Figure 1. User spacecraft coverage from Artemis
Artemis, presently in Phase-C/D with Alenia Aerospazio (I) as Prime Contractor, will carry out two main missions for data relay and L-band mobile communications.
Satellites in a Low Earth Orbit (LEO) are restricted in their ability to communicate with the ground because of the limited contact time. Artemis' location in the geo-stationary orbit will make it possible to relay data from a LEO satellite to Europe while over a large portion of the globe (Fig. 1).
Figure 1. User spacecraft coverage from Artemis
Not only does a data relay service extend the duration of contact time between a LEO satellite and the ground, it also allows the data to be delivered directly to the 'door-step' of the LEO satellite's management organisation. For example the SPOT-4 telemetry and remote- sensing data will be received in Aussaguel near Toulouse close to the operations control centre and image processing centre, while for Envisat data will be received directly at the processing centre at ESRIN, in Frascati (I).
The Artemis data relay payload provides feeder links between Artemis and the ground and inter orbit links (IOLs) between Artemis and the spacecraft in LEO. The feeder links operate at 20/30 GHz, while the inter orbit links can operate at S-band (2 GHz), Ka-band (23/26 GHz) and optical frequencies.
The feeder link, S-band and Ka-band payload elements jointly comprise the SKDR (S/Ka-band Data Relay) payload while the optical IOL payload element is called 'SILEX' (SemiconductorIntersatellite Laser Experiment).
The SKDR payload frequency translates, filters and amplifies the signals passed through it. The section of the SKDR payload used for the forward and return S-band IOLs has a bandwidth of 15 MHz in both directions. The section used for the forward Ka-band IOL has a bandwidth of 30 MHz, while the section used for the return Ka-band IOL has one channel of 30 MHz and two channels of 230 MHz bandwidth.
The centre frequency of these channels can be set to a range of values since the local oscillators are all settable to frequency synthesisers locked to a common 10 MHz highly stable master oscillator. The channels all have a minimum amplitude and group delay ripple across their bandwidth and a minimum phase noise, implying the channels will hardly distort the signals passed through them.
This is a significant characteristic of the SKDR payload since it conserves the resources consumed by the spacecraft in LEO for receiving and transmitting data via Artemis.
The engineering model of the SKDR payload (Fig. 2) has been successfully tested. The flight model is being integrated and some sections of it have also been successfully tested.
Figure 2. Part of the north panel with the engineering-model SKDR return repeater
Artemis is equipped with one IOL antenna having a feed capable of operating at S-band and Ka-band. The IOL antenna is an offset parabolic reflector antenna with a 2.85 m aperture. The antenna is steered in the direction of the LEO spacecraft by rotating the reflector around its focal point by means of a pointing mechanism controlled by an on-board computer. The computer controls the antenna pointing either in open or closed loop mode. In open loop mode the pointing direction is derived from a pointing table loaded by ground command into the computer. In closed loop mode the antenna acquires the LEO spacecraft using a pointing table, and then it corrects the pointing direction and tracks the LEO spacecraft based on error signals derived from the higher order electromagnetic modes in the antenna feed and pre-processed by a track receiver.
When the IOL operates in S-band, the antenna pointing is always performed in open loop, while for a Ka-band IOL the antenna may be pointed in open loop or in closed loop.
To assist the LEO spacecraft in tracking Artemis an unmodulated wide beam beacon signal is broadcast by the latter at 23.540 GHz.
The engineering model of the IOL antenna has been successfully tested and an extensive testing of the closed loop tracking system has been completed. The flight model of the IOL antenna and its tracking systems is being manufactured and tested at unit level.
SILEX is the world's first civil intersatellite data relay system using lasers as carriers for the signal transmission. Two SILEX terminals have been developed and built; one will be launched on the French SPOT-4 Earth observation spacecraft at the beginning of 1998, the other aboard the Artemis satellite.
The two terminals will allow the image data of SPOT-4 to be transmitted at a data rate of 50 megabits per second (Mbps) via the Artemis feeder link to the SPOT-4 earth station near Toulouse. In addition to the data transmission between SPOT-4 and Artemis, the SILEX terminal on board Artemis will also support an experiment between Artemis and the Japanese LEO spacecraft OICETS.
During this experiment, the data rate from OICETS to Artemis will also be 50 Mbps but, in addition, there will also be an optical link from Artemis to OICETS. Via the latter, a link data stream of 2 Mbps can be sent from the ground via Artemis to OICETS.
The principal advantage of using optical wavelengths for the intersatellite link stems from the capability to obtain extremely high 'antenna' gains with relatively modest apertures and, consequently, to use only very limited carrier power. In the case of SILEX, the 50 Mbps transmission over a distance of 42000km can be performed with an optical power of about 60 mW only. As an 'antenna', SILEX uses a 25 cm-diameter telescope on both terminals which provides an 'antenna' gain of well above 100dB.
The disadvantage of these extreme antenna gains is the very narrow width of the transmitted beams and, therefore, the need for very accurate pointing. The communication beam of SILEX has a width of about 8 µrads or about 0.5 mdeg. This results in an illuminated zone of less than 400 m diam. at a distance of 42000 km. To avoid intolerable power losses, the optical beam must be pointed towards the partner with an accuracy of few micro-radians. At this extreme accuracy even the micro vibrations present in normal spacecraft, generated by the operation of the various mechanisms, are disturbing and need careful correction in the pointing loops of the terminal.
Prior to establishing the optical link, the position of the partner spacecraft is not well known. In the case of SILEX, the partner will be in a zone of uncertainty which is seen from 42000 km under an angle of 5500 µrads. Since the optical communication beam has a width of only 8 µrads, a search manoeuvre has to be performed. For this purpose the GEO terminal on Artemis is equipped with an optical beacon which generates, by means of 19 laser diodes, an optical power of typically 10 W which is transmitted within a beam of 750 µrads. To find the partner spacecraft in the 5500 µrads zone, the beacon beam is scanned over the zone of uncertainty until the partner is illuminated by the beacon signal. In this case, the LEO terminal will detect the optical signal and will, in turn, transmit its narrow beam signal towards Artemis. When Artemis detects the LEO signal it will stop scanning its beacon and will centre its transmission direction towards the LEO terminal. Once the pointing is sufficiently accurate both communication beams will be transmitted and the system is ready to start the data transmission.
In order to establish the optical link over a large part of its orbit, the terminal on board SPOT-4 has to have the capability to point its telescope over a large angular range.
For this purpose a two axes gimbal mechanism was developed which carries not only the telescope but also the whole optical head including the optical transmitters, receivers, pointing mechanisms and the control electronics which have to be close to the optical head. In total a mass of about 100kg has to be moved over nearly the full hemispherical coverage.
One further particularity of an intersatellite link with extremely narrow beams is the need for a point-ahead system. As noted above, the optical beam at the partner satellite has a width of less than 400 m. The relative velocity of the two spacecraft is, however, between 0 and 7000 m/s. Taking into account the time needed by the optical signal to travel twice over the distance of 42000km, the spacecraft would be far out of the optical beam when the light arrives at the place where it was two light travel times ago. For this reason it is important to know the orbital parameters of both spacecraft exactly. This point-ahead angle changes continuously during one orbit and has to be updated at short intervals. This procedure requires continues data processing and a very accurate pointing mechanism whose bandwidth can, however, be relatively low. In contrast, the mechanism which provides the fine pointing accuracy of the incoming beams has to have a high bandwidth of several hundred Hertz.
Both terminals are now fully integrated. The LEO terminal (Fig. 3) was delivered to SPOT-4 in March and the GEO terminal to Artemis in June 1997.
Figure 3. SILEX flight-model terminal to be accommodated on SPOT-4, under test at Centre Spatial de Liège (B)
To allow for a proper check-out in orbit of at least the GEO terminal, an optical ground station has been built in Tenerife on the Canary Islands where viewing conditions are among the best in the world. This station is equipped with a one metre telescope which simulates a LEO terminal. Together with the feeder link station in Redu (Belgium), it will be possible to perform a realistic bi-directional data relay experiment involving both optical data links.
In parallel with the development and manufacture of the Artemis data relay elements, preparations are under way with the users of the data relay services.
The main initial European data relay users are SPOT-4 and Envisat. As described above SPOT-4 will carry a SILEX optical terminal and will communicate via Artemis with its control centre in Toulouse. SPOT-4 will also use Artemis for relaying its telemetry signal through an S-band IOL.
Envisat will use Artemis to relay its instrument data to the ground via a Ka-band IOL. The Ka-band terminal has been described in detail in ESA Bulletin No. 88, November 1996.
As part of a collaboration agreement with NASDA, ESA will make part of the Artemis data relay capacity available to Japan. NASDA is planning to use this capacity to provide data relay services to several of their satellites including ADEOS-2, a remote-sensing satellite, and JEM, the NASDA module of the International Space Station, in addition to the OICETS optical experiment described above.
In addition to the data relay payload, Artemis carries a payload which supports the communication of mobile users with fixed partners located anywhere in Europe, North Africa and the Near East. The LLM payload is fully compatible with the EMS payload already developed by ESA and flown onboard the Italsat-2 spacecraft, thereby also providing full redundancy for its mission.
The LLM payload receives the signals transmitted by the fixed users at Ku-band (14.2 GHz) and transmits them at L-band (1550 MHz) to the mobile users. This link is called the forward link. The return link establishes the connection from the mobile user at L-band (1650 MHz) to the spacecraft and at Ku-band (12.75 GHz) from the spacecraft to the fixed user. About 400 bi-directional user links can be established simultaneously.
Since the bandwidth at L-band is a very scarce resource, it is mandatory to make efficient use of the available frequencies. Therefore, the use of Surface Acoustic Wave (SAW) filters was selected. These elements allow two adjacent channels to be positioned next to each other with only 200 KHz guard band between them, enabling a 90% use of the spectrum when the total bandwidth is split into 1 MHz channels and a 97.5% use of the spectrum when the channels have a nominal bandwidth of 4 MHz.
To simplify the coordination of the operation with other satellite systems using the same frequency band the total installed nominal bandwidth of 15 MHz was subdivided into three 1 MHz channels and into three groups of 4 MHz channels. Each channel can be shifted individually and independently from each other within the available 29 MHz band in steps of 0.5 MHz.
To allow a high number of users to exploit the payload simultaneously, it is essential to simplify the mobile user terminal on the ground and to ensure that the EIRP (equivalent isotropic radiated power) per individual user signal is high and also that the total EIRP, i.e. the sum of all signals, is high. This can most economically be achieved by increasing the antenna gain which in turn requires the generation of spot beams on the satellite. The LLM payload provides one large beam covering the whole of Europe, North Africa and the Near East, but also provides three spot beams together covering the same area on the Earth.
The size of the spot beams is constrained by the fact that a solid reflector of only 2.8 m projected aperture could be accommodated on the spacecraft. The L-band power stage applies a butler-like matrix concept. This configuration allows a relatively high efficiency which is independent of the actual traffic distribution over the antenna coverages.
The Artemis platform evolved from previous European telecommunication spacecraft and, hence, is largely based on a classical 3-axis stabilised geostationary concept. The platform has been designed not only to accommodate the payload elements of the Artemis mission but, with minor modifications, also those of other missions.
The platform's central propulsion module (Fig. 4) houses the three propellant tanks (two chemical and one xenon), the liquid apogee engine, the pressurant tanks and the bulk of the pipe-work feeding the reaction control thrusters. The east panel carries the L-band antenna and feed, the west panel the inter-orbit link antenna. The north and south panels are the prime thermal-control radiation areas which accommodate the bulk of the electronic equipment, particularly the highly dissipative units. The Earth-facing panel carries the optical payload (SILEX), TTC antenna, feeder link antenna, as well as various attitude and orbit control sensors. The two batteries are mounted on separate radiator plates, one on the north face and one on the south.
Figure 4. Lower part of the central cylinder of the flight- model spacecraft with the Liquid Apogee Engine
The structural design has followed a largely classical approach utilising aluminium honeycomb material. However, the central cylinder is aluminium honeycomb skinned with carbon fibre. The primary structure provides the load path to the launch vehicle interface and comprises the central cylinder, main platform, propulsion platform and four shear panels. The major elements of the secondary structure are the north and south radiators, the east and west panels and the Earth-facing panel.
The structural model qualification test campaign has been successfully completed and the complete flight structure has now been delivered.
The satellite thermal control also follows a classic approach using primarily passive techniques employing optical solar reflectors on radiator surfaces and multi-foil insulation blankets on the majority of the remaining external surface. In addition, the efficiency of the main radiators is enhanced by the use of heat pipes which are uni-directionally mounted under highly dissipative and/or sensitive equipment.
An innovative aspect of Artemis is the employment of a thermal technique to measure the amount of on-board chemical propellant.
Although already successfully tried on other ESA telecommunications spacecraft, Artemis will be the first spacecraft to have the method designed in from the start. The method has therefore dictated, to some extent, the thermal design around the propellant tanks and dedicated calibration tests are to be performed during the spacecraft solar simulation tests.
Equipping of the flight spacecraft with the thermal hardware is presently progressing well and the analysis campaigns are at an advanced stage. The subsystem will be proven during dedicated solar simulation testing in early 1998.
Power is generated by two identical solar array wings each of four panels made from CFRP (Carbon Fibre Reinforced Plastic) sandwich. Each wing is supported by a yoke, which is attached to the spacecraft via drive mechanisms located on the north and south faces.
The array is partially deployed in transfer orbit by cutting the Kevlar cables of the hold-down mechanism. Full deployment is achieved on reaching geostationary orbit.
The array is designed to deliver just under 3 kW of power during equinox after 10 years in orbit.
The flight solar array wings and drive mechanisms have been built and successfully tested and are now in storage awaiting integration at satellite level.
Power storage is achieved with two identical 23-cell nickel-hydrogen batteries, each with a nominal capacity of 60 Ah. They are equipped to deliver just over 1800 W during eclipses of up to 72 minutes duration at a depth of discharge, with no cell failures, close to 75%.
The flight cells have already been manufactured and are expected to be assembled into batteries by the autumn of 1997.
The spacecraft power is distributed via a single, fully regulated 42.5V bus.
Excess power from the solar arrays is shunted by the Solar Array Regulator which also controls the start of battery charging at the end of eclipse.
The Battery Regulator Unit controls and regulates the power bus during battery discharge, regulates the battery charging and monitors the main battery parameters (voltage, temperature, pressure, current).
The Pyro and Knife Drive Unit, as the name implies, drives all the on-board pyrotechnic devices for appendage release and initiation of the propulsion subsystem.
A Thermal Control Unit provides automatic heater switching based on pre-set thresholds and also conditions the output of temperature sensors. The heater control has a ground over-ride facility.
Each power line is protected individually as close as possible to the power source using redundant fuses. This is managed by the Power Protection and Distribution Unit.
The nominal operation of the subsystem has been successfully confirmed on the engineering model (EM) spacecraft.
The flight hardware has already been built, tested and delivered.
The UPS is a conventional bi-propellant system comprising a single 400 N Liquid Apogee Engine (LAE) and a set of 10 N Reaction Control Thrusters (RCT). The latter are configured into two identical redundant branches, each of eight thrusters.
The UPS will be used for apogee boost, longitude control, wheel off-loading, any re-location manoeuvres and re-orbiting at the end of life. Inclination control will be performed by the Ion Propulsion Subsystem (IPS).
The propellant is stored in two 700 litre Cassini-shaped tanks, one containing the mono-methyl hydrogen fuel and the other the nitrogen tetroxide oxidiser. The total bi-propellant to be loaded will be about 1538 kg. The propellant tanks are pressurised by helium stored in three smaller spherical tanks.
The LAE operates in a pressure regulated mode at about 15.7 bar. Once artemis is in the required orbit the LAE will be isolated and the RCTs will then operate for the remaining spacecraft life in blow-down mode.
All the flight hardware has been delivered and integrated on the flight model spacecraft.
The Artemis platform will be the first ESA satellite to fly electric propulsion technology operationally. It will be used for inclination control throughout the satellite's lifetime.
The IPS consists of two thruster assemblies, one mounted on each of the north and south faces. Each assembly comprises an Ion Thruster Alignment Mechanism upon which two redundant thrusters from different sources are mounted; a Radio-frequency Ion Thruster (RIT) from DASA (Fig. 5) and an Electro-bombardment Ion Thruster (EIT) from MMS.
Figure 5: Engineering model of the Radio Frequency Ion Thruster, courtesy of DASA
Each thruster has its own power supply and control equipment as well as its own flow control/propellant monitoring units. There is a common propellant supply and distribution assembly. The propellant used is xenon, of which 40 Kg is loaded on the satellite.
The main characteristics of this propulsion technology are its high specific impulse (3000 sec) and low thrust level (20 mN), in contrast with chemical propulsion. In the case of Artemis, this results in a net mass saving of about 60 kg which can be advantageously used elsewhere within the satellite.
In operation the system draws about 600 W of power, which is mainly supplied directly from the solar array, augmented later in life by the batteries.
A vast amount of development work has been performed to bring this subsystem to maturity. This effort is now beginning to reap rewards as the individual equipment items successfully negotiate their qualification testing and flight unit delivery gets underway.
The ICDS is the other major novel platform area of Artemis. It departs from the more traditional independent architecture of the On-Board Data Handling (OBDH) and the Attitude and Orbit Control (AOC) subsystems. This subsystem utilises a centralised processing architecture which not only supports both the afore-mentioned functions but also includes a Fault Detection, Isolation and Recovery function (FDIR). The ICDS offers the different spacecraft equipment data collection and command distribution over the standard OBDH bus either by direct connection to this bus or through dedicated interface equipment.
In addition to the On-Board Computer Unit (OBCU) and the dedicated interface equipment mentioned above, the subsystem also includes the more traditional inventory of reaction wheels, momentum wheels, Sun sensors, infra-red Earth sensors, gyros and supporting electronics.
Artemis has followed a three-satellite-model development strategy. Structural and mechanical integrity has been demonstrated with a structural model (Fig. 6) which has successfully been subjected to acoustic and sine vibration testing at qualification levels, while an engineering model of the satellite has been used to verify the electrical compatibility of all of the satellite equipment.
Figure 6. The Artemis structural model under test
Integration of the flight model spacecraft is now underway. Late in 1997 the fully integrated spacecraft will be transported to ESTEC for environmental testing. By late 1998 Artemis will be ready for storage awaiting the start of the launch campaign.
As part of a broad cooperation programme between NASDA and ESA, it has been decided to launch Artemis on the Japanese H-IIA launcher. This launcher is an improved version of the existing H-II and will be able to deliver 2 tons of payload into geostationary orbit. Its design is based on a first stage made of a liquid hydrogen/liquid oxygen engine providing 110 tons of thrust and two solid-rocket boosters providing 230 tons of thrust each. The second stage, based on a liquid hydrogen/ liquid oxygen engine, provides 14 tons of thrust.
Compatibility studies between Artemis and H-IIA started in 1996 and are continuing in 1997. The environmental levels foreseen from H-IIA are similar to those of Ariane-5 for which Artemis was originally designed. A preliminary coupled-load analysis based on a simplified spacecraft and launcher mathematical model has been run and the results are being analysed and compared with the results from Artemis structural model tests.
The launch campaign will start at the end of 1999 with the shipping of the spacecraft to the airport in Kagoshima. From there it will be carried by boat to Tanegashima, a small island in the south of Japan. The operations at the launch site will start with a final check-out of the spacecraft. It will then be loaded with its propellant, in a dedicated area, and a few days before the launch it will be mated to the launcher itself. The full assembly will then be rolled out from the final assembly building to the launch pad some 8 hours before lift-off.
The launch of Artemis, with its circa 3.1 tons of lift-off mass and just under 3 kW of DC power, from Tanegashima in Japan at the beginning of the year 2000 will mark the start of the ESA data relay system which, together with the large number of activities on the users space and ground terminal front, will ensure that Artemis is only the start of ESA's data relay adventure.
ESA Bulletin Nr. 91