Today at 5:00 p.m. I was keen to get home and begin cooking dinner but I just wanted to wait for my colleague Robin to finish processing a set of data that we had begun at 7:30 a.m. this morning. I watched as the long awaited data point appeared on the screen – exactly at the position we had anticipated.
- Our prediction was based on mathematical analysis of the sound field inside a slightly non-spherical cavity, and the effect we were measuring was caused by the fact that at lower pressure molecules travel for longer distances before colliding with other molecules. At half an atmosphere they travel almost 0.001 mm before colliding.
- Our measurements are measurements of the frequency of sound resonances inside a beautifully-crafted, not-quite spherical cavity. The experiment involves flowing ultra-pure and ultra-dry argon gas through the cavity while measuring with microwaves and microphones and pressure meters and flow meters and thermometers. We had to measure more than a hundred resonances in order to be able to average them well enough to see this tiny shift in frequency – it amounts to only around 5 parts in a million of the speed of sound. If we were measuring a length, the shift would be around 5 millimetres in a kilometre – and we could tell that it was 5 millimetres and not 4.9 millimetres.