Final Report Phoenix MECA TECP Humidity Sensor - Flight Model Calibration

1
Final Report
Phoenix MECA TECP Humidity Sensor - Flight Model Calibration
Characterization and calibration of the Phoenix MECA TECP Humidity
Sensor in a Mars Atmospheric Simulation Chamber
S. E. Wood1, M. A. Schneider2, G. Cardell3, C. Knowlen4, A. P. Bruckner4,
D. C. Catling1,5, M. Hecht3, D. Cobos6, and A. Zent7
1
Department of Atmospheric Sciences, University of Washington, Seattle, WA, USA
Aerojet, Redmond, WA, USA
3
Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA
4
Department of Aeronautics and Astronautics, University of Washington, Seattle, WA, USA
5
Department of Earth Sciences, University of Bristol, Bristol, UK
6
Decagon Devices Inc., Pullman, WA, USA
7
Space Sciences Division, NASA Ames Research Center, Moffett Field, CA, USA
2
2
Table of Contents
Final Report ...................................................................................................................................1
Phoenix MECA TECP Humidity Sensor - Flight Model Calibration ...........................................1
Table of Contents...........................................................................................................................2
Summary ..................................................................................................................................3
1.
Introduction...........................................................................................................................4
1.1.
Background ........................................................................................................................4
2.
TECP Humidity Sensor (TECP-HS).....................................................................................5
2.1.
Description.........................................................................................................................5
3.
Mars Atmospheric Simulation Facility (MASF) ..................................................................6
3.1.
MASF Design & Operation ...............................................................................................6
4.
TECP Calibration Chamber (TCC).......................................................................................7
5.
TECP-HS Flight Model (FM) Calibration..........................................................................11
5.1.
Procedure .........................................................................................................................11
5.2.
Results..............................................................................................................................11
References....................................................................................................................................14
Appendix......................................................................................................................................15
1.
Preparation & Set-up ..........................................................................................................15
1.1
Clean Calibration Chamber Components (to reduce outgassing)....................................15
1.2
TECP Mounting Rack Pre-Assembly ..............................................................................15
1.3
TECP Calibration Chamber Assembly ............................................................................15
1.3.1
Flanges C & D ..............................................................................................................15
1.3.2
Flanges A & B ..............................................................................................................15
1.4
TECP Mounting Rack Assembly (continued from 1.1)...................................................16
1.4.1
Assemble Base Plate .....................................................................................................16
1.4.2
Preventing Electrostatic Discharge (ESD)....................................................................16
1.4.3
Mounting TECP(s)........................................................................................................16
1.5
Test TECP data signal connections and control system (Labview).................................17
1.6
Complete Assembly of Calibration Chamber, Connection to MASC ............................17
2
Calibration ..........................................................................................................................17
3
Summary
The Thermal and Electrical Conductivity Probe (TECP) is an instrument in the MECA payload
[1] on the 2007 Phoenix Mars Scout mission [2], mounted on the “wrist” of the Robotic Arm. In
addition to making thermophysical measurements of surface and subsurface materials, the TECP
also contains a capacitive relative humidity sensor, referred to herein as the TECP Humidity
Sensor (TECP-HS). We have performed extensive testing of the TECP-HS at the University of
Washington Mars Atmospheric Simulation Facility (MASF) in order to characterize and
calibrate its performance under Martian environmental conditions of pressure, temperature, and
humidity.
The sensing element of the TECP-HS is a Panametrics MiniCap-2, a commercial capacitive
polymer-based sensor whose performance at temperatures and frost-points below –40 C was not
well known. It is mounted on the TECP electronics board, directly beneath a hole in the TECP
housing which is covered by a permeable Teflon membrane to allow air in while keeping dust
out. The capacitance is proportional to the relative humidity of the air in contact with the sensor.
By assuming that the temperature of this air is equal to the temperature of the electronics board
(Tb), which is monitored with a calibrated thermistor, the absolute humidity (or water vapor
concentration) can be determined.
For testing the TECP-HS sensor in the UW MASF, we built a custom test chamber - the TECP
Calibration Chamber (TCC) - designed to hold four TECPs to perform simultaneous calibration
of the four Flight Models (FMs). The TCC was constructed using stainless steel and ceramic
high-vacuum fittings to minimize adsorption and outgassing of water vapor.
We measured the frost-point temperature (Tf), or absolute humidity, of the air flowing through th
TCC using a Buck CR-1 chilled-mirror hygrometer, which uses liquid nitrogen for mirror
cooling and can measure Tf over the range –90C to +30C with an accuracy of ±0.3C. For the
TECP-HS FM calibration measurements, we also employed an EdgeTech DewPrime I chilledmirror hygrometer. Both hygrometers have a NIST-traceable calibration for Tf ≥ –75C.
The calibration began on March 25, 2006, when the TECP-HS FMs were placed inside the
calibration chamber, and was completed on March 30, 2006. During this period, we collected
more than 50,000 measurements, at 5 to 30 second intervals, covering a wide range of Tf (-80C
to -10C) and Tb (-65C to +30C). Most of the data were collected with the chamber containing
CO2 gas at a pressure of ~5 Torr (7mb), flowing through at ~200 sccm. A calibration curve for 3
of the 4 TECP-HS FM’s was calculated as a six-parameter quadratic function of relative
humidity and temperature. TECP FM#10 stopped functioning on March 28 and therefore was not
calibrated.
4
1. Introduction
1.1.
Background
The Phoenix Mars Lander will be launched in August 2007 for a May 2008 landing in the north
polar region of Mars (latitude 65-72N) during the northern summer season (Ls 75-125).
According to Viking Orbiter MAWD [4] and Mars Global Surveyor TES observations [5], this is
the location and time of year (Fig. 1) when some of the highest amounts of water vapor are found
in the Martian atmosphere due to heating and sublimation of the North polar ice cap – ranging
from 25-75 precipitable microns [6], versus the global annual average of ~10 pr. microns. This
landing site is also considered a likely place to find near-surface ground ice based on the high
concentrations of hydrogen detected by the Mars Odyssey Neutron Spectrometer [7,8,9], as well
as the predictions of thermal-diffusion models [10,11,12].
One of the principal themes of NASA’s Mars Exploration Program is to “follow the water” understanding the past and present distribution and behavior of water on Mars. Water-related
investigations are listed as the highest priority for several of the Mars science goals and
objectives in the MEPAG-2006 report [13] including: ‘establish the current distribution of water
in all its forms’ and ‘...model processes that have caused water to move from one reservoir to
another’ (Goal I, Objective A), ‘determine the processes controlling the present distribution of
water, CO2, and dust by determining the daily and seasonal trends’ (Goal II, Obj. A), and
‘determine the present state, 3-D distribution, and cycling of water on Mars’ (Goal III, Obj. A).
The TECP humidity measurements will be a key part of the multiple, complimentary datasets
obtained by the TECP and other MECA instruments for addressing these questions by looking
for correlations of the humidity with surface and subsurface temperatures, wind, clouds or fog
seen with the LIDAR, and/or regional temperatures and weather patterns observed from orbit.
Figure 1
5
2. TECP Humidity Sensor (TECP-HS)
2.1.
Description
The sensing element of the TECP Humidity Sensor (TECPHS) is a GE Panametrics MiniCap-2 (Fig. 2), a commercial
capacitive polymer-based sensor whose performance at
temperatures and frost-points below –40 C was not well
known. There was also concern that its response time would be
too slow to track diurnal changes. It is mounted on the TECP
electronics board, directly beneath a hole in the TECP housing
which is covered by a permeable Teflon membrane (Fig. 3) to
allow air in while keeping dust out. The capacitance is
proportional to the relative humidity of the air in contact with
the sensor, which may be significantly warmer than the outside
ambient air, but should be very close to the electronics board
(Tb) which is monitored with a calibrated thermistor. From RH
and Tb we can calculate the absolute humidity, and thereby
determine the RH of the ambient atmosphere as accurately as
its temperature (Ta) can be measured.
The humidity circuit delivered by Decagon Devices, Inc., is
nearly identical to the example circuit in the MiniCap 2
datasheet. This circuit modulates the duty cycle of a
rectangular wave as a function of sensor capacitance (Fig. 4).
The circuit operation is as follows:
1. A short pulse drives the SET input of a flip-flop and the
output goes high.
2. The output of the flip-flop charges a capacitor
– the humidity sensor.
3. When the voltage across the capacitor reaches
the flip-flop RESET threshold, the output goes
low and discharges the capacitor (until the next
trigger pulse.)
4. The capacitance CRH sets the duty cycle of the
output rectangular wave Vout since higher
values of capacitance require a longer charging
time to reach the RESET threshold, creating a
larger duty cycle. The humidity measurement
is the low-pass filtered – average – voltage of
the output rectangular wave VRH.
Figure 3
Figure 2
6
Figure 4
3. Mars Atmospheric Simulation Facility (MASF)
3.1.
MASF Design & Operation
This facility provides independent control of humidity, temperature, and pressure, and also
simulates low wind speeds (Schneider 2001). The facility is an open, continuous flow system,
which uses CO2 from a standard high pressure bottle as the atmospheric simulant. The CO2 first
passes through a pressure regulator to reduce its pressure to about 0.1 MPa, and is then directed
to the humidity generator, which contains a saturator, a dryer, and two MKS Instruments mass
flow controllers which regulate the relative mass flows of the dry and wet gas streams, and
further reduce the flow pressure to ~20 kPa.
The desired humidity is set by mixing the wet and dry CO2 flows in the appropriate ratio. The
frost/dew points of the saturated and dry flows are controlled by their temperature. In the
saturator, the CO2 flow passes over liquid water or ice, whose temperature is maintained within
±0.1 C by a Neslab RTE-4 circulating cooler. The dry flow is generated by passing CO2 through
a heat exchanger held at -78.5 C by immersion in a slurry of dry ice and ethanol. Since the
atmospheric pressure on Mars is only ~600-1000 Pa, the flow is expanded to this pressure range
by means of a sonic orifice. This expansion also decreases the concentration of water vapor to its
final desired value.
7
The conditioned flow enters the test chamber located inside a low-temperature freezer (Environmental Equipment So-Low C-85-5) that maintains the desired Mars ambient temperature,
controlled via LabVIEW software. After exiting the test chamber the gas passes through a Buck
Research CR-1 chilled mirror hygrometer (B), which measures the frost point of the flow (Busen
and Buck, 1995). Finally, the flow passes through a -78.5 C cold trap and into the vacuum pump.
All interconnecting tubing (12.7 mm OD) and fittings are made of electro-polished stainless steel
to minimize uncontrolled adsorption or desorption of water from internal surfaces.
The MASF is a flow-through system, so the unavoidable hygroscopic materials such as the
insulated lead wires were placed upstream of the TECP’s to minimize any humidity changes
between the TECP’s and the reference hygrometer. The MASF also has a bypass line to monitor
the humidity of the air before it enters the test chamber.
4. TECP Calibration Chamber (TCC)
8
For testing the TECP RH sensor in the UW Mars Atmospheric Simulation Facility (MASF), we
built a custom test chamber designed to hold four TECP’s for simultaneous calibration of the 4
Flight Units – using stainless steel and ceramic high-vacuum fittings to minimize adsorption and
outgassing of water vapor
9
10
11
5. TECP-HS Flight Model (FM) Calibration
5.1.
Procedure
(see Appendix)
5.2.
Results
During calibration of the four TECP Flight Units, we collected more than 50,000 measurements,
at 5 to 30 second intervals, covering a wide range of Tf (-80C to -10C) and Tb (-65C to +30C).
Most of the data were collected with the chamber containing CO2 gas at a pressure of ~5 Torr
(7mb), flowing through at ~200 sccm. An example of the raw data from one set of measurements
is shown in Fig. 4, which also illustrates the unexpectedly fast response time of the sensor to
humidity changes (< 10 min).
The best-fit response function for the TECP FM#04 (Fig. 5) is a 6 parameter quadratic function
of relative humidity and temperature, and has a standard deviation of 4.5 DN. At -60C the
maximum value of the response function is only 2900, so this standard deviation represents a 5%
error. However at -40C the increased range of response reduces the standard deviation to about
2%. So it will be important to consider the operating temperature of the sensor on Mars, and
ways to keep it as warm as possible.
Table 1
DN = A + B*RH + C*TB + D*RH2 + E*TB2 + F*RH*TB
Coefficients
A
B
C
D
E
F
FM #04
+2820.17
+673.607
-1.25101
-59.9217
-0.0174427
+8.84307
Std. Dev. (DN) 4.45
FM #11
+2811.72
+628.581
-1.22748
-71.6799
-0.0163627
+7.95581
FM #12
+2828.53
+768.985
-1.27763
-119.68
-0.0155112
+10.2918
4.65
4.93
where DN is the raw data output from the TECP-HS analog-to-digital converter, and TB is
calibrated board temperature in °C.
While the form of the calibration function given in Table 1 was suited for finding the best-fit
values of the calibration function parameters, it is not the most useful form for interpreting data
generated by the TECP-HS during operation, i.e. determining the RH from a given DN and TB.
We can obtain a more useful form of the calibration function by rewriting it as a quadratic in RH
and using the quadratic theorem:
12
DN = D*RH2 + (B + F*TB)*RH + (A + C*TB + E*TB2)
DN = X*RH2 + Y*RH + Z
X=D
Y = B + F*TB
Z = A + C*TB + E*TB2
RH = [-Y + sqrt(Y2 – 4*X*Z)] / [2*X]
if RH is real with a value between 0 and 1
Although there are always 2 roots of a quadratic, we found that using the “+” sign in front of the
square root term will produce the correct value of RH. If the RH value obtained is imaginary,
negative, or greater than 1, then one of the input values (DN or TB) must be outside of their valid
range.
13
Fig. 3: Plots of the primary measured variables for the entire calibration run (A) and “zoomedin” section of day 5 (B).The color code for each subplot is given in the title, and the light-gray
shading indicates which data points were usedfor calibration. The dark-gray “bar-code” at the
bottom of each plot indicates the status of several data quality flags used to select the final
calibration dataset. The calibration began on March 25, 2006, when the TECP FMs were placed
inside the calibration chamber, and was completed on March 30, 2006.During this period, we
collected more than 50,000 measurements, at 5 to 30 second intervals, covering a wide range of
Tf (-80C to -10C) and Tb (-65C to +30C). Most of the data were collected with the chamber
containing CO2 gas at a pressure of ~5 Torr (7mb), flowing through at ~200 sccm. A calibration
curve for each TECP FM was calculated as a six-parameter quadratic function of relative
humidity and temperature (see Table 1).
14
References
Boynton W. V. et al. (2002) Science 297, 81-85.
Busen, R., and Buck, A.L., “A High-Performance Hygrometer for Aircraft Use: Description,
Installation, and Flight Data,” J. Atmos .& Oceanic. Tech. 12, 73-84 (1995).
Hecht M. H. et al. (2003) AGU Fall 2003, Abstract #P41B-0404.
Jakosky B. M. and Farmer C. B. (1982) JGR 87, 2999-3019.
Mellon M. T. et al. (2004) Icarus 169, 324-340.
MEPAG (2006) http://mepag.jpl.nasa.gov/reports/
Mitrofanov I. G. et al. (2002) Science 297, 78-81.
Paige D. A. (1992) Nature 356, 43-45.
Schneider M. A. and Bruckner A. P. (2003) Space Tech. and App. Int. Forum – STAIF 2003, ed.
M.S. El-Genk, 1124-1132.
Schorghofer N. and Aharonson O. (2005) JGR 110, E05003.
Smith M. D. (2002) JGR 107, 5115.
Tamppari L. K. et al. (2006) 2nd Wrkshp Mars Atm. Model. & Obs., Granada, Spain, 212. [7]
Feldman W. C. et al. (2004) JGR 109, E09006.
15
Appendix
1.
•
1.1
Preparation & Set-up
Clean Calibration Chamber Components (to reduce outgassing)
Wash with an alkaline detergent in deionized water to remove gross contamination,
machining oils, etc.
• Rinse in deionized water
• · Air dry
•
•
•
•
1.2
1.3
•
•
•
•
•
•
•
•
•
•
TECP Mounting Rack Pre-Assembly
Attach SS-SCA-C2 tapped screw clamps to each of the four 1/8"D x 2"L SS Rods (R8-2) as
shown in Fig. 1.
Place R8-2 rods through Base Plate (BP) holes #1, #5, #7, and #11 so that the screw clamps
rest on the top side of BP (Fig. 2).
Lay down Base Plate on a flat surface so that the R8-2 rods point upward (taking care not to
let the rods slide out of the BP holes as you turn it over)
(The remaining steps in the assembly of the TECP Mounting Rack will be performed after
the assembly of Flanges C & D, described below, because the rods in holes #1 and #7 are to
be used for alignment of the Groove Grabbers with Flange C, and will subsequently be
moved to holes #2 and #8.)
TECP Calibration Chamber Assembly
1.3.1 Flanges C & D
Place Flange D with VCR fitting downward on assembly platform (AP)
Place a metal Conflat gasket on Flange D
Attach a SS-SCA-C3 tapped screw clamp to the center inside edge of both Groove Grabbers
(G), but do not tighten the screws completely.
Attach Groove Grabbers (G) to grabber grooves on diametrically opposite sides of 6"CF
Thin Flange (C), but again, do not tighten the set screws.
Check alignment of Groove Grabbers by sliding Flange C down onto the Base Plate so that
the rods sticking up from holes #1 and #7 go through the center hole of the SS-SCA-C3
screw clamps.
Tighten set screws on Groove Grabbers to firmly attach them to Flange C.
Lift Flange C off of Base Plate
Place Flange C onto Flange D, rotated so that the Groove Grabbers do not block access to
any of the 4 Dsub 15-pin connectors (E)
Place a 1/8"D x 6"L SS rod (R8-6) in the central hole of each SS-SCA-C3 screw clamp on
the Groove Grabbers, sliding it down until the rod touches Flange D
Tighten SS-SCA-C3 screw clamps.
1.3.2
Flanges A & B
16
•
•
•
Connect temperature sensor (PRT) to the downstream 1.33"CF Flange (B1) on the side of the
6"CF Full Nipple (FN), and
Take digital photograph(s) and measurements to record its position inside FN
Attach 6"-2.75" zero-length reducer flange (A) to non-rotatable flange (B_non) of FN, so that
the 6 tapped holes for the 2.75"CF bolts are on the outside. Use a metal Conflat gasket and
the sixteen 2" hex bolts with eight plate nuts
1.4
•
•
•
•
•
•
•
TECP Mounting Rack Assembly (continued from 1.1)
1.4.1 Assemble Base Plate
Remove R8-2 rods from Base Plate holes #1 and #7, and place them in holes #2 and #8
Return Base Plate to its bottom-side-up position
Slide a loosely assembled SS-SCA-C2 tapped screw clamp onto each R8-2 rod (4 ea) until it
rests against the Base Plate (see Fig. 3)
Insert 1/8"D x 2" L ceramic tubes (4 ea) through each SS-SCA-C2, parallel to the R8-2 rods,
so that they pass through the Base Plate holes #3, #6, #9, and #12, as well as the SS-SCA-C2
clamp on the other (top) side of the Base Plate (Fig. 3)
Place one of the halves of a SS-SC-C2 untapped screw clamp against the inside face of each
SS-SCA-C2 tapped screw clamp so that the semi-circular cutouts rest against the rod and
ceramic tube (see Fig. 3) - and place a strip of Kapton tape across the untapped screw clamp
and around each side to hold it against the tapped screw clamp until the TECP(s) are
inserted. (The tape will also help protect the TECP from any abrasion.)
With the combined screw clamps pressed against the Base Plate, tighten the SS-SCA-C2
screws
Cover the inside edges of the rectangular slots in the Base Plate with Kapton tape
• TECP(s) are required at this point in the procedure - the previous steps can be completed
without them
•
•
•
•
•
•
•
1.4.2
Preventing Electrostatic Discharge (ESD)
Check electrical ground, and perform all assembly involving Flight Hardware on the
grounded antistatic mat
Anyone handling Flight Hardware or components in electrical contact with Flight Hardware
must put on ESD wrist-straps, ESD approved (nitrile) gloves and antistatic lab coats
Double check all ground connections
1.4.3
Mounting TECP(s)
After carrying out the SAFE-TO-MATE procedure (JPL Provided), connect the TECP
pigtail(s) to the D-sub 15-pin connector(s) on the vacuum side of Flange D
Label connectors on air-side as to which TECP it connects to (0,1,2,3), and write this down
in notebook (and take photo if possible)
Slide Base Plate down onto R8-6 rods through BP holes #1, #4, #7 & #10
Place base of TECP(s) into the rectangular hole(s) in Base Plate (choose the hole located
over its corresponding D-sub connector on Flange D) with the humidity sensor membrane
(H) facing inward toward the central axis (Fig. X)
17
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
Insert R8-1.25 SS rod(s) through mounting holes in TECP base (U-shaped channel), flush
with the bottom side of the Base Plate. Secure with Lock Rings (LR) on each end of the rod.
Insert R16-6 "J"-shaped SS rods into the ceramic tubes which pass through Base Plate, so
that the rods hook over the top of the TECP between needles 2 & 3. Secure the lower end of
the rod beneath the Base Plate using (?? very small ring clamp?, wire nut?)
Take photos to document assembly and relative location of TECP(s)
1.5
Test TECP data signal connections and control system (Labview)
Power up PC and start Labview VI (more specifics here)
Plug in the power supply for the TECP Interface Box; make sure it is switched OFF
Connect 9-pin serial cable between PC and TECP Interface Box
After carrying out the SAFE-TO-MATE procedure (JPL Provided), connect the 15-pin serial
cable(s) between corresponding connectors on Flange D and the TECP Interface Box
(labeled TECP 0,1,2,3) Run DVM required.
Switch ON Interface Box
Test data system to verify that the TECP(s) are functional and to check correspondence
between TECP ID# labels and data system ID#'s. (This can be checked using the JPLprovided electrical conductivity test adapter connected to an oscilloscope, but visual
inspection may be sufficient. Interface box is clearly labeled.
1.6
Complete Assembly of Calibration Chamber, Connection to MASC
Place a metal gasket on Flange C
Carefully lower the open end of the 6"CF full nipple over the TECP mounting rack until it
rests on Flange C. (It would be useful to have a small "crane" or pulley system for this, as
well as for lowering the assembled Cal. chamber into the freezer for connection to the
MASC)
Use 2.5" hex bolts and plate nuts to seal Flange B_rot to Flange C
Tilt cal. chamber 90 deg. with side ports (B1 and B2) on top
Wrap chamber in antistatic bag.
Move to freezer
Attach grounding strap from assembled cal chamber to facility ground in freezer.
Remove antistatic wrap.
Lower cal chamber into freezer onto pre-positioned support rack
Connect pipe fittings on each side to MASC flow tubing
2 Calibration
Close valve
Pump-down to evacuate system and remove outgassing
Backfill with dry CO2 gas up to Mars pressure (4 Torr)
Perform calibration measurements at room temperature
Turn on freezer, and perform calibration measurements at >=5 different temperatures
between 195K and 274K, and >= 3 different relative humidities at each temperature
Step function response.