-
#120 by SeeShells on 14 Dec, 2015 01:22
-
Question for you antenna experts out there: "How wide is the gap between the close ends of the 1/4 wavelength dipoles used to construct a 1/2 wavelength antenna in the T configuration?"
see gif attached
See .gif attached.
-
#121 by aero on 14 Dec, 2015 01:42
-
Question for you antenna experts out there: "How wide is the gap between the close ends of the 1/4 wavelength dipoles used to construct a 1/2 wavelength antenna in the T configuration?"
see gif attached
See .gif attached.
Thanks Shell - but in the real world of hardware, how far apart are the two vertical feeds leading to the close ends of the 1/4 wl dipoles? Is it really just a half wl wire fed in the center so in reality there is no gap? The graphic doesn't appear to imply that. What is shown in the pink and blue? voltage and current? certainly not RF fields. -
#122 by SteveD on 14 Dec, 2015 01:44
-
Question for you antenna experts out there: "How wide is the gap between the close ends of the 1/4 wavelength dipoles used to construct a 1/2 wavelength antenna in the T configuration?"
see gif attached
See .gif attached.
That's showing the total length. I think what Aero was asking is how large the gap in the middle should be. Could you, for example, make the dipoles of the T 1/16 of a wavelength long and have a large gap of empty space between the two ends? -
#123 by SeeShells on 14 Dec, 2015 02:29
-
I remember tuning a Dipole when I was in school (yes we had electricity then) and made the area between the two tunable with a pair of screws. I remember that a 1/10 WL worked the best. I did a search on the net to find something and there was nothing that jumped out at me to answer the question.Question for you antenna experts out there: "How wide is the gap between the close ends of the 1/4 wavelength dipoles used to construct a 1/2 wavelength antenna in the T configuration?"
see gif attached
See .gif attached.
That's showing the total length. I think what Aero was asking is how large the gap in the middle should be. Could you, for example, make the dipoles of the T 1/16 of a wavelength long and have a large gap of empty space between the two ends?
Shell
BTW it the Magnetic and electrical field you're looking at. -
#124 by rfmwguy on 14 Dec, 2015 02:39
-
Not sure this 1/2 wave dipole will work in a cavity. Normally this would only be used when an rf ground plane were not present. Think of the ham operator stringing up a dipole between a couple of trees.Question for you antenna experts out there: "How wide is the gap between the close ends of the 1/4 wavelength dipoles used to construct a 1/2 wavelength antenna in the T configuration?"
see gif attached
See .gif attached.
That's showing the total length. I think what Aero was asking is how large the gap in the middle should be. Could you, for example, make the dipoles of the T 1/16 of a wavelength long and have a large gap of empty space between the two ends?
In our case, the frustum cavity is the ground plane, only a quarter wave resonator need be present...just like in a magnetron.
To your question of spacing, in a 50 ohm system, the spacing is determined by the dielectric material of the transmission line...whatever the distance is from the center conductor to the braid (in the case of common coax) is all that is needed.
Here's a bunch of stuff related to a center fed dipole:
https://en.m.wikipedia.org/wiki/Dipole_antenna -
#125 by SeeShells on 14 Dec, 2015 03:00
-
That's like 75 ohm?
Not sure this 1/2 wave dipole will work in a cavity. Normally this would only be used when an rf ground plane were not present. Think of the ham operator stringing up a dipole between a couple of trees.Question for you antenna experts out there: "How wide is the gap between the close ends of the 1/4 wavelength dipoles used to construct a 1/2 wavelength antenna in the T configuration?"
see gif attached
See .gif attached.
That's showing the total length. I think what Aero was asking is how large the gap in the middle should be. Could you, for example, make the dipoles of the T 1/16 of a wavelength long and have a large gap of empty space between the two ends?
In our case, the frustum cavity is the ground plane, only a quarter wave resonator need be present...just like in a magnetron.
To your question of spacing, in a 50 ohm system, the spacing is determined by the dielectric material of the transmission line...whatever the distance is from the center conductor to the braid (in the case of common coax) is all that is needed.
http://www.csgnetwork.com/antennaedcalc.html -
#126 by SeeShells on 14 Dec, 2015 03:04
-
Not sure this 1/2 wave dipole will work in a cavity. Normally this would only be used when an rf ground plane were not present. Think of the ham operator stringing up a dipole between a couple of trees.Question for you antenna experts out there: "How wide is the gap between the close ends of the 1/4 wavelength dipoles used to construct a 1/2 wavelength antenna in the T configuration?"
see gif attached
See .gif attached.
That's showing the total length. I think what Aero was asking is how large the gap in the middle should be. Could you, for example, make the dipoles of the T 1/16 of a wavelength long and have a large gap of empty space between the two ends?
In our case, the frustum cavity is the ground plane, only a quarter wave resonator need be present...just like in a magnetron.
To your question of spacing, in a 50 ohm system, the spacing is determined by the dielectric material of the transmission line...whatever the distance is from the center conductor to the braid (in the case of common coax) is all that is needed.
Here's a bunch of stuff related to a center fed dipole:
https://en.m.wikipedia.org/wiki/Dipole_antenna
The kicker is the frustum shouldn't be acting like a ground plane for the antenna as the mode generation nodes are different. This is what Dr. Rodal was talking about many pages ago and I'd agree with him. -
#127 by rfmwguy on 14 Dec, 2015 03:07
-
If you have parallel wires leading towards the dipole launch, air would be the dielectric and used to calculate spacing to maintain 50 ohms match.
That's like 75 ohm?
Not sure this 1/2 wave dipole will work in a cavity. Normally this would only be used when an rf ground plane were not present. Think of the ham operator stringing up a dipole between a couple of trees.Question for you antenna experts out there: "How wide is the gap between the close ends of the 1/4 wavelength dipoles used to construct a 1/2 wavelength antenna in the T configuration?"
see gif attached
See .gif attached.
That's showing the total length. I think what Aero was asking is how large the gap in the middle should be. Could you, for example, make the dipoles of the T 1/16 of a wavelength long and have a large gap of empty space between the two ends?
In our case, the frustum cavity is the ground plane, only a quarter wave resonator need be present...just like in a magnetron.
To your question of spacing, in a 50 ohm system, the spacing is determined by the dielectric material of the transmission line...whatever the distance is from the center conductor to the braid (in the case of common coax) is all that is needed.
http://www.csgnetwork.com/antennaedcalc.html
My thought here is the cavity is already at rf ground potential, so another radiator in the can would not be useful...dead weight so to speak -
#128 by rfmwguy on 14 Dec, 2015 03:11
-
Interesting thought about frustum not being at rf ground potential. Are you planning to isolate signal source ground and frustum ground?
If you are, this will be very interesting. Im not sure I've ever read about a cavity being above rf ground potential...just another example of trying something new...no problem with that -
#129 by aero on 14 Dec, 2015 03:20
-
I remember tuning a Dipole when I was in school (yes we had electricity then) and made the area between the two tunable with a pair of screws. I remember that a 1/10 WL worked the best. I did a search on the net to find something and there was nothing that jumped out at me to answer the question.Question for you antenna experts out there: "How wide is the gap between the close ends of the 1/4 wavelength dipoles used to construct a 1/2 wavelength antenna in the T configuration?"
see gif attached
See .gif attached.
That's showing the total length. I think what Aero was asking is how large the gap in the middle should be. Could you, for example, make the dipoles of the T 1/16 of a wavelength long and have a large gap of empty space between the two ends?
Shell
BTW it the Magnetic and electrical field you're looking at.
hmm 1/10 wl is 1.2 cm isn't it? That seems like a pretty large gap but I'll run it and see. I've been running 1 - 2 mm which is more like the radius of a common coax. But then, what is the diameter of the coax used for 2.5 GHz? Wikipedia gives a range from very small to over 6 inches in diameter. -
#130 by SeeShells on 14 Dec, 2015 03:22
-
If you have parallel wires leading towards the dipole launch, air would be the dielectric and used to calculate spacing to maintain 50 ohms match.
That's like 75 ohm?
Not sure this 1/2 wave dipole will work in a cavity. Normally this would only be used when an rf ground plane were not present. Think of the ham operator stringing up a dipole between a couple of trees.Question for you antenna experts out there: "How wide is the gap between the close ends of the 1/4 wavelength dipoles used to construct a 1/2 wavelength antenna in the T configuration?"
see gif attached
See .gif attached.
That's showing the total length. I think what Aero was asking is how large the gap in the middle should be. Could you, for example, make the dipoles of the T 1/16 of a wavelength long and have a large gap of empty space between the two ends?
In our case, the frustum cavity is the ground plane, only a quarter wave resonator need be present...just like in a magnetron.
To your question of spacing, in a 50 ohm system, the spacing is determined by the dielectric material of the transmission line...whatever the distance is from the center conductor to the braid (in the case of common coax) is all that is needed.
http://www.csgnetwork.com/antennaedcalc.html
My thought here is the cavity is already at rf ground potential, so another radiator in the can would not be useful...dead weight so to speak
Also in this image look at the mode generation. https://drive.google.com/file/d/0B1XizxEfB23tYVNDalhQel9tZ3c/view
If we put an antenna 1/4 from the top or bottom the cavity natural resonance mode generation will not be at the same place as the antenna and cause inter-modal actions decreasing the Q as they build and collapse.
It makes sense to place the antenna center frustum parallel to the bottom in this mode TE013 and for other modes you need to know the mode pattern to properly place if your going to use a dipole.
This way you're adding to the Q not causing a degradation.
Make sense?
Shell
-
#131 by rfmwguy on 14 Dec, 2015 03:28
-
Yes...in the meep sims, is the frustum an isolated element, a pure reflector not having an electrical connection to the radiation source? Pardon my meepneptitude
If you have parallel wires leading towards the dipole launch, air would be the dielectric and used to calculate spacing to maintain 50 ohms match.
That's like 75 ohm?
Not sure this 1/2 wave dipole will work in a cavity. Normally this would only be used when an rf ground plane were not present. Think of the ham operator stringing up a dipole between a couple of trees.Question for you antenna experts out there: "How wide is the gap between the close ends of the 1/4 wavelength dipoles used to construct a 1/2 wavelength antenna in the T configuration?"
see gif attached
See .gif attached.
That's showing the total length. I think what Aero was asking is how large the gap in the middle should be. Could you, for example, make the dipoles of the T 1/16 of a wavelength long and have a large gap of empty space between the two ends?
In our case, the frustum cavity is the ground plane, only a quarter wave resonator need be present...just like in a magnetron.
To your question of spacing, in a 50 ohm system, the spacing is determined by the dielectric material of the transmission line...whatever the distance is from the center conductor to the braid (in the case of common coax) is all that is needed.
http://www.csgnetwork.com/antennaedcalc.html
My thought here is the cavity is already at rf ground potential, so another radiator in the can would not be useful...dead weight so to speak
Also in this image look at the mode generation. https://drive.google.com/file/d/0B1XizxEfB23tYVNDalhQel9tZ3c/view
If we put an antenna 1/4 from the top or bottom the cavity natural resonance mode generation will not be at the same place as the antenna and cause inter-modal actions decreasing the Q as they build and collapse.
It makes sense to place the antenna center frustum parallel to the bottom in this mode TE013 and for other modes you need to know the mode pattern to properly place if your going to use a dipole.
This way you're adding to the Q not causing a degradation.
Make sense?
Shell
-
#132 by SeeShells on 14 Dec, 2015 03:38
-
I believe that is the case in meep, it was one of the issues in modeling the loops.
Yes...in the meep sims, is the frustum an isolated element, a pure reflector not having an electrical connection to the radiation source? Pardon my meepneptitude
If you have parallel wires leading towards the dipole launch, air would be the dielectric and used to calculate spacing to maintain 50 ohms match.
That's like 75 ohm?
Not sure this 1/2 wave dipole will work in a cavity. Normally this would only be used when an rf ground plane were not present. Think of the ham operator stringing up a dipole between a couple of trees.Question for you antenna experts out there: "How wide is the gap between the close ends of the 1/4 wavelength dipoles used to construct a 1/2 wavelength antenna in the T configuration?"
see gif attached
See .gif attached.
That's showing the total length. I think what Aero was asking is how large the gap in the middle should be. Could you, for example, make the dipoles of the T 1/16 of a wavelength long and have a large gap of empty space between the two ends?
In our case, the frustum cavity is the ground plane, only a quarter wave resonator need be present...just like in a magnetron.
To your question of spacing, in a 50 ohm system, the spacing is determined by the dielectric material of the transmission line...whatever the distance is from the center conductor to the braid (in the case of common coax) is all that is needed.
http://www.csgnetwork.com/antennaedcalc.html
My thought here is the cavity is already at rf ground potential, so another radiator in the can would not be useful...dead weight so to speak
Also in this image look at the mode generation. https://drive.google.com/file/d/0B1XizxEfB23tYVNDalhQel9tZ3c/view
If we put an antenna 1/4 from the top or bottom the cavity natural resonance mode generation will not be at the same place as the antenna and cause inter-modal actions decreasing the Q as they build and collapse.
It makes sense to place the antenna center frustum parallel to the bottom in this mode TE013 and for other modes you need to know the mode pattern to properly place if your going to use a dipole.
This way you're adding to the Q not causing a degradation.
Make sense?
Shell
In the real world I'd slap a connector right through the bottom center and feed a coax up to the antenna then branch out horizontally and parallel to the bottom plate with the antenna.
added I'm not a Ham like some here are and antennas are not my cup of tea. -
#133 by rfmwguy on 14 Dec, 2015 03:53
-
Got my extra class in 1981 but inactive since the 90s. Trouble is, a connector will make frustum at rf ground potential with a single center conductor. To isolate frustum from signal source, you'll need a triax connector or a dual center conductor if you want rf ground isolation.
I believe that is the case in meep, it was one of the issues in modeling the loops.
Yes...in the meep sims, is the frustum an isolated element, a pure reflector not having an electrical connection to the radiation source? Pardon my meepneptitude
If you have parallel wires leading towards the dipole launch, air would be the dielectric and used to calculate spacing to maintain 50 ohms match.
That's like 75 ohm?
Not sure this 1/2 wave dipole will work in a cavity. Normally this would only be used when an rf ground plane were not present. Think of the ham operator stringing up a dipole between a couple of trees.Question for you antenna experts out there: "How wide is the gap between the close ends of the 1/4 wavelength dipoles used to construct a 1/2 wavelength antenna in the T configuration?"
see gif attached
See .gif attached.
That's showing the total length. I think what Aero was asking is how large the gap in the middle should be. Could you, for example, make the dipoles of the T 1/16 of a wavelength long and have a large gap of empty space between the two ends?
In our case, the frustum cavity is the ground plane, only a quarter wave resonator need be present...just like in a magnetron.
To your question of spacing, in a 50 ohm system, the spacing is determined by the dielectric material of the transmission line...whatever the distance is from the center conductor to the braid (in the case of common coax) is all that is needed.
http://www.csgnetwork.com/antennaedcalc.html
My thought here is the cavity is already at rf ground potential, so another radiator in the can would not be useful...dead weight so to speak
Also in this image look at the mode generation. https://drive.google.com/file/d/0B1XizxEfB23tYVNDalhQel9tZ3c/view
If we put an antenna 1/4 from the top or bottom the cavity natural resonance mode generation will not be at the same place as the antenna and cause inter-modal actions decreasing the Q as they build and collapse.
It makes sense to place the antenna center frustum parallel to the bottom in this mode TE013 and for other modes you need to know the mode pattern to properly place if your going to use a dipole.
This way you're adding to the Q not causing a degradation.
Make sense?
Shell
In the real world I'd slap a connector right through the bottom center and feed a coax up to the antenna then branch out horizontally and parallel to the bottom plate with the antenna.
added I'm not a Ham like some here are and antennas are not my cup of tea.
Watch out for these and make sure they can handle the power and freq. Also, the triax will need a metal case on the outside of the frustum to minimize rf leakage. -
#134 by SeeShells on 14 Dec, 2015 03:57
-
I'm not going with this design hanging the dipole in the center. You are very correct in the mess it could cause. I may end up with a dual loop modified in the small end but haven't made my mind up yet. So far I like the Q and mode generation a waveguide can provide.
Got my extra class in 1981 but inactive since the 90s. Trouble is, a connector will make frustum at rf ground potential with a single center conductor. To isolate frustum from signal source, you'll need a triax connector or a dual center conductor if you want rf ground isolation.
I believe that is the case in meep, it was one of the issues in modeling the loops.
Yes...in the meep sims, is the frustum an isolated element, a pure reflector not having an electrical connection to the radiation source? Pardon my meepneptitude
If you have parallel wires leading towards the dipole launch, air would be the dielectric and used to calculate spacing to maintain 50 ohms match.
That's like 75 ohm?
Not sure this 1/2 wave dipole will work in a cavity. Normally this would only be used when an rf ground plane were not present. Think of the ham operator stringing up a dipole between a couple of trees.Question for you antenna experts out there: "How wide is the gap between the close ends of the 1/4 wavelength dipoles used to construct a 1/2 wavelength antenna in the T configuration?"
see gif attached
See .gif attached.
That's showing the total length. I think what Aero was asking is how large the gap in the middle should be. Could you, for example, make the dipoles of the T 1/16 of a wavelength long and have a large gap of empty space between the two ends?
In our case, the frustum cavity is the ground plane, only a quarter wave resonator need be present...just like in a magnetron.
To your question of spacing, in a 50 ohm system, the spacing is determined by the dielectric material of the transmission line...whatever the distance is from the center conductor to the braid (in the case of common coax) is all that is needed.
http://www.csgnetwork.com/antennaedcalc.html
My thought here is the cavity is already at rf ground potential, so another radiator in the can would not be useful...dead weight so to speak
Also in this image look at the mode generation. https://drive.google.com/file/d/0B1XizxEfB23tYVNDalhQel9tZ3c/view
If we put an antenna 1/4 from the top or bottom the cavity natural resonance mode generation will not be at the same place as the antenna and cause inter-modal actions decreasing the Q as they build and collapse.
It makes sense to place the antenna center frustum parallel to the bottom in this mode TE013 and for other modes you need to know the mode pattern to properly place if your going to use a dipole.
This way you're adding to the Q not causing a degradation.
Make sense?
Shell
In the real world I'd slap a connector right through the bottom center and feed a coax up to the antenna then branch out horizontally and parallel to the bottom plate with the antenna.
added I'm not a Ham like some here are and antennas are not my cup of tea.
Watch out for these and make sure they can handle the power and freq. Also, the triax will need a metal case on the ojtside of the frustum to minimize rf leakage.
added
The reason I like the symmetrical waveguide is that you precondition the wave before it hits the frustum cavity instead of letting the asymmetrical cavity determine the radiating pattern from the maggie's antenna or an antenna stuck into the sidewalls or end. ( I've just found it's tough listening to the Best of Nickleback, typing and thinking at the same time.
)
Shell -
#135 by zen-in on 14 Dec, 2015 04:12
-
...pretend there is a gif here
Got my extra class in 1981 but inactive since the 90s. Trouble is, a connector will make frustum at rf ground potential with a single center conductor. To isolate frustum from signal source, you'll need a triax connector or a dual center conductor if you want rf ground isolation.
Watch out for these and make sure they can handle the power and freq. Also, the triax will need a metal case on the outside of the frustum to minimize rf leakage.
Since the body of the magnetron is live with RF any balanced drive would need a balun (to drive the fustrum with a dipole). It would have to be pretty beefy and maybe fed coaxially from the magnetron. This intuitively seem wrong because you really want the fustrum to be at ground potential instead of floating. The magnetron already has a very good 1/4 λ drive (or maybe smaller) sticking out of the business end. The problem is one of matching the impedance of the magnetron drive with the fustrum, no matter what drive is used -balanced or unbalanced. But I think the balanced drive is a lot of trouble and would not work. It should be possible to calculate the impedance of the magnetron drive then design an inline matching network that would match the cavity impedance when 1/4 λ driven. Similar idea as calculating stripline impedances. For example reducing the diameter of the feed increases the impedance. The dielectric surrounding the center conductor would also be a factor. There of course has to be a very good gap-free shield around the center conductor and connecting the magnetron body with the Copper shell of the fustrum.
-
#136 by SeeShells on 14 Dec, 2015 04:24
-
It is a 1/4 wave out of the maggie, I measured it after pulling one totally apart (DON'T DO THIS IF YOU DON'T KNOW WHAT YOUR DOING ... BIO-HAZARDS) You can attach directly to the antenna right out of the maggie which will give you the maxim power down a coax into a 1/4 wave in a wavegiude on the frustum.
...pretend there is a gif here
Got my extra class in 1981 but inactive since the 90s. Trouble is, a connector will make frustum at rf ground potential with a single center conductor. To isolate frustum from signal source, you'll need a triax connector or a dual center conductor if you want rf ground isolation.
Watch out for these and make sure they can handle the power and freq. Also, the triax will need a metal case on the outside of the frustum to minimize rf leakage.
Since the body of the magnetron is live with RF any balanced drive would need a balun (to drive the fustrum with a dipole). It would have to be pretty beefy and maybe fed coaxially from the magnetron. This intuitively seem wrong because you really want the fustrum to be at ground potential instead of floating. The magnetron already has a very good 1/4 λ drive (or maybe smaller) sticking out of the business end. The problem is one of matching the impedance of the magnetron drive with the fustrum, no matter what drive is used -balanced or unbalanced. But I think the balanced drive is a lot of trouble and would not work. It should be possible to calculate the impedance of the magnetron drive then design an inline matching network that would match the cavity impedance when 1/4 λ driven. Similar idea as calculating stripline impedances. For example reducing the diameter of the feed increases the impedance. The dielectric surrounding the center conductor would also be a factor. There of course has to be a very good gap-free shield around the center conductor and connecting the magnetron body with the Copper shell of the fustrum.
-
#137 by zen-in on 14 Dec, 2015 04:40
-
It is a 1/4 wave out of the maggie, I measured it after pulling one totally apart (DON'T DO THIS IF YOU DON'T KNOW WHAT YOUR DOING ... BIO-HAZARDS) You can attach directly to the antenna right out of the maggie which will give you the maxim power down a coax into a 1/4 wave in a wavegiude on the frustum.
...pretend there is a gif here
Got my extra class in 1981 but inactive since the 90s. Trouble is, a connector will make frustum at rf ground potential with a single center conductor. To isolate frustum from signal source, you'll need a triax connector or a dual center conductor if you want rf ground isolation.
Watch out for these and make sure they can handle the power and freq. Also, the triax will need a metal case on the outside of the frustum to minimize rf leakage.
Since the body of the magnetron is live with RF any balanced drive would need a balun (to drive the fustrum with a dipole). It would have to be pretty beefy and maybe fed coaxially from the magnetron. This intuitively seem wrong because you really want the fustrum to be at ground potential instead of floating. The magnetron already has a very good 1/4 λ drive (or maybe smaller) sticking out of the business end. The problem is one of matching the impedance of the magnetron drive with the fustrum, no matter what drive is used -balanced or unbalanced. But I think the balanced drive is a lot of trouble and would not work. It should be possible to calculate the impedance of the magnetron drive then design an inline matching network that would match the cavity impedance when 1/4 λ driven. Similar idea as calculating stripline impedances. For example reducing the diameter of the feed increases the impedance. The dielectric surrounding the center conductor would also be a factor. There of course has to be a very good gap-free shield around the center conductor and connecting the magnetron body with the Copper shell of the fustrum.
The problem remains is that going to be a good impedance match. I don't know what impedance the magnetron is suppose to drive. Maybe the tech manuals will say. Using a network analyzer you can measure the input impedance of the fustrum at 2500 MHz when there is a 1/4 λ drive connected to coax. -
#138 by ThereIWas3 on 14 Dec, 2015 05:21
-
Those GIFs show the traditional balanced feedlline going to a dipole, which is itself balanced. Neither side is grounded. If your signal source is unbalanced (Like most ham transmitters with a coaxial antenna connector, then you need a balanced-to-unbalanced converter, a sort of transformer. This is abbreviated balun. You typically run unbalanced feedline (coaxial cable) out to the antenna, where the balan converts it to a balanced feed at the appropriate impedance to match the dipole at the frequency of interest. If you do not get the impedances correct, you get a reflection, power goes back to the magnetron, heating it.
Cable TV cable is 75 ohms impedance at TV frequencies (100-1000MHz). Ham antennas are usually for HF (3-30MHz) and you use 50 ohm cable. 50 ohms does a better match to the sorts of antennas hams use. (There are other kinds than dipoles.)
Any magnetron taken from an oven is not going to have a coaxial feed on it, nor a balanced connection. Instead it will have an unbalanced probe sticking right out of it which is perhaps coupled by some sort of short waveguide to the oven cavity.
Shell proposes to use a coaxial connection from the magnetron to the frustrum. For any given type of coax, power loss increases with distance and with frequency. (see chart) This coax will basically need a tiny antenna at each end - one to pick up the signal from a small cavity that the magnetron's probe sticks into, and the other one to reradiate the signal into the frustrum. It would be more efficient to use a hard waveguide for this, but at only 2.4 GHz it would be large (3.5" by 1.7") and unwieldly. -
#139 by dustinthewind on 14 Dec, 2015 07:12
-
Also in this image look at the mode generation. https://drive.google.com/file/d/0B1XizxEfB23tYVNDalhQel9tZ3c/view
If we put an antenna 1/4 from the top or bottom the cavity natural resonance mode generation will not be at the same place as the antenna and cause inter-modal actions decreasing the Q as they build and collapse.
It makes sense to place the antenna center frustum parallel to the bottom in this mode TE013 and for other modes you need to know the mode pattern to properly place if your going to use a dipole.
This way you're adding to the Q not causing a degradation.
Make sense?
Shell
I wanted to point out that this image is of the group B-field and not the E-field because the maximum touches the metal walls. This paper here: http://cas.web.cern.ch/cas/Denmark-2010/Lectures/Wolski-2.pdf page 25 they point out that at the metal walls E_tangent=0 and but B_tan doesn't have to be zero. The group electric maximum of light inside the cavity should be at the magnetic nodes which look to be at about 1/4 wavelength (or the empty spots between the magnetic fields).
The antenna then does work against this electric field to store energy in the cavity. I think at mid cavity it might do negative work when trying to inject a signal and store energy in the current in the antenna instead of the cavity. Hopefully I am not too far off here.
P.S. I think it is supposed to be desirable to have the shape of the antenna in the shape of the current flow desired to be generated. Transverse electric/TE being a loop or loops parallel to the plate and TM being vertically orientated loops. This appears to be TM.
I guess your antenna looks about right but I am not sure what it should look like in 3 dimensions. Also to match the magnetic field the current would have to be going in at the same time on both antenna ends unlike in the animated video of the antenna.
)