The problem with EM drive testing in general is that you're looking for a signal (thrust) which should be zero, while pumping an enormous amount of energy into

The problem with EM drive testing in general is that you're looking for a signal (thrust) which should be zero, while pumping an enormous amount of energy into a system, and declaring any nonzero reading as success. Physicists know from experience that this is a recipe for fooling yourself. There are a million things that could imitate the purported signal. In the last EM drive test I looked at, the "thrust" was pointing in the wrong direction, and moreover it perfectly matched the temperature curve of the apparatus, even though no theories of the EM drive say it should. That is a dead giveaway that the temperature is causing a fake signal, which e.g. could happen by thermal expansion messing up the meters. In general, the EM drive seems like a playground for sloppy experimentalists, because what anybody else would call a failure can be reported as a success.