For the F20c, there is a definite "twist" to the boards. It's not a straight curve...
Not all curved foils were born equal.
Symetrical section curved foils with different angles of incidence along their span on the F20C, compared to asymetrical section curved foils with fixed angle of incidence along their span on the french maxitris.
It will take a while to fine tune curved foils in cats. And never forget the "S" foils that were barely tested.
Source? Some say the foil is symmetric, some say asymmetric on the F20c. What is on the new F17?
You can get by with 2 mold halves if twist is 0. Otherwise, forget about it. Doubt they are doing that in a production environment anyway, I am sure there are 4 mold halves at an easy $40k for the set.
If you have twist, which is one method of getting elliptical lift distribution, things get complicated really fast. For starters, simply modelling the foil in CAD becomes a nightmare-try 100+ hours of CAD work for the plug and then another 40 or so hours for the molds themselves.
I also realized the broken Nacra foils I own have Rohacell or similar high density foam cores. If the curved boards are similar, you would need to machine this core prior to putting it in the molds. That's probably $200 worth of machine time per core if you own the machine.
One way to look at this is by cost, and surely a foil that costs 30% more isn't going to make your boat 30% faster, hence its not worth it. IMO, $1300 is already a lot of money for a daggerboard, why not just pay the extra $400/board for a curved foil? Its a 3% increase in cost on a brand new F18, for example.
