Instrument Overview
===================
 
See [ISRAELETAL2002]
 
ACP's main objective is the chemical analysis of the aerosols in Titan's
atmosphere. For this purpose, it will sample the aerosols during descent and
prepare the collected matter (by evaporation, pyrolysis and gas products
transfer) for analysis by the Huygens Gas Chromatograph Mass Spectrometer
(GCMS). ACP's products transfer line (PTL) directly interfaces with an
ACP-devoted GCMS feed tube. GCMS is used by ACP for about 20% of its operating
time, based on a cooperative agreement between Service d'Aeronomie du CNRS
(ACP) and NASA Goddard Space Flight Center (GCMS).
 
Scientific Objectives
=====================
 
The approach for selecting ACP's scientific objectives was to rely on the
micro-physics models and on the results of specific laboratory studies.
Two methods of producing Titan-like aerosols in the laboratory have been
investigated. The first involves identifying these aerosols within the complex
organic material that is often observed in the laboratory from simulating the
photolytic and radiolytic processes expected in Titan's atmosphere. This
material roughly corresponds to what Sagan's group labels 'tholins'. The
second route for producing the yellowish aerosols that might correspond to
those of Titan is by polymerising organic molecules using UV. The best
candidate is C2H2, which polymerises more easily than C2H4 or HCN.
These polymers are expected to form the aggregates that are needed to
reconcile polarimetric and photometric results [WEST&SMITH1991;
CABANEETAL1993; RANNOUETAL1994]. The optical properties of tholins and C2H2
polymers are in agreement with Voyager's observations. The chemical processes
leading to these different particle types are not expected to occur in the
same altitude ranges, owing to the diversity of the possible energy sources
and their related vertical distribution. The polymerisation of C2H2 through
the action of solar UV photons is expected, from current modelling, to yield
C2nH2 at altitudes greater than ~ 500 km. The synthesis, in large amounts, of
polymers of the form (C2H2)n in the lower stratosphere may be considered as
rather unlikely, but depends on how the acetylene polymerises long-ward of 190
nm [CHASSEFIERE&CAB1995]. The formation of C-H-N tholin oligomers by
the action of suprathermal Saturn plasma electrons, around 900-1000 km, or of
energetic radiation belt particles at 350-400 km altitude, is shown to be the
most plausible mechanism for explaining the haze's formation. Laboratory
experiments providing the best reproduction of Titan's stratospheric
conditions are needed to test, for example, the gas phase behaviour of tholin
aerosols, and the optical properties of tholin or poly-C2H2 particles.
In the low stratosphere, aerosol particles settling down from the upper levels
may act as condensation nuclei. Below ~ 80 km, this should lead to the
deposition of thin layers (< 0.01 um) of condensed gases (e.g. HC3N, HCN,
C4H10) on the sub-um particles. This first stage is followed by a more
consistent increase in the particle size, up to a few um, where the
condensation of CO2, C3H8, C2H2 and C3H6 occurs, between 67 km and 63 km
[FREREETAL1990]. The main condensation processes occur near the
tropopause, with the condensation of C2H6 below 62 km, and in the troposphere,
where CH4 and N2 condense on the resulting particle, which may yield cloud
droplets ( ~ 100 um, see Toon et al., 1992). Another source of condensation
nuclei may be the particles produced directly at these altitudes by the
radiolysis of organic gases. The energy deposition of cosmic rays peaks near
60 km and, as on Earth, lightning may exist in the low troposphere.
Tholins obtained from the sparking of He-CH4 and N2-CH4 mixtures have been
studied with the ACP experiment in mind [ISRAELETAL1991; RAULINETAL1992].
The pyrolysis gas chromatography (Py-GC) of tholins produced at
77 K shows them to include saturated and unsaturated carbon chains in their
structure; N-containing groups appear in the case of N2-CH4 sparking. The
thermal desorption profile of these tholins clearly showed two peaks, the
first corresponding to the distillation of condensed species, the second to
the pyrolysis. The mass spectrometer (MS) study of the evaporated oligomers
and pyrolysis products showed that a wide range of alkylated aromatic
compounds evolved from the sample, which indicates that such solids contain a
3-D polymer with a high degree of branching.
In other work [EHRENFREUNDETAL1995], the main peaks related to
N-compounds observed in the GCMS are HCN, CH3CN, C2H5CN and longer chain
nitriles. Only the saturated hydrocarbons are observed. The study of the
evolution with temperature shows that HCN is a dissociation product during the
whole Py-GC analysis process, which may imply that the nitriles can form
thermostable structures in the tholins.
From the above and from previous assessment, two classes of scientific
objectives have been adopted:
  Primary objectives, can be satisfied from measurements made during Probe
  entry by ACP coupled with GCMS:
    1. Determine the chemical makeup of the photochemical aerosol, i.e. infer
    the relative numbers of constituent molecules (C, H, N, O) composing the
    aerosol.
    2. Obtain the relative abundances of condensed organics (e.g. C2H2, C2H6,
    HC3N, HCN) in a column average throughout the stratosphere. Compare with
    the abundance of constituent molecules in the aerosol nucleation sites.
    3. Obtain the relative abundances of condensed organics (principally CH4,
    plus organics listed above) in a column averaged over the upper
    troposphere. Compare with the abundances of constituent molecules in the
    aerosol nucleation sites.
 
  Secondary objectives of fundamental interest can also be satisfied with the
  aid of additional information acquired by other Probe instruments (mainly
  the Descent Imager/Spectral Radiometer):
    4. Obtain absolute abundance for all condensed species, averaged over the
    stratosphere and upper troposphere respectively.
    5. Determine the mean sizes of aerosol nucleation sites averaged over the
    stratosphere, and compare with those in the upper troposphere.
    6. Detection of non-condensable species, such as CO, eventually trapped
    in aerosols.
 
Functional Description
======================
 
The gas products transferred from ACP are analysed by GCMS (NIEMANNETAL2002).
A sampling system is required for sampling the aerosols in the 135-32 km and
22-17 km altitude regions of Titan's atmosphere. These altitude ranges refer
to the Probe's nominal descent profile [LEBRETON&MATSON1997].
The sampling system requires an inlet (sampling) tube (ST) extending beyond
the boundary layer (estimated to be few mm thick): the end protrudes 28 mm
from the Probe's fore dome. During sampling, the collecting target's
temperature must be as close as possible to that of Titan's atmosphere in
order to help retain the more volatile aerosol and cloud particle components.
The target is a filter (FIL), made in stainless steel (Beckaert ST10), that
can be moved along the inlet tube. When the filter is in its sampling
position, its front face extends a few mm beyond the inlet tube. This
increases aerosol collection by direct impaction at high altitude, where the
pump does not operate (see below).
Before descent, the filter is held in its storage position inside the oven
(OV). During descent, a mechanism (FIM) can move the filter to its sampling
position and return it into the oven.
The oven is a pyrolysis furnace where a heating element (OH) can heat the
filter and hence the sampled aerosols to 250 deg C or 600 deg C. A motorised
gate valve (GV) can be activated to close the furnace after filter retraction.
Three normally-closed monostable valves (V1, V2, VT) are mounted on the oven's
body. V1 supplies a labelled gas (15N2) to carry the gas sample from the oven
to the transfer lines through V2. The venting valve VT allows the oven's gas
content to be vented to Titan's atmosphere. VT is also usually activated
during certain sequences for purging operations or under pressure control
before transfers (see below). Also, VT can be opened at any time during
descent as a safety measure if the oven pressure rises above 3.2 bar; it
closes when the pressure falls to 2.7 bar.
A drag fan kinetic pump (PU) ensures the flow of Titan's atmosphere through
the filter, at a rate depending on altitude. An exhaust tube (ET), with a
one-shot isolation valve (P2), allows the gas to be vented externally. The gas
flow, on reaching the level of the GV, follows a path perpendicular to the
oven/GV assembly. The result is that the circulation is independent of the GV
status (closed or open). When it is switched off, PU acts as a flow-blocking
device. The pressurisation system for storing N2 gas and controlling its flow
to the oven begins with a gas tank (GT) at 40 bar. Oven filling is controlled
by a pressure transducer (PS) associated with valve V1. A relief
valve (RV, set at 4.1 bar) in the internal gas transfer line protects
GCMS against accidental ACP overpressure.
The whole internal circuit is pressurised during ground operations and the
early part of the flight to Saturn. After the Jupiter flyby, about 3 years
after launch, ACP's internal circuit is evacuated by opening P2. In addition,
valves V1, V2 and VT are opened briefly during each cruise phase checkout. The
gate valve mechanism (GVM) is locked during launch, and unlocked shortly
after. It must be locked again for a short time before Probe entry; this is
done during the last cruise checkout sequence. The inlet tube end is closed by
a sealing cover (SC), which will be opened at the beginning of descent. A
connecting tube (Product Transfer Line, PTL) between ACP and GCMS transfers
the pyrolysis products. Valve P1 isolates the ACP internal circuit from the
product transfer lines. This one-shot isolation valve is opened at the
beginning of descent (T0+2 min) for an initial venting of the internal (ITL)
and the external (PTL) exit transfer lines. The IVA one-shot valve isolates
the PTL at its GCMS extremity. The T1, T2 and T3 special ports are for ground
tests. The V2 and VT electrovalves, P1 and PTL can be heated by special
heaters (HV1 for V1, HV2 for V2, HVT for VT, HP1 for P1 and HPTL for PTL).
In addition, heater HV1 is controlled by a thermostat, activated when the
temperature falls below -5 deg C. This prevents leakage caused by Titan's
low temperatures.
 
Measurement Strategy
====================
 
The operations sequences result from the following requirements:
  1. Determining the compositions of the particle cores (non-volatile and
  volatile components) is conducted mainly in the low stratosphere and down to
  the tropopause (above 30 km). In the higher part of the descent (above 80
  km), it is expected that aerosols will be obtained by direct impaction on
  the filter. Below 80 km, where the pump becomes effective, the samples are
  obtained by filtration.
  2. The second sample must be collected within the troposphere above the deep
  methane clouds (20 km).
  3. Owing to mass constraints, the instrument is equipped with a single
  collector that must be used again after cleaning the oven, filter and
  product transfer lines.
  4. The samples are analysed by using GCMS for a fixed portion of its life
  (approximately 20 min), knowing that this instrument must make at least one
  direct chromatographic analysis of the atmosphere's composition before
  surface impact.
In addition, because of the very short descent profile (120 min minimum), it
was decided to make three transfers for each sample, each transfer using the
direct MS mode. These will use, in turn, the oven at ambient temperature,
250 deg C and 600 deg C. A complete GCMS analysis, requiring about 10 min,
is provided only for the first sample's pyrolysis sequence (transfer after
600 deg C heating). These pyrolysis products have to wait until GCMS has
analysed the contents of the two gas enrichment cells that were sampled in
the upper atmosphere. This implies that the use of the three gas
chromatograph columns for analysing ACP aerosol pyrolysis products obtained
at 600 deg C occurs at T0+73 min (nominal altitude 25 km).
 
 
Operational Sequences
=====================
 
A detailed timeline has been formulated to mesh with GCMS' own. It is based
on the following sequences that ACP must operate during the descent phase.
 
Sequence 1  Initial venting and preparation for the first sampling operation
between ACP initialisation (at T0+1 min 40 s) and the filter in its sampling
position (at T0+6 min 45 s, nominal altitude 130 km).
Sequence 2  First sampling in the low stratosphere. This period ends when the
filter is retracted into the oven (GV locked) at T0+60 min 00 s (nominal
altitude 32 km).
Sequence 3  Heating the filter (ambient, 250 deg C, 600 deg C) and gas product
transfers to GCMS at T0+74 min 00 s (nominal altitude 24 km).
Sequence 4  Oven and transfer lines cleaning, and preparation for the second
sampling operation with the filter in its sampling position at T0+77 min 00 s
(nominal altitude 22 km).
Sequence 5  Second sampling in the upper troposphere. This period ends when
the filter is retracted into the closed oven, at T0+89 min 00 s (nominal
altitude 17 km).
Sequence 6  Heating the filter (ambient, 250 deg C, 600 deg C) and gas product
transfers to GCMS ends at T0+108 min 00 s (nominal altitude 9 km).
 
During sequence 1, after the high entry velocity has been reduced and the
Probe's protective front shield has been jettisoned - eliminating the risk of
contamination from the front shield - the following operations begin at
T0+1 min 40 s. P3 is opened, and flushing and venting the internal circuit
begins by filling the circuit with N2 through V1 and V2 down to P1 (GV is
closed). Venting is obtained by opening VT. The whole process is repeated
after opening valve P1. The mechanism for ejecting the sealing cover (SC) is
then activated, and in-flight calibration of the oven pressure sensor (PS) is
performed to compensate for any offset accumulated during cruise. GV is then
opened and the filter is moved into its sampling position.
 
Sequence 2 begins immediately, with its two sampling modes. The first leaves
the pump unit inactive (until approximately T0+23 min 30 s), so that sampling
is by direct impaction. The second turns the pump unit, so that sampling is
mainly by filtration. Once the pump is switched off, the filter is retracted
and the oven gate is closed. Retraction and oven closing is accomplished in
only 13 s in order to avoid any evaporation of certain condensates in the
aerosols. The whole of sequence 2 takes about 55 min.
 
Sequence 3 deals with:
  1. the preparation of aerosols for producing evaporates and pyrolysis
  products;
  2. the transfer of gas products to the ACP line (feed tube to GCMS);
  3. the analysis either by the sole direct MS mode or the complete mode
  (GCMS + direct MS).
 
The programme for aerosol preparation and transfer consists of three phases:
 
Phase (a)  transfer of the gas products obtained while the filter is in the
unheated oven. At the time of transfer, the filter has considerably warmed
since oven closing, and the temperature gradient is sufficient to produce some
evaporates.
 
Phase (b)  transfer of gas products obtained after heating the filter to
Tf=250 deg C. In order to avoid excessive pressure in the oven and hence in
the ACP feed tube in GCMS, the oven is linked before heating to the atmosphere
by briefly opening VT. In addition, to avoid condensations of gas samples,
transfer lines ITL and PTL (valves and tubing) are heated as much as possible
within the constraint of the allocated power; 90 deg C is the objective.
During the transfer time (~ 1 min), the filter temperature is maintained by
holding Tf at 250 deg C.
 
Phase (c)  transfer of the gas products obtained after heating the filter to
600 deg C. This pyrolysis phase is also preceded by briefly opening VT to
reduce pressure to atmospheric. During transfer, the valves and tubes are
heated within the power constraint as in phase (b); the goal is again 90 deg
C.
 
In order to transfer the gas samples with minimal dilution from the effluent
gas (15N2), each injection into GCMS is done by pressurising the oven to 2.5
bar with N2 (V1 controlled, V2 closed) and then rapidly depressurising it
(fast actuation of V2 and VAB).
The estimated mass flow rate at injection is 3-9 mg/s at 2 bar N2.
The same cycle is used for each analysis that occurs after phase (a), (b) and
(c). As there are two sampling operations (one high altitude, one low
altitude), the complete ACP programme equates to six transfer cycles.
Each transfer cycle comprises two sub-cycles. The 'scientific sub-cycle' is
devoted to analysing the oven contents by a series of six injections (V1
closed) of 0.875 s (V2 and VAB open) each, each followed by an MS analysis
period of 4.750 s. This period for MS analysis is longer than the time
required by GCMS for a full mass scan. After completion of
the sixth MS analysis, VT and VAA are opened (VL4 is closed) for 5 s to
evacuate any residual traces of pyrolysis products in the sampled volume. A
second sub-cycle follows, devoted to a background analysis. Two 0.875 s
injections are each followed by an MS analysis of 4.750 s. A second 5 s
vent (VAA and VT open) completes the transfer cycle.
The oven pressure must be controlled during each transfer for a correct GCMS
analysis. An optimal interface requires that the pressure in the GCMS feed
tube is maintained above 1.9 bar. Also, efficient sample transfer requires
pre-injection pressurisation of the oven to 2.5 bar. During the pressurisation
period (625 ms) a control loop uses the oven's pressure sensor to stay within
the 2.5 bar limit (by closing V1 if needed). During injection, the control
loop ensures a minimum 1.9 bar in the oven by opening V1 if necessary. (Note
that V1 is closed by the software as part of the transfer cycle in any case at
the end of the 625 ms pressurisation period).
Furthermore, because of the unknown aerosol load collected on the filter, we
have to accommodate any unforeseen increase in oven pressure after heating.
The oven pressure is therefore checked before each transfer cycle. If it is
above the nominal 2.7 bar, VT is opened until this value is reached.
Owing to the peak power limit imposed by the Probe system (83 W on the ML3
main power line), there is a conflict between heating the
oven and the transfer lines during the transfer cycle. It was thus imperative
that a temperature control priority be selected by the software. The result is
that heating the oven and holding it constant during the transfer of gas
products to GCMS has first priority. Second priority is given to heating VT,
P1 and PTL. In addition, during the preparation phase sub-sequence before oven
heating, the temperature control loop of the transfer lines activates their
heating for 1-2 min at 120 deg C.
 
For sequence 4, before the second sampling stage, the filter, oven walls and
transfer lines must be well cleaned. The oven and filter are cleaned, first,
by extending oven heating by 1 min in order to complete filter outgassing
(filter temperature is 600 deg C at the end of sequence 3's last transfer).
Immediately after, the oven is filled to 2.5 bar within 3 s by activating the
oven's pressure control loop to open V1. Then, the oven is vented by opening
VT for about 6 s, reducing the pressure down to ambient atmospheric. Cleaning
the transfer lines downstream of V2 is achieved first by switching on heaters
HV2 and HP1 (plus PTL), set at the maximum 120 deg C for 90 s. Then, a
fill-vent phase is started for the oven and transfer lines by:
  1. switching on the oven's pressure control loop (filled at a maximum 2.7
  bar) while valve V2 is opened for approximately 3 s;
  2. venting the two circuits when reducing the internal pressure to the
  external level at the two ends. The oven is vented by opening VT (6 s),
  while the transfer lines downstream of V2 are vented through GCMS by
  opening VAA (few seconds).
This fill-vent operation is repeated once. There is no background analysis
from the MS in order not to contaminate the GCMS instrument unnecessarily. At
the end of sequence 4, ACP is ready for a second sampling sequence.
 
For sequence 5, sampling is performed by filtration in the cloud region. The
altitude range is fixed by the time at which the pyrolysis products analyses
must be finished so that GCMS is freed for direct GCMS analysis of the
atmosphere. It means that the second sampling phase, which
starts T0+77 min (nominal descent profile, 22 km) must be finished at T0+88
min 30 s (17 km nominal descent profile). Then, after about 11 min of
sampling, the filter is moved back into the oven, which is closed for
preparation and analysis.
Before dealing with the next sequence, there is a standby period of 9 min in
order to wait for GCMS as it completes its analysis of sample cell number 3.
 
Sequence 6 is a copy of sequence 3 but during phase (c) there is no time left
for GCMS to analyse the pyrolysis products through the columns - only the MS
mode is used. Purging and background analysis are, as in sequence 3, performed
after phase (a) and again after phase (b/c). At the end of the transfer (phase
c) of the products to GCMS (T0+108 min), ACP is prepared for being turned off
until T0+110 km (8 km nominal altitude).
 
 
Mechanical Configuration
========================
 
The instrument housing is made of aluminium alloy and consists of a single
unit mounted on the lower part of the experiment platform. The rectangular box
carries six attachment lugs on its base. The electronics system (ACPE) has its
own structure attached at six points on one side of the instrument's main
body, in which the mechanical and pneumatic subsystem is located.
ACP's maximum dimensions are 220x200x206 (H) mm. In addition to the inlet and
exhaust tubes, and to the GCMS product transfer tube, there is an extension
accommodating movement of the filter mechanism's rack. ACP's
total mass is 6.7 kg (including 5% margin), shared between 4.5 kg for the
mechanical box and 2.2 kg for the electronics box.
The aerosols sampling inlet tube extends downwards to penetrate the Probe's
fore dome. To provide efficient evacuation of Titan's atmospheric gas after
sampling and filtering, the exhaust tube exits upwards, passing through the
experiment platform and the top platform. By siting ACP close to GCMS, the
inlet tube is close to the Probe's axis and PTL is as short as possible. A
very strict procedure for limiting chemical contamination was followed during
instrument fabrication. The objective values specified for the internal
circuit of the gas transfer subsystem are <50 ppb for the organic compounds
expected during aerosol chemical analysis and for gases such as CO2, CO and
H2O. The internal surfaces of all the transfer tubes have been passivated
('silanised'). A cleanliness plan was followed first at the level of each
equipment element and then at the instrument level. After assembly, the whole
ACP was baked under vacuum for several days at 120 deg C. Also, organic
materials were not used in the GV and filter mechanisms. Their bearings, in
particular, are not lubricated.
In order to maintain the high cleanliness level during the instrument's ground
storage and its first part of the cruise, ACP's internal gas circuit was
filled to 2.5 bar with pure nitrogen. The internal circuit is hermetically
closed off, by three sealing devices, at the level of the sampling and exhaust
tubes and at the interface between the ACP housing and the PTL. The specified
instrument overall helium leak rate of 10^-7 mbar litre s^-1 was found to be
sufficient to hold the gas circuit pressurised for more than 4 years.
The two apertures leading to the atmosphere are sealed by specially-developed
devices: the sealing cover (SC) for the sampling tube and the one-shot P2
valve for the exhaust tube. SC is screwed to the bottom of the sampling tube
(ST); the tightness requirement is satisfied by a stainless steel O-ring
gasket. SC's cap will be spring-ejected towards Titan at 11 m/s; the
requirement was for 8.6 m/s in order to avoid back-collision with Huygens. The
cap and spring, totalling 80 g, are held in position on SC's body by
tin-silver solder. Ejection will be within 2 min of switching on the 53 W
heater (SCH), as the alloy melts at about 150 deg C. Correct ejection is
checked by a Hall effect sensor (HSC). The P2 concept is identical: melting a
brazing allows a spring to eject the sealing cover, freeing the outlet gas
exhaust. The P1 one-shot isolation valve between the ACP housing and PTL.
 
 
Electronics
===========
 
Hardware design
---------------
ACP's electronic system (ACPE) is composed of four functional blocks: DC/DC
converter, control unit, monitoring unit and drive unit. The structure has two
frames and a base plate, both made of aluminium alloy. The first
frame contains two parallel printed circuit boards (PCBs), one for the control
unit (ACPCU) and one for the drive unit (ACPDU). The second frame, which has
an integrated top plate, contains the monitoring unit (ACPMU) PCB and the
components for the DC/DC converter block. After final integration
the two frames are screwed together; the base plate is attached to the first
frame, interfacing with the ACP mechanical main box.
Huygens supplies ACP with three 28 V power lines: main lines ML1 and ML3 and
protected line PL2. ACPDC's main functions are: (1) to provide ML1 input with
DC conversion for supplying other electronics functional blocks; and (2) to
provide all power lines with EMI filters. Conversion of ML1 DC is performed
through a circuit that uses in series an SAE, AFC 461 F/CH for transient
suppression, an SAE AT 02815 TF/CH for DC/DC conversion and one specific EMI
filter to each output of the converter (0, +5, -15, +15 V).
ACPDU drives all the electromechanical devices (electrovalves,
heaters, motors) of the mechanical box. DU is galvanically isolated from the
drive logic in CU. ACPMU conditions and monitors the temperature and pressure
analogue signals coming from the sensors in the mechanical main box. Under
software instruction, ACPCU controls ACP's motors, valves and heaters.
For that purpose, it monitors first, inside ACPE, the transfer
of actuation commands to ACPDU the flow of digital and analogue inputs from
ACPMU. It also monitors all data transfers and synchronisation with ACP
external systems such as the Probe, GCMS and ACP's EGSE.
The data are transferred in two directions: (1) from the Probe's Command and
Data Management Subsystem (CDMS) to ACP - mainly telecommands, descent data
broadcast (DDB) and broadcast pulse (BCP); and (2) from ACP to the Probe -
statusword and packet telemetry data (PTD). Synchronisation pulses are sent to
GCMS for accurate timing of GCMS and ACP valve actuations during transfer
periods. For activating and controlling mechanical and pneumatic components,
the command signals generated within ACPCU are sent to ACPDU through a J21
connector. Status and sensor readings from ACP elements are transferred, after
conditioning, to ACPMU.
ACPCU's main components are the 80C85 8-bit microprocessor, the 80C37 direct
memory address (DMA), a 4.096 MHz oscillator, two PROMS with 8 kbyte memory
each, two RAMS with 8 kbyte memory each, a 12-bit analogue to digital
converter (ADC) with a sample and hold amplifier (S&H), two 8-channel
multiplexers (MUX) and two FPGAs (field programmable gate arrays). In
addition, ACPCU contains memory latches, buffers and optocouplers.
The microprocessor uses eight bits for data lines and 16 bits for address
lines, eight of which are multiplexed with the address bus. The 4.096 MHz
oscillator frequency is divided by one of the FPGAs to provide the
non-maskable microprocessor interrupt with selectable clock pulses of 1, 2, 4,
8 or 16 ms. Two other interrupts of the 80C85 are used for BCP pulses every
125 ms (synchronised with the Probe) and overpressure signals. The 12 kbyte
program is stored in the two 8 kbyte PROMS. When the ACPE main line (ML1) is
powered by the Probe, the 12 kbyte are transferred to the two 8 kbyte RAMS.
The remaining 4 kbyte in the RAM serve as data transient memory.
DMA. driven by the FPGA timer, is used mainly for fast telemetry (TM) and
telecommand (TC) management. Three of DMA's four request inputs are used for
TM/TC data transfer from the 4 kbyte of the RAM to the Probe CDMS. The first
two are used for nominal and redundant TM transfer, and the third for TC
transfer.
The monitoring signals (temperature, pressure and status) coming from ACPMU
are used by the software for controlling the instrument. Analogue signals
(temperature and pressure) are multiplexed and converted to digital signals
into ACPCU. A total of 15 analogue input lines is multiplexed through the two
MUXs. The MUXs output line selection is performed by one FPGA, then the
selected signal is transferred via the S&H to the ADC for analogue to digital
conversion.
One of the two FPGAs controls the exchange of signals with the Probe (BCP,
DDB, TC, TM). It also serves as pump current and speed counter. The other FPGA
provides the microprocessor with a timer and a watchdog, and manages the ACPCU
and ACPMU multiplexers. This FPGA also controls all the output command signals
to ACPDU.
 
Software design
---------------
Onboard software is used by ACP's microcontroller to execute the automatic
sequences, monitor the status and health of the various subsystems, acquire
and interpret TC and format data for TM. Its main functions are to control the
mechanical subsystem, collect data and transmit it to the Probe system,
provide an interface to command ACP via the Probe, and perform descent
measurements synchronised with the rest of the Probe.
ACP's software provides four different operational modes: descent; cruise
checkout; ground test mode: engineering mode. The first three execute
automatically on receipt of the appropriate mission flag and time. The
engineering mode is selectable by a specific TC, and is used mainly for
integration and test phases. In this mode, all ACP functions can be activated
individually by sending the appropriate TC. It is accessible during any
automatic sequence. It can also be used to check specified mechanical devices
during cruise should any problem arise.
 
 
Calibration
===========
Pyrolysis tests
---------------
In order to validate ACP's scientific performance. pyrolysis tests were
conducted at LISA in Paris on solid phase material synthesised from
experimental simulation. The organic synthesis by electrical irradiation of an
N2-CH4 mixture requires a long reaction time (about 20 h). The reactor is a
two-part glass vessel with two tungsten electrodes and a metallic filter in
the lower part. It is filled to a total pressure of 900 mbar with a mixture of
N2 (800 mbar) and CH4 (100 mbar). One of the electrodes is connected to a
Tesla coil fed by a low current (80-100 mA) high frequency (80-200 kHz)
generator. The other electrode is Earthed. Five irradiation periods of about
4 h each were performed and, between each, the gas phase was removed from the
reactor and replaced by a new mixture of N2-CH4 at 900 mbar with the same
ratio as the original. During irradiation, the bottom of the reactor was
cooled by liquid nitrogen. At the end of the synthesis, 2.3 mg of solids were
deposited on the filter.
An ACP model (M3) was specifically developed for this sort of science
validation and laboratory investigation. It is representative of
ACP except for the pump unit and the gas tank, which are not mounted. However,
it contains:
  - the filter transfer mechanism (not motorised) to translate the filter into
    and out of the oven;
  - the gate valve mechanism (not motorised) to close off the oven;
  - the three monostable microvalves used to vent the oven by pumping (VT) to
    fill the oven with pure nitrogen piston gas (V1), and to transfer the gas
    phase from the oven to GCMS (V2).
The filter could be dismounted from its support and easily replaced. In order
to protect the filter's organic solid phase material from oxygen
contamination, the following operations were carried out in a glove box
filled with nitrogen:
  - once removed from the reactor, the filter was mounted in M3;
  - M3's gate valve was closed after enclosing the filter in the oven.
Owing to the oven's low leak rate (<10^-3 mbar l s^-1 ), we can be sure that
no oxygen contamination of the filter occurred during analysis. The analysis
of the pyrolysis products was done at LISA, where a Varian Saturn II GCMS was
used (with helium as carrier gas) at an inlet pressure of 1.6 bar. The
chromatographic capillary column was a CP-Sil-5 CB column of 25 m length, 0.15
mm inside diameter and 1.20 mm film thickness from Chrompack. The column
temperature was controlled thus: 30 deg C for 20 min; raised from 30 deg C to
150 deg C in 30 min at 4.0 deg C/min; 150 deg C for 10 min. The total mass
spectra of the GC peaks were thus collected over 60 min. The ion trap detector
was able to detect masses of 10-226 amu by using the electronic ionisation
mode. The experimental chromatographic conditions were chosen to detect
hydrocarbons from C4 to C10, which excludes any information about the light
hydrocarbons (C2 and C3) and the heavy hydrocarbons (above C10). Using this
experimental approach, we achieved a very good validation of ACP for the 600
deg C pyrolysis sequence that will be performed on the sampled aerosols during
Probe descent.
The chromatogram obtained by GCMS analysis of the synthesised solid phase
yields a pyrogram showing many organic compounds. Some GC
peaks appear clearly from the total mass spectra. The others are identified
only when they are specific ions with a very high MS sensitivity. According to
their mass spectra, 23 gross formulas of more than 25 GC peaks have been
identified unambiguously. The numbered peaks were identified and the
possible isomers according to their mass spectra were suggested. It
immediately appears that no oxygenated organic compounds were detected,
showing that the laboratory procedure prevented oxygen contamination of the
synthesised solid phase material.
The main compounds detected on the pyrogram are mono aromatic hydrocarbons.
The most concentrated is benzene (12), followed by toluene (16) and
C2-substituted benzene (19, 21, 22). One may conclude that these mono aromatic
hydrocarbons are the major constituents of the synthesised solid phase
material. If they are thermally stable, they could be completely desorbed into
the gas phase at 600 deg C with no variation of their structures. Thus the
main constituents are the main compounds desorbed and detected on the
pyrogram.
One argument limits the validity of this result. Since a bi-aromatic
hydrocarbon is observed on the pyrogram (23) and because of the
chromatographic conditions (choice of column), we cannot properly observe poly
aromatic hydrocarbons. In a previous study, pyrolysis of anthracene, which
contains three aromatic cycles, was performed with the same pyrolyser at 600
deg C. The pyrogram showed mainly benzene, toluene, ethenyl-benzene and C8H10
isomers. It shows that poly aromatic hydrocarbons are decomposed at 600 deg C
in benzene and in substituted mono aromatic hydrocarbons. The type of
compounds detected on anthracene pyrograms depends on the pyrolysis
temperature. For example, it was shown that naphthalene (a bi-aromatic
hydrocarbon) was clearly observed at a lower pyrolysis temperature. So it is
quite possible that the mono aromatic hydrocarbons detected on the pyrogram
could be provided by the poly aromatic hydrocarbons present in the solid phase
material and thermally decomposed at 600 deg C into mono aromatic
hydrocarbons.
A small amount of benzonitrile (20) was also detected, which can be similarly
produced by thermal decomposition of poly aromatic compounds containing
nitrogen atoms. Most of the nitriles observed on the pyrogram have also been
detected in the gas phase synthesised by the simulation (3, 4, 7, 8 and 11).
The heaviest nitriles detected (14 and 15) could be provided by thermal
desorption of the same compounds present in the solid phase or by thermal
decomposition of the heaviest molecules. Finally, many hydrocarbons are
detected on the pyrogram (1, 2, 5, 6, 9, 10 and 13). They can be provided by
the two sources cited above, or by the thermal polymerisation of the lightest
alkenes and alkynes.
Before performing the above pyrolysis tests, the procedure for testing the
cleanliness of the ACP Flight Model, after delivery by SEP, was conducted on
M3. The instrumentation includes a very pure nitrogen gas reservoir, a flow
regulator provided by the contractor, and laboratory GCMS. A mixing ratio
threshold of 50 ppb was measured for the significant gas components.
 
Validation tests of the direct Mass Spectrometer mode
-----------------------------------------------------
Tests were run to verify the ACP transfer cycle and to
validate the data obtained by GCMS in the direct MS mode (direct inlet flow
into the dedicated ACP ion source of GCMS). Several nominal sequences
reproducing the ACP transfer cycle were conducted at LISA using a commercial
quadrupole mass spectrometer. The verified transfers correspond to the nominal
descent phase when ambient oven temperature analyses are made after the first
aerosol sampling period. GCMS valve VAB was replaced by a commercial Buerkert
valve to which a commercial Lee-Jeva restrictor, representing FRA, was added.
Flow restriction equivalent to that provided by VL4 on Huygens
was reproduced by using a capillary fused silica tubing of 10 um internal
diameter and 6 cm length. For testing the ACP-MS coupling, the
oven was filled at 30 mbar with pure krypton. The mass spectra obtained (mass
84) is representative of the amount of krypton injected.
Quantitatively, the decrease in krypton's pressure corresponds exactly to the
specifications for transferring the oven's atmospheric content. It confirms
that the calibration of the FRA restrictor used for the M3 model (Lee Jeva
80000 Lohm) gives the correct flow for the ACP-GCMS coupling of the Huygens
experiment.
The observed pressure peaks correspond to the simultaneous opening of V2 and
VAB causing an increase in the total pressure and krypton's partial pressure
in the V2-VAB line. This phenomena is smoothed on the figure by acquiring one
point per second. In the GCMS experiment, a scan of the entire mass spectra
will last about 2 s, so the smoothing of the curve obtained should be greater.
 
The calibration programme
-------------------------
The programme selected for calibrating the flight models is directly linked to
the GCMS calibration plan (see the GCMS paper in this volume). As GCMS will
analyse the gas products transferred from ACP, the calibration system at NASA
Goddard was used to relate (qualitatively and quantitatively) GCMS' mass
spectra to the effluent gases injected from ACP. ACP calibration is a 2-step
process and concerns only the gas phase transfer. This is because we do not
wish to pollute the Flight Models' ACP internal tubing and GCMS analyser with
products of solid samples such as tholins or other complex matter. The tests
allowed the evaluation of GCMS' response to the ACP products.
The first step directly injected several calibrated gas or gas mixtures
containing expected Titan ratios into the ACP feed tube (IVA-VAB).
The second step, which requires the calibration gas to pass through ACP first,
has been delayed because of the very tight GCMS delivery schedule. This part
of the calibration programme will be performed immediately after launch, on
the two spare models.
The second step was similar but the calibrated gas passed through ACP first.
For that, the calibrated gas mixtures were injected from Goddard's calibration
bench into the oven using the special inlet T3 (to be pinched off). During
the investigation, the oven will be heated according to ACP's
timeline before each nominal transfer cycle. The oven pressure will be set to
simulate the conditions after the completion of each of the two aerosol
sampling phases. The oven's gas tightness when the gate valve is locked is
sufficient (leak rate 10^-3 mbar l s^-1) to prevent any change in the
calibrated gas concentrations during the analysis process.
In the ACP configuration at Goddard, isolation valve P1 will be replaced by a
dummy valve with a punctured diaphragm in order to allow passage through to
GCMS. During ACP integration on Huygens, the pneumatic connection to GCMS was
made by attaching PTL between the two. This was followed by a special
procedure for cleaning and filling the PTL with pure N2.
Flight Model cleanliness was assured and controlled by following the same
procedure as with the M3 model. During post-launch calibration, pyrolysis
tests and chromatographic analysis are planned at Goddard using representative
mock-ups of the GCMS and ACP Flight Models.