I have posted the co-ordinates for the best L/D rudder section I can come up with for AR of 4 and Re# 10E6:
http://www.rickwill.bigpondhosting.com/NACA_07-310.txt
It is worthwhile noting that if the AR is not limited structurally or by draft then the best L/D section will get thicker. So if someone wants to play and can handle higher AR then I can go through the exercise.
The best L/D for this section occurs between 0 and 0.5 degrees AoA.
It is VERY important to realise that they need to be able to have relative angle trimming. If the boat goes best on the wind with the combined trim greater than 1 degree into the windward side then the blade area is likely too small to get the best L/D. Typically these rudders will have the combined area similar to a centreboard and rudder of other boats with the same sail area. With the Harry configuration there should be a tendency to round-up with no rudders until the ww hull unloads but these rudders are not going to have much feel so there may be no indication of weather helm. So it may require actually observing the angle of trim between the rudders.
The aim should be to trim the rudders to eliminate leeway. Current aside, the boat should go where it is pointed. So the idea would be to set a course and then slowly trim the rudders relative to each other while maintaining the course to get the best speed. Once this has been done a couple of times the best settings will become intuitive for any particular conditions.
These rudders will not have a cavitation issue for any reasonable speed expectation - Cp less than 0.5 in best working region. If thicker section for higher AR then cavitation would need to be checked and the vortex shedding will be more significant.
The vortex shedding should not be an issue on this section either as the separation is only on the last 1 to 2%.
Rick