9

 20

Dragon/Falon-9 Ready for Roll-out

 21

Dragon/
Falcon-9
Ready for
Launch
 22

Dragon/
Falcon-9
Launch

•  December 8,
2010
 23

Dragon Re-entry
Artists Reconstruction!
 24

Dragon
Descent
 25

Dragon Recovery

Recovered Dragon Spacecraft!

After a “picture perfect” first flight, December 8, 2010 !
 26

So How Did SpaceX Succeed So

Extraordinarily??
 27

So How Did SpaceX Succeed So

Extraordinarily??

They had learned from their mistakes on the

Falcon-1, and the first Falcon-9 launch
(took three Falcon-1 failures to get their first fully
successful flight)
 28

So How Did SpaceX Succeed So

Extraordinarily??

They had learned from their mistakes on the

Falcon-1, and the first Falcon-9 launch
(took three Falcon-1 failures to get their first fully
successful flight)

Everything went their way on this flight

(won’t necessarily be the case for all future flights)
 29

So How Did SpaceX Succeed So

Extraordinarily??

They had learned from their mistakes on the

Falcon-1, and the first Falcon-9 launch
(took three Falcon-1 failures to get their first fully
successful flight)

Everything went their way on this flight

(won’t necessarily be the case for all future flights)

And importantly, by using a different business

model than traditional government aerospace
(a potential game changer)
 30

Traditional Government
Aerospace Business Model
 31

Traditional Government
Aerospace Business Model
•  Modeled on military organizational approaches:
–  Hierarchal, with chain of command
–  Much more focus on control than on efficient use of resources
–  Rely on a large cadre of internal experts and unique facilities
–  Form key alliances with customers, stake holders and specialized
suppliers
–  Follow a fairly rigid requirements driven design approach
 32

Traditional Government
Aerospace Business Model
•  Modeled on military organizational approaches:
–  Hierarchal, with chain of command
–  Much more focus on control than on efficient use of resources
–  Rely on a large cadre of internal experts and unique facilities
–  Form key alliances with customers, stake holders and specialized
suppliers
–  Follow a fairly rigid requirements driven design approach

•  Prefer “Cost-Plus” contracting with the government

–  Covers contractors costs, plus a small profit (~6-7%)
–  Provides flexibility for the government to change requirements
–  Both contractor internal and supplier cost increases can be passed
onto the government customer
 33

Traditional Government
Aerospace Business Model
•  Modeled on military organizational approaches:
–  Hierarchal, with chain of command
–  Much more focus on control than on efficient use of resources
–  Rely on a large cadre of internal experts and unique facilities
–  Form key alliances with customers, stake holders and specialized
suppliers
–  Follow a fairly rigid requirements driven design approach

•  Prefer “Cost-Plus” contracting with the government

–  Covers contractors costs, plus a small profit (~6-7%)
–  Provides flexibility for the government to change requirements
–  Both contractor internal and supplier cost increases can be passed
onto the government customer

•  Proven record for producing custom, complex hardware

and systems
–  With very high performance and reliability
–  That have national security functions or implications
–  Where cost often is not a driver
 34

SpaceX Business Model

 35

SpaceX Business Model

•  Adopted from the Software Development industry:
–  Where Elon got his management and development experience
–  Very flat organizationally
–  Broad and organic collaboration and communication
–  Rely extensively on the internet for technical data, product data, and
procurement of equipment and services
–  Must have multiple suppliers for any critical path components, or will bring in-
house
–  Design approach is collaborative and pursues crawl before you walk before
you run development strategies, rapid prototyping, and identification of low-
cost approaches that allow iterative improvement
 36

SpaceX Business Model

•  Adopted from the Software Development industry:
–  Where Elon got his management and development experience
–  Very flat organizationally
–  Broad and organic collaboration and communication
–  Rely extensively on the internet for technical data, product data, and
procurement of equipment and services
–  Must have multiple suppliers for any critical path components, or will bring in-
house
–  Design approach is collaborative and pursues crawl before you walk before
you run development strategies, rapid prototyping, and identification of low-
cost approaches that allow iterative improvement

•  Prefer “Fixed Price” contracting with customers

–  A fixed price for a fixed set of produced hardware and/or services
–  Minimizes customer requirement changes & insite/oversite
–  Allows for considerable potential profit
–  Relies on very good internal and supplier cost control
 37

SpaceX Business Model

•  Adopted from the Software Development industry:
–  Where Elon got his management and development experience
–  Very flat organizationally
–  Broad and organic collaboration and communication
–  Rely extensively on the internet for technical data, product data, and
procurement of equipment and services
–  Must have multiple suppliers for any critical path components, or will bring in-
house
–  Design approach is collaborative and pursues crawl before you walk before
you run development strategies, rapid prototyping, and identification of low-
cost approaches that allow iterative improvement

•  Prefer “Fixed Price” contracting with customers

–  A fixed price for a fixed set of produced hardware and/or services
–  Minimizes customer requirement changes & insite/oversite
–  Allows for considerable potential profit
–  Relies on very good internal and supplier cost control

•  Goal is to produce hardware and services at large scale

–  For use by government and the general public
–  With very good performance margins and real world use to ensure acceptable
operation
 38

Will the SpaceX Business Model

Continue to Provide These
Extraordinary Results?
 39

Will the SpaceX Business Model

Continue to Provide These
Extraordinary Results?

Too early to say

 40

Will the SpaceX Business Model

Continue to Provide These
Extraordinary Results?

Too early to say

But it certainly is interesting

 41

Will the SpaceX Business Model

Continue to Provide These
Extraordinary Results?

Too early to say

But it certainly is interesting

And quite a contrast to most of our recent

experience with Space
 42

What will SpaceX do next??

 43

What will SpaceX do next??

Perhaps help take us to the Moon and Mars…

 44

Questions
 45

Backup Slides
 46
Historical Perspective:
Ablative TPS
•  TPS Investment in the 60’s - Focused Program -
Technology development with specific mission goal
–  Material Performance, Heat Shield System Development and
Design Architecture
–  Test, Test and more Test
–  Ground and flight test => Material behavior, Analytical capabilities
and model development
EDL

•  Apollo 1960’s - 1970’ Avcoat 5026-39/HC-G

–  Developed honeycomb system due to reliability risk of tiled
approach
»  Needed a lighter weight system compared to DOD TPS
(Carbon- or Quartz Phenolic)
–  Too heavy for Mars entry - Viking

•  Viking (1975) SLA-561 AVCOAT

–  Used low density silicone in honeycomb - similar to Apollo TPS
»  Good insulator with a robust architecture

•  Pioneer-Venus, Galileo
–  NASA didn’t have materials to handle entry conditions
–  DoD investment in carbon phenolic leveraged to these missions
–  But, NASA did not fully explore material performance limits due to
facility capability (e.g., spallation on Galileo)
Arc Jet Testing
 Commercial space is an important
and growing segment of
the US space industry...

...NASA under Gen Bolden will actively

support and advocate its development.

Scaled Composites!

XCOR!

Blue Origin!
SpaceX! Orbital Sciences!
 48

LCA Development History

•  Light-weight Ceramic Ablators (LCA’s), were

conceived and developed at Ames starting in
in the early 1990’s
–  Concept based on Ames’ expertise in low density fibrous
ceramic substrates
»  Developed several fibrous ceramic substrates for TPS used on
the Space Shuttle (AIM-22, FRCI-12, AETB-8)
–  Combined with expertise and advances in ceramic polymer
precursor technology over the past 20 years
»  Selected polymer(s) impregnated into a suitable fibrous
ceramic substrate
»  Innovative impregnation techniques developed at Ames to
maintain low density and good thermal properties
–  Approach maximizes ablation and thermal performance,
and minimizes fabrication costs
 49

PICA Forebody for Stardust

Arc-Jet Testing at Reference
Sample Return Entry Conditions
(qcw = 400 W/cm2, Pstag = 0.25 atm, qload = 24 KJ/cm2) PICA (Phenolic Impregnated Carbon
Ablator):
>> Base lined by Lockheed-
Apollo Shield - Heavy, Martin for the Stardust
with Substantial fore body (single piece)
Recession and Mass heat shield
Loss

Avco-5026

New PICA material -

Lighter Weight with
Reduced Recession
and Mass Loss PICA !

Stardust Sample Return

Probe
PICA-15

Significantly Improved Capability, Reduced Weight and Cost

Compared to Apollo Era Materials -
Enabling Technology for Stardust
 50

PICA Material Performance

 51
Historical Perspective on TPS:
New Ablators, Tiles and Advanced Blankets

•  Modest budget level research and development

continued on ablators (1980’s, 90’s)
–  Light-Weight Ceramic Ablator work at NASA Ames
»  Ceramic substrates with polymer impregnants, yielding
several useful systems (PICA, SIRCA, SPLIT, Black Tile)
–  Polymer based ablator development at Applied Research
Associates
»  Derivatives of Viking Super-Lightweight Ablator (SLA)
–  Silicone ablator development at ITT Industries (formerly
Acurex/Aerotherm)
»  Acusil line of moldable TPS products

•  Modest budget level research and development

on tile and blanket TPS (1980’s, 90’s)
–  Higher temperature tiles (AETB) with tougher coatings
(TUFI, TUFROC) at NASA Ames
–  Higher temperature quilted blankets (Nextel fabrics, Silicon-
carbide fabrics, Saffil batting) at NASA Ames
»  Silicon-carbide fabrics found to be a health hazard
–  Toughened metal (DuraFRSI - NASA Ames) and ceramic
coatings (CRI - Boeing) for blankets
–  Higher temperature felts blankets (PBI, PBO, carbon) at
NASA Ames