-
#100 by TheTraveller on 13 Dec, 2015 15:52
-
states that the Magnetron will excite a transverse ELECTRIC mode for a CYLINDRICAL cavity TE013
The drawing states nothing about the excited mode, only that the frustum has a TE013 resonance at 2.449 GHz as per my EmDrive Design spreadsheet, which has it's resonance predictions match with those of SPR's in house system to +-0.5%
As TE013 and TM113 have the same guide wavelength and frustum resonance in this model, using the magnetron's 1/4 wave stub antenna in the centre of the big end plate will excite the TM113 resonant mode and not the TE013 mode.
Cullen 15 makes it clear that radiation pressure on a end plate in a waveguide is not directly dependent on excited mode but instead is dependent on guide wavelength, which can be effected by excited mode.
If you wish to continue this conversation please join the EmDriveResearch forum as the next comment I make on this subject on this forum will be my final test result report, which I expect to complete in Jan 2016. -
#101 by SeeShells on 13 Dec, 2015 16:11
-
Update on my latest build.
Cutting copper next week. Using a 1.2kW Panasonic maggie, driven by a modified Panasonic maggie inverter power supply. Will be using a 5kv vac switch to remove filament power once the maggie is generating microwaves. Expect 1st Force measurements early Jan 2016.
Report back then.
Am I reading that right that you are using a magnetron feeding directly from a waveguide on top of the frustum. Not sure you arent' being overly optimistic on your Q.
In any event, here's a spreadsheet that I think might help calculate the thermal lift from the air inside the frustum. While the figures on the bottom, energy to heat the air inside the frustum one degree, are still a work in progress -- if you dump 1200watts in there you've got a good chance that things get hot as hell.
BTW welcome back.
Thanks for the feedback. Will have a look at it.
No waveguide used.
Magnetron bolts directly to big end plate launcher with the antenna inside the frustum at the centre of the big end plate to excite TM113 mode. Mounts like Dave's magneton in NSF-1701 as attached.
Magnetron output power is adjustable from around 100W to 1,200W as the full wave voltage doubler inverter power supply control circuits adjust and regulate magnetron current to be constant. This also helps to reduce maggie freq splatter below that measured with simple 1/2 wave voltage doubler power supplies.
Once my spectrum scanner arrives, will be able to measure the freq splatter width generated by both types of maggie & power supplies.
On your post http://forum.nasaspaceflight.com/index.php?topic=39004.msg1456624#msg1456624 you state that the Magnetron will excite a transverse MAGNETIC CYLINDRICAL cavity mode TM113
<<Magnetron bolts directly to big end plate launcher with the antenna inside the frustum at the centre of the big end plate to excite TM113 mode>>
while the image for your post http://forum.nasaspaceflight.com/index.php?topic=39004.msg1456505#msg1456505
states that the Magnetron will excite resonance via a transverse ELECTRIC CYLINDRICAL cavity mode TE013
Completely different modes, and the fact is that the cavity in your proposed experiment is NOT a cylindrical cavity, but it is a frustum of a cone, a truncated cone, excited by a magnetron. A Magnetron can excite modes in a truncated cone that have no equivalence in a cylindrical cavity. Many of the Meep runs discussed in these threads show modes that are asymmetric and have no equivalence in cylindrical cavities. (Your spreadsheet cannot excite such asymmetric modes that do not exist in cylindrical cavities so your spreadsheet will be unable to predict whether such an asymmetric mode will be excited in the actual experiment.)
We look forward to discussion of actual experimental data to determine what actual mode shape will be excited. Looking forward to actual experimental verification of the mode shape (using a thermal camera as done by NASA).
__________________________
(*) Neither Yang nor Shawyer have ever published actual experimental data (with a thermal camera) proving what actual mode shapes were excited in their experiments. NASA verified their mode shape in one of their experiments, as per Paul March discussion in previous threads. More experimental data is needed that verifies actual mode shapes excited in EM Drive experiments, to understand what is going on in these experiments.
I agree with this a lot. The one way to show the mode generated is by a thermal camera. When we got what we thought was the copper defined for the drude model in meep aero ran a series of meep animations.
https://drive.google.com/drive/folders/0B1XizxEfB23tbVYwc1Nsa29yZlk
I need to compare this meep to what we will see in the real world, I didn't get a chance to do the camera before I fried one of my waveguide antennas and re-doing it with a couple minor changes to the antennas that should help from turning my antenna into a match like object.
Shell -
#102 by Rodal on 13 Dec, 2015 16:29
-
states that the Magnetron will excite a transverse ELECTRIC mode for a CYLINDRICAL cavity TE013
The drawing states nothing about the excited mode, only that the frustum has a TE013 resonance at 2.449 GHz as per my EmDrive Design spreadsheet, which has it's resonance predictions match with those of SPR's in house system to +-0.5%
As TE013 and TM113 have the same guide wavelength and frustum resonance in this model, using the magnetron's 1/4 wave stub antenna in the centre of the big end plate will excite the TM113 resonant mode and not the TE013 mode.
Cullen 15 makes it clear that radiation pressure on a end plate in a waveguide is not directly dependent on excited mode but instead is dependent on guide wavelength, which can be effected by excited mode.
If you wish to continue this conversation please join the EmDriveResearch forum as the next comment I make on this subject on this forum will be my final test result report, which I expect to complete in Jan 2016.
Cullen's paper concerning radiation pressure measurements on the end plate of a waveguide are not directly relevant to the EM Drive because the EM Drive is a completely closed resonant cavity (as per Shawyer's "theory"). Cullen's statement pertains to a waveguide that is closed at one end and open at the other end. The resonance of a closed cavity is different from the resonance of an open waveguide. See a number of undergraduate textbooks: Balanis, Collin, Fano, etc.
If you wish to apply Cullen's paper methodology to the EM Drive, you would need to open one end of the EM Drive (but if you do so, you will have to be even more concerned with safety, as the microwave field will be open to escape at the open end -see my previous message regarding the death of monkeys exposed to 100 watts of microwave power).
Also, the whole controversy with Shawyer's EM Drive has to do with the violation of the principle of conservation of momentum if a closed resonating cavity were capable of self-acceleration (without using any external fields) as proposed by Shawyer's "theory".
Cullen's experimental measurements on open waveguides (unlike Shawyer's EM Drive "theory") do NOT violate the principle of conservation of momentum. Cullen became a Professor at a University, Cullen's experiments were wholly consistent with Maxwell's equations and with classical physics.
_______
PS: thank you for your kind invitation to that other forum, but I think it is best to have these discussions at NASA SpaceFlight forum, the home of the free and the brave. NSF's very wide viewership is a counterbalance to the "groupthink" https://en.wikipedia.org/wiki/Groupthink that small forums are prone to have because they lack the "critical mass" of large viewership.
-
#103 by X_RaY on 13 Dec, 2015 16:54
-
states that the Magnetron will excite a transverse ELECTRIC mode for a CYLINDRICAL cavity TE013
The drawing states nothing about the excited mode, only that the frustum has a TE013 resonance at 2.449 GHz as per my EmDrive Design spreadsheet, which has it's resonance predictions match with those of SPR's in house system to +-0.5%
As TE013 and TM113 have the same guide wavelength and frustum resonance in this model, using the magnetron's 1/4 wave stub antenna in the centre of the big end plate will excite the TM113 resonant mode and not the TE013 mode.
Cullen 15 makes it clear that radiation pressure on a end plate in a waveguide is not directly dependent on excited mode but instead is dependent on guide wavelength, which can be effected by excited mode.
If you wish to continue this conversation please join the EmDriveResearch forum as the next comment I make on this subject on this forum will be my final test result report, which I expect to complete in Jan 2016.
@TT
The degenerated state of the TE01p/TM11p is only true for cylindrical cavities (based on the same value of the corresponding Bessel value). If my memory is correct, EMPro showed a mode seperation of these two shapes in a frustrum like cavity resonator. I will recheck this next days if I find the time to do this.
EDIT:
Frank Davies file show what I mean. Please look especially to TE012 and TM112.
-
#104 by X_RaY on 13 Dec, 2015 17:36
-
"... to that other forum, but I think it is best to have these discussions at NASA SpaceFlight forum, the home of the free and the brave. "
Glad you stay here Doc don't switch to the dark side of the force 
-
#105 by Rodal on 13 Dec, 2015 17:37
-
states that the Magnetron will excite a transverse ELECTRIC mode for a CYLINDRICAL cavity TE013
The drawing states nothing about the excited mode, only that the frustum has a TE013 resonance at 2.449 GHz as per my EmDrive Design spreadsheet, which has it's resonance predictions match with those of SPR's in house system to +-0.5%
As TE013 and TM113 have the same guide wavelength and frustum resonance in this model, using the magnetron's 1/4 wave stub antenna in the centre of the big end plate will excite the TM113 resonant mode and not the TE013 mode.
Cullen 15 makes it clear that radiation pressure on a end plate in a waveguide is not directly dependent on excited mode but instead is dependent on guide wavelength, which can be effected by excited mode.
If you wish to continue this conversation please join the EmDriveResearch forum as the next comment I make on this subject on this forum will be my final test result report, which I expect to complete in Jan 2016.
The degenerate state of the TE01p/TM11p is true for cylindrical cavities (based on the same value of the corresponding Bessel value). I my memory is correct, EMPro showed a mode seperation of these two shapes in a frustrum like cavity resonator. I will recheck this next days if I find the time to do this.
EDIT:
Frank Davies file show what I mean. Please look especially to TE012 and TM112.
Frank Davis shows in his frustum of a cone analysis, on page 12, for what he labels as TM112 mode shape ( 1.9355 GHz ):
RED vectors= Electric field
BLUE vectors= Electric fieldQuote"TM112" Note: like TM110 at top, and TM111 at bottom
where he is referring to what mode shapes TM110 and TM111 look like for a cylindrical cavity.
______________
TE012 mode shape (2.1794 GHz) on page 18 looks close to what mode shape TE012 looks like in a cylindrical cavity
______________
What is labeled as TM113 (2.2730 GHz) on page 20 has a little complicated Electric field (red) near the top.QuoteNote: intense fields at top
-
#106 by X_RaY on 13 Dec, 2015 17:49
-
Jep I got the same indication problems while specify the right shape for some of the modes. Also my spreadsheet is not able to predict all of them correctly in comparison with FEM since it is based on formula for cylindrical cavities. That's why I switched to FEM analysis.states that the Magnetron will excite a transverse ELECTRIC mode for a CYLINDRICAL cavity TE013
The drawing states nothing about the excited mode, only that the frustum has a TE013 resonance at 2.449 GHz as per my EmDrive Design spreadsheet, which has it's resonance predictions match with those of SPR's in house system to +-0.5%
As TE013 and TM113 have the same guide wavelength and frustum resonance in this model, using the magnetron's 1/4 wave stub antenna in the centre of the big end plate will excite the TM113 resonant mode and not the TE013 mode.
Cullen 15 makes it clear that radiation pressure on a end plate in a waveguide is not directly dependent on excited mode but instead is dependent on guide wavelength, which can be effected by excited mode.
If you wish to continue this conversation please join the EmDriveResearch forum as the next comment I make on this subject on this forum will be my final test result report, which I expect to complete in Jan 2016.
The degenerate state of the TE01p/TM11p is true for cylindrical cavities (based on the same value of the corresponding Bessel value). I my memory is correct, EMPro showed a mode seperation of these two shapes in a frustrum like cavity resonator. I will recheck this next days if I find the time to do this.
EDIT:
Frank Davies file show what I mean. Please look especially to TE012 and TM112.
Frank Davis shows in his frustum of a cone analysis, on page 12, for what he labels as TM112 mode shape ( 1.9355 GHz ):Quote"TM112" Note: like TM110 at top, and TM111 at bottom
where he is referring to what mode shapes TM110 and TM111 look like for a cylindrical cavity.
______________
TE012 mode shape (2.1794 GHz) on page 18 looks close to what mode shape TE012 looks like in a cylindrical cavity
______________
What is labeled as TM113 (2.2730 GHz) on page 20 looks a little complicated.QuoteNote: intense fields at top
-
#107 by SeeShells on 13 Dec, 2015 17:52
-
"... to that other forum, but I think it is best to have these discussions at NASA SpaceFlight forum, the home of the free and the brave. "
Glad you stay here Doc don't switch to the dark side of the force
He stays here because we have cookies.
Shell -
#108 by Rodal on 13 Dec, 2015 17:55
-
"... to that other forum, but I think it is best to have these discussions at NASA SpaceFlight forum, the home of the free and the brave. "
Glad you stay here Doc don't switch to the dark side of the force
He stays here because we have cookies.
Shell
and Shell
-
#109 by OnlyMe on 13 Dec, 2015 17:57
-
states that the Magnetron will excite a transverse ELECTRIC mode for a CYLINDRICAL cavity TE013
The drawing states nothing about the excited mode, only that the frustum has a TE013 resonance at 2.449 GHz as per my EmDrive Design spreadsheet, which has it's resonance predictions match with those of SPR's in house system to +-0.5%
As TE013 and TM113 have the same guide wavelength and frustum resonance in this model, using the magnetron's 1/4 wave stub antenna in the centre of the big end plate will excite the TM113 resonant mode and not the TE013 mode.
Cullen 15 makes it clear that radiation pressure on a end plate in a waveguide is not directly dependent on excited mode but instead is dependent on guide wavelength, which can be effected by excited mode.
If you wish to continue this conversation please join the EmDriveResearch forum as the next comment I make on this subject on this forum will be my final test result report, which I expect to complete in Jan 2016.
The degenerate state of the TE01p/TM11p is true for cylindrical cavities (based on the same value of the corresponding Bessel value). I my memory is correct, EMPro showed a mode seperation of these two shapes in a frustrum like cavity resonator. I will recheck this next days if I find the time to do this.
EDIT:
Frank Davies file show what I mean. Please look especially to TE012 and TM112.
Don't the actual dimensions of the frustum affect how the different modes are manifest in a particular frustum?
If so, how does this information have any direct connection to a frustum of different dimensions?
I guess what I am asking is, if you cannot use conclusions drawn from a cylindrical cavity to determine modes in a tapered cavity, how can you expect results from one tapered cavity to apply to another tapered cavity of very different dimensions?
It seems to me that before any conclusions drawn from Davies file could be used as general reference, it would need to include a series of different cavity dimensions and just how those dimensional changes affect mode distribution... -
#110 by X_RaY on 13 Dec, 2015 18:01
-
We have done a lot of FEM simulations with different cone dimensions...states that the Magnetron will excite a transverse ELECTRIC mode for a CYLINDRICAL cavity TE013
The drawing states nothing about the excited mode, only that the frustum has a TE013 resonance at 2.449 GHz as per my EmDrive Design spreadsheet, which has it's resonance predictions match with those of SPR's in house system to +-0.5%
As TE013 and TM113 have the same guide wavelength and frustum resonance in this model, using the magnetron's 1/4 wave stub antenna in the centre of the big end plate will excite the TM113 resonant mode and not the TE013 mode.
Cullen 15 makes it clear that radiation pressure on a end plate in a waveguide is not directly dependent on excited mode but instead is dependent on guide wavelength, which can be effected by excited mode.
If you wish to continue this conversation please join the EmDriveResearch forum as the next comment I make on this subject on this forum will be my final test result report, which I expect to complete in Jan 2016.
The degenerate state of the TE01p/TM11p is true for cylindrical cavities (based on the same value of the corresponding Bessel value). I my memory is correct, EMPro showed a mode seperation of these two shapes in a frustrum like cavity resonator. I will recheck this next days if I find the time to do this.
EDIT:
Frank Davies file show what I mean. Please look especially to TE012 and TM112.
Don't the actual dimensions of the frustum affect how the different modes are manifest in a particular frustum?
If so, how does this information have any direct connection to a frustum of different dimensions?
I guess what I am asking is, if you cannot use conclusions drawn from a cylindrical cavity to determine modes in a tapered cavity, how can you expect results from one tapered cavity to apply to another tapered cavity of very different dimensions?
It seems to me that before any conclusions drawn from Davies file could be used as general reference, it would need to include a series of different cavity dimensions and just how those dimensional changes affect mode distribution...
The NASA also.
And at least I have build and studied a large number of different conical cavities inclusive measurements via VNA (K-Band ~24 GHz).
Follow the curves of TE011 and TM111
-
#111 by Rodal on 13 Dec, 2015 18:05
-
Not following the general norm for experimental reports (where geometrical dimensions of tested objects are usually provided by the authors), Shawyer failed to report all the geometrical dimensions of the Demonstrator EM Drive (and of other EM Drives he has reported).states that the Magnetron will excite a transverse ELECTRIC mode for a CYLINDRICAL cavity TE013
The drawing states nothing about the excited mode, only that the frustum has a TE013 resonance at 2.449 GHz as per my EmDrive Design spreadsheet, which has it's resonance predictions match with those of SPR's in house system to +-0.5%
As TE013 and TM113 have the same guide wavelength and frustum resonance in this model, using the magnetron's 1/4 wave stub antenna in the centre of the big end plate will excite the TM113 resonant mode and not the TE013 mode.
Cullen 15 makes it clear that radiation pressure on a end plate in a waveguide is not directly dependent on excited mode but instead is dependent on guide wavelength, which can be effected by excited mode.
If you wish to continue this conversation please join the EmDriveResearch forum as the next comment I make on this subject on this forum will be my final test result report, which I expect to complete in Jan 2016.
The degenerate state of the TE01p/TM11p is true for cylindrical cavities (based on the same value of the corresponding Bessel value). I my memory is correct, EMPro showed a mode seperation of these two shapes in a frustrum like cavity resonator. I will recheck this next days if I find the time to do this.
EDIT:
Frank Davies file show what I mean. Please look especially to TE012 and TM112.
Don't the actual dimensions of the frustum affect how the different modes are manifest in a particular frustum?
If so, how does this information have any direct connection to a frustum of different dimensions?
I guess what I am asking is, if you cannot use conclusions drawn from a cylindrical cavity to determine modes in a tapered cavity, how can you expect results from one tapered cavity to apply to another tapered cavity of very different dimensions?
It seems to me that before any conclusions drawn from Davies file could be used as general reference, it would need to include a series of different cavity dimensions and just how those dimensional changes affect mode distribution...
As shown in the post by X-Ray (immediately above this post), Frank Davis did such an analysis to try to ascertain the geometry (given two dimensions, exploring the whole range of dimensions for the missing dimension) and mode shape present in Shawyer's purported Demonstration EM Drive experiments (since Mr. Shawyer did not either provide experimental confirmation of what mode shape -if any- was excited in his reported experiments, nor did he provide all of the geometrical dimensions).
Frank Davis (NASA) analysis in this regard is outstanding (way superior to anything ever reported by Shawyer or Yang). However, the graphical analysis (discussed in previous threads) was limited to frequency, as a discussion of different mode shapes electromagnetic vector fields (including images for all of them) for different geometries would entail a huge amount of data to publish. -
#112 by rq3 on 13 Dec, 2015 18:20
-
Update on my latest build.
Cutting copper next week. Using a 1.2kW Panasonic maggie, driven by a modified Panasonic maggie inverter power supply. Will be using a 5kv vac switch to remove filament power once the maggie is generating microwaves. Expect 1st Force measurements early Jan 2016.
Report back then.
The filament runs at only a few volts, (usually 3 to5) which "floats" on the high voltage, but many (usually 10-20) amps. Also, unless your "inverter" is actually tightly regulated, the peak voltage may be pushing 10,000 volts prior to the magnetron starting.
Ya'll be safe out there. You're allowed exactly one mistake with a magnetron power supply. There are no second chances, only a funeral. -
#113 by ThereIWas3 on 13 Dec, 2015 18:29
-
My previous report on the meep analysis of SeeShell's project turns out to have a tiny error in the resonant frequency. Through an accident of math, the number printed out by the 'harminv' function of meep looks an awful lot like a frequency in GHz, but it is not. It is the frequency in "meep frequency units", which are scaled by a/c where 'a' is an arbitarary scale factor to make units convenient (consistent throughout all runs) and 'c' is the speed of light in m/s. Since in this case a=0.3 and c is just a hair under 300,000,000, the scaled number comes out looking resonable, until you notice that it in the wrong range by a factor of almost exactly 1E9.
So the corrected figure is 2.494 GHz, not 2.4959 GHz.
I am adding code to the meep program to automatically correct for this and print out the resonance analysis in real-world units. Edit: though a small error now, if we should later choose to change the 'a' scale factor for some reason, results could be considerably off without this correction. -
#114 by SteveD on 13 Dec, 2015 19:03
-
I might be able to spend some time over the holidays putting together the library section of the rfdriven website, if anyone is interested.
-
#115 by SeeShells on 13 Dec, 2015 19:51
-
I might be able to spend some time over the holidays putting together the library section of the rfdriven website, if anyone is interested.
You're welcome to link or use my images of the build. I'll have more to post.
http://s1039.photobucket.com/user/shells2bells2002/library/CE%20Electromagnetic%20Reaction%20Thruster?sort=3&page=1
Shell -
#116 by ThinkerX on 13 Dec, 2015 21:43
-
Quote
I might be able to spend some time over the holidays putting together the library section of the rfdriven website, if anyone is interested.
Perhaps a link to the 'rfdriven' site on page 1, post 1 of this thread is in order? Or will be soon?
-
#117 by aero on 13 Dec, 2015 23:26
-
My previous report on the meep analysis of SeeShell's project turns out to have a tiny error in the resonant frequency. Through an accident of math, the number printed out by the 'harminv' function of meep looks an awful lot like a frequency in GHz, but it is not. It is the frequency in "meep frequency units", which are scaled by a/c where 'a' is an arbitarary scale factor to make units convenient (consistent throughout all runs) and 'c' is the speed of light in m/s. Since in this case a=0.3 and c is just a hair under 300,000,000, the scaled number comes out looking resonable, until you notice that it in the wrong range by a factor of almost exactly 1E9.
So the corrected figure is 2.494 GHz, not 2.4959 GHz.
I am adding code to the meep program to automatically correct for this and print out the resonance analysis in real-world units. Edit: though a small error now, if we should later choose to change the 'a' scale factor for some reason, results could be considerably off without this correction.
The Harminv output is produced, separated by commas, so you can simply copy and paste it from the terminal into your spreadsheet as csv, then add a column to convert by (c/a). Then your spreadsheet maintains a readable record of all of your resonance runs which is quite handy when you need to refer back to a run you made last week. Just make sure that you annotate each run data set so that you can properly identify it after you've forgotten all about it next week. -
#118 by aero on 14 Dec, 2015 00:38
-
Question for you antenna experts out there: "How wide is the gap between the close ends of the 1/4 wavelength dipoles used to construct a 1/2 wavelength antenna in the T configuration?"
See .gif attached.
-
#119 by rfmwguy on 14 Dec, 2015 01:10
-
Good idea...doneQuote
I might be able to spend some time over the holidays putting together the library section of the rfdriven website, if anyone is interested.
Perhaps a link to the 'rfdriven' site on page 1, post 1 of this thread is in order? Or will be soon?
