"Again it showed the same behavior - until the drive was directed towards the top for testing purposes. Usually, the drive is put on a torsion balance in a way that it can best turn the scale left or right.
With the drive directed towards the top nothing should be measurable. [...] However, the drive directed toward the top in this experiment still showed the same force as when directed to the left."
The test result at TU Dresden, Germany: "Tjalmar said: 'I can't say I proved or refuted the EM drive'" (because they didn't have the means to test it with a battery instead of an external power supply)
Reference to a follow-up from a Chinese university:
"But a research group from China did the necessary to refute it. As soon as the EM drive doesn't get its power from a cable outside [i.e., it get's its power from a battery], there is no measurable force anymore."
The final sentence of the abstract:
"Within the measuring range of
three-wire torsion pendulum thrust measurement system，the independent microwave thruster propulsion device
did not detect significant thrust. Measurement results fluctuate within ± 0.7mN range under the conditions 230W
microwave power output，and the relative uncertainty is greater than 80%"
That's basically the thinking behind the laser-propulsion plan. The rocket equation gets a lot less hostile when the engine is not mounted on the rocket, but that has its own practical issues. Less than trying to drag a literal cord, of course.
The Austrian experimental physicist https://en.wikipedia.org/wiki/Martin_Tajmar is a legend already, since he measured in 2002 a gravitomagnetic effect in spinning superconductors at Seibersdorf, but this sensational effect is also probably just bad experimental setup and not a new gravitational theory.
He is the author of the classic book "Advanced Space Propulsion Systems".
I guess a cheap and efficient propulsion that doesn't fit our understanding of physics was too good to be true. Still, it's a bit sad, that would have been beautiful if it were true.
As far as I understand it, it's not quite over yet. Maybe the refusal itself can't be replicated by the other groups. Now that will keep these physicists busy for a while...
That was my thought as well. I'm glad someone spends the time required to refute a claim like this, because it did seem more long-lived and potentially within the realm of possibility than your run-of-the-mill claim that violates known physics.
If we never take the time to explore claims that violate our current understanding, our understanding will stop advancing. Too many scientists are overly dogmatic.
"The theoretical expectation is that you don't measure anything. Now you're measuring something, and you can learn extremely much on this hunt for the error."
I remember that quote too, and the key take aways always seemed like: in science you always learn something.
Taking the EMDrive as an example -- we have measurements that contradict current theory, so we are either going to discover new physics or we are going to learn something about the art of making very precise measurements of force in complex mechanical systems. In no case do we learn nothing.
A third and equally useful possibility is that we might learn to correctly and productively apply the current theory in a situation and configuration where previously we did not know how to do so.
If it's free energy or free momentum, using only physics that has been perfectly described in hundreds of millions of contexts and experiments, for one and a half centuries, using mathematics that don't allow for either, you can rule out this alternative.
The likelihood that free momentum and free energy are possible is indistinguishable from 0. They are empirically impossible in the strongest reasonable sense the word can have.
The critical detail here is that this is neither free energy nor free momentum in this setup. Energy is pumped in, and some momentum is generated - that's incredibly boring and not at all something that should raise any skepticism. The question of what in the hell was being pushed against to convert that energy to momentum, on the other hand, is a really interesting question, but not one to deeply pursue until we're actually sure it's happening. Which seems like the cavity wasn't the interesting bit in the end. A shame and a bit boring after so much careful testing, but nothing to get worked up over.
And the experiments didn't make it obvious that the setup didn't work. I mean, I don't think anyone took the inventor's theoretical claims seriously, but the device did generate some small thrust in a difficult to explain way.
Besides, we know that space isn't nothing, so being able to tractor against whatever is there would be a rather intersting engineering find; physicists would then likely take the experiment and go "oh, there's a neat edge case in this system of equations that explains it". There are a lot of interesting edge cases left to find and engineer.
Free momentum WAS claimed. Hence the space engine applications talked about everywhere.
Momentum is conserved independently, it can not be created from energy without spilling some the other way (into whatever you're pushing against).
The vacuum is a translation invariant state so you can't push against it to get momentum either, but you can create particles out of it and then push against those. That's what electro magnetic thrusters do. They generate photons. The effect is just super tiny.
Sure, something that worked like a Bussard ramjet, but without necessarily funneling all the particles through the craft itself, would still be useful. Grab all the hydrogen (and any bonus dust) within a few kiloclicks and throw it backwards at relativistic speed.
I'm not quite sure what you're suggesting. The universe and the science that describe it are the same thing. Much as I am the same thing as the chemicals and reactions that describe me. However, our understanding of the science may be flawed, and perhaps that's what you're really meaning to say.
A roller coaster is a physical object. A CAD drawing of it is a model. I could make a roller coaster from the model, but the model is not the roller coaster.
On the other hand, the universe is nothing more than the set of constraints that we call "science". The universe is the model, and science is an effort to duplicate that model.
If it was to me, I don't agree. Science is a collection of human ideas that exist in the minds of humans. As another commenter said, it's our model of the universe. There is no difference between science and our understanding of science unless you're talking about differences in understanding between individual people.
It was very much meant for you. I'll just copy from my other response:
> On the other hand, the universe is nothing more than the set of constraints that we call "science". The universe is the model, and science is an effort to duplicate that model.
No I'm not. I'm saying that we know something about how the universe works. And one of the things we know is that there is no free momentum or energy (in flat space time).
The reason that we know this is far stronger than simply: "It hasn't been observed yet". That would allow that any new thing could end up violating the established rules.
It could be that if you write the one hundred millionth Mersenne prime on a cookie, and the one two hundred millionth on a banana they start circling each other and create free energy. This claim has not been falsified. So on what grounds do we rule it out? Because it is incompatible at an incredibly deep level with everything we know about cookies, bananas, and how their physical properties depend on what's written on them.
Every aspect of each of these objects that has ever been observed has been found to be describable in terms of some unbelievably simple mathematical structures. For these unbelievably simple mathematical structures we have prove that anything that can be described by them does not allow free energy.
An observation like the Mersenne, cookie, banana interaction can not violate the mathematical derivation, therefore it implies that out of all the infinitude of observations that have been made of bananas and cookies, that all were describable by energy and momentum conserving physical theories, somehow this one arbitrary one is special. It can not be described by even the type of mathematics we have hitherto used and yet was completely invisible.
How could that be? Flying Spaghetti Monster of course. Just as the FSM hid the true age of the earth from us by interfering with science, so it also conspired to hide the free energy Mersenne Banana Cookie interaction from us.
In other words: If you know that the FSM did not create the earth 4000 years ago, then you even more definitely know that there is no free momentum/energy. If you think that we can't know either of these things, then your concept of knowledge sucks.
I'm not suggesting that anyone spend much time entertaining ideas like the prime cookie or the Flying Spaghetti Monster. I'm merely pointing out that theory does not dictate outcomes. I'm happy that they have disproven the EM drive (or that this appears to be the case), but one of the things that's bothered me about the discourse surrounding this phenomenon has been the notion that we should immediately dismiss it as "impossible". Surely, it was improbable. But that's just what it was: improbable.
I disagree. Calling it improbable rather than impossible muddles language. It's a fake sort of precision that removes an important qualitative distinction, and that is what bothered me the most about this discourse. A lot of naive language prescriptivist that want to discard perfectly good words for the sake of an irrelevant logical/philosophical point.
It's improbable to flip a coin 10 times and get head ten times. It's for all practical purposes impossible to do so 10^10 times. This level of mathematical probability is functionally indistinguishable from 0 thus it is entirely accurate to call it impossible.
It's not logically impossible of course, it just happens not to be possible in our universe. It's physically impossible. Like the prime cookie. The appropriate amount of time spent on investigating prime cookies is not "very small", it's 0.
Now there is of course reason to investigate the EM Drive to some degree: To figure out where the bug was. To learn about potential experimental errors. As a puzzle, to solve the apparent paradox. But any other framing of the issue is dishonest.
EM drive is not an argument such that it might be refuted. It is a device. As such it has fitness for purpose, which depends entirely on the purpose at hand.
If a device doesn't work for the purpose for which you built it, then you built it wrong. I can't help with that. What I can potentially help with is finding the purposes for which the EM drive mechanism can be useful, but I'm not there yet: The phenomenon is not yet understood well enough to determine those purposes.
The OP may not be a native English speaker (located in Europe), so any perceived snarkiness is likely based on an overly literal reading of his word choice.
How about this alternative (which, btw, is how I interpreted the headline): "Principle behind EM Drive refuted" or "Experimental results of EM Drive refuted".
The quote wasn't meant to be snarky. It was meant as an explanation why it is worthwhile to occasionally do experiments, even if you believe the device can't work the way it is supposed to. He also said it was excellent experiment for the purpose of teaching his students.
Judging from the date in the URL of the linked paper [1] and the corresponding section in the Wikipedia article on the EMdrive suggests that you might have better luck looking at somewhat older news articles.
Publication date (and PDF metadata) seems to be 2015 though [2].
That's sort of an unnecessary dichotomy. Enthusiasm doesn't mean hoax, and the mistake was propagated by more than just the inventors.
Everything suggests that it was just science in action. There was an experiment, and it failed to reject the null hypothesis of 'no thrust'. A few groups have repeated the experiment, but not to the experimental power needed to thoroughly reject it. This test seems to be a more definitive rejection of it, but we're still not sure why there was any thrust measured in the first place, though there are some pretty reasonable leads on that.
Those sort of things are always a bit grey and (IMHO) a bit sketchy, but it's worth noting that it's a bit costly to even test the ideas. Starting a business is a decent way to raise the capital to run the experiment, since grant money is likely far harder to get. But, yeah, doing it that way also really makes it basically impossible to be unbiased.
Let a few more rounds of experimentation and refutation go through, and if it's clear it can't work at all, and they stay in business without pivoting, then we can breeze right past hoax and jump to fraud. Time will tell about it, I guess.
Holy mackeral, that is cool. I really appreciate the "well, let's just try it" approach. I can't wait to see what happens when he launches his little satellite. If it shows positive results, lots of people will replicate it.
This is win-win: if it works awesome happens. If it fails then you're left with the experience of having built a little satellite!
Why not run the setup without the cavity, or replace the magnetron with a big ass resistor, and see if you still get deflection. I suspect this is the beginning of a refutation, but it feels like there's a carrot still dangling.
Come on folks. John Baez has dismissed it [1]. For those who don't know who John Baez is, it's as if Terry Tao dismissed your math homework. For those who don't know who Terry Tao is... whatever. Just stop.
Have you read Baez's post that I linked? It's way more harsh than my comment. "Mumbo jumbo", "Failed a course in quantum field theory and then smoked too much weed"...
Also I'm getting seriously creeped out by the amount of EmDrive non-skeptics on HN. Maybe we should make a poll whether 0.999... = 1, and then say that anyone who insists on the correct answer is "condescending".
Ad hominems and appeals to authority are such letdowns.
Almost as those who might even admit they know nothing about it, but feel they're qualified to judge validity of some scientific claim. And provide <insert subculture> mantra as "evidence".
I do know quite a bit about physics. But if I just said the EmDrive is nonsense due to conservation of momentum (which is the 100% correct answer with no caveats), people would shut me down because "NASA scientists know what they're doing". That would be an appeal to authority. So I mention Baez preemptively, but it seems like that's not a winning tactic either. What would convince you, pray tell?
> you'd shut me down because "NASA scientists know what they're doing"
I've said nothing like that.
A working experiment that provides irrefutable data is needed to convince me in any way. Refutation I'd need to leave for others to do, not my competence.
I really like EmDrive as an idea and hope it will work out. It's very inspiring when someone has the guts to push it despite the dominant (scientific) opinion. If it does work out, these people will be some of our greatest heroes.
After all, sometimes people like these have been right in the past in adverse scientific atmosphere. I'm just content to follow it from the sidelines.
If I had to bet on EmDrive, I'd have to bet against it.
The real question is how much you should weigh a single experiment vs. the abstract idea of conservation of momentum. It's tempting to trust the experiment all the way, and imagine that science demands such open-mindedness. But the more physics you study, the more likely you are to trust the abstract idea instead, because you realize just how incredibly well-supported it is. It's a kind of certainty that has almost no analogue in other sciences. Even in a very careful experiment, measurement error is many times more likely than overturning conservation of momentum. (Also measurement error is much more devious than people imagine, but that's another story.)
I'm very confident ($100 to $1 at least) that the EmDrive is a story of measurement error, compounded by ignorance and optimism.
On a philosophical point of view, it worries me that physics might be optimizing towards a local maxima. Which might or might not be the global one. That there even might be something big it has missed. How would we know?
Even more so because physicists seem to be more and more specialized to a narrow sub-field. Everything is already so advanced it takes more and more years to just to master it. Individual contributions appear to be getting smaller and smaller.
Is it possible there's something they don't see because they're too near looking at it? Studying a mountain with a microscope. Is possible we're stuck in a some sort of dead end?
I really can't judge it either way. I just can't stop wondering about that.
Yes, that's exactly it. In science, rationalism (Descartes, Leibniz) has been much more fruitful than empiricism (Locke, Hobbes, Hume...) It's really strange when you think about it. The guys who trusted in divine order ended up discovering tons of things, while Francis Bacon, the father of empiricism and the modern "scientific method", didn't discover anything. That's a big part of the reason why I identify as a rationalist.
By one experiment, you mean one actual experiment, made once, by a single group? Or one single experimental setup, replicated by different groups?
If it's the former, I don't trust it at all, but if it's well practiced and interesting, it ought to be replicated. If it's the second, well, if its results are against theoretical models, theory is wrong. No questions, no exceptions. Now, theory correction does not automatically mean a fundamental change, "hey, this approximation we used has a problem here" is as valid a theoretical correction as "ops, all of our models are wrong, everything is much more complex than we thought".
Specifically about the EmDrive, I'm not aware of anybody that could actually measure an effect yet. So, there's no experiment showing theory is wrong. There's only a lack of experiments confirming it either (or better, now there is one).
> I do know quite a bit about physics. But if I just said the EmDrive is nonsense due to conservation of momentum (which is the 100% correct answer with no caveats), people would shut me down because "NASA scientists know what they're doing".
I wouldn't. However, I would ask why you think violation of conservation of momentum is completely dispositive, rather than just a strong reason for great skepticism?
From Noether's Theorem [1] we know that conservation of momentum is a consequence of physics being invariant under space translation. As far as I've been able to find, we don't know that a universe that results from a big bang must be isotropic. Observationally, ours appears to be, but there are some hints that it might not be, and scientists are taking those seriously.
You were being condescending towards people who don't know about Baez and Tao. Bring critical of crackpot theories is warranted; the flimsier the idea the more pointed the criticism. He isn't nasty though.
87 comments
[ 2.8 ms ] story [ 78.2 ms ] threadThe critical passages - translated by myself
Page 2:
The hint:
"Again it showed the same behavior - until the drive was directed towards the top for testing purposes. Usually, the drive is put on a torsion balance in a way that it can best turn the scale left or right. With the drive directed towards the top nothing should be measurable. [...] However, the drive directed toward the top in this experiment still showed the same force as when directed to the left."
The test result at TU Dresden, Germany: "Tjalmar said: 'I can't say I proved or refuted the EM drive'" (because they didn't have the means to test it with a battery instead of an external power supply)
Reference to a follow-up from a Chinese university: "But a research group from China did the necessary to refute it. As soon as the EM drive doesn't get its power from a cable outside [i.e., it get's its power from a battery], there is no measurable force anymore."
The link to the Chinese paper http://www.tjjs.casic.cn/ch/reader/create_pdf.aspx?file_no=2...
The final sentence of the abstract: "Within the measuring range of three-wire torsion pendulum thrust measurement system，the independent microwave thruster propulsion device did not detect significant thrust. Measurement results fluctuate within ± 0.7mN range under the conditions 230W microwave power output，and the relative uncertainty is greater than 80%"
Unlimited powahhh!
Space travel. Shit's hard.
The Austrian experimental physicist https://en.wikipedia.org/wiki/Martin_Tajmar is a legend already, since he measured in 2002 a gravitomagnetic effect in spinning superconductors at Seibersdorf, but this sensational effect is also probably just bad experimental setup and not a new gravitational theory. He is the author of the classic book "Advanced Space Propulsion Systems".
If we never take the time to explore claims that violate our current understanding, our understanding will stop advancing. Too many scientists are overly dogmatic.
"The theoretical expectation is that you don't measure anything. Now you're measuring something, and you can learn extremely much on this hunt for the error."
"If the result confirms the hypothesis, then you've made a measurement. If the result is contrary to the hypothesis, then you've made a discovery."
Taking the EMDrive as an example -- we have measurements that contradict current theory, so we are either going to discover new physics or we are going to learn something about the art of making very precise measurements of force in complex mechanical systems. In no case do we learn nothing.
Kinda like how discovering the Higgs Boson was boring. It would have been far more exciting if they hadn't found it!
The likelihood that free momentum and free energy are possible is indistinguishable from 0. They are empirically impossible in the strongest reasonable sense the word can have.
And the experiments didn't make it obvious that the setup didn't work. I mean, I don't think anyone took the inventor's theoretical claims seriously, but the device did generate some small thrust in a difficult to explain way.
Besides, we know that space isn't nothing, so being able to tractor against whatever is there would be a rather intersting engineering find; physicists would then likely take the experiment and go "oh, there's a neat edge case in this system of equations that explains it". There are a lot of interesting edge cases left to find and engineer.
Momentum is conserved independently, it can not be created from energy without spilling some the other way (into whatever you're pushing against).
The vacuum is a translation invariant state so you can't push against it to get momentum either, but you can create particles out of it and then push against those. That's what electro magnetic thrusters do. They generate photons. The effect is just super tiny.
Well, if one know how to create such an effect...
We have a model of our perception of the universe
On the other hand, the universe is nothing more than the set of constraints that we call "science". The universe is the model, and science is an effort to duplicate that model.
If it was to me, I don't agree. Science is a collection of human ideas that exist in the minds of humans. As another commenter said, it's our model of the universe. There is no difference between science and our understanding of science unless you're talking about differences in understanding between individual people.
> On the other hand, the universe is nothing more than the set of constraints that we call "science". The universe is the model, and science is an effort to duplicate that model.
The reason that we know this is far stronger than simply: "It hasn't been observed yet". That would allow that any new thing could end up violating the established rules.
It could be that if you write the one hundred millionth Mersenne prime on a cookie, and the one two hundred millionth on a banana they start circling each other and create free energy. This claim has not been falsified. So on what grounds do we rule it out? Because it is incompatible at an incredibly deep level with everything we know about cookies, bananas, and how their physical properties depend on what's written on them.
Every aspect of each of these objects that has ever been observed has been found to be describable in terms of some unbelievably simple mathematical structures. For these unbelievably simple mathematical structures we have prove that anything that can be described by them does not allow free energy.
An observation like the Mersenne, cookie, banana interaction can not violate the mathematical derivation, therefore it implies that out of all the infinitude of observations that have been made of bananas and cookies, that all were describable by energy and momentum conserving physical theories, somehow this one arbitrary one is special. It can not be described by even the type of mathematics we have hitherto used and yet was completely invisible.
How could that be? Flying Spaghetti Monster of course. Just as the FSM hid the true age of the earth from us by interfering with science, so it also conspired to hide the free energy Mersenne Banana Cookie interaction from us.
In other words: If you know that the FSM did not create the earth 4000 years ago, then you even more definitely know that there is no free momentum/energy. If you think that we can't know either of these things, then your concept of knowledge sucks.
It's improbable to flip a coin 10 times and get head ten times. It's for all practical purposes impossible to do so 10^10 times. This level of mathematical probability is functionally indistinguishable from 0 thus it is entirely accurate to call it impossible.
It's not logically impossible of course, it just happens not to be possible in our universe. It's physically impossible. Like the prime cookie. The appropriate amount of time spent on investigating prime cookies is not "very small", it's 0.
Now there is of course reason to investigate the EM Drive to some degree: To figure out where the bug was. To learn about potential experimental errors. As a puzzle, to solve the apparent paradox. But any other framing of the issue is dishonest.
But when it is...
Snarkiness << skepticism.
How about this alternative (which, btw, is how I interpreted the headline): "Principle behind EM Drive refuted" or "Experimental results of EM Drive refuted".
The quote wasn't meant to be snarky. It was meant as an explanation why it is worthwhile to occasionally do experiments, even if you believe the device can't work the way it is supposed to. He also said it was excellent experiment for the purpose of teaching his students.
[1] https://tu-dresden.de/die_tu_dresden/fakultaeten/fakultaet_m...
[2] http://arc.aiaa.org/doi/10.2514/6.2015-4083
Everything suggests that it was just science in action. There was an experiment, and it failed to reject the null hypothesis of 'no thrust'. A few groups have repeated the experiment, but not to the experimental power needed to thoroughly reject it. This test seems to be a more definitive rejection of it, but we're still not sure why there was any thrust measured in the first place, though there are some pretty reasonable leads on that.
Let a few more rounds of experimentation and refutation go through, and if it's clear it can't work at all, and they stay in business without pivoting, then we can breeze right past hoax and jump to fraud. Time will tell about it, I guess.
I think, if nothing else, this is the take-away from the entire EM-drive analysis. I like it because it applies to many things, including software.
https://hackaday.io/project/10166-flying-an-emdrive
This is win-win: if it works awesome happens. If it fails then you're left with the experience of having built a little satellite!
That's called "science." :)
Sure, it is extremely small. But even that is different from zero.
[1] https://plus.google.com/app/basic/stream/z13gzfm4xt2tuxehl04...
Also I'm getting seriously creeped out by the amount of EmDrive non-skeptics on HN. Maybe we should make a poll whether 0.999... = 1, and then say that anyone who insists on the correct answer is "condescending".
Almost as those who might even admit they know nothing about it, but feel they're qualified to judge validity of some scientific claim. And provide <insert subculture> mantra as "evidence".
I've said nothing like that.
A working experiment that provides irrefutable data is needed to convince me in any way. Refutation I'd need to leave for others to do, not my competence.
I really like EmDrive as an idea and hope it will work out. It's very inspiring when someone has the guts to push it despite the dominant (scientific) opinion. If it does work out, these people will be some of our greatest heroes.
After all, sometimes people like these have been right in the past in adverse scientific atmosphere. I'm just content to follow it from the sidelines.
If I had to bet on EmDrive, I'd have to bet against it.
The real question is how much you should weigh a single experiment vs. the abstract idea of conservation of momentum. It's tempting to trust the experiment all the way, and imagine that science demands such open-mindedness. But the more physics you study, the more likely you are to trust the abstract idea instead, because you realize just how incredibly well-supported it is. It's a kind of certainty that has almost no analogue in other sciences. Even in a very careful experiment, measurement error is many times more likely than overturning conservation of momentum. (Also measurement error is much more devious than people imagine, but that's another story.)
I'm very confident ($100 to $1 at least) that the EmDrive is a story of measurement error, compounded by ignorance and optimism.
On a philosophical point of view, it worries me that physics might be optimizing towards a local maxima. Which might or might not be the global one. That there even might be something big it has missed. How would we know?
Even more so because physicists seem to be more and more specialized to a narrow sub-field. Everything is already so advanced it takes more and more years to just to master it. Individual contributions appear to be getting smaller and smaller.
Is it possible there's something they don't see because they're too near looking at it? Studying a mountain with a microscope. Is possible we're stuck in a some sort of dead end?
I really can't judge it either way. I just can't stop wondering about that.
If it's the former, I don't trust it at all, but if it's well practiced and interesting, it ought to be replicated. If it's the second, well, if its results are against theoretical models, theory is wrong. No questions, no exceptions. Now, theory correction does not automatically mean a fundamental change, "hey, this approximation we used has a problem here" is as valid a theoretical correction as "ops, all of our models are wrong, everything is much more complex than we thought".
Specifically about the EmDrive, I'm not aware of anybody that could actually measure an effect yet. So, there's no experiment showing theory is wrong. There's only a lack of experiments confirming it either (or better, now there is one).
I wouldn't. However, I would ask why you think violation of conservation of momentum is completely dispositive, rather than just a strong reason for great skepticism?
From Noether's Theorem [1] we know that conservation of momentum is a consequence of physics being invariant under space translation. As far as I've been able to find, we don't know that a universe that results from a big bang must be isotropic. Observationally, ours appears to be, but there are some hints that it might not be, and scientists are taking those seriously.
[1] http://math.ucr.edu/home/baez/noether.html
I can't find a single one in the comments here...