Antennas and Impedance Matching
I have had a few discussions with a few people recently about antennas and I figured I would post my ideas and thoughts here as well.
The first guy was asking about a portable dipole type antenna that was easy to use for multiband operation from the ground and still be efficient. This seems to be a pretty popular discussion, I have talked about this exact topic so many times I can't count them all in the past several years.
One aspect that I think is really important about dipole-like antennas that people need to understand is a dipole is a dipole because of the electrical characteristics associated with that specific design. A dipole is only truly a dipole when it is 1/2 wave length long and center fed, thus the name dipole (di meaning two and pole meaning side, like a magnet). The 1/2 wave length contains all the sections of the sine wave that are part of the signal you are sending (RF is AC - it oscillates, and the rate of that oscillation is the frequency of operation).
The length of a dipole must be electrically 1/2 wave length. To figure this out you can use the equation 468/f, where f is the frequency in mHz. Note - this works for bare copper in free space. The environment you put your antenna in (proximity to ground, your house, trees, etc) will affect the actual tuning of the antenna.
When you use that dipole antenna on frequencies it was not designed for, say an 80 meter dipole on 40 meters, the current and voltage distribution on that antenna is messed up. This means the antenna is not resonant, and therefore not efficient.
So if the antenna is not resonant why not put a tuner between the radio and antenna? That will solve for the impedance mis-match and everything will be peachy. Right?
NO!!!
Because the voltage and current distribution on the antenna is messed up you can, in theory, use the reactance of the tuner to "tune out" the screwed up reactance of the antenna, thus bringing the impedance back to 50 ohms (which all of our radios are to begin with). Inductive and capacitive reactance adjust the lead/lag time between the voltage and current, allowing those components to mesh with the properties of the antenna in use.
What you don't see is the impedance after the tuner. The impedance of the antenna (and between the antenna and tuner) has not changed - only the impedance between the tuner and the radio is different.
You have to realize that the match between the radio and antenna, and the whole distance, when using coaxial cable, HAS TO BE 50 OHMS! If it is not, the loss characteristics of the cable(s) you are using are no longer valid. You can take Heliax LDF6-50A, with an inherent attenuation of .129dB/100ft @30mHz (AMAZINGLY low attenuation, and at about $4/ft it ought to be), and show an impedance of 400 ohms to it with a mismatched antenna, and your losses would be worse than that of RG-174 (1/8" coax, about 5.5dB loss/100ft @50 ohms). So much for your $4/ft feedline...
To add to it, every length of cable that is not matched to 50 ohms radiates. The larger the mis-match the higher the radiation of the feed line. This can lead to RFI problems.
So, my advice for everyone wanting to put up a dipole-like antenna, and in actuality it is not a dipole, but rather a doublet, is to mount the tuner at the feed point, or as close to it as you can (and this doesn't mean on the back wall of the shack with another 20' of coax to the antenna and 10' to the radios). You need to mount the tuner a couple inches from the antenna, and avoid using coax at this location. Either balanced line or single strand wires should be used.
Another option, if you have to use a feed line between the antenna and the tuner, is to use balanced line. A doublet is a balanced antenna and using a balanced feed line to match it will work, regardless of the impedance shown to the feed line. Balanced feed lines will always work best at their inherent impedance (450 ohms, 300 ohms, 600 ohms, etc), however balanced line does not have the same attenuation characteristics as coax does, coax being an unbalanced cable.
Another antenna circumstance is a vertical antenna. In reality, the vertical part of the antenna is essentially the same thing as the half side of a dipole antenna connected to the center of your coaxial feed line. Instead of the antenna being loaded against the other half of a dipole the other "half" is made up of your ground system. Don't confuse this RF ground with that of an electrical ground, in which case a ground rod could be used. RF ground needs to be efficient because it is part of the antenna. This is a problem with ground mounted antennas because the Earth (dirt) is a really good ABSORBER of RF energy (also known as the proverbial "ground sponge").
A vertical is also not usually a balanced antenna, unlike a dipole and doublet. Because of this, balanced feed line is not the best choice. When you show an unbalanced load to a balanced feed line it begins to radiate like mis-matched coax. Therefore, coax is always the best choice, unless you can ensure that your vertical antenna is balanced (hard to do).
This also requires that your antenna be resonant. This gets much more tricky than a dipole. In fact, a ground mounted vertical will be resonant (according to the electrical properties of RF - the voltage and current) at an impedance of 25-40 ohms, depending on the characteristics of the soil and the efficiency of the ground system. This will be roughly a 2:1 to 1.5:1 SWR. No where in that figure is the value of 50 ohms.
By tuning out the reactance of the impedance mismatch on the antenna (making it longer or shorter) you can bring the impedance of the antenna to 50 ohms, but the antenna is not as resonant as it would be at it's inherent characteristic impedance of 25-40 ohms. Confused yet?
In otherwords, if you strictly tune the antenna (by making longer or shorter) to bring the SWR of the antenna to 1:1, or as close as you can get, the antenna will actually not be as resonant as it would be at it's characteristic impedance. SWR is a ratio of the impedance of your load and the 50 ohm characteristic impedance of your radio. So, 25 ohms is a 2:1 SWR (50/25=2). However, 100 ohms is also a 2:1S SWR (100/50=2).
To match your feed line to 50 ohms you should put a matching network at the feed point of the antenna (basically a fixed tuner - inductor and a capacitor, or however you want to do it) if your antenna is a mono-band antenna. Or, mount a tuner at the feed point and use the reactance of the tuner to tune out the reactance of the antenna and match your feed line to 50 ohms, not passing that impedance mismatch through the coax.
In summary:
Any impedance, other than 50 ohms, that is shown to your coaxial cable will cause ATTENUATION. The higher the mismatch the worse the attenuation, and even the highest of quality coaxial cables are not immune to this. To make the most RF go in to the air your antenna should be resonant, or posses the matching device in close proximity.
If you want to figure out what the exact values of attenuation are (and from what I know it is difficult to figure out) you can research that one on your own. My understanding is what happens, not what exactly it is. As long as my SWR is low through my coax I'm happy.
The first guy was asking about a portable dipole type antenna that was easy to use for multiband operation from the ground and still be efficient. This seems to be a pretty popular discussion, I have talked about this exact topic so many times I can't count them all in the past several years.
One aspect that I think is really important about dipole-like antennas that people need to understand is a dipole is a dipole because of the electrical characteristics associated with that specific design. A dipole is only truly a dipole when it is 1/2 wave length long and center fed, thus the name dipole (di meaning two and pole meaning side, like a magnet). The 1/2 wave length contains all the sections of the sine wave that are part of the signal you are sending (RF is AC - it oscillates, and the rate of that oscillation is the frequency of operation).
The length of a dipole must be electrically 1/2 wave length. To figure this out you can use the equation 468/f, where f is the frequency in mHz. Note - this works for bare copper in free space. The environment you put your antenna in (proximity to ground, your house, trees, etc) will affect the actual tuning of the antenna.
When you use that dipole antenna on frequencies it was not designed for, say an 80 meter dipole on 40 meters, the current and voltage distribution on that antenna is messed up. This means the antenna is not resonant, and therefore not efficient.
So if the antenna is not resonant why not put a tuner between the radio and antenna? That will solve for the impedance mis-match and everything will be peachy. Right?
NO!!!
Because the voltage and current distribution on the antenna is messed up you can, in theory, use the reactance of the tuner to "tune out" the screwed up reactance of the antenna, thus bringing the impedance back to 50 ohms (which all of our radios are to begin with). Inductive and capacitive reactance adjust the lead/lag time between the voltage and current, allowing those components to mesh with the properties of the antenna in use.
What you don't see is the impedance after the tuner. The impedance of the antenna (and between the antenna and tuner) has not changed - only the impedance between the tuner and the radio is different.
You have to realize that the match between the radio and antenna, and the whole distance, when using coaxial cable, HAS TO BE 50 OHMS! If it is not, the loss characteristics of the cable(s) you are using are no longer valid. You can take Heliax LDF6-50A, with an inherent attenuation of .129dB/100ft @30mHz (AMAZINGLY low attenuation, and at about $4/ft it ought to be), and show an impedance of 400 ohms to it with a mismatched antenna, and your losses would be worse than that of RG-174 (1/8" coax, about 5.5dB loss/100ft @50 ohms). So much for your $4/ft feedline...
To add to it, every length of cable that is not matched to 50 ohms radiates. The larger the mis-match the higher the radiation of the feed line. This can lead to RFI problems.
So, my advice for everyone wanting to put up a dipole-like antenna, and in actuality it is not a dipole, but rather a doublet, is to mount the tuner at the feed point, or as close to it as you can (and this doesn't mean on the back wall of the shack with another 20' of coax to the antenna and 10' to the radios). You need to mount the tuner a couple inches from the antenna, and avoid using coax at this location. Either balanced line or single strand wires should be used.
Another option, if you have to use a feed line between the antenna and the tuner, is to use balanced line. A doublet is a balanced antenna and using a balanced feed line to match it will work, regardless of the impedance shown to the feed line. Balanced feed lines will always work best at their inherent impedance (450 ohms, 300 ohms, 600 ohms, etc), however balanced line does not have the same attenuation characteristics as coax does, coax being an unbalanced cable.
Another antenna circumstance is a vertical antenna. In reality, the vertical part of the antenna is essentially the same thing as the half side of a dipole antenna connected to the center of your coaxial feed line. Instead of the antenna being loaded against the other half of a dipole the other "half" is made up of your ground system. Don't confuse this RF ground with that of an electrical ground, in which case a ground rod could be used. RF ground needs to be efficient because it is part of the antenna. This is a problem with ground mounted antennas because the Earth (dirt) is a really good ABSORBER of RF energy (also known as the proverbial "ground sponge").
A vertical is also not usually a balanced antenna, unlike a dipole and doublet. Because of this, balanced feed line is not the best choice. When you show an unbalanced load to a balanced feed line it begins to radiate like mis-matched coax. Therefore, coax is always the best choice, unless you can ensure that your vertical antenna is balanced (hard to do).
This also requires that your antenna be resonant. This gets much more tricky than a dipole. In fact, a ground mounted vertical will be resonant (according to the electrical properties of RF - the voltage and current) at an impedance of 25-40 ohms, depending on the characteristics of the soil and the efficiency of the ground system. This will be roughly a 2:1 to 1.5:1 SWR. No where in that figure is the value of 50 ohms.
By tuning out the reactance of the impedance mismatch on the antenna (making it longer or shorter) you can bring the impedance of the antenna to 50 ohms, but the antenna is not as resonant as it would be at it's inherent characteristic impedance of 25-40 ohms. Confused yet?
In otherwords, if you strictly tune the antenna (by making longer or shorter) to bring the SWR of the antenna to 1:1, or as close as you can get, the antenna will actually not be as resonant as it would be at it's characteristic impedance. SWR is a ratio of the impedance of your load and the 50 ohm characteristic impedance of your radio. So, 25 ohms is a 2:1 SWR (50/25=2). However, 100 ohms is also a 2:1S SWR (100/50=2).
To match your feed line to 50 ohms you should put a matching network at the feed point of the antenna (basically a fixed tuner - inductor and a capacitor, or however you want to do it) if your antenna is a mono-band antenna. Or, mount a tuner at the feed point and use the reactance of the tuner to tune out the reactance of the antenna and match your feed line to 50 ohms, not passing that impedance mismatch through the coax.
In summary:
Any impedance, other than 50 ohms, that is shown to your coaxial cable will cause ATTENUATION. The higher the mismatch the worse the attenuation, and even the highest of quality coaxial cables are not immune to this. To make the most RF go in to the air your antenna should be resonant, or posses the matching device in close proximity.
If you want to figure out what the exact values of attenuation are (and from what I know it is difficult to figure out) you can research that one on your own. My understanding is what happens, not what exactly it is. As long as my SWR is low through my coax I'm happy.
posted by Steve at
5:17 PM
0 Comments
