Who to believe, Kaper or Michlet

[Dan Cahalane]

I'm working on a new design for a racing kayak. I've run Michlet-Godzilla for a recommended hull and have worked up designs based on Kaper. Very different results. The Michlet styled hull is about 7% less total resistance under michlet. The Kaper designed hull is a similar amount "faster" under kaper. The kaper is more swede form with less wetted surface area. The michlet has a much fuller bow and is more symetrical. Any thoughts on which one would be more accurate predicting total resistance for an 18' kayak?

 

[DCockey]

Kaper uses regression: curve fitting using a set of parameters which describe the hull and based on test results of other shapes. No physics is directly involved.

Michlet uses a simplified model of water wave physics and boundary layers, and is based on "first principals".

 

[Dan Cahalane]

I understand how each works, I guess I was looking for some experience on which translates better into real world results. A wave based model that doesn't align with actual test results isn't going to be as good as a model based on regression analysis. I would assume that the Michlet approach would prove to be the more accurate. Is that the case? My Kaper boat looks more what I am used to regarding kayak design. The michlet boat is kind of "Stubby". Do I trust michlet or go with something that looks more "normal" to my eye?

 

[gonzo]

I think you should compare theoretical results based on assumptions and simplifications to successful designs. Also, in the case of small boats like kayaks, you style of paddling and boat handling has a huge effect.

 

[jehardiman]

Gonzo is right here, +/- 7% is totally in the weeds. The answers you have don't even consider real world things like environment and construction which will easily be 10% or more of effective drag, this is the known unknown. Run the final hull shape in both programs, add 3% to the largest drag, stuff the rest in your hip pocket to take care of the known unknowns, and bask in the praise if you win knowing it was your paddling style and not the hull shape that won.

 

[Dan Cahalane]

In a marathon and ultra-marathon kayak race 7% is a huge difference. In a 3 hour race. 12 minutes can make the difference between winning and not even reaching the podium. 7% over a 300 mile race is close to 5 hours.
Both boats will be constructed identically. Both will be paddled in identical conditions. Both will have the same water line length to beam ratio. Both have the same displacement. The question is really do I use the design optimized to the Kaper calculation or the design optimized to the Michlet calculation.
Kaper seems to favor a more swede shape with hollow up front and prioritizes lower wetted surface area. Michlet's simulations seem to favor fore/aft symmetry with a soft V bottom. I could build both boats and test them side by side to determine which calculation produces the most accurate forecast, but I was checking to see if anyone already had this knowledge before I went to the trouble.

 

[gonzo]

You could make two small models of carved styrofoam and tow them side by side. Maybe 18 to 14 inches long. I still think that your style of paddling will have a huge effect. For example, some people lean to the side a lot so a flatter bottom will be more effective for them. If you race in flat water, the lowest wetted surface may be the way to go. I rough water the same design may be slower because it submarines under the waves. I use to row on a four. We used to practice on two different shells. The wider and heavier did better speed on waves (probably up to 1 foot).

 

[wet feet]

I think Gonzo's suggestion has much merit.There is no substitute for getting a model wet to verify a prediction.I would suggest that with a kayak hull form a large scale model at perhaps quarter scale would be easily produced and ballasted to the DWL and once you can arrange an accurate method for moving the model at scale speeds it should soon produce a result.You could even use the method Weston Farmer advocated in his book to carry out a direct comparison by using a yoke connected to both models and observing which one is surging ahead,courtesy of the drag on the other model.