Em drives

[Urwumpe]

Assuming a cavity with f0 = 1GHz and fb = 10kHz, v_max = 3km/s.

v_max in respect to what? Earth? The milkyway? Local group? Your mom? :lol:

What if it is to the Milkyway... does that mean that firing it up results in a decelleration by 247 km/s?

Or do you want to say that actually geocentrism is it?

 

[RisingFury]

The device has to stay consistent in all frames of reference. It doesn't.

It depends on which frame it was in when it started moving...

 

[kamaz]

The device has to stay consistent in all frames of reference. It doesn't.

It depends on which frame it was in when it started moving...

Precisely.

Imagine an em-drive installed on a train moving at 100m/s and thrusting in the direction of motion.

McCulloch's math taken at face value means that the observer on the train and observer on the ground will observe the em-drive moving with different acceleration, because they see it operating at different frequencies, which means different Q-factors. Worse, for the observer on the train the device has a deltaV budget of 3km/s, however for the observer on the ground the device only has a remaining delta-v budget of 2.9km/s, because it's already detuned at power-up.

 

[jangofett287]

So either all of relativity, both special and general and everything built on top of it, is totally wrong, or the emDrive doesn't actually work. Remember for a theory to be replaced, the new one has to explain all the things the old one did and correct some or all of its errors. I've yet to see a theory that explains Mercury and the emDrive.

 

[jedidia]

McCulloch's math taken at face value means that the observer on the train and observer on the ground will observe the em-drive moving with different acceleration

I can't quite wrap my head around it at the moment, but wouldn't that basically break the simultaniety of events?

 

[Urwumpe]

I can't quite wrap my head around it at the moment, but wouldn't that basically break the simultaniety of events?

No, its quite normal in relativity that different observers can see you at different accelerations. Simultanous in relativity is also something different than in Newtonian physics.

Think of exploding stars - one is 200 ly away, the other 400 ly, but you see both explode at the same time. Does that mean that another observer will see them explode at the same time? No: for example an observer at the star 200 ly away might have reported 200 years ago that the other star exploded - but news of the event will not reach Earth until 200 years later.

 

[jedidia]

Simultanous in relativity is also something different than in Newtonian physics.

I am aware of that. simultaniety of events basically just means that if two people are at the same place at the same time, they'll observe the same events (though not neccessarily in the same manner) no matter what frame of reference they're in.

I always figured acceleration as a kind of absolute, though. I mean, it's measurable independant of reference frame, so... how can it be different in individual frames?

 

[Urwumpe]

I always figured acceleration as a kind of absolute, though. I mean, it's measurable independant of reference frame, so... how can it be different in individual frames?

Its the change of the change of position by time... can you imagine how relative this could get?

 

[jedidia]

Its the change of the change of position by time... can you imagine how relative this could get?

That's just the problem, I can't... considering that acceleration can be measured without the help of another reference frame (contrary to position and velocity). Because it's not just change of change of position per time, it's also force per mass. But then again, mass is relative... I just can't wrap my head around it at the moment. Will take some looking into the equations, I guess.

 

[Linguofreak]

Bzzzt, wrong. It requires invisible mass, period. The "weakly interacting massive particle" part is there because search for macroscopic candidates (MACHOs) turned out null.

Not null. "MACHO" is just a catchall term for any baryonic object too dim to be seen at galactic distances. There are plenty of those that we already know about in the solar neighborhood.

The search for MACHOs has nonetheless turned up evidence that there aren't enough of them to explain the missing mass in galaxies, but more importantly, even if those surveys underestimate the contribution of MACHOs to galactic mass, analysis of big bang nucleosynthesis puts a limit on what fraction of the missing mass can be baryonic.

Of course, the search for the particle has also turned out null...

The search for a particle with the right *mass* has turned out null. A particle with the right properties, but too little mass, was already known before dark matter was postulated (the neutrino).

Which theory is simpler: one that involves particles with properties identical or substantially similar to known particles, except for a different mass, behaving exactly according to the laws of gravity and motion now known (keeping in mind that the set of known particles contains several examples of families of particles with identical properties except for different masses already), or one that involves no extra particles, but involves significant changes to gravity and/or the laws of motion?

and one thing they all have in common is claimed modification of inertia.

With which you immediately run into problems with conservation of energy and momentum. I've heard that MOND can also be explained in terms of modifying the force law for gravity from 1/r^2 to 1/r under the appropriate circumstances instead of by modifying inertia, but inverse square forces are the default in 3 dimensional space, if your theory deviates from that, you need to have a mechanism explaining why it deviates. Also, an inverse force, rather than inverse square, has a potential that goes as the log of distance, rather than as the inverse of distance, which means that while an inverse square force's potential approaches a limit as distance goes to infinity, the potential of a straight-inverse force grows without limit, which has all kinds of potentially messy consequences.

---------- Post added at 04:40 ---------- Previous post was at 04:24 ----------

That's just the problem, I can't... considering that acceleration can be measured without the help of another reference frame (contrary to position and velocity). Because it's not just change of change of position per time, it's also force per mass. But then again, mass is relative... I just can't wrap my head around it at the moment. Will take some looking into the equations, I guess.

Acceleration is curvature of your worldline. An observer traveling at a different velocity than you will observe your worldline at a given moment to have a different curvature than what you observe.

There is hyperbolic geometry involved. This, especially section 7, may help you visualize what's going on: http://casa.colorado.edu/~ajsh/sr/sr.shtml

 

[kamaz]

I can't quite wrap my head around it at the moment, but wouldn't that basically break the simultaniety of events?

Worse than that, if the train frame and the ground frame are equivalent (as they should be per Newton and Lorentz) it means that em-drive is in two places at the same time.

White's model (as far as I understand it) does not have that problem, but then he basically introduces aether, and this causes other problems...

---------- Post added 06-07-15 at 11:39 AM ---------- Previous post was 06-06-15 at 09:30 PM ----------

Which theory is simpler: one that involves particles with properties identical or substantially similar to known particles, except for a different mass, behaving exactly according to the laws of gravity and motion now known (keeping in mind that the set of known particles contains several examples of families of particles with identical properties except for different masses already), or one that involves no extra particles, but involves significant changes to gravity and/or the laws of motion?

It does not matter which theory is simpler, it matters which theory can be (dis-)proven by observation or experiment. And you cannot throw out valid experimental data simply because it disagrees with your preferred theory. Yes, it works both ways

The em-drive either moves, or it does not. If it moves, it either involves new physics, or it does not. If it involves new physics, then new physics must be reconciliable with existing physics. Nobody is really questioning that.

Even if em-drive moves, it does not prove that Newton's laws are wrong, any more than a varistor proves that Ohm's law is wrong. But what em-drive critics are doing at the moment, is like saying that a varistor must be a fraud, because R is not constant, and Ohm's law demands that it must be constant. Duh. Ohm's law was formulated based on observations in metals, and semiconductors are not metals. As simple as that. You can of course extend Ohm's law to hold in semiconductors, by saying that 1/R is proportional to carrier density, and carrier density can be modified. But Ohm and his contemporaries did not know that you can modify carrier density, or even that there is such thing as carrier density. All they knew was that there was a dependency of R on material and temperature, and they wrote formulas based on that.

Newton's laws were formulated based on obsevations in absence of strong RF e-m fields. What makes you think that they still hold in presence of such fields? Or, more precisely, what makes you think that there is no electro-gravitic (or electro-inertial) coupling which causes additional forces to be generated in such regime? Thus leading to an apparent violation of Newton's law, simply because you now have unaccounted for forces present in the setup. At one time people believed that electricity and magnetism are separate, because let's face it, it's much more difficult to generate noticeable magnetic field using electric current than using a fixed magnet.

 

[Urwumpe]

Even if em-drive moves, it does not prove that Newton's laws are wrong, any more than a varistor proves that Ohm's law is wrong. But what em-drive critics are doing at the moment, is like saying that a varistor must be a fraud, because R is not constant, and Ohm's law demands that it must be constant. Duh. Ohm's law was formulated based on observations in metals, and semiconductors are not metals. As simple as that. You can of course extend Ohm's law to hold in semiconductors, by saying that 1/R is proportional to carrier density, and carrier density can be modified. But Ohm and his contemporaries did not know that you can modify carrier density, or even that there is such thing as carrier density. All they knew was that there was a dependency of R on material and temperature, and they wrote formulas based on that.

You are still making it too easy. First of all, what has to be established: Does the em-drive move even independent of the experimental set-up. If there are dependencies on the motion of the drive on where or when you activate it and how you install it in the test rig, you have to first check if these dependencies are not much simpler already known physics hiding somewhere in the experiment.

As you can see right now: Even small variations in the input conditions can have big impact on the measured thrust. Yes, that sounds like chaos theory.

Must not mean that the experimenters are guilty. Many things in physics are chaotic. But still, we manage to control them good enough to do something useful with them. But as long as the experiment setup is also a factor there, it has to be researched first before any theory on a em-drive without any contributing factor by the experiment setup is formulated.

Also Ohm did not define that the resistance has to be constant. He assumed it for his law to work, among many other things that had been great in the 18th century but terribly rare today. A slowly changing resistance does not really make his laws useless, but quickly changing resistances are. The dynamic behavior of a transistor is nothing for Ohm. But the static, ideally linearized behavior of a transistor in a circuit does again integrate well into Ohms laws.

 

[kamaz]

There is a strong, definite Poynting vector from the small base towards the big base, which means that the energy flux is from the small base towards the big base. This would mean, that in order to satisfy Conservation of Momentum, the copper cone needs to move in the direction towards the small base to balance the energy flow in the opposite direction. Alternatively, the Poynting vector field may all get dissipated into heat at the big base. But please note that Meep takes into account losses in detail in the copper model.

The Poynting vector seems to be strongly associated with the RF feed from the antenna.

The Big base is at the left and the Small Base is at the right for the xz and the xy plane views.

NEW INFORMATION: We show here that those (Greg Egan, etc.) that pontificate that the electromagnetic fields inside the EM Drive produce a Poynting vector that sums up to zero over integer periods of time are plain wrong. The reason is that the Poynting vector sums up to zero over integer periods of time only when the electromagnetic fields are standing waves (waves that do not travel in the longitudinal direction). The RF feed antenna disturbs what would otherwise be a standing wave frozen in space and results in waves that travel in the longitudinal direction back and forth and a time variation of the amplitude electromagnetic field that is not a simple sinuosoid, as long as the RF feed is on. This results in a non-zero Poynting vector with a net pointing from the small base to the big base over integer periods of time (probably due to geometric attenuation of the travelling waves due to the conical taper). During EM Drive experiments, the RF feed is on: it is only with the RF feed on that forces have been measured.
Notice that the period of this non-sinusoidal variation of the Poynting vector is half the period of the electromagnetic field (as expected from theoretical considerations).

http://forum.nasaspaceflight.com/index.php?topic=37642.msg1399795#msg1399795

Ha!

 

[RisingFury]

If these results were true, then the shape of the device wouldn't matter.

 

[Urwumpe]

It is funny that they use the law of Poynting to explain the production of force. Its a conservation law and basically says (in the lousy student version), that an electromagnetic field can only perform work, if it gets weaker.

 

[Andy44]

Shape matters; my home made perpetual motion machine didn't work until I reshaped the confribulation vector wave manifold into the likeness of Ozzy Osbourne's sunglasses with little lollipops on the edges. I would tell you exactly how it works, but I don't want to upset the world's economy, so for now it's just my little secret.

 

[Thunder Chicken]

We show here that those (Greg Egan, etc.) that pontificate that the electromagnetic fields inside the EM Drive produce a Poynting vector that sums up to zero over integer periods of time are plain wrong. The reason is that the Poynting vector sums up to zero over integer periods of time only when the electromagnetic fields are standing waves (waves that do not travel in the longitudinal direction). The RF feed antenna disturbs what would otherwise be a standing wave frozen in space and results in waves that travel in the longitudinal direction back and forth and a time variation of the amplitude electromagnetic field that is not a simple sinuosoid, as long as the RF feed is on. This results in a non-zero Poynting vector with a net pointing from the small base to the big base over integer periods of time (probably due to geometric attenuation of the travelling waves due to the conical taper). During EM Drive experiments, the RF feed is on: it is only with the RF feed on that forces have been measured. Notice that the period of this non-sinusoidal variation of the Poynting vector is half the period of the electromagnetic field (as expected from theoretical considerations).

Hey, wait a minute...this sounds familiar...

 

[RisingFury]

The Poynting vector shows the direction in which the electromagnetic energy spreads - that's true enough. There's also the relationship with electromagnetic energy and momentum.

Doesn't matter if you interpret light as a photon with set momentum or as a wave, both can carry it just fine.


The argument made by the model is that the Poynting vector decreases over distance in such a way that when it hits the other wall, there's less energy flow and less momentum flow. <--- Herein lies the problem. If the energy flow and momentum flow weakens, where does it go? If we don't want to throw the conservation of momentum out of the window, it has to be reabsorbed by the device or dumped overboard. Dumping overboard would just create a typical photon rocket.

 

[kamaz]

The argument made by the model is that the Poynting vector decreases over distance in such a way that when it hits the other wall, there's less energy flow and less momentum flow. <--- Herein lies the problem. If the energy flow and momentum flow weakens, where does it go? If we don't want to throw the conservation of momentum out of the window, it has to be reabsorbed by the device or dumped overboard. Dumping overboard would just create a typical photon rocket.

Precisely.

This demonstrates that there is a net energy flow from the small end towards the big end. Now, if the energy is flowing into the big end, it must be dissipated there somehow, either by localized heating or production of thrust by whatever mechanism.

Next, observe that radiation pressure of photons hitting a surface (think solar sail) is <S>/c, where <S> is average of Poynting vector over time, and c is the speed of light -- thus the force exerted on the sail with surface A is <S>*A/c. The simulation shows that <S> is zero on the wide end, while it is non-zero on the narrow end. This means that an (inward!) force is exerted on the narrow end, but not on the wide end. Thrust!

But there is still an (apparent?) violation of conservation of momentum. The force comes from the momentum of photons hitting the wall being transferred to the frustum: a photon flying towards the wall carrying the momentum of +p hits the frustum and transfers its momentum to it. So far so good; at this point conservation of momentum holds. But, conservation of momentum should also hold at the antenna, so the act of emission of a photon with momentum +p should have exerted momentum -p on the antenna. So the frustum should have received momentum -p at photon's emission and +p at absorption, resulting in the net momentum of zero.

 

[Urwumpe]

But there is still an (apparent?) violation of conservation of momentum.

Apparent very likely. Its a very tiny factor that is not accounted in the energy balance.