Here's a shit diagram of an alternate set up. It's a top view. using 2metre full length mirrors, 1 metre apart, if you have a reflection every 1cm thats 200 reflections >200m, every 5mm, thats 400 >400m.
The mirrors can be aligned during operation, and it's stationary so no need to worry about misalingning due to movement.
The blue line is a piece of graph paper, the grey square is a flat surface teh laser is just above, a negative result would be unabiguous, if the lowest spot possible for the laser to hit is the same after 400 reflections is the same as with no reflections then thier is no EA. if however it can't hit the same spot, then either there is an EA or the equipment in not well enough aligned.
In the other experiment, there is ambiguity both ways, as there is the posibilty of the mirrors shifting when it is moved. With this tho, a negative result would be conclusive.
in the side view the red line is the laser with no EA and the green is it with an EA.

for a mirror 1m apart
100 reflections give 1mm deflection.
300 reflections is 5mm
400 reflections is 1cm
500 reflections is 2cm
of course the more reflections the more accurate it needs to be to get the result.