"Breaking the sound barrier" was a popular theme as planes flew faster and faster in the late 1940s. It turned out that wind tunnels also ran up against a sound barrier of sorts. At that seemingly magic speed, the velocity of sound, strange things begin to happen. In a wind tunnel, for example, as more and more power is applied to the fans, airflow in the narrowest part of the test section chokes up at Mach 1, the speed of sound. No matter how fast the driving fans turn, the air velocity in this part of the test section remains at Mach 1. The brute-force approach does not work. The same sort of choking occurs in the narrow throat of a rocket engine. Nevertheless, the hot exhaust gases of rocket engines travel faster than sound. They accelerate past Mach 1 as they expand in the rocket engine nozzle. Supersonic wind tunnels employ the same nozzle expansion to reach supersonic speeds.
Apparently contrary to logic, the test models in a supersonic wind tunnel are mounted downstream of the throat section where the choking occurs. Here, in the nozzle, the cross- sectional area of the tunnel is increasing. However, the velocity of the air is not decreasing, rather it is accelerating as all the energy pumped into the air by the fans and stored in the forms of compression and heat energy is converted to kinetic energy. The rocket engine works the same way except that the energy is added by burning fuel rather than by fans. Airflow becomes supersonic once it passes the throat or point of smallest cross-sectional area. This fact of thermodynamics leads to the apparently contradictory situation in which test models are placed at the narrowest part of a subsonic tunnel (where airspeed is logically the greatest) but downstream from the narrow throat of a supersonic tunnel (where common sense says airspeed should be slowing down).
The nozzle or expanding portion of the supersonic test section has a unique shape for each value of the supersonic Mach number. The ratio of test section area to throat area is 1.69 for Mach 2 and 536 for Mach 10. Thus, to encompass a range of different Mach numbers, the shape of the nozzle in a supersonic wind tunnel must be variable. This can be accomplished by interchangeable nozzle blocks, flexible nozzle walls, or some variant thereof. This required change in nozzle shape is the first of three major distinctions between supersonic and subsonic wind tunnels.
The second important difference between subsonic and supersonic tunnels is the magnitude of the energy losses in the air circuit. In subsonic tunnels the fans need only increase air pressure a modest 10 percent or so to compensate for the energy losses induced by the tunnel walls, models, apparatus, turning vanes, and so on. In a Mach 2 tunnel, however, the fan pressure must be increased by approximately 100 percent. Thus the simple fan becomes a compressor consisting of several stages of fans. A Mach 5 tunnel requires a pressure ratio of about 20, necessitating several multistage compressors in series.
Obviously, a much larger amount of power is consumed by these big compressors than by the simple fans in subsonic tunnels, suggesting that the flow losses around the circuit of the supersonic tunnel are much higher for some reason associated with supersonic aerodynamics. The reason is that tremendous energy losses occur in the shock waves immediately downstream from the test section, where the mainstream air decelerates from supersonic to subsonic speeds. These shock-wave energy losses are inherent in all supersonic flow. In the supersonic wind tunnel, the electrically driven fans or compressors must supply this extra energy.
The third and final important engineering difference between subsonic and supersonic tunnels involves the tunnel air itself. Not only must it be clean, that is, free from oil, vapor, dust, and foreign particles, but in addition condensation of the contained water vapor must be avoided. As the tunnel air expands in the nozzle and latent heat is turned into kinetic energy, air temperature falls. Condensation of contained moisture is very likely, but condensation can be avoided by drying the air to extremely low dew points (e.g., -100° F).