Originally published July/August 1998
(A series of explanations/procedures for the COMPLEAT IDIOT)
Since Big Brother Government is constantly "improving" ATC and is always trying to save us from ourselves, whether or not we wish to be saved; it behooves the average airplane driver to be able to talk to these people as they deem necessary (pay attention here, Waldo). Being homebuilders, we should be able to design and install superior devices for enhancing the prospects of clear communications. Historically we don't. Airplane people generally go down to the nearest airplane parts place and fling enormous sums of hard-earned cash for stuff that has the magic incantations "FAA-PMA approved." I wonder what authority the FAA has with FCC radio related things? I guess government is government, like parts is parts.
I find that there is a certain satisfaction of being able to say to ATC that they need to repair their equipment when they cannot talk to me. Of course, being GOVERNMENT employees, they don't listen, but.... I also constantly talk to other aircraft, Approach, Departure, towers from 80 - 150 miles distant. I listen to BFL or WJF towers from 50+ miles and through mountains on a hand held radio. I do it, why not you?
Anyhow, I will attempt to provide an understanding of what some consider a black art WITHOUT using math (not much anyway, mostly because I am required by law and accountants to totally abstain from doing anything remotely resembling math). I will keep this to a level that anyone can understand. If I can understand it, you ought to be able too. The lone exception will be Rocket Scientists because they are so fixated on whatever project they work for that gaining entry to their conscious is difficult. You, the reader, are therefore obligated to not snicker loudly when I give over-simplified explanations. Besides, you are reading this because you think I know something you don't. Maybe you just need Prozac.
I started my informal education of design and installation during the real early days of the CB radio goofiness, prior to, say, 1972. Guess what? Antenna design for those days translates into proper antenna design for aircraft. Who'd a thunk...This early experimentation with handheld 500 milliwatt children's walkie-talkies led to a personal favoritism concerning very long length wire designs, some times many wavelengths long. CB (29 MHz /10 meters) frequency makes for a very long antenna (see following formula) which is impractical for aircraft use. Fortunately, aircraft bands require only about 8 ft (less than 3 meters) for full wave radiators with other designs being significantly shorter.
We do have a serious problem in the Aviation Bands because they occupy the spectrum of 108 to 135 MHz which is a spread of 27 MHz. This spread is equal to several entire amateur radio bands, and those guys use as many as 5 DIFFERENT antennas per band. Obviously, a single purpose antenna is not going to be efficient due to extraneous stray capacitances and inductances affecting antenna lengths. In Idiot: too big or too short to hum along with the song.
Another problem we face is the current annoyance that Government requires ATC to use the higher portions of the Band for ENROUTE COMMUNICATIONS because they were stupid and filled up the lower areas with mandated ground and tower talking. Voltage to Standing Wave Ratio (VSWR) rises sharply with antennas being off-resonance, and efficiency suffers greatly. One solution is to have two dedicated COMM radios and antennas, each tuned to a portion of the band: One that centers on 123.000 MHz and the other centered on 128.000 MHz. Each will function somewhat on the other's assignment, but will be the most efficient where they are tuned. Obviously, you do not want to purchase a radio that has both NAV and COMM functions derived from the same antenna.
One half wavelength (inches) = [ 495 / (frequency in MHz) ] X 12 for free air dielectrics in ether.
Notice the formula is for FREE AIR and bare wire or not covered with insulation? By the way, I wonder how they get free air out there in the ether?
This formula is based on the theoretical distance required to form a sinusoidal wavelength in space (The final frontier. These are the voyages of the Starship Enterskies.).
Everything in antennas is conveniently based on either 1/4 or 1/2 wavelength for spacing, lengths, distances from other antennas. Figure out how long a 1/2 wavelength is for your interested frequency and you have most of your problems solved. By the way, 1/4 wavelength for midpoint of most tower frequencies (123.000 MHz) is 24" more or less [way cool nerd thing to know]. You engineer types just figured out that it is actually 24.146341 inches, but IS IT REALLY? More later.
VSWR (voltage to standing wave ratio): What you get when the antenna does not radiate all the energy you throw at it. Effectively, you get relatively high voltage returning back down the antenna and back into your radio. The lower the VSWR number the more power you are radiating, the better you communicate because the radio is not wasting power. It is possible to have a VSWR ratio of 1:1 using idiot-proof techniques. If you have a real high VSWR, say over 3:1, you can and probably will overheat or destroy the final amplifier stage leaving you without a transmitter. VSWR will increase for frequencies high or low that are not matched to your antenna length. Responsible radio manufacturers usually have a circuit that automatically limits power to the final amplifier if VSWR gets too high (above 2:1).
Responsible antenna manufacturers should make antenna VSWR ratios below 2:1 but THEY USUALLY DON'T !!!
A 2:1 VSWR means you are bouncing at least 3/4 transmitter power you just threw at the antenna right back into the transmitter like a ping pong ball. What does the transmitter do with reflected power? Make heat. Ever notice that the belt clip of your ICOM hand held gets hot when you use the rubber antenna a lot? Do you know how ICOM heat sinks the power amp? That's right, kiddies, it is the metal belt clip. ICOM considers the rubber antenna an emergency antenna only.
Impedance is a restricting force against the free movement of electrons. FREE RANGE ELECTRONS? You betcha, Pilgrim. In techno-babble, impedance attenuates electron flow. In Idiot English, impedance is a drag. Impedance always makes heat, which is considered bad Karma, DUUDE. Big impedance, big heat. Little impedance, much more heat, and, you let the smoke out of your components which is difficult and expensive to put back in.
Impedance can be Inductive (+), Capacitive (-), or resistive (0). Carbon fiber or fiberglass is a combination of all these. Capacitive and Inductive Impedance tend to cancel each other out and make zero for one set of circumstances (usually frequency). Everything has impedance of some sort which is measured in OHMS but you cannot measure these ohms directly (you guess or pay large sums of money to consult a Electrical Engineer). Impedances add together and do weird things in parallel. Impedance is also the thing that makes antennas not have a free lunch because you can't change one thing without affecting overall impedance (always higher).
Capacitance shortens antenna lengths because it makes electrons apparently move slower. WHATT??? HUNH!!? Does Einstein know about this? WHY? Well, without math it is impossible to explain, so just trust me. Everything that is not a perfect insulator to RF energy (nothing known except possibly mega-magnetic fields in magnetic linear particle accelerators or black holes) has capacitance, which is one reason why 'glass planes usually have so many antenna problems. Capacitance is also a dastardly thing that has an annoying tendency of decreasing impedance as frequency goes up.
Inductance is like magnetism or a big fishing net, at least as far as electrons and radio waves are concerned. The more inductance you have the more electrons you capture. The longer the wire you have the more inductance you have (somewhat how coils work). However, the more inductance you have at high frequencies, the more impedance you get.
Velocity Factor is a weird effect caused by capacitance, that makes the required wire actually shorter than the theoretical length at the speed of light. Wire manufacturers publish charts with this data. You just multiply the resonant formula length by the velocity factor to get actual length. Note that this only counts when your wire has a insulating covering which YOU won't use, right?
Note: If you want to get really efficiency nuts, you have your co-ax antenna wire in exact multiples of 1/4 wave, so this is where velocity factor is important because the length will be shorter than theoretical.
IMPORTANT>>>>> There are no known insulators to RF energy. It will penetrate rock, metal, PLASTIC, FIBERGLASS, skin, yada, yada, yada (particularly penetrating to that last yada). You can only "channel" RF energy along conductors (how shielding and wave guides work, but not 1-900-PSYCHICS) The "HAM" guys actually use electrical components like resistors in real high frequency stuff for structural materials.
Things that are at least 1/4 wavelength are always a radiator of energy. This is how things that look like a short aren't a short. A 1/4 wave delta match antenna at 1296 MHz is a dead short less than 2 inches long. I still have my own problems with that concept but use blind faith because I tried it and it works. If I remember correctly, your CUTE LITTLE TEENSIE wire transponder antenna is actually a full wavelength long.
RF energy will collect and flow down metal surfaces (like wires) because most metals have inductance. If the metal is not an even part of a wavelength long, RF will re-radiate itself when it gets to an end of the piece of metal (sometimes useful to know, mostly a problem because it can cause signal cancellation, echoes, loss of signal etc., but it IS how your TV antenna works). If the metal is an even part of a wavelength long, the energy hangs around looking to do something useful before it gets wanderlust.
If the VSWR is low, the antenna receives better. If an antenna receives well it probably transmits real good too. This factor is called reciprocity, which is a somewhat meaningless bit of trivia.
There are two kinds of antennas: balanced and unbalanced. This is only related to the CO-AX feed method rather than the antenna.
>>>> NOTE: REAL IMPORTANT >>>> ALL antennas MUST have two conductor/radiators, it is just that in some antennas one radiator is camouflaged to look like airplane, or car, or dirt. Sort of like the oriental Ying/Yang theory. Fiberglass and that ilk are not conductors to direct current energy; another reason they don't work as a ground plane.
A Monopole or MARCONI antenna (typically used on metal skinned aircraft) has only one apparent radiating element more than 1/4 wavelength long and a characteristic impedance of 50 ohms. Hmmm. This is an unbalanced antenna. You connect the shielded portion of the cable to the antenna mount or ground. My CO-AX radio antenna lead in has an impedance of 50 OHMS per length...Lesson here: Why do you think most metal aircraft have monopole antennas? (hint: 50 OHMs impedance, ease of manufacture...) However, you do get weird impedances an lengths longer than 1/4 wave until you get to 1/2 wave. They can increase or decrease as the electrons see fit. There is actually a mathematical series of equations concerning skin current flows and related yucch, but that stuff makes my head hurt.
New Lesson: Why don't glass aircraft have monopoles? Because there is no provision for the ground portion of the antenna. You do HAVE to provide some sort of a ground plane of reference or the antenna impedance can end up being anything it wants to be. The ground plane can be 1/4, 1/2, 5/8, or 1, 2, 4, 6 multiples of wavelength. This radiator HAS TO BE one of the above even portions of wavelength or it does not work. 5/8 is an exception, but don't worry about it. By the way, each antenna needs its OWN GROUND PLANE REFERENCE.
Monopoles work best when the ground reference is at least 6 wavelength in diameter. Why 6? Those that have done STANDARD DEVIATION/STATISTICS know that when the std deviation gets beyond 6 sn-1 the population becomes so small as to be inconsequential. Funny, the same effect in Electronics. Those like me that flunked math and need Idiot English: 6 wavelengths are so far away that the electrons get too tired by the time they get to the end to cause trouble.
Monopoles can and will work with teensy ground references, or monstrous ground references, or, anything in between, but you get weird radiation patterns and angles which could render the antenna useless. You sometimes get weird impedances, too. Monopoles can be bent into all sorts of weird shapes, but once again you are modifying the radiation pattern which may or may not be a problem. People that say their antenna does not require a ground actually use the coax outside shield as a ground radiator; some times the aluminum radio case too.
Dipoles have TWO active radiating elements (DI: a word from some ancient dead foreign language meaning two, not a dead foreign princess). Dipoles are usually 1/2 wavelength long. It is a BALANCED antenna, which means you should not connect your CO-AX directly to it without some sort of adapter called a BALUN. They have a CHARACTERISTIC IMPEDANCE of 72 to 75 OHMS which is why they are usually not efficient antennas for aircraft. 50 OHMS of your radio does not balance 75 OHMS of the antenna so you lose some efficiency. You need to BALUN your radio because you could be flying over; SAY, all TEN MILLION WATTS of KNX TALK RADIO on your way to Torrance ZAMAPARINI Airport. The shield of your COAX becomes an antenna of sufficient length to receive KNX RADIO, and feeds talk radio energy into your radio. Best case is no effect to you. Middle case is that you listen to Talk Radio instead of the pearls of wisdom that Torrance Tower has regarding your landing clearance. Worst case is you are out the huge sum of money to replace the 1ST transistor receive amplifier (which now-a-days is on a chip and costs more). Fortunately you can still transmit in the blind because that portion was not hooked up to the antenna at the time. A BALUN (acronym for BALanced to UNbalanced) takes the antenna and converts either end to what the other end needs to see. Since one half of the dipole antenna is positive and the other half is negative, KNX RADIO is balanced at neutral. It also provides some increase of efficiency because your antenna is now twice as long as it was previously. Baluns do not solve the 75 OHM IMPEDANCE problem, however. Most people take the increased efficiency (200%) and average it out against the impedance loss (25%), and so are happy with what they get. I use one for my nav portion, but it only works better when flying FROM an OMNI. I also use a 75 to 50 ham impedance/adapter BALUN called a "BAZOOKA" because I like the conceptual thought of "BAZOOKAing" my transmissions.
These guys with their fancy circuit board dipole antennas are just scamming the public on how good they actually are. You don't see me using one do you?
Traditionally, for monopoles, you visualize the ground plane as the direction of most efficient radiation, that is, where you get the best patterns. In Idiot terms, where the biggest piece of airplane points is where you talk the best. I can show you reams of data, but trust me. Therefore, for monopoles, if you want to talk up to the SPACE SHUTTLE in orbit, you put the antenna on the bottom. If you want to talk to towers the antenna should be on the top. My technique is to use a dark area like a hangar with the doors closed and lights off, and a flashlight since light and radio waves behave real similar. Take the flashlight and place on the pointy end of the antenna with the light beam facing the aircraft and running down the antenna length. The shadow approximates which direction will work the best. Did you get a surprise?
Dipoles are different. Since you have both elements radiating, Dipoles work like so: Take a doughnut, place over the antenna so that the hole slides through the center of antenna, and the doughnut is perpendicular to the antenna. Where the doughnut is biggest is the direction you talk best. Notice that if the dipole is in the wing, you talk forward real well (handy for talking to the tower from 20 out) but do not talk well out the sides in the pattern. It is even possible to even not hear the tower in the pattern. Dipoles should be located in things that are vertical, rudders, speed tips, fuselage, etc, but there is another problem which is that dastardly fiend, capacitance. If you mount antennas inside of things; because everything is a conductor to RF energy, you add capacitance. The VSWR goes way up, the antenna length requirements go way down, it is an terrible, terrible, ugly thing to behold. Can you calculate the effects? NOPE, you can only measure the effect after you install. You could encapsulate a thing that fits where you plan to put the antenna and simulates the installation then test the results, trim to resonance, retest, but that is time consuming and I don't know anyone that has a bridge that works on transponder frequencies. Best remedy is to take antennas and hang them outside in the breeze; take the 1/2 knot speed penalty [DON'T GET ME STARTED ON SPEED MODS (LIKE YOU CAN REALLY MEASURE 1/2 KNOT above 100 knots CAS)]. (Can I eat the doughnut now?)
Ask anybody interested in appearance (yuppie scum, dinks, and 2nd Lts): Looking cool, good, and, stylish is not conducive to comfort and ease.
A cool thing about dipoles and monopoles is that you can streamline them in a 45 degree manner and gain electrical performance while lowering drag. You actually end up fooling the electrons into believing that the antenna is shorter, so it works on higher frequencies (this is called bandwidth). However since TANSTAAFL1, you become directional where the open end is, or opposite to the streamlining, which is O.K. for Variezes 'cuz they fly backwards anyways.
By the way, things like gear legs, flying wires, metal spars, strobe leads, fuel lines, all being a mass, alter the pattern you are radiating. If you put any antenna between the spring gear, you could effectively block all the energy and absorb it into the gear. Remember the thing about inductors and wire?
The doughnut TRICK works on monopoles, but gets real bizarre when there are bends.
If the antenna you are talking to has his antenna horizontal, your best efficiency is to have a horizontal antenna because the waves of radio energy match up. MOST FAA/FCC people hang their antennas on long poles in a vertical manner because it gets the antenna higher into the air (works better, but not an issue at 10,000 ft MSL). Guess what? The wing mounted horizontal dipole looses maybe 50 to 75% efficiency talking to towers. Government people get around this problem by throwing around huge amounts of power, say a thousand watts (1,000 X 1% efficiency = 10 watts at 100% efficiency). They can yell at you from very far away, but they cannot hear you yell back. 75 miles is generally their limit. In order to yell back, you will have to consume prodigious amounts of alternator power (about 25 amps, which can be bad Karma), so that when your radio wave arrives at their antenna, it really excites them electrons in the gov't antennae. Gets 'em right agitated y'hear?
The hot new thing in antennas for GPS is CIRCULAR polarization which is a real pain to do. How it works is so: the antenna is many, many, many wavelengths long and is wound in a helix pattern like a spring. Why do the GPS guys do this? 'Cuz their signal is so teensy that they can't afford to waste any signal in polarity losses. A spring will accept polarity in any direction because there is a full wave antenna in a 360 degree pattern. There is a matter of left hand and right hand polarization, but now we are getting really obscure, and since GPS is such a hardship to deal with frequency wise, just let the PROS do all the headache stuff.
Think of antenna resonant length like a tuning fork. The tuning fork only vibrates at one frequency when you excite it. You can excite it with other frequencies, but they have to be some sort of a multiple of the tuning fork frequency, unless you pump massive amounts of energy into the fork, such as a sharp and sudden whack. Same thing with antennas, only they also have a sloppy area of excitement (wide bandwidth). If you are not resonant, it ain't gonna "HUM" properly. Remember that the physical shape of the tuning fork affects the operating frequency just like on an antenna, i.e. Fat forks, like fat antennas have to be shorter for the same frequency as a skinny antenna.
This is a mythical performance factor that uses a dipole as a comparison factor. Gain works funny (logarithmic) so that it takes 6 gains to make one twice as strong power, or twice as good. PLAIN old vanilla straight Dipoles are considered to have a gain of "UNITY" (1). So if someone claims to have a GAIN of 3, his antenna only works 1/2 times better which is not considered to be a big deal. Usually gain gets bigger at the expense of directivity. Omnidirectional antennas have lousy gain. Satellite dishes have humongous gain (like 40 to 75) but teensy little itty-bitty aiming points. Sometimes the antenna guys try to fool you and say their antenna has a gain of 30 dB. dB is just another way of saying "gain". If the antenna guys say their antenna has a 3dBa gain, they are shucking you because the dB(A) scale is modified for only what human beings hear and has no bearing on radio frequencies.
These are little weird things that some math magician invented so he could appear to be smarter than me. In RF, everything works in a logarithmic way. When you stack losses together, you add them, but they are equal to 1/3. 6 in a stack makes either 1/2 of something or two of something. In RF you can't subtract anything, so everything you do adds to signal loss. A connector is usually a 1 dB loss. Most radio systems have 2 connectors, so you have already lost 2/3 of incoming signal strength (why you need 3dB gain to start with). You need to look up how many dB you loose per foot of cheap co-ax wire. That can be scary. Cool thing about dB is that you never really get to zero, but get so tiny that zero and your result are real close together shaking hands.
Rule number 1: keep your cable as short as possible. Not using any cable works the best.
Rule number 2: Do not buy Radio Shack RG58-U plain wire. It has a single wire center conductor which will break, usually just as you are about to enter Burbank ARSA. [What do you mean you do not fly into the Burbank ARSA so this warning won't apply to you??? AAAARRRGH!]
Radio Shack pre-made cable assemblies with male BNC connectors in 6 or 12 ft lengths are made from the "GOOD STUFF" cable which is RG58-AU or RG58-CU. They call the stuff test equipment hook-ups, it has BNC connectors of the right shape, and it's real cheap. RG-8 is real good stuff, but fat. There are more specific types of cable; thinner, thicker, in day-glo colors that have higher "RG" numbers, but mostly the higher the "RG" number, the higher the cable costs, like double, triple, like how much do you have to spend? You need to be a government to afford.... Cable manufacturers are always proud of their product and stamp their name and cable type all over the covering. If someone gives you 300 ft of unmarked RG-178, be suspicious of people in dark suits and sun glasses.
Do not use TV cable, it is the wrong impedance (75 or 300 ohms) and does not balance your radio. Do as I say, not as I do.
Mil spec requirements state that you need to secure the wire every 3" minimum, unless it is in a conduit. I recommend that you procure some real thin wall metal pipe to run your wires through; for: sensors, strobes, tip lights, whatever. Aluminum foil glued on the outside of a tube will also work. You will thank me later when a wire fails inside your wing. You tie any fuselage ends to the negative lead of the battery and WOW all your radio noise will disappear like magic! (do not connect the outside ends to ground) You can also use this conduit as the ground plane for all your antennas, even in 'glass ships. Buy the cheap stuff and get it big enough for more wires later.
Do not bend your CO-AX tighter than a 3" radius for small cable, or 6" radius for large stuff. You say that Radio Shack stuffs their cable into real small packages? Yeah, but they guarantee their stuff to work, and you can take it back if it don't. It is difficult to fix antenna leads when flying along at 10,000 ft. unless you are in the YB-49 flying wing, or HK-1 Hercules.
Note:>>>>> do not EVER touch antennas with ANYTHING when transmitting!
I guess by now you have some way of telling your VSWR is worse than 2:1. Most people use a SWR BRIDGE. Don't use the cheapies unless you know that they actually cover the aircraft bands. Most bridges are for lower bands, and since they have capacitors (UGH), are not reliable above 50 MHz. If you know some amateur radio person (HAM), borrow some of their 2 meter stuff. You will have to go to Radio Shack to get the adapter ends, because Ham's use a different connector for their equipment. By the way, VSWR gets better when the meter needle moves less. You want it so the meter does not move at all, but make sure the meter works, first.
Tune your radio to 123.000. If the VSWR is bad (the needle moves a lot), try 119.000. If the VSWR got lots better, you have: 1.)an antenna tuned for 119.000, 2.) a too much length problem on the radiators, 3.)kinked CO-AX, 4.) a 'GLASS airplane. Tune to 133.000. if the VSWR got lots better, your antenna was too short, or is in a 'GLASS plane. If it did not get lots better high or low, you have kinks, or a crummy installation, OR a scum radio. I can't help you with the glass airplane problem, just throw those nasty fiberglass things away.
If you have too much length, try bending the antenna until ultimately you have a 45 degree gentle bend. If that works, smile and be on your way, ignoring the ugly kink. You can always cut off the extra length, 1/16" at a time, but try bending first. If your antenna has a little ball on the end, DO NOT cut that end off. That little ball is real important electrically, not just to keep from poking your eye out.
If your antenna is too short, well; OH, well. You can make it longer, but it will always be ugly with bumps and gobs. Use house style solid copper wire, about # 12 or #10 gauge, AFTER taking off the insulation. Figure out the length you need (remember formulae pg. 2) and make the length that long over all. If that fixes your problem, smile, if not, sell that stupid VariEze.
For cable kinks, use a new piece of cable or other antenna. If the problem goes away, well, shoot howdy, fix the cable or use the other antenna. This is why conduit in the wings is soooo nice. If the problem stays, you are going to have to spend some real money getting your radio fixed. Try someone else's Handheld on the problem antenna. If the VSWR fixes, get your VISA card out and remember to smile friendly at the radio technician, because they can sense fear or ignorance and it effects your bill.
Another important VSWR problem comes from having your antennas too close together. Anything less than 1/2 wavelength (4 ft) apart, and the antennas will inter-react with unpredictable results. By the way, if you are going to have bunches of antennas all over the ship, do not shorten the leads, just let them dangle in a tidy sort of way. I personally think that more antennas than installed radios is just asking for communications problems. More radios than the one you need is just asking for money problems.
You 'Glass guys can have your internal antennas if you have a standard fuselage. Just imbed some copper or aluminum window screen somewhere, about 4 square feet. Tie this to the ground system or negative battery cable. Go buy a 1/4 wave monopole kit from anywhere, and mount the antenna inside the fuselage, not touching anything. I am working on a solution for backwards flying fiberglass log grazing aircraft that involves Radio Shack Burglar alarm foil and magic.
There is an on-going theoretical war concerning the gain advantages of "stacked" antennas. The gist of the arguments is that when you have bunches of antennas, they all receive, effectively, the same amount of radio energy. Stacking them together is like stacking flashlight batteries to get more voltage. The BIG SKY SETI listening program uses this concept in the Owens Valley. The down side is that impedances in parallel become less so power matching becomes difficult. A folded dipole (300 ohms Z) done 6 ways = 50 ohms, more or less. Insuring that all the antennas are fed the same energy at the same time (co-phasing) is the problem.
1There Ain't No Such Thing As A Free Lunch
Contents of The Leading Edge and these web pages are the viewpoints of the authors. No claim is made and no liability is assumed, expressed or implied as to the technical accuracy or safety of the material presented. The viewpoints expressed are not necessarily those of Chapter 1000 or the Experimental Aircraft Association.
Revised -- 19 March 1999