Dear Dennis,
I counted out, how high can be the dynamic pressure at the inlet of the uniflow tube. Dynamic pressure is: 1/2 specific weight of the air, multiplicated by the speed up to the square. (0.62 kg / m^3 * 27^2 m^2 / sec^2 = 452 (kg m / sec^2) / m^2 , (in other words: N / m^2) or Pascal (Pa). We need to know that the environmental normal athmosferic pressure is round 100.000 Pa. 452 Pa is almost nothing, less than 1/2 % of it.
When you move the tank 1 centimeter towards to the center of circle, the difference of static pressure of the tank, looking from the aspect of spraybar hole, difference will be 0.01 m * 320 m / sec^2 * 700 kg / m^3 = 2240 Pa. Five times more than dynamic pressure making by the airflow! When you put some "shade" ( fender, deflector, obstacle etc.) in front of the inlet of Uniflow tube, eliminating more or less the dynamic airpressure, that equals as moving 1/5 cm = 2 millimeters the tank into outer direction. I am convinced the upwind or downwind does not make nothing, since the inlet is in direct blowing of the prop. I think there is no reason to do nothing.
Some plus: When the model flies round loop at 9 meters radius, 0.8 millimeters of vertical moving of the tank will be equal the full dynamic airpressure...
Excuse me for using metric units, do not think me unpolite, but it is a great advantage to use metric units in physical equations to check the correctness / logics of the equation. E.g. when you know, that the result must be a pressure value, Pa, or N/m^2, you must get finally (kgm / s^2 ) / m^2 , or simply kg / m s^2.
You asked my tank inside arrangement, that is more than simple: Baffle is at 70% longitudinal position of the tank, no holes in it, but both sides are 4 millimeters shorter than the full width of the tank. Fuel line is beginning 4 millimeters aft, uniflow tube ends 4 millimeters before of the baffle.