LOCAL WIND FLOW MODIFICATIONS WITHIN THE VICINITY OF A COMMUNICATION TOWER

The boom length obtained from the International Electro technical Commission (IEC 61400-12-1) standard is based on the centreline velocity deficit expression derived on the assumed incident wind that is perpendicular to the tower face, giving rise to a velocity deficit value that is predicted using the upstream contour profile of the modified flow around the tower. The boom length computed thus does not guarantee that wind observation is accomplished within the recommended industry-accepted accuracy of < 1 % errors between 0° and 360°. In this study, the local flow modification around the 3D complex geometry of two communication towers of different construction details located at Amper-bo and Korabib in Namibia were performed using Ansys Fluent flow solver in other to introduce the safe angle range, where boom length computed based on the IEC reference direction (θ = 0º) is long enough to achieve 99 % of the free stream velocity. In this regard, a full 3D CFD flow analysis was performed to obtain the maximum boom length based on the two commonly encountered boom arrangements (perpendicular and parallel to the face of the tower). The boom length obtained based on the IEC standard boom configuration (θ = 0º) is long enough to keep the speed sensors out of the tower wakes if winds arrive at the tower at incident angles θ = < ± 70º with respect to each boom configuration reference direction (i.e., θ = 0º), which defines the safe angle range. Winds that arrive at the tower between ± 90º ≥ x < ± 150º with respect to each boom arrangement reference direction cast their full shadow on the speed sensors. A boom length of approximately x ≥ 7.5 m and x ≥ 6.5 m at Amper-bo and Korabib respectively are needed to keep the speed sensors out of the tower wakes. However, the suitability of a slender boom of this length which is susceptible to vibration and therefore can introduce more errors to the readings of the speed sensors mounted on them is questionable. A shorter boom length of x ≥ 4.5 m to x ≥ 3.5 m or less at Amper-bo and Korabib respectively is needed to keep the sensors out of the tower wakes when the incident wind angles are within the range ± 150º ≤ x ≤ 180º from the reference directions. This type of arrangement is not encouraged because any slight shift in the direction of the upstream wind will cast shadows of the tower on the speed sensors. The study further revealed that the boom length computed based on the two popular boom arrangements is independent of the scale and turbulence models used in CFD flow analy sis. Also, complex flow interference was reported around the tower structure with no appreciable increase in boom length due to the presence of the secondary support structures. In the second section of the study, a correction method based on the combined field observed wind data is proposed. Based on the accuracy of the correction method, it may no longer be necessary to discard the tower wake affected direction sectors in wind analysis. Finally, the speed sensor to tower separation distance needed to keep the speed sensor away from the tower-induced flow defect was thoroughly investigated using lattice communication towers of different construction details and booms of different arrangements on the towers. A safe angle range (θ = < ± 70º) was proposed. Winds that arrive at the tower at other angles outside this range will likely induce error readings on the speed sensor mounted on the boom and this explains why most towers instrumented according to the IEC standard still fall short of the 1 % speed deficit recommended.