Author Topic: EM Drive Developments - related to space flight applications - Thread 3  (Read 1797168 times)

Offline Paul Novy

  • Member
  • Posts: 12
  • Poland
  • Liked: 11
  • Likes Given: 3
With all the past talks about quantum fields and general relativity as reference frames for explaining EMdrive thrust, I've stumbled on this:

http://www.spacedaily.com/reports/How_spacetime_is_built_by_quantum_entanglement_999.html

not sure if it may bring something to the table...but it appears to be one of the first (theoretical) steps in developing a Theory of Everything.
Cant find the technical papers right away, but those might contain something useful for the theorists inhere? (It's way above my head anyway)


I think that this is what you're looking for:
"Tomography from Entanglement"
http://arxiv.org/pdf/1412.1879v1.pdf

Although math and concepts depicted there are way outside my abilities but someone might find it useful.

Offline frobnicat

  • Full Member
  • ****
  • Posts: 518
  • Liked: 500
  • Likes Given: 151
Err...I don't know if this is a lightbulb moment or a senior moment, but if the cavity generates an asymmetrical pattern of eddy currents in the surface of the frustrum, wouldn't that interact with the Earth's magnetic field to produce a net force?

This was asked before but not thoroughly analysed yet, from my recollection. Also it is known that the set-up at Eagleworks uses a strong permanent magnet in the damping system (back end of the pendulum arm, opposite to frustum). Not sure if there was still this strong magnet present, in particular in the very few vacuum tests, since damping was better on later tests (no longer underdamped responses) and Paul March mentioned an oil bath damping, but with no details (added to magnetic damping or replacing it ?). We don't know if the DC B field due to this magnet would be higher magnitude than earth magnetic field, it depends on how well the fields line reconnects in the U shaped magnetic circuit... back of the envelope guesstimates makes me think this could be same order or higher. With the notable difference that the B field gradient (er, divergence ?) would be much higher across the span of frustum, while the earth magnetic field would be near constant in the concerned span (unless ferromagnetic material around would bend the natural locally homogeneous field). This could be of importance :

With closed current loops (like ones induced by varying B field of an RF excitation on a sub period time) there is no way to obtain a thrust in a locally homogeneous constant B field : the Lorentz force integrates to 0 net force on a loop. Why it's hard to make electrodynamic tether work : the return current path can't be just another wire. But there still could be torque, and since the balance (at Eagleworks) is rotating, this could induce a displacement (even if torque would be applied at end of arm). In a (time) constant inhomogeneous B field, a loop of (DC) current can impart thrust,because of asymmetry of Lorentz force on a forward and return part of the loop (B field stronger here, weaker there).

But : when considering a whole period of the RF excitation, the induced current loops (at skin depth of the copper walls...) are also periodic, AC current loops. This would make the above statements no longer valid, whatever net imbalance of force or torque that could be obtained by pushing with DC current loop on a local B field would cancel with pure AC current loops of non varying geometry, as current going back and forth would average to 0 (Lorentz force) locally at each single part of current path.

So question (to experts !) :

- Can DC current components emerge from possible non linearities in the walls ? Very rough order of magnitude guesstimates (from the dissipated Ohmic losses and copper conductivity and skin depth) makes me think this would be in the ballpark of relevant reported thrusts with only a 1% DC spurious component (1% of RF induced AC current magnitude gets somehow "rectified").

- Can the AC induced "current loops" be of varying geometry (the current loops in a slightly different shape depending on its direction, from one half period to the next) ? Comes to mind Hall effect or something similar to Corbino effect (geometrical magnetoresistance), but carriers mobility in metal would likely make such effect way too small to be relevant (would need semiconductor like mobility).

- Has it been known in some microwave system that the introduction of a DC (or slow frequency AC) B field would alter the functioning inside (detuning or otherwise) ? Not speaking of systems juggling with flying charged particles (CRT ...) but only carriers in metal walls...

@Rodal : would it be possible from your exact solutions to see what the RF induced AC current loops look like at the walls ? My apologies if the request is ill formed, must admit my understanding of skin effect is still very weak at the moment.

Online Rodal

  • Senior Member
  • *****
  • Posts: 5833
  • USA
  • Liked: 5901
  • Likes Given: 5249
...
I just re-read Greg Eagan's paper more thoroughly. He and Yang end with the same result, the integration of T*n over the surface. His "proof" is based on there being ONLY standing waves, of sin(wt) & cos(wt). He does not consider the Evanescent waves that are decaying into the confined volume of the small end, where c/K << c.

These would have exponentially decaying terms, exp[-a*t - b*x], where "a" and "b" are variables dependent on the shape of the cavity, per Zeng and Fan. The part that determines thrust, based on the shape of the cavity, was neglected by Egan and Yang in their analyses. Their resulting closed form equation is correct, the D, E and B, H fields they plug into it are not. Optimize their integral force equation for the correct representation for D, E and B, H, i.e., NOT standing waves, including non-linear materials, then design a system to meet those requirements.

...
The image you show is the Poynting vector field based  on standing waves. (http://forum.nasaspaceflight.com/index.php?topic=37642.msg1381389#msg1381389 )



When you state:  <<Optimize their integral force equation for the correct representation for D, E and B, H, i.e., NOT standing waves>>

are you proposing to only consider the evanescent wave terms and to ignore the standing wave terms in the analysis? (since the standing wave terms perfectly cancel out when averaged over an integer number of time periods). If we ignore the standing waves, then the above image doesn't apply, since it is based on standing waves.

Or do you see any interaction/coupling between the standing wave terms and the evanescent wave terms so that both have to be considered in the analysis?



AERO:   When you considered evanescent waves in the Truncated Cone, modelled as a two-dimensional flat trapezium:

1) Considering only evanescent waves inside the cavity, did MEEP calculate any net thrust force?  If the answer is no, why not? (Did you consider any losses in your analyis? Did you input a tan delta value for the Dielectric polymer insert?)

2) Was attenuation taking into account in the MEEP analysis? If yes, what were the attenuation parameters that you considered?

« Last Edit: 05/29/2015 03:44 PM by Rodal »

Online Rodal

  • Senior Member
  • *****
  • Posts: 5833
  • USA
  • Liked: 5901
  • Likes Given: 5249
...The part that determines thrust, based on the shape of the cavity, was neglected by Egan and Yang in their analyses. Their resulting closed form equation is correct, the D, E and B, H fields they plug into it are not....
If you are correct that Prof. Yang neglected the part that determines the thrust in her analysis, how is she then calculating a thrust force?

Something is amiss.  Greg Egan makes sense, because he concludes that there is no thrust.

But how can we say that on one hand Prof. Yang neglects the part that determines thrust and on the other hand she calculates a thrust force?  Where does her calculated thrust force come from then ?
« Last Edit: 05/29/2015 12:56 PM by Rodal »

Offline v3ngi

  • Member
  • Posts: 1
  • USA, NY
  • Liked: 0
  • Likes Given: 0
Long time lurker here trying to understand all of this in laymens terms if possible...  :o

Is this device creating an unbalanced magnetic field inside the frustum with microwaves then copper pushes off the field inside creating momentum? Am I even close?

Offline JasonAW3

  • Senior Member
  • *****
  • Posts: 2409
  • Claremore, Ok.
  • Liked: 374
  • Likes Given: 10
I'm not certian, but this MAY be related to the current experiments being done;

http://gizmodo.com/sunlight-and-graphene-could-one-day-power-a-spaceship-1707535183

Essentially, graphine seems to have a disproportionate propulsive reaction when exposed to intense light.  If related, perhaps a version of the device should be made from graphene insteaqd of copper for additional testing?
My God!  It's full of universes!

Offline Notsosureofit

  • Full Member
  • ****
  • Posts: 656
  • Liked: 704
  • Likes Given: 1361
With all the past talks about quantum fields and general relativity as reference frames for explaining EMdrive thrust, I've stumbled on this:

http://www.spacedaily.com/reports/How_spacetime_is_built_by_quantum_entanglement_999.html

not sure if it may bring something to the table...but it appears to be one of the first (theoretical) steps in developing a Theory of Everything.
Cant find the technical papers right away, but those might contain something useful for the theorists inhere? (It's way above my head anyway)


I think that this is what you're looking for:
"Tomography from Entanglement"
http://arxiv.org/pdf/1412.1879v1.pdf

Although math and concepts depicted there are way outside my abilities but someone might find it useful.

Yes, thank you.  That is part of the thermodynamic argument I'm trying to pursue.  I've got Kantor's "Information Mechanics" but it's not an easy read.


Offline WarpTech

  • Full Member
  • ****
  • Posts: 1223
  • Do it!
  • Vista, CA
  • Liked: 1290
  • Likes Given: 1739
...
When you state:  <<Optimize their integral force equation for the correct representation for D, E and B, H, i.e., NOT standing waves>>

are you proposing to only consider the evanescent wave terms and to ignore the standing wave terms in the analysis? (since the standing wave terms perfectly cancel out when averaged over an integer number of time periods)

Or do you see any interaction/coupling between the standing wave terms and the evanescent wave terms so that both have to be considered in the analysis?
....


The interaction is, as the standing waves dissipate heat in the copper, they lose energy and shift to longer wavelengths, where attenuation takes over from dissipation and imparts momentum to the copper, not heat. The standing waves provide the stored energy, but the evanescent waves do the work.


Offline WarpTech

  • Full Member
  • ****
  • Posts: 1223
  • Do it!
  • Vista, CA
  • Liked: 1290
  • Likes Given: 1739
...The part that determines thrust, based on the shape of the cavity, was neglected by Egan and Yang in their analyses. Their resulting closed form equation is correct, the D, E and B, H fields they plug into it are not....
If you are correct that Prof. Yang neglected the part that determines the thrust in her analysis, how is she then calculating a thrust force?

Something is amiss.  Greg Egan makes sense, because he concludes that there is no thrust.

But how can we say that on one hand Prof. Yang neglects the part that determines thrust and on the other hand she calculates a thrust force?  Where does her calculated thrust force come from then ?

I'm saying, in her analysis she neglects to "provide" information on what her functions for E, H are. We have no idea what she plugged into her FEA software. Egan on the other hand, doesn't even consider it and assumes all waves are sin(wt) & cos(wt), therefore there is no force. Yang doesn't say what the functions are she uses. If her amplitude for E and H include an exponential decaying factors that are asymmetrical, then they are not periodic and that's all she needs.

Todd


Offline MyronQG

  • Member
  • Posts: 14
  • Liked: 9
  • Likes Given: 5
...The part that determines thrust, based on the shape of the cavity, was neglected by Egan and Yang in their analyses. Their resulting closed form equation is correct, the D, E and B, H fields they plug into it are not....
If you are correct that Prof. Yang neglected the part that determines the thrust in her analysis, how is she then calculating a thrust force?

Something is amiss.  Greg Egan makes sense, because he concludes that there is no thrust.

But how can we say that on one hand Prof. Yang neglects the part that determines thrust and on the other hand she calculates a thrust force?  Where does her calculated thrust force come from then ?

I'm saying, in her analysis she neglects to "provide" information on what her functions for E, H are. We have no idea what she plugged into her FEA software. Egan on the other hand, doesn't even consider it and assumes all waves are sin(wt) & cos(wt), therefore there is no force. Yang doesn't say what the functions are she uses. If her amplitude for E and H include an exponential decaying factors that are asymmetrical, then they are not periodic and that's all she needs.

Todd

Dr. Yang's calculation of thrust is deeply flawed. Although the integration of Maxwell stress tensor over the thruster boundary is the way to proceed, if no EM radiation is allowed (no photon rocket) it is easy to prove that that integration amounts exactly to zero, since nothing is coming out from a surface at infinity and the total electromagnetic field momentum within the volume enclosed by those surfaces is zero, too. As a result no thrust at all.

If, by the other hand, the non-zero thrust resulting from FEA calculation is due to a kind of directional EM radiation, the radiated ouptut power would be around 100 MW, way larger than the total input power, so that thrust result is very likely due to numerical errors.
« Last Edit: 05/29/2015 03:17 PM by MyronQG »

Offline WarpTech

  • Full Member
  • ****
  • Posts: 1223
  • Do it!
  • Vista, CA
  • Liked: 1290
  • Likes Given: 1739
...
Dr. Yang's calculation of thrust is deeply flawed. Although the integration of Maxwell stress tensor on the thruster boundary is the way to proceed, if no EM radiation is allowed (no photon rocket) it is easy to prove that that integration amounts exactly to zero, since nothing is coming out from a surface at infinity and the total electromagnetic field momentum within the volume enclosed by those surfaces is zero, too. As a result no thrust at all.
...

No, this is where you are making an assumption. The EM field momentum within the volume is NOT zero. The symmetry with which it is attenuated and absorbed by the cavity, will determine the forces. It is not "de-facto" symmetrical on all surfaces.

Online Rodal

  • Senior Member
  • *****
  • Posts: 5833
  • USA
  • Liked: 5901
  • Likes Given: 5249
...The part that determines thrust, based on the shape of the cavity, was neglected by Egan and Yang in their analyses. Their resulting closed form equation is correct, the D, E and B, H fields they plug into it are not....
If you are correct that Prof. Yang neglected the part that determines the thrust in her analysis, how is she then calculating a thrust force?

Something is amiss.  Greg Egan makes sense, because he concludes that there is no thrust.

But how can we say that on one hand Prof. Yang neglects the part that determines thrust and on the other hand she calculates a thrust force?  Where does her calculated thrust force come from then ?

I'm saying, in her analysis she neglects to "provide" information on what her functions for E, H are. We have no idea what she plugged into her FEA software. Egan on the other hand, doesn't even consider it and assumes all waves are sin(wt) & cos(wt), therefore there is no force. Yang doesn't say what the functions are she uses. If her amplitude for E and H include an exponential decaying factors that are asymmetrical, then they are not periodic and that's all she needs.

Todd

Dr. Yang's calculation of thrust is deeply flawed. Although the integration of Maxwell stress tensor over the thruster boundary is the way to proceed, if no EM radiation is allowed (no photon rocket) it is easy to prove that that integration amounts exactly to zero, since nothing is coming out from a surface at infinity and the total electromagnetic field momentum within the volume enclosed by those surfaces is zero, too. As a result no thrust at all.

If, by the other hand, the non-zero thrust resulting from FEA calculation is due to a kind of directional EM radiation, the radiated ouptut power would be around 100 MW, way larger than the total input power, so that thrust result is very likely due to numerical errors.
That's only true if there are no sinks or no sources (zero divergence).
If there is a sink (attenuation or losses, for example) then it is easy to prove that the time-averaged (over an integral number of periods) Poynting vector is non-zero.
« Last Edit: 05/29/2015 03:25 PM by Rodal »

Offline MyronQG

  • Member
  • Posts: 14
  • Liked: 9
  • Likes Given: 5
...
Dr. Yang's calculation of thrust is deeply flawed. Although the integration of Maxwell stress tensor on the thruster boundary is the way to proceed, if no EM radiation is allowed (no photon rocket) it is easy to prove that that integration amounts exactly to zero, since nothing is coming out from a surface at infinity and the total electromagnetic field momentum within the volume enclosed by those surfaces is zero, too. As a result no thrust at all.
...

No, this is where you are making an assumption. The EM field momentum within the volume is NOT zero. The symmetry with which it is attenuated and absorbed by the cavity, will determine the forces. It is not "de-facto" symmetrical on all surfaces.

I refer to the volume enclosed by the surface at infinity and the thruster's boundary (topologically equivalent to a hollow sphere).

Offline Notsosureofit

  • Full Member
  • ****
  • Posts: 656
  • Liked: 704
  • Likes Given: 1361
...
When you state:  <<Optimize their integral force equation for the correct representation for D, E and B, H, i.e., NOT standing waves>>

are you proposing to only consider the evanescent wave terms and to ignore the standing wave terms in the analysis? (since the standing wave terms perfectly cancel out when averaged over an integer number of time periods)

Or do you see any interaction/coupling between the standing wave terms and the evanescent wave terms so that both have to be considered in the analysis?
....


The interaction is, as the standing waves dissipate heat in the copper, they lose energy and shift to longer wavelengths, where attenuation takes over from dissipation and imparts momentum to the copper, not heat. The standing waves provide the stored energy, but the evanescent waves do the work.

Minor point, but the standing waves lose energy to the electrons in the copper, which in turn, radiate at SHORTER wavelengths.  IR has shorter wavelengths than microwave.

Offline Flyby

  • Full Member
  • ***
  • Posts: 382
  • Belgium
  • Liked: 447
  • Likes Given: 46
The interaction is, as the standing waves dissipate heat in the copper, they lose energy and shift to longer wavelengths, where attenuation takes over from dissipation and imparts momentum to the copper, not heat. The standing waves provide the stored energy, but the evanescent waves do the work.

Todd, although I understood the general lines of the mechanism you're describing, I'm a bit confused about the frequency shifting.
Initially you said that the shape of the cavity was the reason for the attenuation of the waves, where now you seem to put the wavelength shift to thermal causes?

Furthermore, doesn't the heat have a negative impact on the Q ? with a higher Q, you can store more energy in the standing waves, but more energy means more heat... At some point they must equalize eachother, no?

That would also mean that Shawyer's projections of having Q's running into the millions will probably never materialize?
Unless there are other ways to achieve attenuation without the thermo effects?
« Last Edit: 05/29/2015 04:14 PM by Flyby »

Offline CraigPichach

  • Member
  • Posts: 59
  • Calgary Alberta
  • Liked: 42
  • Likes Given: 4
Ok I am sold on following the data regardless of the cause (EMdrive theory vs. Quantum Vacuum) - the data thus far shows thrust but confidence would be much higher in an experiment where the thrust power is far in excess of any thermal effects (and with increased thrust/disruption of the QV any space warp experiments should be easier to pull off).

If we were going to design an experiment where we generate enough thrust that the results were obvious (i.e. thrust more than overcomes any thermal effects) I see NASA has proposed a 100kW magnetron firing into a 10" tall copper frustum. Some questions relating to that experiment:

1) Water cooled - when they say water cooled is that just the magnetron anode cooling or are they cooling the copper frustrum? Is there any advantage to cooling the copper frustrum - I had heard that with EM-Drive theory that going to cryogenic temps would boost Q factor (and with that thrust). Is there any advantage to say cooling the frustrum to say -5degC by firing this in a liquid jacket. Alternatively do I want to use natural gas and a JT choke and freeze this thing to say -30degC on the outside?
2) Vacuum air - Is there any benefit to putting a vacuum pump on the frustum; my understanding is that the q factor will be higher in a vacuum. If so any thoughts as to how to keep the air circulating out of the frustum at a vacuum but keep the microwaves within the chamber?
3) Silver platting - I think we are all in agreement that silver plating on the copper will boost the q factor? Also are we in agreement that copper will have a higher q than aluminum?

I am an engineer in oil and gas with degrees in Mechanical and Chemical engineering; my brother is an electrical engineering student at UofS. I have access to  fabricators and electrical contractors that I don't see this experiment being that expensive apart from the 100kW magnetron and associated transformers. I am tempted to try this out to see it with my own eyes.



Offline CraigPichach

  • Member
  • Posts: 59
  • Calgary Alberta
  • Liked: 42
  • Likes Given: 4
Apologies, just reading my post had the solution to the item on a vacuum; question #2  should just read - is there any benefit to inducing a vacuum with a vacuum pump prior to a q thruster test? If the frustrum is air tight than by pulling a vacuum we can just disconnect the pump eliminating the problem of lost microwaves. So questions:
#1 - Cooling jacket - worth trying to cool the q thruster?
#2 - internal vacuum - worth trying to run it at a vacuum, say -5kPag
#3 - Silver plated copper - better q factor than aluminum? No dielectric needed with magnetron due to harmonics?


Offline sfrank

  • Member
  • Posts: 8
  • USA
  • Liked: 4
  • Likes Given: 27
...
If we were going to design an experiment where we generate enough thrust that the results were obvious (i.e. thrust more than overcomes any thermal effects) I see NASA has proposed a 100kW magnetron firing into a 10" tall copper frustum. ...

That's news to me. Per the article by Rodal et al:
Quote
Due to these predictions by Dr. White’s computer simulations NASA Eagleworks has started to build a 100 Watt to 1,200 Watt waveguide magnetron microwave power system that will drive an aluminum EM Drive shaped like a truncated cone.

Apologies, just reading my post had the solution to the item on a vacuum; question #2  should just read - is there any benefit to inducing a vacuum with a vacuum pump prior to a q thruster test? If the frustrum is air tight than by pulling a vacuum we can just disconnect the pump eliminating the problem of lost microwaves. So questions:
#1 - Cooling jacket - worth trying to cool the q thruster?
#2 - internal vacuum - worth trying to run it at a vacuum, say -5kPag
#3 - Silver plated copper - better q factor than aluminum? No dielectric needed with magnetron due to harmonics?

A thin copper fustrum would deform if you tried to induce a vacuum inside it (and not outside as well).  Plus it would be really difficult to hermetically seal, especially around the RF input.  A full vacuum enclosure is the best way to go.  NASA did the one test that way, and @zellerium and his colleagues at Cal Poly are planning vacuum tests this summer.  The guys at hackaday.io are also thinking about a vacuum test of their 25ghz thruster.  That one is machined out of an aluminum block and small enough to fit the entire test in a vacuum bell jar.

Cooling is certainly important. Overheating certainly limits the duration of testing right now.  I think if this tech pans out, designing an optimal heat exchanging system will be an important part of creating a successful device.    As for the silver plating, you're right that its been brought up a number of times as a good experimental modification to try.  I added it to the list of proposed experiments on the wiki: http://emdrive.echothis.com/List_of_Suggested_Experiments   

Finally, though I'm not a microwave engineer, I have to say that pumping 100kW of energy in to a 10" copper can doesn't sound very safe.  If you're thinking about building your own, that's great, but I would definitely start with a conventional 800-1200w magnetron like the kind in a microwave.
« Last Edit: 05/29/2015 04:54 PM by sfrank »

Offline mwvp

  • Full Member
  • **
  • Posts: 254
  • Coincidence? I think Not!
  • Liked: 169
  • Likes Given: 30
...
Dr. Yang's calculation of thrust is deeply flawed. Although the integration of Maxwell stress tensor on the thruster boundary is the way to proceed, if no EM radiation is allowed (no photon rocket) it is easy to prove that that integration amounts exactly to zero, since nothing is coming out from a surface at infinity and the total electromagnetic field momentum within the volume enclosed by those surfaces is zero, too. As a result no thrust at all.
...

No, this is where you are making an assumption. The EM field momentum within the volume is NOT zero. The symmetry with which it is attenuated and absorbed by the cavity, will determine the forces. It is not "de-facto" symmetrical on all surfaces.

I refer to the volume enclosed by the surface at infinity and the thruster's boundary (topologically equivalent to a hollow sphere).

Restating my earlier conjecture:

EM Momentum is asymmetric in the frustrum due to copper loss, and selectively thermally dissipated.

The loss is greater in the wide back, more than the narrow front

The wide back is inductive, the narrow front capacitive; the vacuum is a loss-less dielectric.

The wide back gets more low-sideband current and heat, the narrow front more E-field and momentum.

The frustrum selectively dissipates energy supplied at a point on its tuning-slope.

If its accelerated, sidebands appear, one of which may be selectively dissipated, resulting in the acceleration being enhanced or retarded depending on the side of the center-frequency energy is supplied at.

Would be nice if Aero could modify his model to include copper loss in the waveguide. I'm still looking at old thread 2 posts. It would also be nice if Aero could post his .ctl file. Thanks.

Online aero

  • Senior Member
  • *****
  • Posts: 2736
  • 92129
  • Liked: 704
  • Likes Given: 236

AERO:   When you considered evanescent waves in the Truncated Cone, modelled as a two-dimensional flat trapezium:

1) Considering only evanescent waves inside the cavity, did MEEP calculate any net thrust force?  If the answer is no, why not? (Did you consider any losses in your analyis? Did you input a tan delta value for the Dielectric polymer insert?)

2) Was attenuation taking into account in the MEEP analysis? If yes, what were the attenuation parameters that you considered?



I never used a detecter inside the cavity. Meep tutorial advises against that. Hence I have no data from inside the cavity.

There was always a force outside the cavity except when the cavity was completely closed or the gaps were below the resolution of the run. (Same as closed.) The force itself ranged from miniscule to significant but the Force/Power was mostly in the vacinity of 1, typically from 0.9 to 1.1.

I only used a dielectric constant ranging from 1.76 to 2.7. I never did figure out how to convert the loss coefficient (tan delta) to a Drude model that I could use in Meep.

Attenuation - Nothing else. Perfect metal, a dielectric constant and the correct geometry for the EW cavity.

The Meep results were not very repeatable, results changed with a change in resolution. That's a big part of the reason I've quit running cases on the cavities used in the various experiments. The field patterns are pretty though.
Retired, working interesting problems

Tags: