GOMOS InOrbit Operations And Failure Recoveries

1
GOMOS In-Orbit Operations And Failure Recoveries

• Aldo Barbieri
• VEGA Group,
• Mission Operations Department
• ESOC, Darmstadt

 
2
Summary

• ENVISAT overview
• GOMOS description
• GOMOS operation
• Anomaly investigation
• Work-around solution
• Conclusions

3
ENVISAT Overview I

• Earth observation mission in a sun- synchronous
  polar orbit with DNC at 1000 local
• 800 km altitude
• Ca. 14 orbits per day ( Orbit repeat cycle 35
  days)
• Launch at 01.03.2002 (Ariane 5)
• Expected lifetime 5 years
• Weight 8140 kg (Dimensions 25x10x5m)
• Solar Array output power 6,5 kWatt

Europes biggest Earth Observation satellite
4
ENVISAT Overview II

• Mission Objective
• to ensure the continuity of the data
  measurements of the ESA ERS satellites to support
  earth science research and to allow monitoring of
  the evolution of environmental and climatic
  changes

• Instruments an innovative payload comprising 10
  instruments (C-band SAR, radar altimeter,
  spectrometers, microwave radiometer)

• Scientific measurement data
• ca. 250 Gb data products /day

5
ENVISAT Overview III

• Ground segment
• Spacecraft operations at ESOC
• Data processing facility at ESRIN
• Ground Stations at Kiruna (Sweden), Svalbard
  (Norway).

6
GOMOS Objectives and Scientific Requirements

• The scientific objective is indicated by the name
  of the instrument
• Global coverage of the Earths atmosphere with
  respect to
• Ozone and other gases by the
• Monitoring of the
• Occultation of target
• Stars

• The GOMOS instrument provides the capability to
• map ozone and other gases with high accuracy
• all around the Earth (average 45 stars/orbit)
• in a wide altitude range (5km - 100km)

7
GOMOS Measurement Principle
Acquisition of a star
Measurements of initial, un-attenuated star
spectra
Tracking of stars behind the atmosphere
Measurement of attenuated star spectra
through the atmosphere
Comparison of initial and attenuated spectra
to obtain absorption profiles
Determination of vertical profiles of Ozone
and other gas concentration
8
GOMOS Functional Description
Light from Stars
9
The Steering Front Assembly

• The main objective of the SFA is to acquire and
  track selected stars. The SFA is composed of two
  separate units
• the SFM to orientate the mirror both in azimuth
  and elevation
• the MDE which controls the SFM using the
  commands generated by the ICU.

• The SFM is a two stages mechanism
• a rotator which provides the coarse pointing of
  the instrument line of sight towards the star to
  be observed
• a fine stage mechanism mounted on top of the
  rotator which ensures a rotation over a reduced
  angular range for both azimuth and elevation
  angles providing the required pointing stability
  and accuracy during the tracking of the star

10
The Occultation mode
Synchronous Occultation (sequence of stars
repeated over several orbits)

• Occultation mode

Typical Operational Timeline
Asynchronous Occultation (sequence of stars for
one orbit)

• Monitoring modes

In Occultation mode, the instrument autonomously
executes the following sequence

• maintain current position
• coarse rallying
• fine rallying
• detection
• centering
• tracking

11
Azimuth Voice Coil Anomaly I
In March 2003, an anomaly, the so called
"voice-coil command saturation, occurred during
the rallying phase to reach the target
position. This meant that the commanded current
to be sent into the voice-coil actuator, (used
for fine mirror steering) reached the saturation
level causing the instrument to perform a
switch-down to safe mode. This problem
interrupted several times the ongoing operations
until the instrument was recovered from ground.
12
Azimuth Voice Coil Anomaly II

• Initially this anomaly affected only stars with a
  low Azimuth value.
• As a temporary work-around solution, the
  observation planning was modified by reducing the
  azimuth range.
• On 30th June 2003, following consecutive failed
  attempts to recover operations, the SFM appeared
  to be unable to run successfully its
  initialization phase
• In order to resume operations it was decided to
  switch to the redundant part of the instrument.
• In July 2003 GOMOS resumed operations.

13
Azimuth Voice Coil Anomaly investigation result
I
The investigation concluded that the most
probable cause of the problem was a sudden power
cut of the azimuth voice-coil current.
The azimuth angle is approaching its mechanical
zero position as that the consumed current drawn
by the actuator was zero.
14
Azimuth Voice Coil Anomaly investigation result
II
The figure below shows the suspected failing
voice-coil harness.
The harness leading from the fixed part to the
mobile part might have a damaged isolation and
for a certain angle touched the fixed part
resulting in a short circuit.
15
Elevation Voice Coil Anomaly I

• The next anomaly first occurred in January 2005
  after a long period of undisturbed measurements.
• It occurred during rallying of the SFM in
  preparation of a star observation at a certain
  point the current commanded to the voice-coil,
  used for fine elevation steering, reached the
  saturation level and caused the instrument to
  perform an autonomous switch-down to a lower
  Standby mode.
• Since then, GOMOS operations were interrupted
  repeatedly by the same anomaly with a dramatic
  increase in the following weeks.
• This severe failure similar to the one on the
  nominal chain could have meant the instrument end
  of mission.

16
Elevation Voice Coil Anomaly II
A series of in-flight tests were performed to
support the anomaly investigation. The following
anomaly characteristics were observed

• no dependency on orbit position
• no dependency of the voice coil current problem
  with the elevation and/or azimuth angle
• a dependency on the azimuth movement (rallying
  phase duration). The error probability
  significantly increased at azimuth gt 25 deg
• two types of the voice coil current behaviour
  were observed
• swinging (between two saturation levels)
• not swinging (with stable saturation)

17
Elevation Voice Coil Anomaly Swinging behaviour

• the angle and the commanded current started to
  oscillate
• Mirror moves away from its equilibrium position
• the system reacts by increasing the commanded
  current in order to bring the measured angle back
  to its nominal value.
• Gain of control loop seems to increase (the
  consumed current is much lower)

18
Elevation Voice Coil Anomaly Non Swinging
behaviour

• Mirror moves toward its equilibrium position
  (around 5.3)
• No more current in voice-coil
• The measured elevation angle is consistent with
  the consumed current.

19
Possible error sources
4
3
5
Encoder, digital
Encoder, analogue
ICU
MDE
Commanded current, digital
6
Consumed current, analogue
2
7
Elevation Sensor (Inductosyn)
1
Elevation Voice Coil actuator

• ICU input /output buffer failure
• ICU S/W failure
• Cable/connections

• Voice Coil Device failure
• Elevation sensor failure
• MDE A/D conversion failure
• MDE electronic circuit disturbances

20
Failure hypotheses

• Due to the difficulty of the investigations and
  the lack of an on-ground SFM model, it was not
  possible to identify exactly the degraded/failing
  component nor the anomaly triggering mechanism.
• The analysis of instrument operations concluded
  that the origin of the problem is located in the
  Mechanism Drive Electronics (MDE) of the SFM.
• It could be related to certain instability in the
  power provided to the elevation voice coil.
• The most probable cause is a combination of some
  instability of the electronics (Power, EMC)
  induced by the high frequency disturbances of the
  Coarse Rallying Mechanism and the increased
  sensitivity of the Control Loop due to ageing.

21
Operational Work-around Solution ICU patch

• The test results showed that
• the Voice Coil anomaly is not permanent
• there was no evidence that the reappearance of
  the anomaly could further degrade the equipment
  performance.

These observations confirmed that the anomaly
category could be modified without apparent risk.
ICU patch no autonomous switch-down triggered
on the occurrence of the anomaly.
22
Operational Work-around Solution reduced azimuth
FoV

• The test results showed that there was a direct
  correlation between the duration of the azimuth
  rallying and the probability of anomaly
  occurrence.

It was decided to limit the azimuth field of view
(equivalent to limiting the maximum duration of
the rallying) to a maximum of 25 degrees. (the
boundaries of the field of observation have no
constraints and can be adapted according to
scientific needs)
23
Operational Work-around Solution Reducing the
Occultation Sequence

• In order to reduce the potential outage time,
  the operation strategy was changed such that the
  maximum loss of consecutive observations can not
  exceed two orbits.

This was achieved by reducing the Synchronous
Occultation Sequence to two orbits .
24
GOMOS Overall performance
After implementation of the work-around solution
, the GOMOS Instrument is operating with high
availability level.
Elevation anomaly
Azimuth anomaly
25
Conclusions I

• During its operational life, the GOMOS instrument
  suffered two severe failures that could have both
  meant the instrument end of mission.
• A series of in-flight tests were performed to
  support the anomaly investigation and establish a
  failure model.
• The investigation analysis concluded that
• the Azimuth anomaly was due to a sudden power
  cut of the azimuth voice-coil current due to a
  damaged cable.
• the origin of the Elevation anomaly is located
  in the MDE of the telescope, related to a
  certain instability of the power provided to the
  elevation voice coil (telescope elevation
  actuator).

26
Conclusions II

• A satisfactory workaround solution was
  implemented.
• The GOMOS operations were successfully resumed
  although the error could not be completely
  resolved.
• The elevation anomaly does still re-occur
  sporadically but does not significantly disturb
  the science data coverage.

27
http//envisat.esa.int
Aldo.Barbieri_at_esa.int
www.vega-group.com
e