John G. Alden designed hundreds of boats over many decades. MAYAN is design number 356-B and based a design from 1928. Alden's design serves a specific mission: To annually traverse the Intercoastal Waterway from New England to the Carribean in the fall, returning home in the spring.
To succeed in this mission, MAYAN needed a draft no deeper than 5' with her centerboard up to avoid the shoals. She needed a masthead no higher than 65', to clear the bridges without forcing them to open. She needed a broad beam to hold her up against a stiff breeze, and a long waterline to let her make the passage quickly.
|
MAYAN sporting her beam and shallow draft |
Alden's design delivered solutions to all these problems in a graceful seaworthy design: 60' on deck, 16'6" of beam, 4'6" of draft, a 45'6" waterline, and a mainmast head that stands 63' from the waterline. She is a bit short on sail-area, but that's perfect for her new home sailing on Monterey and San Francisco Bays. Here we have consistently strong winds that mask the ill effects of MAYAN's small rig. A side effect of her shallow draft is that her rudder is a bit under 4' tall. When combined with rather primitive ideas about hydrodynamics, it was challenging to keep MAYAN on course in quartering seas and hard to maneuver in close quarters. The advantages of our strong local winds dissipated on the square front and blunt trailing edge of our antiquated rudder shape.
I'm catching up on long-overdue posts. Our significant winter project last year was to re-shape the rudder while MAYAN was ashore to refresh her anti-fouling paint and various other maintenance tasks. Homer Lighthall, who has built hundreds of rudders for racing boats like MERLIN, was recruited to perform the work. Bill Lee, the yacht designer, volunteered to advise on the new shape. A squad of friends formed up from thin air and offered opinions about nearly everything.
The original rudder had a substantial list of problems:
|
MAYAN Original Rudder Shape |
1) The prop aperture took up more than one-third of the leading edge of the rudder. 2) That leading edge was 4" thick, cast of bronze, and sported square corners!
3) The trailing edge of the rudder was 3" thick and finished in a hemisphere; one of the worst shapes when considering drag.
4) The rudder didn't extend up to the counter above, leaving a gap between the rudder and the hull.
5) The rudder didn't extend to the full depth of the keel, leaving it undersized.
6) A gap of 3" ran between the deadwood and the rudder post, allowing turbulence to form along the leeward side of the rudder.
|
Cardboard CAD |
As Bill and I discussed various strategies to re-shape, extend vertically, and reduce the after edge of the rudder, Homer appeared with a sheet of cardboard. He clamped it to the old rudder, lining one side with the trailing edge of the deadwood. Trimming the right angle with a box cutter, Homer slid the top up until it was as close as possible to the counter. Extending a straight edge from the bottom of the keel, he cut the lower side of the cardboard. Finally, after stepping back and estimating the new additional area, he cut the cardboard short of the trailing edge of the old rudder. In ten minutes, Homer had designed the profile of a new rudder, expressed in cardboard. Bill weighed in, "That's about right."
I was still pondering which CAD system I was going to use to draw up the new rudder profile.... but, it was over. Homer's design looked good to me. Cardboard CAD!
|
Old Rudder vs new Cardboard CAD |
"Homer, I'd like to have the same rudder area so that we don't change the boat's handicap rating," I said.
"I'll get my tools," Homer responded. Measure what I've added and what I'm going to cut off. Let me
know if I missed it."
Five minutes with a tape measure confirmed that Homer's eye was nearly perfect. He was off by less than two square inches out of a total of 1,848. "Damn near perfect!" I told him as he returned with his pushcart loaded with a circular saw, power planer, and angle grinder.
"Do you two want to talk about this more, or is this good enough?" Homer asked Bill and me.
"It looks about right to me," responded Bill.
With a little trepidation, I agreed. "Go for it, Homer," I said, sounding more confident than I felt. "We can always screw some back on if it's too small."
Homer just laughed and scribed a line on the old rudder following the trailing edge of
the cardboard. Then, picking up his circular saw, he cut about 7.5" off of the trailing edge of the old rudder along a graceful curve. Seeing a four foot high, by 7.5" wide, by 3" thick slab of wood laying on the concrete under MAYAN, felt incredibly odd. It had happened so fast.
Trading the circular saw for the power planer, Homer went to work, making the trailing edge of the rudder thinner. He worked entirely by eye. Where I would have built templates and patterns, Homer just ran his hand over the shape, stepped back to look at it every few minutes, and kept grinding. My job was to sweep up the rapidly growing mound of teak shavings growing underfoot.
To reinforce the wood and transfer the significant loads from the rudder shaft aft to the blade, the original builders had used two 3" by 1" bronze straps, which extended aft three feet. Homer cut away the last four inches of these straps. Removing the tail end of the bronze straps allowed the after edge of the rudder to be ground down to a far narrower shape. Before the advances in hydrodynamics made in the 1950s, builders and naval architects didn't appreciate the drag created by thick foils. MAYAN was being brought forward into the 21st century just a bit.
With the profile and the foil shape roughed out, Homer headed off to his shop to order some 4" thick mahogany he could use to build up the top and bottom of the rudder to match the cardboard pattern. I set to work grinding all the paint and putty from what remained of the rudder and from areas of the deadwood, which would be shaped and faired to let the water approach the rudder and aperture.
Just aft of the propeller, the square leading edge of rudder was now clean, shiny bronze, highlighting the absurdity of its 4" thick square-cornered shape. As Homer looked over my shoulder, he laughed at the shape.
|
Re-Shaped Leading Edge In The Aperture |
"That's like towing a bucket!" and then continued, "It is terrible in that spot because the water leaving the propeller is moving a lot faster than the boat, It hits that flat surface and goes around those square corners. I'll bet she'll get better fuel economy once we fix that."
Having completed the cleanup, I sat in the shade under MAYAN's port quarter and researched the clearance the MaxProp required aft of the blades and at the tips. We had plenty of space. It seems that MAYAN had once had a much larger propeller. At least she had an aperture large enough to accept one.
"What's the answer?" asked Homer.
"We can make the leading edge round. We've got 5" to work with, and we only need two. We can also bring the bottom of the aperture up about 3". I replied.
"That's what I was hoping," replied Homer as he started cutting the long wooden strip for the bottom of the rudder.
|
Taking Shape - Note: New Wood and Bronze Straps
|
Within an hour, he had cut the bottom edge of the rudder off straight and square, roughly shaped the new addition, dry-fit the new piece with long screws, and then attached it with West Epoxy and a bonding thickener. Turning his focus to the top of the rudder, he installed the new wood in the same manner. Mixing up epoxy and a sanding filler, he built up low spots on the sides of the rudder, which I hadn't noticed.
"We'll start final shaping tomorrow," Homer announced as he packed up his tools. We were only a few days into the project, and MAYAN's rudder looked entirely different.
As always, the final shaping takes much longer than I expected. It is a series of steps repeated over and over. First, the planer and grinder rough out the shape. Then, epoxy filler builds up low spots. Finally, after more shaping with the planer, hand rasps, and a sander, it's back to the first step to continue the shaping. With each pass, the flat wooden plank, which was MAYAN's original rudder, was becoming a much more modern foil shape.
As Homer worked his magic with the power planer, I assigned the task of figuring out how to fill in the large gap between the rudder shaft and the trailing edge of the deadwood. Taking a cue from Homer's work on the rudder, I fashioned wooden blocks which we attached to the deadwood with epoxy and small screws. The trailing edge of these blocks cut to a concave grove to parallel the front of the rudder shaft.
|
Filling The Gap, Rudder On The Left |
We exposed the blunt leading edge of the bronze straps holding the gudgeons to the deadwood. A ramp of thickened epoxy served to fair their sharp corners into gentle bumps, further reducing drag. I ground away the anti-fouling paint to expose the epoxy and epoxy primer sealing the hull and re-coated everything with West Epoxy making it ready for the new primer. Epoxy's ability to survive for decades underwater undamaged and its ability to adhere to high moisture content wood make it the perfect sealer and glue for work on wooden boats.
|
Homer (l) and Matthew (r) Comparing the New Trailing-Edge to the Old |
As Homer wrapped up his tenth day of work shaping the rudder, I fashioned a template of the port side. Having watched Him work by eye for days, I was impressed with how symmetrical the shape of the rudder had become. But, it was time to measure it with something more accurate than my eye. Flipping the template over I, lined it up along the starboard side of the rudder. It fit perfectly. Homer had shaped the 4' broad and now 5' tall rudder perfectly. Using nothing but his decades of experience and skill, he had shaped the two sides of MAYAN's rudder to within a few thousandths of an inch.
As we examined the new rudder shape, one of our long-time crew, Matthew, joined in by holding up the old trailing edge of the rudder for comparison. It wasn't until relatively recently, by schooner standards, that designers realized that the trailing edge should be nearly sharp. As we now know, the hemispherical shape of MAYAN's old rudder trailing edge is just about the worst possible shape, causing more turbulence than a square edge. We predicted this change in the trailing edge would improve MAYAN's performance substantially.
Finally nearing the end of the project, Homer and his team applied two layers of West epoxy and glass cloth to the rudder and deadwood, completely sealing the rudder and bonding the new wood along the rudder shaft to the deadwood. Two coats of epoxy primer covered the West to fill the cloth pattern. Then three coats of anti-fouling finished everything off.
How did it work?
|
New Rudder Shape Completed |
After relaunching MAYAN, we headed out of Santa Cruz Harbor on a calm day. In the channel, idling along at 4 knots, everything felt about the same. But, upon reaching the open Monterey Bay and pushing the engine up to 1750 RPM, the crew and I exchanged glances. It was noticeably quieter.
Gone was the RUMMA-RUMMA-RUMMA of the prop wash smacking into the square flat-fronted rudder just aft of the propeller. Now at 1750 RPM, in a calm sea, MAYAN was making 8.2 knots through the water, a full knot faster than with the old rudder shape.
I pushed the rudder hard over. We watched as MAYAN turned much more quickly and in a far tighter turning circle. Our experienced eyes estimated that at maneuvering speeds, the turning circle was at least one boat length smaller.
Perhaps the most attractive feature of the new rudder shape was expressed by Gene, our most experienced crew member at 80 years old, when he announced, "I used to have to work at turning her. Now I'm driving around with just one hand on the wheel!" His smile was infectious. As each of us drove MAYAN around in drunken circles, backed her up, and spun her side to side, the grins grew wider. The new rudder was a tremendous success.
|
Gene Sofen at the helm, MAYAN
|