- The first development activity kicked off at TNT in
1997, and completed in 2000 with one Engineering Model
(EM) RAFS1 produced [1].
- The updated RAFS1 development started in June
2000 and completed at the beginning of 2002. The industrial
consortium is led by TNT with Astrium Germany as the
subcontractor for the electronics package. In this phase, the
achieved activities include:
- Improved clock stability with inclusion of
thermally regulated base plate. Fig. 1 is the picture
of the updated RAFS1.
- Review of electronics package layout and
components in view of flight production.
- Manufacturing of 5 Engineering Qualification
Models (EQM) for lifetime qualification. Fig. 2
shows 5 EQMs without external cover and 5
vacuum chambers for life test with ‘Picotime’
measurement systems.
- Manufacturing of 1 Qualification Model (QM).
Besides the vibration and EMC/EMI qualification tests,
two radiation tests were carried out at CNES in Toulouse:
one test with Galileo orbit simulation, i.e. 4 cycles of 3rad
per day during one week, and the other with total dose
simulation over the mission duration, i.e. 30 krad continuous
radiation @ 400 rad/h during 3 days. No frequency radiation
sensitivity was observed during the former test. For the latter
test no electronic failure or performance degradation was
observed, but it showed the need for wider compensation of
the drift of the crystal oscillator. The modification has been
implemented on subsequent models. The stability achieved
<2.5*10-14/day in ‘best temperature conditions’ under
vacuum of the RAFS1 model is shown in Fig. 3.
- A third development and qualifications step was
initiated at the end of 2001 and completed at the beginning
of 2003 with the delivery of an EM, which is the baseline
unit for the development of the flight models for GSTB-V2.
Two main objectives were achieved [2]:
Figure 1. Picture of the updated RAFS1 once closed including the
thermally regulated base plate
Figure 2. Five EQMs without external cover and vacuum chambers for life test with ‘Picotime’ measurement systems
Figure 3. RAFS1 EQM frequency data and frequency stability
- Further optimisation of the physics package to
reduce temperature sensitivity resulting better
short/mid term stability with a temperature &
vacuum environment similar to satellite platform
environement ( with +- 1°C temperature changes).
- Inclusion of a DC/DC converter and the satellite
TT&C interface compatible with ESA's new
requirements. Fig. 4 shows the performances
achieved in term of frequency & time stabilities.
Within this configuration RAFS2 shows
capabilities to perform time stability close to 1 ns
over 1 day.
Fig. 5 shows the internal construction consisting in RAFS
core unit equipped with the thermally regulated baseplate &
DC-DC converter.
Figure 4. RAFS2 core model frequency and time stabilities
Figure 5. RAFS2 internal construction
- In the frame of GSTB-V2, one EQM, one Proto-
Flight Model (PFM) and three Flight Model (FM) units
have been delivered (integration and tests on satellites are
on-going). Two FM spare units are under test and ready to
be delivered if required. Table I lists the achieved RAFS
performance for GSTB-V2. Fig. 6 shows the measured
frequency stability of GSTB-V2 PFM and FM1 to FM4.
- Further investigations to improve the flicker floor
and temperature sensitivity are under way . Beside the
‘zero’ temperature coefficient provided by the light shift and
gaz pressure shift into the cell , the lamp has also been
optimized and demonstrates ‘zero’ temperature coefficient.
Nevertheless, still temperature coefficients of 5*10-14/°C
have been observed. By improving the RF atomic
interrogation signal stabilisation circuitry , RAFS has
demonstrated stabilities in a range fo 7*10-15 for half of day
(Fig. 7) or more observation time. Power shift coefficient
has been measured arround 1*10-10/dB change in power.
Therefore, few ppm / °C of atomic interrogation signal is
required to reach stabilities within the 10-15 range. A
carreful worst case analysis of possibles temperature drifts
of parameters associated to the automatic gain control has
been performed and demonstates the feasibility and possible
repeatability of a RAFS having short term stability over one
day lower than 1*10-14.
TABLE I. RAFS FOR GSTB-V2 PERFORMANCE ACHIEVED
| Parameter | Measurement |
|---|
| Frequency stability | < 4*10-14 @ 10’000 sec |
| Flicker floor | < 3*10-14 (drift removed) |
| Thermal sensitivity | < 5*10-14 /°C |
| Magnetic sensitivity | < 1*10-13 / Gauss |
| Mass and volume | 3.3 kg and 2.4 liter |
Figure 6. GSTB-V2 RAFS2 frequency stability
Figure 7. RAFS3 frequency stability
B. Development & Qualification Activities of Passive
Hydrogen Maser
The space hydrogen maser will be the master clock on
the Galileo navigation payload. The first maser development
activity tailored to navigation applications was kicked off in
1998. It was initiated by the development of an active maser
at ON However, at the Galileo definition phase, it became
clear that the accommodation of the active maser on the
satellite was too penalizing in term of mass and volume, and
the excellent frequency stability performances of the active
maser were not required. In 2000 it was re-orientated
towards the development of a PHM based on the industrial
design and ON heritage on active maser studies.
The development of the EM (Fig. 8) [3] was completed at
the beginning of 2003, under the lead of ON with Galileo
Avionica (GA) subcontractor for the electronics package and
TNT supporting the activity in view of the future PHM
industrialisation. The instrument has been under continuous
test since June 2003 for assessment of long term
performance and early identification of reliability and
lifetime problems. This EM model (Fig. 9) shows the
frequency and time stability at first stage. By comparison,
about 5 years of design optimisation and intensive testing has
been necessary to reach such level of performances with the
RAFS.
The industrialization activity aimed at PHM design
consolidation for future flight production was started in
January 2003 [4]. The industrial consortium is led by GA
designing the electronics package with TNT responsible for
the manufacturing of the physical package and the ON
supporting the transfer of technology. The overall structure
of the instrument was reviewed to increase compactness and
to ease the Assembly, Integration and Test (AIT) processes
on the satellite by the inclusion of an external vacuum
envelope. Main efforts in the industrialization frame focused
on the definition of repeatable and reliable manufacturing
processes and fixtures, particularly for the physical package:
Figure 8. Picture of PHM EM. ON&GA