Chapter 5: High-speed Cowlings, Air Inlets and Outlets, and Internal-Flow Systems
[144] The NACA cowling was an obvious subject for research in the new 8-foot high-speed tunnel in 1936. It was the center of much attention in the NACA program and the pressure distributions obtained in the PRT by Theodorsen's group revealed local velocities as high as twice the flight speed for the blunter shapes. Using Jacob's criterion, we estimated that the critical speeds would be as low as 300 mph at altitude, well within the performance spectrum of pursuit aircraft of that period (ref. 175). R. G. Robinson planned a high-speed program which would start by testing five of the cowlings used in Theodorsen's program and then proceed to develop improved less-blunt shapes. I was fortunate to be project engineer on the cowling investigation, my first substantial project assignment.
During design of the cowl models I had noticed one of the then-new [145] DC-3's parked on the Army flight line. It had a very blunt cowl shape and a large fixed exit opening. I picked up some straightedges and clamps from our shop and walked over to the flight line where I found a sergeant servicing the DC-3. He located a ladder and helped me set up my equipment to obtain accurate profile ordinates at several stations on the DC-3 cowls. They turned out to be intermediate to two of the blunter Theodorsen cowls and were therefore not included in our test program. A couple of years later, however, we used them in our DC-3 test project (ref. 31). This is a small illustration of the direct informality with which things were done in those days.
Tests of the blunt cowlings confirmed our low critical-speed estimates, and showed prohibitive drag increases beyond the critical speeds (ref. 173 and fig. 37). They also provided the live demonstration for the 1937 Engineering Conference previously described (page 26).
The next phase of the cowling work provided my first experience at Edisonian tailoring of an aerodynamic body in an effort to obtain a particular pressure distribution. The basic difficulty in subsonic flow...

engine cowl assembly being tested in a wind tunnel
FIGURE 37.-High-speed cowling test setup in the 8-Foot High-Speed Tunnel, 1937.
[146] is that a change in local shape and pressure also causes usually unwanted changes elsewhere on the body. After testing a given profile, I would lay out a speculative change, carry the model back to our machine shop, and often help the machinist in laying out a new nose profile template. When the new shape had been tested and the pressure distribution studied, the process was repeated until eventually we obtained the lowest suction pressure peak possible within the dimensional limits established for the cowling study. These limits were set by the specified diameter of the inlet opening and the distance between the cowl lip and the plane of the engine face which were typical of advanced engines. The resulting high-speed shape, designated cowling "C," had a critical Mach number of 0.64 or 480 mph at sea level, about 170 mph faster than the blunt cowls. The "C" cowling also had somewhat lower drag at subcritical speeds than the blunt shapes and it found use on a wide variety of piston-engine aircraft, including several where the high critical speed was not required.