How PolyITAN-1 was born in the laboratories of KPI



    Design and manufacture


    The satellite was conceived for a long time, although there was no well-thought-out development plan as such, one could say there was a desire to start developing the satellite without official support from the Ukrainian Space Agency and the presence of the satellite at an early stage of development in the plans of our National Technical University of Ukraine “KPI”. The desire arose out of the blue, the Department of NPPs and Engineering Thermophysics at the Heat and Power Department of the KPI has a long history of close cooperation with the space industry. So, for the cooling and thermal stabilization of electronics on artificial satellites Sich and Ocean, space probes Vega and Phobos, several German and one Czech satellites used our heat pipes. But even with the faculty’s efforts, it was impossible to draw the whole spectrum of work, therefore specialists from other faculties were searched for - employees from the FEA, FEL, RTF were invited to develop, information was collected, and various consultations were held at conferences, as this is, nevertheless, the first satellite. Part of the work that could be done under the pretext of profile research of the faculty was done in advance.

    For example, the first version of the satellite’s body, lined with centimeter honeycomb panels, looked like this:



    As information was collected, it turned out that many problems could be avoided by converting the satellite to the CubeSat format and paying for a place in a group launch with other satellites. The thickness of the honeycomb panels decreased by almost three times, the undocking mechanism was not needed, and the satellite itself should freely fit inside the test structure, which is called a canister and was kindly offered to us at ISIS , Delft (Holland).

    Test canister (we prefer to call it a platform) Quadpack to check the dimensions of the satellite inside should slide freely.



    Work on a three-dimensional satellite model is not a problem, only space specifics are added to the drawings with the obligatory indication of the design coordinate system, the associated coordinate system and the orbital coordinate system, and the orientation of the sensors in these coordinate systems is also indicated. Orientation algorithms need the correct transformation of measurement results using transition matrices from one coordinate system to another. In addition, a three-dimensional model is the easiest way to calculate the center of mass and moments of inertia of a satellite with open antennas.

    Onboard computer complex board (BVK board in the photo with the installed GPS receiver from Navis-Ukraine) was developed at our TEFe by a specialist in embedded systems, hired specifically for working with a satellite. During testing, we managed to burn the COM port on the GPS receiver several times.

    The first version of the BVK board:



    The radio lines were developed at the Radio Engineering Faculty of the KPI, two radio channels were laid: one operates in the amateur radio frequency band at a frequency of 437.675 MHz (wavelength ~ 70 cm) and serves to transmit a beacon signal in a telegraph (CW) and telemetry at speed 9600 bps, and the other for data transmission in the amateur radio band of 145 MHz.

    Photos during the development of the software board:



    When it was already working, I was especially pleased with the development of data reception / transmission along with the BVK board. For two months in the laboratory there was a “squeak” from a ground-based receiver. You can listen to the “melody" on the website of the UY2RA radio amateur .

    Students from RTF and TEFA were involved in the installation of antenna equipment on the fifth building of the KPI KPI and laying cables from the roof to the seventh floor, where we have a control center.

    Our antenna economy:



    Power subsystem board with recesses for wires. It was developed by a graduate student of the KPI Electronics Department, he also chose LiFePO3 batteries and wrote software for his board:



    All boards were made at the Kiev Radar Plant .

    Putting the boards together, they worked out various algorithms and satellite modes for quite some time. The usual state of the workplace: a power supply with the help of a semi-disassembled device that simulates solar panels falling into the shadow of the Earth, powers a satellite with batteries. We checked in which modes the batteries have time to charge, and in which not.

    In the photo there are two satellites at once, in different types of assembly:



    Magnetic coils are usually used to control the orientation in such small satellites, which interact with the Earth’s magnetic field and rotate the satellite. Interestingly, for their calculation, an old book by A.P. Kovalenko, “Magnetic spacecraft control systems” by 1975, was used. 1975 At the debugging stage, instead of coils, resistors with two-color diodes were soldered, which can be found in the photo above.

    If we talk about the complexity of the software of BVK, it is easier to list the implemented tasks:
    • GPS real-time clock;
    • satellite modes and their switching based on various data;
    • cyclogram module;
    • work with a GPS / GLONASS receiver using the BINR protocol;
    • MODBUS protocol;
    • work with non-volatile flash memory for storing telemetry;
    • telemetry collection and storage module in the archive;
    • navigation algorithms (calculation and refinement of the orbit according to GPS data);
    • orientation and stabilization algorithm;
    • coil control module;
    • work with sensors of magnetic field, angular velocities, directions to the Sun;
    • interaction with power supply boards and radio lines;
    • a mechanism to put the kernel to sleep mode to reduce power consumption when there are no executable tasks;

    The necessary amount of program code for the needs of the satellite in programming its three boards does not end. To work out navigation, orientation and stabilization, a satellite orbit simulator was written in the matlab, simulating measurements of angular velocity sensors, magnetometers, and direction sensors on the Sun. The ground control center also requires software for parsing and storing archives of the full telemetry of the satellite, for creating cyclograms and sending them to the satellite, for controlling the reconfigurable parameters of the satellite, and all this via the MODBUS protocol. Assigning ground station programming to students 3-4 years before launch was not the best solution. The code was subsequently completely rewritten once from scratch, and students who completed their studies, due to employment in their main job, could not find the time to support the software they created.

    Direction sensor on the sun. We have been developing and manufacturing in our KPI. In a nutshell, the sensor device is described as a small solar battery with four current sensing sections, covered by an opaque panel with two mutually perpendicular slots. The outputs from the photocell go to a chip to amplify the signal, which can be further measured by the ADC on the BVC board. After processing the captured data, you can get a vector pointing to the Sun.

    Photocell of our sensor.


    The sensor itself with a closed panel, on the back there is a signal amplifier chip:



    Oddly enough, the biggest problem in developing the sensor was to find a biaxial rotary stand with a marking of angles for it. For the quality of lighting, our simulator of transatmospheric solar radiation was responsible, the same as for the thermal vacuum tests of our satellite. What will be a little lower in the text.

    Wanted stand:



    During the calibration process, one should not forget that the planet Earth also successfully reflects light and should be taken into account in errors:



    Sensors for constantly monitoring the direction vector to the Sun should be on every facet of the satellite, but in our case there are less of them.

    Test


    Vibration tests. Testing on equipment still Soviet-made. The satellite must withstand vibration and not fall apart. Information on overloads depends on the launch vehicle and in our case had the following meanings: axial longitudinal overload - 7.5 g, transverse overload - 0.8 g, integrated acoustic load - 140 dB.

    Photo of the vibration bench, unfortunately, only from a mobile:



    Thermo-vacuum tests. Conducted entirely by our own department at one of the divisions www.lab-hp.kiev.ua. During the tests, the conditions and influence of space factors were simulated: low temperature, vacuum, solar and terrestrial radiation, blackness of space, which affect the temperature regime of electronic equipment and its reliability. The Thermal Power Engineering Faculty specializes in such, because the tests were carried out on our own with a detailed description of programs, methods and the release of scientific articles. PolyITAN-1 is protected from the cold of outer space by our honeycomb panels, which are a lightweight aluminum honeycomb core (5 mm high cells made of 0.023 mm thick foil), glued on both sides with carbon-fiber plating with a dielectric polyimide film.

    Sotopaneli truth with aluminum covers:



    A solar radiation simulator should provide 1400 W per square meter, for large objects it can sometimes be like a solarium panel and placed in a vacuum chamber with the test object. In our case, the simulator is external. The light source enters the vacuum chamber through a small window and a window in the cryopanel - a nitrogen screen with a temperature of -193 ° C and a degree of blackness greater than 0.93.

    White stickers are just thermocouples for temperature control:



    An open thermal vacuum chamber, you can see the internal nitrogen screen with thermocouples, an old version of the satellite peeps out of the porthole.


    In the photo there is a general view of the equipment, on the left is a cannon simulating solar radiation, and it’s hard not to recognize a vacuum chamber with a porthole.



    The test results showed a temperature fluctuation on the faces of the nanosatellite solar batteries from -32 ° C in the shade and up to 65 ° C under the Sun. The temperature regime inside the satellite was within the tolerance for the equipment. The batteries kept within +5 ... +9 ° С, the microcontroller was heated from +5 to +23 ° С.

    Exposure to radiation.The tests were carried out on the M-30 microtron of the Institute of Electronic Physics of the NAS of Ukraine, Uzhgorod. The degree of degradation of the electronic components and solar cells of the nanosatellite was tested. From what they told me, this is the case: electrons with an energy of 7 MeV fly out of a microtron and knock out gamma radiation from the screen, which already gets on working boards and solar panels. Irradiation lasting about three hours (in two approaches 5578 s + 5578 s) imitates the effect of radiation for 18 months in near-Earth space. The boards began to fail by the end of the first 1.5 hours of testing. The degree of degradation of solar cells was only 3-5% and was calculated on the basis of the measured current before and after the tests with the same luminous flux.

    Photo of the M-30 microtron from the IEF website:



    Calibration of sensors.Absolutely all sensors on the satellite must be calibrated. Some of them are most important, as they are used in orientation and stabilization. These are magnetometers, angular velocity sensors (DOSs), direction sensors to the Sun. The rest relate to a greater degree to the state of the satellite: temperature sensors, as well as conditional sensors for opening the antennas, battery charge, current from solar panels, and current consumption of various subsystems implemented on the boards. About the direction sensor to the Sun was described above, the calibration of the remaining sensors necessary for orientation was carried out outside the KPI. In our satellite, duplication is used, a total of twelve sensors of DOSs and magnetometers. All of them are electronic components on the BVK board. Of the shortcomings, we can mention the extreme inaccuracy of the DOSs at low speeds,

    Testing the GPS / GLONASS receiver and our navigation algorithms. The company "Navis-Ukraine"in the city of Smela provided us with the opportunity to test our navigation subsystem on their equipment. It was important for us to verify the reliability of the mathematical calculation and the orbit refinement on real data from receiving GPS / GLONASS signal on the antenna and to using the obtained calculated data for satellite orientation. The GPS / GLONASS simulator allows you to simulate a signal coming from navigation satellites, as if we were flying in the orbit we had set. Such a device is quite expensive and should be bundled with software that translates the values ​​of consecutive orbit points or ground route points into a data file for the simulator. For a specified time section (orbit or route), the orbits of the selected navigation satellites are calculated, after which outgoing signals and the time of their arrival at the orbit points are calculated for each satellite taking into account SRT. The obtained data for simulation are recorded on a flash drive, which is already used in the GPS / GLONASS signal simulator. The simulator output is connected instead of the receiver antenna and launched for execution.

    GPS / GLONASS simulator:



    Checking antenna notches. The only moving part in the entire satellite structure is the mechanism responsible for opening the antennas after putting them into orbit. The satellite itself, located inside the launch container, is in the off state. But after being thrown into space, the power contact from the batteries is released, the included satellite with some delay should open the antennas. This is a rather important and crucial stage in the life of any spacecraft.

    The antennas of our satellite are like a tape of an ordinary roulette, constantly striving to turn around and straighten up. In the collapsed state, the antennas are held by the limiters; when current is applied, these limiters burn out and release the antennas.

    In the photo a satellite with deployed antennas:



    Antenna Notch Test:



    Assembly and road to launch


    Assembling in the basement of the laboratory together with checking all the connected elements took more than a day. Quite responsible work, performing which, according to assembly drawings, you need to connect the boards, wire all the satellite elements together into a single system (boards, external sensors, batteries, solar panels, antenna stripping mechanism, magnetic control coils), connect the radio and GPS / GLONASS antennas, correctly install directional sensors on the Sun on the panels, screw the panels themselves to the satellite. And at the end to prepare the satellite for transportation:





    Control weighing showed a beautiful number:



    On the road:



    In general, launching a satellite requires strong interaction with various government agencies and paperwork to obtain various permissions. This is the most uninteresting and unpleasant part in its development, which, unfortunately, is impossible to get rid of. If the satellite were not created in the capital of Ukraine, then at least one employee would live on business trips. Obtaining operating frequencies at the national and international levels, as well as paperwork for customs clearance, can take a long time. The international registration procedure takes a lot of time starting from "UkrDtsradiofrezov", "Ukrspetzvzyazku", "Ukrkosmos" and to ITU. In Ukraine, this procedure for university amateur satellites was carried out for the first time. In addition to this, ITU requires that the registration process be started two years before launch. Paperwork for customs is also a serious challenge related to the need to prove the absence of state secrets. However, the presence of these documents does not exempt from looking at the insides of our satellite when crossing the border, one panel had to be unscrewed.

    Such a satellite appeared before the border guards performing their duties:



    A successful moment can be called the repeated transfer of the launch. Six months have passed from the initial planned date for December 2013 to the launch on June 19, 2014, which allowed us to qualitatively refine a couple of subsystems on the satellite, but we still barely managed to do something. The issue of skipping the launch was quite acute.

    Final setup is already in the Netherlands before uploading to the QuadPack platform :



    Official and mandatory photo shoot for the satellite. Photo from blog.isilaunch.com CubeSat



    nanosatellite format means uploading to a standardized QuadPack platformto undock the satellites. In the photo there is a place in the platform for our satellite.



    The loading of 21 CubeSat satellites took place at the ISIS office in Delft in the Netherlands.



    Here are packed 21 satellites of the CubeSat format . Then they went to Russia to install the RS-20 ( Dnepr-1 ) missile on the upper stage .



    Installed QuadPack` and on the overclocking unit.



    Top view, there are more satellites:



    Flight


    On Thursday, June 19, 2014, late in the evening at 19:11 UTC, the Dnepr-1 rocket, on board of which there were 33 spacecraft, was launched from the Yasny Cosmodrome (Orenburg Region, Russia) . Below you can watch a video with animations of the launch and launch of satellites into orbit for this type of rocket from the Koreans.



    One of 33 devices is the first Ukrainian university nanosatellite PolyITAN-1, fully developed by employees, KPI students and enthusiasts. Control Center - UT4UZB is located at the Faculty of Heat and Power Engineering. The main thing is that the satellite was successfully put into orbit at 19:32 and was successfully received at a frequency of 437.675 (+ \ - Doppler). Now the CW beacon and telemetry of the FSK 9k6 format are working. The first to receive the signals was Yegor Kasminin UY2RA .

    On the seventh floor of the fifth building of the KPI in anticipation of the passage of the satellite:


    Also popular now: