A topic in the Open Knowledge Graph — a free, open map of 15,290 topics and the order to learn them in.
The drag coefficient C_D of a bluff body (one whose shape causes flow separation and a large wake) is dominated by pressure drag rather than skin friction drag. For canonical shapes — spheres, circular cylinders, flat plates — C_D depends primarily on Reynolds number and body geometry. At low Re, C_D decreases monotonically (Stokes drag gives C_D = 24/Re for a sphere). At intermediate Re, C_D plateaus (roughly 0.4–0.5 for a sphere, ~1.2 for a long cylinder). At the critical Reynolds number (~3×10⁵ for a smooth sphere), the boundary layer transitions to turbulent before separation, the wake narrows dramatically, and C_D drops by a factor of 3–5 — the drag crisis. Streamlining a bluff body delays separation and reduces pressure drag, though it increases wetted area and therefore friction drag; the net effect is almost always a large reduction in total drag.
Plot C_D vs. Re for a sphere, cylinder, and flat plate on the same log-log chart and compare the curves. Identify the Stokes regime, the plateau, and the drag crisis on each. Calculate the terminal velocity of a falling sphere at different sizes to see how the drag coefficient regime changes the answer. Then compare the drag of a streamlined strut to a circular cylinder of the same frontal area to quantify the benefit of streamlining.
You already know that the drag force on an object is F_D = ½ρV²A·C_D, and that the Reynolds number Re = ρVL/μ measures the ratio of inertial to viscous forces. For bluff bodies — shapes like spheres, cylinders, and flat plates oriented perpendicular to flow — most of the drag comes not from skin friction but from the low-pressure wake behind the object. When the flow separates from the surface, it leaves a large, chaotic recirculation zone. The pressure there is much lower than at the front stagnation point, and this front-to-back pressure difference is pressure drag (or form drag). The shape and size of the wake, and therefore C_D, depends strongly on where and how the flow separates — which is why Re is the controlling parameter.
At very low Reynolds numbers (creeping flow, Re < 1), viscous forces dominate and Stokes derived analytically that C_D = 24/Re for a sphere. As Re increases into the hundreds and thousands, inertia begins to dominate and separation moves earlier on the body's surface. C_D transitions from the Stokes regime into a broad plateau around Re = 10³–10⁵ where C_D ≈ 0.4–0.5 for a sphere and ≈ 1.0–1.2 for a long circular cylinder. In this plateau regime, the laminar boundary layer separates near the widest point of the body, creating a wide, low-pressure wake. This is the regime of most practical engineering interest for moderate-speed objects — falling raindrops, structural cables, sports balls at ordinary speeds.
The counterintuitive exception is the drag crisis, which occurs around Re ≈ 3×10⁵ for a smooth sphere. As Re climbs, the boundary layer itself transitions from laminar to turbulent. A turbulent boundary layer has more momentum and is more resistant to separation — it can "hug" the body further around its back side before detaching. The result is a much narrower wake and a dramatic C_D drop, sometimes by a factor of four. This is the principle behind golf ball dimples: the surface roughness triggers the turbulent boundary layer transition at lower Re than on a smooth ball, putting the ball into the low-drag post-crisis regime throughout its flight. It is crucial to understand that turbulence here *reduces* drag by delaying separation — the opposite of the naive expectation that turbulence means more drag.
Streamlining addresses the same problem from the opposite direction: instead of tricking the boundary layer into staying attached, you reshape the body so there is no abrupt adverse pressure gradient that would cause separation in the first place. An airfoil or ship hull tapers gently to a fine trailing edge, giving the flow a smooth path around the body. The pressure drag is nearly eliminated. The trade-off is more wetted area and therefore more skin friction drag, but for moderate to high Re this exchange is always favorable — a well-streamlined body can have C_D 10× lower than a bluff body of the same frontal area. The engineering lesson is that any time you have a bluff structure (a strut, a wire, a tank) in a high-Re flow, streamlining it or rotating it to reduce its projected area offers large drag reductions, while surface roughness effects are secondary at those conditions.