Home |
Search |
Today's Posts |
|
#1
|
|||
|
|||
Computer model experiment
On May 10, 5:26*pm, K1TTT wrote:
On May 10, 6:49*pm, Art Unwin wrote: On May 10, 1:05*pm, Richard Fry wrote: On May 10, 12:35*pm, Art Unwin wrote: * .... The radiation was 35 db in a shape close to that of a sphere. (when the resistance of the aluminum dipole went to zero the radiation went to a perfect sphere) The radiation was "35 db" compared to what reference value? BTW, a single, linear radiator cannot generate a perfectly spherical radiation pattern, no matter what your model tells you. Even an "infinitesimally" short, center-fed linear dipole has a figure 8 radiation pattern with a directivity (gain) of 1.5 X, or 1.76 dBi -- see any antenna engineering textbook. RF I believe the computer programs to be more up to date than the books! There certainly have been more advances since they have come into being. The programs reflect Maxwells equations which support the presence of particles which is what provide the radiation resistance and not the dipole itself. The dipole will show a donut pattern that will gradually deform to a perfect sphere when resistance drops to zero as per Poynting. I would also point out that the programs support the presence of Gaussian static particles as does mathematics. I would imagine that no matter what programs you decide to use you will get the same results as you increase the element diameter until the impedance is zero.No point in trashing computer programs in advance because of personal intuition. All I have done is removing resistance losses that do not contribute to radiation. the programs are based on the books... but even worse, they are digital approximations of the continuous formulas and as such are not completely accurate. *this is especially true when extremely large or small numbers are used or there are a large number of additions done, as is common in antenna modeling programs. *there are also assumptions made in the development of most of those programs that are often not stated to, or not understood by, the user, such as you. *so when you set something to optimize forever or start making elements extremely skinny, fat, short, or long, or too close together, you are most likely going to get wrong, or physically unrealizable results. Obviously you are very experienced in generating and bug catching in antenna programs having large experiences of finding antenna errors. What exactly in the nature of antenna computer programs, which have been around for some time now, have you found them to be suspect ? In my case the program verified what mathematics show as the presence of particles on the surface and where the total input forces were used for particle propagation. Now I am aware you have taken the position that particles are not involved in radiation and thus you will resist what computer programs arrive at relying on your intuition at all times which requires no personal experience on the subject However, I am taking the program that I purchased on trust especially when it follows the maxwell equations and where I am not adverse to change. I look forward to specific examples that buttress your thoughts in a scientific manner so I may decide what to do with my program purchase. May I recommend you do the same thing with the program of your choice where you can specifically point to the areas of error where they do not meet your expectations. Why not do the same with EZNEC so Roy can learn from your personal experiences and intuitions and institute the appropriate corrections. Never mind the length of the dipole just make the diameter very very fat and see what EZNEC does. |
#2
|
|||
|
|||
Computer model experiment
On May 10, 7:04*pm, Art Unwin wrote:
On May 10, 5:26*pm, K1TTT wrote: On May 10, 6:49*pm, Art Unwin wrote: On May 10, 1:05*pm, Richard Fry wrote: On May 10, 12:35*pm, Art Unwin wrote: * .... The radiation was 35 db in a shape close to that of a sphere. (when the resistance of the aluminum dipole went to zero the radiation went to a perfect sphere) The radiation was "35 db" compared to what reference value? BTW, a single, linear radiator cannot generate a perfectly spherical radiation pattern, no matter what your model tells you. Even an "infinitesimally" short, center-fed linear dipole has a figure 8 radiation pattern with a directivity (gain) of 1.5 X, or 1.76 dBi -- see any antenna engineering textbook. RF I believe the computer programs to be more up to date than the books! There certainly have been more advances since they have come into being. The programs reflect Maxwells equations which support the presence of particles which is what provide the radiation resistance and not the dipole itself. The dipole will show a donut pattern that will gradually deform to a perfect sphere when resistance drops to zero as per Poynting. I would also point out that the programs support the presence of Gaussian static particles as does mathematics. I would imagine that no matter what programs you decide to use you will get the same results as you increase the element diameter until the impedance is zero.No point in trashing computer programs in advance because of personal intuition. All I have done is removing resistance losses that do not contribute to radiation. the programs are based on the books... but even worse, they are digital approximations of the continuous formulas and as such are not completely accurate. *this is especially true when extremely large or small numbers are used or there are a large number of additions done, as is common in antenna modeling programs. *there are also assumptions made in the development of most of those programs that are often not stated to, or not understood by, the user, such as you. *so when you set something to optimize forever or start making elements extremely skinny, fat, short, or long, or too close together, you are most likely going to get wrong, or physically unrealizable results. Obviously you are very experienced in generating and bug catching in antenna programs having large experiences of finding antenna errors. What exactly in the nature of antenna computer programs, which have been around for some time now, have you found them to be suspect ? In my case the program verified what mathematics show as the presence of particles on the surface and where the total input forces were used for particle propagation. Now I am aware you have taken the position that particles are not involved in radiation and thus you will resist what computer programs arrive at relying on your intuition at all times which requires no personal experience on the subject However, I am taking the program that I purchased on trust especially when it follows the maxwell equations and where I am not adverse to change. I look forward to specific examples that buttress your thoughts in a scientific manner so I may decide what to do with my program purchase. May I recommend you do the same thing with the program of your choice where you can specifically point to the areas of error where they do not meet your expectations. Why not do the same with EZNEC so Roy can learn from your personal experiences and intuitions and institute the appropriate corrections. Never mind the length of the dipole just make the diameter very very fat and see what EZNEC does. Groan... Let me tell you the story about 24 dbi gain dipoles... Simple to model.. Then again, no, it's a futile waste of time trying to convince you of the error of your ways.. :/ Continue with fantasy hour... :/ |
#3
|
|||
|
|||
Computer model experiment
On May 10, 7:10*pm, wrote:
On May 10, 7:04*pm, Art Unwin wrote: On May 10, 5:26*pm, K1TTT wrote: On May 10, 6:49*pm, Art Unwin wrote: On May 10, 1:05*pm, Richard Fry wrote: On May 10, 12:35*pm, Art Unwin wrote: * .... The radiation was 35 db in a shape close to that of a sphere. (when the resistance of the aluminum dipole went to zero the radiation went to a perfect sphere) The radiation was "35 db" compared to what reference value? BTW, a single, linear radiator cannot generate a perfectly spherical radiation pattern, no matter what your model tells you. Even an "infinitesimally" short, center-fed linear dipole has a figure 8 radiation pattern with a directivity (gain) of 1.5 X, or 1.76 dBi -- see any antenna engineering textbook. RF I believe the computer programs to be more up to date than the books! There certainly have been more advances since they have come into being. The programs reflect Maxwells equations which support the presence of particles which is what provide the radiation resistance and not the dipole itself. The dipole will show a donut pattern that will gradually deform to a perfect sphere when resistance drops to zero as per Poynting. I would also point out that the programs support the presence of Gaussian static particles as does mathematics. I would imagine that no matter what programs you decide to use you will get the same results as you increase the element diameter until the impedance is zero.No point in trashing computer programs in advance because of personal intuition. All I have done is removing resistance losses that do not contribute to radiation. the programs are based on the books... but even worse, they are digital approximations of the continuous formulas and as such are not completely accurate. *this is especially true when extremely large or small numbers are used or there are a large number of additions done, as is common in antenna modeling programs. *there are also assumptions made in the development of most of those programs that are often not stated to, or not understood by, the user, such as you. *so when you set something to optimize forever or start making elements extremely skinny, fat, short, or long, or too close together, you are most likely going to get wrong, or physically unrealizable results. Obviously you are very experienced in generating and bug catching in antenna programs having large experiences of finding antenna errors. What exactly in the nature of antenna computer programs, which have been around for some time now, have you found them to be suspect ? In my case the program verified what mathematics show as the presence of particles on the surface and where the total input forces were used for particle propagation. Now I am aware you have taken the position that particles are not involved in radiation and thus you will resist what computer programs arrive at relying on your intuition at all times which requires no personal experience on the subject However, I am taking the program that I purchased on trust especially when it follows the maxwell equations and where I am not adverse to change. I look forward to specific examples that buttress your thoughts in a scientific manner so I may decide what to do with my program purchase. May I recommend you do the same thing with the program of your choice where you can specifically point to the areas of error where they do not meet your expectations. Why not do the same with EZNEC so Roy can learn from your personal experiences and intuitions and institute the appropriate corrections. Never mind the length of the dipole just make the diameter very very fat and see what EZNEC does. Groan... Let me tell you the story about 24 dbi gain dipoles... Simple to model.. Then again, no, it's a futile waste of time trying to convince you of the error of your ways.. * :/ Continue with fantasy hour... * :/ What ever program you use let me know the result for a fat dipole. Walk the walk ! Forget the talk! |
#4
|
|||
|
|||
Computer model experiment
What ever program you use let me know the result for a fat dipole.
Walk the walk ! Forget the talk! Are you sure you have not violated the segment length/wire diameter ratio? From Cebik; Intermediate Antenna Modeling: "In NEC-2 it is especially important to keep the segment length (greater than) about 4 times the wire diameter. You may reduce this value by half by invoking the EK command." Also, what does your "Average Gain Test" report show? 73, Frank |
Reply |
Thread Tools | Search this Thread |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Forum | |||
FA: Philbrick GAP/R Model K2-W Early Computer Tube Op-Amp | Boatanchors | |||
FA: Philbrick GAP/R Model K2-W Early Computer Tube Op-Amp | Boatanchors | |||
FA: Philbrick GAP/R Model K2-W Early Computer Tube Op-Amp | Boatanchors | |||
FA: Philbrick GAP/R Model K2-W Early Computer VacuumTube Op-Amp | Boatanchors | |||
FA: Radio Shack Model 100 laptop computer ++ | Equipment |