Military and Aerospace FPGA and Applications (MAFA)

Exploration Technology Development Program’s

Radiation Hardened
Electronics for Space
Environments (RHESE)

Andrew S. Keys, James H. Adams, Donald O. Frazier, Marshall C. Patrick, and

Michael D. Watson
NASA Marshall Space Flight Center, Huntsville, AL 35812
Michael A. Johnson
NASA Goddard Space Flight Center, Greenbelt, MD 20771
John D. Cressler
Georgia Institute of Technology, Atlanta, GA 30332-0250
Elizabeth A. Kolawa
Jet Propulsion Laboratory, Pasadena, CA 91109
RHESE
MAFA–-27
Integrated
NovemberBaseline
2007 Review – 12 September 2007
 Vision for Space Exploration

• The Vision for Space

Exploration (VSE) directs
NASA to pursue a long-
term human and robotic
program to explore the
solar system.

• The VSE is based on the

following goals:
– Return the shuttle to flight
(following the Columbia
accident) and complete the
International Space Station
by 2010.
– Develop a Crew Exploration
Vehicle, test by 2008, first
manned mission no later
than 2014.
– Return to the Moon as early
as 2015 and no later than
2020.
• Gain experience and
knowledge for human missions
to Mars.
• Increase the use of robotic
exploration to maximize our
understanding of the solar
system.
SPACE 2007 – 20 September 2007 2
 Constellation Program

• The Constellation Program consists of multiple projects,

jointly being developed to fulfill the goals of the VSE.
Ares V Launch
Vehicle
Ares I Launch
Vehicle

Lunar Lander

Orion Crew
Exploration Vehicle

Lunar Outpost

SPACE 2007 – 20 September 2007 3

 Surviving the Radiation Environment

• Space Radiation
affects all spacecraft.
– Spacecraft electronics
have a long history of
power resets, safing,
and system failures
due to:
• Long duration
exposures,
• Unpredictable solar
proton activity,
• Ambient galactic
cosmic ray
environment.

SPACE 2007 – 20 September 2007 4

 The Radiation Environment

• Multiple approaches may be employed (independently or in

combination) to protect electronic systems in the radiation
environment:
– Shielding,
– Mission Design (radiation avoidance),
– Radiation Hardening by Architecture,
• Commercial parts in redundant and duplicative configurations (Triple Module
Redundancy),
• Increases overhead in voting logic, power consumption, flight mass
• Multiple levels of redundancy implemented for rad-damage risk mitigation:
– Component level
– Board level
– Subsystem level
– Spacecraft level
– Radiation Hardening by Design,
• TMR strategies within the chip layout,
• designing dopant wells and isolation trenches into the chip layout,
• implementing error detecting and correction circuits, and
• device spacing and decoupling.
– Radiation Hardening by Process,
• Employ specific materials, processing techniques,
• Usually performed on dedicated rad-hard foundry fabrication lines.

SPACE 2007 – 20 September 2007 5

 Radiation Hardened Assurance

• NASA spacecraft developers have defined a Radiation

Hardness Assurance (RHA) methodology process*.
• In general, the process may be described by the following
steps:
– 1) define the radiation hazard,
– 2) evaluate the hazard,
– 3) define the requirements to be met by the spacecraft’s electronics,
– 4) evaluate the electronics to be used,
– 5) engineer processes to mitigate hazard damage, and
– 6) iterate on the methodology, if and when necessary.
• To promote the successful implementation of RHA for
Constellation (and other NASA) missions, the RHESE project
aims to deliver products that assist in mitigating the hazard
damage.

*LaBel, K. A., Johnson, A. H., Barth, J. L., Reed, R. A., and Barnes, C. E., “Emerging Radiation
Hardness Assurance(RHA) Issues: A NASA Approach for Space Flight Programs,” IEEE Transactions
on Nuclear Science, Vol. 45, No. 6, Dec. 1998, pp. 2727-2736.

SPACE 2007 – 20 September 2007 6

 RHESE Overview and Objectives

The Radiation Hardened Electronics for Space

Environments (RHESE) project expands the current
state-of-the-art in radiation-hardened electronics to develop
high performance devices robust enough to withstand the
demanding radiation and thermal conditions encountered
within the space and lunar environments.

The specific goals of the RHESE project are to foster technology

development efforts in radiation-hardened electronics possessing
these associated capabilities:
– improved total ionization dose (TID) tolerance,
– reduced single event upset rates,
– increased threshold for single event latch-up,
– increased sustained processor performance,
– increased processor efficiency,
– increased speed of dynamic reconfigurability,
– reduced operating temperature range’s lower bound,
– increased the available levels of redundancy and reconfigurability, and
– increased the reliability and accuracy of radiation effects modeling.

SPACE 2007 – 20 September 2007 7

 Customer Requirements and Needs

• RHESE is a “requirements-pull” technology development effort.

• RHESE is a “cross-cutting” technology, serving a broad base of multiple

project customers within Constellation.
– Every project requiring…
• operation in an extreme space environment,
• avionics, processors, automation, communications, etc.
…should include RHESE in its implementation trade space.

• RHESE’s products are developed in response to the needs and

requirements of multiple Constellation program elements, including:
– Ares V Crew Launch Vehicle,
– Orion Crew Exploration Vehicle’s lunar capability,
– Lunar Lander,
– Lunar Outpost,
– Surface Systems,
– Extra Vehicular Activity (EVA) elements,
– Future applications to Mars exploration architecture elements.

• Constellation Program requirements for avionics and electronics continue

to evolve and become more defined.

• RHESE will develop products per derived requirements based on the

Constellation Architecture’s Level I and Level II requirements defined to
date.

SPACE 2007 – 20 September 2007 8

 RHESE Tasks

• Specifically, the RHESE tasks are:

– Model of Radiation Effects on Electronics (MREE),
– Single Event Effects (SEE) Immune Reconfigurable Field
Programmable Gate Array (FPGA) (SIRF),
– Radiation Hardened High Performance Processors (HPP),
– Reconfigurable Computing (RC),
– Silicon-Germanium (SiGe) Integrated Electronics for Extreme
Environments.

SPACE 2007 – 20 September 2007 9

 MREE Technology Objectives
• The Main Objective
– A computational tool to estimate radiation
effects in space in support of spacecraft design
• Total dose
• Single Event Effects
• Secondary Objectives
– To provide a detailed description of the
radiation environment in support of radiation
health and instrument design
• In deep space
• Inside the magnetosphere
• Behind shielding

SPACE 2007 – 20 September 2007 10

 Update the Method for SEE Calculation

CREME96 MREE

Integral over
path length
Distribution + Multi-volume Calorimetry +
critical charge Charge-collection models +
Critical charge

SPACE 2007 – 20 September 2007 11

 SIRF Drivers

• Reconfigurable gate arrays form the basis of many adaptable,

scaleable, computing engines
– Add flexibility, capability and robustness to surface and flight systems

SPACE 2007 – 20 September 2007 12

 SIRF Architecture
Based on Commercial Devices

• 5th generation Virtex™ device

• Columnar Architecture enables

resource “dial-in” of
– Logic
– Block RAM
– I/O
– DSP Slices
– PowerPC Cores

Fabrication process and device architecture

yield a high speed, flexible component

SPACE 2007 – 20 September 2007 13

 SIRF Objective
Radiation-hardened Device

• Existing reconfigurable FPGAs are very susceptible to radiation-

induced single event effects
– Significant FPGA resources are currently required to mitigate radiation-
induced single event effects

… …

SIRF technology
Current technology

…
…

Rad-Susceptible SIRF Fabric

FPGA Fabric - Logic function

User invoked TRM required Mitigation implemented at device level

(requires FPGA resources)

Objectives: Eliminate need for user-invoked TMR. Bring a state-of-the-art radiation

hardened reconfigurable FPGA to the space electronics market by ~2010.

SPACE 2007 – 20 September 2007 14

 HPP Drivers

• Problem: Exploration Systems Missions

Directorate objectives and strategies
can be constrained by computing
capabilities and power efficiencies
– Autonomous landing and hazard
avoidance systems
– Autonomous vehicle operations
– Autonomous rendezvous and docking
– Vision systems

SPACE 2007 – 20 September 2007 15

 HPP Technical Approach
Multi-generation Performance Lag

• Radiation-hardened processors lag commercial devices by several technology generations

(approx. 10 years)
– RHESE High performance Processor project full-success metric for general purpose processors
conservatively keeps pace with historical trend (~Moore’s Law)

Core 2

1E4 Pentium IV HPP General

Purpose Processor
Performance Metric
2000
Pentium III
1000
Pentium Pro PPC440
Processor Throughput

LEON3
Pentium PPC603e RAD750 RT ULP ColdFire1
100
80486DX
(MIPS)

Commercial Processor R6000

Radiation-hardened processor Mongoose V
10 80386DX
w
La
’s
re
oo

w
La
M

1.0 ’s
re
80286
oo
M

0.1
8080

1970 1980 1990 2000 2010 2020

~10 yrs. Lunar Lunar
Year EVA Outposts
PDR Dates
CEV LSAM
Block II Lander
M.A.Johnson/GSFC SPACE 2007 – 20 September 2007 16
 Reconfigurable Computing
Subproject
• Subproject Objectives

• Provide reconfigurable computing capabilities as a preferred alternative to

conventional forms
– Processor Modularity
– Interface Modularity
• Reduction of Flight Spares
• Accommodation for Circuit Life Limitations
• Resources where needed, as needed

• Supplement other efforts to mitigate environmental impacts by providing the

capability to detect and work around malfunctioning circuitry
– Fault Tolerance
– Fault Detection, Isolation, and Mitigation
•
• Generally: capitalize on the unique capabilities of RC to adapt in target
systems for changing requirements, performance and environmental
parameters

SPACE 2007 – 20 September 2007 17

 RC Technical Justification
Reconfigurable Computing Subproject

• Flight-Qualified, Multi-String Redundant Hardware is Expensive

– Development, Integration, IV&V, and Flight Qualification
– Space and Weight
– Power Consumption and Cooling
• Custom Design of Computing Resources for Every New Flight System
or Subsystem is Unnecessary and Wasteful
• Requirements for Flexibility are Increasing and Make Sense
– Reconfigurable (Flexible) and Modular Capabilities
– For Dissimilar Spares, and Incremental Changeover to New Technology:
Capacity to use one system to back up any number of others
– General Reusability
• Current Options for Harsh/Flight Environment Systems are Limited
– Custom Hardware, Firmware, and Software
– Dedicated and Inflexible
– Often Proprietary: Collaboration Inhibited
• Modular Spares == Fewer Flight Spares

SPACE 2007 – 20 September 2007 18

 SiGe Technology

The Moon: A Classic Extreme Environment!

Current Rovers / Robotics
Extreme Temperature Ranges:
+120C to -180C (300C T swings!)
28 day cycles
-230C in shadowed polar craters
Radiation:
100 krad over 10 years
single event effects (SEE)
solar events
Many Different Circuit Needs: Requires “Warm Box”
digital building blocks
analog building blocks
data conversion (ADC/DAC)
RF communications
actuation and control
sensors / sensor interfaces

Highly Mixed-Signal Flavor

SPACE 2007 – 20 September 2007 19
 SiGe-Based Remote Electronics Unit (REU)

The X-33 Remote Health Our Project End Game:

Monitoring Node, The SiGe ETDP Remote
circa 1998 Electronics Unit, circa 2009
(BAE)
REU in
connector
housing!

Analog front Digital

Specifications end die control die

• 5” wide by 3” high by 6.75”

long = 101 cubic inches
Conceptual integrated REU
• 11 kg weight system-on-chip SiGe BiCMOS die

• 17.2 Watts power dissipation Our Goals

• -55oC to +125oC • 1.5” high by 1.5” wide by 0.5”
long = 1.1 cubic inches
• < 1 kg
• < 1-2 Watts
• -180oC to +125oC, rad tolerant!
Supports MANY Sensor Types:
Temperature, Strain, Pressure, Acceleration, Vibration, Heat Flux, Position, etc.
Use This REU as a Remote Vehicle Health Monitoring Node
SPACE 2007 – 20 September 2007 20
 RHESE Summary
• All RHESE tasks are “requirements-
pulled” by specific Constellation
Architecture Requirements Document
(CARD) requirements, Lunar
Architecture Team (LAT) technology
needs, and surface systems’ defined
environments.
• An application-dependent trade space
is defined by:
– Radiation Hardening by Architecture
using COTS processors, and
– Radiation Hardening By Design using
Rad-Hard processors.
– Considerations include performance
requirements, power efficiency, design
complexity, radiation
• Radiation and low temperature
environments currently drive spacecraft
system architectures.
– Centralized systems to keep
electronics warm are costly, weighty
and use excessive cable lengths.
– Mitigation can be achieved by active
SiGe electronics.

SPACE 2007 – 20 September 2007 21

 RHESE Summary

• Radiation Environmental Modeling is

crucial to proper predictive modeling
and electronic response to the
radiation environment.
– When compared to on-orbit data,
CREME96 has been shown to be
inaccurate in predicting the radiation
environment.
– The NEDD bases much of its radiation
environment data on CREME96 output.
• Close coordination and partnership with
DoD radiation-hardened efforts will result
in leveraged - not duplicated or
independently developed - technology
capabilities of:
– Radiation-hardened, reconfigurable FPGA-
based electronics,
– High Performance Processors (NOT
duplication or independent development).

SPACE 2007 – 20 September 2007 22