View
5
Download
0
Category
Preview:
Citation preview
Phase Change CellsT. Shane Topham, Gail Bingham, Harry Latvakoski, Mike Watson
1
Outline• Phase change cell introduction• Summary of ISS phase cell experiments• Flight cell design improvements and ground test results
2
Need for Orbital Temperature Reference• Phase transition cells for absolute temperature reference are
key components of future climate monitoring missions• Mission requirement:
“…an SI-traceable standard for absolute spectrally resolved radiance in the infrared with high accuracy (0.1 K 3σ brightness temperature… Each of the interferometers carry, on-orbit, phase transition cells for absolute temperature,… with SI traceability [1].”
• Because temperature uncertainty will only be one of the contributors to the 0.1 K requirement, absolute temperature uncertainty will need to be lower, on the order of 0.01 K or better
3
Phase Transitions as Thermal ReferencesITS90 Phase Change Materials
4
• Large volume of PCM• Long melt times• Deep re-entrant wells• Fragile containers• Detailed manual heating and cooling
proceduresPractical absolute uncertainty, ~0.2 mK or better [2,3]
T90/K t90/°C Material State Uncertainty mK
234.3146 -38.8344 Hg Triple pt 0.2(0.1)
273.16 0.01 H2O Triple pt 0.05 (0.03)
302.9146 29.7646 Ga Melt pt 0.2 (0.03)
Traditional Triple Point of Water Cell
SDL Phase Change Program History
5
2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
(2004) Initial internal studies of potential for NPOESS started
VEGA Intl. approached to collaborate on gallium eutectic work
(2006) SDL & IBMP agree to perform ISS testing of PCM cells.
SDL starts IR&D program to get ISS tests with PCM cells and patent technology
Vega Results show Ga eutectics as viable PCMs for calibrations
ESTO office begins funding support to speed development to benefit CLARREO with space qualification results
Launch opportunities delayed by Russian Calibration Institute approvals with ties to VEGA
Hardware Launched on Soyuz, ISS experiments conducted
FY1 FY2 FY3 FY4SDL IR&D program
Hardware returned on Soyuz to IBMP and then to SDL
Mini Orbital Temperature Reference (MOTR) Flight 1 Experiment Design for ISS• PCM temperature controlled by
2 TECs and a temperature-controlled heater enclosure
• Heat exhausted to cabin through forced ventilation
• Automated experiment data stored on CF card, removable only on ground
• Expanding bellow PCM container ~1 mL volume
• Re-entrant sensor well
6
Sensor
PCM
TEC
Flight 1 Experiment Data Analysis• 19 on-orbit melt curves• Temperature calibration from
bath post-flight• Multiple similar data sets
obtained prior to and after flight
• Use average in plotted window as “melt curve temperature”
7
29.76
29.762
29.764
29.766
29.768
29.77
29.772
29.774
29.776
29.778
0.5 1.5 2.5 3.5 4.5 5.5
Tem
pera
ture
(˚C
)
Time from start of melt cycle (hrs)
Flight Melt Curves smoothed to 2.2 minutes[4]
MOTR Data Collection Timeline
29.771
29.772
29.773
29.774
29.775
29.776
29.777
29.778
8/10/2010 2/26/2011 9/14/2011 4/1/2012 10/18/2012 5/6/2013 11/22/2013 6/10/2014 12/27/2014
3 hr
Mel
t Tem
pera
ture
Ave
rage
s (C
)
Data Collection Date
SDL Predelivery(1)
SDL Predelivery (2)
Moscow Delivery
Moscow Pre-launch (1)
Moscow Pre-launch (2)
ISS Flight
Moscow Postflight(1)
Moscow Postflight(2)
SDL Postflight
Temp Trend(1)
Temp Trend(2)
SDL Post ReCal
MoscowPre-launch
ISS Flight
Moscow Post-flight
SDL Post-flight
8
No continual drift trends observed over 4 years of melt data, indicating PCM contamination not a factorAll data sets are within 6 mK. Individual point within a given set vary by < ±1.5 mK
Moscow Delivery
SDL Pre-delivery
Summary of All MOTR Data Sets
29.771
29.772
29.773
29.774
29.775
29.776
29.777
29.778
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Mel
t Tem
pera
ture
(°C
)
Melt Sequence Number
All Data Sets
SDL Predelivery(1)
SDL Predelivery (2)
Moscow Delivery
Moscow Pre-launch (1)
Moscow Pre-launch (2)
ISS Flight
Moscow Postflight(1)
Moscow Postflight(2)
SDL Postflight
Temp Trend(1)
Temp Trend(2)
SDL Post ReCal 1
9
ISS Data: Average Melt Temperature = 29.7731 °C Standard Deviation = 0.0008 ° CALL Data: Average Melt Temperature = 29.7737 ° C Standard Deviation = 0.0017 ° C Within ~0.6 mK
Orbital PCM Reference Value Criteria• Stable, reliable PCM containment• Miniaturized assembly• Includes multiple PCMs to improve calibration knowledge • Capable of reliably and repeatably freezing and melting PCMs• Capable of accurately measuring temperature during melting to
within 10 mK absolute error• Capable of tracking blackbody temperature to within a few mK
10
Transfer of Calibration
PCMSensor
TEC
Transfer:When the TEC is not
powered it acts as a thermal link to the thermal surface.
If adequately insulated it will come to equilibrium with the thermal surface.
The PCM sensor can be compared to thermal surface sensors’ readings.
Thermal Surface(What you really want to measure)
Calibration:During a
recalibration the TEC is powered and the PCM is controlled to a different temperature than the thermal surface to melt the PCM.
Temperature data collected during the melt allows recalibration of the PCM sensor.
Orbital 3 PCM Cell Design Phase-change cell with three cavities to be used for on-orbit temperature calibration
12
Thermistor
Thermistor locator (Teflon)
Cell body (SS 304)
Phase-change material ~0.3 mL
Threaded primary cap (SS 304)
Secondary cap (SS 304)
Redundant welds contain PCM
Ga 29.7646°CGaSn 20.46°CGaIn 15.63°C
Tem
pera
ture
, °C
Time, hours
Bath
Cells
PCM Freezing
PCM Melting
Bi-phase Equilibrium Temperature
Gallium-Tin Calibration Bath Data
Bath Bi-phase Equilibria Temperatures
• Gallium bi-phase equilibria were within <1 mK of the ITS90 Ga fixed point at 29.7646• Most eutectic melt temperatures are reported to >10 mK absolute accuracies
13
PCM AvgerageBi-Phase EquilibriumTemp (°C)
Bi-phase Equilibrium Temperature for 3 Individual Cells Tested (°C)
Ga 29.7647 29.764, 29.765, 29.765
GaSn 20.4610 20.460, 20.461, 20.462
GaIn 15.6323 15.631, 15.631, 15.635
SDL Vacuum Chamber Test Setup
Vacuum Chamber
Copper Block PCM Cell
Surface Heater
Circulated Liquid Cooled Base
Block Isolator
Outer Block Thermistor
Outer Rad Shield
Inner Block Thermister
TEC
Inner Rad Shield
Cell Mount Plate
14
• Unpowered PCM Cells equilibrated to within 2.5 mK of Inner Block Thermistors at 24 °C and at 1 °C.• Agrees with thermal model delta of ~2 mK equilibrium
Thermal Model of a Ga Melt Curve
15
Condition: Blackbody at 273.15 KTemp sensor between orange and red nodes
Thermistor will read somewhere between
these cell nodesPiece of the melting gallium
Gallium Gradients vs. Melt Time
16
Gradients in cell and duration of melt vary with heating power applied.
Mid-melt offsets in measurements agree with thermal model (~45-75 mK)
Ga Melt Point Repeatability• TEC heating with
constant current• Copper block held
at constant 8.5 ±0.01 °C
• Melt durations very consistent
• Plateau mid-point repeatability ~2 mK
17
GaIn Melt Point Repeatability• Melt durations
varied by ~13%• Inconsistency
possibly due to small constant current variations in the TEC
• Mid-points group to ~5 mK
• Clear correlation between mid-point temperatures and melt durations
18
GaSn Melt Point Repeatability• Melt durations varied by
~20%• Inconsistency possibly due
to small constant TEC variations
• Mid-points group to ~5 mK• Clear correlation between
mid-point temperatures and melt durations
19
Conclusions• Orbital testing of Ga phase transitions showed no change in melt
point under 0G conditions and verified long-term repeatability of melt curves
• SDL improvements of design include smaller cells and addition of two new melt temperatures for GaSn and GaIn eutectics
• SDL ground testing of cells with Ga and eutectics demonstrates 2-3 mK repeatability of three melt temperatures
• Ground testing and thermal modeling improve understanding of:– thermal gradients within cells and their effect on interpretation of melt
curves– true bi-phase equilibrium temperatures of melt materials
Questions?
20
References1. Committee on Earth Science and Applications
from Space: A community Assessment and Strategy (2007). Earth Science and Applications from Space: National Imperatives for the Next Decade and Beyond, pp. 93-94, ISBN 0-309-66714-3.
2. Fluke Corporation,Hart Scientific Division, “5901 Series Triple Point of Water Cells” 2005, [Revised 10/2006 2527367 D-EN-N Rev B].
3. Mangum B. W. & Furukawa G. T. (Eds.). Guidelines for Realizing the International Temperature Scale of 1990 (ITS-90), NIST Technical Note 1265, August 1990.
4. Topham T.S., Bingham G.E., Latvakoski H., etal., Observational study: microgravity testing of a phase-change reference on the International Space Station, npj Microgravity 1, Article number 15009 (2015), doi:10.1038/npjmgrav.2015.9
5. GE Sensing, “Thermometric Ultrastable Probe Thermistors,” SP series Datasheet, [copyright 2006].
6. Heraeus, “Platinum Resistance Temperature Detector,” M 222 Datasheet, document name 30910015 Index B, [June 2010].
21
Additional Slides
Sealed Cells vs. Pressure Dependence of Fixed-Point• For contamination issues PCM containers must be sealed• 1 atm pressure changes melt temperature of water by 10 mK [3]• Container must allow PCM expansion without changing fixed-
point temperature
Flexible container:
-No internal voids-PCM can expand container-PCM vacuum filled-complex filling-complex container-moving parts
Rigid container:
-PCM filled at 1 atm-Internal gas voids compress as PCM expands-location of voids in space?
SDL Temperature Sensor Testing• Heraeus PRT and GE
thermistor excellent size and long term stability [5,6]
• GE Thermistors tracked standards PRT ±3mK, with calibration improvement to ~1mK
• Heraeus PRTs tracked ±10-15mK (worse than larger wire PRTs)
• Heraeus shock resistance 40g at 10-2kHz
Bath Cycling of 4 SP60 Thermistors
(mK)
Drift 0.04% 1000hrs 2.1 x 2.3 mm
Heraeus M222 PRT 1.5 mm
3.2 mmGE SP60 Thermistor
Drift 0.02% /yr
Ga Melts of Various Lengths
1-2 hour melt curves are ~50-60 mK higher than melt point at center of melt due to cell thermal gradientsAgrees with thermal models which predict ~50-80mK. Longer melts within 5 mK of true melt temperature
25
GaSn Melts of Various Lengths
26
1-2 hour melts are only ~30-40 mK above eutectic melt point
GaIn Melts of Various Lengths
27
1 hour melts are 50 mK above eutectic melt point>5 hour melt was within 5 mK of eutectic melt point when it was stopped
Supercooling Behaviors
Ga and eutectics cool below their melt points after reaching temperatures well above as liquids.
28
GaIn Behavior Improvements• Tested in bath at
slower ramp rates
• Ga rich mixture shows slightly less supercooling
29
Acknowledgements• Russian Academy of Sciences IBMP for launch and flight
support• VNIIOFI (Andrey Burdakin) for Gallium eutectic
investigations and preparation training• NASA Earth Science and Technology Office (ESTO) for
funding support• NASA Langley’s CLARREO team for CORSAIR work
Questions?
30
Recommended