Em drives

[Urwumpe]

Yes, because physics as we now it today must be final.

No, mostly because if I have the choice to choose which fantasy I follow, I'll prefer unicorns. I could also have chosen dragons, but who would believe me if the experiment is not even a bit scorched afterwards?

Look, I have nothing against progress. Nothing that could stop it anyway. But progress is not done by fiction. Not by picking more fantastic hypotheses and more fantastic names.

How would such theories read, if you all replace by unicorns and fairies? In a scientific theory, the theory would not become contradictory and unbelieveable, if you replace all the concepts and terms. The words might be funny at first, but the logic and structure behind the theory would still exist.

But what about these? Ever bothered to read what kind of nonsense Unruh radiation actually is? Its absolutely Harry Potter business. Not proven by any experiment. Not even suggested. Not even compatible to previous theories. Compared to it, HQT is a well-based and common theory.

 

[RisingFury]

Alternative theory of operation: http://physicsfromtheedge.blogspot.com/2015/05/emdrive-whence-motion.html

He has to invoke acceleration to make the prediction. Only acceleration will lead him to thrust. That alone is problematic...

He also invokes Unruh effect ( [ame="http://en.wikipedia.org/wiki/Unruh_effect"]Unruh effect - Wikipedia, the free encyclopedia[/ame] ) which is TINY.

Nothing as easily measurable as this.

 

[Urwumpe]

He also invokes Unruh effect ( Unruh effect - Wikipedia, the free encyclopedia ) which is TINY.

Not just Unruh effect - Unruh radiation. Both are related, but only the effect has a proper theoretical base (Based on the Rindler metric). The radiation is mostly a "There oughta be something like Hawking radiation, because we have a virtual event horizon"

Both require pretty strong accelerations BTW. For experiencing a blackbody radiation by a blackbody of just 1 K temperature according to the Unruh effect, you need 2.5E19 g acceleration (Or in full digits: 25,000,000,000,000,000,000 g).

 

[kamaz]

Here is his paper on emdrives: http://www.ptep-online.com/index_files/2015/PP-40-15.PDF

This may be crazy, but this isn't stupid. First he uses modified inertial mass from his own version of MOND ( http://arxiv.org/pdf/1207.7007v1.pdf ):

Theta is radius of the event horizon, which, in case of MOND, is Hubble radius, so the correction is indeed tiny (although observable on galactic scales).

The big idea: since the photons inside the cavity cannot interact with the world outside, that means that the walls form an event horizon for the photons inside. Since the radius of the event horizon is in the denominator, then the smaller the cavity is, the more Unruh radiation you have. The other important idea is that since the Unruh force is generated with each trip through the cavity, so the more times photons bounce around (i.e. the higher the Q-factor is), the more force will be produced. In other world, he basically says that Shawyer has accidentally constructed an Unruh force multiplier. So he ultimately arrives at:

And comparison with measurements:

Given that the error bars are huge, he is uncomfortably close for my taste.

Hmmmmmmmmmmm.

His theory gives two falsifiable predictions:

- thrust increases with increasing cavity length (how is that 24GHz version coming along)?

- thrust increases with Q-factor, and this is also easy to verify by putting a dielectric inside the cavity

 

[Urwumpe]

And comparison with measurements:

Given that the error bars are huge, he is uncomfortably close for my taste.

I would not call that close - there is no system behind the errors that suggest that the theory could converge on the measurements eventually.

Also, the basic assumption that the device forms an event horizon for the photons is nice, but wrong. Its no theoretical perfect reflector or absorber, you still have a pretty huge amount of transmission.

 

[jedidia]

The big idea: since the photons inside the cavity cannot interact with the world outside, that means that the walls form an event horizon for the photons inside.

Didn't NASA's first test include a test with a device without cavities that against expectations also produced a force? I kinda remember something like that, but I'm not sure anymore.

 

[Urwumpe]

Didn't NASA's first test include a test with a device without cavities that against expectations also produced a force? I kinda remember something like that, but I'm not sure anymore.

Not without cavities... I remember that the original design included some cut-outs or holes in the design, that had been declared essential for the thruster to produce thrust. Worked better without. :lol:

 

[kamaz]

Yep, someone thought that this is a photon rocket (photons escaping through the holes in the frustum) so he made test articles with holes and without holes expecting that the one without holes will not produce thrust... it did.

---------- Post added at 11:46 PM ---------- Previous post was at 11:37 PM ----------

I would not call that close - there is no system behind the errors that suggest that the theory could converge on the measurements eventually.

The guy hits within one order of magnitude using an astrophysical model and you are complaining

 

[Urwumpe]

The guy hits within one order of magnitude using an astrophysical model and you are complaining

Would he have used a theological model, would this have been better?

 

[kamaz]

Also, the basic assumption that the device forms an event horizon for the photons is nice, but wrong. Its no theoretical perfect reflector or absorber, you still have a pretty huge amount of transmission.

But the RF leakage is not enough to reconstruct the state of photons inside the cavity, so his assumption holds. ...There was a paper in 2012 which demonstrated that gravity can be viewed as an emergent force caused by information loss (cloaking): http://arxiv.org/pdf/1001.0785v1.pdf

---------- Post added at 12:51 AM ---------- Previous post was at 12:15 AM ----------

Would he have used a theological model, would this have been better?

You mean Lambda-CDM, which requires unobservable mass and violation of conservation laws on cosmological scale? :lol:

 

[Face]

Yes, because physics as we now it today must be final.

Indeed. Unicorn physics - color doesn't matter - must be final. Also in this thread, I'd say.

That doesn't really propagate to other fields of physics, though. Not even with warp speed :lol:.

 

[meson800]

Given that the error bars are huge, he is uncomfortably close for my taste.

What error bars?

It is unclear what the error bars on the observations are,
but they are likely to be wide

They didn't specify any, except that they should be "wide".

Can't the error bars be directly calculated using error propagation? It's easy to take the partial derivatives of the force equation. The sigma values for P,Q,l, W_big, and W_small should be relatively easy to find. The covariance values would be harder to find, but I"m not even sure they matter (power is uncorrelated to the physical dimensions, physical dimensions are uncorrelated from each other, etc).

 

[kamaz]

What error bars?

Errors in force measurement. As you can see on the previous page, thermal expansion of air inside the cavity can generate a force comparable to the claimed effect.

They didn't specify any, except that they should be "wide".

Yes, sloppy writing. It's a theoretical paper though, so you can go through the experimental papers he is taking the values from.

Can't the error bars be directly calculated using error propagation? It's easy to take the partial derivatives of the force equation. The sigma values for P,Q,l, W_big, and W_small should be relatively easy to find.

That gives you error bars for the theoretically calculated value, not for the measured value. Since everything which goes into his calculation is know with decent precision (say 1%) then you can as well say that his prediction is precise.

The covariance values would be harder to find, but I"m not even sure they matter (power is uncorrelated to the physical dimensions, physical dimensions are uncorrelated from each other, etc).

You are overcomplicating it. If the model is true and there are no systematic errors then (F_pred - F_meas)/F_pred follows a normal distribution centered on zero. A metaanalysis of various measurements should also yield a normal distribution centered on zero. Now look:

This is what I call "uncomfortably close".

---------- Post added at 04:45 PM ---------- Previous post was at 04:34 PM ----------

Interesting meta-commentary over at NSF referring to this piece:

 

[Urwumpe]

You mean Lambda-CDM, which requires unobservable mass and violation of conservation laws on cosmological scale? :lol:

Thats wrong - it requires invisible weakly interacting mass (if it would be unobservable, it wouldn't matter to the universe anyway).

Same with violation of conservation laws - it isn't, without the dark energy though, we would be observing a violation of the conservation laws right now.

 

[kamaz]

Thats wrong - it requires invisible weakly interacting mass (if it would be unobservable, it wouldn't matter to the universe anyway).

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

The basic fact is that physics "as-is" cannot explain galaxy rotation. Explaining it requires either introducing invisible mass or modified inertia. The mainstreams keeps looking for the invisible mass and finding nulls. Meanwhile, other anomalous observations keep piling up: Pioneer anomaly, flyby anomaly, Tajmar, Podkletnov, variations in the value of G, now em-drive -- and one thing they all have in common is claimed modification of inertia.

 

[Urwumpe]

Meanwhile, other anomalous observations keep piling up: Pioneer anomaly, flyby anomaly, Tajmar, Podkletnov, variations in the value of G

Are you living in a shed somewhere in the Carpatian mountains, far away from newspapers?

Those have all found boring standard physics explanations, some many months ago for the latest. Maybe they are not that exciting as hoping for new physics. But if the old physics already explain the observations without contradictions, why should there new physics be needed?

[ame="http://en.wikipedia.org/wiki/Pioneer_anomaly"]Pioneer anomaly - Wikipedia, the free encyclopedia[/ame]

(Same with all others. All fantastic flyby anomaly hypotheses for example need to explain why Rosetta did not experience this anomaly)

 

[kamaz]

Who needs heliocentrism when epicycles explain observations perfectly?

Let me ask you one thing: how do you KNOW that gravitational mass is equal to inertial mass?

 

[Urwumpe]

Who needs heliocentrism when epicycles explain observations perfectly?

Because they simply didn't do so perfectly. You need a long Fourier series of epicycles to just approximate a Kepler orbit. And much more epicycles to include the measured perturbances to Kepler orbits in your geocentric epicycle model.

Of course you can do so. Its not impossible and not per se unscientifically. It is just the most complicated and error-prone method to represent astrodynamics.

Let me ask you one thing: how do you KNOW that gravitational mass is equal to inertial mass?

I don't know it. But all measurements end up in equality. If they are not equal, the difference is so small that we fail to detect it today.

But that does not mean, you are free to speculate that there is an invisible unicorn effect that causes small variations of the masses and claim that thats a valid scientific theory. Its mere science fiction.

 

[kamaz]

Excellent! So what exactly makes you think that CDM is preferred to MOND if CDM fails to predict galaxy formation and requires elusive WIMP particles, while MOND models galaxy behavior correctly and the correction to inertial mass required by MOND is below what we can measure?

 

[Urwumpe]

Excellent! So what exactly makes you think that CDM is preferred to MOND if CDM fails to predict galaxy formation and requires elusive WIMP particles, while MOND models galaxy behavior correctly and the correction to inertial mass required by MOND is below what we can measure?

Stop - MOND* does not model galaxy behavior correctly. Thats exactly the fallacy. MOND is "We have no clue, but we just fit the rotation speeds of stars in a galaxy to a function."

The problem is: MOND works for spiral galaxies. MOND does not handle galaxy evolution well, elliptic galaxies are extremely annoying. Motions of galaxy clusters are violating MOND, so some MOND proponents invented a new function that they can fit over the observation data (Others introduced dark matter for fixing MOND in that case, oh the irony). Still lacking a prediction, any newly measured galaxy in the universe can be enough to break MOND. Especially noteworthy is, that the [ame="http://en.wikipedia.org/wiki/Bullet_Cluster"]bullet cluster [/ame]invalidates all current MOND variants, while it confirms the Lambda-CDM model with its observations.

Now some question for you: If you can't see a mass, but can sense its gravity, does this mass exist or not?



* Modified Newtonian Dynamics