I can't add much to this, besides the point that in my experience with expansion tunnels [1][2] from my time at uni, they are very short-term test facilities, l

I can't add much to this, besides the point that in my experience with expansion tunnels [1][2] from my time at uni, they are very short-term test facilities, lasting for fractions of a second. They work by compressing a big vessel of air with a piston, rupturing a diaphragm, and then blowing that highly compressed air through a expansion nozzle (which, unlike subsonic flow, accelerates the flow velocity through the nozzle) before passing it over the test piece. I'm not familiar with the APTU, but it would not surprise me if this is a similar concept, wherein it delivers a short-term high-velocity flow across a test object. I think the main difference is it heats with a combustion air heater, rather than pressurising with a big slug. In any case, because it's quick and requires a small cross-sectional area, much of the surrounds of the machine are all sensors to measure everything, since you don't want to have to do it over and over again. I think that is much of what can be seen in the photo of the APTU. [1] https://en.wikipedia.org/wiki/Expansion_tunnel https://en.wikipedia.org/wiki/Expansion_tunnel [2] http://hypersonics.mechmining.uq.edu.au/facilities http://hypersonics.mechmining.uq.edu.au/facilities (specifically X2 and X3)