R interest in thrust vector control with rigidly mounted thrusters has been revived. The results described in this Note are associated with the side-force generated by a supersonic jet with an exit plane which is oblique to the centerline of the exhaust flow. As shown in Fig. 1, depending on whether the flow is overor under-expanded, the direction of the exhaust flow is deflected either toward or away from the normal to the exit plane by an amount related to the pressure imbalance and the projected area of the exit plane in the lateral direction. Therefore, control over the direction and amount of side-force is available by varying the stagnation pressure and exit plane obliquity. Although the present results are associated with a twodimensional configuration, with the exit plane obliquity established by extending one wall of the nozzle, application to an axisymmetric configuration is conceivable. For example, consider a fixed axisymmetric thruster enclosed by a cylindrical housing which is split longitudinally into various sections (see Fig. 1). At the null position, the end of the cylinder coincides with the exit plane of the thruster. As various combinations of cylindrical sections are translated forward of the exit plane, the radially projected area of the exit plane may be oriented toward any azimuthal angle. The amount of projected area and its orientation may also be changed rapidly, which adds another control feature to the thrust vectoring. The focus of the work described in Ref. 2 involved the extent to which viscous effects affected the performance, particularly at over-expanded conditions. A two-dimensional supersonic jet of aspect ratio (height/width) equal to 1.5 was established over the Mach number range 1.0 to 3.0. One wall of the nozzle contour was then extended a specified fraction of the nozzle height to establish an oblique exit plane. Transparent and parallel lateral sidewalls bounded the nozzle and extensions thereof through which the flow angle could be observed. The side force generated was measured with a strain gage balance to an accuracy of ±2% of the reading, based on the scatter in repeated experiments and comparison to the measured flow angle. In order to assess the degree to which viscous effects impede accurate analytical prediction, the amount of asymmetry was limited to maintain the wall extension within the domain established by the Mach wave originating from the opposite nozzle wall at the exit plane. Figures 2-4 depict the forces measured on three nozzles whose exit plane is offset by an amount 1.0, 4/3, and 5/3 times the jet height, respectively. Comparison with predictions of two-dimensional inviscid flow theory is shown also. These results may be converted into force coefficients by dividing the results by the surface area of the wall extension and the exit dynamic pressure. The corresponding plot of the Fig. 3 data is presented in Fig. 5. These results indicate that, even in the simple case established here, nontrivial deviations between inviscid flow theory and experiment are observed, although the general trends predicted by analysis are correct. It is also clear that the vectoring control may extend well into the overexpanded region before the nozzle boundary layers begin to