swr measurements taken at the transmitter output aren't necessarily erroneous and Rich is right, all reflected power is not lost.
when presented with anything but 50 + j0 ohms at the load we can uncover some vital information if we use an electrically tuned 1/2 line between the transmitter and the load. since the property of this line is such that it repeats the conditions present at the load directly at the transmitter every 180 degrees then we can make this statement:
conditions at the load are duplicated at the transmitter input. with this as a given it's easy to understand that as the impedance at the transmitter is changed the power output is reduced due to the mismatch now existing between the transmitter and the feedline input, to the same degree as conditions existing at the (antenna) load input. the reduction of power delivered by the transmitter to the line is exactly the same amount as the power reflected at the load or to put it another way, the reflection loss at the load can be referred directly back along the line to the transmitter.
so in an electrically tuned 1/2 line with negligible loss if the swr measured at the load is 2:1 then the swr presented at the transmitter input is also 2:1 and values of reactance at both the load and the transmitter input are for the most part identical.
now if we direct our attention to the transmitter mismatch and correct it then the transmitter once again delivers the maximum available matched power towards the load. now the only loss in the system is due to the reflection mismatch presented at the (antenna) load only and its effects at the transmitter input have been cancelled. we have successfully prevented the reflected power from travelling beyond the match point to the transmitter and once again the feedline between the transmitter and the match point re-assumes its characteristic value of Z=50.
remember, under the conditions presented prior to correcting the mismatch between the transmitter and the line input the impedance of the line is no longer the "characteristic" 50 ohms. the impedance of the line has changed from Zc to Z = E/I all along the line.
so if you have a 2:1 mismatch at the antenna the transmitter makes 11% less power available to the load in addition to the simple reflection loss. it doesn't matter to the transmitter if you correct the mismatch there or at the load, the results are the same. if the transmitter can deliver maximum available matched power to the load the antenna will radiate all the power fed to it. any reflected power at that point is of the non-dissipative type and is all eventually absorbed and radiated by the load.
i strongly suspect that there is a reciprocating effect between both ends of a tuned 1/2 wave line under these conditions, (or any random length of line less the identical numbers to a great extent) in other words, if the swr/impedance/reactance presented at the feedpoint of a mismatched load repeats itself identically to the transmitter input then my question is, if we correct the match between the transmitter and the line input is that corrected match repeated back towards the antenna input in like manner?
if Z = E/I in an improperly terminated line then the answer is yes. then it follows that a (antenna) load could be matched well over a wide range of operating frequencies by simply controlling the match point at the transmitter so that it always sees as close to 50 + j0 as possible regardless of what frequency the transmitter is operating at.
if we work with only the real resistance and ignore any values of j for the purpose of example then the value of R required to create a conjugate match between a 50 ohm transmitter and a 2:1 swr representing 25 ohms is easily determined........
if we multiply the two impedance values and take the square root of the product we end up with 35.355 ohms as the value of R required for the correcting conjugate match, sans reactance.
Mismatch
when presented with anything but 50 + j0 ohms at the load we can uncover some vital information if we use an electrically tuned 1/2 line between the transmitter and the load. since the property of this line is such that it repeats the conditions present at the load directly at the transmitter every 180 degrees then we can make this statement:
conditions at the load are duplicated at the transmitter input. with this as a given it's easy to understand that as the impedance at the transmitter is changed the power output is reduced due to the mismatch now existing between the transmitter and the feedline input, to the same degree as conditions existing at the (antenna) load input. the reduction of power delivered by the transmitter to the line is exactly the same amount as the power reflected at the load or to put it another way, the reflection loss at the load can be referred directly back along the line to the transmitter.
so in an electrically tuned 1/2 line with negligible loss if the swr measured at the load is 2:1 then the swr presented at the transmitter input is also 2:1 and values of reactance at both the load and the transmitter input are for the most part identical.
now if we direct our attention to the transmitter mismatch and correct it then the transmitter once again delivers the maximum available matched power towards the load. now the only loss in the system is due to the reflection mismatch presented at the (antenna) load only and its effects at the transmitter input have been cancelled. we have successfully prevented the reflected power from travelling beyond the match point to the transmitter and once again the feedline between the transmitter and the match point re-assumes its characteristic value of Z=50.
remember, under the conditions presented prior to correcting the mismatch between the transmitter and the line input the impedance of the line is no longer the "characteristic" 50 ohms. the impedance of the line has changed from Zc to Z = E/I all along the line.
so if you have a 2:1 mismatch at the antenna the transmitter makes 11% less power available to the load in addition to the simple reflection loss. it doesn't matter to the transmitter if you correct the mismatch there or at the load, the results are the same. if the transmitter can deliver maximum available matched power to the load the antenna will radiate all the power fed to it. any reflected power at that point is of the non-dissipative type and is all eventually absorbed and radiated by the load.
i strongly suspect that there is a reciprocating effect between both ends of a tuned 1/2 wave line under these conditions, (or any random length of line less the identical numbers to a great extent) in other words, if the swr/impedance/reactance presented at the feedpoint of a mismatched load repeats itself identically to the transmitter input then my question is, if we correct the match between the transmitter and the line input is that corrected match repeated back towards the antenna input in like manner?
if Z = E/I in an improperly terminated line then the answer is yes. then it follows that a (antenna) load could be matched well over a wide range of operating frequencies by simply controlling the match point at the transmitter so that it always sees as close to 50 + j0 as possible regardless of what frequency the transmitter is operating at.
if we work with only the real resistance and ignore any values of j for the purpose of example then the value of R required to create a conjugate match between a 50 ohm transmitter and a 2:1 swr representing 25 ohms is easily determined........
if we multiply the two impedance values and take the square root of the product we end up with 35.355 ohms as the value of R required for the correcting conjugate match, sans reactance.
Mismatch
