Index

Reducing the Turbulence of a Mark I Keel.

All too often sailors focus on maximizing sail drive to achieve boat speed without regard to reducing the hull (or other) drag.  Some do however, by using slippery bottom paint.  Still others scrub their hull smooth before a race.  The water line is particularly important in this regard.  To achieve maximum speed, an SJ23 must be heeled not more than 20⁰ and sailed flat on her lines so the turn of the transom is just above the water for smooth flow aft.  I achieve this by adding weight to the bow and keep the crew to the forward end of the cockpit. 
At times I wish the SJ23 had a round transom so it could have more reserve buoyancy to carry crew weight and have a larger cockpit.  Then I would also have a skeg-hung rudder that is more powerful than a transom hung version because it can't stall as easy.  This would force you to have a small inboard engine which would move weight forward and clean up the appearance of the transom.   But let's get back to the subject!  
What about the keel design of an SJ23? The aft end of a Mark I keel is cut square.  Can this create a partial vacuum to produce a significant drag at hull speed?  To get a feeling of how much drag this might be, try pulling a (2x13)" plate sideways through the water at 6 knots.  The drag on the boat won't be as much as this, but what really demonstrates the issue is when you turn the plate sideways, in line with the flow.  The drag goes to almost nothing. 

SOME THEORY - Remember when you were a kid and you stuck your arm out the car window with your hand turned flat to the wind? It had a lot of resistance and eventually you got tired and couldn't hold it any longer.  If you turned your hand  so it was almost level and cupped it a bit, at the critical angle to the wind it acted like an airplane wing and your arm lifted.  This is the effect you want on the keel. I've often wondered what the drivers behind us thought!  To understand this "lift" phenomenon consider the keel as a foil that can generate lift, similar to an airplane wing, but at a slower speed and through a thicker medium.  As a boat sails upwind through the water it also has leeway.  With leeway the keel slides slightly crooked "crab style" through the water, creating lower pressure on the leeward side and higher pressure on the windward side. The difference in pressure creates a difference in flow speed over the two sides of the keel.  The shape of the keel and the differential flow speed creates "lift" on the windward side of the keel.  This is the effect/force that helps to move the boat faster through the water. This same effect may also create turbulence at the aft end of the keel. 

• The front of a keel is rounded for smooth entry and the aft end is tapered for smooth exit.  Ideally the only drag is due to surface resistance, not turbulence. If there is turbulence-induced drag, then the effect of this drag is more pronounced at the rear of a keel than at the front. 

• If the front of the keel is blunt, the still water is compressed against the flat surface.  This deflects the oncoming stream sideways similar to a bow wave.  There is minimal resistance of moving water sliding against still water. But this is no reason to not make a form designed keel. 

• If the rear of the keel is cut off flat, it creates a partial vacuum that is continuously filled with new water flowing into it.  However, the flow creates vortexes that increase the effective cross sectional area of the flat area and the hence the resistance as well. This is the force you want to eliminate. 

Individually these are insignificant factors. When you add them up, they can really create a significant force.  This is probably the reason why one particular boat always goes faster or slower.  Keep in mind, there are no single significant breakthroughs in boat speed, only little ones that add up.  The key is to remove the ones that are slowing you down and to leave the other ones alone.  Finding what slows you down is difficult! 

Consider the following practical examples; 

1. If you sail an SJ23 with the turn of the transom below the surface, the boat goes slower compared to when the turn is just above the surface.  Part of the reason is drag from water having to flow up and forward from under the bow.  Ideally the flow should continue aft without turbulence.

2. An SJ21 is equipped with a swing keel that retracts inside a centerboard slot.  There is a gasket to close the slot to prevent water surging in while the keel is down.  Without the gasket, an SJ21 has drag and is not competitive.  With the gasket this hull has deadly speed! 

3.  The cargo straps on a highway transport truck are usually tied down with one turn in them to minimize wind drag and fatigue.  If the strap is tied straight then it will vibrate with speed.  As the speed increases the magnitude of the vibration increases.  This increases the effective cross sectional area of the strap, increasing the drag.  The same happens in rigging.  Read most any text book on heavy weather sailing if you want more information on this subject. If the strap is twisted, it breaks up the flow, has less cross sectional area and does not vibrate.  It is still not without drag but is the lesser of two evils.

4.  A loss of drive occurs at the leech of a mainsail if the boom is pulled in too tight.  The mainsail is stalled. The effect of this can be seen when the leech tell-tales point forward, instead of streaming aft.  The air flows forward along the leeward side of the cloth instead of flowing aft.  The laminar flow is gone, increasing drag and reducing drive. The vacuum along the leech is literally holding you back. 
   - A mainsail pulled in too tight creates a thrilling ride with lots of heel and little speed.  However you can point higher so in some circumstances it pays to pull the boom in tight.
   - A mainsail released so the telltales stream aft creates less heel and the most speed.  In the heat of battle during a race you have to resist the temptation to pull in the mainsheet.

5. The foil shape of the rudder and keel stub makes them more forgiving while steering the boat.  ("Forgiving" is actually a different word for describing stalling).  I once had the occasion to sail a 22' keel boat that had a flat sheet of aluminum for a rudder when the original was out for repair.  While I could turn the boat if I moved the rudder slowly, a quick turn beyond about 20⁰ would result in a stall, or loss of flow along the leeward side.  At this large angle the boat simply ploughed along in a straight line instead of turning.  This was with full mainsail and 110% jib up.  When we hoisted the 150% genoa the rudder couldn't keep the boat from rounding up.  The rudder could only hold the boat on a straight line with the working jib.  Very interesting learning experience. 

In all five examples listed above, a partial vacuum was created at the back causing turbulence and drag instead of a smooth exit flow aft. 

Still not a Believer? - Around 1985 Bruce Bingham (note 3) rounded the square edges of Sega's propeller aperture to improve the water flow to the propeller for maximum thrust.  By simply rounding the corners of the propeller aperture, the water could flow in and the propeller was no longer starved for water.  Now the total surface of the blades could push the boat instead of just the tips that protruded beyond the edge of the hull.  The engine could easily drive the boat to hull speed with the same throttle setting.  But instead of traveling faster and wasting fuel, he chose to travel at his original speed with about 30% reduced fuel consumption, quieter operation and longer engine life.  It should be noted that during the 1970s the corners of many long keel propeller apertures were cut square without regard to hydrodynamics.  The subtleties of this phenomena was only just understood by manufacturers of the time.  The extra fibreglass added to the strength and it was probably easier to construct.  Besides, the biggest portion of the blade stuck out beyond the keel anyway, so who cares?  This is probably the same thinking that produced the square aft shape of the SJ23 keel, but that is my guess. 

SOLUTION - One might think that the solution for the SJ23 keel is to taper the aft end of the keel so the port and starboard streams of water join together smoothly, without vortex creating drag.  You may well argue that, "I don't race and so I don't need this high tech stuff on my boat."  My answer to that is, "Two sailboats going in the same direction almost always constitute a race!"  The following are techniques for streamlining the aft end of a SJ23 keel.  Please remember that these are educated guesses that are the result of many discussions over beer and coffee by authoritative sailors!

1.      The simplest method is to simply round the aft face.  As an experiment this is fairly easy do and it might give you the biggest gain for your work.  If you want to try this make a temporary jig by cutting a 13" length of 2" diameter ABS tubing in half, lengthwise.  Shape the top end so it fits snug to the hull shape and the bottom end to match the keel.  Plug the open bottom to eliminate turbulence.  Attach it temporarily to the aft end of the keel using foil tape.  This should eliminate most of the keel drag with minimal effort. Foil tape should last one season if it is applied to an acetone cleaned surface. 

2.      The second method is to streamline the keel by tapering the aft to a point, from top to bottom.  This will increase the surface area of the keel but the extra surface drag should be offset by the reduced turbulence.  The longer keel might affect the overall balance of the hull though, which you might be able to correct by moving the centerboard forward. A variation on this method is to create a logarithmic curve to the taper.  This might prove to be the best choice of the three ideas because it should maintain the laminar flow along the hull and minimize the turbulence for the rudder.  The center of balance of the hull shouldn't be affected and it will add the least surface area to the keel area. 
NOTE the following response is from Mike Foreman who tried this method and discovered it does NOT work. 

"If I remember right we were pretty positive on the benefits of this project back in 2003, but there are a couple of things we didn't consider and Mr. Murphy had a bit of a hay day with me after I faired the aft end of the keel. 
The fairing was pretty straight forward.  I clamped a couple of 2x4s to the keel and got a position and size measurement.  Shaped some blue foam and epoxied it to the back of the keel after removing the bottom paint, etc from the area.  Then I added a couple of layers of cloth and more epoxy and it was ready for barrier coat and bottom paint.  I figured this would be my secret weapon on the race course as the modification wasn't immediately obvious when the boat was sitting on the trailer ready to be launched.
WEATHER HELM - 1.  Now let's talk about the rake of the mast.  When I gave up on this project I had the greatest rake of any boat on the dock and it still wasn't enough.  I was going to have to get a new forestay and back stay to get weather helm as I couldn't adjust mine any further.  This resulted in lack lustre acceleration, noticeably after a tack.  Estimated cost of new fore & aft stays about $250 Ca. 
DIRECTIONAL STABILITY - While on the subject of tacks, with a 1 foot longer keel you have more material to twist around through the water.  Not good.  Too slow.
STABILITY - Without realizing it I made Joy IV....MORE...  tender.  This is a bit of a difficult concept to grasp, but here goes.  You can say the keel now has a better grip on the water or as a recent article I read, "an increase in planform area, increase in heeling arm, and increase heeling moment", I read this article several times but I think I got it figured out, right where it says ,"makes the boat more tender."  I didn't immediately notice this phenomena because the modification was done in the Spring just before the seasonal launch.  If I had sailed the boat, then made the modification and sailed again immediately, I would most likely have noticed the tenderness.  The increase in planform area was approximately 1 square foot, or about 12% increase over stock. 
Conclusion.  I have not given up on doing modifications.  However, I view this one as a bust.  When it warms up outside I'll screw with the .......not now, that is a secret."  Mike Foreman.

3,      Consider adding a set of small wings.  They can have a pronounced effect on minimizing vortexes.  One SJ23 was equipped with wings from the factory.  See Tech Tip H11.  Rounding the aft end of the keel and adding small vortex destroying wings might just prove to a have double whammy effect for the overall performance since less turbulence from the keel means the rudder also operates in smoother flowing water.  If the rudder operates in smooth flowing water the power is increased and it needs to be turned less to achieve a result.  Turning the rudder in smaller angles will reduce drag and result in sailing faster. 

      A word of caution, DO NOT grind into the AFT END OF THE keel to taper it.  The ballast is encapsulated inside the Fibreglass and you will expose it, causing corrosion.  The boat WILL LIKELY sink if the top of the ballast is not sealed to the hull! 

Measuring an improvement under sail might be next to impossible due to all the variables of wind and waves. If you could control these factors this type of development would be easy to confirm.  Since I can't afford a test tank, probably the best way to test an improvement is to compare a modified boat to an unmodified boat with the same skippers and sails.  All you are looking for is a change under a given set of conditions. The method of convincing yourself that the boat goes faster simply by the amount of effort expended seldom holds true, especially with sail boats.  I call this the "waxed car syndrome." A car goes faster after you wax it!  A slightly more scientific method, might be to hold a boat in a stream of water and measure the drag on the tow line using a scale.  Another method is to power it at a fixed RPM and note the speed.  Yet another method would be to make before and after polar chart for your boat.  It involves a whole lot more work but usually generates more useful data for other things. It's also a great way to learn about the performance of your boat. 

1. Paying Attention to the Back End Really Pays Off - I've studied this article and my conclusion is that if you have a long keel, such as with a SJ23, cutting the back end off square is OK.  In fact it may be the optimal thing to do.  This is based on the fact that the drag from the turbulence is less than the surface resistance and lift from a tapered end.  It would be nice to prove this but it would be difficult. Some factors in air flow do not necessarily migrate to water, and vice versa. See measuring improvement. 
2. Weighing in on Winglets.  - This is a very good article on the effect that winglets have to control turbulence. 
3. Bruce Bingham is a technical illustrator of nautical articles and lived on board a Fantasia 35, Sega.  He used to publish Sailor's Sketchbook in Sail Magazine.  His hand drawn illustrations dealt with boat improvements that could be performed by most handyman.  His signature was the cat and the mouse, which were always included in some hidden corner. 
4. For some really technical information on shaping keels go to www.computerkeels.com 
5. Center Board Upgrade from SAIL Boatworks - http://sailmagazine.com/boatworks/upgrades/sailboat_centerboard_upgrade/
6. The best solution might be to leave the keel alone and spend your money going to boat shows or drinking beer in the shade somewhere!

Return to Tech Tip Index.  .  .  .  .  .  .  .  .  .  .  .  .  . Have a Question?