SWR has virtually nothing to do with antenna radiation efficiency. once again this is demonstrated best by terminating the load end of the feedline with a value of resistance that is identical to the Zo of the line, as in the case of an rf "dummy" load. it's further demonstrated by the fact that free space has an intrinsic impedance value of approximately 377 ohms. it's funny to see you all pontificating over your myths when the real problem presented by SWR in any system is the reduction in the amount of power delivered to the load, a problem that is easily addressed and remedied once it is realized that whenever the load does not equal the Zo (characteristic impedance) of the line that the line impedance is determined by a widely varying range of values of E/I all along the line which can be used to restore full power output from the (transmitter) source to the load. that's the real problem.
antenna efficiency is determined by BIRP. Balance, Impedance, Resonance and Pattern, only one of which has anything remotely to do with SWR. what has escaped most of you is the fact that once the impedance of the line becomes anything but Zo and if the values of E/I along the line are known that those points can be used to terminate the line and provide the necessary conjugate interference to match the output of the transmitter to a line input impedance that is anything but 50 ohms. this solves the problem created by the load mismatch presented at the other end of the line.
an SWR meter doesn't measure antenna efficiency any more than a flat SWR is a guarantee of optimum antenna performance. if you're feeding a balanced antenna with coaxial feedline (which destroys the pattern) and the antenna is not resonant at the frequency of choice then achieving a flat swr is meaningless. by the same token, antenna systems are floating around in space with swr present that would make any cb'er (and some hams) squeamish while the transmitters feeding them are delivering over 90+% of their available power to those same mismatched antennas.
the principles at work here are the same ones that require the francis line of antennas (under 8') to be used with specific lengths of differing types of feedline for best performance (and lowest swr) when used singly and the same principles involved in a situation that arose recently when one or two customers called me complaining that their PowerStikII antennas (which matched well in their previous tractor rigs) were not seeing the usual low swr in their new Freightliner XL tractors with the original 18' lines. quite the contrary, the swr has risen to 3:1 when relocated to the new tractors. one of the drivers with which i have spent time on several occasions explaining to him how antennas and feedlines operate, particularly under mismatch conditions, decided to take my advice and replace the 18' RG8M lines with 9' RG58 lines and was happy to see that the swr (as measured from the line input) dropped from over 3:1 to just over 1:1 when the 9' lines of RG58 were installed.
in a line with a Zo of 50 ohms and a load that represents anything but 50 ohms the differing readings that are seen using the swr meter when placed in various spots in the line are only fooling you if you don't know what they're trying to tell you and specifically what that is is that since VSWR=EMAX/EMIN what it's trying to tell you is that the impedance along the line varies with the placement location of the bridge in the line, which makes absolute sense since in an unmatched line impedance (Z=E/I) changes from one spot in the line to the next.
the only time a feedline with a Zo of 50 ohms is exactly that occurs only under two sets of conditions: either the transmitter, (source) the antenna (load) and the feedline are all presenting an impedance of 50 ohms or the feedline is unterminated at either end in which case the characteristic impedance of the feedline is solely dependent on its physical characteristics only, length not being one of them. whenever a source and a load are attached where either deviates from the Zo of the line, then and only then does the range of dynamic impedance values appear on the line.
Z (Impedance) = E (Voltage) / (Divided By) I (Current)
for example, given that E and I measured at a particular spot in the line are both equal to 50 volts and 1 amp respectively, the impedance at that point in the line is 50 ohms. i can assure you this is only the case where both the source and the load are both equal to the Zo (characteristic impedance) of the line.
where it is otherwise the line reverts to a rather wide ranging set of values that are easily visible at *multiple spots along the line, especially if that line is *longer than 180 degrees at the operating frequency in question. real resistance values are known to be present minimally anywhere from 33 - 78 ohms wholly dependent on the length of the line for absolute terminating values.
as to the question of the tuned 1/2 wave line. do you want to use it? if you do then know this. when you do the exact values of resistance and reactance that are present at the feedpoint of the antenna will be mirrored almost exactly at the transmitter input to the line. that's not a bad idea if you can measure and know that the resistance at the feedpoint is + or - 50 ohms and the reactance is near a zero net value. if they're say 33 ohms and j-15 ohms capacitively reactive you have to ask yourself a question. do you really want those exact conditions to exist at the transmitter? only if the transmitter is equipped with its own adjustable PA matching circuit. with a fixed 50 ohm transmitter output and 33 j-15 at the antenna the last thing you want to use is a tuned 1/2 wave line, unless you want reduced transmitter power output or maybe you just want to be able to see the conditions at the antenna feedpoint as it may be inconvenient to make measurements at the antenna for one reason or another. if the transmitter is fixed 50 ohm output then you may want to consider the fact that what's going on with the match between the transmitter and the line might be of equal if not somewhat more concern since it can be managed from either end of the line anyway. that's what the AT in the nomenclature in TS850SAT does. it's not really an antenna tuner per se since what it really does is makes sure that the transmitter is always working into as close to a 50 ohm line input as possible at all times, regardless of the mismatch at the load
up to a certain point. ranges of matching values achieved by these "antenna tuners" are anywhere from 16 - 150 ohms and wider. the same thing to a lesser extent (33 - 78) can easily be done by knowing where these values of impedance can be found along an unmatched line under similar conditions.
every length of feedline is a matching transformer....Cebik.
Yes, you can see a "difference" in the readings. "Yes", there may be some cases where doing so is necessary. But! It isn't as important as some of us make it out to be! 