Author Topic: EM Drive Developments - related to space flight applications - Thread 3  (Read 1876600 times)

Offline ElizabethGreene

  • Member
  • Posts: 69
  • Nashville, Tennessee
  • Liked: 138
  • Likes Given: 3
I love Mr. Feynman, and I would love to have time with him.  I _did_ the sprinkler experiment.  :)  It spins backwards a smidgen, then stops.  If you get the flow rate just right, the flow spirals into the nozzles like the swirl of a toilet or drain. 

I will, one day, have a go at the charged top and collapsing B field paradox.  One day.

Offline WarpTech

  • Full Member
  • ****
  • Posts: 1311
  • Do it!
  • Vista, CA
  • Liked: 1350
  • Likes Given: 1816
The two are not equal, because the wave decays in the time longer time t + dt, before it can transfer its momentum to the other side.
In the most extreme case, where the photon is completely absorbed, all its momentum is transferred to the endplate on the first contact. But even in this case, the momentum applied to the endplate will be equal and opposite to the momentum applied to the power source when the photon was generated, so there would be no net thrust.

The most efficient way to achieve thrust from photon momentum is to remove the resonator entirely and simply emit the photons away from the spacecraft in a coherent beam.

Your last statement is not news to me. However, what you said above is not what I've proposed. I've decoupled the thrust from the magnetron. The magnetron simply charges the resonator to an enormous energy level Q and is then turned off. Simultaneously, release the docking clamps that held the frustum in place while charging.

Now, which way will it move as the energy stored inside decays to zero, considering that the attenuation is not symmetrical?

It started from 0-NET momentum but inside it has a stored oscillation that is rapidly decaying. The time-average of these decaying waves exert unequal pressures on the walls due to constructive and destructive interference. Thrust will depend on rate of decay of the pressures in both directions, and which one decays faster, like a game of tug-of-war. Whoever can hold out the longest wins.

I'm thinking that the pressure will decay faster at the big end because destructive interference will starve it of the return feedback. Where the small end receives the reflections from the big end and converts that energy into "work", F = qv x B, depending on the phase angle between the field and the eddy currents, which determines the relative Power Factor. The tapered cavity allows there to be phase angles other than pi/2, which exist for reflection from a flat end plate.
Todd

Offline TheTraveller

Wait a minute, I missed something… and I can't find it searching the thread. Where did we obtain Yang's frustum dimensions from? A few pages ago we complained she didn't published them.

Try here:
http://www.emdrive.com/NWPU2010translation.pdf

Dimensions and modes all over the original paper.
"As for me, I am tormented with an everlasting itch for things remote. I love to sail forbidden seas.”
Herman Melville, Moby Dick

Offline WarpTech

  • Full Member
  • ****
  • Posts: 1311
  • Do it!
  • Vista, CA
  • Liked: 1350
  • Likes Given: 1816
... you seem to violate CoM but don't let that be a brick wall just because someone tells you it does ...

As it relates to the EmDrive I've put Conservation of momentum on a shelf for later examination. Candidly, I have quietly wondered if it might have a loophole or two for someone who was clever enough to find them. I seriously doubt I'm that clever.

Playing in that vein, I have a related thought experiment.

Imagine the space shuttle in orbit. The cargo bay doors are open. An astronaut equipped with an MMU fires his gas jets, accelerates from space outside the shuttle, and smacks into the back wall of the cargo bay. The astronaut transfers his momentum to the shuttle, and recieves a concussion for his trouble.

The momentum flow is Gas from the Jets go left, shuttle goes right. CoM is satisfied.

Two asprin later the astronaut+mmu is in the cargo bay with the bay doors closed and sealed. The bay is a hard vacuum. The astronaut reluctantly fires the MMU jets and smacks into the back wall again. A small amount of momentum is transferred to the shuttle, and the astronaut rethinks his "glamorous" career with Nasa.

For conservation of momentum to be satisfied the force of the gas striking the inside of the cargo bay must exactly balance the force of the shuttle to the right for there to be no net momentum change. This exact balance of gas pressure does not match my understanding of gas behavior at all. I expect instead to see the all kinds of non-Newtonian action in the gas as it expands randomly into the bay in all directions. Turbulence and brownian motion will rob energy out of the gas literally left and right.

I also don't see what would prevent the astronaut from pulling out a vacuum pump and compressing it back into the MMU's cylinders for another shot.

I have an idea about how to test this here on earth, but the EmDrive work seems much more urgent and promising.

Could someone point out the obvious flaw in my thought experiment?  Despite having it drawn on my bathroom mirror since October, I've still not managed to see it.

I had the same idea! If the astronaut were smart, he would close the bay doors and put his back up against the wall before firing his MMU. The shuttle would start moving instantly, before the gas could reach the other side. When it finally reaches the other side, turbulence and heat will have dissipated much of it's energy and nothing needs to leave the shuttle. I can't imagine why it would not work.

Now if we had 2 astronauts, one on either side. Each with his own MMU device opposing each other, and one of them has a longer burst than the other....

Offline Prunesquallor

  • Full Member
  • *
  • Posts: 174
  • Currently, TeV Brane Resident
  • Liked: 157
  • Likes Given: 73

I had the same idea! If the astronaut were smart, he would close the bay doors and put his back up against the wall before firing his MMU. The shuttle would start moving instantly, before the gas could reach the other side. When it finally reaches the other side, turbulence and heat will have dissipated much of it's energy and nothing needs to leave the shuttle. I can't imagine why it would not work.

And then run a vacuum pump to harvest the gas, compress it and put it back in the backpack. Repeat as needed.
Retired, yet... not

Offline rfmwguy

  • EmDrive Builder (retired)
  • Senior Member
  • *****
  • Posts: 2165
  • Liked: 2681
  • Likes Given: 1124
DIY Challenge...lost in theories? Skeptical? Go retro and construct something yourself. Hewlett and packard did. Jobs did. Wright brothers, etc. Many others started small and dreamed big. Theoriticians need practical data. Braintrusts here at nsf need data...just do it. You are hereby challenged in the spirit of innovation. Who's in? Succeed or fail, u can help....I am and will be happy to prove or disprove, for all that is important is the effort...snarky comments have zero value, imo.

Offline SeeShells

  • Senior Member
  • *****
  • Posts: 2335
  • Every action there's a reaction we try to grasp.
  • United States
  • Liked: 2982
  • Likes Given: 2601
... you seem to violate CoM but don't let that be a brick wall just because someone tells you it does ...

As it relates to the EmDrive I've put Conservation of momentum on a shelf for later examination. Candidly, I have quietly wondered if it might have a loophole or two for someone who was clever enough to find them. I seriously doubt I'm that clever.

Playing in that vein, I have a related thought experiment.

Imagine the space shuttle in orbit. The cargo bay doors are open. An astronaut equipped with an MMU fires his gas jets, accelerates from space outside the shuttle, and smacks into the back wall of the cargo bay. The astronaut transfers his momentum to the shuttle, and recieves a concussion for his trouble.

The momentum flow is Gas from the Jets go left, shuttle goes right. CoM is satisfied.

Two asprin later the astronaut+mmu is in the cargo bay with the bay doors closed and sealed. The bay is a hard vacuum. The astronaut reluctantly fires the MMU jets and smacks into the back wall again. A small amount of momentum is transferred to the shuttle, and the astronaut rethinks his "glamorous" career with Nasa.

For conservation of momentum to be satisfied the force of the gas striking the inside of the cargo bay must exactly balance the force of the shuttle to the right for there to be no net momentum change. This exact balance of gas pressure does not match my understanding of gas behavior at all. I expect instead to see the all kinds of non-Newtonian action in the gas as it expands randomly into the bay in all directions. Turbulence and brownian motion will rob energy out of the gas literally left and right.

I also don't see what would prevent the astronaut from pulling out a vacuum pump and compressing it back into the MMU's cylinders for another shot.

I have an idea about how to test this here on earth, but the EmDrive work seems much more urgent and promising.

Could someone point out the obvious flaw in my thought experiment?  Despite having it drawn on my bathroom mirror since October, I've still not managed to see it.
I love your thought experiment. Nobel prize winner Ernest Rutherford said: "If you can't explain your physics simply, it's probably not very good physics." I've lived by this and I believe our visual imaginary is so much more powerful than just cranking out and solving formulas, they should be a tool for our imagination.

Simple to solve your MMU problem as you go from a open entropy environment with your doors open to a closed entropy with the doors closed. No matter how you disperse your MMU jets and no matter how or what conversion to another form of energy, the thrust is contained within the Shuttles bay and your momentum banging into a wall will not equate to the outside.

I have this air tank I use to fill up tires if they go flat on my cars (and don't ask I have too many ;) ). Lets say I put a set of speakers in the tank with my Ipod with 0 PSI, then call my Ipad and it starts playing some Led Zeppelin ... full volume. The tank sits there, not moving.  I pressurize the tank to 200 psi and do the same, make a call and Led Zeppelin plays again. Nothing happens, the tank will not move.

Opening the tank to get my Ipad and speakers I forget to depressurize the tank. Zoowie off it goes and I hope it doesn't hit my 36 Pontiac!

Someone is playing Led Zeppelin in the Em Frustum and it's moving.

I also shelved CoE and CoE (for now) and I'm embracing quantum effects as standard physics is butting heads with it and will not give us answers we need.

Good to see you here.

Shell




Offline Rodal

  • Senior Member
  • *****
  • Posts: 5895
  • USA
  • Liked: 6045
  • Likes Given: 5325
I had previously posted a mathematical proof that one cannot define the attenuation and phase constant solely on the basis of the cone angle.  Here is a geometrical proof that Zeng and Fan are misleading in their text (their equations and figures are correct, though).  If one uses the cone angle, one also needs to define what is the distance from the small base to the apex of the cone.

Imagine two different truncated cones, one blue and one red, both having the same length  and the same small diameter.



The only difference is that the blue truncated cone has a larger big diameter than the red truncated cone. Therefore the blue cone has a significantly larger cone angle than the red cone.

In the blue cone, one hase other modes that get cut off.   All the modes that get cut off in the red cone also get cut of in the bluecone.

So, why should the geometrical attenuation of the blue cone be worse than the geometrical attenuation of the red cone?


That's the question I posed in a previous message.  The answer is that it is incorrect to state that the blue cone has worse attenuation than the red cone, solely on the basis of the cone angle.  One has to also take into account the distance between the small base and the apex of the cones (which is given by their spherical radii r1) in order to determine their attenuation.  This is evident when showing the distance to the apex as in the following picture:



which shows that the blue cone has the small base much closer to the apex of the cone than it is the case for the red cone.  The distance to the apex of the cone is critical for attenuation and the phase constant and disregarding it leads to absurd conclusions.

In the figures of Zeng and Fan, when comparing different cone angles at the same spherical radial distance r, one is really considering a truncated cone with a much smaller small base, like this green truncated cone that has the small base at the same distance to the apex as the blue cone:



Looking at that green cone it becomes now apparent the benefit of a small cone angle and small distance to the apex: it leads to a very pointy geometry with a lot of attenuation.

Here is a comparison showing the blue and the green truncated cones by themselves:



Observe that this green truncated cone looks very different from the EM Drive geometries that have been tried up to now.

So, in a few words: to put what Zeng and Fan write about "geometrical attenuation" in the language that people familiar with the EM Drive are familiar with:

the ideal geometrical attenuation is achieved by an EM Drive geometry that has BOTH a minimum distance from the small base to the apex of the cone and a small cone angle, what this means, effectively is that the small diameter of the truncated cone should be as small as possible

This is nothing new to people already familiar with the following formulas:

*Shawyer's Design Factor

*McCulloch's formula

*@Notsosureofit's fomula

All these formulas show that the best geometry is the one that maximizes the difference between the small base and the large base of the truncated cone.

Zeng and Fan show precisely the same: the best geometry is the one that maximizes the difference between the bases.

Also observe that the General Relativity theory of Marco Frasca also shows the same

There is a singularity at the apex of the cone.  Attenuation is maximized the closer the small base is to the apex of the cone.  The General Relativity effect is maximized the closer one gets to the apex of the cone.  Unruh wave effect is maximized the closer the small base is to the apex of the cone.  Notsosureofit's dispersion gets maximized the closer the small base is to the apex of the cone. Shawyer's group velocity difference is maximized the closer the small base is to the apex of the cone.  All these theories agree.

So why hasn't this been tried?  It has to do with what Todd calls "the tug of war" between energy  storage Q and evanescent waves.  Up to now researchers have been concerned with maximizing Q and concerned that if they make the small base too small, they will not have a high Q cavity.   But it should be tried, because indications are (i.e. the low effective Q used by the Chinese to achieve high thrust) that the optimal design may enjoy a smaller small diameter than the ones tried up to now.
« Last Edit: 06/14/2015 02:31 AM by Rodal »

Offline deltaMass

  • Full Member
  • ****
  • Posts: 955
  • A Brit in California
  • Liked: 671
  • Likes Given: 275
So what's the big picture wrap up of 2D vs 3D?

Offline aero

  • Senior Member
  • *****
  • Posts: 2785
  • 92129
  • Liked: 724
  • Likes Given: 249
...
Three dimensions         
flux1:   2.25E+000   7.13E-009   -4.8601123862
force1:   2.25E+000   -3.46E-008   
Two dimensions         
flux1:   2.25E+000   2.24E-010   -4.4050439326
force1:   2.25E+000   -9.85E-010   

Check the file names to determine what was a 2D run and what is a slice of a 3D run. Note the files ending in 02.png show the antenna location. That is, 2 time steps after start

So what's the big picture wrap up of 2D vs 3D?
the distribution of the electromagnetic field inside and outside look very different

and the calculated values look quite different too

Which way is the force ? is it pointing towards the small end forward? or is the opposite way?

In the attached, the small circle is at the 0, 0 orign, and the large circle is at +1, 0. There are two detectors located at +1.5, 0 and -1.5, 0. The Flux at the detectors is summed (its expended energy), the forces at the detectors are the differenced. Perhaps I have the subtraction backwards. Using EW test results, the force is commonly in the minus direction. But then, no body has ever ran an experiment that is anything like this with a dielectric penetrating the frustum side walls. At least the external wave pattern agrees with the force direction, I think. If necessary, I could make runs detecting the forces individually so that we could add them by hand, or I could replace the frustum with a planewave source, then detect the forces individually. In that case we would know what the force signs and magnitudes should be analytically.

As for the difference in the field patterns, ?? don't know. I know that the resonant frequency is commonly different in meep between 2D and 3D but in this case I did not check resonance, but just used the identical Gaussian center drive frequency and bandwidth. Maybe one of the other meep users could provide insight?

Edit Add 06/14/15
After sleeping on it I realize than I'm going to need to do this all over again, checking the boundary layers and sign conventions as a minimum.

This control file started out as a new file, structured to be a simple, closed, fast running cavity model suitable for posting. Opening it with the dielectric out the sides violates more than one of my original assumptions. I'll get a new set of data today or this week, depends on the challenges.
« Last Edit: 06/14/2015 03:17 PM by aero »
Retired, working interesting problems

Offline ThinkerX

  • Full Member
  • ***
  • Posts: 311
  • Alaska
  • Liked: 120
  • Likes Given: 59
Ok...it be way past time for one of the DIY types here to take a shot at duplicating the Yeng / Fan EM Drive design. 

Offline Wetmelon

Could someone point out the obvious flaw in my thought experiment?  Despite having it drawn on my bathroom mirror since October, I've still not managed to see it.

I think that as soon as the container (shuttle) starts to move, it creates a slight pressure gradient within the container which creates forces that oppose that motion, even in a situation with a rarefied gas.  It might take a long time for everything to go back to "zero", but it eventually would.  That's my prediction, anyway.

Offline WarpTech

  • Full Member
  • ****
  • Posts: 1311
  • Do it!
  • Vista, CA
  • Liked: 1350
  • Likes Given: 1816
Could someone point out the obvious flaw in my thought experiment?  Despite having it drawn on my bathroom mirror since October, I've still not managed to see it.

I think that as soon as the container (shuttle) starts to move, it creates a slight pressure gradient within the container which creates forces that oppose that motion, even in a situation with a rarefied gas.  It might take a long time for everything to go back to "zero", but it eventually would.  That's my prediction, anyway.

Yup, that's why we put the vacuum pump on the "other" side of the shuttle. :)

Offline ThinkerX

  • Full Member
  • ***
  • Posts: 311
  • Alaska
  • Liked: 120
  • Likes Given: 59
Quote
I think that as soon as the container (shuttle) starts to move, it creates a slight pressure gradient within the container which creates forces that oppose that motion, even in a situation with a rarefied gas.  It might take a long time for everything to go back to "zero", but it eventually would.  That's my prediction, anyway.

Yet, at least initially, it should move the shuttle, right?


Offline Star One

  • Senior Member
  • *****
  • Posts: 8265
  • UK
  • Liked: 1339
  • Likes Given: 168

Offline SeeShells

  • Senior Member
  • *****
  • Posts: 2335
  • Every action there's a reaction we try to grasp.
  • United States
  • Liked: 2982
  • Likes Given: 2601
Ok...it be way past time for one of the DIY types here to take a shot at duplicating the Yeng / Fan EM Drive design.

I am and have been working on it.

Shell

Offline madsci

  • Member
  • Posts: 22
  • Liked: 4
  • Likes Given: 3
Data from second Baby EmDrive test

https://hackaday.io/project/5596-em-drive/log/19468-torsion-test-2-data

  It seems that they no longer use the magnetic suspension as in their first test.
  That's good since the magnetic field could have coupled with the rest of the assembly and transmit momentum to it.

  Did they explain their reasons for abandoning the magnetic suspension ? I can't find anything on their web site.

Offline TMEubanks

... you seem to violate CoM but don't let that be a brick wall just because someone tells you it does ...

As it relates to the EmDrive I've put Conservation of momentum on a shelf for later examination. Candidly, I have quietly wondered if it might have a loophole or two for someone who was clever enough to find them. I seriously doubt I'm that clever.

Playing in that vein, I have a related thought experiment.

Imagine the space shuttle in orbit. The cargo bay doors are open. An astronaut equipped with an MMU fires his gas jets, accelerates from space outside the shuttle, and smacks into the back wall of the cargo bay. The astronaut transfers his momentum to the shuttle, and recieves a concussion for his trouble.

The momentum flow is Gas from the Jets go left, shuttle goes right. CoM is satisfied.

Two asprin later the astronaut+mmu is in the cargo bay with the bay doors closed and sealed. The bay is a hard vacuum. The astronaut reluctantly fires the MMU jets and smacks into the back wall again. A small amount of momentum is transferred to the shuttle, and the astronaut rethinks his "glamorous" career with Nasa.

For conservation of momentum to be satisfied the force of the gas striking the inside of the cargo bay must exactly balance the force of the shuttle to the right for there to be no net momentum change. This exact balance of gas pressure does not match my understanding of gas behavior at all. I expect instead to see the all kinds of non-Newtonian action in the gas as it expands randomly into the bay in all directions. Turbulence and brownian motion will rob energy out of the gas literally left and right.

I also don't see what would prevent the astronaut from pulling out a vacuum pump and compressing it back into the MMU's cylinders for another shot.

I have an idea about how to test this here on earth, but the EmDrive work seems much more urgent and promising.

Could someone point out the obvious flaw in my thought experiment?  Despite having it drawn on my bathroom mirror since October, I've still not managed to see it.

This is really pretty simple. Gas in a vacuum behaves ballistically (i.e., the molecules travel in straight lines until they hit something, or some other force bends their trajectory). When you say, "Turbulence and brownian motion will rob energy out of the gas literally left and right," you are thinking of a dense gas, not a vacuum. Note also that turbulence does not "rob" energy - the kinetic energy (in a thick gas) gets converted to random motions of the gas, and eventually to random motions of the atoms (i.e., the gas gets a higher temperature). The energy is still there, it's just disordered. Any net momentum is conserved, just transferred between gas molecules.* But, if you agree that sending molecules out a nozzle impart momentum, you should be able to see that molecules impacting somewhere also imparts momentum. At a molecular level, the processes are the same.

Suppose there is 1 kg of gas expelled at 100 m/sec, the shuttle bay is 20 meters end to end, the astronaut starts in the middle, the astronaut + suit weighs 100 kg, and the Shuttle weighs 10 tons (10^4kg). When the astronaut fires the MMU, the gas molecules go left at 100 m / sec, and the astronaut goes right at 1 m / sec. For order 100 milliseconds, the shuttle does not respond, and then gas molecules start hitting the far (left) end. Some are absorbed, some reflected, and the shuttle starts moving left at 1 cm / sec (or thereabouts, depending on just how the molecules inside the bay are bouncing around). This continues for about 10 seconds, until the astronaut hits the right wall. The astronaut is just another particle here, more massive than the gas molecules, but following the same dynamics. In your model (assuming the astronaut goes to sleep, and doesn't do anything else) it could take a good while for the system to settle down - the astronaut could keep bouncing from wall to wall, with the shuttle moving about 20 cm in the opposite direction in the 20 seconds or so between each such collision.

Also note that if the Shuttle bay was open, say facing the Earth, firing the MMU would impart a thrust to the shuttle in the direction opposite the bay direction (in this case, away from the Earth), even if all of the gas momentum was originally perpendicular to that direction - this is really just a (very inefficient) rocket engine. In a rocket engine reaction chamber, molecules are heated and move every which way, bouncing off each other and the walls of the chamber. Only the ones moving in the direction of the nozzle escape, and so the net momentum imparted to the chamber is in the opposite direction. (Generally rocket exhaust has a high enough pressure - i.e., enough gas molecule collisions - that the momentum transfer on the walls of the nozzle is also significant, but this doesn't change any of the principles here.) In your example, the molecule move originally one way, but are randomized in collisions by the walls (and the astronaut!) and the ones that have velocity vectors pointing out the bay escape, leaving the momentum transfer of their last collision with the bay walls, and thus moving the shuttle in the opposite direction.     

*All of this continues to work in denser air, but the motions are more complicated. Suppose you blow up a balloon, and hold it with the throat pinched off. The gas molecule motions inside the balloon are complicated, but they all balance, and the balloon has no thrust. Once you let go, the throat opens, molecules start pouring out of the throat, and the balloon goes zooming in the opposite direction. Think about what's going on at a molecular level. Molecule 1 hits molecule 2, sending 1 towards the throat while 2 goes in the opposite direction (with no net momentum for the sum of the two molecules). Molecule 1 then escapes, leaving a net momentum (in molecule 2's velocity). It will not strike the wall of the balloon, but other molecules, until that net momentum is imparted to the far wall of the balloon and it starts moving in the direction opposite the throat. The mean free path in air is about 70 nanometers, so it takes millions of gas molecule collisions for this to happen, and we prefer (are forced) to take an averaged viewpoint, and say that there is pressure on the far wall which is imbalanced and thrusts the balloon, but it's really just billions and billions of molecule collisions that do this, the same physics as in your example. 

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5895
  • USA
  • Liked: 6045
  • Likes Given: 5325
I added the TE012 NASA test without dielectric insert that registered no thrust force to the EM Drive Experimental Results wiki:

http://emdrive.wiki/Experimental_Results

including a note reading: "@TheTraveller made an argument that the test may have been conducted at the wrong (too low) frequency for resonance"

QUESTION: does anybody have a link to TheTraveller's message where he points out that the test may have been conducted at the wrong frequency, so that the note can link to the message?   Thanks

Tags: