Using a popular airfoil analysis code (XFOIL, by M. Drela & H. Youngren, see http://web.mit.edu/drela/Public/xfoil/) we analyzed the performance of the stock centerboard of a San Juan 23. As this board was a thin flat plate, this performance was rather poor, consistent with the excessive leeway (115? tacking angles) and noticeable lee helm of the vessel. A thicker, more effective centerboard would not fit the narrow centerboard slot of the vessel. However, again using XFOIL, we developed an add-on leading edge for the stock board, which greatly improved the upwind performance (95? tacking angles). Having the add-on leading edge remain outside the slot when the board was raised, allowed that edge to be thicker than the board. As shown in Figure 1, the leading edge is elliptical, making the resultant centerboard cross section tadpole-like.
We present here selected XFOIL results for the stock centerboard, a NACA 0010 cross-section, and the tadpole cross-section. The first three figures show pressure coefficient data for a 6? angle of attack and a Reynolds number corresponding to 5 knots. These figures also reflect forced transition to turbulent boundary layers at 5% of chord (1% for the stock case). Both the NACA and tadpole cross-sections perform well with either forced or computed boundary transition. Forced transition merely reflects the reality of non-polished board surfaces. For the stock board, however, it is only by such measures that the board is not completely stalled at more than a 3? angle of attack.
Figures 4 and 5 present lift and drag data for the three cross-sections, again assuming 5 knots and forced transitions.
As shown by Figures 5 and 6, the tadpole cross-section significantly outperforms the stock flat plate and nearly equals the performance of the NACA0010 section.