Quote from: frobnicat on 03/29/2015 10:13 am....Few surprising facts and data points for this set of hypothesis :- That was known from beginning (and always was a difficulty for purely thermal mechanistic) : for some modes there is "no significant net thrust" without dielectric while same mode with dielectric exhibits thrust. This is known for TE012. The absence of thrust without dielectric was for "some very early evaluations", the experimental plot is not published. A TE012 mode with dielectric and thrust is reported in Brady's report.....What is new is the reporting of Poynting vector plots, to show the Poynting vector field for different modes and their significance.http://forum.nasaspaceflight.com/index.php?topic=36313.msg1352243#msg1352243http://forum.nasaspaceflight.com/index.php?topic=36313.msg1352368#msg1352368Although Dr. White published several papers predicting that the force on the EM Drive would depend on the Poynting vector, for example,Quote from: p.10 of Brady et.al.'s "Anomalous Thrust Production from an RF Test Device Measured on a Low-Thrust Torsion Pendulum"Consideration of the dynamic fields in the ¼ wave resonance tube shows that there is always a net Poynting vector meaning that the RF launcher tube assembly with dielectric cylinder common to both the slotted and smooth test articles is potentially a Q-thruster where the pillbox is simply a matching network.I have not seen any published calculations by Dr. White and his group of the Poynting vector field for the EM Drive. There are no Poynting vector plots shown in the paper "Anomalous Thrust Production from an RF Test Device Measured on a Low-Thrust Torsion Pendulum" by Brady et.al. and there have been no Poynting vector plots shown by Paul March in our threads either.The Poynting vector (its divergence and its vorticity) plays a prominent role in Dr. John Brandenburg's paper (see attachment to http://forum.nasaspaceflight.com/index.php?topic=36313.msg1349142#msg1349142 ) but again, there is no computation and plots of the Poynting vector field for the truncated cone for different electromagnetic modes.Shawyer has not published any Poynting vector calculations either. Prof. Juan Yang and her colleagues mention the Poynting vector in their calculations, but I have not seen any plots of the Poynting vector field reported in Prof. Juan Yang's papers either.Greg Egan did not publish any plots of the Poynting vector field either, in his article: http://gregegan.customer.netspace.net.au/SCIENCE/Cavity/Cavity.htmlConsideration of the boundary conditions for the Poynting vector seem to also have not been mentioned in the published literature of the EM Drive.I have not seen the following previously mentioned in any EM Drive report:QuoteSince the components of the electric field E parallel to a copper surface (either the wall or the bases) must be zero at the surface, the Poynting vector component perpendicular to a copper surface (either the wall or the bases) must be zero at the copper surface (either the wall or the bases) .Let me repeat that: the Poynting vector component perpendicular to the small and the big bases of the truncated cone must be zero at those surfaces (must be zero at the small base and must be zero at the big base).For a Transverse Magnetic (TM) mode the Poynting vector parallel to the surface doesn't have to be zero. Actually, as the images show, in some cases the maximum Poynting vector occurs at the wall for a TM mode, and for a TM mode the Poynting vector at the wall must be parallel to the wall.On the other hand, for Transverse Electric (TE) modes both components of the Poynting vector (parallel to the wall and perpendicular to the wall) must be zero at the copper surfaces (either the wall or the bases) . For TE modes the Poynting vector is zero at all copper surfaces: zero at the walls and zero at both of the truncated cone bases.The following points do not appear to have been previously made either:Quote from: Rodal Examination of the Poynting vector radial component shows that for this particular mode (TE012) without a dielectric, the Poynting's vector is self-cancelling and hence it is not a surprise that NASA measured no thrust force for this TE012 mode without a dielectric, since according to NASA Eagleworks' own theory (relying on Poynting's vector as per Dr. White's papers) there should not be a thrust force without a dielectric for mode TE012 because Poynting's vector self-cancels for this mode. ....However, for other modes (TM311 for example), Poynting's vector is not self-cancelling, but it is pointed towards the small base. This justifies the fact that Shawyer communicates that he is presently not using a dielectric, since a dielectric does not appear necessary for certain modes.Actually, Prof. Yang writes the complete opposite conclusion (http://www.emdrive.com/NWPU2010translation.pdf ):Quote from: Yang Juan,Yang Le,Zhu Yu,Ma NanUsing finite element numerical method to numerical analyse the classical Maxwell equation of electric field of the idealised conical resonator, to obtain the model and practical of the distribution of the electric field of the cavity under 1000W. By analyse the properties under different modes and the different properties. Calculation show that under the four modes, TE011, TE012, TE111 and TM011, the quality factor of TE012 is highest and with highest thrust, followed by TE011. With the Small End of the cavity unchanged, the quality factor and thrust decrease with the increase in the Large EndProf. Juan Yang writes that her Finite Element calculations show mode TE012 as having the highest thrust (without dielectric). My exact solution calculations show that the Poynting vector fields are self cancelling for TE012 without dielectric.Furthermore NASA Eagleworks experiments confirm the self-cancellation of the Poynting vector field for mode TE012: NASA, using the truncated cone without dielectric, and excited at the TE012 frequency, "measured NO significant net thrust": the complete opposite of Prof. Juan Yang's conclusion, but in full accordance with my calculations of the Poynting vector distribution for mode TE012.
....Few surprising facts and data points for this set of hypothesis :- That was known from beginning (and always was a difficulty for purely thermal mechanistic) : for some modes there is "no significant net thrust" without dielectric while same mode with dielectric exhibits thrust. This is known for TE012. The absence of thrust without dielectric was for "some very early evaluations", the experimental plot is not published. A TE012 mode with dielectric and thrust is reported in Brady's report.....
Consideration of the dynamic fields in the ¼ wave resonance tube shows that there is always a net Poynting vector meaning that the RF launcher tube assembly with dielectric cylinder common to both the slotted and smooth test articles is potentially a Q-thruster where the pillbox is simply a matching network.
Since the components of the electric field E parallel to a copper surface (either the wall or the bases) must be zero at the surface, the Poynting vector component perpendicular to a copper surface (either the wall or the bases) must be zero at the copper surface (either the wall or the bases) .Let me repeat that: the Poynting vector component perpendicular to the small and the big bases of the truncated cone must be zero at those surfaces (must be zero at the small base and must be zero at the big base).For a Transverse Magnetic (TM) mode the Poynting vector parallel to the surface doesn't have to be zero. Actually, as the images show, in some cases the maximum Poynting vector occurs at the wall for a TM mode, and for a TM mode the Poynting vector at the wall must be parallel to the wall.On the other hand, for Transverse Electric (TE) modes both components of the Poynting vector (parallel to the wall and perpendicular to the wall) must be zero at the copper surfaces (either the wall or the bases) . For TE modes the Poynting vector is zero at all copper surfaces: zero at the walls and zero at both of the truncated cone bases.
Examination of the Poynting vector radial component shows that for this particular mode (TE012) without a dielectric, the Poynting's vector is self-cancelling and hence it is not a surprise that NASA measured no thrust force for this TE012 mode without a dielectric, since according to NASA Eagleworks' own theory (relying on Poynting's vector as per Dr. White's papers) there should not be a thrust force without a dielectric for mode TE012 because Poynting's vector self-cancels for this mode. ....However, for other modes (TM311 for example), Poynting's vector is not self-cancelling, but it is pointed towards the small base. This justifies the fact that Shawyer communicates that he is presently not using a dielectric, since a dielectric does not appear necessary for certain modes.
Using finite element numerical method to numerical analyse the classical Maxwell equation of electric field of the idealised conical resonator, to obtain the model and practical of the distribution of the electric field of the cavity under 1000W. By analyse the properties under different modes and the different properties. Calculation show that under the four modes, TE011, TE012, TE111 and TM011, the quality factor of TE012 is highest and with highest thrust, followed by TE011. With the Small End of the cavity unchanged, the quality factor and thrust decrease with the increase in the Large End
....BTW what would be the heating profile for TE012, with and without dielectric ?.....
... it seems like you're saying is that we should discard the Poynting vector? ...
.....In your opinion, do you see any correlation between magnetic field strength at the large/small end as seen in the Comsol plots and the measured thrust levels? .....
....A separate question, do you think ponderomotive forces are important here?
Doctor Rodal -Could the discrepancy between your calculations and the Chinese results be due to a typo or translation error? Given that both happen, and keeping in mind the now resolved dispute between your numbers and those of Eagleworks.
I don't think that the specific pointed discrepancy (that mode TE012 without a dielectric resulted in no significant force in the NASA experiment and that it also has a self-cancelling Poynting vector field, while Prof. Juan Yang wrote in her 2010 paper that TE012 was the mode shape giving the highest thrust force) is due to a translation error. I also doubt that it is a typo because that statement is made in the body of the article as well as in the conclusions and it can also be ascertained from the plots shown in the paper.
QuoteI don't think that the specific pointed discrepancy (that mode TE012 without a dielectric resulted in no significant force in the NASA experiment and that it also has a self-cancelling Poynting vector field, while Prof. Juan Yang wrote in her 2010 paper that TE012 was the mode shape giving the highest thrust force) is due to a translation error. I also doubt that it is a typo because that statement is made in the body of the article as well as in the conclusions and it can also be ascertained from the plots shown in the paper.Wild speculation, then:The Chinese, if I recollect correctly, are pumping a lot more energy into their device than Eagleworks.I have seen repeated mention here before this results in a lot more heat, and the frequency changes with heat. So maybe this morphs TE012 into something else?
Quote from: ThinkerX on 03/31/2015 02:25 amQuoteI don't think that the specific pointed discrepancy (that mode TE012 without a dielectric resulted in no significant force in the NASA experiment and that it also has a self-cancelling Poynting vector field, while Prof. Juan Yang wrote in her 2010 paper that TE012 was the mode shape giving the highest thrust force) is due to a translation error. I also doubt that it is a typo because that statement is made in the body of the article as well as in the conclusions and it can also be ascertained from the plots shown in the paper.Wild speculation, then:The Chinese, if I recollect correctly, are pumping a lot more energy into their device than Eagleworks.I have seen repeated mention here before this results in a lot more heat, and the frequency changes with heat. So maybe this morphs TE012 into something else?That's the first thing that jumps to mind, if one has to speculate. Also she uses different equations to calculate the force. Her equations also lead to another contrarian statement she makes in her 2010 paper (and in the conclusions of that paper) that keeping the small base at constant diameter and keeping the same axial length, increasing the diameter of the big base decreases the thrust. (I wonder whether she still believes those calculations as her later experiments involve truncated cones with increased -rather than decreased- cone angle of the EM Drive).
Based on what you're saying, I'm going to make another (hobbled) sense antenna, for now it will be just a solder cup. Then I will repeat the same VSWR test to see how things change.
....So my most likely candidate for unloaded testing is 2413.5mhz. I have no clue what mode is being excited here.....
Do Hall thrusters use Xenon or can they use a more common fuel since Xenon is rare. Massive tonnage to Mars or the Moon over time will deplete a lot of Xenon. I was thinking something like hydrogen, oxygen, or some other more common element for mass use in space.
...Have you attempted to cast the dimensions in terms of the wavelength of the drive frequency, in closed form? ...
...In discussing the Poynting vector you rely on the ideal boundary conditions being zero at the cavity walls. To what extent do the shear forces resulting from copper being "not ideal" modify these assumptions? 0%, .001%, 1%, 10%? And how large might those left over shear forces be?