[Bobboau]

http://www.nasaspaceflight.com/2015/04/evaluating-nasas-futuristic-em-drive/

 

[Dragon]

It's been talked about for quite some time. Not only does it work in vacuum, it also produces noticeable results with Warp Field Interferometer - the gizmo dr. White uses for his Alcubierre Drive research. It's one of the more bizzare inventions I've ever seen, the potential implications are incredible. Extraordinary claims require extraordinary evidence, sure, but there you have it.

[AdmiralRalwood]

It's been talked about for quite some time. Not only does it work in vacuum, it also produces noticeable results with Warp Field Interferometer - the gizmo dr. White uses for his Alcubierre Drive research.

The article mentions both the EM Drive working in a vacuum and the Warp Field Interferometer detecting measurable, repeatable results; the two are not linked beyond both being researched by NASA Eagleworks. EDIT: I missed a vital sentence.

[Cyborg17]

This is really exciting!

[Mongoose]

And we still have no idea how the damn thing works.

[esarai]

And we still have no idea how the damn thing works.

Yep.  We may have accidentally proven the existence of quantum vacuum virtual plasma before we even figured out how to test if it exists.

[Dragon]

That's assuming this thing's operating principle isn't even more bizzare. From what I've gathered, we might have a sub-light warp drive on our hands. Although it producing force seems to imply something else, it does register quite strongly on White's Warp Field Interferometer. Or maybe "QV virtual plasma noise" and "random warping of ambient spacetime" are two ways of saying the same thing? Wouldn't be the first case of that in physics. Eh, I really of wish my institute had some sort of partnership with NASA. The new particle accelerator we have is nice, but what I wouldn't give to have a chance of working with White's WFI and the EM drive tests... There will be some really exciting things happening in there in the near future.

[Bobboau]

would be hilarious if we found out that supersaturating an area of space with photons had an effect indistinguishable to negative mass or something equally bizarre.

[watsisname]

If true, that could explain how core collapse of a star leads to supernova explosion.   But I am pretty sure it is not true.  If a few kilowatts (or whatever it is now) is making an experimentally observable effect through such a mechanism, then you'd expect pretty obvious decrease in the gravitational field of stars.

Relativistically, filling space with photons actually increases the gravitational field, because of their momentum.

[Bobboau]

yeah, nothing about this thing makes any form of sense. a few kilowatts of power shouldn't be able to do anything unexpected, we'd be seeing lots of other weird things in nature. maybe it could be an effect of very strong standing waves? those don't happen very often in nature, everything about this smells other than the fact that it keeps getting positive results in multiple experiments at multiple labs. the fact this happens to now have a connection with alcubirre drives is just way to ****ing convenient.

I guess we'll see what happens a few months from now.

[AdmiralRalwood]

If true, that could explain how core collapse of a star leads to supernova explosion.

Isn't that explained quite thoroughly by the energy output of the heavy metal fusion resulting from the gravitational collapse? I wasn't aware of there being an unexplained portion requiring an "anti-gravity" effect.

[watsisname]

Not completely thoroughly.  Ultimately, the energy source of the explosion is conversion of gravitational potential energy into kinetic, plus energy released from fusion.  But exactly how that energy is transferred into the outgoing shockwave is not a trivial question.

You might think that the radiation pressure from all the photons produced by the fusion would easily explain it, but it doesn't.  It's not nearly enough.  It turns out that neutrinos actually play a big role, which might be surprising since they are so weakly interacting.  But the core density during collapse is very high, and there are a lot of neutrinos produced.  As an XKCD What If shows, if you were close to a core collapse supernova, the neutrino radiation alone is enough to kill you!  So they must supply a fair bit of pressure to the surrounding layers of the star, but even this is still not enough to completely explain the shockwave.  This is an astrophysics problem which has been studied for the last several decades, and it is not yet completely solved.  Fairly regularly a new paper will show up on ApJ on the subject.


I'm pretty sure that whatever the missing details are, absolutely none of them involve antigravity.

[Herra Tohtori]

The whole thing is very, very exciting. Especially because we really have no idea what's going on with the thing, only that it seems to work!

It's like a prehistoric human figuring out that a hollow tree trunk floats on water much better than a raft made of fully intact logs. No idea how it works out, just that it does.


By the way, just to clear up a slightly ambiguous part in the article:

The concept of an EM Drive as put forth by SPR was that electromagnetic microwave cavities might provide for the direct conversion of electrical energy to thrust without the need to expel any propellant.

Any asymmetric light source produces thrust without "propellant" per se - the thrust of a monodirectional source of electromagnetic radiation is F = P/c where P is basically the sum of the radiation going in one direction, and c is of course the speed of light. *

The problem is that this ratio is, of course, quite small since the reversal of the equation tells that the power requirement for attaining a given thrust is basically the thrust multiplied by speed of light - so to get 1 Newton of thrust, you need about 300 MW light source and that's, quite frankly, a lot. So the basic ratio is 300 MW/N for anything that just uses electromagnetic radiation; this doesn't break the conservation of momentum.


On the other hand, the advertised power to thrust ratio for this new mystifying device is 2.5 kW for 720 mN of thrust, which comes out to 3472 W/N - which is about 86,405 times as effective, or in the order of five magnitudes better. Enough so to make it considerably interesting as a means of propulsion and *definitely* appears to go against the conservation of momentum as we know it.


In fact, I would feel tempted to hypothesize that the device is using ambient dark matter as its active propellant... if it didn't basically go against the whole concept of "dark matter". The problem is that dark matter is supposed to be impervious to electromagnetic interaction - there would have to be some unknown interaction going on to facilitate this. There are some problems with assuming that it's "pushing against quantum plasma" - mostly because it would require a substantial re-formulation of the conservation of momentum, and there's probably going to be some messy stuff happening to relativity as well. On the other hand perhaps quantum plasma *is* a form of dark matter...


The more tantalizing possibility is actual warping of space-time, causing an acceleration-like effect which we measure as a force.


Needless to say, either way it's a really, really exciting discovery. Especially as it seems to be increasingly unlikely to be a measurement anomaly or other problem with the experiment setup.



*Math to derive the thrust of a monodirectional light source with known power P:

Power is defined as

P = dE/dt

Force is defined as

F = dp/dt

A photon's energy and momentum are:

E = hf
p = hf/c

so photon's momentum can be written as

p = E/c

which means that

dp = dE/c

and

dp/dt = dE/dt*c

which is

F = P/c  <=> P = F*c

[Phantom Hoover]

Basically, a giant spotlight works exactly as you'd expect a rocket engine with an exhaust velocity equal to the speed of light to (using mass-energy equivalence on the power draw). The EmDrive has an effective exhaust velocity of eighty-six thousand times the speed of light.

Is it still likely that it's all just measurement error?

[Herra Tohtori]

It doesn't work quite like that because relativity. Assuming superluminal velocities never really works out.

The "giant spotlight rocket" uses photons as its "propellant", and while they carry momentum, they have no mass - all of their energy is in the form of electromagnetic radiation.

If you give the propellant some mass of its own, then the equation for the momentum of individual propellant particles changes. Especially if you're looking at relativistic ejection velocities, which by the way are not a practical concern at the moment.

Propellant with mass ejected at light speed would require infinite power - and would have infinite momentum as well. Doesn't happen.


Basically the more mass you move, the less ejection velocity you need. This means that the propellant flow is not as energetic. In many applications of reaction principle, the reaction mass is abundantly available, so it is more efficient to use more reaction mass (like air) which allows using less power. This is why, for example, high bypass turbojet engines are much more efficient (at subsonic speeds) than straight turbojet engines with no bypass air at all.


Spacecraft are in an unique position because they must transfer their reaction mass with them. That's why rocket engine always seek the highest ejection velocity possible - to gain the most energetic exhaust, while using as little propellant as possible.



The big thing with the EM drive is that if it turns out to be right, space travel may no longer be limited by the cold facts of the rocket equation. There may be useable propellant *everywhere* in the form of quantum plasma, as long as you can energize your drives!


And to answer your question, it's starting to look increasingly unlikely that it's just a fib or measurement error or anything of the kind. It's being reported in different, independent experiments, with consistent results.


Mad exciting.

[Phantom Hoover]

That's a nice wall of text but there's a reason I used words like 'equivalent' in my post. I'm comparing this to the familiar (to anyone who plays KSP) world of rocketry under Newtonian mechanics to give some sense of perspective.

[Herra Tohtori]

But it's specifically *not* equivalent.

Think of it this way.



As you start with massive propellant and slow ejection velocity, your rocket uses very little power (energy of exhaust) to produce a given thrust force.

As you reduce propellant particle mass and increase ejection velocity, your rocket's exhaust becomes more energetic, you need more power to produce the velocity, but your propellant flow reduces.

As you ramp up the propellant velocity and reduce the mass flow, your rocket's exhaust power increases, while the force of thrust remains equal.

As you approach relativistic speeds, your propellant flow will be minimal, but your energy demands will be very high.



Now as you move from propellant with mass to *photons*, you basically extend the curve to where ejection velocity is light speed and propellant mass flow is zero, because you of course need no propellant.

However, your thruster's energy consumption has basically reached the highest possible value per unit of force.


And that's where things stop. Exhaust velocity doesn't exceed light speed, if it did you'd get some fairly odd causal violations.


Aside from that, the analogy or "equivalency" you presented is simply incorrect, because ejection velocity is just one component of the impulse given to the exhaust - you need to know the mass as well.

All we can really say is that the thrust per electrical power ratio is about 86,000 times as much as expected from a "spotlight rocket". You can't draw any conclusions as to exhaust velocity (real or equivalent) without knowing the propellant mass flow (real or virtual).

[Phantom Hoover]

Jesus christ I know that it doesn't actually work that way, what I was saying was essentially that if you modded an engine into KSP with a 3*10^8m/s exhaust velocity it would behave the same way as a photon engine IRL. Honestly, I realise I didn't really make that clear in my original post but it feels like you're only skimming what I say so you can write lengthy essays on physics that I already understand.

[Herra Tohtori]

Yes, if you also modded the engine to only use electrical power as a resource, and if the vehicle had enough power to run the engine.

At any rate, since you didn't actually mention KSP at all, my understanding was that you were referring to real world and real physics with this:

Basically, a giant spotlight works exactly as you'd expect a rocket engine with an exhaust velocity equal to the speed of light to (using mass-energy equivalence on the power draw). The EmDrive has an effective exhaust velocity of eighty-six thousand times the speed of light.

...hence I responded with a real physics explanation on why it doesn't work like that.

[Phantom Hoover]

Honestly, I realise I didn't really make that clear in my original post

And no, if you use mass-energy equivalence to get a mass flow from that 300MW per Newton figure you gave, then calculate an exhaust velocity from that you find that, as I said initially, a photon drive works exactly as you'd expect a Newtonian reaction engine with an exhaust velocity of c to.