-
#2060 by rfmwguy on 24 Jan, 2017 17:33
-
I know you and another overly-active non participant in emdrive experimentation find it difficult to comprehend something new or unexpected might be occurring with the emdrive. That being said, no one here has been appointed as a the expert in emdrive discussions despite their posting style. I suggest this forum remain open to theoretical possibilities beyond textbook references. I happen to believe the emdrive project will be resolved by experimentalists, not by self-appointed experts in google-fu or outdated college courses. Thus is blunt but so are you.I agree with you that Maxwell is a brick wall. It assumes perfection in reflection
to balance out equations and make the universe whole again.
In a recent post, I already explained how Maxwell's equations handle not perfect reflection. If you want to make false statements like that Maxwell requires perfect reflection from conductors, or that metals are somehow transparent to oscillating magnetic fields (there is a wealth of data saying they aren't if you bothered to look), you are the one acting like a brick wall.
I encourage theorists to think beyond Maxwell and look for other possibilities. Who could argue with that accept conservative science fundamentalists? Most of the universe is unknown energy and matter and there is no grand unification theory. What is happening with the emdrive could be a tiny crack in the brick wall of existing knowledge. -
#2061 by meberbs on 24 Jan, 2017 18:04
-
I know you and another overly-active non participant in emdrive experimentation find it difficult to comprehend something new or unexpected might be occurring with the emdrive. That being said, no one here has been appointed as a the expert in emdrive discussions despite their posting style. I suggest this forum remain open to theoretical possibilities beyond textbook references. I happen to believe the emdrive project will be resolved by experimentalists, not by self-appointed experts in google-fu or outdated college courses. Thus is blunt but so are you.I agree with you that Maxwell is a brick wall. It assumes perfection in reflection
to balance out equations and make the universe whole again.
In a recent post, I already explained how Maxwell's equations handle not perfect reflection. If you want to make false statements like that Maxwell requires perfect reflection from conductors, or that metals are somehow transparent to oscillating magnetic fields (there is a wealth of data saying they aren't if you bothered to look), you are the one acting like a brick wall.
I encourage theorists to think beyond Maxwell and look for other possibilities. Who could argue with that accept conservative science fundamentalists? Most of the universe is unknown energy and matter and there is no grand unification theory. What is happening with the emdrive could be a tiny crack in the brick wall of existing knowledge.
Before I can look for other theories, I would need an experiment telling me what to look for. You have recently been promoting ideas that contradict countless experiments. You haven't pointed out a single piece of experimental data that would make college classes outdated. -
#2062 by LowerAtmosphere on 24 Jan, 2017 18:45
-
Edit: Dustinthewind surprises me yet again with an interesting post he made about Copper and meta-materials back in 2015
A definite must-read for those considering advanced concepts and further advanced designs!
https://forum.nasaspaceflight.com/index.php?topic=36911.msg1460370#msg1460370
https://arxiv.org/ftp/arxiv/papers/1009/1009.5663.pdf
https://arxiv.org/pdf/1605.07487v1.pdf
------
Well this was a huge sidetrack!
My original comment is redacted. It was misunderstood entirely and that limits its usefulness. It was only implied that the walls will never be a perfect shield in the real world, not that they are entirely transparent, nor that they somehow filter out only the E field effectively - that would be new physics right there!
The entire point was that even if you super-cooled the walls, the cavity is an open system because it is slightly permeable in places. That is just the nature of thin Copper when exposed to oscillating and temporarily static fields* plus areas where sidewall pressure is negligible. Not to mention areas where many holes will occur randomly. Arguing about permeability is pointless unless you suggest that it is a compass or somehow interacts with other electromagnetic fields to produce the thrust effect.
This is simply a continuation of the observation that phase-overlapped photons would escape (remember the Finnish theory!) and so would accelerated neutrinos. All of this boils down to the incontrovertible fact that the EM drive is not a textbook closed system people, so close your textbooks and pick up your radiometers and interferometers and point them close to a resonant copper cavity if you disagree.
This does raise an interesting question however, what could be used to degrade its shielding capability temporarily? It would be interesting to have variably translucent walls, say something like this https://phys.org/news/2007-01-switchable-mirror-glass-energy-efficient.html but for micro waves. Could be a feasible way to pulse the thrust as others have suggested?
*This becomes a major point of contention when you have to decide whether to include QED in the equation. Many argue that virtual pairs imply that no truly static fields exist at all, raising the point that how static an EM field is is in fact a matter of probability. Essentially you reenter the debate as to the nature of the Quantum vacuum. Manipulating the QV will determine how oscillatory your system is and how much it will penetrate the walls. -
#2063 by Rodal on 24 Jan, 2017 18:51
-
Well this was a huge sidetrack!
To clarify, the comment https://forum.nasaspaceflight.com/index.php?topic=41732.msg1634718#msg1634718 was referring to the statement made by somebody else (not you) that proposes that the EM Drive enables only a time-varying magnetic field (actually he wrote H field) to escape with no time-varying electric field making it to the outside.
My original comment is redacted. It was misunderstood entirely and that limits its usefulness. It was only implied that the walls will never be a perfect shield in the real world, not that they are entirely transparent, nor that they somehow filter out only the E field effectively - that would be new physics right there! ...
The statement "Maxwell is a brick wall. It assumes perfection in reflection" was also written by somebody else (not you)
The emphasis and what was addressed was the validity of such statements, not who made them. The statements stand or fall on their own. -
#2064 by rfmwguy on 24 Jan, 2017 19:24
-
You have several experiments with observed displacement forces using an asymmetric resonant cavity. Ask star-drive if he thinks we need to look beyond maxwells basic equations or assumption that emdrive is a closed system. Better yet, build one yourself and conduct an experiment. You seem to have such a strong position and interest in the topic. Same applies to others attempting to leverage opinion on emdrive. It's up to experimentalists now...theorists had their chance imho.
I know you and another overly-active non participant in emdrive experimentation find it difficult to comprehend something new or unexpected might be occurring with the emdrive. That being said, no one here has been appointed as a the expert in emdrive discussions despite their posting style. I suggest this forum remain open to theoretical possibilities beyond textbook references. I happen to believe the emdrive project will be resolved by experimentalists, not by self-appointed experts in google-fu or outdated college courses. Thus is blunt but so are you.I agree with you that Maxwell is a brick wall. It assumes perfection in reflection
to balance out equations and make the universe whole again.
In a recent post, I already explained how Maxwell's equations handle not perfect reflection. If you want to make false statements like that Maxwell requires perfect reflection from conductors, or that metals are somehow transparent to oscillating magnetic fields (there is a wealth of data saying they aren't if you bothered to look), you are the one acting like a brick wall.
I encourage theorists to think beyond Maxwell and look for other possibilities. Who could argue with that accept conservative science fundamentalists? Most of the universe is unknown energy and matter and there is no grand unification theory. What is happening with the emdrive could be a tiny crack in the brick wall of existing knowledge.
Before I can look for other theories, I would need an experiment telling me what to look for. You have recently been promoting ideas that contradict countless experiments. You haven't pointed out a single piece of experimental data that would make college classes outdated. -
#2065 by Monomorphic on 24 Jan, 2017 20:13
-
Hello people -What mode is this? - I'm playing with a pretty standard frustum and a loop antenna in the time domain with meep. I have been for sometime working with this and other similar geometry. Some may recall my previous posts inquiring of the proper excitation of a loop antenna. My current loop antenna incorporates the lessons learned from that prior discussion.
I still have difficulty calculating an identifiable resonant mode. For a long time I couldn't get any clear images of the modes. I thought the reason was because these configurations had two resonances side by side. So close together that in fact meep could not separate them unless the source bandwidth was so narrow that the run would take weeks. I did finally find a geometry where the two resonant frequencies could be individually isolated. The attached gifs are of those two frequencies individually.
The resonant frequencies are 2.45246 GHz (the spinny one) and 2.43589 GHz (not spinny). The images are of the ex field component. The modes look nothing like the static images posted here on NSF. Does anyone care to explain why the fields are so "not typical" of images created by FEKO and COMSOL?
What are your frustum dimensions? It's hard to tell without the poynting arrows, but the top one looks like TE211. -
#2066 by Rodal on 24 Jan, 2017 20:15
-
Hello people -What mode is this? - I'm playing with a pretty standard frustum and a loop antenna in the time domain with meep. I have been for sometime working with this and other similar geometry. Some may recall my previous posts inquiring of the proper excitation of a loop antenna. My current loop antenna incorporates the lessons learned from that prior discussion.
Suggestion for any analyst to debug a numerical model (input, discretization mesh, boundary conditions, etc.) of a resonant cavity:
I still have difficulty calculating an identifiable resonant mode. For a long time I couldn't get any clear images of the modes. I thought the reason was because these configurations had two resonances side by side. So close together that in fact meep could not separate them unless the source bandwidth was so narrow that the run would take weeks. I did finally find a geometry where the two resonant frequencies could be individually isolated. The attached gifs are of those two frequencies individua
lly.
The resonant frequencies are 2.45246 GHz (the spinny one) and 2.43589 GHz (not spinny). The images are of the ex field component. The modes look nothing like the static images posted here on NSF. Does anyone care to explain why the fields are so "not typical" of images created by FEKO and COMSOL?
where "MEEP" stands for any numerical code
1. First run the simplest problem a MEEP model of a problem that has an exact solution, for example a cylindrical cavity, that has well known solutions https://en.wikipedia.org/wiki/Microwave_cavity#Cylindrical_cavity.
Only when one has a MEEP numerical model that coincides with the exact solution (eigenfrequencies and eigenmodes) then proceed to the next step
2. Run a MEEP model for a truncated cone for which we have good experimental comparisons. We have compared exact solutions, and COMSOL, FEKO, EmPro and ANSYS numerical models with NASA's experiments (for example TE012 without a dielectric).
See: https://forum.nasaspaceflight.com/index.php?topic=39214.msg1470390#msg1470390
Only at that point one has the confidence that the Meep numerical model is robust and valid, to then be able to discuss the validity of mode shapes not encountered in other numerical simulations. (Not because of a problem with Meep itself, but a problem with the input discretization, boundary conditions, etc. in the modeling itself)
QUESTION: Did you run Meep models that coincided with the exact solution for a cylindrical cavity, and that coincided with NASA's truncated cone (TE012 with no dielectric) experiment ?
-
#2067 by aero on 24 Jan, 2017 20:23
-
This is your Integer frustum - 30x18x24 cm. And just in case the antenna spec's got confused in the communication, the antenna is a 2 cm diameter, (one cm radius) loop located 1 cm from the center of the small end.Hello people -What mode is this? - I'm playing with a pretty standard frustum and a loop antenna in the time domain with meep. I have been for sometime working with this and other similar geometry. Some may recall my previous posts inquiring of the proper excitation of a loop antenna. My current loop antenna incorporates the lessons learned from that prior discussion.
I still have difficulty calculating an identifiable resonant mode. For a long time I couldn't get any clear images of the modes. I thought the reason was because these configurations had two resonances side by side. So close together that in fact meep could not separate them unless the source bandwidth was so narrow that the run would take weeks. I did finally find a geometry where the two resonant frequencies could be individually isolated. The attached gifs are of those two frequencies individually.
The resonant frequencies are 2.45246 GHz (the spinny one) and 2.43589 GHz (not spinny). The images are of the ex field component. The modes look nothing like the static images posted here on NSF. Does anyone care to explain why the fields are so "not typical" of images created by FEKO and COMSOL?
What are your frustum dimensions? It's hard to tell without the poynting arrows, but the top one looks like TE211. -
#2068 by aero on 24 Jan, 2017 20:58
-
Dr. Rodal,
If there were money riding on this I might take the time to do as you suggest in detail. But do either of us think that MIT, not to mention their government sponsors of meep development, would actually pawn off a faulty FDTD code on their students or the world? No, it is safe to say that with over 10,000 users, meep is thoroughly debugged. It does what it does, just as I expect FEKO and COMSOL do what they do.
As for verifying eigenfrequencies and eigenmodes - the images aren't associated with the Harminv module so frequencies are mute, although I did use Harminv to determine the drive frequency initially. The cavity pretty obviously resonates. Will the eigenmodes tell me what the values of m,n,p are? The visual difficulty IMO results from the extended, time dependent source of a loop antenna. Do you know what the images in a cylindrical resonant cavity should look like using an extended, time dependent source? That is really the heart of the problem.
Steve
-
#2069 by Rodal on 24 Jan, 2017 21:05
-
Dr. Rodal,
Not at all. MIT is my alma mater. I have my S.B., S.M. and Ph.D. degrees in Aeronautics & Astronautics from MIT, and have developed new Finite Difference and Finite Element algorithms and computer programs throughout my work at MIT, and after that, for very nonlinear problems in the time domain (using conditionable stable and unconditionally stable methods). I was not referring to anything faulty with Meep, on the contrary I was suggesting you to do what I (and all of my colleagues involved in numerical analysis) have always done to check any numerical model: I was specifically addressing the input, discretization, antenna excitation modeling and boundary conditions imposed by you in your model. Enough said.
If there were money riding on this I might take the time to do as you suggest in detail. But do either of us think that MIT, not to mention their government sponsors of meep development, would actually pawn off a faulty FDTD code on their students or the world? No, it is safe to say that with over 10,000 users, meep is thoroughly debugged. It does what it does, just as I expect FEKO and COMSOL do what they do....
-
#2070 by Monomorphic on 24 Jan, 2017 21:20
-
This is your Integer frustum - 30x18x24 cm. And just in case the antenna spec's got confused in the communication, the antenna is a 2 cm diameter, (one cm radius) loop located 1 cm from the center of the small end.
I'm not sure. Don't quote me on the TE211 as I would need to see more vector components to say for sure. When I sweep that frustum in FEKO I get TE013 and another mode at 2.474Ghz.
-
#2071 by WarpTech on 24 Jan, 2017 22:30
-
Hello people -What mode is this? - I'm playing with a pretty standard frustum and a loop antenna in the time domain with meep. I have been for sometime working with this and other similar geometry. Some may recall my previous posts inquiring of the proper excitation of a loop antenna. My current loop antenna incorporates the lessons learned from that prior discussion.
I still have difficulty calculating an identifiable resonant mode. For a long time I couldn't get any clear images of the modes. I thought the reason was because these configurations had two resonances side by side. So close together that in fact meep could not separate them unless the source bandwidth was so narrow that the run would take weeks. I did finally find a geometry where the two resonant frequencies could be individually isolated. The attached gifs are of those two frequencies individually.
The resonant frequencies are 2.45246 GHz (the spinny one) and 2.43589 GHz (not spinny). The images are of the ex field component. The modes look nothing like the static images posted here on NSF. Does anyone care to explain why the fields are so "not typical" of images created by FEKO and COMSOL?
The only way I can see getting this pattern is if the frequency is much higher than you think it is, or the antenna length is much longer than you think it is. Like 8X longer. -
#2072 by aero on 24 Jan, 2017 23:27
-
Quote
The only way I can see getting this pattern is if the frequency is much higher than you think it is, or the antenna length is much longer than you think it is. Like 8X longer.
It's neither of those. Now that you mention it though, I do note that the circumference of the frustum at the big end is exactly 15 times larger than the circumfernce of the loop. (30/2) That means the current in the loop passes a fixed point on the circumference of the big end 15 times while while the What? goes around the cavity once. But it is 14 cm from that point on the loop to the point on the circumference. I guess I'm not up to ray tracing it out today. -
#2073 by WarpTech on 25 Jan, 2017 00:21
-
Quote
The only way I can see getting this pattern is if the frequency is much higher than you think it is, or the antenna length is much longer than you think it is. Like 8X longer.
It's neither of those. Now that you mention it though, I do note that the circumference of the frustum at the big end is exactly 15 times larger than the circumfernce of the loop. (30/2) That means the current in the loop passes a fixed point on the circumference of the big end 15 times while while the What? goes around the cavity once. But it is 14 cm from that point on the loop to the point on the circumference. I guess I'm not up to ray tracing it out today.
Just thinking out loud here... You are plotting Ex (_ex) correct? Yet, the TE modes will have E as circles around the frustum. So the (amplitude) of the plot you need to make is,
E = sqrt(Ex2 + Ey2)
Then you should see circles for the E fields, and these should oscillate as a whole at, 2.449 GHz.
-
#2074 by Envelopesqueeze on 25 Jan, 2017 00:30
-
I read the NIST article (https://www.nist.gov/news-events/news/2017/01/nist-physicists-squeeze-light-cool-microscopic-drum-below-quantum-limit) regarding the alleged observation of a squeezed microwave laser reducing the quantum noise below the supposed quantum limit. Could this effect induce a force in the "real" world? Is the EMDrive, as designed, causing the squeezing of EM waves resulting in this effect? I'm interested if the "reduced quantum noise" effect occurs in proximity to the squeezing or where the squeezed wave impacts something. Any thoughts?
LOL
You mean induce a force in the realm of common experience? Oh, yes!
https://en.wikipedia.org/wiki/Radiation_implosion
But first the premise; that the 'quantum' world isn't the "real" world. If we apprehend the science, we understand the "real" world far more accurately than our common, net experience, our delusional meta-narrative. Alas, Feynman, one of its inventors, says neither he, nor anyone else does.
But anyways, the article's title implies the subject is a class of opto-mechanic phenomena, which I alone here, it seems, have the honor and distinction of purporting the EM Drive mechanism to also be.
Here is another video I got by searching youtube for "optomechanics":
See about 9:50 - 20:00 for a clearer description than in the video I linked a few days ago.
Squeezing light is mostly, AFAIK, about improving sensitivity or resolution limited by Heisenberg uncertainty, by finessing parameters.
Closest other applications I can think of are optical tweezers used in biology, levitation and cooling in physics.
The EM Drive, for space and transport, is a novel application of optomechanics. AFAIK.
Well, now you're not alone mwvp. Thanks for the reference and fantastic video. I need to get smart on this stuff before I start shooting lasers at copper planters.
The quantum squeezing is actually limiting the wave to one direction. I know it reduces the noise significantly and can be accomplished with an Optical parametric oscillator. I'm thinking manipulation of the quantum back action is increased by reducing the noise of the incoming microwave into the resonant cavity, as described in the NIST article.
I'm wondering if the filtering of the microwaves going into current prototypes showed differences that would support the optomechanical explaination. -
#2075 by aero on 25 Jan, 2017 01:05
-
Quote
The only way I can see getting this pattern is if the frequency is much higher than you think it is, or the antenna length is much longer than you think it is. Like 8X longer.
It's neither of those. Now that you mention it though, I do note that the circumference of the frustum at the big end is exactly 15 times larger than the circumfernce of the loop. (30/2) That means the current in the loop passes a fixed point on the circumference of the big end 15 times while while the What? goes around the cavity once. But it is 14 cm from that point on the loop to the point on the circumference. I guess I'm not up to ray tracing it out today.
Just thinking out loud here... You are plotting Ex (_ex) correct? Yet, the TE modes will have E as circles around the frustum. So the (amplitude) of the plot you need to make is,
E = sqrt(Ex2 + Ey2)
Then you should see circles for the E fields, and these should oscillate as a whole at, 2.449 GHz.
Is it as simple as that?
Yes, my notation ex is Ex. The underscore is a separator.
Not totally simple because I'll have to convert the big end slices to csv files, use Octave to square the files and add them together and take the square root of the summed data, then use Octave again to plot the files, and convert to a gif, but that is a viable path to images of E(x,y). Should I be concerned about the non-zero Ez component? And should I do the same type of thing for the X-vu of the cavity, ie.. ey^2 + ez^2 ?
Thanks!
A thought - actually, Python might be a more flexible tool to code this in? -
#2076 by mwvp on 25 Jan, 2017 02:09
-
I read the NIST article (https://www.nist.gov/news-events/news/2017/01/nist-physicists-squeeze-light-cool-microscopic-drum-below-quantum-limit) regarding the alleged observation of a squeezed microwave laser reducing the quantum noise below the supposed quantum limit.
...
...
Squeezing light is mostly, AFAIK, about improving sensitivity or resolution limited by Heisenberg uncertainty, by finessing parameters.
Well, now you're not alone mwvp. Thanks for the reference and fantastic video. I need to get smart on this stuff before I start shooting lasers at copper planters.
Thanks, and you're welcome. Too often I have read a little about something and found later, after much thought and work, I should have studied the matter. I'm afraid, IMHO that's too common here. But considering the body of knowledge on EM, electronics, photonics, physics and chemistry is so vast and deep, perhaps I've wasted time reading and researching too much, and building experiments and running models too little. There's a golden mean. The road is narrow....
The quantum squeezing is actually limiting the wave to one direction. I know it reduces the noise significantly and can be accomplished with an Optical parametric oscillator. I'm thinking manipulation of the quantum back action is increased by reducing the noise of the incoming microwave into the resonant cavity, as described in the NIST article.
I'm wondering if the filtering of the microwaves going into current prototypes showed differences that would support the optomechanical explaination.
As an engineer, I tend to think in terms of interfacing to sensors and actuators. If I want to listen to vibrations from a very sensitive microphone, I would take measures to well filter the power supply to it, and following amplifier stages.
Gravitational wave sensors, such as LIGO, are optomechanical devices. There are a number of youtube videos about them (search youtube "radiation pressure gravitational wave") that discuss quantum noise, shot noise, et.
To get the extreme sensitivity, light is "squeezed" in amplitude but only at the expense of phase, because, as I understand it, poor 'ol Heisenberg can be tortured, but not broken.
But if I'm building a power supply for a motor, I don't care much about how "clean" the power supply is. (Unless the motors put so much interference on the supply, it messes up the receiver or video on, say, a quadcopter, et.). Filtering the supply won't move an actuator or motor any better.
I see, as an optomechanic system, the EM Drive as a sort of motor. The frustrum is a cavity resonator. As has been discussed here recently (see the archive), different tuning points to feed it at, how to feed it, what to feed it with, et.
So it sounds kind of absurd, putting a hepa filter on a truck carburetor, a sub-micron fuel filter. Overkill. The author of the NIST article shouldn't use "magic" WRT photons.
I'm finding that another characteristic of fringe science; if you don't understand something well, and it has "exciting" potential touted by the illiteratti, it can lead to off the wall conclusions. Today it's squeezed light, yesterday new particles, dark model evanescence, extraordinary spin. We should keep a running list.
But then again, lately I've been considering orbital angular momentum, and those spinny modes Aero just posted, and has in the past. Perhaps light is being twisted and squeezed, acquiring extraordinary mass along with greater than h-bar/2 spin momentum?
Hmmmmmmmmmmm
-
#2077 by M.LeBel on 25 Jan, 2017 02:15
-
...
There have been other strange statements about electromagnetism made in the last few pages. For example, to pick just one of them, the proposal that only a time-varying magnetic field will show up outside without any time-varying electric field !!
In a recent post, I already explained how Maxwell's equations handle not perfect reflection. If you want to make false statements like that Maxwell requires perfect reflection from conductors, or that metals are somehow transparent to oscillating magnetic fields (there is a wealth of data saying they aren't if you bothered to look), you are the one acting like a brick wall.
Certainly White, McCulloch or Shawyer have not even remotely proposed something so, what are the words ?, bizarre like that.
It has been known for a century that special relativity makes the existence of time varying electric fields an inevitable consequence of the existence of time varying magnetic fields and vice versa.
In the inertial system A moving relatively to the inertial system B, purely magnetic fields from B will look like a combination of electric and magnetic fields in A. According to relativity, both frames are equally fit to describe the phenomena and obey the same laws. If my memory is correct this is discussed in the undergraduate Lectures on Physics by Feynman.
If for the EM Drive claims to be true something as bizarre (and against the accumulated knowledge on electromagnetism) would be needed, that goes against electromagnetic relativity, then such an argument becomes (even if not intended), an argument against the reality of the EM Drive claims, because it contradicts all the accumulated knowledge on electromagnetism and copper walls.
.... It has been known for a century that special relativity makes the existence of time varying electric fields an inevitable consequence of the existence of time varying magnetic fields and vice versa. which makes inevitable that:
A) Time is real stuff, as real as E & B
B) That E & B are variations on the same theme: Time
There is nothing wrong with the statement or the equations. IMO, we simply stopped short of looking for what the equation means.
dE/dT = B dB/dT= E : how can it be otherwise?
So, we have EM waves.... [ E & B.... time rate variation ] ..... gravity ..... emDrive
Time rate variations or gravity are not bound inside the frustum ....
Food for thought... -
#2078 by demofsky on 25 Jan, 2017 02:40
-
Though I see no particular reason to expect dramatic results, I do agree with rfmwguy that simulation could be more complete. As far as I know, we have not seen simulations with a fine mesh through the frustum walls, nor any meaningful period of time dependent simulation. The classic boundary condition of no parallel electric field at a conducting surface is an approximation - electron response is at finite speed- albeit a very good one. There is also the question of a quantum treatment of the metal.
I think this cannot be emphasized enough. To further add to this (and all the furniture flying around!) there is also a question of what exactly is being modelled in the various simulations being ran in software like FEKO and the underlying assumptions/approximations inherent in the model. One should always be wary of over interpreting what a simulation represents and conscious of what it doesn't handle.
For instance, what are the implications of any coatings/oxidation on the outside of the fustrum and their associated dielectric properties. No models I am aware of in this forum have made allowances for dielectric differences between the inner and outer surface of the fustrum. This is certainly not required for tuning antennae within the fustrum and explains why no one has gone through the effort to model complex surface conditions.
I think Dave raises a good point about interaction between EM conditions on the outside of a thin copper fustrum. It would be very interesting to see if this can be measured more definitively experimentally or with a more comprehensive model such as mentioned by RERT above. -
#2079 by WarpTech on 25 Jan, 2017 04:58
-
Just thinking out loud here... You are plotting Ex (_ex) correct? Yet, the TE modes will have E as circles around the frustum. So the (amplitude) of the plot you need to make is,
E = sqrt(Ex2 + Ey2)
Then you should see circles for the E fields, and these should oscillate as a whole at, 2.449 GHz.
Is it as simple as that?
Yes, my notation ex is Ex. The underscore is a separator.
Not totally simple because I'll have to convert the big end slices to csv files, use Octave to square the files and add them together and take the square root of the summed data, then use Octave again to plot the files, and convert to a gif, but that is a viable path to images of E(x,y). Should I be concerned about the non-zero Ez component? And should I do the same type of thing for the X-vu of the cavity, ie.. ey^2 + ez^2 ?
Thanks!
A thought - actually, Python might be a more flexible tool to code this in?
If you have a non-zero Ez component, then the full magnitude of the field will be;
|E| = sqrt(Ex2 + Ey2 + Ez2)
Like I said, "IF" the mode is TE01x, then the frequency will be uniform for the circle in the xy plane. You can plot the above in each plane, in the way you said.
|E|xy = sqrt(Ex2 + Ey2 )
|E|yz = sqrt(Ey2 + Ez2 )
|E|zx = sqrt(Ez2 + Ex2 )
It is likely that what you are seeing is an artifact of using cartesian coordinates for the field, inside a circular cavity. I don't know since I don't program meep, but have fun!
to balance out equations and make the universe whole again.
