LAST UPDATE : January 16 2021


A tiny Radioamateur LEO Satellite project

After the emergence, around 20 years ago, of the Cubesats and COTS (Commercial Off-the-shelf components in space) paradigms many efforts have been made to simplify and try to reduce the price of hardware, and especially the "astronomical" costs of launch. We have seen satellites really reduced to the bones, like the Sprite credit-card size "chip-sats" (not to be confused with NASA's CHIPSat). Some actors have been the luck to "hitch" free trips to outer space. A recent format is emerging, the femto-sats of the PocketQube class, whose dimensions don't exceed 5x5x5 cm. and the weight 250 grams. Such a small form factor offers opportunity to reduce cost and have a standard, in exchange poses issues and difficulties due to the very small space available. This is not a lilliputian piece, with just a beacon inside. This is a project of a radioamateur sat with a decent function and performance, inexpensive and whose circuitry fits into 0.5 or 0.75 unit of a PocketQube. As much as possible using common components ubiquitous and easy to find everywhere. No tech-divide, please. PREAMBLE Actually, it's very complicated and/or expensive, to send a sat in orbit (of course). Anyway, if you don't succeed or abandon the idea from the start and you can't operate the stuff "on-space" you still can use my project "off-space" to learn (or teach) the basis of RF and the nano satellite technology. This is not a software simulator but an actual satellite device, based only on traditional RF techniques (anyway, I'm working on embedded SDR solutions) and analog operation, highly useful for educational purposes. There's plenty of theoretical papers on smaller sats around but very little practical stuff. It could also be modified as a terrestial transponder/beacon, Il ne faut pas rever... Don't imagine to build such stuff just with a solder iron and a multimeter. Anyway, I strived to make something simple, because my self I got no racks of Expen-sight or Agi-BankLoan equipement. No need thousands $ of spatial measure and RF equipement, but you must anyway have a decent RF Lab, and you must be skilled in SMD techniques and manage miniature devices. NOTICE If you really got launch opportunities please remember that's COMPULSORY to apply to ruling boards, like FCC or equivalent, the IARU and the like, for authorizations. Especially U.S. FCC don't appreciate at all people that launch without follow the guidelines or don't apply IARU coordination (and could fine thousands of $ !!). Such a sat is intended to operate into radioamateur frequencies segment, so you must be a (or, rely on a) licensed radioamateur to operate the device in terrestrial or outer space environment (of course if you don't actually transmit ie you connect the PA to a dummy load, you don't need a ham licence) THE CIRCUIT It includes a main CW beacon, a secundary voice beacon, a minimal V/U (VHF to UHF) transponder, with the possibility to add a data module for telemetry. The block schematic is showed here The trasponder performance should be tailored to allow the use as ground stations of simple handheld crossyagi antennae and Baofeng style handheld VHF-UHF RTX's TRASPONDER The VHF input frontend is classic, based on a double-gate mosfet like BF966 or similar. The second gate is useful as AGC/Limiter. V.H.F. input frequency is sent to a glorious NXP (Philips) SA636 for conversion to shared 10.7 Mhz intermediate frequency (I.F.). The SA636 consiste of a Gilbert-cell multiplicating mixer, and a high performance I.F. chain. The SA636 output is switched with the beacon output, so when the beacon is operating, the transponder is offline, to avoid mess. SA636 have no frequency synthetizer onboard, and it is notorious that operate the internal oscillator with a crystal over 120 Mhz is tricky and pesky. No matter, of course, of LC tanks with varactors and the like. To assure the mandatory frequency stability and precision we must use some external PLL synth. For various reasons, the "obvious" use of PLL's with internal VCO and fractional divisor with frequency settings by registers, like the famous Analog ADF43xx series is not practical here. Then I adopted compact fixed frequency PLL's. This approach have a drawback for actual space use, that you need special XTAL's made for purpose but if you can afford the cost of a launch you can for sure pay the price of custom made XTAL's. For testing purposes and academic applications other useful value of frequency could be employed. The VHF L.O. is made out of a Dallas DS1080L, a 4X PLL that gives, with a standard 33.360 Mhz XTAL the LO frequency of 133.440. If we add the I.F. value of 10.7 Mhz, we fall well into the "ham" range of 2 meters @ 144.140 Mhz For UHF converter I used another device from Maxim-Ic a miniature chip performing a 32x PLL multiplication, the MAX7044. Value of 444.000 Mhz is obtained with a 13.875 Mhz standard XTAL, that minus the usual 10.7 Mhz I.F. gives the frequency of 433.300 into the segment of 70 cm. band. An alternative is a less common but standard XTAL of 13.824 Mhz. BEACONS Time to time, as customary, a beacon signal must be sent. The CW beacon code is performed by a tiny microcontroller, like Microchip 12F675. In addition there's a voice beacon sending a short pre-recorded voice message. To fit into the form factor of PocketQube, I used an inexpensive module with a recorder/reproducer that can be easily operated by onboard computer. It's a tiny "nude" module, without press-buttons and the like. A 10.7 Mhz DSB is generated from voice module output with another glory of RF electonics, the SA602. A slightly more complex but improved alternative involve a proper modulator, a MC1496, able to generate clean DSB with an accurate carrier rejection. OUTPUT STAGES The signal coming from selector comes to filtering, and further (up-)conversion There's a filtering performed by a 10.700 Mhz crystal filter. The goal of filter is dual: to clean the trasponder signal into a narrow bandwidth, and precisely filter the DSB beacon' signal to make it SSB. It's the simpler method, to avoid dealing with DSP's and SDR's. Another SA602, whose ports are working at 450 Mhz (official max. is 500 Mhz) performs the conversion to UHF 70 cm. space segment The power is provided by a chinese module dubbed Nwzd RF-PA 2.0 whose dimensions of 44x22 mm. nicely fits into main PCB. Some 25 dBm (300 mW) on 70 cm. should be sufficient for purpose. The module is inexpensive and a good aspirine against headaches involved in design and building an efficient amplifier for UHF. (but i'd like someone help to develop more efficient solutions) After testing I discovered that 9.6 volts is a reasonable limit to operate reliably without a bulky heatsink (remember, outside the terrestrial atmosphere there is quite little dissipation of heat) and give a 20dbm output for a 0 dBm input, quite constant from 50 to 438 Mhz. Current is slightly less than 200 mA. POWER SUPPLY Again, the room is very small, in practice we can put only one compact 3.7 volts flat profile rechargeable battery element and a limited amount of solar cells to power the device. As we need a supply rail @ 5 volts, and another one @ 3.3 volts plus 8 volts for power amplifier I found the best solution with a step-up converter. There's a multitude on market, I used a Maxim MAX756, that I had in the junkbox. I tested stable operation down to around 2.0 volts,so there is a good margin to manage battery discharge. Now we need only a chassis, the solar cells and an antenna. A first prototype is already on test bench. Here the block layout A. Frontend and downconverter mixer B. IF and AGC C. Voice reproducer D. DSB modulator E. VHF L.O. F. Upconverter mixer section G. UHF L.O. + beacon H. On board computer I. Modules interconnect On top : Power amplifier unit The main module top side, with the PA piggyback-ed on top. Detail of front-end + IF The assembly is composed of a main double face single board, and a single face PCB board for the power supply unit, stacked. This is a preliminary work, in progress. Any serious and competent advices, criticisms and contributions are welcome. sat at