SPEDE instrument paper
Draft / 08.04.2004

Anssi Malkki, Walter Schmidt, Maria Genzer, FMI/ Space Research
More authors TBA.

Abstract

This paper describes the SPEDE (Spacecraft Potential, Electrons, and
Dust Experiment) instrument on-board the SMART-1 satellite of European
Space Agency. SPEDE is a Langmuir probe plasma instrument, making
measurements from which plasma parameters in the surroundings of the
spacecraft can be derived. In the analysis and interpretation of data,
knowledge of many instrument details are important. The aim of the
paper is to describe the SPEDE implementation to the accuracy that
most questions of the users of the data are answered.  In addition to
general parameters such as probe geometry, this also implies a quite
thorough description of the electronics. The digital electronics is
implemented using a novel design with a voltage-to-frequency
converter, performing essentially integrating measurements over a
given time window. Known features of the implementation are also
discussed, where appropriate. 

1 Introduction

SMART-1, Europe's first mission to the Moon, has a dual task of at the
same time of being a technology demonstration and a science
mission. The most important technology experiment for ESA is for the
first time using a Hall type electric propulsion thruster as the main
engine. The advantage of the electric propulsion (EP hereinafter),
compared to traditionally used combustion thrusters, is the low mass
that is required to be carried as propellant. The propellant is Xe
gas, which is ionised, and accelerated in the thruster by energy from
the solar panels. The engine can be operated for very long times
without interruption, and the thrust can be controlled very
accurately. The drawback of the electric propulsion engine is the
small thrust (approximately 70 mN for the SMART-1 engine), which means
that the acceleration that the engine provides is very small. Thus, on
a short trip (like the mission to the Moon), this is not the most
cost-efficient solution. However, on a longer (inter-planetary)
mission, the EP approach is very attractive.

The ESA is preparing a mission to Mercury in year 2012 timeframe. This
mission, called BepiColombo, will use a Hall thruster as the
propulsion engine. In order to validate the technology and gather
experience on any possible side effects the engine and its exhaust may
have on the spacecraft and the payload, the SMART-1 mission is serving
as a testbed for BepiColombo.

The target of SMART-1 is the Moon. The spacecraft is carrying a suite
of imaging instruments designed for measuring the Moon using different
methods and wavelengths. In addition, there are technology
experiments, testing novel communication methods, and two plasma
instruments to monitor the plasma environment of the spacecraft. 

The two plasma instruments of SMART-1, SPEDE and EPDP (Electric
Propulsion Diagnostics Package) complement each other. EPDP is
designed specifically to monitor the dense plasma that is the result
of the operation of the EP engine. The instrument design parameters
are optimised for relatively dense plasmas (electron and ion densities
between 1010 and 1014 m-3). EPDP has one Langmuir probe (LP) for
electron measurements and a retarding potential analyser (RPA) for ion
measurements. Both these probes are located very close to the EP exit,
however not in the plasma beam itself. (See figure 1.) In addition to
the plasma sensors, EPDP has sensors monitoring possible contamination
on the solar panel and spacecraft body surfaces. The plasma sensors of
EPDP are not operated in a monitoring mode in the sense that they
would sample the plasma continuously. Both the LP and the RPA perform
data acquisitions by (time-tagged) command. This is different than the
operation of SPEDE, where nominally data is collected in a low
sampling rate continuously as described later.

This paper is organised as follows. In section 2 we present the
objectives of the SPEDE instrument, and how they have been translated
into long-term operations plans for the duration of the SMART-1
mission. Section 3 contains the instrument description, starting with
a brief introduction to Langmuir probes, and continuing with probe
design, electronics design and electronics readout description. Known
instrument features are also included, to the extent known at the time
of the writing of this document. Section 4 presents the operations:
basic mode configurations and the data acquired. List of data products
is also included in this section. The paper concludes with a summary.

2 SPEDE Instrument Objectives

The SPEDE (Spacecraft Potential, Electron and Dust Experiment)
experiment, consisting of two electric sensors and an electronics
unit, will measure the electron flux and wave electric fields. The
cylindrical sensors are mounted on the tips of two 60-cm booms,
located at the +X and -X faces of the spacecraft (see Figure 1.). The
sensors are connected to the electronics unit via a single triaxial
cable each without any active electronics outside the board; the SPEDE
electronics is housed on two electronics boards, located in a box
inside the spacecraft body.

Insert figure 1: SMART-1 with EPDP and SPEDE probes indicated.

The mission of the SPEDE experiment is two-fold: it will monitor
- The disturbances (electron flux, wave electric fields, and
spacecraft potential variations) induced by the propulsion system, and

The variability of the electron density and wave electric fields
during the Earth spiralling and cruise phases and during the Moon
phase

2.1 Monitoring of disturbances produced by the SMART-1 propulsion
  system

Gas releases, most commonly from thrusters used to control the
spacecraft velocity and attitude, can disturb observations of some
instruments as well as contaminate the spacecraft structure. Because
of possible interference effects and spacecraft contamination can be
detrimental to sensitive instruments, it is important that the
disturbances produced by the thruster operations in the spacecraft
environment are monitored. This is especially important for the
SMART-1 mission, for the first time using electrical propulsion on an
ESA mission. Charged clouds expanding from the propulsion system may
introduce a variety of phenomena when interacting with the ambient
plasma and the spacecraft surface. These include 

- variations in the spacecraft potential and electron flux, 
- contamination of the spacecraft surfaces for an extended period of time,
and 
- generation of wave electric fields.

Especially, large effects in the spacecraft potential can be
immediately observed if, for some reason, the exhaust ions are not
properly neutralized by the cathode electron emitter. It is also
important to gain knowledge of the reactions of the plasma environment
to changes in EP engine parameters of operation. 

2.2 Monitoring of electron density in the inner magnetosphere

In the first part of the SMART-1 mission, the Earth spiralling phase,
the spacecraft will be accelerated by the ion propulsion engine and
remain in the inner magnetosphere. In addition to monitoring the
effects of the propulsion as described above, the SPEDE observations
are used for measuring the distribution of thermal plasma of the
plasmasphere whenever the EP is not operating. Particularly the
measurements aim at detecting the position of its outer boundary, the
plasmapause, usually located at a distance of 3-7 Earth's radii at the
equator.  

As long as the perigee of the orbit is less than 3-4 Re (20 000 km),
the plasmapause is crossed twice per orbit. When the perigee is
between 20 000 and 40 000 km, the plasmapause is not always
encountered, particularly so during magnetic storms, when the
plasmasphere becomes smaller in size. 

2.3 Monitoring of plasma density and waves in the Earth's
  magnetosphere and in the Solar Wind

After the perigee of the orbit is raised outside of the plasmasphere
boundary (see above), SPEDE observations will concentrate on low-rate
monitoring of the magnetospheric and solar wind plasma. These regions
have been extensively investigated on earlier missions with plasma
instrumentation optimised for tenuous plasmas, and no scientific
break-throughs are expected. The measurements will consist of
monitoring variations of plasma density by operating the instrument in
a constant-bias low-sampling mode.

2.4 Monitoring of space weathering of the Moon

The target of the SMART-1 mission is the Moon which has no magnetic
field and atmosphere. Therefore, it is continuously exposed to the
interplanetary space environment. The fast solar wind stream hits the
dayside lunar surface and is possibly capable of lifting up small dust
grains from the surface. Behind the moon, the solar wind produces a
wake that is more tenuous than the solar wind. 

On lunar orbit, SPEDE observations are used for studying solar wind -
moon interaction processes. The uplifted dust particles can be
detected as variations in the spacecraft potential, as the particles
are ionised when hitting the spacecraft surface.  A high sampling rate
will be used at the region of the predicted wake boundary, to obtain
the best data both for dust impact detection and studies of the plasma
density and turbulence at the wake. Optimised modes initiated by
time-tagged commands will be used.

3 Instrument description

The general philosophy of the SMART-1 mission follows the principles
of "faster, cheaper, better - SMARTer". As SMART-1 is a testbed for
new technologies, also many payload instruments are implemented using
novel technologies for relatively conventional measurements. The
technology aspect also holds for the SPEDE instrument.

From the measurement principle point of view, SPEDE is a conventional
Langmuir probe instrument. The two SPEDE probes are covered with
titanium nitrite (TiN), in order to ensure minimal side effects due to
contamination and ageing of the surfaces. The electronics is
implemented using as much digital electronics as possible. The final
result is an instrument with a wide dynamic range, good resolution,
and at the same time low mass, low power consumption, and low cost. 

3.1 Measurement principle

The principle of plasma measurements using Langmuir probes is as
follows. Since plasma consists of free charges (electrons and ions),
it seeks for quasineutrality: Even a slightest imbalance in charge
results in a quick neutralising current, normally carried by the more
mobile electrons. The system consisting of the spacecraft, the
probe(s), and the plasma, seeks for equilibrium through generating
electric currents either towards or away from the probe. At the same
time, all currents to the spacecraft-probe system must balance each
other, which defines the equilibrium potentials the spacecraft and the
probe must be in. The current between the probe and the plasma can be
measured, and with modeling, plasma parameters such as electron and
ion densities, electron temperature, and the potential of the
spacecraft, can be derived. The theory of the measurement has been
well developed and thoroughly tested, and the general behavior of
Langmuir instruments is well understood.  However, each
spacecraft-instrument combination is unique, and adjustments to theory
are necessary to accurately model the behavior, and obtain best
possible results.

The SPEDE instrument - like any Langmuir probe instrument - does not
directly measure any plasma parameters. There are only two directly
measurable parameters: the current to or from each probe, and the
potential difference between a probe and the reference voltage at the
instrument. Depending on the configuration of the instrument, we can
measure either the current or the voltage. All other results are
derived from these parameters. The two measurements provide
possibilities for different derived products, as will be described
below. 

3.2 Basic modes of operation

The basic modes of operation are 
- measure current as function of bias voltage (Langmuir sweep), 
- measure current with constant bias voltage (Langmuir mode), 
- measure voltage with respect to (spacecraft ground) reference
voltage (voltage mode).
- plasma wave measurements (in voltage mode).
These modes will be discussed below.

3.2.1 Langmuir sweep

In the Langmuir sweep mode, the bias potential is varied stepwise,
with programmable voltage step size and voltage range. This
measurement provides the characteristic Langmuir response curve, to
which one can fit plasma parameters (Figure 2.).

 Insert Figure 2: Textbook Langmuir curve.

From the Langmuir response, one can use a parametric fit, and derive
electron density, electron temperature, ion density, ion temperature
and spacecraft potential. The accuracy of the estimates depends on the
performance of the instrument, but also on background knowledge of the
plasma. Not all parameters are independent, but depend for example on
ratio of temperature and density, and thus a good fit may not be
unique. However, in most cases ambiguities can be solved with a priori
knowledge of what can be assumed.

The shape of the curve is very sensitive to electron density and
temperature (actually the ratio n/?T), around the region where ion
current (negative in the plot) changes to electron current. This is
the region one attempts to sample in order to obtain the most from the
measurement.

The SPEDE instrument can (by hardware design) sample a voltage range
from -13 V to +14,4 V with reference to the spacecraft ground. The
voltage can be varied in 255 steps with spacing that is smallest
around zero, and sparser closer to the extremes. The range between 4,4
V and + 14 V is covered by one step (see Figure 3.). This decision was
made to have the best sampling close to zero, where the curvature of
the Langmuir response was expected to be tightest. The last point
close to +15 V would then give the (saturation) electron current.

Insert Figure 3: SPEDE voltage step spacing.

In practice, the SMART-1 spacecraft potential with respect to the
surrounding plasma tends to be between 17 and 24 V negative. In Figure
2 the curve crosses I=0 at X V, corresponding approximately to the
plasma potential. To the left of this point, the net current is
towards the probe (emitting electrons or collecting ions). If the
spacecraft potential is negative, the curve shifts to the right in
order to the crossing point to arrive at the plasma potential. In the
case of SMART-1 and SPEDE, this point moves to around 18 volts, beyond
the range of bias voltages SPEDE can reach. This reduces the
possibilities for deriving parameters from SPEDE sweeps, as discussed
more in detail later.

3.2.2 Langmuir mode
The second mode (Langmuir mode) provides data of plasma flux
variations. Depending on bias voltage, either ion (negative bias with
respect to the plasma) or electron (positive bias) current is
collected. Basically one attempts to select a working point on the
Langmuir curve that is most sensitive to variations in either plasma
density or spacecraft voltage. As discussed above, the ability of
SPEDE is reduced by the SMART-1 spacecraft potential. In figure 4 a
family of Langmuir curves has been plotted, assuming a plasma with Xe+
as the dominant ion species. The red rectangle indicates the range of
voltages and currents that can be obtained with SPEDE, assuming a
spacecraft potential of -17 V. We see that in this situation, with any
parameter setting that can be obtained for SPEDE, we sample the ion
side of the Langmuir curve, providing information about the
temperature and density of the ions. 



Figure 4: Family of Langmuir curves, with SPEDE measurement range indicated.

3.2.3 Voltage mode

In the third basic mode, the probe is in a passive mode measuring
variations in the potential difference between the probe and the
reference voltage. As the probe potential should not be modified by
the measurement itself in this mode and as also small scale variations
should be detected, the voltage is sampled with very high input
impedance and capacitive offset suppression 

In SPEDE, there is not bias current generator, and thus in the voltage
mode the potential of the probe is floating freely. As discussed by,
e.g., Pedersen et al. (1984), the probe potential is closer to the
plasma potential, when an appropriate bias current is applied. Thus
the SPEDE probes cannot be taken as to accurately float in the plasma
potential. On the other hand, the probes are very close to the
spacecraft, and thus photoelectron effects, as well as the effect of
the spacecraft structure on the electrostatic potential of the
surrounding plasma would introduce uncertainties in interpreting DC
voltage measurements. As a consequence, voltage mode measurement is
used only for measuring wave fields generated by plasma fluctuations.

3.2.4 Wave measurements

TBW, mode not used in current data set.

3.3 Probe design

The probes are made of two 0.1mm Titan foils at the end of the 60cm
long slightly conical carbon fiber booms. The foils are sputtered with
TiN (as mechanically and chemically inert surface material), insulated
from one another and the boom carbon fiber material by 0.05mm adhesive
Kapton foils. The inner layer has a length of 150 mm, the outer layer
of 100 mm (acting as the probe), leaving 25 mm free on both sides as
guard. The purpose of the guard is to prevent capacitive leakage
current between the sensor foil/wire and the spacecraft ground
including boom structure. The boom has a diameter of 22mm at the tip,
which is covered by a lid from the same sensor foil material,
electrically connected to the guard layer. For low density plasma the
10 cm x 2.5 cm probe cylinder can be considered spherical. The
mechanical layout of the boom and probe is presented in Figure 5.

Insert Figure 5: SPEDE probe mechanical design. 

The probe is connected to the electronics via a double shielded wire
(Triax) with the outer shield at the electronics end connected only to
the analog ground as electrical shield. The inner shield is connected
to the inner guard layer of the probe foils, which is actively kept at
the sensor potential. The innermost cable wire connects the outer
sensor probe foil with input of the first amplifier stage and the
current measurement resistor.

The mechanical design of the booms was determined by spacecraft
constraints: a very stiff, but with 128g lightweight design protruding
only 60 cm from the s/c panel was enforced by the satellite's launch
configuration. As also no moving parts were allowed, the SPEDE probes
are located rather close to the s/c surface compared to the original
suggested locations at e.g. the tips of the solar panels. This limits
the possibilities of monitoring natural plasma, but enhances SPEDE's
capability of monitoring possible impacts of the thruster on
scientific instruments mounted along the s/c panels. 

3.4 Electronics design

The electronics is built around one Field Programmable Gate Array
(FPGA) containing all processor functions, and two separate analog
channels with operational amplifiers and
Voltage-to-Frequency-Converter (VFC). The processor functions in the
FPGA include the boot code, real-time clock and the analog control and
measurement functions. For Langmuir measurements, the bias voltage for
each channel is generated by an FPGA-driven resistor network dividing
the channel-specific reference voltage in 256 different ways,
corresponding to digital control values between 0 and 255. A buffered
level shifter provides low-impedance bias voltages in the range of
-13V to +14V. The resistive network is selected in such a way that
small step sizes are possible between -3V and +3V, with increasing
steps beyond this range up to +4.5V for control value 254. One
additional bias voltage of +14V is provided for measurements in the
expected electron saturation current regime (electron density
measurements), assuming spacecraft potentials inside the +/- 3V range.

The probe current is connected through the measurement resistor and
the bias generator to instrument ground, causing a voltage drop across
the resistor. This voltage drop is monitored as potential difference
with respect to reference ground by high-impedance amplifiers. The
resulting voltage difference is converted to a frequency in the
VFC. The range of the VFC corresponds to 100 Hz for large positive
voltages to 150 kHz for large negative voltages with 0V close to
50kHz. 

For potential measurements, the connection between the 200 Ohm
resistor and the rest of the electronics is decoupled by a
relay. Pre-amplifier 1 connects directly to a specialized amplifier
which in this case also drives the guard. The final amplification
stage with connection to the VFC is separated at its input by a
capacitor to suppress offset voltages. The functional electronics
diagram is shown in Figure 6.




Figure 6: SPEDE measurement electronics diagram.

The measurement result is the frequency coming from the VFC, compared
with the internal fixed 16MHz clock frequency. Using a measurement
window defined by the 16 MHz clock as e.g. 1 sec the number of
observed VFC pulses correspond directly to the VFC frequency averaged
over the measurement window. Defining the measurement window by
e.g. 150 VFC pulses, the number of observed 16 MHz clock pulses give
the average pulse length of the VFC signal. The nominal measurement
windows are 998 ms and 4 ms, or alternatively 20 or 150 reference
clock pulses.

3.4.1 Technology aspects

Using a Voltage to Frequency conversion to digitalize the analog
signals offers several advantages: Covering the full dynamic range of
the front end electronics, the resolution can be optimized to the
signal range of interest. Opposite to the sample measurement method of
standard ADCs, a VFC is an integrating device covering all short
duration signals if they deviate significantly from the background. A
relevant example in the case of SMART-1 is the EP valve operation and
the same advantage is expected to be very useful for dust impact
detection in Lunar orbit. In addition, signals that deviate only
little from zero current can be measured with high resolution and
short integration times, as they result in frequencies close to the
center frequency of the device (50 kHz). The bit-resolution can also
be optimized for the observation objectives, as there is no principle
restriction to the number of bits generated per conversion. SPEDE uses
between 8 and 22 bit data depending on the observation mode and
measurement data and integration times between 4ms and 1 sec.

In SPEDE design, as few analog components as possible were used as
possible. This was because in many cases their availability, radiation
resistance, size, mass, flexibility and power consumption are in
contradiction to the design drivers for the instrument. Implementing
most of the electronics in a programmable digital device also allows
major design changes after most of the device is already designed or
even built. This approach makes future miniaturization possible,
including implementing all components on a single die carrier. 

Replacing a commercial Digital-to-Analog-Converter against a digitally
controlled resistor network offers the possibility to design for
different resolution in the various voltage ranges. The SPEDE approach
is also considered better in terms of power consumption and radiation
sensitivity.  For Langmuir Probe measurements one large negative bias
voltage for ion-density and one large positive voltage for electron
density would be sufficient, while the bias voltage step width close
to the area with fast current changes defines the precision of the
electron temperature and composition analysis. The resistor network
that is used in SPEDE consists of one resistor in series with 8
parallel resistors, which can be connected to ground via 8 separate
control lines. Using 8-bit control values to drive these lines, a wide
range of resistor ratios can be defined. With the network connected to
a 2.5V reference, an inverting level shifter moves the resulting
voltages between 2.5V (all lines open) and 0V (all lines plus
short-circuit connected to ground) to -14.4V and +14.4V. Due to
restrictions of the amplifiers, the lowest useful voltage is -13V,
corresponding to a control value of 5.


4 Operations and Data

4.1 General description

SPEDE has ten different operational configurations ("modes") stored
simultaneously in the instrument memory. Mode 0 is idle mode for Safe
or no telemetry configuration. Mode 1 is the housekeeping mode, which
is activated when SPEDE is ON, if not other mode is commanded. This
mode collects data in Langmuir mode from both probes, with one sample
per minute.

All modes can be activated by simple spacecraft command. Each mode
definition consist of 4 parameters plus a control byte for each of the
two probes, defining the mode of operation. The two probes (+/- X) can
be programmed individually, with timing of measurements being
identical for the two probes. The details are given in the SPEDE User
Manual (add ref here). The mode configuration parameters define
whether a Langmuir or Voltage measurement is used, the time step
interval, number of measurements in one measurement sequence, time
interval between measurement sequences, and in the Langmuir mode also
the voltage step, and whether a two-way measurement (sweep with
hysteresis measurement) is performed. The voltage range can also be
adjusted. For each measurement and probe, different sampling
philosophy (pulse length or frequency measurement; see electronics
description above) can be used. 

As discussed above, SPEDE performs measurements within a time window,
thus defining the frequency output from the voltage to frequency
converter. Thus the time between samples cannot be shorter than this
time window. In practice, the shortest time windows are of the order
of 4 milliseconds, and this is not a serious limitation. For each
measurement, one needs to consider the balance between measurement
reading accuracy (better with a long window) and measurement
duration. For sweep measurements, shorter windows are preferred,
whereas for Langmuir measurements, a long window with 1 second is
usually used. The integrating measurements record even small signal
variations during the measurement time window, which makes them
suitable for detecting small variations in Langmuir current.

Description of wave measurement TBA.

Parameters for each measurement are included in the header of the data
file, and actual mode configuration tables or mode numbers do not need
to be available for data analysis.

SPEDE is operated using Input Timelines (ITLs), which are converted
to Payload Operation Requests by Smart-1 Science Operations Centre, and
uploaded to SMART-1 during ground contact. For most operations,
standard timelines with SPEDE operations related to orbital or
spacecraft events (especially EP operations), are used. For specific
activities such as calibration measurements and configuration table or
software updates, dedicated timelines are developed. Timelines during
different phases of the mission will be discussed in below.

4.2 Operations timelines

During the first two weeks of the mission, instrument commissioning
was performed. The commissioning activities consisted of health checks
and in-flight calibration measurements coordinated with the EPDP
instrument. The result of the commissioning was that SPEDE was fully
functional, and that satisfying agreement between the results of the
two plasma instruments could be reached. Mode configuration tables
were also adjusted, in order to improve the scientific performance. 

After the initial commissioning period, the SMART-1 mission entered a
more than three month phase of almost continuous EP thrusting. This
was performed to raise the apogee of the orbit above the Earth's
radiation belts, in which the energetic particles are known to be
harmful for both the spacecraft subsystem electronics and the
instruments. During the thrusting phase, SPEDE was operated almost
continuously, collecting data according to the EP monitoring
timeline. These data need to be correlated with simultaneous
measurements of the EP engine housekeeping parameters, as well as EPDP
measurements.

As an example, for EP operations monitoring the nominal sequence is to 
- measure the ambient plasma before EP switch-on with a full-range
Langmuir sweep, 
- monitor switch-on transients in a fast sampling Langmuir mode with
constant bias,
- perform another sweep to obtain plasma parameters when EP is on,
- measure Langmuir current during EP operation with a constant bias
and slower sampling,
- before EP switch-off perform the switch-on measurements in a
reversed order, and
- return to ambient plasma measurement mode.

After the initial thrusting phase, SMART-1 spends a lot of time in a
spiralling orbit in the Solar Wind and magnetosphere, and the EP
engine is only used during part of the orbit. When the EP is off,
SPEDE measures the ambient plasma. The selected operational timeline
uses the housekeeping mode during most of the orbit. Faster sampling
has been used on five 2-hour periods located at regular intervals
around orbit, and complemented with Langmuir sweeps at the beginning,
middle, and end of each fast sampling period. As discussed above,
SPEDE is not fully optimised for measuring low density plasmas, but
data is collected to monitor the performance of the instrument, and
compare measurements with other satellites at the vicinity when
possible.

<Sample ITL TBA>

After insertion to Lunar orbit, the planned operations will include
high frequency sampling in Langmuir mode plus wave measurements around
the predicted location of the boundary of the kinetic wake behind the
Moon. At these regions, dust impacts may be detected, in addition to
novel data on plasma turbulence behind the moon. A dedicated timeline
will be developed for that purpose.

At suitable intervals and occasions, in-flight calibration and
performance monitoring measurements are performed. In order to monitor
the instrument response in vacuum, a background calibration is
performed when SMART-1 is in eclipse, and its location is predicted to
be inside the extremely tenuous lobe part of the magnetotail. A
corresponding reference measurement has been performed on ground.

4.3 Measurements to products: sweep, voltage, and current

The Langmuir measurements produce data with probe current as a
function of time and probe bias voltage. If the probe is operated with
constant bias, the result is a time series of probe current,
reflecting the plasma flux variations. If in sweep mode, time is not
an important parameter to the same extent, and voltage vs. current
gives the Langmuir response in the plasma.

For voltage measurements, the data provides probe voltage vs. time and
for plasma wave measurements, wave energy in a given frequency
band. For data products, see Section X below.

4.4 Derived quantities

The plasma parameters that can be derived from the measurements,
grouped by measurement modes as described above in section 3, are as
follows. 

4.4.1 Langmuir sweep

From the Langmuir sweep, one can derive
- electron (and ion) density and
- electron temperature, and
- spacecraft potential
assuming that the plasma composition is known. Electron temperature
and density are related, with the ratio Te/ne basically determining
the shape of the curve in the range where the current is most
sensitive to the bias voltage. The best fit may also in some cases not
be unique. Often, however, one can fix one parameter to a limited
range, using a priori information, and obtain a solution. For this
evaluation it is assumed that the probe current is given by the
instrument. In fact the instrument compares the voltage drop caused by
the probe current through a measurement resistor with the voltage
difference between reference voltage and reference ground. While the
reference is rather stable, a temperature dependence in the order of
one per mille can be observed and has to be taken into account.

The SMART-1 spacecraft equilibrium potential is strongly negative
(typically between 17 and 24 V, as derived from EPDP Langmuir sweeps),
and thus goes beyond the bias voltage range of SPEDE. This means that
we can only seldom measure anything else than an ion current, reducing
our possibilities to derive electron parameters or the spacecraft
potential. With the SPEDE range of voltages and currents illustrated
in Figure 4, we can that for the plasma used in modeling for that
figure (mi = 54 amu), the range of densities that can be measured is
1010 to 1014 m-3. In Figure 4, a spacecraft potential of 17 volts has
been assumed. 

The measurements are thus almost always in the ion current regime, and
up to relatively high densities result in probe currents within the
SPEDE accepted range. At low densities, the measurements are limited
by measurement noise, which is of the order of 20 nA, resulting in
lowest reliable ion density of 1010 m-3. For comparison, the
saturation photoelectron current for a proton plasma is of the order
of our measurement noise (~ 20 nA).

4.4.2 Langmuir mode

In the Langmuir mode, the data is interpreted as electron flux
variation (or ion flux, due to strongly negative spacecraft
potential), which is proportional to electron (ion) density
variation. Again, this interpretation is not unique, since the current
is coupled with plasma temperature and s/c potential variations, the
sensitivity depending on the bias voltage used.

4.4.3 Voltage mode

The voltage mode of SPEDE provides measurements of 
- s/c potential variations, and
- wave electric fields.
The sampling frequency can be set as discussed in Section 4.1
above. The shortest time between samples that can be set is 4 ms,
resulting in 250 samples/ second. This means that fluctuations above
125 Hz are undersampled, which needs to be taken into account in the
analysis and interpretation of the data. This limitation of course
does note apply to the wave mode measurements, but there the waveform
is not transmitted to ground.

This mode is intended for monitoring fast varying probe potential as a
consequence of thruster parameter changes, impact of dust particles
with the s/c or interaction with varying solar wind. While the actual
measurement is independent of slowly varying s/c potential, an offset
beyond 13V can drive the input amplification stage into
saturation, making any measurement impossible. Therefore this mode was
not implemented for scientific purposes on a routine basis.

4.5 Data products
4.5.1 Level 1b
o probe readings in raw units 
4.5.2 Level 2
o probe current
o probe voltage
o wave spectrum
4.5.3 Level 3 derived parameters
o only from sweep mode

5 Summary
We have presented the SPEDE instrument on board SMART-1. At the time
of writing this paper, SPEDE has been operating for 6 months, and a
lot of data has been collected. Analysis is underway, and calibrated
and quality checked data will be made available at the archive of
ESA as it becomes available.

6 References

Pedersen, A., C.A.Cattell, C.-G. Falthammar, V. Formisano,
P.-A. Lindqvist, F. Mozer, and R. Torbert, Quasistatic electric field
measurements with spherical double probes on the GEOS and ISEE
satellites, Space Sci. Rev., 37, 269-312, 1984.

More references TBA.

1 / 11