In 2011 I was invited up to St Andrews University to talk about MiHsC (The title of my talk: Can inertia be modified electromagnetically?). Their physics department has a superb reputation so I was a little nervous. I met some of the academics and sat down to have lunch with them and one of them asked incisively: "If MiHsC is true then why don't we see the inertial mass of something in a metal box reduce?". Well, at the time I nearly choked on my tuna sandwich, but in fact this was one of the first questions I asked myself in the early days, and, when I got my voice back, I explained that for the accelerations of objects we are familiar with, the Unruh waves are extremely long. For example, an object with an acceleration of 9.8 m/s^2 will see Unruh waves 7x10^16 m long (a few light years) and Faraday cages do not affect such long EM waves (submarines can receive long EM waves). Also, as some of the lecturers there helpfully pointed out: Unruh waves are not solely EM waves, they're waves in all the fields.
How about light in a box though? If you have photons in a box whose inner sides are mirrored, then they do contribute an inertial mass to the box because if you move the box one way, then the photons bash into the mirror on the backwards side and so contribute to the inertia that opposes the box's motion. So light or photons in a box, have inertial mass. The interesting thing about the photons in a metal box is their fast speed, so that the mirrors are forcing them to accelerate rapidly backwards and forwards. That means the Unruh waves associated with their inertia are now of a similar wavelength to the box's size and they can be damped by its walls, at least their EM component. So MiHsC predicts a loss of inertial mass for the light in the metal box, just the same as it predicts a much smaller loss of inertia for objects inside the Hubble volume.
Now what if the box is a cone? (EMdrive) The photons are resonating within it so the Unruh waves they see are of a similar size to the cone and typically fewer Unruh waves will fit or 'be allowed' at the narrow end of the box than at the wide end. One way of thinking about this is that the photons going from the narrow to wide end gain inertial mass in a new MiHsCian way. This turns out to be a bit like the old rocket method of blasting hot gas out of the wide end, but now we are blasting 'virtual' mass. To conserve momentum (mass*velocity) the whole system has to move the other way. Hence the typical motion of the Emdrive towards its narrow end. MiHsC predicts the results quite well without any tuning parameters, see earlier blogs or my paper on MiHsC and the Emdrive here (an introduction to MiHsC is here).
Note that, if the EM waves' frequency is tuned so that the Unruh waves fit better within the narrow end, then the Emdrive might actually move the other way, and it would be interesting to know whether this was the case for the recent NASA experiment where it did actually move the other way. I'm now working on a second paper, that takes into account individual Unruh waves, to be submitted..
Reference
McCulloch, M.E., 2015. Can the Emdrive be explained by quantised inertia? Progress in Physics, Vol. 11, 1, 78-80 (Pdf).