Author Topic: EM Drive Developments - related to space flight applications - Thread 3  (Read 3130834 times)

Offline SeeShells

  • Senior Member
  • *****
  • Posts: 2442
  • Every action there's a reaction we try to grasp.
  • United States
  • Liked: 3186
  • Likes Given: 2708
The Yang/Shell dimensions with cone half angle at 6 degrees had TM113 and TE012 both resonate at near 2.45 GHz.

This geometry (now officially entered into the World Book Of Paranormal EM Drive Phenomena) resonates with TE012 at 2.45 GHz but the TM11 modes are further apart:

TM112= 2.23227 GHz
TM113= 2.64095 GHz

So aero will have to model it in Meep with a loop antenna or we have no resonance at 2.45 GHz
Go ahead make me feel bad Jose',  I'm about ready to take a brick to the first frustum I'm so frustrumed. :)
Just as long as you don't throw the brick at me :)
Well it's just another brick on the wall. I'd have to strap a EMDrive to it to get it that far anyway. :)
Make it better, give me a number for resonance close to 2.45GHz extending the small plate out down the frustum about 50% of the size of the large plate. My confidence is a little rattled and my pen went on strike. :)

Offline deltaMass

  • Full Member
  • ****
  • Posts: 955
  • A Brit in California
  • Liked: 671
  • Likes Given: 275
When I looked at the design equations, I reckoned that the big/small diameter ratio should be as big as possible, which means maximising the half-angle. No idea if this has any effect.

Offline SeeShells

  • Senior Member
  • *****
  • Posts: 2442
  • Every action there's a reaction we try to grasp.
  • United States
  • Liked: 3186
  • Likes Given: 2708
The Yang/Shell dimensions with cone half angle at 6 degrees had TM113 and TE012 both resonate at near 2.45 GHz.

This geometry (now officially entered into the World Book Of Paranormal EM Drive Phenomena) resonates with TE012 at 2.45 GHz but the TM11 modes are further apart:

TM112= 2.23227 GHz
TM113= 2.64095 GHz

So aero will have to model it in Meep with a loop antenna or we have no resonance at 2.45 GHz
Go ahead make me feel bad Jose',  I'm about ready to take a brick to the first frustum I'm so frustrumed. :)
Just as long as you don't throw the brick at me :)

Like I said, I don't build anything until I understand what it is I'm building. I don't wing-it, I'm a tinker-er. I'll build a prototype of something I can predict and then tinker with it until it works or I understand why it doesn't. Although, I must admit it's a lot easier simulating a printed circuit board than it is building mechanical models. Don't get too frustrated Shell, it's a journey.
Todd
Microwave cavities are not my forte, I can design a heck of a test rig but the old math is taking so much longer than it did the first time. sigh.

Thanks...
Shell

Offline WarpTech

  • Full Member
  • ****
  • Posts: 1407
  • Do it!
  • Statesville, NC
  • Liked: 1453
  • Likes Given: 1925
When I looked at the design equations, I reckoned that the big/small diameter ratio should be as big as possible, which means maximising the half-angle. No idea if this has any effect.
True, but if you look at the impedance plots from Z&F, there is very little room along k*r to operate on a slope for any angle over 15 deg, (pi/12). More than that and it starts to look like a flat plate. IMO, Shell's 6 deg frustum is probably not a bad design if she can get the k*r at the small end around 18 and the k*r at the big end around 30.
Todd
 

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5911
  • USA
  • Liked: 6124
  • Likes Given: 5564
When I looked at the design equations, I reckoned that the big/small diameter ratio should be as big as possible, which means maximising the half-angle. No idea if this has any effect.
True, but if you look at the impedance plots from Z&F, there is very little room along k*r to operate on a slope for any angle over 15 deg, (pi/12). More than that and it starts to look like a flat plate. IMO, Shell's 6 deg frustum is probably not a bad design if she can get the k*r at the small end around 18 and the k*r at the big end around 30.
Todd
Shawyer and McCulloch both have (for different reasons) a constraint on length:  the length cannot be too long.  Independently, that's what the calculations show.  If you have a cone with a low angle (6 degrees) and you extend it, you end up with a large region near the small base that is just sitting there not resonating.  It becomes useless volume.  Like you said: you want to get to the point where the force at the small base is zero.

But hold it, at that point, not much further, any further is wasted.

Too low angle == bad.  If not extended, looks like a cylinder.  If extended to the apex, it is an extremely long cone with a large portion of the volume sitting there doing nothing good: no Q resonance near the apex.  Modes persist but they become evanescent way before reaching the small base.  So the ending portion just sits there doing no good.


Here is the insectoid overlord again.  The useful part are the eyes and the brain.  The resonance takes place there.  Too long a pointy chin (too much purple) is wasted volume:


« Last Edit: 08/02/2015 02:09 am by Rodal »

Offline WarpTech

  • Full Member
  • ****
  • Posts: 1407
  • Do it!
  • Statesville, NC
  • Liked: 1453
  • Likes Given: 1925
When I looked at the design equations, I reckoned that the big/small diameter ratio should be as big as possible, which means maximising the half-angle. No idea if this has any effect.
True, but if you look at the impedance plots from Z&F, there is very little room along k*r to operate on a slope for any angle over 15 deg, (pi/12). More than that and it starts to look like a flat plate. IMO, Shell's 6 deg frustum is probably not a bad design if she can get the k*r at the small end around 18 and the k*r at the big end around 30.
Todd
Shawyer and McCulloch both have (for different reasons) a constraint on length:  the length cannot be too long.  Independently, that's what the calculations show.  If you have a cone with a low angle (6 degrees) and you extend it, you end up with a large region near the small base that is just sitting there not resonating.  It becomes useless volume.  Like you said: you want to get to the point where the force at the small base is zero, not much further, any further is wasted.

Too low angle == bad.  If not extended, looks like a cylinder.  If extended to the apex, it is an extremely long cone with a large portion of the volume sitting there doing nothing good: no Q resonance near the apex.  Modes persist but they become evanescent way before reaching the small base.  So the ending portion just sits there doing no good.

Understood, so at Shell's cone angle what is the shortest section that can be made to resonate at 2.45GHz? And how small does the small end need to be, such that nothing reaches the small end?
Todd

Offline rfmwguy

  • EmDrive Builder (retired)
  • Senior Member
  • *****
  • Posts: 2205
  • Liked: 2713
  • Likes Given: 1134
NSF-1701 3rd static temp test video. I learned the IR gun is not useful at 3 feet. Also the quick drop of temp when mag cycles off is false reading. Regardless, saw some interesting arcing at full power at 1 minute duration. Frustum itself remained at low temp. Mag went to about 160°C.


Offline Rodal

  • Senior Member
  • *****
  • Posts: 5911
  • USA
  • Liked: 6124
  • Likes Given: 5564
When I looked at the design equations, I reckoned that the big/small diameter ratio should be as big as possible, which means maximising the half-angle. No idea if this has any effect.
True, but if you look at the impedance plots from Z&F, there is very little room along k*r to operate on a slope for any angle over 15 deg, (pi/12). More than that and it starts to look like a flat plate. IMO, Shell's 6 deg frustum is probably not a bad design if she can get the k*r at the small end around 18 and the k*r at the big end around 30.
Todd
Shawyer and McCulloch both have (for different reasons) a constraint on length:  the length cannot be too long.  Independently, that's what the calculations show.  If you have a cone with a low angle (6 degrees) and you extend it, you end up with a large region near the small base that is just sitting there not resonating.  It becomes useless volume.  Like you said: you want to get to the point where the force at the small base is zero, not much further, any further is wasted.

Too low angle == bad.  If not extended, looks like a cylinder.  If extended to the apex, it is an extremely long cone with a large portion of the volume sitting there doing nothing good: no Q resonance near the apex.  Modes persist but they become evanescent way before reaching the small base.  So the ending portion just sits there doing no good.

Understood, so at Shell's cone angle what is the shortest section that can be made to resonate at 2.45GHz? And how small does the small end need to be, such that nothing reaches the small end?
Todd
Too tired to calculate that now.  In the coming days...

All machines occupied with $$$ paying work
« Last Edit: 08/02/2015 02:21 am by Rodal »

Offline rfmwguy

  • EmDrive Builder (retired)
  • Senior Member
  • *****
  • Posts: 2205
  • Liked: 2713
  • Likes Given: 1134
The Yang/Shell dimensions with cone half angle at 6 degrees had TM113 and TE012 both resonate at near 2.45 GHz.

This geometry (now officially entered into the World Book Of Paranormal EM Drive Phenomena) resonates with TE012 at 2.45 GHz but the TM11 modes are further apart:

TM112= 2.23227 GHz
TM113= 2.64095 GHz

So aero will have to model it in Meep with a loop antenna or we have no resonance at 2.45 GHz
Go ahead make me feel bad Jose',  I'm about ready to take a brick to the first frustum I'm so frustrumed. :)
Just as long as you don't throw the brick at me :)

Like I said, I don't build anything until I understand what it is I'm building. I don't wing-it, I'm a tinker-er. I'll build a prototype of something I can predict and then tinker with it until it works or I understand why it doesn't. Although, I must admit it's a lot easier simulating a printed circuit board than it is building mechanical models. Don't get too frustrated Shell, it's a journey.
Todd
Microwave cavities are not my forte, I can design a heck of a test rig but the old math is taking so much longer than it did the first time. sigh.

Thanks...
Shell
Shell, I went thru what u are, ideas abound, so just pick one and run with it. There is lots to learn beyond frustum dimensional issues. Go with ur gut and git er done. frustum dimensions are a very small part of the experiment...don't let it mess up the bigger pic, thermal, electrical and test stand management. frustums can change, I'm sure mine will.

Offline aero

  • Senior Member
  • *****
  • Posts: 3628
  • 92129
  • Liked: 1145
  • Likes Given: 360
I made what I hope is a loop antenna out of 20 gauge perfect metal wire. The loop outside diameter is 14 mm, and 20 gauge wire diameter is 0.814 mm.

I proceeded by making a hollow cylinder 14 mm diameter, with 0.814 mm thick sidewalls and 0.814 mm long. I then cut a section 0.814 mm long from the cylindar side. I placed a current source across this gap 1.628 mm long.

I placed this “thing” in the center of the lattice with absorbing boundary layers on all 6 sides and excited it with an ez current source at 1.93 GHz. The lattice is 1.8 wavelengths on a side and the boundary layer is 0.5 wavelengths thick.

I made a meep run and made images of all 3 field views of all 6 EM field components.

Only I don't know what the field pattern should look like. Do the images look like they were generated by a loop antenna?

Views are here as is my control file (its also attached):
https://drive.google.com/folderview?id=0B1XizxEfB23tfk9TOE9HV29EeGJDQkVucm9RY2Fxb3RxaGI0RDFuMzh6MXhoSWN2aU9Lanc&usp=sharing
Retired, working interesting problems

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5911
  • USA
  • Liked: 6124
  • Likes Given: 5564
I made what I hope is a loop antenna out of 20 gauge perfect metal wire. The loop outside diameter is 14 mm, and 20 gauge wire diameter is 0.814 mm.

I proceeded by making a hollow cylinder 14 mm diameter, with 0.814 mm thick sidewalls and 0.814 mm long. I then cut a section 0.814 mm long from the cylindar side. I placed a current source across this gap 1.628 mm long.

I placed this “thing” in the center of the lattice with absorbing boundary layers on all 6 sides and excited it with an ez current source at 1.93 GHz. The lattice is 1.8 wavelengths on a side and the boundary layer is 0.5 wavelengths thick.

I made a meep run and made images of all 3 field views of all 6 EM field components.

Only I don't know what the field pattern should look like. Do the images look like they were generated by a loop antenna?

Views are here as is my control file (its also attached):
https://drive.google.com/folderview?id=0B1XizxEfB23tfk9TOE9HV29EeGJDQkVucm9RY2Fxb3RxaGI0RDFuMzh6MXhoSWN2aU9Lanc&usp=sharing

Sorry I cannot give you any feedback without examining all the usual csv files because to understand the mode shapes I need to have access to numerical data. The Meep image output do not have numerical data associated with the contour levels, and color contours get repeated and hence not way for me to understand what is going on.

Offline aero

  • Senior Member
  • *****
  • Posts: 3628
  • 92129
  • Liked: 1145
  • Likes Given: 360
I made what I hope is a loop antenna out of 20 gauge perfect metal wire. The loop outside diameter is 14 mm, and 20 gauge wire diameter is 0.814 mm.

I proceeded by making a hollow cylinder 14 mm diameter, with 0.814 mm thick sidewalls and 0.814 mm long. I then cut a section 0.814 mm long from the cylindar side. I placed a current source across this gap 1.628 mm long.

I placed this “thing” in the center of the lattice with absorbing boundary layers on all 6 sides and excited it with an ez current source at 1.93 GHz. The lattice is 1.8 wavelengths on a side and the boundary layer is 0.5 wavelengths thick.

I made a meep run and made images of all 3 field views of all 6 EM field components.

Only I don't know what the field pattern should look like. Do the images look like they were generated by a loop antenna?

Views are here as is my control file (its also attached):
https://drive.google.com/folderview?id=0B1XizxEfB23tfk9TOE9HV29EeGJDQkVucm9RY2Fxb3RxaGI0RDFuMzh6MXhoSWN2aU9Lanc&usp=sharing

Sorry I cannot give you any feedback without examining all the usual csv files because to understand the mode shapes I need to have access to numerical data. The Meep image output do not have numerical data associated with the contour levels, and color contours get repeated and hence not way for me to understand what is going on.

Well, I copied the EW antenna geometry so I could put it into the Brady model, but before I do that I was hoping for some little confirmation or guidance from the forum. I'll give it some time.
Retired, working interesting problems

Offline LasJayhawk

Aero, back in the late 60's I would have paid good money for some of those as posters. Groovy Man. :)

Offline SeeShells

  • Senior Member
  • *****
  • Posts: 2442
  • Every action there's a reaction we try to grasp.
  • United States
  • Liked: 3186
  • Likes Given: 2708
I made what I hope is a loop antenna out of 20 gauge perfect metal wire. The loop outside diameter is 14 mm, and 20 gauge wire diameter is 0.814 mm.

I proceeded by making a hollow cylinder 14 mm diameter, with 0.814 mm thick sidewalls and 0.814 mm long. I then cut a section 0.814 mm long from the cylindar side. I placed a current source across this gap 1.628 mm long.

I placed this “thing” in the center of the lattice with absorbing boundary layers on all 6 sides and excited it with an ez current source at 1.93 GHz. The lattice is 1.8 wavelengths on a side and the boundary layer is 0.5 wavelengths thick.

I made a meep run and made images of all 3 field views of all 6 EM field components.

Only I don't know what the field pattern should look like. Do the images look like they were generated by a loop antenna?

Views are here as is my control file (its also attached):
https://drive.google.com/folderview?id=0B1XizxEfB23tfk9TOE9HV29EeGJDQkVucm9RY2Fxb3RxaGI0RDFuMzh6MXhoSWN2aU9Lanc&usp=sharing

Sorry I cannot give you any feedback without examining all the usual csv files because to understand the mode shapes I need to have access to numerical data. The Meep image output do not have numerical data associated with the contour levels, and color contours get repeated and hence not way for me to understand what is going on.

Well, I copied the EW antenna geometry so I could put it into the Brady model, but before I do that I was hoping for some little confirmation or guidance from the forum. I'll give it some time.
http://www.ws6x.com/ant_calc.htm
Good source for dims and your radiation patterns look a little different. Should look more like a donut. Great idea of slicing a cylinder to get a loop.

Little frazzeled out myself. Nice work aero!

Offline SteveD

  • Full Member
  • ***
  • Posts: 312
  • United States
  • Liked: 83
  • Likes Given: 10
Pardon me but if the long end of an optimal frustrum is twice the size of the short end then doesn't that imply that the the inverse square law dictates the optimal design of the frustrum?     




Offline SteveD

  • Full Member
  • ***
  • Posts: 312
  • United States
  • Liked: 83
  • Likes Given: 10
Shawyer and McCulloch both have (for different reasons) a constraint on length:  the length cannot be too long.  Independently, that's what the calculations show.  If you have a cone with a low angle (6 degrees) and you extend it, you end up with a large region near the small base that is just sitting there not resonating.  It becomes useless volume.  Like you said: you want to get to the point where the force at the small base is zero.

But hold it, at that point, not much further, any further is wasted.

Too low angle == bad.  If not extended, looks like a cylinder.  If extended to the apex, it is an extremely long cone with a large portion of the volume sitting there doing nothing good: no Q resonance near the apex.  Modes persist but they become evanescent way before reaching the small base.  So the ending portion just sits there doing no good.


Here is the insectoid overlord again.  The useful part are the eyes and the brain.  The resonance takes place there.  Too long a pointy chin (too much purple) is wasted volume:



I'm not sure it's wasted volume.  Lets say all of this involves some type of weekly interactive particle that can only interact with areas in resonance.  Having one end not in resonance might increase thrust by creating an area of the frustrum where the particles cannot interact with normal matter.   (Implying that the rectangular frustrum with the superconductor might work by somehow creating conditions where our mystery particle can only interact with one end of the frustrum).  Or to put it another way, if you already have a frustrum already it might be worth testing, but expect a high probability of things not working.

Offline SteveD

  • Full Member
  • ***
  • Posts: 312
  • United States
  • Liked: 83
  • Likes Given: 10
NSF-1701 3rd static temp test video. I learned the IR gun is not useful at 3 feet. Also the quick drop of temp when mag cycles off is false reading. Regardless, saw some interesting arcing at full power at 1 minute duration. Frustum itself remained at low temp. Mag went to about 160°C.


Digital cameras can be modified to record in the IR spectrum.  http://www.lifepixel.com/  IR is normally filtered out so that you can't go peaking under peoples clothes.

Offline deltaMass

  • Full Member
  • ****
  • Posts: 955
  • A Brit in California
  • Liked: 671
  • Likes Given: 275
weekly interactive particle
you can't go peaking under peoples clothes.

Sometimes poor spelling is an advantage  8)

Offline TheTraveller

Yang's frustum has the following dimensions:

- cavity length (m): 0.24
- big diameter (m): 0.201
- small diameter (m): 0.1492

In particular, the big end is smaller than the height, not larger.

There is no resonance I can find for those dimensions at 2.45GHz. Can get TE012 resonance at 2.51GHz.

Shawyer's frustum design rules are:

1) Small end to be as small as possible to have cutoff just below it's guide wavelength. For TE01 mode and 2.45GHz the min small in diameter, in air, is 148.7m diameter. As you can see the Prof Yang small end diameter at 149.2mm is 0.5mm bigger than the minimum. That gives the frustum a bit of breathing room if the external freq needs to increase for tracking or a wide magnetron output bandwidth.

2) Big end to be a big as possible.

3) Length to be determined by desired external frequency and mode so to fit the desired 1/2 effective tapered cavity guide wavelengths in the cavity.

Effective guide wavelength and Df changes as you alter big dia, small dia and external frequency. To do this effectively and interactively you need to either use my spreadsheet or develop such a tool yourself.

It Is Time For The EmDrive To Come Out Of The Shadows

Offline TheTraveller

Pardon me but if the long end of an optimal frustrum is twice the size of the short end then doesn't that imply that the the inverse square law dictates the optimal design of the frustrum?     

Shawyer's Df dictates the big to small end ratios. Design for 1.0.

Best I can get in a doable design is Df = 0.925. Is a big ass 400mm dia big end with a 148.7mm small end and a length of 267.5mm. Spherical end plates. Gives TE013 resonance at 2.45GHz
« Last Edit: 08/02/2015 08:17 am by TheTraveller »
It Is Time For The EmDrive To Come Out Of The Shadows

Tags:
 

Advertisement NovaTech
Advertisement Northrop Grumman
Advertisement
Advertisement Margaritaville Beach Resort South Padre Island
Advertisement Brady Kenniston
Advertisement NextSpaceflight
Advertisement Nathan Barker Photography
1