So apparently NASA just built an impossible propulsion device:

Sawyer's engine is extremely light and simple. It provides a thrust by "bouncing microwaves around in a closed container." The microwaves are generated using electricity that can be provided by solar energy. No propellant is necessary, which means that this thrusters can work forever unless a hardware failure occurs.

Why is this device considered impossible?

  • $\begingroup$ The link does not provide how NASA is involved. $\endgroup$
    – Anixx
    Oct 22, 2014 at 3:29
  • $\begingroup$ @Anixx Yes it does! "Now, American scientist Guido Fetta and a team at NASA Eagleworks—the advanced propulsion skunkworks led by Dr Harold "Sonny" White at the Johnson Space Center—have published a new paper that demonstrates that a similar engine working on the same principles does indeed produce thrust." $\endgroup$
    – uhoh
    Nov 15, 2023 at 22:55

2 Answers 2


It's a violation of Newton's 3rd law of motion? This is a fairly complete answer to the question in the absence of any deeper argument by the proponents of the device, but I'll get to those in a minute. Firstly, it's important to recognize that our universe follows a set of predictable and consistent laws. This might have been controversial in the 1700s. One of those laws is that forces must be accompanied by an opposite and equal force an another object. Now there are way that modern physicists can revise this in creative ways. For instance, there could be many bodies acted upon by a force that can't be divided up into 1-to-1 interactions, but these are all objections that don't change the bottom line. You can't push against nothing.

Another revision might be massless particles. Radio waves are photons, and they have momentum like photons. You can propel yourself by emitting any kind of light - but the forces are miniscule compared to what is being claimed here. It also requires tremendous amounts of energy.

So let's move on to the "modern physics" adjustments to Newton's 3rd law, and possible violations of it. It is sometimes claimed that general relativity doesn't preserve Newton's 3rd law. Maybe this is semantics. Practically, it's obvious that gravitational waves are scientifically verified and predicted by general relativity. Much like photons, these can carry momentum along with them. Since this is a "ripple in spacetime" and not momentum carried by some matter, there's certainly room for argument. However, momentum is still conserved. Also like photons, this is mostly useless in terms of the actual thrust involved and the difficulty of building a practical machine.

Instead, most of what I'm reading about this subject makes reference to "new" physics, mostly quantum gravity type stuff. There are many theories that would predict that some sort of reactionless drive is possible. However, proving such a principle through a reactionless drive prototype is contextually bizarre. If you could show even the slightest variation from classical physical theory, then that's interesting by its own merit. What I'm saying is that any such machine should first be interesting to physicists, then be interesting to spaceflight. It's entirely possible that we will discover some novel way to break Newton's 3rd law, but then we'll have no practical way to use it in spaceflight. After all, this was the exact case with my above 2 historical examples.

More than likely, they're not on to anything. Dr. Harold White uses real physics and makes claims backed up by genuine science, in addition to having access to a large body of peer scientists to back things up. However, his statements are difficult to parse correctly, and this combines with irresponsible reporting to produce headlines which are often blatant falsities. In his paper abstract, he refers to force produced by virtual particles. Great, that's fine. This is not tremendously surprising, because this is within established physics. It might just be that they achieve "propulsive momentum transfer via the quantum vacuum virtual plasma" without diverging from current theory. But that doesn't make it a reactionless drive! Current theory still preserves momentum. Maybe no one has done a detailed study to see how this particular case balances momentum, but we have every reason to believe that it does. Even if they are onto something, that lends no credibility to the bogus experiments that are searching for some true reactionless drive.

  • 10
    $\begingroup$ I've only had a chance to nibble on the attached paper, but it's quite likely that this engine, if it does work, does not work the way we thought it worked. One of the negative control engines produced thrust, and critically tests were performed at atmospheric pressure. It's possible/likely they've accidentally made a terrible ion drive, using the air in the test chamber as reaction mass. $\endgroup$ Aug 1, 2014 at 20:39
  • 3
    $\begingroup$ +1 to @JeremyKemball. Extraordinary claims need extraordinary evidence. Don't make the same mistake we made with cold fusion -- rather than leaping to an unsupported assumption, present the facts, invite others to repeat the experiment, and THEN decide whether you've got anything (and what you've actually got, if so). $\endgroup$
    – keshlam
    Aug 1, 2014 at 21:18
  • 7
    $\begingroup$ From the attached 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). They can't reject the null hypothesis. Move along, move along, there's nothing to see. $\endgroup$ Aug 2, 2014 at 1:42
  • $\begingroup$ Actually, David, given difference of output, they can reject the null. The "working" produces more more? The null of "There shall be no measurable difference in thrust between the null-machine and the working machine." If they get a significant difference, then they've failed the null. $\endgroup$
    – aramis
    Aug 2, 2014 at 20:41
  • 3
    $\begingroup$ They have to account for uncertainties. Did they do that? There's no indication from the paper. Torsion balances are notoriously touchy devices. People who work with them for years get inconsistent results. To wit, look at the inconsistent results in measurements of G. These guys are amateurs in comparison. There's nothing to see here. Until this is reproduced widely, it's best to look at this as garbage physics. $\endgroup$ Aug 2, 2014 at 23:54

Newton's 3rd law is empirical, and this observation could in theory be the first black swan. That by itself is unlikely, given the amount of empirical data gathered so far (the historically more likely hypothesis is measurement error; wouldn't be the first case).

What I consider more damning is that per Emmy Noether's first theorem, spatial symmetry is associated with the conservation of impulse. And that symmetry hasn't been disproven yet. This is a stronger statement, because we know Newtonian physics are wrong (see General Relativity) but Noether's theorem still holds in GR.

  • 1
    $\begingroup$ I'd say the microwaves escape the containment and produce light pressure; it's no different than laser propulsion or solar sail except for using microwave wavelengths. $\endgroup$
    – SF.
    Oct 22, 2014 at 9:06

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.