The encyclopedia · R&D & Science · Strategic decision · 2001–2021
Shawyer's EmDrive: twenty years of selling an engine that broke physics
A British engineer spent two decades selling a reactionless drive; the only thrust anyone could confirm was a warm scale.
Satellite Propulsion Research Ltd · 2002-12
What happened
Roger Shawyer founded Satellite Propulsion Research Ltd in 2001 on a single claim: a sealed metal cone that bounces microwaves around inside can produce thrust with no propellant — a reactionless drive that would break conservation of momentum. In December 2002 he demonstrated 0.02 newtons of thrust from an 850-watt rig for a few dozen seconds before it overheated. The result was never published and never independently replicated; he sold the idea anyway.
For fifteen years the EmDrive was Shawyer's pitch: no fuel, no reaction mass, Mars in weeks. Boeing looked in 2009 and walked away by 2012. The claim found its best marketing by accident in 2014, when slides from NASA's tiny Eagleworks lab leaked early and made headlines around the world; in November 2016 Eagleworks published a peer-reviewed paper reporting 1.2 millinewtons per kilowatt in vacuum — 'peer-reviewed' being the word that did the work.
Independent labs never saw real thrust. In 2018 Martin Tajmar's group at Dresden found the drive 'pushed' just as hard in configurations where physics says it cannot — an artefact of Earth's magnetic field pulling on unshielded power cables. In 2021 they rebuilt the NASA experiment exactly and found the scale itself: the engine warms, the fastening elements warp, and the balance settles on a new zero point. 'Our measurements refute all EmDrive claims by at least 3 orders of magnitude.'
Twenty years after the first demo, nothing had flown, no buyer had signed, and the company's name had become a byword for the impossible. The debunk earned a fraction of the headlines the leak earned, and the peer-reviewed paper became the meme's best evidence. The real bill landed on the founder: a lifetime claim that a warm scale had made briefly, gloriously credible.
Why it happened
- The demo never repeated: 0.02 newtons for a few dozen seconds, unpublished and unreproducible, was the evidence the whole company stood on.
- The claim was marketed before it was demonstrated — two decades of promotion with no independent replication, and Boeing's 2012 walk-away as the best due diligence the drive ever faced.
- The legitimacy was borrowed: NASA's peer-reviewed paper was the company's best advertising, and it was built on a scale that warped when warm.
- There was no exit: no contract, no flight, no admission — only a meme that outlived the claim.
The lesson
A claim physics forbids needs a demonstration physics can trust — and a demo that overheats before it repeats is not evidence, no matter whose peer-reviewed paper quotes it.
Aftermath
Satellite Propulsion Research never flew anything and never closed a licensing deal; the EmDrive survived as a meme and as the reference case for 'peer review is not physics'. NASA never retracted the paper but wound down the Eagleworks line; Dresden's result explained the signal, mount by mount. Shawyer kept selling after the debunk — the drive still has a website and a mailing list. The genuine casualty was the field's patience: every real exotic-propulsion proposal since has had to answer for the warm scale first.
Sources
- Wikipedia — EmDrive (full timeline: Shawyer 2001–2002, SMART grant, Eagleworks 2014 leak and 2016 JPP paper, Dresden 2018 and 2021 null results)
- VICE — 'NASA's Peer-Reviewed Paper on the EmDrive Is Now Online' (Nov 2016, 1.2 mN/kW claim, pilot-wave theory, peer-review caveats)
- New Scientist — "'Impossible' EM drive doesn't seem to work after all" (Jun 2018, Dresden 4 µN signal, Earth's magnetic field, unshielded cables)
- Universe Today — 'In a Comprehensive New Test, the EmDrive Fails to Generate Any Thrust' (Apr 2021, thermal warping of scale fastenings, refuted by 3 orders of magnitude)
spotted an error? The club wants to know.
More like this
LK-99: the 'world-first' superconductor that was copper sulfide and iron
CERN's 750 GeV bump: 500 papers about a statistical wiggle
The smoking gun of cosmic inflation was a smudge of galactic dust
Somewhere, someone solved the problem this company failed at. 2nd Opinion →

Comments · 0
Sign in to join the comments.