small stirling technology exploration power for future

National Aeronautics and Space Administration

Small Stirling Technology Exploration Power for Future Space Science Missions

Scott Wilson, Nicholas Schifer, NASA Glenn Research Center

March 4, 20192019 IEEE Aerospace Conference

2019 IEEE Aerospace Conference

National Aeronautics and Space AdministrationNational Aeronautics and Space Administration

Michael Casciani, Vantage Partners, LLC.


Small nuclear power systems could provide electricity to power probes, landers, rovers, or communication repeaters for space science and exploration missions• Would convert heat to electricity for powering spacecraft sensors and communication

• Fractional GPHS (General Purpose Heat Source) output: ~60 watts thermal output (good for 10-Watt Generator)

• LWRHU (Light Weight Radioisotope Heater Unit, often called RHU) ~1 watt thermal output for each heater unit

Low Power RPS for Small Spacecraft

GRC Development Goals• Sufficient power for small spacecraft functions

• Long-life and low degradation to ensure sufficient power at EOM

• Robust to critical environments (vibration, shock, constantacceleration, radiation, etc.)

• Thermal capability and high efficiency

• Operates in vacuum or atmosphere

Dynamic Power Conversion System Efficiency• 1 watt generator efficiency projected to be 12-13%

• 10 watt generator efficiency projected to be 16% Conceptualization of Seismic MonitoringStations Being Deployed from Rover

JPL Pub 04-10, Sept-2004: 51 mission concepts, 14 RHU-based, 13 fractional-GPH

2


• Subassemblies under development include convertor, controller, and insulation

• Need to formalize requirements later this year (Fundamental Research Project to help)

• Goals are based on existing DPC requirements

GRC 1-W Dynamic RPS Concept - Goals

Category Goals Current Best Estimate

Design life 20 years 20 years

Heat input 7 to 8 watts to convertor 8 watts

Power output At least 1 We DC from controller > 1 We DC

Heat source surface temperature TBD < 450 ºC

Stirling hot-side temperature 325 to 375 ºC 350 ºC

Stirling cold-side temperature -100 to 50 ºC -100 to 50 ºC

Robustness Overstroke tolerant

Random vibration level TBD

Environment vacuum or atmosphere vacuum and atmosphere

Constant acceleration 20 g 19 g

3


In-house dynamic RPS from 1-10 We DC power output• Demonstrate practicality of 1 watt power level by maturing subassemblies

and interfaces

• Complete engine design at 10 watts power output (62 Wth heat source)

Initial Demonstration • Controller breadboard• Free-piston Stirling convertor• Multi-layered metal foil insulation (using Stirling thermal simulator)

Increased Fidelity • Controller brassboard fabrication and test• Stirling convertor durability testing

System Testing • Stirling convertor + controller• Electrically heated prototype system (includes insulation)

Technology Development at GRC

4

Pneumatic Testing

Controller Testing

Insulation Testing (at vendor)


GRC 1-W Dynamic RPS Concept

Electrical Controller

Stirling Engine

Multi-Layer Metal Insulation

Heat Source

Linear Alternator

Stirling Convertor

Radiative Coupling Heat Rejection Flange

5

in-line concept for Stirling convertor not yet complete

Table 1: smal/STEP Concept Characteristics ,--------------------------------------------------1 ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' [ _ - - - - - - - - - - - - ----- - - - - - - - - - - - - --- - - - - - - - - - - - - ----- I

Item smallSTEP

Nuclear Fuel 8 LWRHU

Thermal Inventory 8 watts (thermal) Beginning of Life ,(SOL) -when the RPS is fueled

Beginning of Mission 1 watt (BOM) Power - up to 3

vears after fuelinq End Of Design Life 0.83 watts

(EODL) Power - 14 years after BOM

Power Conversion Stirling Converter

Environment Multi-Miss ion Capable

Voltage Range 4 to 5 V DC

Degradation Rate 1.2%/yr

Efficiency 13%

Mass (including 6 pounds (3 controller) kilograms)

Volume 4.3 inch (11 centimeter) X 12.5 inch (32

centimeter) cube


Light Weight Radioisotope Heater Unit (LWRHU)• Long history of use on many space missions for

heating spacecraft electronics

• Aeroshell designed to survive reentry into Earth’s atmosphere for safety

• Diameter: 1.0 inch, Length: 1.3 inch (1.1 watts of thermal energy at BOL)

Generator concept uses 8x LWRHU• 8 Wth Heat to 1 We DC electrical power

GRC testing will use electric heaters to simulate the LWRHUs• Designed to provide similar thermal gradients

compared to LWRHU

• There are four resistance cartridge heater total, each one simulates two LWRHUs

Heat Source

LWRHU Assembly

LWRHU Simulator uses electric heaters

Welded wire bus

LWRHU volume

electricheater

cross-section view Test Hardware

Graphite block

6

Generator concept uses 8x LWRHU

o • •


Objectives • High performance is critical to minimize losses

• Peregrine Falcon Corp. provided Multi-Layered Metal Insulation (MLMI) package in October 2018

• MLI design considerations

• 8 watts thermal input, targeted >90% efficient

• Evacuated to enable low thermal losses

• Lab unit allows disassembly for inspection

• Thermal simulator design considerations

• Reject 7 watts

• Radiation coupling to heat source

• Instrumented with TCs along the length to calculate heat transfer, also enables model comparison

• Efficiency is flat with achievable emissivity values

Insulation Design

7

MLI Package

Thermal Analysis Study

Thermal Simulator Thermal SimulatorFabricated

J ~O

,,.

2,0

liO

.,

1.)

f' 1.0

-~ 0.8

~ 0.6 C

~ OA ;;; C 0 ,2

00

111

00

0

•

-300

Varying Fm issivity

rel;;ati'-'fl'Yfl~

~ ·~- .. Harder

0.0~ 0 .1

• • Easler

0 .15

Environm~nt al Te mperature, C

Varyi ne [nvironme nta l Tem perat ure

deep space

• e~rth

02

• Lunar cr.u@r

Loss:&s: inue:;;as;e as eteterior t&mperature decre.ises

-200 -1 00 1 00

Env1ronm e-ntal Temperature. c


As-is• Radiative coupling

• No additional layers

• Heat source temperature = 643 C

• Stirling hot-end temperature = 450 C

Minor modification• Compliant thermal interface

• Heat source temperature = 450 C

• Stirling hot-end temperature = 450 C

• Appears it could work for a 10-W DRPS with minor modification

Exploratory cases for 10-W DRPS

8

Thermal Simulator

• 643 450

600

400

500 350

300

400

250

300

200

200 150

100

100

so "' 50

Insulation Evolution

9

0.83 watts Hot RHU Surlace 500 C

Consistent Emissivities for 20 Year Life of Generator

Multilayer Insulation Overall Effective Thermal Conductivity-0.001 W/m-K

Sink Temperatures: Microporous Insulation Case A: -270 C (Space) Case B: -50 C (Cold Mars) Case C: +50 C (Hot Mars)

figure 1: MLMI in Cross Section

Coocept Only

Analysis needed for Deflnitizatkm

Cartridge

Heaters

Ef fective

Thermal

Conductivity

... 0.001

W / mK

Case B : -5()"( (Cold Mars)

Case C: •SO"C (Hot Mar5)

10-LAYERED PAIR ML! PEREGR!Nf COt.F•GURA r•O,\ '

RHU HOLDER '

l 0

400•c.-·e1 _.._ .. .....

Elcure 1; MLMI ln CrOH Section

-~--


Convertor ModelingConfidence in Predictions• 1D Sage vs. 3D CFD

• Modeled domain was truncated at the piston face and displacer rod (no seals, no bounce space, no displacer gas or radiation)

• Sage connects fixed temperatures directly to ends of the displacer, which artificially elevates displacer temps and associated axial parasitic heat transfer losses, while Fluent model resolves complex thermal and fluid flow fields

• Sage assumes no motion by the displacer when resolving heat flux while Fluent resolves temperature gradients and heat flux by moving components and deforming gas volume meshes

Codes agree well • The PV power output agree within 2%

• Indicated power predicted at 1.5 watts

10

Tcmpora1uro 625.0 609.2 500 4 577.6 581.8 548,1 530,3 514,5 498.7 482,9 487. 1 451 .3 435.5 419.7 403,9 388.2 372.4 3~.6 3'10.8 325.0

IKJ

875,000

850,000

825,000

_800,000 &. - 775,000 ~ ~ 750,000 ~

<>- 725,000

700,000

675,000

650,000 0.00

~ Piston Face

Gas Duct

Rejector Displacer Acceptor

I - PP -> Bl (Fluent) - CSD ·> BL (Fluent) - ESD ·> BL {Fluent} --- PP-> BL (Sage) --- CSD --> BL (S.igc) --- ESD -> BL {Sage)

0.10 0.20 0.30 0.40 0.50

Volume ( cm3 }


Objectives• Rectify AC power for load focused on controller design with generic load• Provide 1 We DC on 5 Vdc bus for rechargeable battery system and sensors• Charging battery enables house keeping power and periodic burst transmission of

measured data for communication

11

Controller Design

• Keep engine at a constant power level

• Shunt excess power when battery is fully charged and power is not required by the load

Stirlin,g converter

sense

AC to DC rectification

House keeping supply

Shunt

Battery

Transmitter

Load power regulation

RS422 to Controller vehicle or

data in GSE

Microprocessor

Load Sensors

c:::::J Cont roller hardware

c:::::J Load hardware

*optional ci rcuit if it proves to be energy efficient


Initial testing met the power output goals• Demonstrated linear AC regulator controller using a MOSFET H-bridge • Analog circuit controlling FETs for AC to DC rectification and alternator current control to improve

power factor • Load voltage monitoring allows for load control and shunting of unused power

12

Controller Breadboard Testing

Value Ideal diode rectifier

Wave form smoothing

Alternator Voltage, Vp-p

25.3 25.4

Alternator Power, We 1.37 1.34

Controller voltage, Vdc 11.7 11.5

Controller Power, We 1.11 1.22

Controller efficiency 80% 91%

CH3 5. OOV/d iv Position - 1.00 div

1: 111 / . ', t~I\

Output voltage

I

Estimated alternator

voltage

'v

Marn:10,111.

AcqModo 1 ; i Norr.tHl -500kS/s ?~s/rl Iv

[--l'ILE S<1ve

I • Filn I ist

-~ I

Dutil Type

-------------- ;..---··--=-""-=-- -~ ~~- -- ----- - -- ------- 4 rormdt

pp Avg

,R~,--, ' Slopped

Alternator voltage

l. t'.l~l°I

Cll1 0 .620A CH9 -' ,, .,. 6.51622V Cll:l ,, r, .,,.-. 7 .13393V

pp Avg Out y

.,-------

:Cll3 :CHI O :CII?

Alternator current

20 .18V 163.7 10nA 47 80X

Av11 :Cll4 RMS :CHI

3 .58%4nV 193 .14 3nA

Frame

l:xecute Save

3?8? -----------""' Fd~ Cll1"'.f"-------------•,!t:Fil•;;------2018/09/11 15: 30:23. 80529113 Auto 0.020A 2018/09/11 15:31 :01


Component Testing

13

Flexure Testing • FEA predicted static stress @deflection, corresponding stress noted for each

amplitude to develop S-N curve (fatigue stress vs. number of cycles to failure)

• Both flexure designs easily exceeded 10 million cycles operating at 100 Hz, high end of typical range used as threshold for identifying transition from high cycle fatigue to infinite life. The boundary between the finite-life and infinite-life regimes, or endurance limit, lies somewhere between 1 million and 10 million cycles for steels

• Displacer and piston flexures demonstrated over 100 million cycles with margin over the nominal amplitude without fracture

• Sufficient confidence our test effort will be failure free

Alternator Testing • Successfully demonstrated (2x) styles of 1-watt moving coil linear alternators,

(motored up to ~4 mm amplitude)

• Successfully demonstrated conduction path across flexure stacks and connections

• Demonstrated non-contacting operation of power piston

• Completed alternator design in concert with controller design in order to tune the motor inductance and ensure flight rated components (including capacitors)

Sandvik 7c27mo2 1095 spring steel

1400

1200

1000

i_ ,oo

j 600

400

200

HardUmll Amplitude:,3:m::;m;---------- I Nomlnal Amplitude, 2 mm

0 1.E+OO 1.E+02 1.E+06

1.E+0,4 des to F.ailure Number of Cy

1.E+08 1.E•10

National Aeronautics and Space Administration 14

Convertor DescriptionDesign Description

• Split-Stirling, gas duct between engine and alternator compression space

• Moving coil alternator

• Gap regenerator

• Flexure bearings for piston and displacer

• Laboratory design emphasized flexibility (not low mass)

Parameters• Hot-end temperature, 350 ºC

• Cold-end temperature, 0-50 ºC

• Operating frequency, 100 Hz

• Operating pressure, 110 psig

• Pressure amplitude, 13 psig

• Piston amplitude, 4.5 mm

• Displacer amplitude, 2 mm

Lab Test Setup(insulation not shown)

Alternator

Cartridge heater

Heat rejection loop

Gas duct

Engine heater head


Measurements• Hot and cold end thermocouples (8x)

• Dynamic CS pressure transducer (1x)

• Mean pressure transducer (1x)

• Hall effect sensors (2x)

• Power meter (power, voltage, current) for electrical heat input and alternator power output

Fabrication and Testing• Fabrication is complete

• Assembly is complete

• Pneumatic testing is complete

• Motor tests have started, some iterative changes are in progress

Convertor InstrumentationHot-end Temperature

Cold-end Temperature

Dynamic Pressure

Mean Pressure

Hall Effect Sensor

Hall Effect Sensor

Electric power input

Electric power output

15

-I -

I I

\ , I , -~

' I ' I ' 't,f !

' I ~

I I ' w - I , I

[ ~ ' ' I ' I - -

L _j . .

National Aeronautics and Space Administration 16


2019 NETS (Y. Lee, et al.)• JPL’s Small RPS A-Team Study focused on mission

concepts for (1 mWe to 40 We)

• Concepts were generated and reviewed as part of a JPL Innovation Foundry Architecture Team (A-Team) study

• Provided recommendations to RPS Program development activities

• Power, Mass, and Size (of interest for us)

• 1 to 10 watts

• 2.5 to 4 kg

• 3000 cm3 to 4000 cm3 (3U to 4U)

• Best fit for GRC 1-W DRPS1. Lunar Geophysical Network (10 We)

2. Pluto Lander (10 We)

3. Magnetosphere Study Fleet (1.5 We)

An Exploration of Mission Concepts That Could Utilize Small RPS

17

Mission Concepts Power Spectrum

Lmw. 1omw. 100mW. 10w. 40W.

Mission Concepts Mass Spectrum

0.5 kg 1kg 1.5 kg 2 kg 3 kg 3.5 kg 4 kg 4.5 kg

1000 cm• (•1u •) 6000 cm• ("6u• i 12000 cm:1r12U")


Conclusion• Small RPS concepts are being studied by NASA for use on small spacecraft

• GRC is working toward demonstration of a 1-watt dynamic RPS• Testing is in progress on all subassemblies (convertor, controller, insulation)

• Completed controller breadboard design and test, Brassboard is next

• Completed convertor assembly and alignment of moving components, 1st operation is imminent

• Completed insulation fabrication, test setup in progress

• GRC is also working on a paper design of 10-watt dynamic RPS

Small Stirling Technology Exploration Power or “smallSTEP”

18

----- - - -


Special thanks to:• RPS Program and Project Management• 1W Stirling Team

• Scott Wilson (Technical Lead)• Nick Schifer (Convertor Fabrication & Test)• Steve Geng (Alternator Design & Test)• Mike Casciani (Controller Design & Test)• Daniel Goodell (Thermal Management Design & Test)• Terry Reid (Advanced Modeling)• Barry Penswick (Engine Modeling & Design)• Paul Schmitz (Requirements)• Roy Tew (Sage Analysis)

Presenter

Presentation Notes

Takes a big team to make a small Stirling!

small stirling technology exploration power for future

Documents