Feeding the Loop - I found that the type of Gamma match described by Ted Hart in the ARRL ANTENNA handbook is strong, has no loss and is
very easy to adjust. It will hold the SWR to almost 1:1 from 1.6MHz to 8MHz. It may well be that using a coupling loop could make adjusting the
SWR easier compared to the gamma match if the loop has loss caused by poor construction, low Q capacitor or loop location. I find if my loop has
been compromised in any way then the amount of bend needed in the gamma match has a direct correlation to the Q or Loss in the loop to the
point where it will not match without moving (welding) the tap point further up the side of the loop and lengthening the matching section.
People have tried to adjust the matching system with the loop horizontal and just above the ground, don't do this! The induced Loss and resultant very low Q will make any adjustments useless or impossible. The aerial should be placed vertically and at least one loop diameters away from any structures that could absorb RF during the test. Don't let anyone tell you that feeding or adjusting a correctly made loop is difficult.
If the resonant SWR changes radically between bands (cannot be bought down by resonating the loop) then look for something nearby including very poor ground which could be inducing loss into the loop and causing the loop impedance to change with frequency. A loop mounted in a good location, even one less than one loop diameter above a good ground will resonate and match to give almost a 1:1 SWR over the frequency range.
The Gamma match - This is an unbalanced feed system. There is a case for common mode currents. I have not found RF on the coax even under HI power in this installation with the short coax run straight down from the loop through the metal roof and with separate grounds for the loop and for the support mast. A separate earth should be used for the loop and for the mast in any case. It should be noted that the gamma match will skew the radiation pattern causing a front to back ratio of around 6dB favouring the direction of the of the gamma match mount.
I have not found induced noise on the feed line as the signal to noise ratio of my big copper loop is the same as a small shielded RX loop. It has been suggested that using a shielded coupling loop will also stop noise being picked up by the coupling loop. If there is that much impulse noise around then I feel the noise induced into the main resonant loop would swamp anything induced into the small coupling loop.
Note: I am not talking about noise induced into the feed line which can be reduced by using loop coupling to preserve the balance of the feed system and the main loop. - In my situation I have not encountered this problem.
Update - I have just tried a shielded coupling loop. There was absolutely no change to the noise level. I rotated the antenna and increased power line noise above S9. Once again there was no difference between the gamma match and the shielded loop coupling. The matching is hard to adjust without making the coupling loop completely self supporting. The base of my antenna is 2m above the roof making the location and shape of the coupling loop difficult to reach or adjust from this position. I was unable to get the SWR below 1.8:1 on 40m in the time spent testing, although I had 80m and 160m close to 1.1. I will retest when I can do meaningful far-field signal strength readings. Someone had reported a drop in output with loop coupling unless you get the size just right - apparently you can get the SWR correct but still have less output if everything is not set up correctly - I will be able to test this and also plot any change in radiation pattern in the near future.
Construction - Wire versus Copper - Some loop builders have said that they didn't note any difference between a thick copper loop and a thinner wire loop. Other builders claim their wire loop design works as well as a copper loop. Why? Well first a quick refresh:
Radiation resistance (Rr) is that part of an antenna's feed point resistance that is caused by the radiation of electromagnetic waves from the antenna. The radiation resistance is determined by the geometry of the antenna, not by the materials of which it is made. It can be viewed as the equivalent resistance to a resistor in the same circuit. Loss resistance (Rloss) is in series with Rr and consumes/wastes power - Rloss includes capacitor, wire, component and construction losses. Then there are environmental losses caused by ground and other induction losses which all conspire to lower the efficiency of any aerial.
The first thing to look at is the construction of the loop, particularly the connections to the tuning capacitor and the type of capacitor used.
If you introduce any loss with the capacitor or the loop construction then there will be very little difference between a copper and a wire loop.
This is why some builders in their haste to get a small copper loop built underestimate the loss caused by construction, components and location and then find that the wire loop they built seems to perform as well as the copper version they tried.
The radiation resistance (Rr) is around 0.005 Ohms in a 3.6m diamiater copper loop at 1.8Mhz. Any loss in any connection, like short coax braid connections to the cap will destroy the loop Q / efficiency by introducing resistive losses (Rloss) that are equal to (if you are lucky) or greater than (99% of the time) the radiation resistance of the loop. A capacitor with low Q and / or contact resistive loss will have the same effect.
Most people can't seem to get their heads around such low values of Rr and the high current flowing in a loop - even with low power - and they completely underestimate the loss in materials and construction techniques that they employ.
Good builders know that any aerial designed for portable use will have some compromises and they state this in their construction projects like
Alex PY1AHD - These portable loops should not be compared in preformance to a much bigger loop or dipole as some writers and builders do, why? They should rightly be compared to other portable aerials and in that case the loop, if built correctly, is usually superior.
Stated again - Any resistive loss you introduce with your construction and components could be in series with 0.005 Ohms or less of Rr.
The ¼ wave lenght loop -The second case is a loop that is a ¼ wave length in circumference. This loop length will work when made of wire.
It will not be as efficient as a ¼ wave copper loop (unless the proponents of such loops can tell me how introducing a resistive loss (wire) in series with the radiation resistance of the loop can increase current flow through the loop into the radiation resistance) but even with a loss of 50% (1/2 an S point) over a copper loop you will not notice the difference on air some of the time. This is an easy loop to try and it works.
A few loop builders are confusing radiation resistance (Rr) with feed-point resistance. Because of the low value of Rr in a small loop antenna the conductor loss is a significant part of the total resistance. In a large dipole or vertical antenna the feed-point resistance may be almost equal to
the high value of radiation resistance. In a small loop the feed-point resistance must include the significant value of conductor loss resistance.
Experts claim the current is uniform around a loop - No they don't. It has been stated again and again that the current is ONLY uniform in very small loops - NOT a ¼ wave length loop. (ARRL Antenna handbook 18 years ago) - And again a 1/4 wave loop does not have tiny values of Rr when compared to a small loop, so you don't have to build a ¼ wave loop with copper tube - nothing new here.
Some builders are suggesting that by using wire (which lowers the Q of the loop) to increases the bandwidth has no effect on performance - Yes it does! Others state that the voltage across the capacitor is less then the experts predicted - Well yes! But I find this applies to most loops.
It is well known that as the Q of a loop goes down due to loss, the voltage across the capacitor is less. A wire loop has a lower Q than a copper loop of the same size (more loss) so the voltage across the tuning capacitor in a wire loop will be even less again - nothing new here.
You can trade efficiency / Q for bandwidth in a ¼ loop and it will still work - Just don't get confused when reading about ¼ wave length wire loops and some new loop theory (because of incorrectly applying small loop data to large wire loops) and then try and transfer that information to a true small loop - you will come unstuck!
People have tried to adjust the matching system with the loop horizontal and just above the ground, don't do this! The induced Loss and resultant very low Q will make any adjustments useless or impossible. The aerial should be placed vertically and at least one loop diameters away from any structures that could absorb RF during the test. Don't let anyone tell you that feeding or adjusting a correctly made loop is difficult.
If the resonant SWR changes radically between bands (cannot be bought down by resonating the loop) then look for something nearby including very poor ground which could be inducing loss into the loop and causing the loop impedance to change with frequency. A loop mounted in a good location, even one less than one loop diameter above a good ground will resonate and match to give almost a 1:1 SWR over the frequency range.
The Gamma match - This is an unbalanced feed system. There is a case for common mode currents. I have not found RF on the coax even under HI power in this installation with the short coax run straight down from the loop through the metal roof and with separate grounds for the loop and for the support mast. A separate earth should be used for the loop and for the mast in any case. It should be noted that the gamma match will skew the radiation pattern causing a front to back ratio of around 6dB favouring the direction of the of the gamma match mount.
I have not found induced noise on the feed line as the signal to noise ratio of my big copper loop is the same as a small shielded RX loop. It has been suggested that using a shielded coupling loop will also stop noise being picked up by the coupling loop. If there is that much impulse noise around then I feel the noise induced into the main resonant loop would swamp anything induced into the small coupling loop.
Note: I am not talking about noise induced into the feed line which can be reduced by using loop coupling to preserve the balance of the feed system and the main loop. - In my situation I have not encountered this problem.
Update - I have just tried a shielded coupling loop. There was absolutely no change to the noise level. I rotated the antenna and increased power line noise above S9. Once again there was no difference between the gamma match and the shielded loop coupling. The matching is hard to adjust without making the coupling loop completely self supporting. The base of my antenna is 2m above the roof making the location and shape of the coupling loop difficult to reach or adjust from this position. I was unable to get the SWR below 1.8:1 on 40m in the time spent testing, although I had 80m and 160m close to 1.1. I will retest when I can do meaningful far-field signal strength readings. Someone had reported a drop in output with loop coupling unless you get the size just right - apparently you can get the SWR correct but still have less output if everything is not set up correctly - I will be able to test this and also plot any change in radiation pattern in the near future.
Construction - Wire versus Copper - Some loop builders have said that they didn't note any difference between a thick copper loop and a thinner wire loop. Other builders claim their wire loop design works as well as a copper loop. Why? Well first a quick refresh:
Radiation resistance (Rr) is that part of an antenna's feed point resistance that is caused by the radiation of electromagnetic waves from the antenna. The radiation resistance is determined by the geometry of the antenna, not by the materials of which it is made. It can be viewed as the equivalent resistance to a resistor in the same circuit. Loss resistance (Rloss) is in series with Rr and consumes/wastes power - Rloss includes capacitor, wire, component and construction losses. Then there are environmental losses caused by ground and other induction losses which all conspire to lower the efficiency of any aerial.
The first thing to look at is the construction of the loop, particularly the connections to the tuning capacitor and the type of capacitor used.
If you introduce any loss with the capacitor or the loop construction then there will be very little difference between a copper and a wire loop.
This is why some builders in their haste to get a small copper loop built underestimate the loss caused by construction, components and location and then find that the wire loop they built seems to perform as well as the copper version they tried.
The radiation resistance (Rr) is around 0.005 Ohms in a 3.6m diamiater copper loop at 1.8Mhz. Any loss in any connection, like short coax braid connections to the cap will destroy the loop Q / efficiency by introducing resistive losses (Rloss) that are equal to (if you are lucky) or greater than (99% of the time) the radiation resistance of the loop. A capacitor with low Q and / or contact resistive loss will have the same effect.
Most people can't seem to get their heads around such low values of Rr and the high current flowing in a loop - even with low power - and they completely underestimate the loss in materials and construction techniques that they employ.
Good builders know that any aerial designed for portable use will have some compromises and they state this in their construction projects like
Alex PY1AHD - These portable loops should not be compared in preformance to a much bigger loop or dipole as some writers and builders do, why? They should rightly be compared to other portable aerials and in that case the loop, if built correctly, is usually superior.
Stated again - Any resistive loss you introduce with your construction and components could be in series with 0.005 Ohms or less of Rr.
The ¼ wave lenght loop -The second case is a loop that is a ¼ wave length in circumference. This loop length will work when made of wire.
It will not be as efficient as a ¼ wave copper loop (unless the proponents of such loops can tell me how introducing a resistive loss (wire) in series with the radiation resistance of the loop can increase current flow through the loop into the radiation resistance) but even with a loss of 50% (1/2 an S point) over a copper loop you will not notice the difference on air some of the time. This is an easy loop to try and it works.
A few loop builders are confusing radiation resistance (Rr) with feed-point resistance. Because of the low value of Rr in a small loop antenna the conductor loss is a significant part of the total resistance. In a large dipole or vertical antenna the feed-point resistance may be almost equal to
the high value of radiation resistance. In a small loop the feed-point resistance must include the significant value of conductor loss resistance.
Experts claim the current is uniform around a loop - No they don't. It has been stated again and again that the current is ONLY uniform in very small loops - NOT a ¼ wave length loop. (ARRL Antenna handbook 18 years ago) - And again a 1/4 wave loop does not have tiny values of Rr when compared to a small loop, so you don't have to build a ¼ wave loop with copper tube - nothing new here.
Some builders are suggesting that by using wire (which lowers the Q of the loop) to increases the bandwidth has no effect on performance - Yes it does! Others state that the voltage across the capacitor is less then the experts predicted - Well yes! But I find this applies to most loops.
It is well known that as the Q of a loop goes down due to loss, the voltage across the capacitor is less. A wire loop has a lower Q than a copper loop of the same size (more loss) so the voltage across the tuning capacitor in a wire loop will be even less again - nothing new here.
You can trade efficiency / Q for bandwidth in a ¼ loop and it will still work - Just don't get confused when reading about ¼ wave length wire loops and some new loop theory (because of incorrectly applying small loop data to large wire loops) and then try and transfer that information to a true small loop - you will come unstuck!
Small loop calculations are being revised and show higher Rr and more efficiency then previously thought.
From VK5KLT Leigh Turner - All of the software programs and formula under-estimate the loop's true radiation resistance; sometimes by huge discrepancies! This is because the traditional equations relied upon in small loop parameter calculators do not factor in all of the complex modes that contribute to EM field creation and radiation. My good friend Prof. Mike Underhill, G3LHZ and his graduate students have done much valuable empirical work in this important area trying to characterize and better understand these modes and why the small loop is so inexplicably efficient on the low bands. This excellent efficacy goes against loop folklore and traditional wisdom and has therefore created a controversy amongst certain antenna experts!
Again on Capacitor voltage: Generally the radiation Q and associated capacitor voltage is significantly lower in practice than that predicted by most calculators due to their underestimate of Rr, i.e. the Q is lower because additional EM radiation is occurring. - Thanks Leigh.
If we assume a good ¼ wave copper loop at 40m has an Rr of about 1 Ohm + Rloss of about 0.12 Ohm or so at 7 MHz (conventional calculations) then with 400W of power applied, the loop current will be: I = Sqrt P/R = (400/1.12)1/2 = about 19 amperes. With 1000W of drive, the loop current will be about 30 amperes, on 160m the loop current would be almost 90 amperes.
Do you think that the surface area / skin depth of 38 feet of thin wire can pass a 20A to 90A current with less resistive loss and less loss via
skin effect as effectively as 38 feet of 3/4" copper tube? Remember this is RF, the skin effect and not just the DC resistance is a major factor.
If you make the loop out of wire, the wire loss will be in series with Rr so less current will go into Rr and more will be wasted in Rloss.
Again - The radiation resistance (Rr) is determined by the geometry of the antenna, not by the materials of which it is made.
It is all about tradeoffs - You trade one of the following: 1 Small-size, 2 High-efficiency, 3 Wide-bandwidth. When you build an antenna you can have any two of the above but not all three. My 3.6m dia copper loop is a compromise. It would have been more efficient (but less bandwidth) if larger diameter copper had been used, but how far do you go? If you are building your first loop or trying a different construction then you need to question everything you read - including everything written here - listen to both sides, read other opinions go back and look at basic antenna and electromagnetic laws and then make an informed decision on what you want to build and what compromises you are willing to make.
I note that some wire loop sites now say that a wire loop is a compromise and does not produce the radiated field strength of a copper loop.
It also could be said that a ¼ wave copper loop is a compromise compared to a ¼ wave wire loop with respect to bandwidth and cost.
However the cost of multiple wire loops to cover the bands that one copper loop can operate over would in the end, be more expensive. Unless you build a 1/4 wave 128 foot wire loop for 160m and operate it as a halfwave loop on 80m and a fullwave loop on 40m - remembering that the loop operation modes completely changes at a 1/2 and 1 wavelength and the aerial is no longer a small magnetic loop aerial.
Wide Bandwidth - Again because of the resistive losses lowering the Q and therefore the efficiency of the loop, a wire loop does indeed exhibit
a wider bandwidth than a copper loop. But they trade away some of the main advantages of a copper loop. These are the ability to rotate the loop and null local interference (unless the wire is mounted on a frame - so why not use copper tube instead). Don't underestimate this ability.
In a good loop it really does work well, with nulls up to 25db attainable. A ¼ wave wire loop will not work on a lower frequency than the one it was cut for without horrendous loss. A wire loop is basically (not always) a mono band aerial; a correctly made copper loop is a multi band aerial.
Narrow Bandwidth - Think again before trading bandwidth. Narrow bandwidth it is not a liability as some would suggest. If the loop tuning unit
is correctly designed then retuning is quick and easy. It is an advantage in noisy band conditions, in the presence of strong nearby stations and from receiver frontend overload. It reduces harmonic radiation and can help reduce your bandwidth on crowed bands. There is usually no need to retune when just listening across the band, only when you decide to change your transmit frequency. On 40m and 80m this takes from 1 to 4 seconds.
From VK5KLT Leigh Turner - All of the software programs and formula under-estimate the loop's true radiation resistance; sometimes by huge discrepancies! This is because the traditional equations relied upon in small loop parameter calculators do not factor in all of the complex modes that contribute to EM field creation and radiation. My good friend Prof. Mike Underhill, G3LHZ and his graduate students have done much valuable empirical work in this important area trying to characterize and better understand these modes and why the small loop is so inexplicably efficient on the low bands. This excellent efficacy goes against loop folklore and traditional wisdom and has therefore created a controversy amongst certain antenna experts!
Again on Capacitor voltage: Generally the radiation Q and associated capacitor voltage is significantly lower in practice than that predicted by most calculators due to their underestimate of Rr, i.e. the Q is lower because additional EM radiation is occurring. - Thanks Leigh.
If we assume a good ¼ wave copper loop at 40m has an Rr of about 1 Ohm + Rloss of about 0.12 Ohm or so at 7 MHz (conventional calculations) then with 400W of power applied, the loop current will be: I = Sqrt P/R = (400/1.12)1/2 = about 19 amperes. With 1000W of drive, the loop current will be about 30 amperes, on 160m the loop current would be almost 90 amperes.
Do you think that the surface area / skin depth of 38 feet of thin wire can pass a 20A to 90A current with less resistive loss and less loss via
skin effect as effectively as 38 feet of 3/4" copper tube? Remember this is RF, the skin effect and not just the DC resistance is a major factor.
If you make the loop out of wire, the wire loss will be in series with Rr so less current will go into Rr and more will be wasted in Rloss.
Again - The radiation resistance (Rr) is determined by the geometry of the antenna, not by the materials of which it is made.
It is all about tradeoffs - You trade one of the following: 1 Small-size, 2 High-efficiency, 3 Wide-bandwidth. When you build an antenna you can have any two of the above but not all three. My 3.6m dia copper loop is a compromise. It would have been more efficient (but less bandwidth) if larger diameter copper had been used, but how far do you go? If you are building your first loop or trying a different construction then you need to question everything you read - including everything written here - listen to both sides, read other opinions go back and look at basic antenna and electromagnetic laws and then make an informed decision on what you want to build and what compromises you are willing to make.
I note that some wire loop sites now say that a wire loop is a compromise and does not produce the radiated field strength of a copper loop.
It also could be said that a ¼ wave copper loop is a compromise compared to a ¼ wave wire loop with respect to bandwidth and cost.
However the cost of multiple wire loops to cover the bands that one copper loop can operate over would in the end, be more expensive. Unless you build a 1/4 wave 128 foot wire loop for 160m and operate it as a halfwave loop on 80m and a fullwave loop on 40m - remembering that the loop operation modes completely changes at a 1/2 and 1 wavelength and the aerial is no longer a small magnetic loop aerial.
Wide Bandwidth - Again because of the resistive losses lowering the Q and therefore the efficiency of the loop, a wire loop does indeed exhibit
a wider bandwidth than a copper loop. But they trade away some of the main advantages of a copper loop. These are the ability to rotate the loop and null local interference (unless the wire is mounted on a frame - so why not use copper tube instead). Don't underestimate this ability.
In a good loop it really does work well, with nulls up to 25db attainable. A ¼ wave wire loop will not work on a lower frequency than the one it was cut for without horrendous loss. A wire loop is basically (not always) a mono band aerial; a correctly made copper loop is a multi band aerial.
Narrow Bandwidth - Think again before trading bandwidth. Narrow bandwidth it is not a liability as some would suggest. If the loop tuning unit
is correctly designed then retuning is quick and easy. It is an advantage in noisy band conditions, in the presence of strong nearby stations and from receiver frontend overload. It reduces harmonic radiation and can help reduce your bandwidth on crowed bands. There is usually no need to retune when just listening across the band, only when you decide to change your transmit frequency. On 40m and 80m this takes from 1 to 4 seconds.
Click Image to
enlarge
In Practise - The Loop in use and information that matches my real world findings.