I will be running under special callsign 233RC103/QRP whenever the propagation allows it, with only 3W output into either vertical 1/4 GP antenna or testing the new Delta Loop. Working frequencies 27.555 or 27.610 MHz USB.
A transciever is basically a box with alot of nice buttons and displays that transforms your voice in radio waves, and the radio waves back to sounds you can hear. More or less, this is done trough a complex process that includes many stages that take care of filtering, mixing, switching, amplifying, detection, modulating or demodulating and so on. Every one of these stages had it’s own role and required a certain complexity in order to achieve the required performance, and generally the better performance you want, the more parts you need and the more money you’ll have to pay.
But, we ARE in the 21st century, everybody knows what a computer is and probably has at least one in their house, and also knows what … magic a computer can do with the right software. So, why not implement all these functions required by a transciever into a piece of software running on a microprocessor, and let it filter, mix, switch, amplify, detect, modulate, demodulate or whatnot ? You would just need to find a way to interface the signal between the antenna and the software and you should be set. The advantages coming from using software instead of hardware would be many:
– hardware complexity reduced to a simple interface for basic functionality;
– performance of software filters, mixers, modulators etc is better and simpler to implement than that of hardware solutions;
– limitless functions, options, types of modulation etc, it only depends of the software used;
– posibility to monitor a wide frequency band (limited only by interface), not just one frequency like in classic trancievers;
Don’t let me make it sound like it’s my idea though because it’s not, it has been running around since computers were getting close to enough power to attempt this, back in the 80’s, and the first working models of such transcievers were used for military communications in the 90’s. How would a SDR transciever work exactly ? Go to the next page to find out.
After a long break of 2 years, i’m back on air as 233RC103, this time making an ambition of using only home-made equipment. And as we are in the 21st century, I have decided classic transciever design is outdated and the future is Software Defined Radio, so I went ahead and started building a SDR transciever, complemented with a home-made external power amplifier and antenna. As I am still testing and developing the new equipment and taking care of problems that obviously come up when constructing such a thing, but I have managed to make a few nice contacts these days: 3AT213, 91RI105, 113RC/DX, 119AT014, 309EX303 are just a few of the many.
I have developed the protection circuit for the EB104 amplifier I am working on, after I finally had some time to design and test a few models. The main requirements have been:
– protection in case of high temperature;
– protection in case of high SWR;
– protection in case of wrong output filter selection;
– simple design (i’m a fan of the whole K.I.S.S. rule of thought), able to work in strong electromagnetic fields, reliable, inexpensive.
Because i will be using the same directional coupler i have used in the SWR meter (the one made on PCB) wich is directly influenced by the signal frequency, and because i want full HF coverage, i cannot just measure the reflected signal and make a circuit cut the amplifier when it goes over a limit; on 28Mhz the coupler generates roughly 4 times more voltage that let’s say in 7Mhz. So a system that compares direct and reflected signal and triggers when the latter is percentually too high was needed, therefore an operational amplifier was the natural choice. This will solve the SWR problem, and because the directional coupler will sit between the amplifier output and the low-pass filters, it will also trigger when a wrong band is selected.
The thermal problem will be even more easy to fix, i will use a NTC thermistor in a resistive divider; the second operational amplifier from the LM358 IC I have chosen to use will just compare the voltage from the resistive divider to a preset one, and will trigger when the thermistor’s value gets too small.
As usual, it’s much easier to understand when pictured:
The 5V stabilizer LM7805 is there to ensure good separation, and the BC107 transistor is used to increase current capabilities on the output. Parts list:
C1 – 100uF / 16V
C2 – 470uF / 6.3V
C3 … C10 – 10nF ceramic
R1, R4 – 470 ohm
R2 – 1 Kohm
R3 – 10 Kohm
VR1, VR2 – 10 Kohm
D1 … D4 – 1N4148
The circuit will trigger once one of the two described conditions will take place, and will remain like this until power is removed for 10 seconds, due to D1 or D4 diodes. A LED connected between LED1 + and LED1 – points will signal SWR protection enabled, and a LED connected between LED2+ and LED2 – will signal thermal protection enabled. TH+ and TH- will be connected to a 1Kohm NTC thermistor wich will be placed on the heatsink, as close to the amplifier’s power transistors as possible, and the FWD and REF points will be connected to a SWR sensing board like the one described in the SWR meter article. For reliable operation, low-pass filters on both FWD and REF lines might be needed, made from a series 1Kohm resistor and a 10 uF / 16V capacitor to ground.
There are many ways in wich the amplifier might be stopped from working once these protections trigger. Switching back the RX/TX relay while in full operation might be dangerous (for a second both transmitter and amplifier will work without a load) plus the relay might be damaged. The simple way is to cut down the power of the amplifier by removing the gate bias, by simply connecting the BC107’s collector (BIAS point) to the bias voltage regulator’s reference circuit (pin 5 of MC1723CP in the Eb104 schematic). This will still allow you to remain on the air, the transmitter will see the correct impedance on the amplifier’s input and the amplifier’s finals will be able to handle even infinite SWR and the heatsink will get the chance to cool down due to running in low power mode. SSB or AM work will be a problem, because the amplifier will work in C class now.
This has been tested with 100W on both antenna and dummy load, it’s working OK, the real test will be when the rest is finished though.
Since yesterday I am a member of the Romeo Charlie international DX Group, with the callsign 233RC103; you can find more information about Romeo Charlie by clicking this link.
From now on 233SP129 will be retired.
Meet you on the air ! 73s.
After we finally got some propagation back for a few days, tonight (14.12.2009) i got the chance to have a very interesting contact. Interesting because it was done at 02:00 local time and the 800Km distance between Bucharest and Zagreb is too big for direct wave and rather small for F-layer reflexion. Plus, 328MIC001 comes here barely over the noise constantly for about 3 hours now, so it’s not a sporadic event.
Signals are very weak on both of our ends (R 2-4, S 0-1), but cluster.dk helped us to get in contact.
73s and all the best Rocky !
A QSO I had on 31.10.2009 with Rudi – 91RSN300 from Java Island, Indonesia, about 10.000Km away.