Instrument Overview:
   ====================
   PEPE is designed to be a general- purpose plasma sensor capable of
   measuring electrons and mass-resolved ions over most of particle phase
   space from 8 eV to 33,500 eV. Because 3-axis stabilized spacecraft
   pose a problem for complete coverage of all possible plasma arrival
   directions, PEPE is also designed with a electrostatically scanned
   field-of-view (FOV) that covers a solid angle of 2.8 pi steradians
   once each 65 s. PEPE 'heritage' comes largely from work at the two
   collaborating institutions: SwRI's development of miniaturized
   electrostatic analyzer optics [YOUNGETAL1998] and LANL's development
   of linear electric field (LEF) time-of flight (TOF) mass spectrographs
   [MCCOMASETAL1998; NORDHOLTETAL1998;].
 
 
   PEPE contains three sensors integrated within a single housing.
   PEPE contains a number of components such as electron
   multiplier detectors, carbon foils, and high ohmic surfaces that are
   sensitive to particulate and chemical contamination, The most
   sensitive instrument, the ion mass spectrometer, was protected with a
   cover that was deployed in flight. The two other sensors were flown
   without covers. Our calculations indicated that PEPE could survive the
   launch environment without a deployable cover, thereby saving cost,
   complexity and mass. Throughout ground handling, PEPE was fitted with
   sealed cover through which N2 purge gas was flushed. The cover was
   removed before the rocket fairing door was closed, but the purge
   continued through launch. PEPE in-flight performance indicates that
   the instrument did not suffer any deleterious effects caused by this
   approach.
 
 
   The PEPE optical system is rotationally symmetric about the central
   vertical axis and consists entirely of electrostatic elements. The
   optical system consists of three main functional elements: Angular
   deflection optics, energy/charge analyzers, and the LEF TOF optics.
   PEPE is located on a bracket at the upper (+Z) end of the +X-Y panel
   of the spacecraft such that the central plane of the PEPE
   FOV lies in the spacecraft XZ plane. This gives PEPE access to both
   the nominal solar wind direction and to the DS1 Miniaturized
   Camera/Spectrometer (MICAS) boresight
   direction located along the spacecraft +Z axis. PEPE FOV coordinates
   are illustrated in Figure 1. The 3-axes of the PEPE coordinate
   system are parallel to those of the spacecraft.  Arrows in the
   figure indicate the sense of the numbering of angular pixels that
   make up the PEPE FOV. The angular coordinates for azimuth
   are defined between 0 and 359 degrees and correspond to the direction
   of arrival for both positively and negatively charged particles (e.g.,
   +90 deg azimuth indicates particles arriving from the direction of
   the spacecraft -Z axis). The angular coordinates for elevation
   are defined as the direction of arrival of positively
   charged particles. Because of the nature of the deflection electric
   field that controls the PEPE elevation viewing direction, the
   direction of arrival of negatively charged particles is the opposite
   of that shown in Figure 1.
 
 
   The entire PEPE ion mass analysis system weighs less than 1 kg, uses
   less than one watt of power, is extremely compact and is designed to
   be integrated very tightly with the power supplies that drive it.
   Close integration with the power supplies is important to keep the
   mass and volume of PEPE down.  Often, TOF mass spectrometers using
   high voltages require large standoff distances that translate into a
   large volume and thus a high mass device.  The 15kV and ion MCP high
   voltage power supplies are designed to plug directly into the
   time-of-flight cylinder without the use of conventional high voltage
   connectors or cables.
 
   High Voltage Power Supply Subsystem. The PEPE electrostatic optics and
   MCP detectors nominally require 8 high voltage sources ranging in
   maximum absolute values from 3600 V for the MCPs to 15,000 V for the
   TOF. Four of these voltages
   (two for MCPs and positive and negative 15,000 V for the LEF TOF) are
   obtained from independently programmable high voltage power supplies
   (HVPS) that are adjustable over an 8-bit range but are otherwise
   static. Each deflection electrode requires a programmable bi-polar
   stepping supply (i.e., one that simultaneously creates equal negative
   and positive potentials) while the ion and electron ESA's are operated
   from two independent supplies. The most efficient way to generate the
   interrelated deflection and ESA voltages is to employ a so-called
   'bulk' supply that biases the two bi-polar deflection supplies and the
   ESA supplies into a coarse operating range from which finer voltage
   steps can be set. The result is a power and component
   efficient system with 12-bit programmable high voltage control between
   2500 V for the ESA's and ? 5000 V for the deflection electrodes.
 
 
   Because of the large number of HVPS in PEPE, efficient packaging of
   the supplies was critical. The method chosen eliminated HV cables and
   bulky connectors entirely. As mentioned above, the ion MCP and the two
   TOF supplies are each built into small easily installed metal-plated
   plastic boxes that insert into the central portion of the PEPE body.
   High voltage connectors are standard pin and socket connectors with
   insulating sleeves that interleave as the pins mate. The electron MCP
   supply is located on the same circuit board as the MCP and amplifiers.
   Because of the need for close proximity of supplies to the deflection
   and ESA electrodes, each ESA supply is located on a single circuit
   board that also carries the inner ESA toroid, inside of which is the
   ESA supply. The bi-polar deflection supplies are imbedded within the
   structure that supports the deflection electrodes. The
   interiors of all HVPS housings were parylene coated to passivate high
   voltage components and increase the dielectric strength of the housing
   insulation.
 
 
   Scientific Objectives:
   ======================
   The primary scientific objective of the PEPE investigation is to study
   the interaction of the solar wind with small solar system bodies. On
   July 29, 1999 DS1 flew past Asteroid Braille (formerly 1992 KD) at a
   closest approach distance of 27 +/- 2 km (K. Fleming, personal
   communication). Unfortunately the encounter miss distance was well
   outside the planned value of  ~7 km. The encounter geometry was also
   outside nominal values. Those circumstances, coupled with the apparent
   hard mineral makeup of Braille (e.g., pyroxene or similar), limited
   the charged particle signal that could be expected at DS1. Indeed, the
   PEPE data failed to show any evidence of solar wind interaction with
   the asteroid.
 
 
   Following the encounter with Braille, PEPE will have had the
   opportunity to sample the comas of two very different comets in 2001.
   Comet Wilson Harrington is thought to be a low activity highly
   outgassed body with little coma activity. Although a gas outburst was
   observed in 1949 [FERNANDEZETAL1997], it is presently unclear what
   the gas production rate will be during the planned encounter in early
   2001.  The second object, Comet Borrelly, is a relatively young
   object and was should be relatively active. The combination of
   observations near these two objects will allow PEPE to break new
   ground in the comparative study of the evolution of cometary comas and
   their composition.
 
 
   A second important PEPE scientific objective is the study of solar
   wind ion composition and ion and electron velocity distributions.  DS1
   and PEPE will provide a solar wind observation platform well away from
   the Earth during most of the mission. Data from PEPE during these
   periods can be applied to 3-dimensional studies of large scale
   structures in the solar wind, for example the evolution of coronal
   mass ejections from the sun. Simultaneous studies early in the DS1
   mission will combine data from PEPE, Wind, ACE and Cassini plasma
   instruments to calibrate PEPE response. Solar wind data will also
   support demonstration of PEPE's operational capabilities in the plasma
   environment of the DS1 spacecraft. The quality, accuracy, and
   consistency of scientific observations of the solar wind made with
   PEPE will help establish the appropriateness and utility of new
   technologies incorporated into its design.
 
 
   Calibration:
   ============
   Ions and electrons enter the PEPE optics through a grounded toroidal
   grid that defines the external acceptance aperture of the instrument.
   The grid also terminates the electric fields within PEPE. Immediately
   inside the grounded grid, ions and electrons entering PEPE experience
   an electric field that deflects them into the acceptance volume of the
   top-hat electrostatic analyzers. The deflection electric field is
   generated by two opposing, cylindrically symmetric toroidal
   electrodes. Equal voltages of opposite polarity are applied to the
   deflection electrodes so that, for example, ions coming from above the
   symmetry plane are deflected in the direction of the top-hat
   apertures, while electrons coming from below the plane are also
   deflected toward the top-hat apertures. The top-hat analyzers, one for
   ions and one for electrons, are positioned so that their apertures are
   located symmetrically above and below the central plane of the
   aperture. As ions and electrons enter the region of the top-hat
   analyzer electric fields they are deflected down or up into their
   respective electrostatic energy analyzers. Up to this point the
   optical systems are identical for ions and electrons.
 
 
   Elevation Deflection Analysis. The deflection optics consists of
   matching upper and lower toroidal deflectors whose shape is an
   exponential given by the equation:
 
 
       z = k1 +/- k2 exp[k3(r - k4)]                          (1)
 
 
   where z is the height of a point above the aperture mid-plane and r is
   the corresponding radius from the central axis. The value of k4
   determines the radius of the inner edge of the deflection toroids and
   is 6.00 cm in PEPE. Although rectilinear deflector profiles can give
   nearly the same performance, an exponential shape was chosen because
   it offers the highest deflection sensitivity for a given plate voltage
   [ZHIGAREV1975). The particular configuration of the deflection plates
   themselves was also designed to create space underneath the electrodes
   for the placement of high voltage supplies. This improves engineering
   architecture at the expense of the deflection constant and hence
   magnitude of the applied voltage.
 
 
   The relationship between the angle of charged particle deflection
   (Gamma), the deflection voltage (VD) in volts placed on the
   electrodes, and the particle energy/charge (E/Q) in eV was obtained by
   ray-tracing and is given by:
 
 
       Gamma(deg) = k5 Q*VD/(E/Q) + 0.9deg.                   (2)
 
 
   Both VD and Q are to be taken as signed quantities. Thus, for
   electrons Q = -1 and if VD is positive then Gamma is a negative
   quantity in Figure 1. The nominal value of k5 = 134.5 deg per charge
   with a slight variation of < 10ver the elevation range of +/- 45deg.
   The constant of 0.9 deg represents an offset in the top-hat field of
   view. Voltages up to a maximum of +/- 5000 V can be applied to the
   electrodes which places a limiting value on the maximum angle through
   which particles of a given energy can be deflected:
 
 
       Gamma-max(deg) = 6.725 x 10**5/(E/Q).                  (3)
 
 
   This means that the maximum FOV deflection at the upper limit of the
   PEPE energy range (33,500 eV) is 20.0 deg per charge [Then how do we
   get +- 45o?]. The angular resolution of the instantaneous FOV is 4 deg
   FWHM for electrons and 3.6 deg FWHM for ions. These values are set by
   the separation of the deflection electrodes and the intervening
   collimation and are designed to match the acceptance aperture of the
   ESA optics.
 
 
   Different angular directions are accepted by stepping the deflection
   voltages through a program that causes the central FOV angle of regard
   to move from -45 to +45 deg (or vice versa depending on charge sign)
   in 16 linearly spaced steps. Equation (2) shows the relationship
   between particle energy/charge and the deflection angle. This implies
   that the voltage VD is dependent on the voltage VE that is applied to
   the energy/charge analyzer and determines the energy of particles
   accepted.
 
 
   Energy/charge and Angle. After passing through the angle deflection
   electrodes, ions and electrons enter an electric field created by two
   facing toroidal top-hat ESA's. Each ESA consists of concentric
   toroidal electrodes with a cutout in the outer electrode which
   functions as the ESA entrance aperture. Toroidal shapes are chosen
   because they result in a larger aperture area per unit of ESA
   electrode area [YOUNGETAL1989]. When a voltage is applied to the inner
   electrode (positive for the electron ESA, negative for the ion ESA),
   charged particles with the appropriate energy/charge and angle travel
   through the electric field to the exit of the ESA. Different energies
   are sampled by stepping the voltage on the ESA inner toroids. At the
   same time that the ESA voltage is stepped, so is the deflection
   voltage according to the relation given by (5).
 
 
   The open area formed by cutouts in the outer, grounded ESA toroids
   expose the oppositely charged inner toroids. This sets up an electric
   field that deflects ions and electrons into their respective energy
   analyzers. The toroidal apertures are set back 15 deg from the
   vertical. The ESAs are designed so that electrons travel through a
   bending angle of 90 deg, 75 deg of which are spent in the ESA. Ions
   travel through a total bending angle of 115 deg of which 100 deg are
   in the ESA. The combination of toroidal electrode separation and ESA
   bending angle sets the energy passband at (Delta)E/E = 0.085 (FWHM)
   for the electron ESA and 0.050 for the ion ESA.
 
 
   The relationship of voltage on the ESA to ion and electron E/Q is
   given by:
 
 
       E/Q = k6 * VE                                          (4)
 
 
   Where k6 = 13.07 is an unsigned constant of the ESA geometry and is
   identical for the ion and electron analyzers. Substitution of Equation
   (4) into (2) gives the control equation for the combined
   deflection/energy analyzer system:
 
 
    Gamma(deg) = 10.27 VD/VE.                                 (5)
 
 
   Electrons exiting the electron ESA are accelerated by +200 V onto a
   conversion dynode. Electric fields in the dynode/MCP volume focus the
   electrons in two directions so that beam image size is kept to a
   minimum and azimuthal angle of arrival information is preserved. The
   purpose of the dynode optics is to convert incoming primary electrons
   into secondary electrons of a few eV that can be easily focused on to
   25 mm active diameter MCP. The toroid major radius at which the
   electrons exit the ESA is 41.66 mm, whereas the area onto which they
   are focused on the MCP is an annulus 6 to 12 mm in radius. The chief
   considerations in using the conversion dynode were reduction in the
   volume and complexity of a larger diameter MCP and holder, and
   reduction in overall detector background. The secondary electron yield
   of the nickel-plated Noryl(TM) plastic dynode material for penetrating
   energetic electrons is low relative to the efficiency of MCP's for
   these particles. The relative background improvement is calculated to
   be a factor of ~3 [YOUNGETAL1998].
 
 
   The LEF TOF mass analyzer processes neutralized and charged ions
   differently.  In the case of neutralized ions, time-of-flight analysis
   of neutralized ions is performed in the following manner: Ions exiting
   the carbon foil eject one or more secondary electrons. These are
   focused and accelerated into an annular outer ('start') section of the
   MCP located at the far end of the time-of-flight section.
   When the electrons are detected in the outer portion of the MCP, a
   signal is sent to the PEPE time-to-digital converter to start a 83 MHz
   clock.  The LEF optics are designed so that the neutralized ion
   reaches the center portion of the micro-channel plate. When detected
   by the MCP, the time-to-digital converter clock is stopped, a vernier
   is applied to the time measurement, and the time of arrival as well as
   the information that the center ('stop') section of the MCP was struck
   is recorded.  Because energy per charge of the ion, the length of the
   time-of-flight section, and the time-of-flight is now known, the mass
   per charge of the ion can in principle be deduced from the following
   equation:
 
 
       M/Q=2(E/Q \373 Eloss/Q + Vacc)(T/D)2                   (6)
 
 
   Where M is the mass of the incoming ion with energy/charge E/Q exiting
   the ESA, Eloss is the amount of energy lost due to ion scattering in
   the foil, Vacc is the TOF acceleration voltage, D is the average
   length of the time-of-flight section and T is the time-of-flight. This
   is called 'straight through' or linear mass analysis because the
   electric fields inside the mass-analysis section do not affect
   particles neutralized in the foils.  The mass resolution of this type
   of analysis is typically low, primarily because of variations in path
   length and particle energy introduced by the ESA, TOF geometry (finite
   foil and MCP stop region sizes), and scattering in the foil. Both the
   foil and the MCP have finite dimensions and the geometric paths of
   neutrals through the TOF analyzer are not constrained. Therefore a
   relatively large range of path lengths is possible. Which path is
   taken depends on the entry point of the ion on the foil and on the
   angle through which individual neutrals are scattered in the foil.
   Secondly, the energy of ions striking the foils varies within a range
   of may vary by up to 5%, set by ESA resolution. This energy range is
   significantly reduced by accelerating the ions through a 15 kV
   potential before they strike the foil. Furthermore, during traversal
   of the foils energy variations are introduced by scattering. Lastly,
   the finite resolution of the time-to-digital converter introduces
   relatively small errors. Thus the finite size of the LEF TOF optical
   elements (ESA exit aperture, foil and MCP) and the random scattering
   in angle and energy introduce a number of uncertainties into the TOF
   measurement. The resulting theoretical mass per charge resolution can
   be expressed by adding the RMS values of these quantities [BETTS1979]:
 
 
 
 
       (Delta)(M/Q)/(M/Q) = [SUM (Delta E) {E[(Delta)E/(E/Q -
                            Eloss/Q + Vacc)]2 + [2(Delta)D/D]2 +
                            [2(Delta)T/T]2}]1/2.               (7)
 
 
   Where (Delta)E represents each uncertainty in energy introduced by
   resolution and energy scattering effects, the sum (Delta E) is over
   all such energy terms, E is the total particle energy entering the
   foil, (Delta)D/D represents path length variations, and (Delta)T/T is
   the electronic timing uncertainty.
 
 
   In the case where the ions emerge from the carbon foil with a charge
   state other than zero, they will be deflected by the electric field
   induced in the TOF section.  This electric field is produced by a
   -15kV potential applied to the entrance to the TOF section and a
   matching +15kV potential applied to a spherical-shaped grid at the
   exit of the region. The potentials create a well-defined voltage drop
   of 30 kV along the sides of the TOF section induced by a resistive
   coating.  The coating is applied to different sections of the time-of-
   flight region with different methods. Deposition of a constant
   thickness coating produced a constant drop in voltage.  Deposition of
   a varying thickness that decreases proportionally to the distance from
   the foils squared produces a comparable variation in the voltage
   distribution and a linearly increasing electric field with distance
   from the foils. This carefully constructed electric field creates a
   restoring force on the ions in the direction of the axis asymmetry of
   symmetry. The force varies linearly along the axis of symmetry and is
   pointed towards the entrance (carbon foil) end of the TOF section.
   Ion trajectories are confined to the center of the TOF cylinder so
   that the linear electric field is not disturbed by any slight
   variations or irregularities in the coatings. Ions below an energy of
   approximately 18 keV will bounce in the field, executing simple
   harmonic motion as they do so. As a consequence, ion energy and flight
   path in the LEF portion of the TOF do not affect flight times with the
   result that mass resolution is increased over that of the straight
   through paths described above.  The ions act as though they were a
   simple mass on a spring and thus the time of an oscillation in the
   linear electric field region is proportional only to the mass per
   charge of the particle. [MCCOMAS&NORDHOLT1990]. Ions with energies
   above ~18 keV will exit the LEF without undergoing a bouncing motion.
   Similarly, ions that leave the foil negatively charged will be
   accelerated by the LEF down on to the stop region of the MCP and
   analyzed with intermediate mass resolution.
 
 
   At the end of their oscillation nearly 1000f the ions strike a
   secondary emitter located at the center of the foil-end of the
   time-of-flight cylinder.  Because of the properties of the cylindrical
   electric field, ions that do not pass close to the center of the LEF
   are deflected towards the cylinder walls and do not contribute to the
   signal. Electrons ejected by the secondary emitter are focused onto
   the stop region of the MCP and are processed in exactly the same
   manner as the straight-through events described above. Because the
   energy, angle, and variable path to length are no longer in the
   equation for mass per charge of particle, mass resolution improves to
   M/(Delta)M = 20. Since the same MCP is used for both the ST and LEF
   stops, the results is a single spectrum with two peaks per species
   (three peaks per species, in the case of species which may leave the
   carbon foil with a negative charge.)
 
 
   Detectors:
   ==========
   The electron detector consists of a stack of 3 individual MCP's
   arranged in Z- configuration. The MCP is operated with the entrance
   plate near ground potential and the exit at up to +3600 V. Large
   (~ 0.5 cm**3) high voltage capacitors are needed to couple MCP pulses
   to signal amplifiers whose inputs are near ground. The azimuthal FOV
   is divided into 16 sectors that require 16 anodes and corresponding
   signal chains and coupling capacitors. In order to save volume, the
   printed circuit board itself (made of Rogers Corp. TMM3 (TM)) was
   incorporated as the capacitor dielectric material sandwiched between
   the 16 anodes and the corresponding Amptec A111 amplifier inputs.
 
 
   The ion ESA bending angle of 100 deg was chosen to match the conical
   shape of the top of the LEF time-of-flight cylinder to which it is
   interfaced. Ions that exit the ESA are accelerated by voltages between
   -8 kV and -15 kV onto ultra-thin (nominally 0.5-1.0 microgram/cm**2)
   carbon foils located at the entrance to the LEF TOF analyzer.
   (Although PEPE can operate at acceleration voltages below -15kV, the
   latter is the nominal designed operating potential and will be used as
   the standard value throughout the paper.) A slit electrode biased at
   500f the ESA inner toroid voltage is located 1.5 mm past the ESA
   exit. This electrode prevents fringing fields created by the -15 kV
   acceleration potential from penetrating back into the ESA and causing
   unwanted trajectory deflections.  A uniform electric field inside the
   acceleration region leading up to the carbon foils is ensured by the
   use of a high resistance coating on the inside wall of the ceramic
   cylinder in the portion of the TOF section above the -15kV electrode
   (see discussion below).
 
 
   LEF TOF Analyzer. Ions enter the LEF TOF analyzer after being
   accelerated by up to 15kV. Acceleration by 15kV ensures that even the
   lowest energy ions will pass through the carbon foils with high
   probability and relatively small scattering angles.  When (positive)
   ions pass through the carbon foils their charge state may be changed.
   Usually this means that the ion will attach an electron upon exiting
   the foil and will be neutralized as a result.  A relatively small
   percentage of ions (usually a few percent to ~30 0epending on species
   and incident energy) will exit the foil positively charged.
   Furthermore, a varying percentage (from a few to ~30 0epending on
   energy and species) of incident ion species with finite electron
   affinities (primarily H, C, and O) will exit the foils in a negative
   charge state [BURGIETAL1990; FUNSTENETAL1993;].
 
 
 
 
   Electronics:
   ============
   LEF TOF Electronics. The MCP housing for the time-of-flight section is
   also a new design.  A chevron configuration of two high-gain MCPs is
   used instead of a standard z- stack of three MCPs.  The MCP holder is
   highly integrated and fits directly into the bottom of the
   time-of-flight stack.  The capacitors and resistor needed to bias the
   MCP stack and decouple signals are built directly into the MCP holder.
   The anode is integrated behind the MCP holder and pins
   from it couple directly into the amplifier and discriminator
   electronics (termed the front-end electronics (FEE)) board.
 
 
   All time-of-flight information is processed on the FEE board via
   high-speed discriminators that encode the start channel that was hit
   as well as initiating the time-to- digital (TDC) clock. In PEPE, three
   constant fraction discriminators are coupled together to read out 8
   individual anodes in a coded form.  Each amplifier/discriminator is
   built as a chip-on-board (COB) module and 2 sets of 3 each are used to
   read out a total of 16 anodes.  Eight of the anodes are devoted to
   coarse measurement sectors, 6 that cover 41 deg each and 2 that cover
   20.5 deg each.  Eight fine sector anodes cover 5.125 deg each.  The
   symmetry of the PEPE entrance optics requires that all anodes have a
   boundary at least every 45 deg.  Anode information is used to
   determine from which azimuthal sector of the sky the ion entered.  The
   voltage settings for the elevation and energy analyzers are also
   reported so that all of the energy and angular information for a given
   particle is known before any time-of-flight measurement is made.  All
   signals from the start anodes are recorded as ion 'singles' data.  The
   angular mapping of the PEPE field of view as seen from PEPE's position
   on the spacecraft can be seen by combining the look angles with
   the coordinate definitions given in Figure 1.
 
 
   The central section of the stop MCP has its own high-speed
   discriminator which, when triggered, stops the TDC clock.  Each
   amplifier/discriminator consumes only 35 mW with transition times < 1
   ns. The amplifier/discriminators use balanced input differential
   voltages and have a 25,000 electron (25ke-) RMS equivalent noise with
   a 150ke- threshold level (equivalent to 75mV).  A minimum signal of
   300ke- is required from the MCP to achieve < 1 ns timing error.
 
 
 
 
   The TDC is similar to that used in the Cassini CAPS ion mass
   spectrometer (IMS).  To save power, the base oscillator runs at only
   83 MHz.  The timing resolution of 750 ps is achieved by routing the
   start and stop signals into a vernier delay line which determines the
   ion time-of-flight to full resolution.  Bin widths may vary as much as
   100ps in a repeating pattern because of variations in the width of the
   taps in the delay line.  The TDC range is 40 ns to 1536 ns with a
   pulse-pair resolution of 40 ns. The dead-time of the TOF read-out and
   sorting electronics is 1.74 microseconds. Singles data, on the other
   hand, are recorded with a dead time of 150 ns. The TDC includes its
   own built-in test (BIT) that allows all of the COBs to be tested by
   injecting a charge pulse at their inputs.  A variable start-to-stop
   delay is also incorporated so that timing measurements can be checked.
    This makes it possible to test the entire signal chain from the anode
   to the DPU with the BIT. Correct operation can be verified at
   integration time or in flight.  The TDC/FEE system also has its own
   ground support equipment (GSE) and simulator so that the TDC/FEE/TOF
   ion analysis section can be tested independently of the DPU and the
   DPU can be run with known inputs from a TDC simulator.
 
 
   Location:
   =========
   The PEPE optical system is rotationally symmetric about the central
   vertical axis and consists entirely of electrostatic elements. The
   optical system consists of three main functional elements: Angular
   deflection optics, energy/charge analyzers, and the LEF TOF optics.
   PEPE is located on a bracket at the upper (+Z) end of the +X-Y panel
   of the spacecraft (Figure 3) such that the central plane of the PEPE
   FOV lies in the spacecraft XZ plane. This gives PEPE access to both
   the nominal solar wind direction and to the DS1 Miniaturized
   Camera/Spectrometer (MICAS) boresight
   direction located along the spacecraft +Z axis. PEPE FOV coordinates
   are illustrated in Figure 1. The 3-axes of the PEPE coordinate
   system are parallel to those of the spacecraft.  Arrows in the
   figure indicate the sense of the numbering of angular pixels that
   make up the PEPE FOV. The angular coordinates for azimuth
   are defined between 0 and 359 degrees and correspond to the direction
   of arrival for both positively and negatively charged particles (e.g.,
   +90 deg azimuth indicates particles arriving from the direction of
   the spacecraft -Z axis). The angular coordinates for elevation
   are defined as the direction of arrival of positively
   charged particles. Because of the nature of the deflection electric
   field that controls the PEPE elevation viewing direction, the
   direction of arrival of negatively charged particles is the opposite
   of that shown in Figure 1.
 
 
 
 
   Operational Modes:
   ==================
   Energy/angle scan programs are controlled by
   re-programmable voltage tables stored in on-board memory. The ESA
   voltage is held constant at one of 120 levels while the deflection
   voltage is stepped over a program of 16 linearly-spaced voltages that
   follow Equation 5. Each step is held for 32 ms so that a full
   elevation scan of +/- 45 deg is covered in 512 ms. Out of the 32.0 ms
   interval, 4.0 ms is set aside as settling time for the high voltage
   supplies. After a deflection voltage sweep is completed, the ESA
   voltage is advanced to the next step. ESA steps are spaced at
   logarithmic intervals such that the energy range is covered in 120
   contiguous steps. Eight additional steps, during which the ESA is held
   at zero voltage, are used for detector background measurements. The
   complete elevation angle/energy range of PEPE is covered in 16 x 128 x
   0.032 s = 65.536 s. In order to enhance PEPE performance during the
   flyby of Asteroid Braille, a fast-scanning mode was created that
   covered a smaller region of interest (ROI) in energy/angle space (16
   deflection x 16 energy steps) using a shorter sampling interval of 8.0
   ms.
 
   Measured Parameters:
   ====================
   During each 32.0 ms sample
   interval, data from the 16 angle-sorted electron and ion
   counter/registers as well as valid ion TOF event and related logical
   event registers are accumulated in a double-buffered dual-ported
   memory in the data acquisition subsystem (DAQ). Data continue to be
   accumulated throughout one full angle/energy scan (65.5 s) at the end
   of which they are made available to the Harris RTX2010 microprocessor.
   The ion and electron data are stored in the MQ, TOF, and SINGLES
   memories that are operated in ping-pong fashion in order to allow the
   processor to move and operate on the previous cycle's data while new
   data are being acquired. An entire cycle of science data, including
   housekeeping and status information, is stored in processor resident
   memory to permit compression prior to formatting into CCSDS packets
   for transmission to the spacecraft data handling system.
 
 
   The PEPE data array contains a total of about 2.1 Mbytes of data
   acquired over 65.5 s. These data are transmitted to the ground at a
   maximum rate of 1024 bits/s thus requiring a compression ratio of
   roughly 32:1. Compression is achieved primarily by summing over
   adjacent elevation, azimuth and energy channels. In the solar wind
   survey mode, 4 adjacent elevation and azimuthal angular channels and
   over two adjacent energy channels are summed, thus reducing the
   angle/energy array from 16 x 16 x 120 elements to 4 x 4 x 60 elements.
   The TOF array is also summed over 4 adjacent channels.  This
   compression scheme can be re-programmed in many ways in order to place
   the emphasis of the transmitted data on specific regions of interest
   within the data array. A second option for controlling the transmitted
   data products is to reprogram the angle/energy voltage scan table,
   creating a special region of interest (ROI) within the scan space of
   the instrument. Summing over these data then produces a higher
   resolution energy/angle scan that is averaged over the same 65.5 s
   time interval. This option was used for observations during the
   close-approach section of  the asteroid flyby. Still a third option is
   to transmit the entire uncompressed data matrix over a period of 32 x
   65.5 s = 35 minutes.
 
 
   --------------------------------------------------------------------
                      Table 1. PEPE Performance Summary
   -------------------------------------------------------------------
   Parameter Range/Resolution       Performance             Units
   --------------------------------------------------------------------
   Energy    Range                  8.0 to 33,500           eV/e
             Range scan             1280 steps, log-spaced
                                    (8 used for background
                                     measurement)
 
 
             Resolution (electrons) 0.085                  (Delta)E/E
             Resolution (ions)      0.046                  (Delta)E/E
             Analyzer constant      13.07
 
 
   Mass      Range                  1 to 135                amu/e
             Resolution
             (straight thru)        ~4                     M/(Delta)M
             Resolution (LEF)       ~20                    M/(Delta)M
 
 
   Angle     Range (EL)             -45 deg to +45 deg
             Range scan (EL)        16 steps, linear
             EL deflection constant 6.7 x 10**5/(E/Q)       (deg/V)
             Range (AZ)             360 deg
             Solid angle coverage   8.9                     sr
             Resolution
              (electrons, EL x AZ)  256 pix @ 5 deg x 22 deg
             Resolution
              (ions, EL x AZ)       128 pix @ 5 deg x 5 deg
             Resolution
              (ions, EL x AZ)        32 pix @ 5 deg x 22 deg
             Resolution
              (ions, EL x AZ)        96 pix @ 5 deg x 45 deg
 
 
 
 
   Temporal  AZ x TOF                0.008/0.032            s
             AZ x EL  x TOF          0.128/0.512            s
             AZ x EL x EN x TOF      16.38/65.536           s
 
 
  Sensitivity*
             Electrons
             (deg x 22 deg pixel)  1.5 x 10**-4        cm**2 sr eV/eV
             Ions
             (5 deg x 22 deg pixel)  8 x 10**-5        cm**2 sr eV/eV
             Ions
             (5 deg x 22 deg pixel
              TOF @ +/-8kV)          3 x 10**-5        cm**2 sr eV/eV
 -----------------------------------------------------------------------
 *Estimated based on ray-tracing.
 
 
 
 
 
 
-----------------------------------------------------------------------
   Table 2. PEPE Resources
 
-----------------------------------------------------------------------
   Parameter              Resource                    Units
 
-----------------------------------------------------------------------
   Mass                   5.5                         kg Power
   (average)              9.6                         W Volume
                          7.25                        liters Density
                          0.83                        g/cm**3 Telemetry
   (commandable)      1024, 512, 250, 100, 50, 25     bits/s Operating
   range              -20 to +35                      C
 
 
 
-----------------------------------------------------------------------
   References: ===========
   Betts, R.R., Time of flight detectors for heavy ions, Nuclear
   Instruments and Methods; 1- 15 June 1979; vol.162, no.1-3, pt.II,
   p.531-8.
 
 
   Burgi, A., M. Oetliker, P. Bochsler, and J. Geiss, Charge exchange of
   low energy ions in thin carbon foils, J. Appl. Phys. 68, 2547-2554,
   1990.
 
 
   Fernandez, Y. R., L. A. McFadden, C. M. Lisse, E. F. Helin, and A. B.
   Chamberlin, Analysis of POSS images of comet-transition object
   107P/1949 W1 (Wilson- Harrington), Icarus 128, 114-126, 1997.
 
 
   Funsten, H. O., D. J. McComas, and B. L. Barraclough, Ultrathin foils
   used for low- energy neutral atom imaging of the terrestrial
   magnetosphere, Optical Engr. 32, 3090-3095, 1993.
 
 
   Harten, R. and Clark, K., The design features of the GGS Wind and
   Polar spacecraft, Space Science Reviews 71, 23-40, 1995.
 
 
   McComas, D.J., J.E. Nordholt, S.J. Bame, B.L. Barraclough, and J.T.
   Gosling, Linear electric field mass analysis: A technique for
   three-dimensional high mass resolution space plasma composition
   measurements, Proc. Nat. Acad. Sci., USA, 87, 5925-5929, 1990.
 
 
   McComas, D.J. and J.E. Nordholt, A new approach to 3-D, high
   sensitivity, high mass resolution space plasma composition
   measurements, Rev. Sci. Inst., 61, 3095-3097, 1990.
 
 
   McComas, D.J., J.E. Nordholt, J.-J. Berthelier, J.-M. Illiano, and
   D.T. Young, The Cassini Ion Mass Spectrometer, in Measurement
   Techniques for Space Plasmas, AGU Monograph Series, 102, ed. by R. F.
   Pfaff, J. E. Borovsky, and D. T. Young, AGU, Washington, DC, 187-194,
   1998.
 
   Nordholt, J.E., J.J. Berthelier, D.M. Burr, H.O. Funsten, R.
   Goldstein, J.M. Illiano, D.J. McComas, D.M. Potter, and D.T. Young,
   The Cassini Ion Mass Spectrometer: Performance metrics and techniques,
   in Measurement Techniques for Space Plasmas, AGU Monograph Series,
   102, ed. by R. F. Pfaff, J. E. Borovsky, and D. T. Young, AGU,
   Washington, DC, 209-214, 1998.
 
 
   Wang, J., D. Brinza, R. Goldstein, J. Polk, M. Henry, D. T. Young, J.
   J. Hanley, J. E. Nordholt, D. Lawrence, and M. Shappirio, Deep Space
   One investigations of ion propulsion plasma interactions: Overview and
   initial results, AIAA Paper 99-2971, 35th AIAA/ASME/SAE/ASEE Joint
   Propulsion Conference, June 1999.
 
 
   Young, D.T., S.J. Bame, M.F. Thomsen, R.H. Martin, J.L. Burch, J.A.
   Marshall, and R. Reinhard, 2?-radian field-of-view toroidal
   electrostatic analyzer, Rev. Sci. Instrum., 59, 743-751, 1988.
 
 
   Young, D. T., et al., Cassini Plasma Spectrometer investigation,
   Measurement Techniques for Space Plasmas, AGU Monograph Series, 102,
   ed. by R. F. Pfaff, J. E. Borovsky, and D. T. Young, AGU, Washington,
   DC, 237-242, 1998a.
 
 
   Young, D. T., et al., Miniaturized Optimized Smart Sensor (MOSS) for
   space plasma diagnostics, in Measurement Techniques for Space Plasmas,
   AGU Monograph Series, 102, ed. by R. F. Pfaff, J. E. Borovsky, and D.
   T. Young, AGU, Washington, DC, 313-318, 1998b.
 
 
   Young, D. T., J. E. Nordholt, and J.-J. Hanley, Plasma Experiment for
   Planetary Technology: Technology Validation Report, to be published as
   JPL Tech. Rept., 2000.
 
 
   Zhigarev, A., Electron Optics and Electro-beam Devices, Mir Publ.,
   Moscow, 1975.