Surface Roughness Application in Star CCM+?
 

Hi, i'm currently simulating flow over a seamed cricket ball in Star CCM+ as part of a University project. One section of my investigation involves modelling surface roughness and the effect it should have on aerodynamic side-force generation by the ball. I've therefore configured the ball to be pointing directly forward (so that the prominence of the stitched seam doesn't result in any flow asymmetry and induce a side-force) with one side of the ball left smooth and the opposite side assigned a surface roughness of 0.3mm (ball diameter = 0.072m) within the Star CCM+ model (as opposed to modelling roughness explicitly).

I ran the simulation as described above at two speeds (24m/s and 42m/s) which theoretically should produce differing results (the low(er) speed should result in a sideforce toward the rougher surface whereas the high(er) speed should result in a force toward the smoother surface). Unfortunately, my results oppose this theory and predict the opposite. Also, when evaluating the skin friction coefficient along the ball's surface, the roughness appears to have made no difference toward the skin friction (whereas in a previous section of my investigation - looking at an angled seam rather than surface roughness - a similar amplitude of force yielded a significantly noticeable contrast). My supervisor advised me to be cautious using the surface roughness application within star ccm+ as, in his words, it alters the log-law region of the boundary layer. I'm struggling to suggest why this would provide me with seemingly inaccurate results; does the lack of influence on the viscous portion of the boundary layer result in unaffected skin friction? Why might this calculate a force opposed to that predicted by theory (specifically theory pertaining to viscous properties and the influence of a perturbation on the boundary layer separation)? I apologise if my description is unclear / incomplete but i'm struggling to get my head around the specifics of how the star ccm+ surface roughness application functions and how that affects my investigation.

PS: Using a low y+ grid and a RANS solver (SST + Gamma ReTheta transition).

I'm not really understanding what's going on here.

You've been told by your supervisor to investigate surface roughness to which he advises you to not use the surface roughness model?

I think you need to review literature on what it means to have surface roughness. If the surface is rough, why are you resolving down to the viscous sublayer? It doesn't make any sense to do so.