A method is presented for calculating the stresses produced by rectangular cutouts of any size in torsion boxes. The problem is divided into a "box problem" and a "cover problem." The box problem is a special case of the general method of analyzing torsion boxes without cutouts. In the cover problem, simple shear-lag theory is used to obtain "key" stresses; the final distributions are obtained from these key stresses by means of simple rules or empirical distribution curves. Comparisons with the results from three series of tests in which the dimensions of the cutouts varied over a wide range are shown.

The problem treated in this paper is that of the stresses in a torsion box with a large rectangular cut-out. The theoretical treatment is confined to stresses termed the "primary stresses". Comparison of the theoretical results with strain-gage data for a series of cut-outs indicates that the primary stresses are probably adequate for designing the major part of the structure, the only important exception being in the design of the cover sheet in the full section adjacent to the cut-out.

Stresses in the covers of torsion boxes containing large cutouts are found by changing half of the symmetrical uncut portion of the cover into three stringers carrying direct stress and intermediate cover material carrying shear stress. Illustrative examples are solved and the results are compared with experimental values. For one example, the results are compared with a more detailed numerical method of analysis. The agreement between theory and experimenbt is satisfactory in all cases except those with very large cutouts and flexible bulkheads.