|
Eznec modeling loading coils?
It's not practical, and often not possible (due to inability to model
proximity effect), to model a loading coil on a turn-by-turn basis. Where a loading coil is physically long enough to have a current gradient across it, I've always believed it's adequate to model it as a wire whose outer diameter is the same as the outer diameter of the coil. Usually, a single lumped inductive/resistive "load" placed in the center of that wire is adequate, although if the coil is particularly long, you might want to split the load into several equal pieces distributed along the fat wire. This approach assumes that the current gradient across a coil is the same as for a wire of the same outside diameter as the coil. I don't, however, have any measured data to confirm this. It would be a fairly simple matter to put toroidal current transformers just above and below a coil and compare their values to what EZNEC or other programs report. If anyone has done this or is interested in doing it, I'd love to see the result along with a detailed description of the methodology. Roy Lewallen, W7EL Yuri Blanarovich wrote: Question to Roy, can EZNEC model loading coils in loaded antennas, like mobile whips including the physical properties of the coil? I just had some exchange with W8JI on eHam.net http://www.eham.net/articles/5998 where he is ignoring the fact that loading coil is part of the radiator and its physical properties, where length to diameter ratio have impact on the current distribution in the coil. He is claiming that modeling programs confirm his statement that the current is the same at the both ends of the loading coil (close in short coils) . I am having hard time to see how the modeling program is capturing physical properties of the coil besides inductance, Q etc. W9UCW has done considerable amount of work on the subject and has measured that the current diminishes in the loading coil away from the feedpoint. Thanks! Yuri, K3BU |
Why not seperate the inductance into
four separate loads over a length that equals one load? oSaddam (Yuri Blanarovich) wrote in message ... Question to Roy, can EZNEC model loading coils in loaded antennas, like mobile whips including the physical properties of the coil? I just had some exchange with W8JI on eHam.net http://www.eham.net/articles/5998 where he is ignoring the fact that loading coil is part of the radiator and its physical properties, where length to diameter ratio have impact on the current distribution in the coil. He is claiming that modeling programs confirm his statement that the current is the same at the both ends of the loading coil (close in short coils) . I am having hard time to see how the modeling program is capturing physical properties of the coil besides inductance, Q etc. W9UCW has done considerable amount of work on the subject and has measured that the current diminishes in the loading coil away from the feedpoint. Thanks! Yuri, K3BU |
This approach assumes that the current gradient across a coil is the same as for a wire of the same outside diameter as the coil. I don't, however, have any measured data to confirm this. It would be a fairly simple matter to put toroidal current transformers just above and below a coil and compare their values to what EZNEC or other programs report. If anyone has done this or is interested in doing it, I'd love to see the result along with a detailed description of the methodology. Roy Lewallen, W7EL Barry, W9UCW has done extensive measurements using termocuples and other methods minimizing disturbance to measured system. He found that current in (esp. longer) coils is diminishing towards the other end. If you read the thread and some more velleyball on Topband reflector last month, there is more info. Yuri, K3BU |
That's what EZNEC would predict, modeled as I suggested.
Roy Lewallen, W7EL Yuri Blanarovich wrote: This approach assumes that the current gradient across a coil is the same as for a wire of the same outside diameter as the coil. I don't, however, have any measured data to confirm this. It would be a fairly simple matter to put toroidal current transformers just above and below a coil and compare their values to what EZNEC or other programs report. If anyone has done this or is interested in doing it, I'd love to see the result along with a detailed description of the methodology. Roy Lewallen, W7EL Barry, W9UCW has done extensive measurements using termocuples and other methods minimizing disturbance to measured system. He found that current in (esp. longer) coils is diminishing towards the other end. If you read the thread and some more velleyball on Topband reflector last month, there is more info. Yuri, K3BU |
Yuri Blanarovich wrote:
Barry, W9UCW has done extensive measurements using termocuples and other methods minimizing disturbance to measured system. He found that current in (esp. longer) coils is diminishing towards the other end. If you add 1/4WL to the top of the antenna, the current in the coil will *increase* towards the other end. In a standing wave antenna, the total current is the sum of the forward current and the reflected current. Whether the total current increases or decreases depends on the phasing of the two currents at the point where the coil is placed. -- 73, Cecil http://www.qsl.net/w5dxp -----= Posted via Newsfeeds.Com, Uncensored Usenet News =----- http://www.newsfeeds.com - The #1 Newsgroup Service in the World! -----== Over 80,000 Newsgroups - 16 Different Servers! =----- |
When I open the load dialog window, it presents R in some form or another (the answer to your Q provision). Are you concerned the program will treat the load as an infinitesimally small object? If it does, you could decimate the coil inductance and distribute it across 10 segments whose span is equivalent to your anticipated physical length. Perhaps a lot of trouble (and perhaps there's another, better way), but it seems to be the price of your query. 73's Richard Clark, KB7QHC That sounds reasonable. Including physical properties of the coil (Length/diameter ratio, physical size) in the calculations of the model could help to optimize the system, allow more precise calculations and participation in the optimization routines. If the load is treated as point, the current is considered the same at the entrance and exit of the coil, which is not the case especially in longer loading coils. Question is to what extent is program capable of capturing and then massaging it and what is the correlation to real life situations. I got some RF ammeters, it will be interesting to see wasaaap. Yuri, K3BUm |
If you add 1/4WL to the top of the antenna, the current in the coil will *increase* towards the other end. In a standing wave antenna, the total current is the sum of the forward current and the reflected current. Whether the total current increases or decreases depends on the phasing of the two currents at the point where the coil is placed. -- 73, Cecil http://www.qsl.net/w5dxp No standing or laying waves allowed. We are talking about the (mostly) electrical quarter wave loaded radiators (aka mobile antennas) as used in shootouts. Yuri |
Yuri Blanarovich wrote:
No standing or laying waves allowed. We are talking about the (mostly) electrical quarter wave loaded radiators (aka mobile antennas) as used in shootouts. Those mobile antennas are standing-wave antennas. The forward current is relatively constant and is reflected at the end of the whip. The reflected current adds to the forward current to give the total current. A standing- wave antenna is similar to a transmission line with standing waves except that it is designed to radiate and the transmission line is not. The major reason that the current drops through the loading coil is the same as for a 1/4WL monopole without a loading coil. The total current is maximum at the feedpoint and drops to zero at the end of the whip because the forward and reflected currents add to those values. The forward current through the coil is virtually the same value at each end of the coil. The reflected current back through the coil is virtually the same value at each end of the coil. The phasing between those two currents is what determines the total current which is indeed different at each end of the coil. -- 73, Cecil http://www.qsl.net/w5dxp -----= Posted via Newsfeeds.Com, Uncensored Usenet News =----- http://www.newsfeeds.com - The #1 Newsgroup Service in the World! -----== Over 80,000 Newsgroups - 16 Different Servers! =----- |
Yuri,
I think you would better understand what is happening if you view the inductance in question as half of a transformer, where the other half is the aether. This allows you to bring in the formular NxI one side = NxI the otherside. This way you can see that I is constant. If it wasn't a constant then the lines of flux would have a very distorted shape at one end of the inductance. The fact that transformers have impedances that are not totally resistive is because of other factors than Inductance. If I am incorrect with this analogy I would apreciate any corrections as I am sure Yuri would to Art oSaddam (Yuri Blanarovich) wrote in message ... Question to Roy, can EZNEC model loading coils in loaded antennas, like mobile whips including the physical properties of the coil? I just had some exchange with W8JI on eHam.net http://www.eham.net/articles/5998 where he is ignoring the fact that loading coil is part of the radiator and its physical properties, where length to diameter ratio have impact on the current distribution in the coil. He is claiming that modeling programs confirm his statement that the current is the same at the both ends of the loading coil (close in short coils) . I am having hard time to see how the modeling program is capturing physical properties of the coil besides inductance, Q etc. W9UCW has done considerable amount of work on the subject and has measured that the current diminishes in the loading coil away from the feedpoint. Thanks! Yuri, K3BU |
There are a couple of problems with the analogy. First, of course, is
that there is no secondary winding to convert the primary's magnetic field to the current I you reference. The "aether" has no N and no I, and doesn't do this. The second problem is that a simple transformer model, or any other lumped-element circuit model, is valid only if its dimensions are very small compared to a wavelength(*). A mobile whip loading coil is often long enough that a current gradient can occur from one end to another, so a single lumped element model might be inadequate. Roy Lewallen, W7EL (*) A particular problem with the mobile loading coil in an electrically small whip is that the current changes a great deal over even a very short distance, as it goes from zero at the end of the whip (assuming no top loading) to a large value at the base. It sounds like sort of a circular argument, but a lumped element model loses accuracy any time there can be a current gradient across the component. Interestingly, the problem becomes less and less severe as top loading is increased, because it reduces the current gradient along the whip. You *could* quite accurately model a loading coil as a single lumped element if you had enough top loading. Art Unwin KB9MZ wrote: Yuri, I think you would better understand what is happening if you view the inductance in question as half of a transformer, where the other half is the aether. This allows you to bring in the formular NxI one side = NxI the otherside. This way you can see that I is constant. If it wasn't a constant then the lines of flux would have a very distorted shape at one end of the inductance. The fact that transformers have impedances that are not totally resistive is because of other factors than Inductance. If I am incorrect with this analogy I would apreciate any corrections as I am sure Yuri would to Art |
Roy Lewallen wrote in message ...
There are a couple of problems with the analogy. First, of course, is that there is no secondary winding to convert the primary's magnetic field to the current I you reference. The "aether" has no N and no I, and doesn't do this. The second problem is that a simple transformer model, or any other lumped-element circuit model, is valid only if its dimensions are very small compared to a wavelength(*). A mobile whip loading coil is often long enough that a current gradient can occur from one end to another, so a single lumped element model might be inadequate. Well Roy, I was trying to remove the inter coil and end to end capacitance from the equation so a simple analogy could be made. Thus if there was a current change the cause could be removed from the itemizerd inductance alone so that other reasons would have to be researched for energy changes or losses. ( simi;ar to transmission lines and filter circuits) Still I bend to your superior experience in this matter and withdraw my comments so that I do not mislead others. My comments emanate from lumped circuit theorems where the other half of the transformer shows up at the receiving end as being discussed in another posting initiated by Dr Slick. The difference being the environnment, one of specialised steel laminates and the other being the eather. Looking at things that way forces one to review the different coupling to ground effects and other anomolies of this particular situation Regards Art Roy Lewallen, W7EL (*) A particular problem with the mobile loading coil in an electrically small whip is that the current changes a great deal over even a very short distance, as it goes from zero at the end of the whip (assuming no top loading) to a large value at the base. It sounds like sort of a circular argument, but a lumped element model loses accuracy any time there can be a current gradient across the component. Interestingly, the problem becomes less and less severe as top loading is increased, because it reduces the current gradient along the whip. You *could* quite accurately model a loading coil as a single lumped element if you had enough top loading. Art Unwin KB9MZ wrote: Yuri, I think you would better understand what is happening if you view the inductance in question as half of a transformer, where the other half is the aether. This allows you to bring in the formular NxI one side = NxI the otherside. This way you can see that I is constant. If it wasn't a constant then the lines of flux would have a very distorted shape at one end of the inductance. The fact that transformers have impedances that are not totally resistive is because of other factors than Inductance. If I am incorrect with this analogy I would apreciate any corrections as I am sure Yuri would to Art |
| All times are GMT +1. The time now is 07:55 PM. |
Powered by vBulletin® Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
RadioBanter.com