Author Topic: EM Drive Developments - related to space flight applications - Thread 3  (Read 1801983 times)

Offline mwvp

  • Full Member
  • **
  • Posts: 254
  • Coincidence? I think Not!
  • Liked: 169
  • Likes Given: 30
Looking for Harminv Q calculation formula - here
https://github.com/stevengj/harminv

I found this,
Quote
You can get the "quality factor" Q (pi |freq| / decay) by:

does that make any sense to anyone?

I'd rather not guess what it means, but there is a hint that what is now labelled as the imaginary frequency was previously labelled "decay".

I'll guess. Imaginary frequency is modulated frequency. If the frequency is only real, it is continuous (CW). And apparently, negative imaginary is decaying.

I re-read your Harminv links; been rebuilding my desktop linux system, accelerated with an Nivida GPU; got CUDA installed last night after a minor driver conflict. What point, or over what area/volume integral the data for Harminv is coming from?

Would I be correct in assuming the dipole feed terminals, after the power is cut?

Is the dipole source a resonator too? Is it a superconducting (loss-less) resonator? Maybe that's why the Q is so high?

I would think the method is sound for a 2 pole/order network. But trouble and even chaos might be found where tuned-length waveguide couplers like Yang & EW's 100kW design. Energy can slosh around, pass-band ripple, giving odd results depending on when the data is analyzed.

I worked on a couple systems, a holographic acoustic microscope, and cellular antenna feeds, that were both complex signal, narrow band, and had to be analyzed with a narrow-band (FM) chirp, which was transformed to the time-domain with an inverse Fourier transform by the VNA. Accurate impulse response (wide band) data was impossible.

Hopefully, I can get my %#& ^%@! *$^#) box of pain and grief up soon and can help.

Online Rodal

  • Senior Member
  • *****
  • Posts: 5838
  • USA
  • Liked: 5919
  • Likes Given: 5260
...

Is the dipole source a resonator too? Is it a superconducting (loss-less) resonator? Maybe that's why the Q is so high?

...
It is either because


1) aero is modeling the Copper as a Drude material model.   But there is no readily available data for a Drude material model for copper at 2.45 GHz.  The Drude material model at optical frequencies is inapplicable.
 An idea would be to try different Drude material constants that effectively bring up the resistivity such that a reasonable Q=45,000 is obtained

and / or

2)  it is due to the fact that Meep is attempting to obtain Q from a time decay from the time response.

At this early time in the transient with the RF feed ON there is no decay, on the contrary, there is magnification.

To get decay he has to turn the RF feed OFF
  and calculate the Q based on the decay.

This is what the comment from Steve Ng is hinting at.

Otherwise Meep needs to perform volume and surface integrations of the fields, in order to calculate Q during this transient period.
« Last Edit: 07/25/2015 02:07 AM by Rodal »

Offline mwvp

  • Full Member
  • **
  • Posts: 254
  • Coincidence? I think Not!
  • Liked: 169
  • Likes Given: 30
...
1) aero is modeling the Copper as a Drude material.  The Drude material constants he is using are equivalent to a material of unrealistically small resistivity.  An idea would be to try different Drude material constants that effectively bring up the resistivity such that a reasonable Q=45,000 is obtained

That sounds like cheating, a bad idea if it masks a flawed premise, like using an over-coupled dipole where a loosely coupled 1/4 wave probe or loop should be used. Or even an eigenmode excitation from Mpd. Just saying. I know, I need to stop criticizing and do it. I want to. My computer isn't amenable.

I would think if the Drude model came from Meepers, the units, being normalized to u0 and e0, would be natural and not the problem. But I speak from ignorance I'm eager to relieve.

2)  it is due to the fact that Meep is attempting to obtain Q from a time decay from the time response.

That should be entirely valid. However, think of the whispering gallery, with start-up transients. It matters where, as well as when the measurement is made.

Otherwise Meep needs to perform volume and surface integrations of the fields, in order to calculate Q during this transient period.

Perhaps Mpd (Meeps eigenmode harmonic solver) could do that real fast?

Online Rodal

  • Senior Member
  • *****
  • Posts: 5838
  • USA
  • Liked: 5919
  • Likes Given: 5260
...
1) aero is modeling the Copper as a Drude material.  The Drude material constants he is using are equivalent to a material of unrealistically small resistivity.  An idea would be to try different Drude material constants that effectively bring up the resistivity such that a reasonable Q=45,000 is obtained

That sounds like cheating, a bad idea if it masks a flawed premise, like using an over-coupled dipole where a loosely coupled 1/4 wave probe or loop should be used. Or even an eigenmode excitation from Mpd. Just saying. I know, I need to stop criticizing and do it. I want to. My computer isn't amenable.

I would think if the Drude model came from Meepers, the units, being normalized to u0 and e0, would be natural and not the problem. But I speak from ignorance I'm eager to relieve.

2)  it is due to the fact that Meep is attempting to obtain Q from a time decay from the time response.

That should be entirely valid. However, think of the whispering gallery, with start-up transients. It matters where, as well as when the measurement is made.

Otherwise Meep needs to perform volume and surface integrations of the fields, in order to calculate Q during this transient period.

Perhaps Mpd (Meeps eigenmode harmonic solver) could do that real fast?


<<I would think if the Drude model came from Meepers, the units, being normalized to u0 and e0, would be natural and not the problem. But I speak from ignorance I'm eager to relieve.>>

No, take a look at the Drude model, its properties are very dependent on frequency.  The Drude model constants available are for the optical range and wholly inapplicable to the 2.45 GHz range. The Drude model constants are not u0 and e0



« Last Edit: 07/25/2015 01:41 AM by Rodal »

Offline aero

  • Senior Member
  • *****
  • Posts: 2743
  • 92129
  • Liked: 704
  • Likes Given: 237
As Todd Desiato "WarpTech" said, why don't you just scale your Drude model with a different Drude constant such that the Q output is a reasonable number like the ones I provided for Yang/Shell and rfrmwguy/NSF-1701 based on my solution?  :)

Seems like the most expedient thing to do...

Love to do it that way. Who knows the Drude model to use? The skin thickness to use? A way to parameterize the Drude model using multiple runs to find the value that gives a valid Q?  There is only one term in my model as long as we leave er ~=1 as a metal. But I'd guess, I don't know, but I'd guess that we should leave the frequency term alone (adjusting for er, the divisor) and increase er. I'll make a run or two that way and let you know.
Retired, working interesting problems

Online Rodal

  • Senior Member
  • *****
  • Posts: 5838
  • USA
  • Liked: 5919
  • Likes Given: 5260
As Todd Desiato "WarpTech" said, why don't you just scale your Drude model with a different Drude constant such that the Q output is a reasonable number like the ones I provided for Yang/Shell and rfrmwguy/NSF-1701 based on my solution?  :)

Seems like the most expedient thing to do...

Love to do it that way. Who knows the Drude model to use? The skin thickness to use? A way to parameterize the Drude model using multiple runs to find the value that gives a valid Q?  There is only one term in my model as long as we leave er ~=1 as a metal. But I'd guess, I don't know, but I'd guess that we should leave the frequency term alone (adjusting for er, the divisor) and increase er. I'll make a run or two that way and let you know.

Do NOT change or play with the skin thickness.  The skin thickness is well defined and understood.  Square root of resistiviity ...

calculate the skin thickness:

http://www.microwaves101.com/calculators/869-skin-depth-calculator

http://www.rfcafe.com/references/calculators/skin-depth-calculator.htm

And yes, you have to, you must change the frequency term too.

The frequency term is called the relaxation time, and it is the relaxation time that is related to resistivity !

http://optics.hanyang.ac.kr/~shsong/27-Metals.pdf

But if Meep is trying to calculate Q from a time decay,  all  of this is FUTILE.

You have to turn the RF feed OFF to get a time decay
« Last Edit: 07/25/2015 01:58 AM by Rodal »

Offline WarpTech

  • Full Member
  • ****
  • Posts: 1228
  • Do it!
  • Vista, CA
  • Liked: 1295
  • Likes Given: 1749
As Todd Desiato "WarpTech" said, why don't you just scale your Drude model with a different Drude constant such that the Q output is a reasonable number like the ones I provided for Yang/Shell and rfrmwguy/NSF-1701 based on my solution?  :)

Seems like the most expedient thing to do...

Love to do it that way. Who knows the Drude model to use? The skin thickness to use? A way to parameterize the Drude model using multiple runs to find the value that gives a valid Q?  There is only one term in my model as long as we leave er ~=1 as a metal. But I'd guess, I don't know, but I'd guess that we should leave the frequency term alone (adjusting for er, the divisor) and increase er. I'll make a run or two that way and let you know.

Do NOT change or play with the skin thickness.  The skin thickness is well defined and understood.  Square root of resistiviity ...

calculate the skin thickness:

http://www.microwaves101.com/calculators/869-skin-depth-calculator

http://www.rfcafe.com/references/calculators/skin-depth-calculator.htm

And yes, you have to, you must change the frequency term too.

The frequency term is called the relaxation time, and it is the relaxation time that is related to resistivity !

http://optics.hanyang.ac.kr/~shsong/27-Metals.pdf

But if Meep is trying to calculate Q from a time decay,  all  of this is FUTILE.

You have to turn the RF feed OFF to get a time decay

What I was trying to say is, we can estimate the resistance per square meter (square inch) on the surface of the copper, at the frequency. Once we have the Rho = Ohms/m^2, it's just a matter of applying it to the E field on the surface using Ohm's law, for each cycle. In the simulation, this loss would be subtracted from the gain on each cycle.

Power loss = E^2/Rho,

You just need a value for Rho at the given frequency. I saw one in a paper a while back. I'll try to find it if @Rodal doesn't have it.

Todd

Online Rodal

  • Senior Member
  • *****
  • Posts: 5838
  • USA
  • Liked: 5919
  • Likes Given: 5260
This table gives resistivity in Ohm-m for many copper alloys:

http://eddy-current.com/conductivity-of-metals-sorted-by-resistivity/

Offline aero

  • Senior Member
  • *****
  • Posts: 2743
  • 92129
  • Liked: 704
  • Likes Given: 237
As Todd Desiato "WarpTech" said, why don't you just scale your Drude model with a different Drude constant such that the Q output is a reasonable number like the ones I provided for Yang/Shell and rfrmwguy/NSF-1701 based on my solution?  :)

Seems like the most expedient thing to do...

Love to do it that way. Who knows the Drude model to use? The skin thickness to use? A way to parameterize the Drude model using multiple runs to find the value that gives a valid Q?  There is only one term in my model as long as we leave er ~=1 as a metal. But I'd guess, I don't know, but I'd guess that we should leave the frequency term alone (adjusting for er, the divisor) and increase er. I'll make a run or two that way and let you know.

Do NOT change or play with the skin thickness.  The skin thickness is well defined and understood.  Square root of resistiviity ...

calculate the skin thickness:

http://www.microwaves101.com/calculators/869-skin-depth-calculator

http://www.rfcafe.com/references/calculators/skin-depth-calculator.htm

And yes, you have to, you must change the frequency term too.

The frequency term is called the relaxation time, and it is the relaxation time that is related to resistivity !

http://optics.hanyang.ac.kr/~shsong/27-Metals.pdf

But if Meep is trying to calculate Q from a time decay,  all  of this is FUTILE.

You have to turn the RF feed OFF to get a time decay

No, I miss spoke, not skin thickness, rather wall thickness. Skin thickness as used in EM physics is not an available parameter. But a different material, and that is what we are discussing, will have it's own skin thickness.

As for where Harminv is sampling the field decay, here (vector3 0.05 0.05 0.05). It samples after the gaussian source shuts off but I have no control over that using the published information. There must be a way to turn it off because when running Harminv, source is supposed to turn off, propagate waiting for  the fields to decay, then Harminv samples the remaining fields for frequency and amplitude assuming sin waves.
« Last Edit: 07/25/2015 02:32 AM by aero »
Retired, working interesting problems

Offline aero

  • Senior Member
  • *****
  • Posts: 2743
  • 92129
  • Liked: 704
  • Likes Given: 237
Quote
What I was trying to say is, we can estimate the resistance per square meter (square inch) on the surface of the copper, at the frequency. Once we have the Rho = Ohms/m^2, it's just a matter of applying it to the E field on the surface using Ohm's law, for each cycle. In the simulation, this loss would be subtracted from the gain on each cycle.

I think that is what the Drude model is supposed to do, or at least one of the things it is supposed to do. Because the material model is the only way meep allows the user to access the fields. Short of "write your own function", that is.
Retired, working interesting problems

Offline WarpTech

  • Full Member
  • ****
  • Posts: 1228
  • Do it!
  • Vista, CA
  • Liked: 1295
  • Likes Given: 1749
Quote
What I was trying to say is, we can estimate the resistance per square meter (square inch) on the surface of the copper, at the frequency. Once we have the Rho = Ohms/m^2, it's just a matter of applying it to the E field on the surface using Ohm's law, for each cycle. In the simulation, this loss would be subtracted from the gain on each cycle.

I think that is what the Drude model is supposed to do, or at least one of the things it is supposed to do. Because the material model is the only way meep allows the user to access the fields. Short of "write your own function", that is.

I see. So if this is what you have to work with,



and the conductivity is already entered correctly, and the frequency is determined elsewhere. Then perhaps there is a way to set a maximum value for tau? The problem is as was said, it's not calculating the decay time because at this time it is in a transient where the signal is growing. So tau is probably infinite. If you set a realistic maximum value, then there should be losses starting from the beginning. Hopefully, you can scale tau to match @Rodal's number for Q?

FYI: I used to program simulations of DC to AC and AC to DC converters using Turbo Basic on an XT 286, back in the 80's. I had to program the output display to my orange screen, pixel by pixel. When I finally upgraded to a 486 and 64 colors, I had to add time delay and stop commands to the output display, because it went by so fast I couldn't see it. LOL!
Todd

Offline TheTraveller

Eagleworks has a Facebook page

https://www.facebook.com/eagleworksnasa

Maybe stop by and make a comment or so.

You can also search the NASA data base for all the papers with Dr. White and/or Paul March as authors. Lots of interesting stuff there.

http://ntrs.nasa.gov/search.jsp?R=20110023492

Enjoy
"As for me, I am tormented with an everlasting itch for things remote. I love to sail forbidden seas.
Herman Melville, Moby Dick

Online lmbfan

  • Member
  • Posts: 55
  • Phoenix
  • Liked: 42
  • Likes Given: 42
For those looking for output functions for meep, I finally found them in a file called "meep.scm".  In my build it is in "/usr/local/share/meep".

I know very little about Guile/Scheme (the .scm files) and so I'm not entirely sure I modified this file the best way.  I have attached a text file that can be added to the end of "meep.scm" anywhere after the "ouptut-png+h5" function (line 990).  Usage is similar to the following:

(output-txt Ez "-0 -z 0 -o output_directory/ez.csv" )

The "Ez" term can be changed to "Ex", "Hx", etc for electric and magnetic fields.  I think it will do "s" and "b" and "d" terms as well, similar to output-png.

Please note that this implementation, similar to the output-png function, appears to write a temporary h5 file, export a .csv, then delete the h5 file.  I can't imagine this is very fast, so using this may slow down simulation time dramatically.  I would expect there is a quicker/more efficient way to do this in native C++ code, but that's beyond me right now.

The geometry functions (block, cylinder, etc) are over in "/usr/local/share/libctl/utils" and the files are "geom.c" and "geom.scm" for those looking to perhaps import some type of geometry.  These files are part of the libctl install (a required lib).

Offline ThinkerX

  • Full Member
  • ***
  • Posts: 303
  • Alaska
  • Liked: 112
  • Likes Given: 59
Last post I saw on the Eagleworks Facebook page is from July 3rd by Paul Harstad.

Don't know enough to evaluate that posts contents:

Quote
The EM Drive produces movement in the surrounding area in the same way that buoyancy works. By distorting the electro hydrodynamic pressure in the surrounding area, it will move towards the lower pressure area and away from the higher pressure area.

It's basically an inverse Doppler effect. Using the Doppler effect to propel something by inducing the difference in field strength on the surrounding area.

The main reason this hasn't gotten much attention over the years is that equating "space-time" with "localized EMF fields" is normally consider invalid, even though Einstein himself presented a paper on relativity being mathematically equivalent to localized another theory.

But all these devices (Cannae Drive/EM Drive) work by manipulating "space-time" with "EMF", logically what we know as space-time can also be modeled as an EMF Field tensor towards gravity sources that creates a centripetal force on any travelling EMF.

Also all the relativity experiments are just as consistent with a tensor caused by EMF flow towards mass. You can just as easily explain it as "The boat goes through water traveling east" as "The water to the east of the boat takes up less space-time". They are two ways of saying the same thing, but the first explanation is better for logical application.

I corrected the spelling, otherwise its a straight copy.

To me, it seems a little like the 'inertial ratchet' theory put forth here now and again.

Offline demofsky

  • Full Member
  • *
  • Posts: 165
  • Liked: 96
  • Likes Given: 1662
Last post I saw on the Eagleworks Facebook page is from July 3rd by Paul Harstad.

Don't know enough to evaluate that posts contents:

Quote
The EM Drive produces movement in the surrounding area in the same way that buoyancy works. By distorting the electro hydrodynamic pressure in the surrounding area, it will move towards the lower pressure area and away from the higher pressure area.

It's basically an inverse Doppler effect. Using the Doppler effect to propel something by inducing the difference in field strength on the surrounding area.

The main reason this hasn't gotten much attention over the years is that equating "space-time" with "localized EMF fields" is normally consider invalid, even though Einstein himself presented a paper on relativity being mathematically equivalent to localized another theory.

But all these devices (Cannae Drive/EM Drive) work by manipulating "space-time" with "EMF", logically what we know as space-time can also be modeled as an EMF Field tensor towards gravity sources that creates a centripetal force on any travelling EMF.

Also all the relativity experiments are just as consistent with a tensor caused by EMF flow towards mass. You can just as easily explain it as "The boat goes through water traveling east" as "The water to the east of the boat takes up less space-time". They are two ways of saying the same thing, but the first explanation is better for logical application.

I corrected the spelling, otherwise its a straight copy.

To me, it seems a little like the 'inertial ratchet' theory put forth here now and again.

So now we have the 'EMF Boyancy' theory.  In some ways this should be (very slightly) less controversial than the 'Inertial Ratchet' theory since they are not invoking inertia and much less controversial than the 'QV Propeller' theory.

On the other hand this espouses an open system approach, some form of which I believe will ultimately prevail.

Also, this is an interesting theoretical approach from the perspective that it aligns with some of Einstein's thinking in the 20's.

Offline mwvp

  • Full Member
  • **
  • Posts: 254
  • Coincidence? I think Not!
  • Liked: 169
  • Likes Given: 30
Last post I saw on the Eagleworks Facebook page is from July 3rd by Paul Harstad.

Don't know enough to evaluate that posts contents:

Some principles must be understood, before descriptions make sense. Mathematics, for instants.

Quote
The EM Drive produces movement in the surrounding area in the same way that buoyancy works. By distorting the electro hydrodynamic pressure in the surrounding area, it will move towards the lower pressure area and away from the higher pressure area.

The area surrounded is the area in the frustrum. The pressure is the energy/frequency of the photon gas, which expands to lower frequency by the doppler shift.

Quote
It's basically an inverse Doppler effect. Using the Doppler effect to propel something by inducing the difference in field strength on the surrounding area.

Well said. The currents in the frustrum are reacting against the absolute-rest frame of space time and the absolute energy frame of the EM field at C. Since those frames are absolute, they might as well be outside the frustrum. Its an open system.

Quote
The main reason this hasn't gotten much attention over the years is that equating "space-time" with "localized EMF fields" is normally consider invalid, even though Einstein himself presented a paper on relativity being mathematically equivalent to localized another theory.

That's obscure. It makes sense, in the context of stating there are two absolute inertial frames; the vacuum at rest, and energy at C.

Quote
But all these devices (Cannae Drive/EM Drive) work by manipulating "space-time" with "EMF", logically what we know as space-time can also be modeled as an EMF Field tensor towards gravity sources that creates a centripetal force on any travelling EMF.

I would have said the EM Drive frustrum is manipulated (translated) through space by the induced currents reacting against the trapped EM mode, when the system is destabilized (forced from equilibrium) by doppler shift. Space and energy at C are the absolute stator, the frustrum is the rotor of a motor.

Quote
Also all the relativity experiments are just as consistent with a tensor caused by EMF flow towards mass. You can just as easily explain it as "The boat goes through water traveling east" as "The water to the east of the boat takes up less space-time". They are two ways of saying the same thing, but the first explanation is better for logical application.

I think of the frustrum as an accelerated frame, distorting the confined EM mode from its natural energy state. When its accelerated, the mode relaxes, falls to a lower energy state, or to a higher energy state in the opposite direction. Light takes the path of least-action. Doppler shifted photons find the frustrum either longer or shorter, so the photon-mode either is relaxed or compressed, gaining or losing energy to the frustrum, which either increases or decreases momentum.

I corrected the spelling, otherwise its a straight copy.

To me, it seems a little like the 'inertial ratchet' theory put forth here now and again.

Yep. Perhaps more than simple a ratchet, an unstable, precariously balanced ratchet that can accelerate by "falling" forward, dissipating energy. But I still have doubts  :-\
« Last Edit: 07/25/2015 07:32 AM by mwvp »

Offline SeeShells

  • Senior Member
  • *****
  • Posts: 2326
  • Every action there's a reaction we try to grasp.
  • United States
  • Liked: 2956
  • Likes Given: 2589
MWVP you said...
Yep. Perhaps more than simple a ratchet, an unstable, precariously balanced ratchet that can accelerate by "falling" forward, dissipating energy. But I still have doubts  :-\

--------------------------------

I like this one,
 
"This paper will discuss the current viewpoint of the vacuum state and explore the idea of a "natural" vacuum as opposed to immutable, non-degradable vacuum. This concept will be explored for all primary quantum numbers to show consistency with observation at the level of Bohr theory. A comparison with the Casimir force per unit area will be made, and an explicit function for the spatial variation of the vacuum density around the atomic nucleus will be derived. This explicit function will be numerically modeled using the industry multi-physics tool, COMSOL(trademark), and the eigenfrequencies for the n = 1 to n = 7 states will be found and compared to expectation."
http://ntrs.nasa.gov/search.jsp?R=20150006842&qs=Nm%3D4294903350%7CAuthor%7CWhite%2C%2520Harold%26N%3D0

Makes so much sense in ways that even I can draw a parallel between all of them, eigenfrequencies, the casimir effect and a natural vacuum state that really seems to be mutable.

The EMDrive has shown a red flag by
Air=thrust, No air=little thrust

Doesn't this come close to what Todd has been thinking? What if through the wave actions of the cavity we alter the natural mutable vacuum by modifying the eigenfrequencies  of the matter in the cavity and even the walls of the frustum creating an area where the flustrum falls forward because it mutates the immutable Q vacuum and interacts with the "outside" QV.

Air=thrust
No air=little thrust ... the only thing still around is the QV and the copper of the cavity. The inside copper skin of the cavity ~5um is still interacting with the harmonic wave actions including evanescent and ghost modes inside of the cavity so the ~5um depth may be a little deeper.
What I'm wondering we have a thrust of (Ta) Thrust air and Thrust (Na) no air, what is the percentage of difference in mass between the two? Can a close ratio be derived between the thrust values?

My SO is telling me I need to go to bed. So I'll leave you all with this crazy Eddie thought. Be back in the morning over a cuppa joe.

Shell

Offline Chrochne

  • Full Member
  • *
  • Posts: 182
  • Liked: 104
  • Likes Given: 247
The latest article by Wired about the EmDrive hit the space community hard (already 200 shares of the article). I see all the channels I check buzz with the news. Even some hardcore sceptics are looking with some kind of interest to see what Prof. Tajmar results will be (as far as I can tell). Many "enthusiasts" as me are dejected, because of possibly very low thrust, but I want to believe that all new technology need to start somewhere and can develop. And hope that maybe we can use it for terrestrial application in some not so distant future. And of course even low thrust can be the start of the new era in space for humanity and perhaps colonization of our solar system in very distant future.

By the way just curious. This Prof. Tajmar has really this good reputation? Or rather he has a good laboratory to prove / disprove EmDrive works?

Thanks and have a nice weekend because this Monday will really be interesting.

PS: Get ready the forum can be flooded by posts for a while before it calms down after the Prof. Tajmar results.

Offline Star One

  • Senior Member
  • *****
  • Posts: 8015
  • UK
  • Liked: 1280
  • Likes Given: 168
The latest article by Wired about the EmDrive hit the space community hard (already 200 shares of the article). I see all the channels I check buzz with the news. Even some hardcore sceptics are looking with some kind of interest to see what Prof. Tajmar results will be (as far as I can tell). Many "enthusiasts" as me are dejected, because of possibly very low thrust, but I want to believe that all new technology need to start somewhere and can develop. And hope that maybe we can use it for terrestrial application in some not so distant future. And of course even low thrust can be the start of the new era in space for humanity and perhaps colonization of our solar system in very distant future.

By the way just curious. This Prof. Tajmar has really this good reputation? Or rather he has a good laboratory to prove / disprove EmDrive works?

Thanks and have a nice weekend because this Monday will really be interesting.

PS: Get ready the forum can be flooded by posts for a while before it calms down after the Prof. Tajmar results.

I've read that even with low thrust it would still be useful for getting around the Solar System?
« Last Edit: 07/25/2015 10:27 AM by Star One »

Offline TheTraveller

The EMDrive has shown a red flag by
Air=thrust, No air=little thrust

Air = acoustic vibration on big end plate greater than small end plate (enables switch from IDLE to MOTOR mode).

No air = no acoustic vibration (hard to get out of IDLE mode).
"As for me, I am tormented with an everlasting itch for things remote. I love to sail forbidden seas.
Herman Melville, Moby Dick

Tags: