Experiments on 6cm
Bernard Nock G4BXD takes his first tentative steps onto the 6cm band.
Having sorted out my receiving and transmitting operations on the 23, 13, 9 and 3cm bands my attention was directed at the 6cm amateur allocation, basically 5.7GHz. There are three allocations in the 6cm band but the main narrowband activity is at 5760MHz. While there is a lot of equipment available for both the 23cm (1296MHz) and the 13cm (2.3GHz) bands there seems slightly less for the 9cm (3.4GHz) and 3cm (10GHz) sections.
There seem to be even fewer commercial amateur units for the 6cm band although it is very popular with many entrants during the RSGB UKAC contests and the UK Microwave Groups events. The usual supplier in Germany, Kuhne Electronics, does make a suitable transverter although it is quite costly, I feel. Certainly too costly for the amount of use I thought I would be getting out of the band but probably ideal for those interested in EME (earth-moon-earth or moonbounce) communications. While I am a great fan of ‘plug and play’ I felt I was getting a bit more adventurous regarding building stuff from scratch or at least smaller units.
What is Available?
This led me to consider just what I either already had in my component boxes, or was available from China via eBay or other suppliers. The amateur section is close to the 5.8GHz Wi-Fi bands so some of that equipment can be repurposed. There are many little amplifiers, oscillators and the like coming out of China but unfortunately the quality or indeed specifications leave a lot to be desired.
I recently bought some quoted 1W amplifiers, rated to 2.4GHz, from a Chinese eBay seller only to find I get about 200mW out of them at 2.3GHz. Many others have found similar results, Chinese Watts being a different standard to Western Watts apparently. It seems the Chinese are not averse to simply copying the silicon chips, printing them with the names of reliable manufactures and passing them off as genuine.
So, just what would I need to make a 5.7GHz receiver to start with? I found I had a rather nice double-balanced mixer marked up at 2.5 to 10GHz and, being US made, was probably reliable. This would need a local oscillator (LO) and some RF amplification to produce an output at the IF frequency chosen.
I have several of the very useful ADF4351 PLL oscillator boards, Chinese of course, to which I have added extra decoupling on the various voltage regulators to reduce the phase noise generated. The common ADF4351 unit only goes up to about 4400MHz and if I were choosing a 70cm (432MHz) IF, say, I would need a 5328MHz local oscillator. The ADF5355 chip does go up to 6GHz but those boards, even from China, are quite costly.
I decided to test the theory using the local oscillator set to 4000MHz, which would give me an IF of 1760MHz, meaning I could use my RSP2 software defined radio to act as the IF receiver. I now needed some RF amplification at 5.7GHz to feed into the mixer. I have to admit I am attempting this basically at random. I am not taking into account LO levels, how many dBm the mixer needs, how much RF gain I need or such. It really is plug and play.
Again, there are various little amplifier boards offered on eBay coming out of China. I have bought several over the past few months but none were suitable at 6GHz. If you enter Microwave Amplifier into the eBay search line, you’ll see many proper, professional type amplifiers, usually of limited bandwidth and upper limits. I managed by luck to drop on a UK seller offering a 6GHz unit, which I bought, and he offered me a second one, which was very handy.
These little units had about 13dB gain and were designed for the 6GHz band so were ideal as the front-end of the receiver lash-up. I arranged the two amplifiers connected with barrel connectors and then to the RF input of the double-balanced mixer. The ADF board was connected to the LO input at the same time.
I used one of the little Chinese made amplifier boards on the output of the mixer unit, quoted as 50 to 4000MHz and using an SPF5189 chip, the spec sheet of which says it has 18.7dB at 900MHz. I was only using the device at 1760MHz so it should work fine.
In Practice
So, Fig. 1, I had some RF amplification at 5.760GHz, a mixer, a local oscillator on 4GHz and some IF amplification at 1760MHz. I now needed an antenna of some sort. Obviously in a finished state the goal would be to use a parabolic dish with some sort of feed, either a horn or dipole type. I remembered I had a PCB-mounted log periodic antenna (LPA) that covered 2 to 11GHz. Basically, it acts as a two-element beam at this frequency range, low gain but worth a try I thought.
Luckily, I have my office workroom window facing east, directly towards the South Birmingham microwave beacons, 21km away, on 6 and 9cm. I get the 9cm beacon 5 and 7 with an indoor antenna. I balanced the breadboarded setup on some books in the window with the LPA pointing in the right direction, set the RSP2 to 1760.900MHz and switched on my project. Imagine my surprise when up popped the 6cm beacon keying away, Fig. 4, only slightly off where it was expected.
I could swing the board with LPA attached left and right and could peak the signal from the beacon. It was weak but fully readable. Interestingly, during my 9cm
experimentation I could receive the South Birmingham 9cm beacon on the same heading and see the Telford 9cm beacon, 10kHz higher or so, obviously by reflection as its direct path is through the house and behind me.
I couldn’t see any sign of the Telford 6cm signal during the first tests. Later in the day though I did see the Telford signal, receiving it via rain scatter of storm clouds southeast of my location. Receiving the weak signal led me to think about what I could do to improve the antenna – the RF amplifiers can only amplify what you feed into them.
Remembering how simple it was and how well it worked, I decided to build a double bi-quad antenna, Fig. 2, using PCB and 1mm wire. The antenna is easy. The back plate is 11 x 5.6cm with a 1.3cm skirt each side. The element, 1mm wire, is 1.3cm on each side mounted 0.5cm off the reflector. I used an SMA connector with the element soldered directly to it.
A short time later a very nice little antenna had been constructed and was connected to the front of the setup, Fig. 3, pointed at
the beacon. On switch-on there was indeed a good amount of extra gain, Fig. 5, off the bi-quad proving it was a little better than the wideband LPA used before.
The truth is though the real game changer would be a dish antenna and luckily I had a small ex-satellite TV dish, which was light and small enough to be tested in the confines of the office workroom. With the breadboard clamped to a tripod and with the bi-quad facing away from the beacon, I attempted to place the dish in the right place, distance and angle-wise to reflect the beacon signals into the bi-quad.
The results were very encouraging, Fig. 6, with a large increase in the signal received, both the Birmingham and Telford beacons coming in loud and clear. In fact, while experimenting I was discussing the results online and a fellow, more experienced microwaver, Graham G3VKV, invited me to listen for his signals, some 57km away to the south of me.
With Graham transmitting dots from his station I was just about able to see the dots on the SDR waterfall though of course very weakly. This is very encouraging and rather pleasing to think how easy it was to assemble these units into a near working receiver. Of course, carrying a breadboard around clamped to a tripod is hardly convenient but it has shown me that I am on the right track.
With this in mind I have ordered a purposebuilt local oscillator from a German supplier,
Dieter DF9NP, which will give me a stable 5328MHz signal with enough power to split to supply the receiver and transmitter mixers. A flat panel 5.8GHz Wi-Fi 24dBi antenna is on order, which will make testing a lot easier than trying to hand balance a dish. The remaining problems will include amplifying the transmit RF to a level to make it usable on the air.
One solution for the transmit RF side might be to use one of the FPV (first person video) amplifiers sold to boost the 5.8GHz signal used with Drones and the like. Various units are around apparently delivering 2 to 4W, probably Chinese watts though. Hopefully I can bring you the tale of the transmit side of things in the near future along with news of my first proper QSO.
Of course, these are very simple experiments. Those more experienced in the black arts will no doubt be laughing all the way to their re-flow stations and CNC lathes but as a relative newcomer to these frequencies I think it shows that even with a little knowledge you can get things to work and, as I seem to be told often, it’s the fun of the hobby. Enjoy.