Charles Buhler reactionless drive math error?

April 22, 2024 3:24 AM

A story about a NASA engineer discovering a new force of nature which enables reactionless propulsion popped up some days ago. I was intrigued at first but this all sounds very scammy. One commenter had to say this about the math in the underlying patent: "Lol the math is so bad. He "derives" the formula mv=t*dU/dx, but U switches from the total potential energy in the first equation to something like the potential energy density in that equation. The total potential energy U is not a function of x so dU/dx is zero and his whole argument falls apart." The equations can be found in the patent on page 20. And also at the end of the article I linked to in the beginning. But I have no knowledge of physics, is this piece of math indeed obviously false?
posted by SweetLiesOfBokonon to Science & Nature (17 answers total)

Best answer: From the abstract of the patent:

Asymmetries in the resulting electrostatic pressure force vectors result in a net electrostatic pressure force acting on the object. ... A non-limiting use case example is the use of electrostatic pressure force apparatus as a thruster to propel a spacecraft through a vacuum.

The accompanying diagram makes it perfectly clear that this "thruster" is self-contained, and that in operation it doesn't eject any mass and nor does it interact with any external field, electric or otherwise. So we don't even need to get into the weeds of the maths, because if the device worked as claimed it would violate conservation of momentum, which basic high school physics says is a thing that doesn't happen. The overwhelming likelihood, then, is that the associated equations have simply been made convoluted enough for their author to get lost in them.
posted by flabdablet at 4:59 AM on April 22 [10 favorites]

I also note that this is a patent application rather than a granted patent. Which makes sense, patent offices having a longstanding policy against granting patents to perpetual-motion boondoggles.
posted by flabdablet at 5:15 AM on April 22 [4 favorites]

Best answer: I am a theoretical physicist (though I am not your theoretical physicist) and I agree with flabdablet. The important point here is not so much "we know momentum is conserved", but that "we know that Maxwell's equations of electrodynamics conserve momentum". The latter can be rigorously proven in a very simple way using Noether's theorem.

The claim here is that the author has derived a momentum-nonconserving result from electrostatics (which is a subset of the physics described by Maxwell's equations). This means, incontrovertibly, that the momentum-nonconservation is a result of an error in the calculation.

There's apparently a clique of NASA engineers who have been memetically infected with this electrostatic/electrodynamic reactionless drive nonsense after a few of them fell for the EmDrive idea, which was similarly based on a complicated calculation of radiation pressure using the standard formalism described by Maxwell's equations but with errors in the calculation.
posted by heatherlogan at 5:36 AM on April 22 [28 favorites]

It's true that there is Wierd Shit that happens when fields are involved, see the Casimir effect, but none of that Wierd Shit will produce usable thrust outside the apparatus because that's Not a Thing. It's like Maxwell's Demon as described in Mr Tompkins in Wonderland: any mechanism that could e.g. sort molecules by heat would itself produce more entropy than it was removing from the affected / enclosed system. Net effect: trombone sound.
posted by seanmpuckett at 7:46 AM on April 22

One fun thing is that if we really had a reactionless drive that could put out 1G of acceleration indefinitely, you could get a probe to Alpha Centauri within a decade. Which, incompatibility with known physics aside, gives you a sense I think of the magnitude of what is actually being claimed when people say they have a working reactionless drive.

It's like finding a free, clean, limitless source of energy by rubbing two balloons together, or something.
posted by BungaDunga at 7:54 AM on April 22

Best answer: A reactionless electrostatic drive is not only like a free, clean, limitless source of energy, it is a free, clean, limitless source of energy.

Being electrostatic, charge doesn't move around inside it; therefore, maintaining its internal arrangement of charges requires no energy input. But if it makes thrust, then all you have to do to get usable energy out of it is hook it up to the end of a crank attached to a generator. Which would also require some form of braking to stop it trying to wind itself up to infinite speed when run at under 100% capacity.

So yeah, it breaks conservation of energy as well as conservation of momentum. The thing is a canonical perpetual motion boondoggle.
posted by flabdablet at 8:31 AM on April 22 [2 favorites]

Best answer: Ok, there's one other thing I gotta say here. There is no such thing as "electrostatic pressure force". There is electrostatic force; there is also electrostatic pressure, which is force per unit area defined for a particular surface. What seems to be being described in the patent application is simply electrostatic force.

Electrostatic force can always (always!!!) be described as a sum of Coulomb forces between point charges. (You have to work out where the point charges are, which can be complicated when conductors are involved, but they are definitely somewhere, and then the total force is just the vector sum of a bunch of Coulomb forces.) And we know, for every such pair of charges, that the force on point charge A due to point charge B is equal in magnitude and opposite in direction to the force on point charge B due to point charge A (this is just Newton's third law as manifest in Coulomb's law). The net force vector upon the whole assemblage is the sum of all the individual forces, which conveniently cancel pairwise between each pair of point charges, guaranteeing that the total force is zero. QED.
posted by heatherlogan at 8:31 PM on April 22 [1 favorite]

Best answer: Oh, and the math... Eqn. (7) is equal to zero, because the integral of E^2 is over all space and then there's no x dependence left to differentiate with respect to. (Also, it looks like somebody forgot that force is a vector; Eqn. (11) is nonsensical because the right-hand side is a scalar.)

How badly does an engineer have to screw up in order to lose their PEng?
posted by heatherlogan at 8:41 PM on April 22 [1 favorite]

There are a LOT of really smart guys* who have been taking this really seriously for a really long time. (*it's gonna be all men, right?) I know a whole forum subsection full of them, in fact a discussion of Buhler's work is at the top of that forum right now. (To be clear, I'm not one of these smart guys, just aware of the churning discussion.) There are probably pure-physics forums where it's being discussed as well.

So while I absolutely agree that this whole thing smells of physics violations, and it has for a very long time, I personally am just withholding judgement and letting them enjoy doing their thing ... For all we know, they could very well be on the trail of some new physics, and they just have to find the right way to jimmy open that crack. (And I hope that bit of gentle accommodation doesn't cause the engineering police to break in my front door and revoke my PE license.)
posted by intermod at 9:10 PM on April 22

Smart doesn't equate to well-informed, and Engineers' Disease is called that for good and sufficient reason.
posted by flabdablet at 11:46 PM on April 22

it looks like somebody forgot that force is a vector

I think you're forgetting that this is a pressure force. Typical woman, amirite guys?
posted by flabdablet at 11:48 PM on April 22

I absolutely agree that this whole thing smells of physics violations, and it has for a very long time, I personally am just withholding judgement

It doesn't just smell of physics violations, it reeks. It screeches and howls and gibbers and glares with physics violations. This is a bunch of smart guys indulging in a circle jerk, just like the US Government Is Concealing Alien Visitors push, and upon whom withholding judgement merely reflects poor judgement.
posted by flabdablet at 11:57 PM on April 22 [1 favorite]

Best answer: The more I think about this "theory of operation", the dumber it gets.

He starts with conservation of energy -- it's equation (1) in the patent application. This just says that as the device moves around, its kinetic energy increases or decreases to compensate for the decrease or increase of its potential energy.

But the only potential energy being ascribed to the device is its internal electrostatic potential energy, which stays the same no matter where in space the device is located. No change in potential energy with position means the kinetic energy must be the same no matter where the device moves to; if it starts at rest, it stays at rest. As the one commenter quoted in the OP says, "The total potential energy U is not a function of x." This is high-school level physics. I can assure you that it is not being discussed in pure-physics forums.

There's no crack here; there's nothing to jimmy open; it's just one dude who has clearly forgotten anything he ever learned in the physics course required for his engineering degree and a bunch of people who want to believe so much harder than they want to understand.
posted by heatherlogan at 5:47 AM on April 23 [3 favorites]

Elsewhere Dr. Buhler explains this is a variety of the EmDrive technology. NASA tested the EmDrive in 2014, according to Wikipedia.
posted by cthlsgnd at 6:07 AM on April 23

NASA tested the EmDrive in 2014

and... (drumroll, badum tish!) it didn't work.

people who want to believe so much harder than they want to understand

are responsible for more earthly misery than I care to think about.
posted by flabdablet at 8:36 AM on April 23

Crap.

I guess this means no juice, right? Not even a hoverboard?
posted by mule98J at 9:06 AM on April 24

I guess this means no juice, right? Not even a hoverboard?

Sadly, our only recourse is to console ourselves with memes.
posted by heatherlogan at 3:23 PM on April 25 [1 favorite]