· Hullo A couple of weeks ago I made a video about the development of the space shuttle and its aerodynamics, and in the process I found this amazing photo, which I thought looked like the result of a computer simulation. But IN fact it’s a 1970’s photograph of a test of a shuttle model in a Mach 20 helium wind tunnel, using electron beam fluorescence. And there were a lot of questions about this - how do you make a hypersonic wind tunnel, why would you use helium and what’s the electron beam fluorescence all about. So… let’s start with How you make a hypersonic wind tunnel? If you’re only casually familiar with wind tunnels you might know enough that you think of them as a tunnel with big fans pushing air through them and a location to mount test objects so you can observe aerodynamic flow around them. The tunnel might be open to the world OR it could be arranged in a loop with the air circulating to reduce power needs, and there’ll be all sorts of vanes and grid type devices to direct and control airflow and try to make it as smooth as possible. Either way if you know anything about aircraft you’ll know that aircraft with propellers aren’t going supersonic, The Republic XF-84H Thunderscreech was an attempt to get a prop driven plane close to mach 1, but it was much more successful at converting aviation fuel into noise. Propellers are good at adding a small amount of speed to a large amount of air and getting supersonic flows might be possible but prohibitive. Anyway, a wind tunnel isn’t just about fans pushing air around, there’s a lot of other engineering required to make air flow the way you need it to, and a big part of wind tunnel design is the shape of those tunnels that the wind flows through. If you’re even casually aware of physics you’ve probably heard of bernouli’s laws, these are a set of relationships that explain how pressure and velocity of a fluid changes as it moved through pipes with differing cross sections. If a fluid flows from a wide pipe into a narrow pipe then conservation of mass requires that the velocity of the fluid increases so that the same amount of mass per second flows through the fat section and the thin section of the tube. And wind tunnels can use use this, after all a tunnel can just be considered like a pipe from the point of view of Bernoulli’s laws. So you’ll see wind tunnels with wide sections that narrow down just before the test section with the models, increasing the speed of the air where it’s tested. So you might think that’s the problem solved, just shrink the tube down until you get supersonic flows and shrink it more to get hypersonic flows. But this doesn’t work, the commonly cited version of bernoullis law is for incompressible fluids, and you can only get away with approximating gasses as incompressible up to speeds of about Mach 0.3. You can still make the air go faster, but it’s not longer a simple linear relationship between cross section and flow rates. The air gets compressed, the density rises and so you don’t get the same increase in speed that you’d expect in the incompressible regime. But if you push hard enough the flow reaches Mach 1, and pushing it even harder doesn’t increase the flow speed, it just increases the density, pressure and temperature. This is a condition called choked flow. So the next step is then to allow this gas flow to expand out of the choked flow condition through a widening tunnel section, and what happens is the stored energy in the form of pressure pushes the gas to higher and higher speeds, past mach 1 and getting faster as the cross section of the tunnel expands. So for supersonic gas flows the intuitive form of bernoullis law is reversed - increasing the area accelerates the flow and decreasing the area slows the flow. Now, many of my long time viewers might be thinking they’ve seen me explain this before, and you’re right, because it’s exactly the same phenomena used in rocket engine nozzles to maximize the exhaust speed. You have the high pressure combustion chamber, a throat to generate the choked flow and an expanding nozzle section to accelerate the gas. And then there is the example of the thermal structure test tunnel at Langley that wanted to simulate reentry conditions, and they did this by effectively building a hypersonic wind tunnel that was a methane fueled rocket engine, generating high temperature flow conditions at Mach 7. One difference is that the throat and expansion geometry in rockets has a circular cross section to reduce hot spots, but it’s common to see wind tunnels only contract and expand in one dimension, some of them even make the throat and nozzle adjustable in real time, others can vary it between tests by opening the tunnel up and replacing blocks that construct the flow. Anyway, we’re getting ahead of ourselves, with this supersonic expansion phenomenon you can now see how it’s possible to turn high pressure subsonic flows into supersonic, or even hypersonic flows. But in driving this you want high pressures, and fans aren’t good at that, so those get replaces by pumps to compress the gas more effectively. Sustaining large tunnels at high flow rates can require huge amounts of power - hundreds of megawatts, but with rapid response instrumentation you can perform all sorts of tests quickly. So many small tunnels are driven by compressed gas where there’s a high pressure reservoir that gets pumped up to pressure and then a valve is opened and the gas flows for a short time, sometimes minutes, sometimes fractions of a second. In this image these large spheres are actually vacuum spheres that the high pressure gas exhausts into, to get the best pressure gradient. So, now we’ve explained how you can make a hypersonic tunnel, why then is this tunnel using Helium instead of air? Well there is a downside to this expansion process, the expanding gas cools down, this is just conservation of energy as heat and pressure are converted into kinetic energy. The faster you go the cooler the gas gets. If there’s moisture in the air it’ll condense out, this is a similar phenomena to the condensation cones seen around transonic aircraft and rockets. Air in wind tunnels need to be dried to avoid this, and that can be a serious technical challenge if the tunnel has to continually draw in fresh air. But as you go faster the temperature can drop to the point where air itself liquifies, and that sets the limit on how fast a tunnel can simulate, for air this is between mach 4 and 5. One way around this problem is heating the air before the throat so it remains above the liquefaction point when expanded, it only takes modest heating to 50 Celsius to make a mach 5 tunnel work, but for mach 10 the air temperature needs to be about 900 Celsius and mach 20 would demand temperatures of over 4000 Celsius, necessitating some exotic systems for heating air and stopping it from destroying the tunnel. But another alternative is to use a gas with a lower boiling point, and Helium is about as good as you can get in that department. So you can push a helium wind tunnel up to mach 20 easily without spending all your time solving thermal management problems. Helium obviously isn’t a perfect stand-in for air, it’s monatomic which alters its physical properties compared to diatomic gasses like Oxygen and Nitrogen., but also realize that heating air to 5000 Celsius changes its chemistry compared to regular air so a mach 20 air tunnel wouldn’t be realistic either. Its one of many compromises engineers have to accept for tests like this. This isn’t the only way to solve these problems - shock tunnels, expansion tunnels, light gas guns, arcjets all offer different techniques, some of which address some parts of the problem better than others, there’s no one best technique short of actually flying a heavily instrumented test article in the test conditions. So anyway the final part is about the electron beam fluorescence which gives the image its beautiful appearance. Wind tunnels are used to analyse the aerodynamic qualities of models, and there are many techniques for measuring how the airflows are disturbed in the vicinity of the model. You might have seen some schlieren imaging which captures density changes in the flows by exploiting the changes in refractive index of the gas, under the right conditions I’ve even seen transonic shocks on top of wings on passenger planes. However in the case of these hypersonic helium tunnels this doesn’t work well because helium has a much lower refractive index than air, and, the process of expanding the gas to get hypersonic flows means the gas density is lower, and the low density lowers the refractive index even more. So instead an electron beam is fired through the helium and this excites the atoms into a higher energy state, then when the atoms drop back to their base state they emit light. Now in a gas the rate at which the atoms do this is higher when the gas is higher density. So, the result is a brighter glow in higher density regions. Depending on the test it’s also common to include small amounts of nitrogen which does a much better job at emitting light and illustrating the changes in air density near an object. So, now you know what’s going on in this 50 year old photo - and hopefully you‘rave learned a bit about hypersonic wind tunnels in the process. I'm Scott Manley, Fly Safe
