Earthquakes are not the only thing that can cause a building to move. Simple expansion and contraction as a building heats and cools can move walls around. Not so you’d notice with the naked eye, but certainly noticeable if you have mirrors attached to the walls to guide a laser beam around a lab.
That’s what we have in one of our teaching labs for measuring the speed of light. We use a time-of-flight approach – basically we modulate the intensity of a laser beam so that it flips on and off a few million times a second – we send the light on a fifty metre path round the lab, and time how long it takes to complete the path. There’s a little more to it than that, but that’s the guts of it.
But the problem we have in our new lab, that didn’t affect us too badly in the old one, is that the mirrors on the walls keep moving. It’s a real pain; one week the equipment can work beautifully, the next week it is almost useless. To adjust the mirrors involves getting up on a ladder – while at the same time someone else watches an oscilloscope output – that’s not something we’re going to do for the sake of a couple of undergraduates getting a good result. The movement in the walls is tiny, but it is enough to mean that the returned laser beam no longer hits the detector.
The equipment is also complicated by the fact that the heating for the lab (and this lab is COLD) consists of ducted hot air blown out of the ceiling, straight above where the laser beam goes. Putting turbulent hot air into the beam doesn’t do much for its ability to travel in a straight line and remain a tight beam. Hot air has a different refractive index to cold air, and this means that the beam takes a bit of a shaky path from one mirror to the next, and spreads out a bit enroute; neither of which is particularly ideal. The solution to this one is relatively easy – let my students freeze – but its not a popular one.
Today, the experiment is working just fine, so I’m happy. What it will do tomorrow is anyone’s guess.