When sailing my latest model i noticed (with some fear!!!) that it made a lot more leeway than my old advanced sharpie proa. At first i thought it might be because of the wrong way the main hull (vaka) leaned. It was leaning in toward the wind by a couple degrees, not very much.
I still hope that is the case, but im starting to think the vaka has to little rocker to give the lift it needs to counteract the leeway.

Im using my Paradox-plans as some kind of realitycheck. The paradox has 4,2/0,23 = 18,26:1 in length/rocker ratio. My latest vaka has 14,3/0,45 = 31,77777:1 in length/rocker ratio. I know its not correct to compare a very slender vaka with a sharpie-monohull, but i need some kind of reference when trying to understand whats going on.
If i scale the rocker to a Paradox it would only have 13,2 cm rocker instead of 23 cm. I dont think it would work at all. I know it a very flawed comparison, but as i said before, i need to have some kind of reference. What im doing is very new and i cant lean on othe peoples work very much.

The rocker of the big flat bottom and sides establish a even flow of water around the hull, with very little pressure-difference between the bottom and the sides. It should work much like Matt Laydens Paradox and Enigma - type of boats, where the hull is a very low aspect lifting foil. Because of the much more slender shape of my vaka i believe i dont need any chinerunners, and my previouse model confirmed that way beyond any doubt.
I have never seen a sailboat without high-aspect finkeels make that kind of very shallow angle towards the wind.
I have seen it do 30 degrees uppwind in flat water, with great speed. I was very suprised when my new model was much worse, with lots of leeway. Its strange.

Johannes.

Johannes,

it might have to do with the sail area/immersed hull cross section.  if the model is very light and so has very little hull in the water this might occur.  I think more rocker might tend to favor a better ratio in this regard, or maybe it bights “deeper” where the water isn’t so easily moved.

Tom

Thanx Tom!

I tested with 0,5 kg rock in the Vaka and it sailed much better. The leeway was much reduced. It tracked much better.  I rebuilt the the crossbeams so the vaka and mast did not heel anymore and that made some difference to.
Its still not as fast as my old vaka though…. Strange!

I still think there is an optimum amount of rocker, and this one has to little. I should have built it with the same rocker as my old one. I have to test with a slightly larger sail and bigger and heavier outrigger before i can say anything conclusive about this.

Johannes.

Downwind-sailing. Better balance.

We were having a discussion over on the Proa-file group recently about how much lateral area is required relative to the sail area. Apparently there is a general rule of thumb that the lateral surface area (e.g. keel or centreboard) should be about 4% of the sail area.

I did some maths and discovered that one important factor is the boatspeed to windspeed ratio, the higher the ratio, the less lateral area you need. I wrote a spreadsheet for calculating the lateral area required: http://www.users.on.net/~malcolmandjane/boatstuff/LateralAreaCalc.xlsx . To use it, you need to make a bit of a guess as to the lift coefficients for both the sail and the hull lateral surfaces.

For a hull without a centreboard or keel the lift coefficient would probably be quite low, say around 0.1. Even with a centreboard it would still not be much more than 0.4 as the lift coefficient goes up by about 0.1 for 1 degree of increase in angle of incidence, so a lift coefficient of 0.4 for an efficient centreboard would equate to about 4 degrees of leeway, roughly speaking.

This also highlights one of the major problems with using models to test ideas. Models generally have lower boatspeed to windspeed ratios that full size boats, due to the way that hull resistance scales, so a model will generally need more lateral area than the full sized vessel for the same leeway angle. If you are aware of this, you can compensate by increasing the lateral area of the model, but of course, this effects other factors like the wetted surface area, so it’s not so easy to make a direct comparison.

Cheers,

Mal.