I presume these vanes are intended to reduce or eliminate swirling flow in the suction piping to the pumps. Whether suction nozzle orientation is top, bottom or side shouldn't matter with deswirl vanes. What is important is that enough is known about the the swirl intensity, particularly the the swirl flow angles across the radius of the suction piping just upstream of the Cheng vanes. My recollection is that these are not variable pitch vanes but rather have a fixed pitch that requires knowledge of inflow incidence angles for proper setting. The hydraulic forces involved with radial inlet flow splitters or non-radial deswirl vanes can be damaging to the vanes if fixed vane angles are mismatched with the inlet flow angles. Generally, it requires pitot tube velocity traverses across the pipe inside diameter to correctly determine the inlet flow conditions. Broken vanes pieces entering a pump impeller could have disastrous consequences. If vane trailing edges are located too close to the impeller blade inlet edges, then vortices shed from deswirl vanes interacting with impeller blades can cause or add to impeller blade passing vibration levels with potentially adverse consequences. Flow blockage effects can be a concern for some pumping applications like primary reactor coolant pumps. Cheng vanes were water-flow tested for deswirl of pump piping multiple bend induced swirl flow in such an application several years ago by a Westinghouse division. I understand they were deemed to be unsatisfactory for the application but I don't know what the criticisms were.