Skip to content

“Either the results are completely wrong, or Nasa has confirmed a major breakthrough in space propulsion.”

Daniel Lakeland points us to this news article by David Hambling from 2014, entitled “Nasa validates ‘impossible’ space drive.” Here’s Hambling:

Nasa is a major player in space science, so when a team from the agency this week presents evidence that “impossible” microwave thrusters seem to work, something strange is definitely going on. Either the results are completely wrong, or Nasa has confirmed a major breakthrough in space propulsion. . . .

He has built a number of demonstration systems, but critics reject his relativity-based theory and insist that, according to the law of conservation of momentum, it cannot work.

According to good scientific practice, an independent third party needed to replicate Shawyer’s results. As reported, this happened last year when a Chinese team built its own EmDrive and confirmed that it produced 720 mN (about 72 grams) of thrust, enough for a practical satellite thruster. . . . a US scientist, Guido Fetta, has built his own propellant-less microwave thruster, and managed to persuade Nasa to test it out. The test results were presented on July 30 at the 50th Joint Propulsion Conference in Cleveland, Ohio. Astonishingly enough, they are positive. . . .

OK, that was 3.5 years ago. Any followups? A quick google search revealed this article by Guilio Prisco from 2017, “Theoretical Physicists Are Getting Closer to Explaining How NASA’s ‘Impossible’ EmDrive Works: The EmDrive propulsion system might be able to take us to the stars, but first it must be reconciled with the laws of physics.”

If I wanted to be snarky, I’d say they could do a 2-for-1 deal and power the Em-drive with cold fusion. But my physics knowledge is weak, so I’ll just say . . . who knows, maybe this is the interstellar drive we’ve all been waiting for! I’ll believe it once it appears in PNAS.


  1. Bob says:

    The Wikipedia article on the EmDrive does not use the phrase “A kangaroo, a feather, and a scale”, but it refers to the same idea. Specifically, it states:

    In all of the experiments set up, a very large amount of energy goes into generating a tiny amount of thrust. When attempting to measure a small signal superimposed on a large signal, the noise from the large signal can obscure the small signal and give incorrect results.



  2. Anoneuoid says:

    This em drive stuff is a classic case of no one doing direct replications. As a result none of the measurements from different groups are directly comparable, and there is always an excuse for why they were different “this time”. Saw it in preclinical research all the time.

    Check out the “Facilities of Research-Spinal Cord Injury” project, which was NIH funding direct replications and leaving no room for such excuses. Not sure if this is all the studies but afaict nothing replicated:

    After that the “solution” seems to have been “don’t fund direct replications” since nothing changed in practice. That seems to be the strategy in play here as well. Intriguing idea plus sloppy methodology equals black hole for funding.

  3. Jaime Frailey says:

    As far as I’ve read, and I am interested in this, it seems to have something to do with the power generated by earth’s magnetosphere. I suspect when NASA (or another space-fairing country) tests this far enough away from a planet, there won’t be this phenomena.

  4. From Where is My Flying Car (by yours truly):

    The main problem with reactionless thrusters is not so much that some smart person hasn’t managed to invent them yet, but that they are completely incompatible with generally accepted physical law. Unlike cold fusion, where the energy is available according to standard physics if we could only figure out a workable pathway, a space drive flat-out violates the conservation of momentum. Conservation of momentum is such a basic part of the laws of physics as currently understood, that the math (Noether’s Theorem) tells us that if it isn’t universally true, then we can’t even expect the laws of physics to remain the same when we do an experiment in a different place (i.e. to remain symmetric to space translation).

    It is then purely in the spirit of trying to avoid Failure of the Imagination that I, gathering with frail and failing hands the tattered shreds of my scientific credibility, examine the possibility that we might actually get a space drive any time soon, or ever. Even so, the only reason that I feel justified even mentioning it is that there has been published a peer-reviewed paper in a reputable journal [White, March, Lawrence, Vera, Sylvester, Brady, and Bailey: “Measurement of Propulsive Thrust from a Closed Radio-Frequency Cavity in Vacuum”, AIAA Journal of Propulsion and Power, December 2016] by NASA experimenters claiming exactly that.

    The EM Drive, invented by Roger Shawyer in 1999, is conceptually quite simple. It consists of a frustum (truncated cone) of electrically conducting (and thus microwave-reflecting) material, with flat ends. An office trash can, if it had a flat lid, would approximate the shape. The test article in the experiments was 9 inches long, 11 inches across on the big end, and 6 inches across on the small end. There was a 2-inch thick slab of polyethylene fitted into the small end. In operation one directs a stream of 1.9 GHz microwaves to bounce back and forth between the ends, setting up a standing wave pattern with two nodes along the axis of the cone and 4 across. This is supposed to generate a thrust toward the big end.

    The tests were carried out in vacuum, with fairly elaborate precautions taken to prevent a wide range of possible confounding factors. The instruments had rated measurement precision in the nanoNewtons, 10,000 times smaller than the actual measurements. Even so, the measured thrusts are tiny: if you could somehow scale it up to high power at the best thrust-to-power ratio achieved, you would need 2.2 megawatts (3000 horsepower) to produce a pound of thrust. Even so, the thrust is three orders of magnitude higher than a photon rocket, producing 1 to 2 \mu N/W compared to 3 to 6 \mu N/kW. This is one reason that the authors are confident that the thrust is not due to a leaked radiation effect. On the other hand, it is about a factor of 50 less than a Hall thruster (ion rocket), but of course the Hall thruster uses reaction mass.

    What can possibly be going on? By far the most likely explanation, of course, is that someone made a mistake, either in measurement, or in not excluding some perfectly conventional effect, or the like. 100 nanoNewtons is the sideways force you get on a 10-pound weight when you put it on a frictionless surface tilted by one inch per the diameter of the Earth. By far the least likely explanation is that the the EM Drive really is a reactionless thruster in the basic physics sense. If there is any daylight at all between these extremes, it would have to be that there is something against which the thruster is getting traction: dark matter, perhaps, or the mass/energy implicit in the quantum field.

    Once upon a time, I had the opportunity to listen to an artist who made bells of glass. These were quite beautiful and produced the same tinkling sounds you would expect. For contrast, he had also made similar bells out of pure quartz: the same silicon dioxide substance, but a crystal instead of the disordered molecular structure of glass. The quartz bells rang with a pure tone that went on, and on, and on—audible perhaps ten times longer than the glass ones.

    An electronics engineer would have said that the quartz bells had a higher Q factor than the glass ones. Q in a resonator is a measure of how fast energy at the resonant frequency dissipates out into other frequencies and is lost. The EM Drive is essentially an electronic bell. The higher the Q, the better the efficiency of force produced per watt of input power, according to Shawyer’s theory. Note that Shawyer’s theory may not be the right one even if the device does produce thrust; but the part about the Q factor seems much more reasonable than the rest of it. Shawyer’s (and NASA’s) copper, almost steampunk-looking, experimental machine has a Q of 35 thousand; the state of the art in superconducting microwave resonators used in particle accelerators, for example, is 50 billion. Call it a factor of a million better efficiency.

    If this turns out to be a real effect and Shawyer’s theory turns out to be right—and we are going far afield into la-la Failure of the Imagination land here—we should expect to be able with nanotech to build atomically precise, optical frequency thrusters that get on the order of Newtons per watt. In which case your two-ton flying car is supported against gravity by a 30-horsepower engine.

    The perspicacious reader will instantly smell a rat. If we put the same thruster in only a one-ton car, the car would accelerate upward at one G (its occupants experiencing 2G), gaining energy at a rate that would very quickly surpass 30 HP. Well, no. Even Shawyer’s theory says the effect disappears with acceleration. But Einstein tells us that gravity and acceleration are the same; maybe the gadget can’t even work against a gravity field? After all the experiments, as far as I know, were all done with the gadget (a) standing still, and (b) oriented sideways to the local gravity.

    One might imagine it being used to maintain a static force, e.g. to overcome air resistance, when following an orbit-like path. But even that quickly runs afoul of the conservation of energy, given the heat being dumped into the air. So it is quite difficult to conceive of a law of operation for the EM Drive that is both consistent and useful.

    [Added in press: a later set of experiments makes the “they made a mistake” explanation of the effect look a lot more probable. [Martin Tajmar, Matthias Kößling, Marcel Weikert, Maxime Monette: The SpaceDrive Project – First Results on EMDrive and Mach-Effect Thrusters, Space Propulsion Conference, Seville, Spain, May 2018] My faith in physics is restored.]

    • Robert Virkus says:

      ” a space drive flat-out violates the conservation of momentum”

      I’m a great fan of your work on Drexlarian Nanotechnolgy but I’ll have to disagree with that statement above.

      That’s the strawman everyone sets up to claim the laws of physics as currently understood must eliminate such possibilities but in reality, I believe it’s simply a lack imagination. Not all proposed or experimental space drives are the same. The work of Woodward has both theoretical and practical aspects which explain why it works and it why doesn’t violate momentum conservation. Yet it’s a “space drive” and has serious funding and support.


      Robert Virkus

  5. Christopher Hall II says:

    @J Storrs Hall : RE:1.19.19 11:32 am “From Where is My Flying Car (by yours truly)”

    Bravo—a well-crafted and thoughtful response.

    Off-topic: you have a fantastic surname, if I do say so myself ;-)

    — C. Hall II

  6. Daniel H. says:

    that’s basically the reference Gavin’s article above mentions, right?
    It was also featured on Slashdot

    “Unfortunately, the EM drive also generates the thrust when the thruster is directed so that it cannot produce a torque on the balance (e.g., the null test also produces thrust). And likewise, that “thrust” reverses when you reverse the direction of the thruster. The best part is that the results are the same when the attenuator is put into the circuit. In this case, there is basically no radiation in the microwave cavity, yet the WTF-thruster thrusts on. So, where does the force come from? The Earth’s magnetic field, most likely. The cables that carry the current to the microwave amplifier run along the arm of the torsion bar.”
    “The researchers’ conclude by saying: “At least, SpaceDrive [the name of the test setup] is an excellent educational project by developing highly demanding test setups, evaluating theoretical models and possible experimental errors. “

  7. Physicist Mike McCulloch proposes a theory of quantised intertia from which he has calculated estimates of propulsion that agree with a number of (varying) observations of EM drive thrust:

  8. Michael Nelson says:

    This is an even more recent article that says the German team did a poor replication–not just because it varied from the protocol, but also because they just did bad science:

    The site Next Big Future appears to be legit. There are several news articles, also post-May 2018, that say DARPA is continuing to fund the research, and that McCulloch is collaborating with scientists from Spain as well as from the German lab:

    Here’s the key quote: “A German team at Dresden is evaluating the EmDrive and will report next year, though early results suggest thrust measurements could instead be stray magnetic fields.”

    That line links to an earlier PM article, and the key quote there is:

    “‘We identified errors that others may have missed to far,’ Tajmar [the German scientist] says. ‘However, we are an order of magnitude below the power levels from the NASA tests and we need to test different geometries and frequencies.’ That is to say: The magnetic effect could account for some part of the weird EmDrive results, but it can’t account for everything, at least not yet.”

    It looks like the most recent reports debunk the debunking.

    • Anoneuoid says:

      Thanks Michael. This is exactly what I found when looking into the EM drive (see my other post in this thread):

      But for an unknown reason Tajmar and his team didn’t use the mandatory stepup/isolation transformer: Therefore they operated the device at the wrong frequency, one that could never trigger any thrust signature.

      For some reason no one is reporting actual direct replications that should be free of such excuses. I find it quite ridiculous and it casts doubt on the integrity of the whole thing for me.

    • Ooops, wardrobe malfunction! Sorry, Andrew Gelman, for my prior comment, but WHY can’t anyone use acronyms correctly anymore?! NASA = National Aeronautical and Space Administration, not “Nasa”. The BBC is twice as guilty, often referring to NATO as Nato. What is “Nato”? Some guy named Nathan with a catchy nickname?

      I am not complaining about the English usage of our august blog host, the venerable Professor Gelman. Behold him here in all his glory.

      How can one not respect this man?

      I just hate getting confused by sloppily-rendered acronyms. Words *ARE* important.

      • Ambassador for Nato and Nasa says:

        Words are important; hollering about details of no consequence at all not so much.

        When I was in school, I was good–top of the class–with grammar and rules, the so called correct way of using language. This lead to an unhealthy obsession with inconsequential details, but now-a-days I’m more concerned with how well the text communicates its meaning, as long as it is sufficiently well formatted.

        I expect people to be able to read a string of letters such as Nasa or Nato and use contextual information to infer what is meant by them. It is inconceivable to me that anyone would be confused; but if they are, then the problem is with something else than what is capitalized and what is not.

        P.S. Also, it is this sort of contextual reasoning that I’m expecting people to be able to employ when deciding what “sufficiently well formatted”.
        P.P.S. Regarding the anecdote about being good with grammar: obviously I am not talking about English, but my native tongue.

        • Terry says:

          More important than making text “correct” is making it uniform. If everyone spells them as NASA and NATO, then, a reader will be slowed down a split second when coming across Nasa and Nato. So using your rule that a text should communicate well, the more standard usage is preferable. Don’t annoy the reader unneCCessarRilly.

    • Dzhaughn says:

      That’s a tough one. ARGH: Acronym Rigor’s Gotta Happen?

  9. David Hambling says:

    Keep an eye on Mike McCulloch’s work —

    Martin Tajmar is also pursuing this. And the Cannae team are likely to report this year too. Also, NASA’s Eagleworks may finally get a paper out.

    Meanwhile, the Chinse have gone quiet in the last year or two, having earlier promised a space launch ‘soon’, but I am reliably informed that their work continues.

    THanks for your interest!

  10. Seeing would be believing.
    To see any results that prove the concept you need:
    1) video camera.
    2) a large air table.
    3) marker-tape
    4) the prototype EM device with portable power supply.
    5) a bubble level and plumbline.
    Put number 4 on number 2 and see what happens.

  11. Chris says:

    No to beat a dead horse here, but the NASA peer reviewed paper about this is (officially) out:

    One such theory/explanation for the unknown thrust given by the scientists was:
    “If the vacuum is indeed mutable and degradable as was explored, then it might be possible to do/extract work on/from the vacuum, and thereby be possible to push off of the quantum vacuum and preserve the laws of conservation of energy and conservation of momentum.”

    Which, sounds eerily similar to a recent patent granted by a Navy physicist (and endorsed by the Chief Technical Officer of the Naval Aviation Enterprise):

  12. Warthog says:

    And yet “everyone knows” that solar sails work. The only difference between the EMDrive and a solar sail is source and wavelength of photons involved.

  13. Batacurium says:

    If ı am remembering correct,the German research team at the University of Dresden explained the small amount of force is actually the expansion of the material created by the microwaves hitting to the surface.I don’t know a lot of about this so if someone has any information please enlight me and all the other curious people about this subject

Leave a Reply