Site Construction -- Details, Good Practices
As the TARPN infrastructure is developed, some of the point to point paths (potential links)
are marginal so there will be a natural tendency to want to increase the power levels to make robust RF links.
A worthwhile point to remember is to make reasonable efforts not to apply too much overkill when establishing any given link.
Remembering that the FCC rules tell us to only use as much RF power as is necessary for any given communication need,
here's a few technical mechanisms to increase link performance:
1. As already stated, limit transmitter power as much as is possible to what it needed to make a reliable link to the next station.
2. Try to apply good RF design practices as is reasonably possible, including:
a. Keep away from cheap feedlines. Use well shielded coax, double shielded coax or even hardline whenever possible
b. Separate same-band antennas as much as possible to eliminate "near field" radiation between antennas. A good rule of thumb if mounting space permits is to keep vertical polarized antennas directly above/below each other or aimed away from each other and with significant horizontal separation. Ideally give at least 10' of vertical space between UHF antennas and 20' of space for VHF antennas. Two VHF vertical dipoles mounted with 20' of vertical separation have MORE isolation than the same omni-directional antennas mounted 60 or 80 feet from each other horizontally. Planting yagis such that they are aimed away from each other is also good.
c. If doing vertical separation, give the priority top antenna slot to your weakest link. Generally speaking, better antenna placement almost always is more beneficial even if there is more feed line loss.
d. Long feed line runs can benefit from sharing the line run by using diplexers or triplexers and investing in low loss coax or hardline.
e. Make sure VSWR is kept as low as possible. High VSWR can create problems of its own, not only by reducing antenna performance but by also potentially putting RF energy (signal) places you don't want it. Some of this effect, when experienced and VSWR cannot be improved, can be mitigated by putting RF beads and coax ferrite snap on bands on the coax. It is best however to use a properly resonant antenna if possible.
f. Use good grounding practices, including use of lightning arrestors. If even one outside antenna is not well grounded and arrestor protected, your entire station can become TOAST by a lightning strike, or even a near strike. (Tadd & I have lost our share of sites to the combined efforts of Mother Nature and Murphy's Law. Trust us when we say that good grounding and lightning protection are a must!)
g. Unless you are running coax to separate points of exit from a residence or site, us a single point of entry design to maximize (f.), even if it means adding a few feet of feedline for a given antenna.
h. If using more than one yagi antenna at a given site for directivity and path gain, and the antennas are on the same band (eg: 449.875 and 430.875 mhz), you can gain an additional 20 db or so near field isolation by alternating between vertical and horizontal polarization. Very effective!
i. Folded dipole, DC grounded yagi antennas have proven to be the most resistant to weather related antenna detuning. Some of the most robust RF links that were established for the NEDA network back in its day used a ham reproduced clone of a commercial folded dipole fed yagi. These antennas have been proven to keep both resonance and radiation pattern intact even when coated with 1/2 inch of ice.
j. Select operating frequencies carefully, and if running more than 1 link from any given site, use an intermodulation program to make sure you do not inadvertently create receiver blocking through harmonics or mixing products. (yep..sure nuff....your 144.93 link will knock the socks off your 434.790 link if you are so brazen as to try it!) Also, the higher the power level, the greater the mixing products. You can also minimize both out of band and in band RF loading and receiver blocking by using simple pass band filtering if available. My favorite filters are the 6" 1/4 wave 2m cans. They also work on UHF very nicely in 3/4 wave resonance mode and can be adjusted for thru loss settings between .5 and 3 db.
k. Remember that RF can float around on equipment, TNC's, computers, monitors, power supplies, audio lines, etc etc. Use good grounding (preferably common grounding), dressing and cabling. RF beads and snap on gizmos at your DC distribution bus and on computer leads can help reduce random signals. Have you got a Pi that's in one of those nifty clear PLASTIC enclosures? Consider spraying the inside of the enclosure with conductive paint and adding a grounding screw to the case. Another thing to do is listen to a very weak signal (or look at, if you have the technology to peek at your receivers detectors), then power on and off all the electrical devices you have to see if any of them are creating spurious RF that your link receivers can hear. In the early days of networking with computer based hardware, it was not uncommon for some motherboards and peripherals to radiate intolerable amounts of broadband RF hash. FCC rules for computing enclosures eventually improved things, but its good to be aware of devices that generate RF noise. I just took out of service a Chinese manufactured CFL light whose power source was radiating so much crap it was jamming a third of my home wireless power controls. Wow!
Related point, make sure all power AC connections have surge/noise/hash protection & filtering. I think by now most of us have learned this lesson (plus the hardware is super cheap if you know where to look), but lest we fail to remember the obvious....here it is!
--- Dana, WA2WNI April 2014