A source close to NASA Eagleworks has leaked the test results of the 'impossible' EM Drive. While it's important to note that the results that have been leaked haven't been published in an academic journal, they do suggest that the system works and is capable of generating force of 1.2 millinewtons per kilowatt in a vacuum. ScienceAlert reports: The paper concludes that, after error measurements have been accounted for, the EM Drive generates force of 1.2 millinewtons per kilowatt in a vacuum. That's not an insignificant amount -- to put it into perspective, the super-powerful Hall thruster generates force of 60 millinewtons per kilowatt, an order of magnitude more than the EM Drive. But the Hall thruster uses fuel and requires a spacecraft to carry heavy propellants, and that extra weight could offset the higher thrust, the NASA Eagleworks team conclude in the paper. Light sails on the other hand, which are currently the most popular form of zero-propellant propulsion, use beams of sunlight to propel them forward rather than fuel. And they only generate force up to 6.67 micronewtons per kilowatt - two orders of magnitude less than NASA's EM Drive, says the paper. The NASA Eagleworks team measured the EM Drive's force using a low thrust pendulum at the Johnson Space Centre, and the tests were performed at 40, 60, and 80 watts. They were looking for any sign that the thrust could be a result of another anomaly in the system, but for now, that doesn't appear to be the case. "The test campaign included a null thrust test effort to identify any mundane sources of impulsive thrust, however none were identified," the team, led by Harold White, concluded in the paper. "Thrust data from forward, reverse, and null suggests that the system is consistently performing with a thrust to power ratio of 1.2 +/- 0.1 millinewtons per kilowatt." But the team does acknowledge that more research is needed to eliminate the possibility that thermal expansion could be somehow skewing the results. They also make it clear that this testing wasn't designed to optimize the thrust of the EM Drive, but simply to test whether it worked, so further tweaking could make the propulsion system more efficient and powerful.