...
Wyngaard (JAM, 1981) calculated the flow distortion caused by a sphere with a potential flow (laminar) model. Note that for a turbulent flow, this model is adequate only upwind of the sphere. Downwind of the sphere, the turbulent wake complicates the flow around the sphere (not to mention periodic vortex shedding of the wakes). Upwind of the sphere, the distortion of the mean flow is on the order of 3/2 (a/r)^3 where a is the radius of the sphere and r is the distance from the center of the sphere. Thus for a 36" diameter sphere, the flow distortion is estimated to be 1e-3 on the order of 6e-4 in the worst case and 7e5e-4 in the best case. These are certainly acceptable levels of flow distortion, but these estimates are only valid upwind of the sphere. For a 24", or 600mm, diameter ball the flow distortion is decreased by a factor of (2/3)^3 = 0.3.
The most desirable compromise would be to place these 36" diameter balls only on the two guy wires 'opposite' to the sonic, i.e. those at 90 degrees and 150 degrees from the sonic boom. The next best would be to use three 24" diameter balls.