Circuitry

Oscillator

The original beacon used the internal oscillator in a G4DDK004 source with a 60C ovened crystal. This proved to have inadequate frequency stability. Experience with the G8ACE MK II temperature controlled oscillator  in my 24GHz transverter suggested that this would improve stability. The big problem was getting the extra volume of  the oscillator box into the beacon box. By judicious squeezing  and chassis reorganisation this was finally achieved.  The G8ACE oscillator employs a Dutch made 108.00396MHz crystal cut for 60 Degrees Centigrade. The oscillator is calibrated to put the oven temperature at the turning point of its characteristic.  This then drives a G4DDK designed DDK004 source, used as a multiplier which produces around 5mW at 2592.095MHz. The total multiplication is times 96.  The G8ACE oscillator has on board tuning voltage facilities and this is used to provide FSK. The oscillator is contained in a tinplate box with lid.

Multiplier

The next stage is a Surplus Qualcom 2.5 to 13GHz multiplier retuned to 10GHz output. The 10V and -5V supplies are obtained from the PSU board for the Amplifier. It is mounted on a 5mm thick aluminum plate for support, mainly of the input/output sma connectors.

Amplifier

This is a converted Qualcom Omnitrack Amplifier, originally tuned for 14GHz output but retuned using the W1RIL notes to 10GHz output. It produces 1W output. A compact PSU board designed by WW2R is mounted on the back of the heatsink. The PA alone  takes 1.2A from the 12V supply and gets very warm in continues operation, hence the very large heatsink mounted on the box. A lot of time was spent trying to get the heat away from the Amplifier when building the beacon to improve reliability, whilst trying to prevent water getting into the box from that route!

Keyer

This uses the WW2R intelligent keyer which has the following sequence:-

NT5NT/B EM12PX <mark>

NT5NT/B <mark>

NT5NT/B Richardson TX<mark>

NT5NT/B <mark>

This is followed by the following variable sequence of four 3 digit Morse numbers which indicate the state of the beacon

1. Output power.

 For the mathematically minded the formula is:-

Output power (dBm) = -0.0006*count² + 0.1953* count + 14.253

for the rest of us 150 is equal to 1W into 50 ohms at the output of the Amplifier.

2. Power supply voltage,

This is measured using a potential divider (maximum input to A/D is 5.0V!) measured at input to PA regulator board

PSU Voltage = Channel 2 count/10 (i.e. 138 means 13.8V)

3. PA Chassis temperature

This uses an LM34DZ chip in thermal contact with the Qualcom amplifier chassis, mounted as shown in the picture

PA temp (F) = Channel 3 count * 2 (i.e. 50 means 100F)

4 Oscillator temperature

This uses an LM34DZ chip in thermal contact with the heater plate in the G8ACE oscillator. Its 5V supply is generated by a separate 78LO5 regulator mounted in the oscillator box.

Oscillator temp (F) = Channel 4 count * 2 (i.e. 50 means 100F)

Output Spectrum

The wideband spectrum is very clean. However the close in spectrum has a couple of signals either side of the main carrier at -65, -33, +33 and +65kHz from the main carrier, as the following Spectrum Analyser plot shows. They were found to be generated  by the negative bias chips in the beacon. Much time was spent trying to filter out these signals but to no avail, however provide a secondary receive sensitivity test, as they are all 70dB down from the main carrier (~-40dBm)

How does that all sound?

Listen for yourself!  An .mp3 audio file of the beacon as recorded at WW2R, 12 miles North of the beacon, on a USB receiver, using open waveguide is an antenna is available for download nt5nt_10g.mp3

However for something more exciting, another recording made on Dec 30 2002 using an 18" dish pointing at southern OK ( i.e. away from the beacon) illustrating what rain scatter (or was it Hail scatter?) signals sound like is available for download as an .mp3 file  rainynt.mp3

Antenna

The antenna, donated by WA5VJB is a wide-sector horn, originally designed for the ALQ-117 airborne ECM system by GTE Sylvania, Mountain View, CA .  It provides at least 180 degrees of coverage (probably 200 deg) and has a four-hole WR-90 flange on it.  It is fully sealed from the weather. (thanks info K5VH)
 

Power

The beacon runs off 12-15V. Power is connected through the N type connector (mounted on the bottom of the box in normal mounting position). 25' of FSJ4-50 Andrews superflex cable is used supply the 13V 1.5A to the unit. It was chosen for its robustness, waterproofness and availability. The whole unit is mounted to the 2" mast with a pair of 2" muffler clamps. The box is rated waterproof to IP45, but copious amounts of aluminum loaded sealant were added for additional waterproofing of waveguide interfaces etc, as the photos show.

In  Aug 2007 the beacon went off the air. The power supply was providing 14V to the power cable but it was found not to be getting to the Beacon. Investigation revealed that the centre pin of the top (exposed) heliax connector had been corroded into a pile of white power, hence no connectivity! Considering the same cable has been exposed to the elements on a roof 17 stories up, I am surpised it lasted the 6 years it did! Looking inside the beacon enclosure revealed no signs of water ingress there.

To the Nortel Amateur radio club for hosting the beacon and to WA5VJB for donating the antenna.

Reception reports

These should be sent to WW2R at the email address at qrz.com. Reports received are recorded in the following table to give an indication of possible coverage to others.

 

Call	Grid	Location	Date	Distance (km)
K5VH	EM00XE	Dripping Springs, TX	04/08/03	335
W5LUA/R	EM24TQ	Mena, AR	17/07/03	289
WW2R/R	EM03XD	Jacksborough, TX	30/11/03	126