A review of my NEC data will not show such...
The NEC current across the 71.7 ohm resistor at the center terminals of the receive antenna is 0.015641 amperes.
P = (0.015641)^2 * 71.7 = 0.017541 watts = ~17.5 mW.
Your EZNEC evaluation should show about the same, if your model setup is similar.
//
The main limiting factor of the demo version of EZNEC is that it allows the use of a maximum of only 20 segments. This is not enough for complex antenna structures, like five-element Yagis, but it enough for getting good results with a simple system consisting of two dipoles.
I used 18 segments, 9 for each dipole. I used an odd number of segments in each dipole to get the source in the center for the transmitting dipole, and to be able to use the current of the center segment (segment 5) in the receiving dipole as its short-circuit current.
To get resonance, the dipole length was slightly shorter than a half wavelength. I adjusted the dipole length for minimum reactance (a couple of milliohms) in the input impedance. The input resistance was slightly higher than 73 ohms. I used the same length for both dipoles. I applied 270 V to the transmitting dipole, which gave me a radiated power of 1001 watts. The dipoles were at the same height, parallel, and separated by 3000 meters.
I got 96.65 mA in the center segnent of the receiving dipole. I did not place any load resistor in the model of the receiving dipole. This current corresponds to an open-circuit voltage of .09665 X 73 = 7.05545 V. The power to a 73 ohm load would be [(7.05545)^2]/[(4)(73)] = 170.5 mW. This is close to the result using Kraus.
Since my results are close to Kraus, I think they are at least roughly correct. I'm sure that if you check over your NEC model you'll get about the same results I did.
Additional Note:
To make my NEC model as close to yours as possible, I added a 73 ohm load resistor to the center of the receiver dipole. Now, the center segment of the receiver dipole has a current of 48.27 mA. This results in the power applied to the resistor being [(.04827)^2] X 73 = 170.1 mW. This is close to what I obtained above. Adding the load resistor, which is close to the receiver dipole resistance, causes the current in the center segment to be about half of the short-circuit current.
The main limiting factor of the demo version of EZNEC is that it allows the use of a maximum of only 20 segments. This is not enough for complex antenna structures, like five-element Yagis, but it enough for getting good results with a simple system consisting of two dipoles.
I used 18 segments, 9 for each dipole. I used an odd number of segments in each dipole to get the source in the center for the transmitting dipole, and to be able to use the current of the center segment (segment 5) in the receiving dipole as its short-circuit current.
To get resonance, the dipole length was slightly shorter than a half wavelength. I adjusted the dipole length for minimum reactance (a couple of milliohms) in the input impedance. The input resistance was slightly higher than 73 ohms. I used the same length for both dipoles. I applied 270 V to the transmitting dipole, which gave me a radiated power of 1001 watts. The dipoles were at the same height, parallel, and separated by 3000 meters.
I got 96.65 mA in the center segnent of the receiving dipole. I did not place any load resistor in the model of the receiving dipole. This current corresponds to an open-circuit voltage of .09665 X 73 = 7.05545 V. The power to a 73 ohm load would be [(7.05545)^2]/[(4)(73)] = 170.5 mW. This is close to the result using Kraus.
Since my results are close to Kraus, I think they are at least roughly correct. I'm sure that if you check over your NEC model you'll get about the same results I did.
Additional Note:
To make my NEC model as close to yours as possible, I added a 73 ohm load resistor to the center of the receiver dipole. Now, the center segment of the receiver dipole has a current of 48.27 mA. This results in the power applied to the resistor being [(.04827)^2] X 73 = 170.1 mW. This is close to what I obtained above. Adding the load resistor, which is close to the receiver dipole resistance, causes the current in the center segment to be about half of the short-circuit current.
I checked my (non-EZNEC) NEC model carefully, and don't see any errors. The NEC average gain test shows it to be highly accurate. The gain, impedance and pattern of the transmit antenna are "textbook."
But when I use my "other means" to calculate the power in a 73 ohm resistor at the center of the receive dipole for these conditions, that gives me about 170 mW, the same as you report.
So I don't know what is going on with it at this point.
//
I checked my (non-EZNEC) NEC model carefully, and don't see any errors. The NEC average gain test shows it to be highly accurate. The gain, impedance and pattern of the transmit antenna are "textbook."
But when I use my "other means" to calculate the power in a 73 ohm resistor at the center of the receive dipole for these conditions, that gives me about 170 mW, the same as you report.
So I don't know what is going on with it at this point.
//
It's too bad that we can't compare our simulation results, since we then can't compare our results for two monopoles over a ground plane, which is the whole point of our discussion. With the two monopoles separated by 3000 meters, and applying 1 kW to the transmitting monopole, knowing how much power is applied to a 36.5 ohm load resistor will give the answer about whether there is a 6 dB failure in reciprocity for surface waves when a vertical monopole over ground is used. If the receiver produces about 170 mW in the 36.5 ohm resistor, the same power as for a pair of dipoles, it demonstrates that the failure in reciprocity is real, since, with a 3 dB gain over a dipole in both the transmitting and recieving dipoles, the maximum power available to a resistive load at the receiving antenna output should be about 4 X 170 = 680 mW.
I have a suggestion for troubleshooting this discrepancy: Why don't you download the EZNEC demo program from www.eznec.com? It's free, of course. By setting up the problem on EZNEC, you will see whether the difference in results is the particular NEC program, or the way the problem was set up. The "cardioid " antenna in the EZNEC antenna library, which comes with the demo program, can be easily modified to two monopoles over ground or two dipoles in free space. The cardioid antenna consists of two transmitting monopoles over ground. One of of the transitters can be converted into a receiver by setting the RF source to the voltage mode, and setting the voltage to zero. The two monopoles over ground can be converted to two dipoles by setting the ground description to "free space," and extending the former monopoles below where the ground plane was by the same length as the monopoles. If you get the same result as you got with your NEC-2 program, you will have to check how the problem was set up. If you get about 170 mW at the receiving dipole load of 73 ohms, there must be a difference in the two types of NEC programs.
It's too bad that we can't compare our simulation results, since we then can't compare our results for two monopoles over a ground plane, which is the whole point of our discussion. With the two monopoles separated by 3000 meters, and applying 1 kW to the transmitting monopole, knowing how much power is applied to a 36.5 ohm load resistor will give the answer about whether there is a 6 dB failure in reciprocity for surface waves when a vertical monopole over ground is used. If the receiver produces about 170 mW in the 36.5 ohm resistor, the same power as for a pair of dipoles, it demonstrates that the failure in reciprocity is real, since, with a 3 dB gain over a dipole in both the transmitting and recieving dipoles, the maximum power available to a resistive load at the receiving antenna output should be about 4 X 170 = 680 mW.
I have a suggestion for troubleshooting this discrepancy: Why don't you download the EZNEC demo program from www.eznec.com? It's free, of course. By setting up the problem on EZNEC, you will see whether the difference in results is the particular NEC program, or the way the problem was set up. The "cardioid " antenna in the EZNEC antenna library, which comes with the demo program, can be easily modified to two monopoles over ground or two dipoles in free space. The cardioid antenna consists of two transmitting monopoles over ground. One of of the transitters can be converted into a receiver by setting the RF source to the voltage mode, and setting the voltage to zero. The two monopoles over ground can be converted to two dipoles by setting the ground description to "free space," and extending the former monopoles below where the ground plane was by the same length as the monopoles. If you get the same result as you got with your NEC-2 program, you will have to check how the problem was set up. If you get about 170 mW at the receiving dipole load of 73 ohms, there must be a difference in the two types of NEC programs.
Ermi wrote: ...With the two monopoles separated by 3000 meters, and applying 1 kW to the transmitting monopole, knowing how much power is applied to a 36.5 ohm load resistor will give the answer about whether there is a 6 dB failure in reciprocity for surface waves when a vertical monopole over ground is used. If the receiver produces about 170 mW in the 36.5 ohm resistor, the same power as for a pair of dipoles, it demonstrates that the failure in reciprocity is real, since, with a 3 dB gain over a dipole in both the transmitting and recieving dipoles, the maximum power available to a resistive load at the receiving antenna output should be about 4 X 170 = 680 mW.
_________
Even if one is willing to accept an assertion that a receive monopole over a perfect ground plane in this complete r-f system receives NO ground reflections from/in the horizontal plane, such would reduce the overall gain of that path only by 3 dB, not the 6 dB you state, because ....
... the transmitting part of this dual monopole system unquestionably acquires the 3 dB gain benefit of its reflections from such a ground plane, which is predictable by radiation theory, and which has been proven in thousands of groundwave field strength measurements of AM broadcast stations just beyond their near-field radii, for many decades.
//
Ermi wrote: ...With the two monopoles separated by 3000 meters, and applying 1 kW to the transmitting monopole, knowing how much power is applied to a 36.5 ohm load resistor will give the answer about whether there is a 6 dB failure in reciprocity for surface waves when a vertical monopole over ground is used. If the receiver produces about 170 mW in the 36.5 ohm resistor, the same power as for a pair of dipoles, it demonstrates that the failure in reciprocity is real, since, with a 3 dB gain over a dipole in both the transmitting and recieving dipoles, the maximum power available to a resistive load at the receiving antenna output should be about 4 X 170 = 680 mW.
_________
Even if one is willing to accept an assertion that a receive monopole over a perfect ground plane in this complete r-f system receives NO ground reflections from/in the horizontal plane, such would reduce the overall gain of that path only by 3 dB, not the 6 dB you state, because ....
... the transmitting part of this dual monopole system unquestionably acquires the 3 dB gain benefit of its reflections from such a ground plane, which is predictable by radiation theory, and which has been proven in thousands of groundwave field strength measurements of AM broadcast stations just beyond their near-field radii, for many decades.
//
I tried to get Rich to discover the truth about the subject under discussion for himself by using his NEC program, but that did not happen. If Rich had gotten his NEC program to work correctly with the problem he had posed himself, he would have seen that, if two parallel half-wave dipoles are in free space, at the same height, operating at 1 MHz, separated by 3000 meters, with 1000 watts of radiated power from the first dipole, the maximum power available from the second dipole is about 170 mW. Values close to 170 mW were obtained by Kraus's method and Rich's own "other method." My own simulation with the EZNEC demo program also gave about 170 mW. I think that if Rich were to check the setup of his NEC-2 simulation, he would get 170 mW also. If he can't get this result, he should use another version of the NEC program.
After the simulation of the two dipoles, the next simulation to be performed with NEC is to find out how much is the maximum power available from a receiving quarter-wave monopole over a ground plane when a transmitting quarter-wave monopole over the same ground plane, separated from the receiving monopole by 3000 meters, radiates 1000 watts. I did the simulation with with my EZNEC demo program, and, as I expected, the maximum available power was about 170 mW, the same as for two half-wave dipoles in free space, separated by the same distance, and the transmitting dipole radiating 1000 watts.
I knew that two half-wave dipoles in free space would give the same maximum power transfer as two quarter-wave monopoles over a ground plane because that is a well-known theorem in antenna theory, which was proved many years ago. The earliest proof of this theorem that I know of is by Kenneth A. Norton, "System Loss in Radio Wave Propagation," Journal of Research of the National Bureau of Standards-D, Vol. 63D, pp. 53-73, 1959.
As I explained in my previous post, if two half-wave dipoles in free space and two monopoles over a ground plane have the same power transfer ratio for the same separation distance, this proves that the receiving monopole has 6 dB less gain than the transmitting monopole. It is known that the transmitting monopole has 3 dB gain over a dipole. To make the system gain for two monopoles over a ground plane equal to the system gain of two dipoles in free space, the receiving monopole must have 6 dB less gain than the transmitting monopole.
This phenomenon has been explained as the absence of ground reflection at the receiving monopole when receiving groundwaves, which causes the effective height of the receiving monopole to be half of what it would be if skywaves were received.
I have my own intuitive explanation of the phenomenon, which I have not seen stated in print before, but it is more satisfactory to me than the "missing reflection" explanation. I hope that it also provides a convincing explanation to others:
I is well-known that the ground plane causes a quarter-wave transmitting dipole to have 3 dB more gain than a half-wave transmitting dipole in free space. This is because all of the radiated power from the transmitting monopole must be above the ground plane. Similarly, all of the radiated power intercepted by the receiving monopole must be above the ground plane. Thus, a quarter-wave monopole above ground must have only half of the capture area of a half-wave dipole in free space. So, the ground plane causes a -3 dB gain in the receiving monopole compared to a receiving dipole.
To repeat, the ground plane causes a +3 dB gain (compared to a half-wave dipole in free space) in the transmitting monopole because all of the radiated power of the monopole must be above the ground plane. At the same time, the ground plane causes a -3dB gain in the receiving monopole because it cuts off half of the incident radiation applied to it, compared to a half wave dipole. The + 3 DB gain of the transmitting monopole compared to the dipole, and the - 3dB gain of the receiving monopole compared to a dipole, combine to cause the system gain of two dipoles to have the same system gain as two monopoles. Thus, the receiving monopole gain is 6 dB less than the transmitting monopole gain when groundwaves are received.
The many thousands of field strength measurements Rich mentions of monopole antennas generating groundwaves are not relevant to this discussion because the transmitting monopole gain is not in dispute. There is only a dispute about the receiving monopole gain, and only when receiving groundwaves, and not sky waves.
It's too bad that the antenna books mentioned do not deal with the interesting and well-established fact that reciprocity fails for receiving monopoles over ground that are intercepting groundwaves. I hope that Rich will be able to set up the dipole and monopole simulations on his NEC program so that he will be able to check out this fact for himself.
I tried to get Rich to discover the truth about the subject under discussion for himself by using his NEC program, but that did not happen. If Rich had gotten his NEC program to work correctly with the problem he had posed himself, he would have seen that, if two parallel half-wave dipoles are in free space, at the same height, operating at 1 MHz, separated by 3000 meters, with 1000 watts of radiated power from the first dipole, the maximum power available from the second dipole is about 170 mW. Values close to 170 mW were obtained by Kraus's method and Rich's own "other method." My own simulation with the EZNEC demo program also gave about 170 mW. I think that if Rich were to check the setup of his NEC-2 simulation, he would get 170 mW also. If he can't get this result, he should use another version of the NEC program.
After the simulation of the two dipoles, the next simulation to be performed with NEC is to find out how much is the maximum power available from a receiving quarter-wave monopole over a ground plane when a transmitting quarter-wave monopole over the same ground plane, separated from the receiving monopole by 3000 meters, radiates 1000 watts. I did the simulation with with my EZNEC demo program, and, as I expected, the maximum available power was about 170 mW, the same as for two half-wave dipoles in free space, separated by the same distance, and the transmitting dipole radiating 1000 watts.
I knew that two half-wave dipoles in free space would give the same maximum power transfer as two quarter-wave monopoles over a ground plane because that is a well-known theorem in antenna theory, which was proved many years ago. The earliest proof of this theorem that I know of is by Kenneth A. Norton, "System Loss in Radio Wave Propagation," Journal of Research of the National Bureau of Standards-D, Vol. 63D, pp. 53-73, 1959.
As I explained in my previous post, if two half-wave dipoles in free space and two monopoles over a ground plane have the same power transfer ratio for the same separation distance, this proves that the receiving monopole has 6 dB less gain than the transmitting monopole. It is known that the transmitting monopole has 3 dB gain over a dipole. To make the system gain for two monopoles over a ground plane equal to the system gain of two dipoles in free space, the receiving monopole must have 6 dB less gain than the transmitting monopole.
This phenomenon has been explained as the absence of ground reflection at the receiving monopole when receiving groundwaves, which causes the effective height of the receiving monopole to be half of what it would be if skywaves were received.
I have my own intuitive explanation of the phenomenon, which I have not seen stated in print before, but it is more satisfactory to me than the "missing reflection" explanation. I hope that it also provides a convincing explanation to others:
I is well-known that the ground plane causes a quarter-wave transmitting dipole to have 3 dB more gain than a half-wave transmitting dipole in free space. This is because all of the radiated power from the transmitting monopole must be above the ground plane. Similarly, all of the radiated power intercepted by the receiving monopole must be above the ground plane. Thus, a quarter-wave monopole above ground must have only half of the capture area of a half-wave dipole in free space. So, the ground plane causes a -3 dB gain in the receiving monopole compared to a receiving dipole.
To repeat, the ground plane causes a +3 dB gain (compared to a half-wave dipole in free space) in the transmitting monopole because all of the radiated power of the monopole must be above the ground plane. At the same time, the ground plane causes a -3dB gain in the receiving monopole because it cuts off half of the incident radiation applied to it, compared to a half wave dipole. The + 3 DB gain of the transmitting monopole compared to the dipole, and the - 3dB gain of the receiving monopole compared to a dipole, combine to cause the system gain of two dipoles to have the same system gain as two monopoles. Thus, the receiving monopole gain is 6 dB less than the transmitting monopole gain when groundwaves are received.
The many thousands of field strength measurements Rich mentions of monopole antennas generating groundwaves are not relevant to this discussion because the transmitting monopole gain is not in dispute. There is only a dispute about the receiving monopole gain, and only when receiving groundwaves, and not sky waves.
It's too bad that the antenna books mentioned do not deal with the interesting and well-established fact that reciprocity fails for receiving monopoles over ground that are intercepting groundwaves. I hope that Rich will be able to set up the dipole and monopole simulations on his NEC program so that he will be able to check out this fact for himself.
Ermi wrote: Thus, the receiving monopole gain is 6 dB less than the transmitting monopole gain when groundwaves are received. The many thousands of field strength measurements Rich mentions of monopole antennas generating groundwaves are not relevant to this discussion because the transmitting monopole gain is not in dispute.
By Ermi's analysis, the receive monopole has 3 dB less groundwave gain than a dipole in free space, not 6 dB. The additional 3 dB claimed in Ermi's 6 dB value is the result of the proven 3 dB gain of the monopole transmit antenna above that of a dipole in free space. So the gain of the transmit monopole is not irrelevant in my previous posts as stated in Ermi's quote above, or to this analysis.
Ermi has attributed the net gain/loss of this transmit-receive monopole system compared to dipoles in free space as belonging to the receive monopole antenna, alone.
As Ermi has stated that this gain difference for receive monopoles does not apply to received skywaves, maybe he will post his views on the gain values of the elevation pattern of the receive monopole in the region between zero degrees elevation and the elevation angle where he expects that reciprocity takes affect.
Regarding Ermi's comments about the effective height of the receive monopole I will repeat my statements from my earlier post.
Quoting from Section 11, paragraph 8 of Radio Engineers' Handbook, 1st edition, by F. E. Terman:
"...the relative response of the antenna to waves arriving from different directions is exactly the same as the relative radiation in different directions from the same antenna when acting as a radiator. Also like the antenna directivity, the effective height and the impedance of the antenna are the same in reception as in transmission. These reciprocal relations between transmission and reception properties make it possible to deduce the merits of a receiving antenna from transmission tests, and vice versa."
There is no qualification -- this applies to all forms of antennas.
As far as I have discovered, there is no authoritative reference stating and conclusively proving that reciprocity fails for monopole antennas -- even for the groundwave.
My regrets for introducing the system performance of two dipoles. The subject here is the reciprocity of a single monopole, and not how two monopoles compare to two free-space dipoles.
//
Ermi wrote: Thus, the receiving monopole gain is 6 dB less than the transmitting monopole gain when groundwaves are received. The many thousands of field strength measurements Rich mentions of monopole antennas generating groundwaves are not relevant to this discussion because the transmitting monopole gain is not in dispute.
By Ermi's analysis, the receive monopole has 3 dB less groundwave gain than a dipole in free space, not 6 dB. The additional 3 dB claimed in Ermi's 6 dB value is the result of the proven 3 dB gain of the monopole transmit antenna above that of a dipole in free space. So the gain of the transmit monopole is not irrelevant in my previous posts as stated in Ermi's quote above, or to this analysis.
Ermi has attributed the net gain/loss of this transmit-receive monopole system compared to dipoles in free space as belonging to the receive monopole antenna, alone.
As Ermi has stated that this gain difference for receive monopoles does not apply to received skywaves, maybe he will post his views on the gain values of the elevation pattern of the receive monopole in the region between zero degrees elevation and the elevation angle where he expects that reciprocity takes affect.
Regarding Ermi's comments about the effective height of the receive monopole I will repeat my statements from my earlier post.
Quoting from Section 11, paragraph 8 of Radio Engineers' Handbook, 1st edition, by F. E. Terman:
"...the relative response of the antenna to waves arriving from different directions is exactly the same as the relative radiation in different directions from the same antenna when acting as a radiator. Also like the antenna directivity, the effective height and the impedance of the antenna are the same in reception as in transmission. These reciprocal relations between transmission and reception properties make it possible to deduce the merits of a receiving antenna from transmission tests, and vice versa."
There is no qualification -- this applies to all forms of antennas.
As far as I have discovered, there is no authoritative reference stating and conclusively proving that reciprocity fails for monopole antennas -- even for the groundwave.
My regrets for introducing the system performance of two dipoles. The subject here is the reciprocity of a single monopole, and not how two monopoles compare to two free-space dipoles.
//
Just wanted to toss in kudos to Rich and Ermi Roos on this topic/discussion. A lot of interesting debate, and well presented on both sides. Sure, of necessity it may not *all* be 100% immediately applicable to the average part 15 hobby station. But the principles are fascinating and I wanted to pass compliments on this excellent tech debate in progress. Many good points raised and answered, and I've been finding it interesting.
Daniel
Just wanted to toss in kudos to Rich and Ermi Roos on this topic/discussion. A lot of interesting debate, and well presented on both sides. Sure, of necessity it may not *all* be 100% immediately applicable to the average part 15 hobby station. But the principles are fascinating and I wanted to pass compliments on this excellent tech debate in progress. Many good points raised and answered, and I've been finding it interesting.
Daniel