making a sailboat rudder

[PaulRowntree]

This past weekend I broke my sailboat's rudder in half near the waterline... good thing the wind was down and we were close to shore! The boat is 25' long, and the 40"x18" rudder is just hung off the stern.

Anyway, I am thinking of machining a new one from two thicknesses of marine plywood epoxied together, then seal it somehow. I am trying to figure out how to shape the submerged foil properly from a mathematical model (based on the NACA-00012 airfoil shape) without hand-coding gcode. I don't think flutes will work because of the curvature. It may be possible to inlay some steel reinforcing strips running top to bottom between the layers to keep this from happening again. As long as it still floats, weight shouldn't be a real issue.

Has anyone done something like this with VCP? It may be similar to cutting airfoils for wind generators and the like.

Would you folks trust to the epoxy or would you add some T-nuts to bind the layers of ply together?

Any advice would be much appreciated.
Cheers!

[tmerrill]

Hi Paul,

I've done NACA airfoils in Aspire without any issues, and helped one person set up the toolpaths in Cut3D, but can't recall what program was used to generate the model. I really can't think of how to do this with just VCP that would be time efficient and produce good results. If I really needed this and only had VCP, I'd probably create a series of thin profiles, cut them out with 2D toolpaths and use them as a guide for hand planing.

One thought about the metal reinforcements (especially if you figure out an easy way to duplicate the shape and could repeat it) is that if your rudder hadn't been the sacrificial component, could more severe damage been done? I'm not a sailboat person, so it is just a curiosity question.

Tim

[PaulRowntree]

Hi Paul,

I've done NACA airfoils in Aspire without any issues, and helped one person set up the toolpaths in Cut3D, but can't recall what program was used to generate the model. I really can't think of how to do this with just VCP that would be time efficient and produce good results. If I really needed this and only had VCP, I'd probably create a series of thin profiles, cut them out with 2D toolpaths and use them as a guide for hand planing.

One thought about the metal reinforcements (especially if you figure out an easy way to duplicate the shape and could repeat it) is that if your rudder hadn't been the sacrificial component, could more severe damage been done? I'm not a sailboat person, so it is just a curiosity question.

Tim

Thanks Tim, it looks like I will have to be creative. You have a good point about the sacrificial aspects, but the wood was clearly hit by dryrot or the like. The fibres at the break feel pulpy. Not the stuff to trust your life and well-being to!
Cheers!

[Paul Z]

Paul,

The drag of an object is largely due to its cross section in the direction of travel. If you compare your hull cross section to the rudder cross section, you will see that that drag of the hull is orders of magnitude greater that the rudder. My point is that if you left the rudder as a rectangular solid and it would have very little effect on the boats performance ... but it would be ugly.

I think you could use a large diameter round over bit on the front of the rudder and 120 degree V bit on the back and you would have a very serviceable rudder.

I’d bond the two pieces of marine grade ply with marine epoxy.

Paul Z

[PaulRowntree]

Paul,

The drag of an object is largely due to its cross section in the direction of travel. If you compare your hull cross section to the rudder cross section, you will see that that drag of the hull is orders of magnitude greater that the rudder. My point is that if you left the rudder as a rectangular solid and it would have very little effect on the boats performance ... but it would be ugly.

I think you could use a large diameter round over bit on the front of the rudder and 120 degree V bit on the back and you would have a very serviceable rudder.

I’d bond the two pieces of marine grade ply with marine epoxy.

Paul Z

Of course Paul, you know that you are bordering on heresy there, and in any other forum a fanatical argument would erupt. But fortunately, not here.
From what I have been reading, many folks would agree with you, many not. Some insist that the way a rudder should work is as an airfoil providing lift to one side or another, and not just a water deflector.

I am thinking that if I want to get back on the water anytime soon, your approach may be the most practical, since my sailing skills are far from the theoretical limits of what my boat could do. Or any boat for that matter!

Cheers!

[RichardF]

Paul,

take a look here : http://mothboat.tripod.com/CMBA/Building/foils.htm

Might give you an idea

Richard

[Rick O.]

For durability & strength, you might look into WEST system epoxy, along with some of the many guides on using it for marine plywood.

[PaulRowntree]

Thanks Richard, Rick. That is good reading. I am now leaning towards generating an STL file to describe the sides, based on the math here

http://www.pdas.com/naca456thick4.html

It is an unusual shape of rudder however, so the STL creation will be ... interesting. My wife is rolling her eyes. What could possibly go wrong, I ask?

I underestimated the size of the rudder though, and if I end up CNC'ing the piece, it may be using John Murphy's expertise and his AXYZ. Roger Foster of Carlisle Canoe Co. just south of here is a canoe craftsman, and will help with the epoxy and glass.

All that is required now is the courage to start ...

Cheers!

[PaulRowntree]

Here is a progress report on the rudder project. The overall dimensions will be 66"x28"x1.5".

I made a sheet of graph paper large enough for the rudder, then marked ~40 key (x,y) points to define the outline of the rudder. I then wrote a Delphi program to read these coordinates, and scaled them to make the units inches. This shape can then be exported to an EPS file that can be used to cut out the profile vectors.

The trick was the foil shape. Normal NACA foils become thinner as the chord (the distance from leading to trailing edge) gets shorter, which would put a fine taper down at the rudder tip, where it is most likely to get banged about on rocks (not by me of course). So I adapted the NACA formulae to maintain a constant thickness as the chord changes, and wrote a scripting code to allow different cross-sections to be created for different pieces of the rudder, and then output a binary STL file for each side of the rudder. The image shows the starboard piece, with the NACA-like foil below a more rectangular using MeshLab to view the STL file. This piece has three different sections; some sanding is needed to ease the transitions between the different sections, but the shape should be good now. The waterline is where the rectangular stock first flares out coming down.

Next step is making chips and then assembling the sides with epoxy and applying fibre glass.

Cheers!

[Rick O.]

Hey, Paul.. you've put a lot of work into this already, looks great.

Just a thought, though: have you done any weight/buoyancy calculations on your materials? Many stock rudders have a foam or honeycomb core, with fiberglass skin. They are often designed to have only slight negative buoyancy. If this is going to be made of heavier material than the original, the boat's balance might be affected. More mass might require more effort to move, especially if the boat is heeling and the rudder needs to be moved quickly. A heavier rudder might need stronger hardware, and perhaps a longer tiller.

[PaulRowntree]

Hey, Paul.. you've put a lot of work into this already, looks great.

Just a thought, though: have you done any weight/buoyancy calculations on your materials? Many stock rudders have a foam or honeycomb core, with fiberglass skin. They are often designed to have only slight negative buoyancy. If this is going to be made of heavier material than the original, the boat's balance might be affected. More mass might require more effort to move, especially if the boat is heeling and the rudder needs to be moved quickly. A heavier rudder might need stronger hardware, and perhaps a longer tiller.

Good thinking Rick. The original was mahogany or mahogany ply, so the only mass change will be the fibreglass and epoxy, which should be only a few pounds. Minus the extra weight that I have been carrying with the old waterlogged wood! I have decided against using internal steel reinforcing, but vertical carbon cords between plys is still an option. I have access to the mount points on the transom and could easily reinforce them or broaden the mount plates if the weight was excessive.
Thanks for the additional thoughts!
pr

[Paul Z]

Another Delphi programmer??? I thought I was the only one left.

Paul Z

[PaulRowntree]

Hi Paul ... guilty as charged. It is the only language that I can still understand months and years after writing the code.

The rudder wood work is now done, and I am very pleased with the results. I ended up modifying the NACA airfoil shape to give more thickness on the trailing edge by adding into the formulas a wedge with the thin edge at the thickest portion of the foil; with the two halves together the trailing edge is now ~0.3" instead of 0.02", so it is much stronger. The fibreglass adheres better as well. Plus, I refaired the transition from the handle to the foil shape, which cuts down on the manual shaping required. Then it was off to John Murphy's shop to cut the sides into 3/4" marine plywood. Thanks again John! I then joined the halves with thickened WEST epoxy, did some final shaping and drilling for mounting hardware, then did the glassing. Roger Foster of Carlisle canoes is an expert and an artist, and he provided a lot of guidance and insight. Thanks Roger!
Tomorrow I mount up the hardware and see how it fits to the boat. I decided to reinforce the tiller by adding 1/8" bands of steel on each side of the yoke holding the rudder. Short of a major catastrophe, it should be smooth sailing from here!

It has been interesting learning how to produce precise complex shapes via the STL files.

Cheers!
pr

[John Murphy]

Turned out great Paul!

Glad you didn't paint it