-
#1900 by Rodal on 18 Jan, 2016 19:41
-
It is the change in energy content that matters, using E = m c^2 therefore m = E /(c^2), which is extremely small due to the huge value of c^2. You also have the special relativity effect (gamma correction) which is tiny due to the extremely small velocities in EM Drive experiments.
Doc, has anyone done the math to show what the equivalent force would be at 1kW using Einstein's cylinder shape? Also, are there any reasons to assume an asymmetrical shape might not add to the effect?Einsteins paper...posted elsewhere, linked here as well. A cylinder in space and imparted motion due to radiation applied:
http://einsteinpapers.press.princeton.edu/vol2-trans/214?ajax
This was precisely the point made by NSF user WarpTech, (Todd we miss ya
) in classic exchanges he had in these threads with NSF user DeltaMass (we miss you too ). Unfortunately the EM Drive's claimed self-acceleration does not appear to be explainable on this basis because of...the relatively puny energy change involved in the EM Drive cannot be used to justify the claimed "anomalous force" (off by orders of magnitude...)
Back to the whiteboard and keep your eyes on the Notsosureofit hypothesis, who remains here (from time to time
)...
As per Einstein's paper, what matters is the center of mass (in the translation in the paper is called the center of gravity) and not the shape of the body. EM Drives's without dielectric inserts are mundane in this respect: homogeneous medium (air or vacum) inside the EM Drive, just a very minor change in geometry, so center of mass as per Einstein should still apply.
NASA's EM Drive with dielectric insert is more interesting: EM Drive becomes inhomogeneous due to the asymmetric presence of the dielectric and you have the Minkoswki-Abraham paradox still unresolved.
Also concerning the dielectric insert you also have several of the papers on CP-symmetry violation uncovered by Mulletron.
Various people have tried to explain the EM Drive as an open system using Minkowski's formulation, and you also have Woodward's Mach effect consideration. All of them subject to controversy, but at least not as mundane as Shawyer's EM Drive without a dielectric insert. -
#1901 by rfmwguy on 18 Jan, 2016 19:47
-
OK, if we take M = delta E/C2, and you have a CW source, it would be 1 instantaneous mass change. Sooo, pulsing multiple signals would be additive?
It is the change in energy content that matters, using E = m c^2 therefore m = E /(c^2), which is extremely small due to the huge value of c^2.
Doc, has anyone done the math to show what the equivalent force would be at 1kW using Einstein's cylinder shape? Also, are there any reasons to assume an asymmetrical shape might not add to the effect?Einsteins paper...posted elsewhere, linked here as well. A cylinder in space and imparted motion due to radiation applied:
http://einsteinpapers.press.princeton.edu/vol2-trans/214?ajax
This was precisely the point made by NSF user WarpTech, (Todd we miss ya ) in classic exchanges he had in these threads with NSF user DeltaMass (we miss you too ). Unfortunately the EM Drive's claimed self-acceleration does not appear to be explainable on this basis because of...the relatively puny energy change involved in the EM Drive cannot be used to justify the claimed "anomalous force" (off by orders of magnitude...)
Back to the whiteboard and keep your eyes on the Notsosureofit hypothesis, who remains here (from time to time )...
As per Einstein's paper, what matter's is the center of mass (in the translation in the paper is called the center of gravity) and not the shape of the body. EM Drives's without dielectric inserts are boring in this respect: homogeneous medium inside them, just a very minor change in geometry, so center of mass as per Einstein should still apply.
NASA's EM Drive with dielectric insert is more interesting: EM Drive becomes inhomogeneous due to the asymmetric presence of the dielectric and you have the Minkoswki-Abraham paradox still unresolved.
Also concerning the dielectric insert you have several of the papers and work uncovered by Mulletron.
Various people have tried to explain the EM Drive as an open system using Minkowski's formulation, and you also have Woodward's Mach effect consideration. All of them subject to controversy, but at least not as boring as Shawyer's EM Drive without a dielectric insert. -
#1902 by Rodal on 18 Jan, 2016 19:52
-
...
Mmmm. I think we need somebody like WarpTech to come back here to discuss this, as WarpTech was much more daring and open minded about looking at such explanations than I am -at the moment
OK, if we take M = delta E/C2, and you have a CW source, it would be 1 instantaneous mass change. Sooo, pulsing multiple signals would be additive? .
WarpTech is still a NSF member. Last post in another thread...
https://forum.nasaspaceflight.com/index.php?topic=36911.msg1455270#msg1455270 -
#1903 by rfmwguy on 18 Jan, 2016 19:59
-
OK, the only way I could see this doing anything meaningful is to have multiple frequencies with amplitude and phase change states almost instantaneously...kinda like a magnetr....OK, enough daydreaming for now...
Mmmm. I think we need somebody like WarpTech to come back here to discuss this, as WarpTech was much more daring and open minded about looking at such explanations than I am -at the moment
OK, if we take M = delta E/C2, and you have a CW source, it would be 1 instantaneous mass change. Sooo, pulsing multiple signals would be additive? .
WarpTech is still a NSF member. Last posted in another thread...
-
#1904 by graybeardsyseng on 18 Jan, 2016 20:40
-
Motivation alert - While some may be discouraged by the lack of any new info on emdrive, there is a "noble" reason to continue the search for new propulsion...while it may take decades, if we never start the race, we'll never finish it.
So, I give you yesterday's video link to the barge landing yesterday of SpaceX's Falcon 9. Carrying heavy propellant does have its disadvantages:
http://www.space.com/31653-spacex-rocket-landing-crash-droneship-video.html
There was a time when such concerns did not deter engineering studies of much more powerful space propulsion, performing actual testing with conventional explosives to test the concept (instead of microNewton testing) (look at 8 min, for example), and famous physicists like Freeman Dyson were intimately involved:
20 Astronauts, all-up mission. Return time from Mars: 42 days. Mars surface payload: 150 metric tons
https://en.wikipedia.org/wiki/Project_Orion_(nuclear_propulsion)
Thanks Doc - Brings back very fond memories and some wistful feelings of "what might have been".
Orion was a very viable and workable concept. It would just have taken Courage and Will and we would own the solar system by now. We could have beaten Clarke to Saturn (or Kubrik/Clarke to Jupiter if you prefer) well before 2001.
As someone who has personally worked and armed "physics packages" several times larger than Orion, it doesn't take magic or superhuman ability - just good careful physics and engineering.
I miss those days. Again - thanks and thanks for the great jpg. I had a physical poster similar to that one back in the 70's but it is long gone.
Herman
-
#1905 by Rodal on 18 Jan, 2016 21:42
-
...
Thanks Doc - Brings back very fond memories and some wistful feelings of "what might have been".
Orion was a very viable and workable concept. It would just have taken Courage and Will and we would own the solar system by now. We could have beaten Clarke to Saturn (or Kubrik/Clarke to Jupiter if you prefer) well before 2001.
As someone who has personally worked and armed "physics packages" several times larger than Orion, it doesn't take magic or superhuman ability - just good careful physics and engineering.
I miss those days. Again - thanks and thanks for the great jpg. I had a physical poster similar to that one back in the 70's but it is long gone.
Herman
https://forum.nasaspaceflight.com/index.php?topic=39004.msg1479057#msg1479057
Thanks, notice that the mission featured in the image that I attached was for one of the later iterations of the Orion design: a nuclear pulse propulsion module launched into earth orbit by Saturn V-25 (http://www.astronautix.com/lvs/satv25su.htm) rockets.
So the intention was not to detonate the nuclear bombs in the Earth's atmosphere. Even though it relied on Saturn V-25's to go into orbit, the mission's specific impulse was still several times that of a nuclear thermal system like NERVA.
A second Saturn V-25 launch would put the Mars spacecraft with the crew of 20 into orbit. After rendezvous and checkout, the combined spacecraft would set out on a round trip to Mars - with a total mission duration much shorter than presently envisioned under present projects !
-
#1906 by dustinthewind on 18 Jan, 2016 21:48
-
This was just posted on the other site. Things are now making a little more sense.I'll just have to make sure I fix that so I can have something other than questionable models, this includes FEKO and meep.
Shell
Nicely said.
TE01x mode has a unique end plate characteristic. The end plate currents do not cross over to the side walls & there is only 1 annular current ring. See attached.
No other TE & TM mode does this. Should be very easy to see this on any sim.
Have also read building in a thin but totally insulating gap between the end plates to the side walls suppresses all modes that need end plate to side wall currents to exits (so the mode can form). This should stop the degenerative but equal guide wavelength TM11x mode from trying to form / forming.
Anyway just look at the end plate for a single annular current ring. If it is not there, it is not TE01x mode.
<<Anyway just look at the end plate for a single annular current ring. If it is not there, it is not TE01x mode.>>
No, that can not be an electric current ring at the end plate of an EM Drive, that would only be true for an open waveguide without any metallic ends !!!
Again, the tangential electric field, parallel to a metal, must be zero at the metal end plates.
The only field that can show up at the end plates of the EM Drive having metal end plates for mode TE01p is a magnetic field.
People, we learned this at school: BOUNDARY CONDITIONS for Electric Fields, remember
NASA Eagleworks does not have this misunderstanding. NASA got the fields correct, including the boundary conditions from the very first COMSOL analysis run by Frank Davis.
The EM Drive metallic ends cannot be ignored. The EM Drive is not a waveguide with open ends !!!!
There are boundary conditions at the ends due to the metal !!!
The electric field at the metal end plates in mode TE012 is a big zero (see plot below)
Quote
Some quick notes.
The movies are mislabelled.
The TE01 mode should be TE10. It is not any reference to the frustum, but the dominant mode of the waveguides that are excited by the RF source. I should have not have mentioned the rectangular waveguide mode, it is confusing.
The Feko solver I'm using is method of moments.
The E-field magnitudes shown are not taken on the surface of the frustum (Where they are zero) but 2mm inside.
Feko calculates the standing waves. The animation in these movies are made my changing the phase of the RF sources so affecting the instantaneous E-field magnitude displayed. At 2.47 Ghz, this would be pretty fast!
The frustum walls are perfect conductors.
Note that the E-field scale is logarithmic. This can be misleading or helpful, I'm not sure. Will maybe try a linear scale next time.
I'll do an update with an improved model using copper and S-port measurements soon.
End Quote
In summary the image of the circular current ring is correct and unique to Edited: TE011 modes. TE being an electric field some where in the cavity circles around the concentric axis. The electric fields stored inside the cavity tend to match the shape of actual currents some where within the cavity. The circular current in the plate is due to approaching light where in Si units E/c = B or in CGS units E = B for light, and by (v x B) where v = c the approaching B field (light) induces an electric field which excites a current in the plate, during current build up. As a result of this "approaching light" the current accelerates generating another B field which constructively adds to the approaching light, but the B field generated by the excited current is now propagating away from the current or plate. So we now have counter-propagating waves. This wave leaving has the same B field but opposite velocity away from the big plate so by E = v x B or the right hand rule, the Electric field of departing light, cancels out any electric field from the Approaching light. Hence the electric field at the base plate is exactly zero. Any current in metal excited by light reacts this way as current is free to move and hence it is a boundary condition that the tangential electric field at the surface of a metal (which reflects light) is zero.
It is tempting to assume an electric field though because it is hard to imagine current accelerating with out first experiencing some electric field. On the other hand the light in the cavity is in perfect resonance with with the cavity wall current, so at equilibrium (maximum energy storage inside the cavity) this is the case. -
#1907 by Rodal on 18 Jan, 2016 21:56
-
This was just posted on the other site. Things are now making a little more sense.I'll just have to make sure I fix that so I can have something other than questionable models, this includes FEKO and meep.
Shell
Nicely said.
TE01x mode has a unique end plate characteristic. The end plate currents do not cross over to the side walls & there is only 1 annular current ring. See attached.
No other TE & TM mode does this. Should be very easy to see this on any sim.
Have also read building in a thin but totally insulating gap between the end plates to the side walls suppresses all modes that need end plate to side wall currents to exits (so the mode can form). This should stop the degenerative but equal guide wavelength TM11x mode from trying to form / forming.
Anyway just look at the end plate for a single annular current ring. If it is not there, it is not TE01x mode.
<<Anyway just look at the end plate for a single annular current ring. If it is not there, it is not TE01x mode.>>
No, that can not be an electric current ring at the end plate of an EM Drive, that would only be true for an open waveguide without any metallic ends !!!
Again, the tangential electric field, parallel to a metal, must be zero at the metal end plates.
The only field that can show up at the end plates of the EM Drive having metal end plates for mode TE01p is a magnetic field.
People, we learned this at school: BOUNDARY CONDITIONS for Electric Fields, remember
NASA Eagleworks does not have this misunderstanding. NASA got the fields correct, including the boundary conditions from the very first COMSOL analysis run by Frank Davis.
The EM Drive metallic ends cannot be ignored. The EM Drive is not a waveguide with open ends !!!!
There are boundary conditions at the ends due to the metal !!!
The electric field at the metal end plates in mode TE012 is a big zero (see plot below)
Quote
Some quick notes.
The movies are mislabelled.
The TE01 mode should be TE10. It is not any reference to the frustum, but the dominant mode of the waveguides that are excited by the RF source. I should have not have mentioned the rectangular waveguide mode, it is confusing.
The Feko solver I'm using is method of moments.
The E-field magnitudes shown are not taken on the surface of the frustum (Where they are zero) but 2mm inside.
Feko calculates the standing waves. The animation in these movies are made my changing the phase of the RF sources so affecting the instantaneous E-field magnitude displayed. At 2.47 Ghz, this would be pretty fast!
The frustum walls are perfect conductors.
Note that the E-field scale is logarithmic. This can be misleading or helpful, I'm not sure. Will maybe try a linear scale next time.
I'll do an update with an improved model using copper and S-port measurements soon.
End Quote
In summary the image of the circular current ring is correct and unique to TE01 modes. TE being an electric field some where in the cavity circles around the concentric axis. The circular current in the plate is due to approaching light (v x B) where the approaching B field (light) induces an electric field which excites a current in the plate. As a result of this "approaching light" the current accelerates generating another B field which constructively adds to the approaching light, but the B field generated by the excited current is now propagating away from the current or plate. So we now have counter-propagating waves. This wave leaving has the same B field but opposite velocity away from the big plate so by E = v x B or the right hand rule, the Electric field of departing light, cancels out any electric field from the Approaching light. Hence the electric field at the base plate is exactly zero. Any current in metal excited by light reacts this way as current is free to move and hence it is a boundary condition that the tangential electric field at the surface of a metal (which reflects light) is zero.
It is tempting to assume an electric field though because it is hard to imagine current accelerating with out first experiencing some electric field. On the other hand the light in the cavity is in perfect resonance with with the cavity wall current, so at equilibrium (maximum energy storage inside the cavity) this is the case.
In summary, it is very misleading to keep posting images about open waveguides when referring to a closed metallic cavity like the EM Drive, particularly when one is discussing the fields at the metal end plates.
As we learnt in school, mode shapes like TE01 are mode shapes for open waveguides.
It is misleading to characterize a closed cavity like the EM Drive as an open waveguide.
The EM Drive has metal plates at each end, hence it is not an open waveguide, it is a closed resonant cavity.
As such there is no such thing as a mode TE01 in a closed resonant cavity like the EM Drive.
As we learnt in school a closed cylindrical or conical resonant cavity has modes TE011 or TE012 or TE013, etc , for which there can not be an electric current ring produced by the electric field at the metallic end plates !!!
You are correct, the tangential electric field, parallel to a metal, must be zero at the metal end plates.
-
#1908 by dustinthewind on 18 Jan, 2016 22:19
-
In summary the image of the circular current ring is correct and unique to TE01 modes. TE being an electric field some where in the cavity circles around the concentric axis. The circular current in the plate is due to approaching light (v x B) where the approaching B field (light) induces an electric field which excites a current in the plate. As a result of this "approaching light" the current accelerates generating another B field which constructively adds to the approaching light, but the B field generated by the excited current is now propagating away from the current or plate. So we now have counter-propagating waves. This wave leaving has the same B field but opposite velocity away from the big plate so by E = v x B or the right hand rule, the Electric field of departing light, cancels out any electric field from the Approaching light. Hence the electric field at the base plate is exactly zero. Any current in metal excited by light reacts this way as current is free to move and hence it is a boundary condition that the tangential electric field at the surface of a metal (which reflects light) is zero.
It is tempting to assume an electric field though because it is hard to imagine current accelerating with out first experiencing some electric field. On the other hand the light in the cavity is in perfect resonance with with the cavity wall current, so at equilibrium (maximum energy storage inside the cavity) this is the case.
In summary, it is very misleading to keep posting images about open waveguides when referring to a closed metallic cavity like the EM Drive and one is discussing the fields at the end plates.
As we learnt in school, mode shapes like TE01 are mode shapes for open waveguides.
It is misleading to characterize a closed cavity like the EM Drive as an open waveguide.
The EM Drive has metal plates at each end, hence it is not an open waveguide, it is a closed resonant cavity.
As such there is no such thing as a mode TE01 in a closed resonant cavity like the EM Drive.
As we learnt in school a closed cylindrical or conical resonant cavity has modes TE011 or TE012 or TE013, etc , for which there can not be an electric current ring produced by the electric field at the metallic end plates !!!
Again, the tangential electric field, parallel to a metal, must be zero at the metal end plates.
Sorry TE011. Are the heat rings not induced by current flow and the resistance in the metal?
http://forum.nasaspaceflight.com/index.php?topic=36313.msg1352878#msg1352878
-
#1909 by Rodal on 18 Jan, 2016 22:24
-
In summary the image of the circular current ring is correct and unique to TE01 modes. TE being an electric field some where in the cavity circles around the concentric axis. The circular current in the plate is due to approaching light (v x B) where the approaching B field (light) induces an electric field which excites a current in the plate. As a result of this "approaching light" the current accelerates generating another B field which constructively adds to the approaching light, but the B field generated by the excited current is now propagating away from the current or plate. So we now have counter-propagating waves. This wave leaving has the same B field but opposite velocity away from the big plate so by E = v x B or the right hand rule, the Electric field of departing light, cancels out any electric field from the Approaching light. Hence the electric field at the base plate is exactly zero. Any current in metal excited by light reacts this way as current is free to move and hence it is a boundary condition that the tangential electric field at the surface of a metal (which reflects light) is zero.
It is tempting to assume an electric field though because it is hard to imagine current accelerating with out first experiencing some electric field. On the other hand the light in the cavity is in perfect resonance with with the cavity wall current, so at equilibrium (maximum energy storage inside the cavity) this is the case.
In summary, it is very misleading to keep posting images about open waveguides when referring to a closed metallic cavity like the EM Drive and one is discussing the fields at the end plates.
As we learnt in school, mode shapes like TE01 are mode shapes for open waveguides.
It is misleading to characterize a closed cavity like the EM Drive as an open waveguide.
The EM Drive has metal plates at each end, hence it is not an open waveguide, it is a closed resonant cavity.
As such there is no such thing as a mode TE01 in a closed resonant cavity like the EM Drive.
As we learnt in school a closed cylindrical or conical resonant cavity has modes TE011 or TE012 or TE013, etc , for which there can not be an electric current ring produced by the electric field at the metallic end plates !!!
Again, the tangential electric field, parallel to a metal, must be zero at the metal end plates.
Sorry TE011. Are the heat rings not induced by current flow and the resistance in the metal?
The heat rings at the metal end plates of resonant cavities with TE01p mode shapes are not the result of electric currents (charge) produced by the electric field . They are instead the result of induction heating, resulting from the magnetic field inducing eddy-currents on the metal.
-
#1910 by Rodal on 18 Jan, 2016 22:30
-
In summary the image of the circular current ring is correct and unique to TE01 modes. TE being an electric field some where in the cavity circles around the concentric axis. The circular current in the plate is due to approaching light (v x B) where the approaching B field (light) induces an electric field which excites a current in the plate. As a result of this "approaching light" the current accelerates generating another B field which constructively adds to the approaching light, but the B field generated by the excited current is now propagating away from the current or plate. So we now have counter-propagating waves. This wave leaving has the same B field but opposite velocity away from the big plate so by E = v x B or the right hand rule, the Electric field of departing light, cancels out any electric field from the Approaching light. Hence the electric field at the base plate is exactly zero. Any current in metal excited by light reacts this way as current is free to move and hence it is a boundary condition that the tangential electric field at the surface of a metal (which reflects light) is zero.
It is tempting to assume an electric field though because it is hard to imagine current accelerating with out first experiencing some electric field. On the other hand the light in the cavity is in perfect resonance with with the cavity wall current, so at equilibrium (maximum energy storage inside the cavity) this is the case.
In summary, it is very misleading to keep posting images about open waveguides when referring to a closed metallic cavity like the EM Drive and one is discussing the fields at the end plates.
As we learnt in school, mode shapes like TE01 are mode shapes for open waveguides.
It is misleading to characterize a closed cavity like the EM Drive as an open waveguide.
The EM Drive has metal plates at each end, hence it is not an open waveguide, it is a closed resonant cavity.
As such there is no such thing as a mode TE01 in a closed resonant cavity like the EM Drive.
As we learnt in school a closed cylindrical or conical resonant cavity has modes TE011 or TE012 or TE013, etc , for which there can not be an electric current ring produced by the electric field at the metallic end plates !!!
Again, the tangential electric field, parallel to a metal, must be zero at the metal end plates.
Sorry TE011. Are the heat rings not induced by current flow and the resistance in the metal?
The heat rings at the metal end plates of resonant cavities with TE01p mode shapes are not the result of electric currents (charge) produced by the electric field . They are instead the result of induction heating, resulting from the magnetic field inducing eddy-currents on the metal.
How are eddy-currents in the metal not current flow?
Why is it important to make this distinction between the electric fields at the end plates of the EM Drive vs the magnetic fields at the ends of the EM Drive?
Because there is a huge state of confusion as to what mode shapes are being excited in Meep and FEKO models.
FEKO and Meep images show the electric (E or D) and the magnetic (B or H) fields in the EM Drive resonant cavity
Confusing what FEKO and Meep display as electric fields (E) for eddy-currents is completely misleading and leads to huge confusion. -
#1911 by dustinthewind on 18 Jan, 2016 22:34
-
Why is it important to make this distinction between the electric fields at the end plates of the EM Drive vs the magnetic fields at the ends of the EM Drive?
Because there is a huge state of confusion as to what mode shapes are being excited in Meep and FEKO models.
FEKO and Meep images show the electric fields in the EM Drive resonant cavity
Ok sorry. I think we both agree that for a TE011 the E field at the end plate is zero. -
#1912 by Rodal on 18 Jan, 2016 22:36
-
Why is it important to make this distinction between the electric fields at the end plates of the EM Drive vs the magnetic fields at the ends of the EM Drive?
Because there is a huge state of confusion as to what mode shapes are being excited in Meep and FEKO models.
FEKO and Meep images show the electric fields in the EM Drive resonant cavity
Ok sorry. I think we both agree that for a TE011 the E field at the end plate is zero.
Yes we absolutely do agree.
But unfortunately others keep posting images of the electric field (E) for an open waveguide mode shape TE01 and producing a state of confusion as to what the electric field (E) should look like for mode shape TE012 in a resonant cavity with metal end plates.
Another unfortunate thing is that Meep displays the electric E field and magnetic H field components in Cartesian coordinates instead of cylindrical polar or in spherical coordinates, and that there are 3 components for each, so that there are 6 components to inspect in Meep. If you add to that that Meep images are shown without numerical values so that people cannot tell what is practically zero from what is significant you end up with a state of confusion as to whether Meep is showing a TE or a TM mode
And that's why you have the last series of pages dealing with this confusion.
Things are much better when COMSOL or FEKO images are shown (the fields numerical vector resultant are shown instead of Meep showing separate Cartesian components). Humans have poor abilities to vector add Cartesian components in their heads and translating that into intrinsic coordinates.
But even when using COMSOL or FEKO, confusion ensues if people post the electric field TE01 for open waveguides because the electric field E should be zero for TE012 at the metal end plate but it is not zero for an open waveguide TE01 field -
#1913 by rfmwguy on 18 Jan, 2016 22:36
-
Thought provoking quote O' the day -
"The more important fundamental laws and facts of physical science have all been discovered, and these are so firmly established that the possibility of their ever being supplanted in consequence of new discoveries is exceedingly remote." - Light Waves and Their Uses. By Albert A. Michelson. Published by The University of Chicago Press, 1903, pp 23-25.
https://en.wikiquote.org/wiki/Albert_A._Michelson -
#1914 by ThinkerX on 18 Jan, 2016 22:38
-
Quote
OK, the only way I could see this doing anything meaningful is to have multiple frequencies with amplitude and phase change states almost instantaneously...kinda like a magnetr....OK, enough daydreaming for now
Which is what we're playing with, right? So maybe this device 'works' largely because magnetrons are so noisy? (But in that case, doesn't 'tuning' become almost an exercise in futility?)
-
#1915 by Rodal on 18 Jan, 2016 22:49
-
Quote
OK, the only way I could see this doing anything meaningful is to have multiple frequencies with amplitude and phase change states almost instantaneously...kinda like a magnetr....OK, enough daydreaming for now
Which is what we're playing with, right? So maybe this device 'works' largely because magnetrons are so noisy? (But in that case, doesn't 'tuning' become almost an exercise in futility?)
The problem is precisely the key word <<almost instantaneously>> since a magnetron is not changing <<almost instantaneously>> for Relativity effects to play a significant role regarding inertia when you use a magnetron.
Wish that it would be as simple as to use a magnetron to have propellant-less space propulsion
There must be more than that
-
#1916 by rfmwguy on 18 Jan, 2016 23:00
-
Its hard for me to judge, but IF a delta energy level change equates to a delta mass change per Einstein, Doc is right, C2 is a huge denominator to overcome UNLESS you have additive, rapid, almost chaotic high energy level changes from multiple sources of GHz photons (a spectral spray so to speak).QuoteOK, the only way I could see this doing anything meaningful is to have multiple frequencies with amplitude and phase change states almost instantaneously...kinda like a magnetr....OK, enough daydreaming for now
Which is what we're playing with, right? So maybe this device 'works' largely because magnetrons are so noisy? (But in that case, doesn't 'tuning' become almost an exercise in futility?)
The thing I get out of this is a single, CW shot of power only changes mass once (as long as the CW signal is there). Seems like the delta E on/off forces the delta M changes. Like I've said many times, theoretical physics is not my strong suit, so I can only rely on the basic fundamentals.
This being the case, it is not violating physical laws as we know them today. -
#1917 by RFPlumber on 18 Jan, 2016 23:16
-
RFPlumber -
Just had a browse over the COMSOL Frustrum data you posted. What I'm reading in these plots is:
3.13*10^5 V/m peak Electric Field (circulating about a quarter way down from the small end)
1.62*10^-3 Tesla peak Magnetic Field (axial in about the same location), and something in the region of 0.6*10^-3 T at the wall near the peak fields, which is presumably what drives the induced current.
At least 1Kw/M^2 power density in a roughly 10cm ring round the frustrum around the location of the peak fields, and at least 350 A/m over the same 10cm ring.
The total power in the ring looks to me to be about 60W, which is a good match for the 100W input power, given there is power elsewhere. Seems to me the frustrum would be a few tens of degrees above ambient at that rate of heating.
More speculatively, one can turn 350 A/m into 35A by multiplying by the width of the current 'ring'.
If I've got the units or anything else wrong please let me know.
One last question, since one would expect a standing wave rather than static fields, it would be good to know whether your plots are a measurement at a specific point in time, or whether they are a cycle average of some kind.
Thanks again,
R.
[Modified to add missing word.]
My understanding is that all those numbers are amplitudes, not rms. During my first time simulation it took me a while to figure out the 50% difference in the power COMSOL was giving me for the cavity vs the one I thought I was entering for the feeding port... Turned out the feeding port voltage to enter was supposed to be an amplitude, and not the familiar RMS.
-
#1918 by RFPlumber on 18 Jan, 2016 23:25
-
Major RF Leak
...
This RF leak issue has now been understood but unfortunately not resolved. The leak is from all those gaps between the main body and side walls. And my initial covering these gaps with adhesive copper tape made the issue worse on average. It looks like where copper tape does not get a solid electrical contact all across the gap, it then couples capacitively to the leak and essentially becomes a transmitting antenna for it. All adhesive copper tape has since been removed.
While experimenting with it, turns out all it takes is a sub-mm gap to show leakage. You live, you learn.
How much energy is leaking vs. going to heat losses?
It is hard to tell. The worst case would appear to be somewhere around 94% leaking, based on the number for the ideal unloaded Q of this cavity being ~70,000 and the actual unloaded Q being ~4000, which indicates a 94% loss somewhere. It could all be radiation, but this is not very likely. Another estimate is based on taking into account a very high directivity of these leaks and taking a reasonable 10x gain factor for some combination of slot and horn antenna which this leaking frustum appears to resemble. Combining this with the 0.4 to 2x power factor which an RF receiver which ¼ wave monopole antenna is showing for leaking frustum vs the 1x value the receiver is showing for ¼ wave monopole instead of the frustum suggests the leaking power may be between 4% to 20% of the total.
How important is the leak for what is being tested?
My humble understanding is that while the leak is obviously bad for a number of reasons (being a potential health hazard, adding extra noise to all the measuring systems involved, degrading quality of the XBee wireless link to test platform), it should not be impacting the expected test outcome (thrust or no thrust) simply because the leak is already reflected in the number for the Q factor as just one particular source of losses in the cavity. If the leak were to be completely eliminated, the Q factor is likely to increase, potentially significantly. However, and it is very important to mention, the guiding theory for the experiment (that of Shawyer) does not impose any lower limit on the required Q factor, quite the opposite it predicts the thrust to scale linearly with the Q factor value. The obtained Q value of 2000 is supposed to result in thrust levels of at least a few hundred uN which is already more than enough to detect on the experimental setup.
Can the leak be fixed?
The only available solution appeared to be to replace one or both FR4 side walls with solid copper plates and then solder the whole assembly shut. I didn’t want to do this just yet, as it would make it very hard to proceed to the dielectric test.
Hence the frustum has been tested with the leak. I will provide and summarize test results in the next post.
-
#1919 by dustinthewind on 18 Jan, 2016 23:27
-
Dr. Rodal, RFPlumber -
Many thanks for posting the COMSOL data for the 100W frustrum. I'm only phone in a cafe now, and will do more justice to reviewing it when I get to my desk tomorrow.
FWIW, I'm counting 498 A/m versus my SWAG of 1900 (for some random frustrum!) as at least a score-draw. Also interesting to note that the 498 might be nearer 1000 if anyone managed to get half of the 800w of magnetron power into a frustrum.
Why is the number interesting? Well, two papers, Furza and Tuval's have pointed to arrangements if current loops as having potentially significant effects.
The first says that current loops can produce (barely) detectable gravitational effects.
The second points out that current loops with varying current can produce a non-zero net force on the total system. In any case, the Lorentz forces on internal current loops are relevant to the matter at hand.
Obviously, the two are unrelated, and not likely in themselves to explain any EMDrive effect.
However, between them, they are more than sufficient to motivate me to want to know the rough size of the currents we are creating in the frustrum walls.
I hope that helps, I've probably been brief enough earlier to make my interest seem entirely random.
Thanks again,
R.
I have been meaning to ask about this in regards to the image of the frustum below. I apologize as I don't remember where the link is to the name labeling of this image so I am uncertain if the field is of the magnetic field on the side walls (TE mode) or if it is the perpendicular electric field from charge separation (TM mode). As Rodal has mentioned above this is a source of confusion. However, I believe my question is indifferent to this and instead I am pointing out the traveling behavior of the wave as it moves. Does this traveling behavior, (I think this behavior would be hard to see in any FEA analysis?) indicate that we have currents which are out of phase in space and time?
The point being to consider a phased array and I will post an image below that shows how the static electric fields from charge separation induce a force in opposition to relativistic magnetic effects so that a phased array is an inefficient propulsion mechanism. The + and - symbols are from charge separation and two currents in the same direction denote magnetic attraction. Opposing currents = magnetic repulsion. All information is delayed in time and space. The idea being as seen lower in the image of the phased array that we loop the wire so that charge separation doesn't happen as it does in a phased array. This is what I believe happens inside the cavity when a TE011 mode is excited (circular currents). As a result we should "not have any perpendicular electric field on the side walls" (no charge separation) in addition to the other "absolute constraint" of "no electric field parallel to the metal side walls". As a result we have eliminated the opposing electric propulsion of the phased array leaving only the magnetic phased array propulsion.
One would think that this would some how enhance any propulsion received (as we now no longer have the opposing propulsion from the electric force) from having the currents out of phase, like a phased array. The problem is some how exciting this to actually happen inside the frustum and I am wondering if this is possible to excite inside the cavity and if "oneguide.gif" could be an indication that this may be possible that we are looking at currents separated in space and time in such a way as to behave like a purely magnetic phased array or maybe a purely electric phased array?
RERTS comment triggered this back to my mind as I have been thinking about it off and on.
Click on the phased array image to expand it and make it more visible.
) in classic exchanges he had in these threads with NSF user DeltaMass (we miss you too 
)...
...