[stargazer-3]

In an unpublished NASA report they are seeing over 100uN when feeding 80W power to the drive. Assuming a spacecraft to weigh a ton, that would accelerate it by ~3 m/s per year. So it's not only a question of EmDrive working, but also a question of whether it's possible to scale up the technology.

[VLM]

Isn't that an incredibly tempting figure? Almost like the result was picked because it almost has to be commercialized?

You'll get various war stories about what it really took in practice for satellite XYZ, but geosynch satellites burn along the lines of fifty or so m/s of delta-v per year for station keeping. To a first order simplification its true that you launch a geosynch comsat into space and it stays over the same general point on the earth for infinite time, blah blah but in practice there is always slow drift and impact of solar winds and lunar gravitation and whatever, so its not actually motionless and requires continuous, eternal, fine tuning.

Anyway using a couple hundred watts to completely eliminate the stationkeeping thruster fuel completely seems a fair enough trade...

And yes I know stationkeeping isn't a perfect continuous demand for thrust all year long, etc. Still, its interesting, and of the right magnitude, more or less.

I don't think it would be terribly useful for human transport, true. But its suspiciously right at the mass, power, and thrust values required to be commercially viable for comsat station keeping. That's either incredibly lucky, sometimes the universe does smile on us, or ...