RR-1RGR 3-1June 2015
Cryocoolers for Space Applications #3
R.G. Ross, Jr.
Jet Propulsion LaboratoryCalifornia Institute of Technology
Topics
� Space Cryocooler Historical Overview andApplications
� Space Cryogenic Cooling System Design andSizing
� Space Cryocooler Performance and How It'sMeasured
� Cryocooler-Specific Application and IntegrationExample: The AIRS Instrument
Copyright 2015 California Institute of Technology. Government sponsorship acknowledged. CL#15-2287
2015 CEC Cryocooler Short Course
RR-2RGR 3-2June 2015
� Cryocooler Technical Performance Data Requirements• Operating needs of typical space detectors
• Space cryocooler technology and reliability challenges
� Thermal Performance Measurements• Example performance & parameter dependencies
• Spatial distribution of power dissipation
� Effect of Pulse Tube Gravity Orientation on Performance
� Generated Vibration and Vibration Suppression Techniques
� Launch Survivability
� Electrical Interface Compatibility• Magnetic and electric fields
• Inrush and reflected ripple current
Topics
Session 3—Space CryocoolerPerformance and How It's Measured
RR-3RGR 3-3June 2015
References
� Ross, R.G., Jr., “Chapter 11: Cryocooler PerformanceCharacterization,” Spacecraft Thermal Control Handbook,Vol. II: Cryogenics, The Aerospace Press, El Segundo, CA,(2003) pp. 217-261. (23 references).
� Ross, R.G., Jr. and Johnson, D.L., “Effect of Heat RejectionConditions on Cryocooler Operational Stability,” Advancesin Cryogenic Engineering, Vol. 43B (1998), pp. 1745-1752.
� Ross, R.G., Jr., Johnson, D.L. and Rodriguez, “Effect ofGravity Orientation on the Thermal Performance of Stirling-type Pulse Tube Cryocoolers,” Cryogenics, Vol. 44, Issue:6-8, June - August, 2004, pp. 403-408.
� Ross, R.G., Jr., “Vibration Suppression of Advanced SpaceCryocoolers — An Overview,” Proceedings of theInternational Society of Optical Engineering (SPIE)Conference, San Diego, CA, March 2-6, 2003.
RR-4RGR 3-4June 2015
� Johnson, D.L., Collins, S.A. and Ross, R.G., Jr., "EMIPerformance of the AIRS Cooler and Electronics," Cryocoolers10, Plenum Publishing Corp., New York, 1999, pp. 771-780.
� Ross, R.G., Jr., “Chapter 6: Refrigeration Systems for AchievingCryogenic Temperatures,” Low Temperature Materials andMechanisms, Y. Bar-Cohen (Ed.), CRC Press, Boca Raton, FL(Scheduled to be published in Nov. 2015). (79 references)
� Ross, R.G., Jr., “Appendix A: Constructing a CryocoolerMultiparameter Plot,” Spacecraft Thermal Control Handbook,Vol. II: Cryogenics, The Aerospace Press, El Segundo, CA,(2003) pp. 605-608.
� http://www2.jpl.nasa.gov/adv_tech/ JPL website with 103 JPLcryocooler references as PDFs (R. Ross, webmaster)
References (Con't)
RR-5RGR 3-5June 2015
• THERMAL PERFORMANCE• Complete parametric thermal performance map including
compressor stroke, expander stroke, coldtip temperature, inputpower, coldtip load, and compressor and expander rejecttemperature
• Compressor and expander heat dissipation fractions andthermal resistances from source to heat sink
• Cooler electronics input power vs compressor input power
• ALLOWABLE HEATSINK TEMPERATURE RANGE
• EMI PERFORMANCE
• Mil Std 461 AC and DC magnetic and electric fields
• Reflected ripple current
• GENERATED VIBRATION (vs axis and suppression systemmode)
• LAUNCH VIBRATION SURVIVABILITY (with interface mass oncold finger; with piston motion suppression?)
Typical Cryocooler PerformanceData Requirements
RR-6RGR 3-6June 2015
Functional Schematic
Cryocooler CalorimetricThermal-Vacuum Test Facility
RR-7RGR 3-7June 2015
JPL Cryocooler Thermal-VacuumCharacterization and Lifetest Chambers
RR-8RGR 3-8June 2015
TRW 1W-35K Pulse Tube Cryocoolerduring Thermal Testing at JPL
RR-9RGR 3-9June 2015
Heat Sink Temperature = 20°C
TRW 1W-35K Pulse Tube Cooler
No-LoadTemperature
Sensitivity of Thermal Performanceto Compressor Stroke
RR-10RGR 3-10June 2015
BAe 50 to 80 K Stirling Cryocooler
Sensitivity of Thermal Performanceto Compressor Stroke
RR-11RGR 3-11June 2015
Sensitivity of Thermal Performanceto Compressor Stroke
%DRIVE
AIRS 55K Pulse Tube Cryocooler
RR-12RGR 3-12June 2015
TRW 1W-35K Pulse Tube Cooler
DriveFrequency
Sensitivity of Thermal Performanceto Drive Frequency
RR-13RGR 3-13June 2015
Sensitivity of Thermal Performanceto Fill Pressure
Stirling Technology 80K Stirling Cooler
FillPressure
RR-14RGR 3-14June 2015
Sensitivity of Thermal Performanceto Displacer Stroke
BAe 50 to 80 K Stirling Cryocooler
DisplacerStroke
RR-15RGR 3-15June 2015
Sensitivity of Thermal Performanceto Heat Sink Temperature
TRW 1W-35K Pulse Tube Cooler
Heat SinkTemp.
RR-16RGR 3-16June 2015
Algorithm for Predicting Effect ofHeatsink Temperature Change
Ref: Ross, R.G., Jr. and Johnson, D.L., “Effect of Heat Rejection Conditions on CryocoolerOperational Stability,” Advances in Cryogenic Engineering, Vol. 43B (1998), pp. 1745-1752.
Based on the empirically derived findings, one can derive thecooling power P¶(T
A, Ð
A) at heatsink temperature T
A and coldend
temperature ÐA and as equal to the cooling power P¶(T
0, Ð
B) at
the baseline heatsink temperature T0 and
coldend temperature ÐB ,
i.e.
P¶(TA, Ð
A) =P¶(T
0, Ð
B); where Ð
B= Ð
A– (T
A¶-¶T
0)¤�
where
TA = Operating heatsink temperature (°C)
ÐA = Operating Coldtip temperature (K)
T0 = Reference heatsink temperature (°C)
ÐB = Effective Coldtip temperature (K) at Ref heatsink temp (T
0)
� = Measured change in heatsink temperature required to shift thecoldend performance by 1 K. � » 5 to 7 for many coolers
RR-17RGR 3-17June 2015
Thermal Performance Plot forDirect Mount to Radiator
CO
OL
ER
HE
AT
DIS
SIP
AT
ION
, w
att
s
RR-18RGR 3-18June 2015
The Carnot Refrigeration Cycleand its Efficiency
COPCooler
= heat absorb ¤ (heat reject - heat absorb)
COP
Cooler = cooling power
¤ input power
COPCarnot
= Tcold
¤(Thot
- Tcold
)
RR-19RGR 3-19June 2015
Sensitivity of %Carnot COPto Compressor Stroke
= 100 × ____________________________
(cooling power @ Tcold
) (Thot
-Tcold
)
Input electrical power × (Tcold
)
%Carnot COP = 100 × ________COPCooler
COPCarnot
TRW 1W-35K Pulse Tube Cooler
RR-20RGR 3-20June 2015
BAe 80 K Stirling Cryocooler
JPL Cryocooler CalorimetricThermal-Vacuum Test Facility
RR-21RGR 3-21June 2015
BAe 80 K Stirling Cryocooler
Stirling Cooler Input Power andThermal Dissipation Characteristics
RR-22RGR 3-22June 2015
BAe 80 K Stirling Cryocooler
Effect of Heatsink Temperatures onHeat Rejection Location
RR-23RGR 3-23June 2015
Effect of Gravity Orientation onPulse Tube Thermal Performance
RR-24RGR 3-24June 2015
Pulse Tube
Cold Tip
Pulse Tube
Cold Tip
Warm End
Warm End
Pulse Tube Temperature Regions vs Construction Method
RR-25RGR 3-25June 2015
Gamma-Ray 80 K Pulse TubePerformance vs Power and Load
For 0° Orientation
•
24 watts/watt at 80K
RR-26RGR 3-26June 2015
•4 WattAddedLoad
Gamma-Ray 80 K Pulse TubeConvective Load vs Angle
RR-27RGR 3-27June 2015
IMAS 55 K Pulse TubePerformance vs Power and Load
For 0° Orientation
•
22 watts/watt at 80K
RR-28RGR 3-28June 2015
IMAS 55 K Pulse TubeConvective Load vs Angle
0.6 WattAddedLoad
•
RR-29RGR 3-29June 2015
Effect of Gravity Orientation onPT Performance: Conclusions
� Key Conclusions:
• When the PT hot end is oriented UP (+/- 80 degrees) the PTperformance is normal (reflects the nominal non-convectionconductivity of the PT)
• When the PT is horizontal or the hot end is tilted down the PTperformance can be impacted by large convection loads internalto the PT.
• The level of convection loads has been found to be a strongfunction of the aspect ratio of the PT geometry. Long-slim PTshave minimal effect, whereas short squat PTs can have verylarge effects
• Gravity Orientation can be an important constraint duringcryogenic system ground testing
RR-30RGR 3-30June 2015
6-DOF Vibration-Force Dynamometer
JPL Exported VibrationCharacterization Facility
RR-31RGR 3-31June 2015
Typical Generated Vibrationfrom Oxford-Style Compressor
RR-32RGR 3-32June 2015
BAe 50-80K Cryocooler
Vibration Force Spectrum forSingle Piston Oxford Cooler
RR-33RGR 3-33June 2015
Vibration Spectrum forIntegral Dual-Piston Cooler
RR-34RGR 3-34June 2015
Cryocooler Drive Electronics
• Adaptive feed forward algorithms used to null measuredacceleration or vibration force by tailoring individual harmonicamplitude and phase on one of the two compressor halves
• Generally implemented digitally in cryocooler driveelectronics, some nulling as many as 16 harmonics
Accelerometeror load cell
Approach to CryocoolerActive Vibration Suppression
RR-35RGR 3-35June 2015
2005 Jly: Suzaku: XRS GSFC ADR cooled by SHe/solid Ne cryostats
Dual Compressor Vibration ForceSpectra with Harmonic Nulling
RR-36RGR 3-36June 2015
Cryocooler-Generated Vibration Conclusions
� Large quantities of exported vibration data have been acquiredon a broad cross-section of Oxford-style coolers. The datareflect a high degree of similarity between machines
� Key Conclusions:
• Head-to-head mounting of coolers can do a good job acancelling the fundamental and 2nd harmonic (100x reduction)
• Higher harmonics are typically not improved with head-to-headmounting unless active vibration suppression is used
• With active vibration suppression, cross-axis harmonicsgenerally create the worst case exported vibration levels
RR-37RGR 3-37June 2015
Fragile Cold Finger
Unconstrained Pistons
Another Challenge —Surviving Launch Vibration
RR-38RGR 3-38June 2015
• REQUIREMENTS
• Random Vibration on the order of 0.16 G2/Hz from 50 to 800 Hz
• Sinusoidal Vibration from 10 to 100 Hz (3 to 8 G, mission specific)
• CHALLENGES
• Most Oxford-style compressors have little trouble passing therandom vibration requirement
• Stirling and PT coldfingers are quite vulnerable to Random Vibe
• The low-frequency sinusoidal environment can be troublesomefor integral back-to-back Oxford-style compressors because oftheir very low frequency piston slosh mode
• The low-frequency sinusoidal environment can also betroublesome for Stirling displacers and counter-balancers
• TEST METHODS
• Typical aerospace vibration test facilities
• Piston/displacer/balancer stroke measurement during test runsvia supplementary electronics
Launch Vibration Requirements,Challenges, and Test Methods
RR-39RGR 3-39June 2015
Typical Space Launch VibrationRequirements (from GEVS)
0.16 g2/Hz
50Hz
Design Limit Loads = Use Mass Acceleration Curve (MAC)Flight Acceptance Levels = 1 minute per axis at (Qual Levels/two)
Protoflight Levels = 1 minutes per axis at Qual LevelsQualification Levels = 2 minutes per axis at Qual Levels
Typically, Lowest Resonant Frequency > 50 Hz (hard for coolers to meet)
Qu
alif
ica
tio
n T
est
Le
ve
l
Most Coolers
RR-40RGR 3-40June 2015
BAe 55 K Cooler UndergoingLaunch Vibration Testing at JPL
RR-41RGR 3-41June 2015
Compressor Stroke during Sine VibeTest vs Coil-Shorting Resistance
Single-Piston Compressor ( BAe 50-80K Cryocooler)
(3-g Sine sweep at 2 Octaves/minute)
RR-42RGR 3-42June 2015
TRW 1W-35K Pulse Tube Cryocooler
Drive Coils
• Inphase piston response isvery high Q and wellcoupled to launchexcitation
• Vibration suppressioninvolves shorting the drivecoils to provideelectrodynamic braking
Piston Slosh Mode
Example Resonant Response ofIntegral Two-Piston Compressor
RR-43RGR 3-43June 2015
Example Cryocooler ColdfingerBumper Assembly
80K
(300K)
RR-44RGR 3-44June 2015
Lead orTungsten Shot
DisplacerBody
CryocoolerColdtip
Factor of 4 Vibration ReductionAchieved with Particle Damper
Example Cryocooler ColdfingerParticle Damper
RR-45RGR 3-45June 2015
Cryocooler Launch Vibration Testing Conclusions
� A significant number of cryocoolers have been tested forrobustness with respect to launch vibration tolerance
� Key Conclusions:
• Most compressors have little difficulty passing typical launchrandom vibration Qual test levels
• However, most coldfingers and pulse tubes are marginal attypical launch random vibration Qual test levels. Most requireadd-on supports (bumper ass'y) or added damping
• Most compressors have difficulty passing typical low-frequencylaunch sine vibration Qual test levels (20 to 40 Hz). Most requireadditional piston restraint such as by shorting motor windings
RR-46RGR 3-46June 2015
Cryocooler EMI Requirements,Challenges, and Test Methods
• REQUIREMENTS
• Magnetic Fields below Mil Std 461C RE01 & 462 RE04
• Electric Fields below Mil Std 461C RE02
• Ripple Currents below Mil Std 461C CE01/03.
• In-Rush Current Limits
• Must pass Susceptibility to External EMI
• CHALLENGES
• Most Oxford-style compressors have very high Magnetic Fieldsat their fundamental operating frequency
• Most Oxford-style compressors have very high Ripple Currentsat twice their fundamental operating frequency
• Inrush currents and electric fields need to be managed withproper circuit design and shielding
• TEST METHODS
• Mil Std 461 in screen room
• Need means (vacuum bonnet) to allow cooler to operate outsideof vacuum chamber
RR-47RGR 3-47June 2015
Low-Frequency AC Magnetic FieldTest Setup with TRW PT Cooler
RR-48RGR 3-48June 2015
Note typical exceedances
Compared with Mil Std 461 RE01 Requirements
Historical Cryocooler CompressorAC Magnetic Field Emissions
RR-49RGR 3-49June 2015
High-Frequency AC Electric FieldTest Setup with TRW PT Cooler
RR-50RGR 3-50June 2015
Early Cryocooler ElectronicsAC Electric Field Emissions
Note typical exceedances
RR-51RGR 3-51June 2015
AIRS Cryocooler ElectronicsConducted Ripple Current
RR-52RGR 3-52June 2015
• Measurement and test techniques for space cryocoolers arequite well developed and documented in the literature
• Thermal performance as a function of drive parameters
• Heat dissipation quantities and locations
• Coldhead gravity effects on performance
• Generated vibration as a function of drive parameters
• Launch vibration robustness
• Generated EMI and Susceptibility to External EMI
• Typical test data are also readily available in the literature
• Means of bringing coolers into conformance with typicalspace requirements are also documented in the literature
Cryocooler Measurement Summary