 Göttingen 387

Göttingen 387 is a named cambered airfoil profile defined by tabulated upper- and lower-surface coordinates. The number identifies a particular section; it is not a general formula from which arbitrary airfoil camber or thickness can be read as decimal digits. A primary NACA investigation reproduces specified and measured ordinates for this section, making the geometry used in that experiment explicit.^[ref]

A practical reconstruction begins with Göttingen 387 coordinate normalization. Coordinates stated in percent of the reference chord become lengths by multiplying by $c/100$. Preserve the table's origin and reference line: a vertical coordinate offset is not automatically an airfoil camber measured from a newly chosen chord line. Distinguish the specified design coordinates from measured model coordinates rather than mixing their columns.

The section provides a concrete example of why a profile name alone does not specify a wing experiment. Ward's 1931 apparatus used a rectangular duralumin model $5$ inches in chord and $30$ inches in span, tested at an average Reynolds number of $3.4\times 10^6$ in the variable-density tunnel at a working pressure of $20$ atmospheres. The model was tested alone and with struts mounted at different upper- and lower-surface positions.^[ref]

From those dimensions, the model's span is $0.762\mathrm{m}$, chord $0.127\mathrm{m}$, area $0.096774{\mathrm{m}}^2$ and aspect ratio $b^2/S=6$. An original coefficient conversion illustrates the reporting convention: if a test of that geometry at stipulated dynamic pressure $q=1000\mathrm{Pa}$ gave $C_L=0.8$, then $L=qS{C}_L=77.4192\mathrm{N}$. This coefficient and dynamic pressure are illustrative inputs, not values transcribed from Ward's tables.

The reported comparison found that attaching struts changed the aerodynamic behavior, with larger effects for the tested upper-surface arrangements. The strut-interference drag could be comparable with the isolated-strut contribution.^[ref] That observation concerns the complete tested arrangements, not an intrinsic drag coefficient of every object carrying the same airfoil profile.

A proposed reproduction records the coordinate source, manufactured surface, span, end conditions, Reynolds number, angle of attack and support arrangement. Measure the wing alone before and after adding a support, and keep reference area and tunnel corrections consistent. A finite-wing result, even after a wall correction, should not be silently treated as an isolated two-dimensional section polar.