Don't Get Too Excited About NASA's New Miracle Engine

News spread quickly this week that a propulsion system said to violate the laws of physics had been validated by no less an authority than NASA itself. Could a fuel-less space drive propelled by microwave thrust –something once deemed impossible – actually be possible? We spoke with some experts. Let's just say they're skeptical.

Above: Spaceship concept art by L.A.-based artist Cuba Lee

Earlier this week, Wired reported on an unusual engine designed and tested by researchers at NASA's Johnson Space Center. Dubbed the "Cannae drive," the propulsion system is similar to the so-called EmDrive, a "reactionless" engine proposed years ago by British engineer Roger Shawyer and popularized in a 2006 writeup in New Scientist. Both space drives are designed to convert electric power into thrust by bouncing microwaves around in a closed container, thereby eliminating the need for onboard propellant. The concept has been roundly criticized for appearing to violate the law of conservation of momentum.


And yet, Chinese scientists claimed last year to have built an EmDrive capable of producing thrust. Now, in a paper presented at last week's 50th Joint Propulsion Conference in Cleveland, Ohio, a team of NASA researchers claims its Cannae drive has also achieved thrust. Hence Wired's headline: "Nasa validates 'impossible' space drive."

But according to Drexel University physicist Dave Goldberg (io9's friendly neighborhood "Ask a Physicist" columnist), it's important to keep three things in mind, when claiming that NASA has validated the Cannae drive's "quantum vacuum virtual plasma," the hand-wavy concept by which the engine is purported to exert thrust:


  1. A team of NASA researchers ≠ NASA, the organization. NASA is not a monolithic entity. It has hundreds of research groups that do their own work.
  2. Extraordinary claims (for instance, that the Cannae drive violating conservation of momentum) require extraordinary evidence.
  3. We simply don't have the data to make a judgment here yet.

CalTech physicist Sean Carroll, who, earlier this week, equated the notion of "propulsive momentum transfer via the quantum vacuum virtual plasma" to "nonsensical sub-Star-Trek level technobabble," had more to add, telling io9:

The business about "quantum vacuum virtual plasma" (the physics of which they "won't address" in this paper) is complete bullshit. There is a quantum vacuum, but it's nothing like a plasma. [The researchers] hook up a gizmo with all sorts of electromagnetic fields fluctuating around, then claim to measure an extremely tiny thrust (about the weight of a single grain of sand), which occurs even for the test article that wasn't supposed to produce any thrust at all.

Carroll's final point – that the researchers measured thrust not only when the drive was configured to produce it, but also when set up to do nothing at all – may be the most important takeaway of all. It's a point Mika McKinnon expands upon in her explanation below.

A New Thruster Pushes Against Virtual Particles! ...or is a Lab Error

Which is more likely: that NASA broke the laws of physics, or that an early experiment on propellant-free microwave thrust technology has a measurement flaw? If you opted for the latter, you have a solid grasp of the cycle of elation and chagrin that characterizes the practice of science.

Above: Spaceship concept art by L.A.-based artist Cuba Lee

Do you remember the "faster-than-light neutrinos" story a few years ago — in which repeated experiments kept turning up neutrinos that violated the universal speed limit? At least, they did until determined researchers finally tracked down a jiggly wire, plugged everything in firmly, and determined that whoops, no, the neutrinos travelled at perfectly ordinary speeds. How about in the 1990s, when NASA tested an antigravity system, only to realize that the positive test results were from interference from the testing equipment? Those are valuable stories to remember when reading excited headlines about NASA breaking the laws of physics with new engine technology.


During an episode of Cosmos in 1980, Carl Sagan taught, "Extraordinary claims require extraordinary evidence." Despite all the advances in science and technology that have happened since then, that basic philosophy is still incredibly useful when assessing crazy new breakthrough technology that fundamentally redefines how we approach decades to centuries of scientific understanding.

The EmDrive is a concept for producing thrust by bouncing microwaves around a waveguide, with no propellant necessary. If it worked, it'd simplify spaceflight in all sorts of amazing ways.

The EmDrive. Ooooh, pretty!


One small problem: thrust without propellant should violate Conservation of Momentum, a conservation principle that states momentum (mass times velocity) stays constant. It's a fundamental concept, one of the basic inviolate ideas taught in introductory physics classes right along with Conservation of Energy and the universal speed limit of light in a vacuum. Sometimes obeying the various conservation laws can be mind-warpingly complex as you realize just how strange and wonderful our universe is in its myriad of quirks — but if you take it carefully, sooner or later you find that bit you were missing to make everything fall into place. Claims that something violates Conservation of Momentum are met with just as much justified scepticism as any new plan for a perpetual motion machine, or a method of extracting free energy.

Now, electromagnetic waves (or "light," if you have no patience for unnecessary complication) do carry momentum. The easiest way to wrap your head around that is to think of the particle half of the wave-particle duality, where its momentum is just mass x velocity. The velocity is the speed of light (c), but photons are massless. Turn to Einstein's famous equation, E = mc2, do some algebraic jiggling to solve for mass, and drop that into the momentum equation. It simplifies down to momentum = Energy / speed of light. And thus, we get a quite-literally-shiny way of transferring momentum to obey the conservation laws, and everyone is happy, right?

...except that the system is supposed to bounce light around a closed waveguide, which is a closed system. That's the physics equivalent of pulling yourself up by your bootstraps: a nice concept, but fatally flawed in execution.

So, next came the Cannae drive, a quantum vacuum plasma thruster. It takes the same concept, but opens up the system to shove off against virtual particles, theoretical particle and anti-particle pairs that spontaneously pop into existence, mutually-annihilate, and disappear. Freaky, weird, still not entirely solid physics that we're confident is real, but getting more plausible. This is the new propellant-less microwave thrust system that NASA recently subjected to a round of preliminary testing.

A Cannae drive pushing off against virtual particles. Trippy!



NASA is a well-respected institution; they aren't some wild-eyed cranks shouting techno-babble gibberish at passing physicists. If they say something is real, it's worth taking a closer look.

Except NASA doesn't say the Cannae drive worked, and they very specifically avoided looking at the physics of it at all. A few researchers at NASA just wrote a conference paper about eight days of testing of the engine, with a bunch of adjusting, modifying, and rejiggering to see if they could get the device to function.


A quick note about the practice of science: conference papers are the work-in-progress presentations of the research world, where people announce preliminary results, neat effects, and other things they think are worth pursuing in more detail to confirm it's real before tackling the peer-reviewed journal publication process. To say something was announced in a conference paper is equivalent to a scientist going, "Hey, I'm working on this thing that I think is neat. Do you have ideas for what else I should check while I'm investigating it?"

They announced in their conference paper that they spent just over a week testing a Cannae drive. It wasn't finalized, rigorous testing, but preliminary tinkering to see if it was worth pursing. If you read the 1-page abstract, it specifically mentions they were constantly moving the device into the test chamber, back onto the bench to fuss with it, then pop it back into the test chamber. In those eight days of testing, they measured thrust.

Here's where it the results move from plausible to suspicious. From the paper:


Thrust was observed on both test articles, even though one of the test articles was designed with the expectation that it would not produce thrust. Specifically, one test article contained internal physical modifications that were designed to produce thrust, while the other did not (with the latter being referred to as the "null" test article).

Restating: the researchers measured thrust when the drive was all set up to produce thrust, but they also measured thrust when it was set up to do nothing at all. That the null test article produced thrust is really suspicious. Either it's a measurement error, or this drive produces thrust by some mechanism that isn't explained by the semi-plausible physics backing it up and this breakthrough in spacecraft propulsion is working by some not-even-theoretical mechanism.


Now, which is more likely?

  1. Eight days of initial tests on a piece of controversial technology in a NASA lab have proven all-new, extraordinary, physics-revolutionizing spacecraft propulsion in a manner so spectacular that the drive works even when it isn't set up to do anything at all; or
  2. Somewhere in the testing process is some sort of procedural, mechanical, or interference error producing false results.

As someone who has done my fair share of novel research that didn't go exactly as expected, this conference abstract reads like the researchers were looking for extra eyeballs to figure out what about their testing rig might be flawed — not a grand announcement of a spectacular breakthrough. This has the potential to be cool, but at the moment, about the strongest thing that it's scientifically responsible to say about these test results is that the researchers need to revise their testing setup.


If you came here directly, I highly recommend this intro with comments from another pair of physicists. Thank you to Ryan Carroll of Central Oregon Community College for dissecting the physics with me! I had a hard time picking a top image for this story. What's your favourite implausible, high-technology, far-out spacecraft?

Reply317 replies
Leave a reply