I found my old article recently regarding my home-brew Power Over Ethernet Project. James and will shortly be putting internet at the Scout Hut around one mile away and having my router as close to the antennas as possible will be mandatory. Here’s an old project that I wrote up in 2009:
Using very high quality coax from my vertical antenna to a router in the shack is a very expensive option due to the very high losses at 2.4 GHz. We could use Ecoflex 15, but at £6 per meter and £9 for a connector, it’s a bit rich for a kids experiement.
Last year, I became inquisitive and managed to take apart one of my Andrew Grid Dishes. They’re rather well made so it takes some patience and some sharp Stanley blades – along with an excellent first aid box! I nearly cut my finger off. Not the first time. My original intention was to replace the feeder directly with heliax to trim down the losses but looking at how they assemble these in the factory, I needn’t have worried.
You will see from the pictures (see link below) that these dishes are well made. I wonder what the goo that completely blocks up the tube that supports the dipole centre? It’s clearly some sort of water seal and I must say it’s completely over the top. It took a heck of a struggle to get the coax out.
These dishes are shipped in two halves along with the coaxial dipole mounting centre that bolts to the grid from behind. It’s possible to get the polarisation incorrect from the beginning. Notice in the picture that this one is vertically polarised. To change polarisation, you DO NOT change the plane of the dipole feepoint, instead you must rotate the dish through 90 degrees. Note that all fasteners are imperial dimensions, not metric. Don’t start losing nuts and bolts if you live in a metric country.
With a 23dbi gain and an 8 degree lobe of RF pointing in the appropriate direction, this is a serious bit of kit but please use low-loss coax; LMR-400, Westflex or if you must, RG213. Don’t use RG58, this is actually very HIGH-loss coax and shouldn’t be sold as low-loss at all. The losses at 2.4GHz are tremendous and you could easily see most of your signal wasted in the coax. The dish is terminated in an N-Type connector on a flying lead with RG-213 type coax. Point-to-point, you probably can’t do better unless you start to spend astonishing amounts of cash.
James and I will try and give a pair of these a test in the summer from Monument hill to the Scout hut – about 10 miles. Apparently a breeze. The world record is for an un-amplified signal is something like 150 miles!
I’ve had loads of fun tonight putting together the Andrew Grid antennas (well one of them) and duly installing it on an old brass curtain rail in the shack, pointing towards the kids PC in the “little lounge” (as we call it). The signal shot through the roof when I got the lobe on the sweet spot, bearing in mind that the lobe is only a few degrees wide, so get it wrong and you don’t see a signal at all! This is good news though and proves that my 2.4Ghz antenna designs were utterly rubbish
I tell you, it’s even bigger than this picture! Like all good antennas, they are bloody huge if you muck about with them inside a shack. Clearly designed for a roof on a rotator (in my view!!), I have decided that it would be a waste to put a wireless router on the roof, instead I am going to see if I can bring some Andrews Heliax down the wall and straight into the shack. Perhaps a kind amateur may help with a 20 meter run of scrap heliax? That way, I can play wireless networking, general purpose radio and anything else that tickles me!
The only thing that worries me is the polarisation. Should I mount this vertical or horizontal? I was going to go horizontal for a point-to-point wireless bridge but now I’m not so sure. I’ll have to mull that one over – I have a feeling that horizontal polarisation works better over long distances, we’ll see.
One problem though, the guy that sold these to me seems to have “lost” one of the antennas. He thinks he shipped 5 grid antennas according to his letter. However, I received 4. Something odd has happened. [Later] He’s emailed me saying that he’ll investigate on Monday because he didn’t pack them himself. [Later again] He only sent the four and now the fifth is missing so he’s sending a cheque for £10 as compensation.
Not being able to resist the pull of the decibals, I raided the larder tonight for a Pringle tin to build a wave-guide antenna from scrap parts. This is the story of that project.
Manufacture: I soldered a 30.5mm (quarter wave) element to a UHF bulk-head connector and drilled out a hole in the Pringle for hot-glueing. Exactly how far away from the base of the tin I should fit the element took ages – and lots of conflicting web pages. In the end, I aimed for 1/4 wave from the back of the tin. Someone is going to tell me that this is probably the most awful place to stick it – I can believe you
My multimeter couldn’t get a reading on the foil inside the can so I used kid’s water based glue to stick sheets of tin-foil to the outside. This tinfoil was grounded to the base of the tin, although I had some difficuty in doing this because there appears to be some sort of laquor applied to the base. The copper wires seen in the photo were to act as a physical and an electrical assistant; to provide a good ground to the outside of the can.
Ater connecting the Pringle-Wave-Guide to a buffalo wireless access point, I turned it in the general area of the kids room that was running NetStumbler on their network card. I waggled the can around in my shack, and went to the kids room to check if the signal strength had risen at all during the waggling. No luck. So far, it’s as much use as a dummy load.
It just goes to show that some experiments just fail.
I needed to get an antenna higher for the kids computer room but a quarter wave vertical didn’t seem to have the gain I wanted, in fact from the kids room I couldn’t even see the network. I hunted for a Pringle can to make a “Can-Tenna” (see http://www.m0mcx.co.uk/?p=17) but apparently we were out of stock(!). A small yagi was the answer:
I made the driven element directly out of the lossy RG58 coax so that the centre conductor became one side of the driven element and the braid became the other. I used solder to give the wires some strength.
A good test but don’t be fooled. The standard 5/8th antenna shipped with most routers are probably just as good. This yagi needs to be a 5 or maybe 8 element to work better.
I also built a quarter wave last night directly from Westflex W-103, no other components – and although it compared well to the shipped vertical, it only started to match the retail antenna when I gave it some height. Another good (but failed!) experiment!
I have just bought some MC to N-type pigtails for the wireless project but now I’m trying to find equipment enclosures to house the router up the mast right next to the grid antenna. Can I find one? Nope. There’s a company paying good money to Google for first place in the rankings for “non metallic enclosure” but the site is so bad(!) and the pages are even named, “NewPage 1″ and stuff like that. The MD’s son’s summer project by the look of it. Need to get the pros in!
After much searching, I found a few sources:
I have a first class project in tow to put a 2.4GHz link in to the club about 4.5 miles away. I’ve just secured 5 grid dishes with 24dbi of gain for £125. What is a Fresnel zone and how do I spell it?
Anyway, there’s this great site to calculate the potential likelihood of succeeding in putting a link up between one site and another:
It also gives you the dreaded Fresnel Ellipse data which I think I’m slowly starting to understand – essentially, you can block around 40% of the ellipse in one place and the link should still work, but no more. Google it for more.