Wi-Fi Cables
I have been really struggling with my Wi-Fi antenna system, mostly the cables leading down from the antenna. I may be an electronics engineer, but I am not really an RF (Radio Frequency) engineer so I am making lots of mistakes along the way.
You may recall that I am using an Omni and a Yagi antenna combined with a splitter. The real difficulty is with the cables. The whole thing is mechanically based on a Weingard TV antenna mast system that allows me to manually raise and point the antennas, and also lay them down flat for travel.
My original thought was to bring down two separate antenna wires and use a switch in the living quarters to select between them. I then decided for various logistical reasons to mount both antennas on the same mast. That allowed me to place the splitter (combiner) on the antenna mast outside and then run a single cable through the roof.
The basic trade off with antenna cable (Co-Ax) is between flexibility and low loss. Large diameter stiff cables can have low loss, and therefore run relatively long distances, but they don’t bend very well. Even if they can be bent to a reasonably small radius, they are not intended to be bent and straightened and bent and straightened etc. Small diameter cables can be very flexible, but they are also very high loss, so you don’t want to run them very far. Not all cable types are meant to be outdoors in the weather either.
For example, I started off assuming that I would use LMR-400 cable as recommended by the web site that I was using,
http://www.radiolabs.com/index.php. LMR 400 is good cable, so it’s not Radiolabs fault that it won’t work for me. It would be perfect if you wanted to put a Wi-Fi antenna up on your roof and just leave it there. It’s 0.405” OD and looses 6.6db per 100 ft at 2.4Ghz. I am still using 2 each 2 ft. LMR 400 cables from my two antennas down to the splitter (combiner). That chunk of cable doesn’t need to bend and the 3% power loss is negligible.
The cable from the splitter (combiner) that goes down through the roof of my living quarters does need to bend, and repeatedly so. At first I was thinking to use LMR 100, which is very thin and flexible. It is 0.11” OD but it looses 15.8db per 100 ft. at 450Mhz. It isn’t even rated at 2.4Ghz, where presumably the rate of loss is even higher. Despite this, LMR 100 is often used for very short Wi-Fi cables like 1 ft. or less from a laptop module to a clip on antenna. I am planning to place the transmitter/receiver module just under the base of the antenna, so the cable only needs to go a couple of feet, but LMR 100 isn’t recommended for outdoor use.
Connectors are another issue to consider. The connector at the transmitter/receiver module end of the cable is an SMA type, which means it has an outer ferrule of about 0.315” OD, with an internal thread and a central socket. The RP-SMA connector on the transmitter/receiver module has a matching external thread and a center pin. This sort of small threaded connector is often used for small laptop type modules with short little antennas that plug directly into the module. There are many other small and medium size RF connectors and I am not qualified to tell you about all of them.
The connectors most often used for large outdoor Wi-Fi antennas are type-N, with a type-N Female at the antenna and a type-N male on the mating cable connector. A type-N female is a large, robust, industry standard connector with external threads and a center socket. The type N Male has an outer ferrule with internal threads and a center pin. The OD on the outer ferrule is about ¾”, so you would need a rather large hole in the roof to feed one of those through. In my case the splitter (combiner) also has type-N female connectors.
I eventually settled on LMR 240 Ultra as the cable of choice for the two feet of cable that has to flex and enter through the ceiling of my living space. It is 0.24” OD with a Polyethylene jacket that is suitable for outdoor use. The inner conductor is stranded (not solid) which is why it is “ultra” flexible. It is rated at 12.7 db loss per 100 ft. at 2.4Ghz, which means I will loose just under 6% of my signal strength in 2 ft.
There are probably stores or web sites out there that will make this specific cable to order, but I will just be making it myself. I do not recommend this unless you have good soldering skills, and the proper equipment. The SMA connector requires a special crimping tool and also requires that you solder a very small pin. The type N connector is much larger and easier to solder but is still not trivial.
You can easily buy the necessary connectors from on line sources such as DigiKey
http://www.digikey.com/, but you need to be very careful to use cables that are made for the specific cable type that you are using, otherwise the pieces just won’t fit. Another source that I have used to buy the LMR 240 Ultra (among other things) is
http://www.universal-radio.com/.
Yes, RF cables can get very complicated. Actually, it’s much more complicated than I can explain here. Suffice it to say that it is much more like plumbing that wiring. Also, be warned, Wi-Fi uses 50 ohm cable, while broadcast TV uses 75 ohm cable. If you don’t know what that means, and not many people do, then don’t worry about it, but just don’t mix the two. If you do, then it probably won’t blow anything up, but it won’t work properly either.
I promised to tell you about all of my mistakes so here it is… I have thus far managed to waste about $75 on cables I can’t use. Learning can be expensive.
Beyond cables, I have many other concerns about my Wi-Fi design. Combining the signal from two antennas can produce unexpected results if you aren’t careful. If the two antennas receive the same signal out of phase with one another then they can cancel each other out. What I am hoping to achieve is to be able to receive a pretty strong signal from any direction, and a really strong signal in the direction that the Yagi antenna is pointed. That means my Omni and Yagi antenna need to add their signals together in that direction, not subtract. At 2.4Ghz the wavelength is only about 5” so the relative position of the two antennas is critical to a ½” or so. I understand the theory (at some level) and I won’t try to explain it here, but I have no way to know for sure if I have achieved my goal. A real RF engineer would need an anechoic chamber and other expensive equipment to know for sure, so I will just have to hope for the best.
I also have concerns about the long term durability of this setup. I will seal up all the connections with “Coax Seal” which is a tape like, tar like, sticky material commonly used to try and keep moisture out. It’s a harsh environment though, and I have 5 separate type N connections, two antennas, and a splitter all exposed to the weather.
There is also the issue of lightning strikes. Normally an antenna system would use a lighting protector, which is a device in line with the antenna cable which attempts to short the unwanted energy to ground. There is no really good ground on the truck though. It sits on 4 rubber tires and its not like I can drive a big copper grounding rod down to the water table and then run a big heavy wire to the lightning protector.
On the other hand, serious lightning involves millions of volts traveling hundreds or thousands of feet through the air, so those insulating rubber tires are like next to nothing. I think a lightning protector might provide some protection from static electricity but a direct strike could easily blow a hole straight through my roof. Fortunately we don’t get many serious lightning storms here in the San Francisco bay area, though I do expect that I might end up in Texas or the Midwest from time to time.
Beyond all the RF and antenna issues it remains to be seen just how accessible Wi-Fi hotspots will be on the road. Clearly I am putting a lot of time and energy into making this work. It is important part of my plans. The truth is, I still have a lot to learn though.
To be continued…