David Tyler wrote:

The standard textbook Principles of Yacht Design reckons on the NACA 0012 foil for a rudder to be robust against stalling. An end plate will probably help, but adding one has made my rudder hum at high speed (it didn't before).

 A humming rudder is not the end of the world. My Frøken Sørensen's rudder sometimes did that when the speed exceeded 6.3kts. That rudder was a deep foilshaped thing, attached as a swing-up dinghy rudder (no endplate). My armchair guess is that the rudder on Zebedee is less likely to start humming, as it is supported at the lower end.
No guarantees, though.

Arne

Many thanks!  The hole between the keel and rudder hasn't been filled in.  Your ideas exactly coincides with what Paul Thompson suggested including the 15 % NACA profile and end plates which I've only just heard of . Hauling out tomorrow.  I'll take photos of the various stages.  

Alan

I am with David Webb that end plates will help, no matter what rudder foil you go for. This has been tried by me and many others ( a Phil Bolger idea), and it really cuts the turning radius with such a lo-AR rudder. Make the end-plate with a ‘wing span’ of  some 50-70% of the chord. In your case, I also suggest you add a similar ‘cavitation plate’ just below the waterline.

As for foil shape, it depends: If you have filled up the propeller aperture with deadwood, so there is no gap between keel and rudder, there will be little help in making any fancy rudder section. If you on the other hand have built the original space between keel and rudder, I suggest a NACA 00-series section with 15% thickness to chord, that is 125.7mm or 4.95 inches thickness  -  if you keep the chord of 33 inches. These blunt sections are not prone to stalling. We built one such spade rudder for a 34’/4ton boat here and gave it 18% balance. We gave it over 17% thickness to make room for the rudder stock (freestanding). That rudder ensured powerful steering with light tiller forces.

Now I twisted that rudder foil into one with 15% thickness and 33” chord, and added the numbers at each station. Showing a resolution of 0.1 mm is a bit overkill. Be happy if you approach 1mm resolution.

Cheers,
Arne

PS: Just for the cause of science, I suggest you make the new rudder's profile and size as close as possible to the original.

2 weeks ago Zebedee failed to tack again but this time landed on the rocks. Foolishly I I hadnt left room to wear round.   I'm hauling out in 2 days time to do repairs and change the rudder . Zebedee has the keel and rudder of the long keel version as shown in Voyaging on a Small Income.  The current rudder is shaped but only 1.75 ins thick with a chord of aprox 33 ins.  I will have to check the chord.  I  don't want to change the rudder mounting ie the stern post.  Paul Thompson very kindly is working on it but it would be great to hear some ideas.  My first thought is that 1.75 ins is only 6% camber so that if the rudder is turned more than a few degrees there is turbulence and a braking effect.  So perhaps increasing the thickness to 12% or even more would help.