Slashdot Mirror
Final NASA Eagleworks Paper Confirms Promising EM Drive Results (hacked.com)
An anonymous reader quotes a report from Hacked: Earlier this month Hacked reported that a draft version of the much expected EmDrive paper by the NASA Eagleworks team, had been leaked. Now, the final version of the paper has been published. The NASA Eagleworks paper, titled "Measurement of Impulsive Thrust from a Closed Radio-Frequency Cavity in Vacuum," has been published online as an open access "article in advance" in the American Institute of Aeronautics and Astronautics (AIAA)'s Journal of Propulsion and Power, a prestigious peer-reviewed journal. The paper will appear in the December print issue of the journal. The final version of the paper is very similar to the leaked draft. In particular, the NASA scientists confirm the promising experimental results: "Thrust data from forward, reverse, and null suggested that the system was consistently performing at 1.2 +/- 0.1 mNkW, which was very close to the average impulsive performance measured in air. A number of error sources were considered and discussed." The scientists add that, though the test campaign was not focused on optimizing performance and was more an exercise in existence proof, it is still useful to put the observed thrust-to-power figure of 1.2 mN/kW in context. "[For] missions with very large delta-v requirements, having a propellant consumption rate of zero could offset the higher power requirements. The 1.2 mN/kW performance parameter is over two orders of magnitude higher than other forms of 'zero propellant' propulsion, such as light sails, laser propulsion, and photon rockets having thrust-to-power levels in the 3.33--6.67 uN/kW (or 0.0033--0.0067 mN/kW) range."
In other words, a modest thrust without having to carry fuel can be better, especially for long-distance space missions, than a higher thrust at the cost of having to carry bulky and heavy propellant reserves, and the EmDrive performs much better than the other "zero propellant" propulsion systems studied to date.
They say that they have looked at outgassing, and assumed that its not relevant due to slow temp rise not producing rising force. But that does not cover possibility that the electromagnetic resonances are somehow vaporising and ejecting structure at much higher speeds. At .0012N thrust with 1kW input (and 100% efficiency) a rocket would need exhaust velocity of 1.6e6 m/s and consume around 0.8ng per second - damned difficult to weight with required sensitivity and hard to spot except by looking for evidence in the gases within the chamber as metals will condense out quickly.
... are more or less the same ones than before (= very unclear setup, situation very unlikely to represent the claimed break of the conservation laws, highly restricted conditions not telling much, etc.).
Summary of my impressions after quickly reading this paper:
- The actual methodology generating the thrust isn't clearly explained, 95% of this paper is about the testing conditions (measurements, sources of error, assumptions, etc.). Although I assume that detailed explanations on this front might drive to a level of clarity similar to the one of the tests, as explained in the next point.
- Complex testing setup which is very difficult to be adequately understood from outside. It seems that only people with actual experience under these specific conditions (and, ideally, with physical access to an equivalent setup) are in a position to critically analyse these tests and be specific about the (very likely IMO) source(s) of error.
- Even by ignoring the two aforementioned points, plainly believing that everything is fine and just analysing the results, there are various issues which are somehow against the reliability of this experiment and related out-of-proportion assumptions. Examples in fig. 9: a maximum displacement below 0.005 micrometres (extrapolating such a top performance to interstellar travels is sensible?!); assuming that the error in the measurements remain constant under different conditions (?!); testing just 3 different scenarios (40, 60, 80 W) and getting counter-intuitive results (30/40 = 0.75; 106/60 = 1.76; 76/80 = 0.95; 60 W delivering the best performance?!).
Fig. 19 is even more descriptive by showing a tremendous variability of the measurements; in the best-performing 60 W scenario, they vary from 130 to 45 micronewtons!! With only a few cases being similar enough (85 and 92); out of all the about 20 cases, there are only a few which are identical under the given conditions.
Custom Solvers 2.0 = Alvaro Carballo Garcia = varocarbas.
IAATP working on quantum electrodynamics (QED) and other theories.
The fundamental problem with this experiment is that it appears to violate conservation of momentum. This violation is not something that can be discarded easily: it has been confirmed directly and indirectly in millions of experiments over decades.
Momentum conservation is also a cornerstone of quantum field theory (QFT) and it is a symmetry which survives quantization. The entire Standard Model (SM) is a momentum-conserving QFT. The SM has been confirmed to a high accuracy in particle accelerators for many years. Any violation of momentum conservation would have been quickly noticed. You cannot simply invoke 'quantum mechanics', 'zero point', 'vacuum fluctuations', etc. to explain excess thrust. Momentum conservation is fundamental, both classically and quantum mechanically.
So what about the EM drive results? There is a possibility that some new physics is at play, however it is vastly more likely that there is a systematic error which has not been eliminated. (If I had to guess I would imagine that because a large amount of RF energy is being pumped into large metal cavities, the apparatus is resting at the bottom of a standing wave potential.)
The way to finally confirm or refute this is to take the drive into space. In this case, it is almost certain that the net thrust would be equal to the momentum of the photon flux leaving the drive.
Apart from the open process and independently verified results
This is the problem though the results are not at all verified. Have you actually read the paper? It shows an appallingly low level of scientific methodology for a paper claiming to observe a phenomenon which violates the currently know fundamental laws of physics. For example at one point it is quoting a fit to 7 significant figures without giving any uncertainty range which suggests a position accuracy of ~1nm which is less than the size of an atom. I am unconvinced that they measure the position this accurately. While this ultimately will not affect the result they claim it shows sloppy practice which is not a good for inspiring confidence.
However most importantly when considering errors at no point do they see to consider charged particle emission as a source of thrust. They do worry about the components becoming charged which they say they fix by grounding but if you are emitting electrons grounding the engine just ensures that there is no charge build up which will allow the engine to continue to operate. Since you cannot ground a craft in space the charge would build up their until the engine's thrust stops.
So it's great that they publish their results openly but what there is to see there in no way inspires any confidence that they have observed some new, fundamental physics phenomenon. Instead of engineers they need to get some scientists involved because the paper shows a total focus on simply measuring the thrust and zero scientific investigation to investigate the cause of the thrust.
Umm, you realize we needed to make nuclear power plants *before* we could make the bombs, right? Granted they were designed specifically to enrich the natural fissiles into weapons-grade isotopes, but the reactors still came first.
And no, looking at the current state of a technology based on physics that we don't yet understand in no way helps us understand how much improvement that technology may undergo. At best it gives us a glimpse at what it might enable.
In that context though, the best initial application is likely to be deep space probes. Satellite maintenance might get included as well, but that's more a convenience than an enabling technology - there's already orbital refueling vehicles under development. Deep space though - that's where constant low thrust acceleration pays off big time.
From what I can find, current RTGs for space applications top out at about 5W/kg. Assuming a 1kW RTG is half the total mass of the probe, that gives us a 400kg probe with 1.2mN of thrust using the current unoptimized EM drive tested. That translates to 3um/s^2 of constant acceleration. You're not going see much change right away. Starting from rest, in one minute it will travel a grand total of 5.4mm. But acceleration adds up:
Displacement as a function of time:
1day: 11km. 1 week:550km 1month:10,000km 1 year:1.5Mkm (yay, 0.5% of the way to Mars!) 10 years: 150Mkm (Wait, we're still not to Mars?). 100 years: 1.5Bkm (3x the distance to Pluto). 5200 years: ~4.2 light years(we've reached Proxima Centauri!)
So yeah, with current technology it's not actually much good for deep space probes, and I haven't even factored in the losses of climbing out of the sun's gravitational well. If you're operating close enough to the sun to use solar though you can up your power to 300W/kg (near Earth orbit), and assuming the same 50/50 power to payload ratio that will get you ~60x the thrust (and thus 60x the distance per unit time). Then the numbers look a bit better: You might get most of the way to Mars in a single year for example. And more importantly be able to turn around and repeat the journey indefinitely.
So I suppose inner-system scouting probes and perhaps interplanetary cargo transportation could be early applications. And if optimizations could yield a 10-fold improvement in engine thrust/W, well then things start getting really interesting. Travel to and from Mars in a month, with no need for refueling? That's the stuff science fiction is made of.
--- Most topics have many sides worth arguing, allow me to take one opposite you.
The paper does not state that the power to thrust ratio is constant at all speeds but simply that in the tests that were performed changing the power level changed the thrust in a linear manner.
That would be easy to test: the earth is moving continuously and changing direction all the time in a predictable manner.
What if the cavity simply exhibits a force on dark matter?
It's not dark matter if it interacts with the electromagnetic field, it's something else literally by definition.
SJW n. One who posts facts.
False.
Trinity, the very first man-made nuclear explosion, used Plutonium from the Hanford Engineering Works, created in the B, D, and F reactors. You do know that Plutonium doesn't exist in nature, right? It's either created in a reactor via neutron bombardment of U238, or in a cyclotron.
More than that, the Chicago Pile was the first man-made self-sustaining nuclear reaction (in 1941), and the basis of all reactor design that followed to support the Manhattan Project, which made bombs detonated in 1945.
Reactors very much came before the bombs.
Slashdot still doesnâ(TM)t support Unicode after it was added to the HTML standard in 1997.