> The arrows literally moved down! On the top part, you've got a supersonic free-stream deflected with an expansion fan to a supersonic parallel flow over a pl

> The arrows literally moved down! On the top part, you've got a supersonic free-stream deflected with an expansion fan to a supersonic parallel flow over a plate, deflected back and slowed-down to free-stream conditions through an oblique shock. The only thing the upper-surface "sees" is a parallel, supersonic flow. On the lower part, you've got a supersonic free-stream deflected to a lower-speed supersonic flow through an oblique shock, creating a parallel supersonic flow over the lower surface of the plate, deflected back and re-accelerated to free-stream conditions through an expansion fan. The only thing the lower surface "sees" is a parallel, supersonic flow. Now, unless you can come up with a force component emanating either from the oblique shocks or from the expansion fans and contributing to the lift vector, it fair to say that the flow deflection is not directly what is causing lift on the angled plate.