A Simple 2.4 GHz Signal Source
An essential item for Mode S development is a reliable signal source. One that does not require either a receiver on 2.4 GHz or special test equipment to align. Of course the easiest is to obtain something already tested and working, or a reliable kitset. The G0MRF unit from AMSAT-UK seemed the best answer but at the stage of greatest need it was being redeveloped and would not be available for some time.
Consequently sources were checked for a fail-proof design. What evolved and is presented here proved to be excellent with plenty of signal and the additional advantage of output on most of the VHF/UHF/SHF amateur bands. (Check G0MRF Signal Source for a report on the AMSAT-UK kitset.)
The best idea came up on the AMSAT-BB under the title "$2 S-band signal source." This used the TTL oscillator from an old "486" Mother-Board feeding a diode multiplier to a 2.4 GHz tuned line. Performance is entirely satisfactory. I am using a 24 MHz and a 20 MHz oscillator listening (respectively) to the 100th and 120th harmonic.
Look for the TTL oscillator on an old 486 MB. It will be marked with its frequency, 20, 24, 25, 33, 50 or even 80 MHz. Supply this with 5 volts. Solder a short length of wire on the output terminal and you have your 24000.00 MHz test signal. It may not be exactly on frequency, but it will be close.
Note: The actual oscillator frequency is non-critical, however the higher the better. Ensure the 7805 is bypassed at the IC, any lead lengths can give rise to destructive oscillation.
Most components were recovered from the same 486 MB. The low power 5 volt regulator (78L05) is suitable and low cost.1000 uF feed-through capacitors are available at the Auckland VHF group trading table.
This circuit can be viewed at source:-
A suitable container can be fabricated from PC board or just use an old small tin. The output connector is not necessary, a dipole constructed from a short length of coax pigtailed out is more than adequate. I made one using coax and a short length of brass tube. Pick a tube that will slide neatly over the coax once the braid and outer insulation are removed.
RFC is 3.5 turns on a ferrite bead. Almost any small signal silicon diode can be substituted for the 1N914 - a 1N4148 gives plenty of output. Spacing L3-L4 and L4-L5 is about 3 mm.
Testing. A strong signal should be heard on harmonics of the TTL oscillator frequency Replacing the shorting link with a milliammeter should give about .5 milliamps. If you have a 2.4GHz receiver adjust C3 (4BA screw) for maximum signal.
NOTE: This adjustment may be very broad and is unnecessary for output - its just a nice touch! Good earthing of the lid to the case is essential for peaking C3. For most folk, only the 7805 will need to be bought. Its likely that many will have even this in there spares' draw so cost should be minimal.
An even simpler circuit soon appeared. The diode multiplier is unnecessary. The TTL oscillator alone into a short bit of wire gives a good signal on 2.4 GHz. If you have a good 5 volt supply, even the regulator can be eliminated. For slightly improved results using this simple idea try making a quarter wave 2.4 GHz whip for the output or even a half-wave dipole. A 9 volt battery permits the signal source to be moved a distance away for antenna checking. This circuit is ideal for checking other bands and frequencies.
Odd and even harmonics of the oscillator will be heard above its fundamental. Just multiply the marked frequency to find its harmonic. Strong signals should be heard well beyond 2.4 GHz. If you find one of these TTL oscillators on 24 MHz you are in luck. This gives markers at 144, 432, 1296, 2400, and in the 9cm and 5cm bands. The above-described units are in constant use for antenna comparison & tuning, and as weak signal sources for checking receive performance. While the output is around -105 dBm, that will produce a signal 35 dB above the noise in a reasonably good converter. If the output from any of these sources cannot be heard with both the source and converter on the bench a short distance apart with just a few cm of wire in the antenna then the receive system will not hear AO-40 without a huge antenna - perhaps a 20 metre dish!
"Check out NZART's Break In magazine of November/December 2001 for a comprehensive article on 2.4 GHz signal sources."