|
Using an RF amp to overcome feed line loss works fine. For wi-fi type radios, they rapidly switch between transmit and receive and you need a Bi-Directional Amplifier to work well. It does little good to boost the transmit signal, if you cannot hear the ACK packet that comes back to complete the data transfer. So a BDA amplifies both transmit and receive. But this means that on the receive side, you not only boost the signal, you boost the noise as well. If somebody is running a high power RF amp in the area, they can cause you and all their neighbors unnecessary problems, without helping their own links work well. The receive side of your BDA will boost noise as well as signal and the absolute signal level being higher may do you no good if there is too much noise.
Signal to noise ratio as shown on the Broadband-Hamnet(TM) signal screen is often more important a gage of a quality connection than absolute signal strength. If you have a strong signal but also a lot of noise, your radio may have trouble correctly receiving data. As a rule of thumb, we find if you can get a S/N reading of 10 or more on the mesh nodes, you have a workable signal. Below 10 any slight drop and you may have problems receiving the data. So with little noise in the feed line, an RF amp can work well. I have done it, with an existing antenna at the top of a tower and 300 feet of feed line. But check the actual loss for the feed line - it does not usually take an 8 watt amp to overcome the feed line loss. The good path analysis software always allows you to plug in values for feed line loss when you are filling in all the blanks for an analysis. If you go by the calculations, you will be in the ball park and things will usually work as expected. Commercial microwave has been in use for decades and if slapping on a big amp would work well for the pros, they would have found a way to use them. They have nice path planning software and expensive test gear to go measure exactly how the calculated results worked. One of our local hams worked military then commercial microwave for decades and he told me most commercial links commonly run 3 - 4 watts of power. They have anti-torque guy wires added to arms on microwave towers to keep them from twisting. They get good LOS paths, or they move to a place that has a LOS path, or they create one with a tall tower - but they get LOS. Then they very carefully adjust their directional antennas at both ends of a point to point link, then use really stout mounting hardware to make sure the antenna does not move. Thats what works. The commercial gear often puts the radio on the back of the dish. Overcoming feed line loss works, and is good to know about, but some would consider it a patch for a less than optimum equipment configuration. Amps are not perfect. They add noise and can amplify a marginal signal to a point where unwanted parts of the signal are raised to a level where they cause problems. This was a classic problem with cheap CB RF amplifiers. The original radio had some noise in it but with the factory design, these unwanted parts of the signal were below an acceptable level. Amplify the whole signal and the parts of the signal that used to be below problem levels are way above acceptable levels. If these are not filtered, the signal going to the antenna has all sorts of unwanted noise in it. Feed line can go bad and fill up with water vapor that attenuates microwave signals. Wave guide is sometimes pressurized with nitrogen to keep the water vapor out. Putting the radio on the back of the antenna usually gets around having to pressurize your feed line with nitrogen.
In Austin, we have found that with good sites, simply putting a 24dB BBQ grill parabolic antenna at each end of a link, gets us 10 miles, at will, using a WRT54G with no added RF amplifier. If there is a tree in the way, move your antenna or the tree. If a building is in the way, you probably need to move, but it is sometimes fun to reflect signals off some buildings and make a "bank shot". This is usually not real stable as the reflective surface moves a lot and sends your signal all over the place. I have used the rounded belly of an old style 4 legged water tower (or as they are called by the makers "elevated tanks") to shoot basically straight up at the belly of the tank from below and then adjust the antenna to make contact with another mesh node, when the angle of the antenna is just right. This works might work if you are portable and socked in by trees and buildings. I used a 24dB dish, and no amp to link with AD5OO, years ago in early mesh field tests. At night a LASER pointer taped to the feed support helps in fine tuning the antenna - and you really have to get it fine tuned for a decent link, reflected shot or not. No waving the antenna at the birds and calling it aiming. Take a compass and get a vector that is the path, then slowly move the antenna very slightly left right / up down until you link. At a distance of miles, very slight moves of the antenna make a big difference with a highly directional antenna.
Get out in the field and try the gear. Do some calculations with free software. A tripod and compass rose for it are common for portable microwave systems. I have a compass I got back in Boy Scout days and it still works fine for my microwave work.
-Glenn KD5MFW
|
