-
#1040 by OnlyMe on 31 Dec, 2015 17:06
-
....
(I thought that Shell and aero were expecting TE01 instead of TE12)(completely different mode !!! )
Actually the mode shown by the Meep model does not look like any of these modes, really...
One of those possibly stupid questions...
Wouldn't how the resonant microwaves are introduced affect what resonance looks like?
Shell, uses a dual waveguide insertion.., rfmwguy put the magnetron right into the large end plate...
Could you have the same mode excited in two different ways that then winds up producing different patterns, at any specific location, relative to the insertion point(s)? -
#1041 by SeeShells on 31 Dec, 2015 17:13
-
Things can get quite complicated when you must consider that the frustum isn't simply a cone with the resonance between the endplates alone but between the dual waveguide injectors. You have to look at it as almost two separate resonating cavities one playing off on the other. Microwave modes in a cavity can reflect of walls but also reflect off other modes to create a TE01 or about anything else you desire. It isn't a surprise when other modes can be present.
What model are you running?
The geometrical axisymmetry of the frustum of the cone is presently not being exploited. However, assuming full axisymmetry would only produce fully axisymmetric electromagnetic modes. One would not be able to get the TM212 mode that NASA obtains in their experimental measurements for example.
A number of modes are not fully axisymmetric but display m-fold symmetry (where "m" is the first quantum number in TEmnp or TMmnp modes). The following images shows the lowest TEmn and TMmn modes, for arbitrary "p":
Only modes with m=0 are fully axisymmetric (for example TE012) . For m=1 one has to model half of the circular cross-section (and one can impose symmetry on the boundaries). For higher m>1 one has to model "smaller pie slices". So, it looks like one could at least save 50% of the mesh by exploiting axisymmetry
More problematic, it would preclude non-fully axisymmetric modes. To exploit axisymmetry one would have to determine what is the maximum number of poles around the circumference one wants to model: it would effectively set a pre-defined limit on the "m" and "n" quantum numbers that the model could model for TEmnp and TMmnp modes.
This is further complicated by the fact that Meep has revealed asymmetric modes not present in cylindrical cavities. Imposing axysymmetry would get rid of such asymmetric modes. For example, an asymmetric placement of an antenna, or an asymmetric placement of a waveguide can excite asymmetric modes in a real cavity, and it is useful for the designer to know this.
Actually, one of the greatest contributions of Meep analysis has been to make this asymmetric modes evident, and show how difficult it is to achieve axisymmetric resonant modes with antennas.
Well...here's my first shot at computing an endplate 'energy' picture from the meep data.
I hesitate to call this the thermal signature because I'm guessing at the equation.
This picture was generated with the following algorithm:
(where each frame element is the <x,y,z> H field 3 vector)
for each frame 1..112
for each row
for each col
sum[row][col] += abs( VCos_Angle(data[frame][row][col] , z) ) * vlength(data[frame][row][col]);
This sum is then output as the height of a cylinder and colored as in the picture. Elements whose resulting value < 1e-6 are ignored. The "MAX" shown is automatically computed as the highest value of the sum array. The MIN is 0 to 3 decimal points as shown. Values are in 'meep units'.
This was my best guess at a reasonable way to compute this. If a different algorithm is desired, let me know.
This meep model file is from aero modeling SeeShells' device. It is the same data as used to generate the H field animations I've been posting.
It looks pretty darn close to both TM12 and TE12!
Doing this for the small end is pretty trivial. Doing it for the conical section is a son-of-a-gun, but I'm working on it
QuoteThis meep model file is from aero modeling SeeShells' device. It is the same data as used to generate the H field animations I've been posting.
(I thought that Shell and aero were expecting TE01 instead of TE12)(completely different mode !!! )
Actually the mode shown by the Meep model does not look like any of these modes, really...
That's the reason why I had suggested making this verification comparison
All Meep runs need verification to compare with reality The problem is not with Meep, the issue is with the particular models
-
#1042 by Rodal on 31 Dec, 2015 17:38
-
...
The issues appear to be with the Meep model. The electromagnetic field distribution in the Meep model has not been verified vs. exact solutions, experiments and independent COMSOL models insofar as the electromagnetic fields are concerned. To start with, 1% of a microsecond is thousands of times less than the tens of microseconds required to establish the standing waves for resonance. There are also many issues surrounding pre and post-processing of the data and convergence of the Meep model.
Things can get quite complicated when you must consider that the frustum isn't simply a cone with the resonance between the endplates alone but between the dual waveguide injectors. You have to look at it as almost two separate resonating cavities one playing off on the other. Microwave modes in a cavity can reflect of walls but also reflect off other modes to create a TE01 or about anything else you desire. It isn't a surprise when other modes can be present.
The exponential growth of the electromagnetic fields has not been addressed either.
Kudos to VaxHeadRoom for looking at this.
One cannot take for granted the output of a numerical model: it needs to be verified.
The huge error of 113 times on the material input leading to an unreasonable Q (and hence to unreasonable damping of electromagnetic modes) was only addressed during the past few days...
-
#1043 by rfmwguy on 31 Dec, 2015 17:49
-
I'll add to that with some silly thoughts of my own. Visually, a cavity "in tune" I think has static TE/TM fields in an analogy to the acoustic resonance of a cavity. Not the same, I know, but its how my mind works....
(I thought that Shell and aero were expecting TE01 instead of TE12)(completely different mode !!! )
Actually the mode shown by the Meep model does not look like any of these modes, really...
One of those possibly stupid questions...
Wouldn't how the resonant microwaves are introduced affect what resonance looks like?
Shell, uses a dual waveguide insertion.., rfmwguy put the magnetron right into the large end plate...
Could you have the same mode excited in two different ways that then winds up producing different patterns, at any specific location, relative to the insertion point(s)?
Asymmetry adds layers of complexity that the meepers have had to overcome, slowly but surely. Considering this I would postulate that any variation of antenna placement, frequency, dimensions (due to heating) could throw it into another mode or prevent one from stabilizing.
Now...this is simply my opinion as nothing like this size and shape has been evaluated to my knowledge in the past. Its counterintuitive, a closed asymmetric cavity...I would have considered something like this defective
The only shape I have ever used like this is a 50 to 75 ohm rigid hardline TAPERing matching adapter, but that was not an open-ended system.
So, when I first discovered emdrive, it appeared to me a 50 to 75 ohm closed adapter/reducer, which would cause chaos to the load source (if that makes sense).
-
#1044 by Rodal on 31 Dec, 2015 17:58
-
This speculation reminds one of people speculating at previous Meep output results:
1) FRACTALS due to extremely small numbers. A well known numerical artifact to people experienced with Finite Difference and Finite Element methods. People unfamiliar with Finite Difference and Finite Element methods in these threads started to speculate that the fractals in the solution meant something exciting about the EM Drive. The fractals are a numerical artifact associated with the coarse mesh and the fact that the post processed images were hiding the numerical value of the fields (which were close to zero).
2) ELECTROMAGNETIC FIELDS OUTSIDE THE EM Drive due to extremely small numbers in those regions. A well known numerical artifact to people experienced with Finite Difference and Finite Element methods. People unfamiliar with Finite Difference and Finite Element methods in these threads started to speculate that the electromagnetic fields outside the EM Drive, known to be 20 orders of magnitude smaller than the inside fields meant something exciting about the EM Drive. The non-zero fields are a numerical artifact associated with matrix inversion in the simultaneous solution of equations in the numerical solution.
3) QUALITY OF RESONANCE (Q) in the tens of millions. Some people started to speculate that this meant something exciting about the EM Drive. The huge Q (in the superconducting range) was due to an error in the material input to Meep of a factor of 113 times. It was a well known issue of Garbage In - Garbage Out, to people experienced with Finite Difference and Finite Element methods.
Now people are speculating about the electromagnetic field mode shapes. They are looking at only 1% of a microsecond of the Finite Difference solution, which is thousands of times smaller than the time required to establish standing waves for resonance The fact that this early transient response is sensitive to initial conditions and the mesh that may not necessarily have a physical significance.
Finite Difference models need to be verified vs. exact solutions, experimental data and independent numerical models to eliminate garbage-in, garbage-out modeling issues and all kinds of issues related to the Finite Difference mesh in time and space, as well as post-processing issues.
-
#1045 by ThereIWas3 on 31 Dec, 2015 18:01
-
I think this would be an interesting shape to investigate. It is an exponential horn with a flat plate at the small end (where the sides are nearly parallel) and spherical at the other, with radius chosen so that there is a right angle where it meets the flaring horn.
RF to be introduced at the center of the small end in the center of the small plate. The plate could be moved in and out to adjust for resonance.
Fabrication might be difficult, unless you had access to a manufacturer of brass instruments.
-
#1046 by SteveD on 31 Dec, 2015 18:02
-
It looks like a cylindrical cavity resonator. The basic result for both orientations is almost the same (increasing weight measurement of the counterweight at the balance).
Looks strong like thermal effects like convection / ballooning...
Edit: This is an important test and must be compared with a similar but conical cavity! Maybe one can eliminate (most of) the thermal component going this way.
It looks like a frustum to me, not a cylinder.
But you're right that he obtains very similar values of thrust with the frustum pointing in opposite directions.
Therefore, it's not a confirmation of the EM drive effect.
If I remember correctly Iulian obtained different thrust numbers for the up/down directions.
However, if the thermal effect thrust with such a setup is indeed of the order of 1g, that casts a lot of doubts on Iulian's results too: they might also be explained by thermal effects.
Pardon me but what (mod - removal) are the two of you talking about? I just watched that video and it seemed to me that on the first test there was a, slowly increasing (!), actual downward force followed by a swift move upward when thermal expansion put the frustum out of resonance. The second run is inconclusive (I'm espeically worried that the device seems to get heavier due to changes in the balance beam setup) but seems to show an increased lift that falls back when the thing goes out of resonance.
My suggestions to the testers would be 1. see my earlier post about junking the balance beam and moving to some form of old fashion, analogy, scale and 2. if he's turning it on and office by manipulating the power at a distance, implement an led or other light in the camera's field of vision that will go on and off to show the power settings of the device.
If you're claiming a different interpretation, then list scale readings with timestamps (and preferably translation). -
#1047 by SeeShells on 31 Dec, 2015 18:07
-
I will agree there are serious questions remaining in the meep analysis to be answered but it is the only real simulation tool we have. I puzzeled for a long time trying to make sense of some of the questions meep opened up and if this was our only thing to question I'd be a very happy lady....
The issues appear to be with the Meep model. The electromagnetic field distribution in the Meep model has not been verified vs. exact solutions, experiments and independent COMSOL models insofar as the electromagnetic fields are concerned. To start with, 1% of a microsecond is thousands of times less than the tens of microseconds required to establish the standing waves for resonance. There are also many issues surrounding pre and post-processing of the data and convergence of the Meep model.
Things can get quite complicated when you must consider that the frustum isn't simply a cone with the resonance between the endplates alone but between the dual waveguide injectors. You have to look at it as almost two separate resonating cavities one playing off on the other. Microwave modes in a cavity can reflect of walls but also reflect off other modes to create a TE01 or about anything else you desire. It isn't a surprise when other modes can be present.
The exponential growth of the electromagnetic fields has not been addressed either.
Kudos to VaxHeadRoom for looking at this.
One cannot take for granted the output of a numerical model: it needs to be verified.
The huge error of 113 times on the material input leading to an unreasonable Q (and hence to unreasonable damping of electromagnetic modes) was only addressed during the past few days...
I remember just a few short months ago aero and you trying to even get meep to display any resonance that could be made sense of without looking like a painting from Timothy Leary. With MaxHeadroom and a few others stepping up to the plate I believe we will have a good tool we can make good basic models.
Verification will come with time and we have to wait for it. Looking at what and how meep works with a dose of objectivity is smart.
Shell
PS: On another note. Got my data back all 20,000 pieces of it and will be working on getting the software re-installed.
HAPPY NEW YEAR as this is going to be a Smokin' New Year! -
#1048 by ThereIWas3 on 31 Dec, 2015 18:25
-
I remember just a few short months ago aero and you trying to even get meep to display any resonance that could be made sense of without looking like a painting from Timothy Leary.
I find looking at the pictures more informative about when resonance occurs than the mathematical analysis of "Q" done by Meep. When you can just count an integral number of full cycles from one end to the other, and the amplitudes peak, you know that something is going on. -
#1049 by SteveD on 31 Dec, 2015 18:27
-
I think this would be an interesting shape to investigate. It is an exponential horn with a flat plate at the small end (where the sides are nearly parallel) and spherical at the other, with radius chosen so that there is a right angle where it meets the flaring horn.
RF to be introduced at the center of the small end in the center of the small plate. The plate could be moved in and out to adjust for resonance.
Fabrication might be difficult, unless you had access to a manufacturer of brass instruments.
Rf insertion in the small end might produce a null as oer Rfmwguy's first run. -
#1050 by SeeShells on 31 Dec, 2015 18:29
-
I think this would be an interesting shape to investigate. It is an exponential horn with a flat plate at the small end (where the sides are nearly parallel) and spherical at the other, with radius chosen so that there is a right angle where it meets the flaring horn.
Like it...
RF to be introduced at the center of the small end in the center of the small plate. The plate could be moved in and out to adjust for resonance.
Fabrication might be difficult, unless you had access to a manufacturer of brass instruments.
Shorten the small section and hit a TE01 in the large cavity. Make it tuneble with the small plate coupled to the large via a Quartz rod through the center. The horn can now grow with thermal expansion and not loose resonance.
Shell
Please excuse my PC Paint drawing, I still have to install my graphics programs.
-
#1051 by Rodal on 31 Dec, 2015 18:35
-
...
Now...this is simply my opinion as nothing like this size and shape has been evaluated to my knowledge in the past. Its counterintuitive, a closed asymmetric cavity...
<< nothing like this size and shape has been evaluated to my knowledge in the past.>> ?
Huh ?
Closed asymmetric cavities as well as truncated cone waveguides have been analyzed in peer-reviewed papers prior to the advent of the EM Drive .
Previous threads discussed exact solutions of a frustum of a cone going back to the late 1930's (prominently Schelkunoff (*), including pictures of Schelkunoff standing at Bell Labs next to truncated cone microwave guides in the early 1930’s)
as well as a number of papers dealing with applications of truncated cone cavities (most prominently the paper by Mayer, Reccius and Knochel in the IEEE transactions 1992). All this work at a much higher level than the one published for the EM Drive.
I have quoted papers related to the fact that the waveguide cut-off frequency is known to be not applicable to closed cavities, including truncated cones.
Mayer, Reccius and Knochel in the IEEE transactions 1992 even remarkQuoteA conical cavity is described. The new cavity is superior to the often used cylindrical cavity, because it intrinsically does not suffer from mode degeneration
Title: Conical cavity for surface resistance measurements of high temperature superconductors
Authors: Mayer, B.; Reccius, A.; Knochel, R.
Publication: IEEE Transactions on Microwave Theory and Techniques, vol. 40, issue 2, pp. 228-236
http://adsabs.harvard.edu/abs/1992ITMTT..40..228M
Also the papers by:
De Villiers and Meyer (including De Villiers PhD thesis)
Zeng and Fan for conical waveguide (discussed at length with WarpTech, including my finding. pointing out, and correcting errors in Zeng and Fan's paper)
Davis and Scharstein paper for an open-ended conducting frustum
Amin et al paper for multilayer conical dielectric waveguides
Greg Egan's solution in the web
Patents on conical cavities, truncated cones and conical waveguides (unrelated to the EM Drive) discussed in previous threads
__________
(*) Schelkunoff was born in Samara, Russia in 1897, attended the University of Moscow before being drafted in 1917. He crossed Siberia into Manchuria and then Japan before settling into Seattle in 1921. There he received bachelor's and master's degrees in mathematics from the State College of Washington, now the University of Washington, and in 1928 received his Ph.D. from Columbia University for his dissertation On Certain Properties of the Metrical and Generalized Metrical Groups in Linear Spaces of n Dimension.
After receiving his degree, Schelkunoff joined Western Electric's research wing, which became Bell Laboratories. In 1933 he and Sally P. Mead began analysis of waveguide propagation discovered analytically by their colleague George C. Southworth. Their analysis uncovered the transverse modes. Schelkunoff appears to have been the first to notice the important practical consequences of the fact that attenuation in the TE01 mode decays inversely with the 3/2 power of the frequency. In 1935 he and his colleagues reported that coaxial cable, then new, could transmit television pictures or up to 200 telephone conversations.
He authored another IRE paper “Transrnission Theory of Pure Electromagnetic Waves,” published in November 1937. He treated the theory of spherical waves in a 1938 paper in the Transactions of the American Institute of Electrical Engineers and followed this with a 1939 IRE paper on the induced electromotive force method of computing radiation from antennas. In his September 1941 IRE paper, Schelkunoff addressed the ambitious topic of the “theory of antennas of arbitrary size and shape.” He explained that his mathematical analysis of antennas was “precisely the analysis appropriate to wave guides and electric horns.” He observed that:
We may also think of the antennas as the wall of an electric horn with an aperture so wide that one can hardly see the horn itself-just like a Cheshire cat: only the grin can be seen. Schelkunoff suggested that the physical picture which emerged from his mathematical analysis was “attractive to an engineer.” He began his analysis with Maxwell’s equations and hypothetical conical antennas and went on to show how to apply the results to antennas of other shapes although they were “definitely more complicated.” He concluded that he believed that “the antenna theory is in such a shape that accurate results can be calculated if all visible factors such as base capacitance and antenna shapes are taken into consideration.”
During his 35-year career at Bell Labs, Schelkunoff's research included radar, electromagnetic wave propagation in the atmosphere and in microwave guides, short-wave radio, broad-band antennas, and grounding.
-
#1052 by SeeShells on 31 Dec, 2015 20:11
-
Bingo!!!....
(I thought that Shell and aero were expecting TE01 instead of TE12)(completely different mode !!! )
Actually the mode shown by the Meep model does not look like any of these modes, really...
One of those possibly stupid questions...
Wouldn't how the resonant microwaves are introduced affect what resonance looks like?
Shell, uses a dual waveguide insertion.., rfmwguy put the magnetron right into the large end plate...
Could you have the same mode excited in two different ways that then winds up producing different patterns, at any specific location, relative to the insertion point(s)?
That is what I'm seeing in the meep models. From Antennas to waveguides. -
#1053 by ThereIWas3 on 31 Dec, 2015 20:25
-
Rf insertion in the small end might produce a null as oer Rfmwguy's first run.
Actually, I was thinking circular polarization, fired down the Z axis. I think interesting things might happen at places where the horn circumference is an integral multiple of the wavelength...
I think how the RF is injected could be a major factor in efficiency of the "effect".
This sort of guessing is what my EM prof told us was going through his head as he was inventing the helical beam antenna. "I figured if some dimension was related to a wavelength, it might work better." The fact that the beam came off the end actually surprised him, when he built the thing in his basement, using an oatmeal box as the form. -
#1054 by SeeShells on 31 Dec, 2015 20:43
-
that's not A magnetron on the end but a waveguide. The magnetron is a 1/4 wavelength snub antenna with a cap launcher on the end.I think this would be an interesting shape to investigate. It is an exponential horn with a flat plate at the small end (where the sides are nearly parallel) and spherical at the other, with radius chosen so that there is a right angle where it meets the flaring horn.
RF to be introduced at the center of the small end in the center of the small plate. The plate could be moved in and out to adjust for resonance.
Fabrication might be difficult, unless you had access to a manufacturer of brass instruments.
Rf insertion in the small end might produce a null as oer Rfmwguy's first run.
He might have been better off using the waveguide attached to the magnetron to the top like in the picture with a Z-match hole in the top plate. IMHO.
-
#1055 by SeeShells on 31 Dec, 2015 20:52
-
You can do a search for Post by : SeeShell, subject words: helicalRf insertion in the small end might produce a null as oer Rfmwguy's first run.
Actually, I was thinking circular polarization, fired down the Z axis. I think interesting things might happen at places where the horn circumference is an integral multiple of the wavelength...
I think how the RF is injected could be a major factor in efficiency of the "effect".
This sort of guessing is what my EM prof told us was going through his head as he was inventing the helical beam antenna. "I figured if some dimension was related to a wavelength, it might work better." The fact that the beam came off the end actually surprised him, when he built the thing in his basement, using an oatmeal box as the form.
You will find some information where it was searched before. The big stopper was it couldn't be modeled in meep but hasn't been put to sleep on my end.
Shell -
#1056 by rfmwguy on 31 Dec, 2015 21:19
-
Upload for a friend's reference only...
-
#1057 by SteveD on 31 Dec, 2015 21:28
-
that's not A magnetron on the end but a waveguide. The magnetron is a 1/4 wavelength snub antenna with a cap launcher on the end.I think this would be an interesting shape to investigate. It is an exponential horn with a flat plate at the small end (where the sides are nearly parallel) and spherical at the other, with radius chosen so that there is a right angle where it meets the flaring horn.
RF to be introduced at the center of the small end in the center of the small plate. The plate could be moved in and out to adjust for resonance.
Fabrication might be difficult, unless you had access to a manufacturer of brass instruments.
Rf insertion in the small end might produce a null as oer Rfmwguy's first run.
He might have been better off using the waveguide attached to the magnetron to the top like in the picture with a Z-match hole in the top plate. IMHO.
How badly would using a resonance absorption isolator for the waveguide break things? -
#1058 by SeeShells on 31 Dec, 2015 21:41
-
Don't know, maybe a Lot of excess heat. Isn't it better to clean up the magnetron output before insertion?
that's not A magnetron on the end but a waveguide. The magnetron is a 1/4 wavelength snub antenna with a cap launcher on the end.I think this would be an interesting shape to investigate. It is an exponential horn with a flat plate at the small end (where the sides are nearly parallel) and spherical at the other, with radius chosen so that there is a right angle where it meets the flaring horn.
RF to be introduced at the center of the small end in the center of the small plate. The plate could be moved in and out to adjust for resonance.
Fabrication might be difficult, unless you had access to a manufacturer of brass instruments.
Rf insertion in the small end might produce a null as oer Rfmwguy's first run.
He might have been better off using the waveguide attached to the magnetron to the top like in the picture with a Z-match hole in the top plate. IMHO.
How badly would using a resonance absorption isolator for the waveguide break things?
Shell -
#1059 by SeeShells on 31 Dec, 2015 21:42
-
Why I'll not be playing around with my magnetron on New Years Eve.
Happy New Year Everyone!!!!
Shell
Huh ?