Back in the 2013 I was visiting my friend and near neighbour Briain Wilson, GM8PKL, who has a very good view from his from living-room and patio to the north over the City of Edinburgh and beyond. He told me that on very clear winter days he could see snow covered peaks which he thought were in the Cairngorms which are at a distance of around 140km away. A few weeks later I was visiting Chris Tran GM3WOJ (a well known HF DXer/contester) who lives near Tain in Ross and Cormarty and he happened to remark that from his patio he could see Ben Macdui on a clear day. I began to realise that these summits were probably LoS from both ends of the path between my home location in Edinburgh (IO85JV)and Chris’s location (IO77WS). I suggested that it would be interesting to try a 3cm microwave contact scattering signals over the peaks between the two locations. In the back of my mind was the thought that both locations could be done with more than the usual comfort of portable operation, and as we could probably run about 20 watts at each end of the path it should work. A bit of path profiling confirmed that in fact both ends were in fact LoS to these peaks and it was likely that a signal could be scattered over them on 10GHz.

Figure 1. Path between GM3WOJ and Edinburgh

Figure 2. Path profile from Edinburgh to the summit of Ben Macdui.

Figure 3. Path profile from Ben Macdui to GM3WOJ.

Figure 4: Direct path profile from my QTH to GM3WOJ.

At this point Chris suggested that we could do one better and he would be very happy to host a 3cm personal beacon on either his house or one of his towers provided I built it and set it up. He was happy to keep an eye on it and mind it. I jumped at this and said I was very happy to provide the necessary hardware.

Over the winter of 2013/14 I assembled the necessary bits and pieces. The exciter I used one of Andy JNT’s LMX2541 frac-N synthesiser PCBs programmed by a PIC microcontroller to do the keying as per Andy’s recipe. The JT4G keying is timed from GPS. The frequency reference is a well-aged 10 MHz OCXO. It was felt that the extra complications of GPS frequency locking were probably not worth the effort from experience of running the 23 cm beacon GB3EDN

This generated a signal on 3456 MHz which is tripled in frequency (With a bit of recycled Sat TV PCB) to drive a modified QUALCOMM Omnitrak PA (courtesy of Chris Towns, G(M)8BKE). The output is a few hundreds of mW which is fed to a small 35cm BSB dish (remember those?). This gives a beam width of around 7 degrees which from IO77WS gives good coverage of a lot of the UK southwards. The all important housing is from a surplus DMC 22 GHz link unit which had already yielded lots of useful stuff for 24GHz.

Figure 5: Block Diagram of hardware.

Figure 6: ODU hardware.

After an initial trip to Chris,GM3WOJ’s in mid May to reconnoitre and get some measurements to fix the beacon to one of his smaller towers ( at 11m AGL) I got all the hardware together and tested to my satisfaction. This tower was chosen as it had a clear view south and was not obstructed by trees. It all had to be pretty rugged and weather-proof as I did not want to be driving up and down the A9 to effect repairs. In mid June I returned and between us we installed the beacon. Remarkably all went to plan and we had it on the air within that day. Monitoring the signal locally all sounded good but it remained to be seen if it would get past the mountains to the south which incidentally are the highest in the UK second only to Ben Nevis. An interesting problem was ensuring that the dish was actually pointing in the correct direction when the tower was elevated as there was no received signal to pan it on.

Figure 7.The beacon hardware with builder.

Figure 8: The beacon installed on tower and the take-off south over the Black Isle.

The next day I set off south down the A9 heading home to Edinburgh. I had a Horn antenna fixed to the car as all the best microwave geeks have and monitored it going south. The signal was good on backscatter from the Black Isle but largely disappeared in the vicinity of Inverness. Further down the A9 just before Aviemore where the Cairngorms are in good view from the road it returned and at reasonable strength which was reassuring. I drove up to the Ski slope at Cairngorm and it was not audible LoS as the area is shielded by the direction of the glen. Pointing the antenna at any of the peaks in the vicinity however gave a reassuringly good signal. After that I headed south back on the A9 home with no more signals. It worked with back-scatter but it remained to be seen if forward scatter would work. The next day I was keen to see if it was audible in Edinburgh and I pointed a small (40cm) sat TV dish with a modified LNB as a pre-amp out of my shack window to the north. After a bit of waving it about in the right general direction I was amazed to find the signal was clearly audible at 216 km with an 1800 m obstructed path. The signal was always audible after fixing the antenna to a bracket on the roof.

Figure 9: The 40cm receive dish on my roof in Edinburgh (IO85JV). Note the blue sky! It blew away a couple of weeks later.

This arrangement worked well until the winter but in January gales the dish was ripped off the roof and landed in the back garden complete with LNB. Very kindly Briain Wilson GM8PKL offered to host the receiver at his QTH which is LoS to some of the grampian peaks. Figures 10 and 11 show the new dish installation. It uses a re-purposed Sky Dish and a modified LNB with an external LO feed. The LO is locked to the 10 MHz output of a Morion DOCXO and comes out on a 432 MHz IF. The receiver is a FT790R2. The audio from this receiver is streamed onto the internet and is at and click on “M3U” if it does not open automatically.

The stream is generated on a Raspberry Pi computer using Darkice and Icecast is the web-server` programme. These programmes are open-source and intended for internet “radio” stations.

Figure 9: The “gorse green” dish at GM8PKL's QTH.

Figure 10: Briain, GM8PKL panning the dish.

Viewing the Streaming Audio with Spectran.

The streaming audio can be monitored with Spectran on your PC. To do this you need to install a “virtual audio cable” program to pipe the audio directly from your media player programme output to Spectran. I use the standard Windows media player programme with VB Cable for this purpose. VB Cable is available here at no cost, . Instructions for setting it up are also provided at that URL. To setup Spectran click Setup and then Select soundcard  as shown below. Spectran should see the cable input and output as sound card ports.

Figure 11

Select Cable Output for the Spectran input sound card and Speakers High Definition Audio as shown and click OK. Start Spectran and you should see the signal on the waterfall display. The plot will lag behind the actual signal because of latency on the internet making it difficult to decode with WSJT software (but it can be done!)