Instrument Overview
  ===================
    The Atmospheric Structure Instrument and Meteorology Package was
    an engineering subsystem of the Mars Pathfinder spacecraft which
    provided data for scientific analysis.  It was implemented as a
    facility experiment in an in-house mode at NASA's Jet Propulsion
    Laboratory, taking advantage of the heritage provided by the
    Viking mission.
 
    ASI/MET data were collected during different phases of the
    mission by accelerometers and wind, temperature, and pressure
    sensors mounted in various places on the spacecraft.
 
    Data acquired during the entry, descent, and landing (EDL) phase
    of the mission permitted the reconstruction of profiles of
    atmospheric density, temperature and pressure from altitudes in
    excess of 100 km to the surface.  After the landing, day to day
    variations in temperature, pressure, and wind speed were
    monitored.
 
 
  Scientific Objectives
  =====================
    The Mars Pathfinder mission has provided a third set of in-situ
    measurements of the Martian atmosphere, after the two Viking
    Landers 20 years ago.  The ASI/MET package was designed to
    collect density, temperature, and pressure information about the
    structure of the Martian atmosphere during the spacecraft's
    passage through the atmosphere, from about 150 km altitude down
    to the surface.  Once on the surface, it was also designed to
    regularly monitor the temperature, pressure, and wind speed and
    direction at the landing site for the duration of the mission.
 
    Combined with information collected from the Imager for Mars
    Pathfinder about atmospheric aerosols and dust, the amount of
    water vapor at the landing site, and wind information derived
    from observation of the MET mast-mounted wind socks, these data
    should contribute to a better understanding of the overall
    structure of the Martian atmosphere, and its variation over time.
    A further comparison with data collected by the Viking Lander
    spacecraft can also reveal information about the longer-term
    evolution of the atmosphere.
 
    For more information about the scientific objectives of the
    ASI/MET investigation, please see [SEIFFETAL1997].
 
 
  Location
  ========
    The lander is located in an ancient flood plain in the Ares
    Vallis region of Chryse Planitia at 19.17 degrees north, 33.21
    degrees west with respect to the U.S.  Geological Survey
    cartographic network.  The landing occurred on July 4, 1997, at
    which time it was late northern summer (Ls = 143 deg) on Mars.
    See [GOLOMBEKETAL1997B] for further details.
 
 
  Subsystems, Detectors, and Platform Mounting Descriptions
  =========================================================
    The ASI/MET experiment is performed by accelerometer and
    meteorology (MET) instruments using hardware distributed
    throughout the Pathfinder lander.  The accelerometer instrument
    consists of three science and three engineering accelerometers,
    with supporting electronics, mounted on two boards in the
    Pathfinder Integrated Electronics Module (IEM).  The MET
    instrument consists of pressure, temperature and wind sensors and
    a single IEM electronics board.  The pressure sensor is mounted
    outside the IEM, but within the lander thermal enclosure and is
    connected to an aperture in the lander vehicle via a pitot tube.
    All the temperature sensors and the wind sensor are mounted on a
    meteorological mast 1 cm in diameter and 1.1 m high, which is
    deployed after landing.  The mast base is located on the end of a
    lander petal to minimize the thermal contamination of temperature
    and wind measurements by the spacecraft.  In addition to
    pressure, temperature, and wind sensor electronics, the MET board
    also carries the signal processing electronics for the
    temperature sensors of the Aeroshell Instrumentation Package
    (AIP).
 
 
    Accelerometers
    --------------
      All six science and engineering accelerometers are Allied
      Signal QA-3000-003 units.  The three science accelerometers are
      mutually orthogonal and are oriented parallel to the lander X,
      Y, and Z axes, whereas the three orthogonal engineering
      accelerometers are oriented in the X direction and at +/- 45
      degrees to the Z-axis (+YZ and -YZ).  Because of limitations
      imposed by the Pathfinder system design, some accelerometers
      are displaced significantly from the entry vehicle center of
      mass.
 
      The primary function of the engineering accelerometers is to
      monitor and contribute to the timing of important events during
      entry and descent, such as parachute deployment.  They are also
      used after landing to right the spacecraft and point the lander
      high gain antenna accurately.  Science accelerometer
      observations focus on measuring atmospheric accelerations
      throughout entry and descent with maximum resolution, although
      the optimum measurement strategy has been modified to provide
      redundancy for the engineering accelerometers.  Data from all
      six accelerometers are sampled at the same rate and will be
      available to ASI/MET investigators.
 
 
    Pressure Sensor
    ---------------
      The ASI/MET pressure sensor is a Tavis variable reluctance
      diaphragm device similar to the pressure sensors flown on the
      Viking mission.  This device has proven long-term stability and
      reliability over four martian years of operation.  It is
      designed to make measurements of dynamic pressure during entry
      and static pressure on the surface, and is connected to the
      outside world by a pitot tube approximately 1 m long and 2 mm
      inside diameter.  Its response time is expected to vary from
      1.5 seconds at 3 mbar to 0.45 seconds at 10 mbar.  The tube
      entry port lies in the plane of the aperture between the lander
      instrument shelf and two petals, and is oriented perpendicular
      to the anticipated airflow during descent.  Because no objects
      were allowed to extend beyond the lander profile during descent
      the entry port location is not ideal.
 
 
    Temperature Sensors
    -------------------
      Atmospheric temperatures during the descent and landed phases
      of the mission are measured by fine wire chromel-constantan
      thermocouple sensors mounted on the mast.  It consists of three
      thermocouples wired in parallel in a fiberglass support
      structure.  The thermocouple wires are 75 um in diameter and
      the junction diameter is approximately 200 um.  The fine wires
      provide conductive isolation for the junctions, and ensure that
      they reach equilibrium with the tenuous martian atmosphere
      fairly rapidly, although radiative corrections will have to be
      modeled to obtain the best results from these sensors.  Three
      wires provide redundancy against damage, either during descent,
      or due to dust storms during the landed mission.
 
      A single thermocouple is assigned to temperature measurements
      during the descent phase of the mission, when the entry vehicle
      is falling by parachute after the heat shield has been
      jettisoned.  The descent sensor is located near the top of the
      mast, immediately below the wind sensor.  Its thermocouple
      wires are approximately 4.0 cm from the mast axis and are
      oriented perpendicularly to it.  During descent the mast is
      stowed and the sensor is close to the center of an aperture in
      the corner of the lander, although it is set back significantly
      due to concerns over interference with the lander airbags.
      Accurate descent temperature measurements rely on undisturbed
      air flowing through the aperture and over the descent sensor.
      Wind tunnel measurements on models have shown that this flow is
      highly disturbed, so that descent temperature measurements are
      likely to be unreliable.  Currently a flexible duct linking the
      external flow to the sensor is planned to improve the quality
      of the data.
 
      Three thermocouples are designed for temperature measurements
      during the landed phase of the mission when the mast is
      deployed.  These sensors are located on the mast 0.273, 0.508
      and 1.038 m above the plane of the petal.  For each sensor, the
      thermocouple wires are oriented vertically approximately 2.6 cm
      from the mast axis.  All three sensors and mounts are identical
      and are clocked at the same angle on the mast.  Temperature
      measurements are subject to thermal contamination when the mast
      is downwind from the spacecraft or the sensor wires are
      downwind from the mast.  The sensors are therefore oriented
      away from the spacecraft when the mast is deployed so that
      these conditions coincide, minimizing the range of unfavorable
      wind directions.  The orientation on Mars is 130 degrees east
      of north (pointing roughly southeast).
 
 
    Wind Sensor
    -----------
      ASI/MET uses hot wire resistance thermometers to measure wind
      speed and direction.  The sensor consists of six 50 um diameter
      platinum- iridium wire elements arranged symmetrically around a
      cylindrical core approximately 2.6 cm in diameter.  Each
      element consists of a 20 cm wire wound around insulating
      formers to give eight closely spaced lengths oriented parallel
      to the axis of the mast roughly 3 mm from the core.  The sensor
      is mounted at the top of the mast, at a height of 1.096 m above
      the lander petals.  All six sensor elements are connected in
      series and are heated by a constant current source.  In still
      air at typical martian surface pressures, the elements are
      heated to approximately 40 degrees Celsius above ambient.  Wind
      blowing round the core of the sensor cools the elements, and
      the cooling produced at an individual element depends on its
      position relative to the core and the wind direction.  Overall
      cooling is dependent on wind speed and the pattern of cooling
      for the six elements is a indicator of wind direction.
      Sensitivity is greatest for low winds and varies roughly
      inversely with wind speed.
 
      A single chromel-constantan thermocouple junction is mounted
      within the core of the wind sensor.  Its function is to provide
      a temperature boundary condition measurement needed to
      interpret the wind measurements.
 
 
    Wind Socks
    ----------
      Also mounted on the MET mast were three aluminum wind socks,
      used to determine wind speed and direction.  Since data
      collection from these wind socks was done with the Imager for
      Mars Pathfinder, they were not technically considered to be
      part of the ASI/MET package.  A separate instrument description
      exists for them, under the name Mars Pathfinder IMP Windsocks.
      (A copy of that document, WINDINST.CAT, is included on the
      ASI/MET CD.)
 
 
  Operational Modes and Measured Parameters
  =========================================
    The sampling frequency of all the sensors during the entry and
    descent portion of the mission was governed by the vertical rate
    of descent through the atmosphere.  After landing, pressure and
    temperature measurements were made to establish the diurnal
    variations and the day-to-day variations over the operating life
    of the mission.  Details are provided in the individual data set
    descriptions.
 
 
    Accelerometers
    --------------
      The accelerometers had several gain states to cover the wide
      dynamic range from the micro-g accelerations experienced upon
      entering the atmosphere to the 18g peak experienced during
      deceleration and landing.  The gain states provided dynamic
      ranges of 16 mg, 800 mg, and 40 g.  (The accelerometer
      calibration was done relative to the gravitational force on a
      concrete plinth in building 150 at the Jet Propulsion
      Laboratory.  The acceleration due to gravity at this location
      (1 g) is equal to 9.795433 m/s**2.) With 14 bit digitization of
      the signals, these ranges had respective resolutions of 2
      micro-g, 100 micro-g, and 5 mg.  The noise levels of each gain
      state have been measured at 1 to 2 counts or less, implying
      that accelerations of 10 micro-g should be significant with 20%
      uncertainty.  Low pass filters in the accelerometer electronics
      were used to attenuate signal frequencies above 5 Hz.  This was
      done to suppress the effects of noise and spacecraft dynamic
      motion.
 
 
    Pressure Sensor
    ---------------
      The pressure sensor obtains data in two ranges simultaneously;
      0 - 12 mbar for descent and 6 - 10 mbar for surface
      observations.  Signal digitization to 14 bits provides
      respectively 0.75 ubar and 0.25 ubar resolution over these
      ranges.  Again, system noise levels were measured at 1 to 2
      counts or less.  As Tavis pressure sensor measurements are
      temperature dependent, the temperature of the device is
      monitored in flight by an accurate platinum resistance
      thermometer (PRT).  The PRT covers the 200-335 K temperature
      range with a resolution of 0.01K.
 
 
    Temperature Sensors
    -------------------
      All ASI/MET thermocouples are referenced to 'cold' junctions
      mounted within a small isothermal block located inside the
      lander petal close to the base of the mast.  The temperature of
      the block is monitored by an accurate PRT with a range of 140
      to 330 K and a resolution of 0.01 K.  Each thermocouple has a
      dynamic range of roughly +/- 100 K about the PRT temperature,
      and a resolution of 0.01 K.  On the surface of Mars, the
      thermocouple junction time constant is of order 4-5 seconds.
 
 
    Wind Sensor
    -----------
      The wind sensor has two operating modes, high and low power.
      In the high power mode, a current of 51.5 mA flows continuously
      through the sensor producing the overheat required for wind
      measurements.  In the low power mode, 20.6 mA flows for 3 msecs
      whilst element temperature is being sampled.  Low power
      produces insignificant wire heating and provides an accurate
      measure of ambient temperature at the sensor.  The resolution
      of the wind sensor element temperature measurement is 0.11
      Kelvin for the low power and 0.04 Kelvin for the high power
      setting.  Under martian conditions, the sensor responds to
      power changes with a short term time constant of 1-2 seconds.
 
 
  Calibration
  ===========
    All the ASI/MET sensors have been subjected to a series of
    calibration measurements.  In many cases these are satisfactory
    but in some cases resource and schedule limitations in the
    hardware development effort left significant gaps.  Preliminary
    analysis of the calibration data has been completed, but some
    work remains to be done by ASI/MET science team members.  A
    summary of the measurements is given below.
 
 
    Accelerometers
    --------------
      Accelerometer calibration measurements can be divided into two
      categories; end-to-end response calibration for each sensor,
      and alignment and location calibration relative to the lander
      spacecraft coordinate system.
 
      End-to-end response measurements using the assembled flight
      accelerometer board on a precision dividing head have
      established the linearity, gain and offset of all six
      accelerometers, at each gain setting over a temperature range
      of -5 degC to +40 degC.  All the sensors are highly linear, and
      gains and offsets vary linearly with temperature.  Calibration
      measurements are limited to 1g on the 40g gain setting, and
      have not covered the full temperature range expected for the
      IEM electronics, so that some extrapolation is required.  These
      measurements have also established the relative orientation of
      the accelerometer heads in the flight board coordinate system
      to a few hundredths of a degree.
 
      Further optical measurements on the spacecraft in its cruise
      configuration have established the shape of the heat shield and
      the orientation of its symmetry axis in the spacecraft
      coordinate system with similar accuracy.  The relative
      orientation of these two coordinate systems will not be
      measured, but they are specified to be coincident to better
      than 0.25deg and are expected to be considerably better in
      practice.  During the spin balancing of the spacecraft, the
      location of each head relative to the entry vehicle C of M will
      be determined to about 2 mm in the X & Y directions.  In the Z
      direction the location of the C of M will not be measured and
      will be uncertain to approximately 2.5 cm.
 
      Detailed data conversion expressions used to calibrate the
      accelerometer data are provided in [SCHOFIELD1996A] and
      [SCHOFIELD1997A].
 
 
    Pressure Sensor
    ---------------
      The calibration of the ASI/MET pressure sensor used in
      [SCHOFIELDETAL1997] has been revised significantly on this
      CD-ROM, for the reasons described below.
 
      The Tavis pressure sensors used by ASI/MET are sensitive to
      temperature and must be calibrated to correct for this
      dependence.  During pre-launch calibration, both the flight and
      flight spare pressure sensors were inadvertently exposed to
      temperatures 30 K below their design limits.  This produced
      changes in sensor offset of several tenths of a mbar, and
      increased the variation of offset with temperature by a factor
      of 7 relative to the original calibration.  These changes were
      not noticed until spacecraft thermal vacuum testing and by the
      time they were understood, it was too late to procure new
      sensors.  It was therefore decided to fly the flight sensor and
      use cruise phase health check measurements to calibrate the
      temperature dependence of offset.
 
      As sensor gain calibration was not possible during the mission,
      it was assumed the gain had not changed.  Limited pre-flight
      measurements and more extensive calibration of the flight spare
      sensor suggested that the effects of gain changes on pressure
      measurements at 7 mbar were 6-8 times smaller than those of
      offset changes.
 
      During the nominal landed mission, the ASI/MET pressure sensor
      experienced diurnal temperature variations of 265-300 K.  These
      introduced large, spurious, components to the diurnal pressure
      cycle, due to the temperature variation of sensor offset, which
      had to be removed using calibration data.  In cruise, it was
      only possible to calibrate flight sensor pressure offset in the
      270-280 K temperature range.  These data indicated a
      temperature dependence of -0.019 mbar/K.  This was the
      temperature dependence assumed in [SCHOFIELDETAL1997].
 
      However, after the mission, the offset calibration was
      revisited.  It was found that the cruise offset measurements
      were influenced by long term changes (over 6 months) as well as
      temperature dependent changes.  Offset measurements over the
      270-280 K range made during the 90 minutes before entry
      eliminated the long term drift and indicated a temperature
      dependence of -0.0135 mbar/K - a smaller dependence than the
      cruise offset measurements.
 
      Late in the landed mission when the spacecraft was off
      overnight and took data only during the day, the sensor
      temperature and measured pressure cycles were observed to be
      very similar from day to day (with the exception of the slow
      seasonal increase in pressure).  However, during the four 24
      hour, 4 second sampling sequences (sols 32, 38, 55, and 68),
      pressure sensor temperatures often differed by 15 K for the
      same time of day.  This is because at the beginning of the
      sequence the spacecraft had been off all night, while at the
      end of the sequence it had been on all night.  The -0.019
      mbar/K temperature coefficient did not produce repeatable
      pressure cycles under these conditions but the -0.0135 mbar/K
      coefficient did, providing additional support for the pre-entry
      result.  The pressure cycle data also suggested that this
      offset coefficient was valid over the 260-310 K temperature
      range.
 
 
    Temperature Sensor
    ------------------
      Both the thermocouple and PRT temperature sensor channels have
      undergone end-to-end temperature calibration.  In addition,
      independent sensor calibration data are available, and the
      gains and offsets of the electronics have been measured over a
      range of -40 degC to 50 degC.
 
      End-to-end temperature calibration was performed with the
      sensors in a nitrogen atmosphere within an isothermal copper
      chamber.  At a pressure of one atmosphere, chamber temperatures
      were moved between stable plateaus covering the range 140 to
      360 K in 20 K steps whilst data were logged.  Spatial
      temperature variations within the chamber were less than a few
      hundredths of a degree when data were taken.  The thermocouple
      isothermal block was located outside the chamber at a
      temperature different from the sensors, to provide
      representative thermocouple voltages.  Both block and chamber
      temperatures were monitored by National Institute of Standards
      and Technology (NIST) calibrated PRT thermometers, accurate to
      0.01 K.
 
      In addition to the end-to-end calibrations, accurate
      calibration curves are available for the individual PRT sensors
      used by ASI/MET.  The temperature vs voltage characteristics of
      chromel-constantan thermocouples are also well documented and
      are expected to vary little from thermocouple to thermocouple.
 
      The end-to-end temperature calibration was performed with the
      flight electronics at ambient temperature.  To correct for
      electronics board temperature variations expected during the
      mission, the response of each channel to fixed input voltages
      was measured over the -40 degC to 50 degC temperature range,
      allowing accurate gain and offset temperature coefficients to
      be derived.  For the thermocouple channels, gain is also
      dependent on lead resistance.  This effect was present in the
      end-to-end calibrations but was not part of the electronics
      gain measurements.  Lead resistance for each thermocouple was
      therefore measured at ambient temperature using an AC technique
      to eliminate errors produced by thermocouple voltages.
 
 
    Wind Sensor
    -----------
      The wind sensor is effectively an array of PRT temperature
      sensors and its calibration can be divided into two parts; the
      calibration of the temperature sensors and the calibration of
      the temperature response of these sensors to wind speed and
      direction (wind calibration).  In addition the geometrical
      properties of the sensor elements must be measured accurately.
 
      The temperature calibration of the flight wind sensors is very
      similar to, and was conducted at the same time as, the
      thermocouple and PRT temperature calibration discussed above.
      It consisted of end-to-end temperature calibration from 140 to
      360 K and flight electronics gain and offset calibration over
      the range -40 degC to 50 degC.  In addition the constant
      currents supplied by the flight electronics to heat the wind
      sensor windings were measured accurately in both low and high
      power modes at ambient temperature.  In one wind sensor
      channel, only ambient temperature gain and offset data have
      been obtained.  This is not a serious problem as the
      temperature dependence of gain and offset variations is very
      similar for all six channels.
 
      Three nominally identical wind sensors were constructed and
      subjected to end-to-end temperature calibration measurements.
      The first was mounted on the flight mast, the second was held
      in storage as a flight spare, and the third was delivered to
      the AMES research center as a flight test unit for wind
      calibration investigations.  The wind calibration was conducted
      in the Mars wind tunnel at AMES using nitrogen at pressures of
      8, 12, and 18 torr, for wind speeds in the range 0 - 50 m/s and
      wind directions covering 360 degrees in steps of 30 degrees and
      60 degrees in steps of 5 degrees.  Wind direction changes were
      simulated by rotating the sensor in the wind tunnel.  Wind
      calibration was not performed directly on the flight sensor to
      avoid contamination by the fine dust used to measure wind speed
      in the tunnel experiment chamber.  The assumption underlying
      the wind calibration is that it can be transferred from the
      flight test unit to the flight unit given accurate temperature
      calibrations and geometrical measurements for both units.
 
      As of the time of writing (Oct.  1998), reliable wind speeds
      and directions had not yet been derived from the data received
      from the Martian surface.  Re-calibration of the wind sensor
      continues, and the wind results should be released soon.
 
 
    Aeroshell Instrumentation Package
    ---------------------------------
      The Aeroshell Instrumentation Package (AIP) data were recorded
      via the ASI/MET data stream for a portion of the descent
      through the Martian atmosphere.  The data calibration
      expressions for these data are provided in [SCHOFIELD1996B].
 
 
  Operational Considerations
  ==========================
    The sensor test masses of the three science accelerometers were a
    few centimeters away from the lander's center of gravity and
    could therefore receive contributions from angular accelerations
    due to the lander's pitch and yaw.
 
    The science accelerometers first detected the Martian upper
    atmosphere 160 km above the surface.  The descent rockets fired
    5.2 minutes later at an altitude of 0.1 km, ending the direct
    measurement of aerodynamic decelerations.
 
    Data collected from the engineering accelerometers were not used
    in the generation of atmospheric profiles.  These accelerometers
    were kept in their least sensitive 40g measurement range because
    they were being used to control the deployment of the parachute.
 
    The EDL phase pressure sensor wasn't able to begin unobstructed
    measurements of the atmosphere until the heat shield separated
    from the lander 3.4 minutes after entering the atmosphere.  The
    spacecraft was 7.4 km above the surface at this time.  The
    measurements continued for 1.7 minutes until at 0.3 km above the
    surface, the inflation of the airbags again obstructed the
    pressure measurements.