The Project consists in making a modulator/demodulator for AFSK at 1200 bps complying with the Bell 202’s rules.
Satellite’s boards intercommunication
The satellite has to be able of receiving and resending digital frames to Earth’s stations when in view, as well as full control and telemetry data decoding. All this information has to be passed through a modem well capable of modulating and demodulating these signals.
The parts that will be in connection with de modem board will be the CPU, the TX and the RX. The CPU will send all data in RS-232 serial format to be later converted into AX.25 packets. In the same way, this board will send the data received to the CPU in an AX.25 format previously converted to RS-232 levels.
The communication between the modem board and the transceiver (RX and TX) will be a direct one. The data coming from the receiver will be sent to the modem for demodulation. In the same way demodulated frames will go directly to the transmitter to be sent to the Earth.
The AFSK modem receives the AX-25 packet frames from the RX in AFSK modulation. Once demodulated through the FX614 chip, the frames will be passed to an uControler that will translate them into TTL levels, in order to enable the communication between the modem’s board and the satellite’s CPU. At the same time the modem has to be able to receive TTL frames from de CPU and convert them into the AX-25 protocol through the uController. Once again the frames will be passed through the modem where they will be modulating in AFSK the TX carrier that is o be transmitted later.
These frames we are speaking about content the information packets that have to be processed at their destination. These information results in APRS packets info with position, data, telemetry, etc…
The mean modems speed will be 1200 bps, because the modem/chip (FX614) comes so configured. In future projects we’ll study the possibilities of a 9600 bps speed. This speed could be reached through a complex filter system that would use the A/D and D/A input and output ports from the uController. Please take note that we are time and space limited. We are speaking about a Cubesat satellite with a size that cannot surpass 10 x 10 x 10 cm and a mass of 1 Kg.
By the other side, we’ll consider the possibility of inserting a DCD system that would allow us to keep in a memory’s buffer different frames that can be present at the same time by putting them in a queue.
Módem AFSK
After quite a lot of considerations, we have succeeded in finding the right IC (the FX614) complying the required specifications, so it will be the main chip in this essential part of the AFSK modem.
The chip current requirements are quite low and it resists quite well the temperature margins required in space. Further to this, the crystal needed by this IC is easily found on the market.
This IC will allow us detecting and decoding very small amplitude signals quite well. It has to be said that we’ll place some decoupling capacitors in TX and RX to avoid external signals could enter through our pass band. The TX/RX’s tracks will be made as short as possible and designed very carefully with many mass points, in order to refuse interference signals coming outside the pass band.
The external components required by this IC are very few, something that interests quite a lot to us due to the very limited internal space.It needs an external oscillator with a crystal on 3.580 kHz quite common in this type of designs. It also needs a decoupling capacitor in the power supply in order to avoid unnecessary “offsets”.
Here it is not recommended the use of an electrolytic capacitor, because its inherent instability in so extreme ambiances as is space could produce an explosion that will put in danger the whole mission.
Logical control modes
In Zero-power mode we have the “mode sep” that keeps the modem in an state of inactivity. In all other control systems, we have to wait 20 ms for getting a reading coming from the RX or for sending a frame to de TX, in order to stabilize the levels, filters and oscillator.
Internal functioning
The received signal passes through an amplifier to be later filtered. In this way, the smaller signals are well detected. The filter function is to reduce the group’s latency and attenuate out of band frequencies.
The energy detector compares the received signal with a reference level. If during some defined time the signal do not surpasses this level, the detector will place a “1” and will tell us what is happening.
The AFSK demodulator will convert a 1200 Hz AFSK signal into a logical “1” and the 200HZ AFSK signal into a logical “0”.
The AX.25 frame
The final AX.25 frame that carries all the satellite communication will be as follows:
It has to be remarked that all these data will be processed in the uController included in the Modem board. The final objective of the software is deciding what kind of frame has been received on the satellite. And it has to find out the right address which is going to be resent by any telemetry frame.
Other important function is the calculation of the checksum and the validation of all the frames received by means of the microprocessor software. If it is busy, it has to send a waiting signal.
Once validated and checked all this information, it will be sent to the CPU. If it is the CPU who sends some data, its mission will be the encapsulation of data in frames to be sent later by the TX.
Testing board
One of the main objectives of this project is developing a final testing board. It has to be designed carefully because one of the requirements of Sallesat satellite is that the board has to be always functional and so one has to use in it an already well known circuit and well tested technology.
There is a PIC-ENCODER that exists from almost a decade. This modem can receive APRS frames and send them to the Windows Hyperterminal. This is a well tested and functional chip that can be easily improved.
As the processor in this board is a 16f84 (well studied during the present lecture year) it’s quite small for our objectives. This is because its memory is only 1kB. So, the frames received are sometimes been cut. We need a chip with greater capacity.
The 16f87 chip will be the one that gives us a good solution to this. It has 4kB of memory and this should be enough to receive a complete frame without any cuts. This will be the next objective to reach: changing the software in order to adapt it to this new chip.
At the same time, I am pretending to decipher the assembler code that uses the PIC-ENCODER in order to adapt it to our needs because it has been used during many years. This is the most difficult part, but it is not something impossible and I am sure I’ll succeed it.
Another possible IC is the FX614. It has been considered in the project start and it will be the best option for this board.
I have to note you that the image board included does only include the reception part of the PIC-ENCODER and has been adapted to the Sallesat needs, so many jumpers and connections have been omitted. This board is already functional and will be the base of a future one.
Satellite segment
