| |
|
|
Translation by Georges Mathgen, LX1BB
|
|
|
SATEDU is planned with several objectives in mind. The first one is educational, as large parts of the studies, primarily those involving the structure, can only be conducted by gathering sufficient competences and time. The second one is to set a goal that will be relatively easy to reach. The third one imposes reduced design constraints which can be managed module by module. The fourth one is to build a structure that can be adapted to a maximum number of launchers while maintaining a certain "comfort" for the flown electronics at all low orbits and at altitudes between 400 and 1000 km. The fifth one is an easy access to ground stations ("Easy Sat"), for traffic as well as for telemetry which will be massive for demonstration purposes. The last one is to minimize the load on the command stations by designing a satellite capable of getting itself out of a delicate situation, either automatically or through the action of space mechanics. In all cases, if satellite control is lost, a stable setup, usable by the payload will be activated in order to operate it up to "the last crumbs". The structure Height of page
The network of POLYMECA schools, including CESTI in St. Ouen, ENSMA in Poitiers and ENSMM in Besançon, were chosen for designing the structure and for carrying out the thermal studies related to this project. These schools had the advantage of a good team work experience and of many projects carried out in cooperation with the industry. Therefore, it was possible to come up with a good design, integrating all aspects required to achieve the objectives as well as their practical realization. The structure was designed to house module boxes of identical dimensions stacked one on top of the other. They can contain a 160 x 100 mm circuit board. Their walls are 5mm thick and have a double function:
The structure is a cube of 300mm side length composed of a 20mm thick NIDA sole carrying three vertically stacked, 8mm thick aluminum beams which support the module boxes and provide a thermal path in all directions. These beams are kept rigid by four mounts upon which five solar panels will be fixed. The latter have no influence on the structural rigidity so that the conceptual constraints are lighter. An adapter, specific for each the type of launcher, can be attached to the sole. The structure was dimensioned to withstand Ariane 5 qualification standards with a safety coefficient of 2. Thanks to its thermal inertia, the assembly can function under all attitudes and for any orbit, either "cold" (with eclipse) or "hot" (without eclipse), while guaranteeing good electronic performance over a very acceptable temperature range that simulations made by ENSMA with THERMICA situate between 0 and 35 °C. A typical curve relating to a 146 MHz permanent transmission on a cold orbit is shown in the figures. Three assembly versions were studied to offer multiple configuration options. The first one, shown in the figures at the end of this paper, is the simplest. It features 2 x 7 identical module boxes mounted in stacks with the two batteries at the bottom. The second one features an emplacement on top of the structure, occupying all the room of that section. This emplacement is designed to house either a large circuit board or various modules that wouldn’t fit in a standard module box. The third one also features the emplacement at the top, but leaves some empty room, on the centre beam along its axis and very close to the center of inertia. An active stabilization unit is supposed to be mounted here. The number of standard boxes is reduced to ten. The modular structure allows the boxes to be positioned in a way to achieve the best thermal compromise. The antennas Height of page
They have already been described in a JAF article. The following conditions should be met: From a functional standpoint, it is necessary to ensure a good radiation pattern for all attitudes, especially on 146 MHz and 435 MHz. For 2.4 GHz, pointing is required because it was decided to use only one antenna in order to avoid antenna switching. From a mechanical point of view, the antennas must fit into the assembling structure of the different foreseen launchers and comply with the aerodynamic pressure constraints during the flight phase after separation of the protection cap. While this is easy for a patch antenna, it is not for large antennas such as the 146 MHz one. After numerous studies, the following configuration was adopted:
This configuration offers the best results in the largest number of cases. Above all, it offers the best decoupling between the 435 and 146 MHz antennas, which is favorable for good receiver performance. The radiation charts were presented in a former issue of JAF and the reader is invited to refer to that. It is to be noted that this antenna solution was also retained for PCSAT. Electrical Power Supply Height of page
The primary electricity generator is made up of four solar panels with a nominal power of 12 W each. The secondary generator is composed of eight 6Ah SAFT VR6 NiCd elements connected in series and supplying 9.6 V. The power supply feeds 3 regulated busses: one at 5V / 3A and two at 9V / 1.3A. The power supply was designed in view of maximizing its efficiency. It also comprises a charge regulator that switches automatically from nominal charge to maintenance charge by controlling the battery voltage. A time delay mechanism activates the power supply 30 seconds after separation. In case of non-detection of the separation or a failure in the time delay mechanism, a circuit comprising a lithium battery with controlled discharge will activate the power supply after approximately 3 weeks, guaranteeing its start-up in all circumstances. The average available power for the payload is 6.8 W for a heliosynchronous orbit at 500 km. This power takes the current necessary to recharge the battery into account while the main load is supposed to consume its nominal power permanently. The typical battery charge level is estimated at 50 %. It includes the charge as well as the discharge efficiencies. The following formula calculates the useful power Put as a function of the available power, Pdis (solar generators), the battery charge efficiency, n, the illumination time, Ti, and the eclipse, To. Put = Pdis * n * [Ti / (To + Ti * n)] An article by F5DKJ describes the results. The measured values are:
Attitude Determination Height of page
Several systems were envisaged for determining the satellite’s attitude. As the constraints were not very strong (5°), the choice turned towards a basic system measuring the incidence of solar flux on the satellite’s faces. The selected sensors were described in a LAF article this year and F6CGJ was in charge of manufacturing them. There is one sensor per face producing a voltage proportional to the cosine of the solar incidence angle. After normalization, the voltage values are directly used to determine the attitude since they are the cosines of the direction to the sun at the satellite’s reference point. From these elements, the right ascension, the declination of the rotation axis and the satellite’s orientation with respect to the Sun and the Earth are calculated. The entity is completed by vibrating piezoelectric gyrometers from MURATA measuring the rotation speeds along the three axes and providing the rotation vector at the satellite’s reference point. By combining both indications, all attitude elements are obtained with the necessary redundancy. The measured data are transmitted to ground stations by telemetry and fed into a prediction program which compares them with the expected values. This allows the sensor drift to be monitored and the coefficients to be adjusted in order to maintain the desired precision. Attitude Control Height of page
All three structure types house magnetic couplers. The main principle has been described in a LAF article. Their mechanical layout accounts for integration constraints, consumption and magnetic moment necessary to achieve pointing in a given direction in less than 30 minutes. The validation of these calculations is currently underway. Satedu is not planned to be permanently pointed towards the Earth, but only on demand. This applies particularly for the pointing of the 2.4 GHz antenna and for reorienting the satellite in situations where its rotation would place the sun on the Z- face, which is not equipped with solar cells. The satellite will therefore be reoriented before it gets into such a situation. The geomagnetic field is not at all uniform and the previously cited program uses a model calculating the values of the field components at the satellite’s reference point as a function of its attitude and position in orbit. Thus, the program can determine the instructions to be sent to the magnetic couplers by taking the following elements into account:
The map below shows the shape of the geomagnetic field at 800 km. The vectors indicate the orientation of the local horizontal component while the size of the squares is proportional to the component’s magnitude at the nadir.
It appears that not all maneuvers are possible at any time. The strategy for pointing the satellite must take the envisaged maneuver into account in order to choose that part of the orbit which allows the pointing to be done, as well as the satellite’s attitude before the maneuver. There is one magnetic coupler on each face, except for the Z- face which has none. Therefore, the one on the Z+ face must have a double magnetic moment in order to achieve the same torque as the others. The magnetic couplers are composed of 100 turns of copper wire wound in the shape of a square of 25 cm side length. The nominal current needed amounts to 0.7 A. The couplers are controlled in pairs, X+ and X-, Y+ and Y-, Z+. The control unit comprises a MOSFET transistor bridge as well as a safety and a current measuring circuit. To validate the calculations, we created a model that enabled us to verify that the geomagnetic field fulfilled its role correctly. (We really knew that, but it’s so much fun to check it, HI!) The model is shown in the following photo.
The magnetic couplers are not placed in their nominal configuration, but this has no influence on the validation of the results we are looking for. Hams can easily reproduce this experiment which is also a good demonstration for QRPs. The amount of material involved is small: 200 m of 0.3 mm enameled wire, a 2 or 3mm thick wooden chipboard, a domino, 2m of very thin and flexible multi-thread wire (0.1mm thick, for example earphone wire), 3m of wire, as thin as possible, but strong enough for suspension. The forces involved are very weak and friction has to be minimized. The wooden board is suspended from the ceiling with the feed wires pointing downwards, vertically in relation to the board in order to reduce parasitic coupling. The model will be fed with 12V, the two coils being mounted in parallel. Depending upon the direction of the current in each coil, the board will be oriented facing the local magnetic north, east ort west. Even the slightest air movement has to be avoided. With the assembly shown in the photo, orientation will occur in about 3 minutes after the oscillations cease. Active stabilization Height of page
Here again, the principle of stabilization upon demand applies. The available power being limited, it is impossible to have the 2.4 GHz antenna permanently pointed towards the Earth. The emplacement in the centre of the satellite can house three reaction wheels. These are normally at rest and direction pointing is achieved by sequential rotation of each wheel by controlling the angle of rotation at a fixed speed. Up to now, no particular solution has been chosen. ENSICA has developed a unique design that is particularly suitable for nanosatellites and picosatellites. Unfortunately, lack of time only allowed the construction of a demonstration model which nevertheless permitted us to prove the validity of the hypotheses. Anyway, the satellite will be spun at low speed after separation from the launcher. The speed will be around 5 rpm. This value is sufficient to obtain a good kinetic momentum, compatible with the required couples to allow pointing. It also ensures a good distribution of the thermal fluxes on the faces and inside the structure as well as a small enough deviation of the rotation axis. The Payload Height of page
The operating modes will be as follows:
At least two functions can be simultaneously activated. The end-of-life mode is the linear transponder mode (which will also pass the FM), 435MHz -> 146MHz. One possible functional diagram (among the possible configurations) is shown at the end of this article. The two 435 MHz receivers are fed by the antenna coupler at the input stage. They each possess a demodulated AF output and a linear output at 10.7 MHz. The output level is 0dBm. The telemetry elements are: Output level on 10.7 MHz, RSSI RX, and VCO voltage. The receivers use double conversion with a first IF at 50 MHz. Each receiver’s consumption is lower than 200mW. They are permanently switched on. The AF and 10.7 MHz outputs are connected to a switching matrix which comprises also the modems, the FM modulators at 10.7 MHz and the voice recorder. Depending on the selected mode, the 10.7 MHz signal for the transmitters is generated either by the receivers or by the FM modulators. An FM transponder on 2.4 GHz and a linear tranponder on 146 MHz can be switched on simultaneously. In digipeating mode, phone traffic operation is switched off. On 146 MHz, the normal output path of the transmitter is multiplexed with the telemetry transmitter to the antennas. The output level of the FM modulators is measured. The board is controlled by 8 bits in order to produce different configurations. The 146 MHz transmitter has simple conversion with emphasis on global transmission efficiency and reliability. Its output power can be varied in 2 steps of –3 dB and is nominally rated at 3W. The objective of reducing the third harmonic’s level is important because it is close to the RX input frequency. The decoupling between the antennas favors the isolation. The measured values are:
The TX is controlled by 4 bits. The 2.4 GHz transmitter output is directly linked to the antenna. It is a double conversion TX (70 MHz for the second IF). The second conversion uses reduced sideband in order to simplify the filtering and increase the output power. The nominal output level amounts at 23 dBm at the antenna. However, if an architecture with good efficiency can be found, the level could be increased to 2W. Its level is adjustable in 2 steps of –3dB each. The measured values are the same as those of the 146 MHz transmitter with the OL2 level added. It is also controlled by 4 bits. A wide band input (10 MHz) at 70 MHz is planned for the case a 1.3GHz receiver could be added. This would eventually allow ATV operation by satellite. However, the link balances will necessitate sophisticated reception equipment. (D= 2.5m on receive for 2 Watt transmission level at the satellite.) The interface board for the structure accomplishes the following functions:
The board was built by the "Trois Bassins" high school at Réunion under the direction of Jean Marie Vacheron. Jean Paul, FR5CY, was responsible for the interface and the coordination. The computer Height of page The computer will soon be presented in a separate article.
Connections between modules Height of page
There are three types of connections:
This configuration minimizes the connections, guaranties a good reliability of control signals and allows interfaces without too much complexity. It also permits each module to be controlled through a PC’s parallel port during laboratory tests. Concerning the control of the modules, the default parameter bits are all set to zero. This way, in case of a computer failure and thus the loss of control of the satellite, the payload is set by default in a usable mode so that the satellite can remain operational up to its definitive end of life. Conclusion (by FAQ) Height of page The launching ???? : WHEN WE WILL BE READY !!! The launcher ??? : Ariane 5 Lifetime ???? : 3 years minimum, 100 years if all goes well (HI) Who’s working on it ???? never enough people !!! May I participate ??? : Yes !!! (on the menu: sweat, blood and tears) The color photos will be placed on the Satedu website in several days. F1HDD/ON1RG |
