Here it was, the moment we had been waiting for. As Carey bore off to starboard, I jumped the main up, came aft, took the mainsheet, and slowly loaded up the main. Both of us were nervous. Every thing checked outno unusual sounds, and the mast was bending as it should.
As the boat started moving nicely through the water, we glanced at each other. We couldn’t believe this was finally happening. The excitement and not a little relief finally broke through and, suddenly chattering away, we noticed for the first time how really beautiful the day was. Two years of planning, worrying, and building would pass before this day arrived. Whatever else happened, it was a day worth the wait.
I’ve been interested in freestanding masts for 20 years, since studying the history of American sailing craft. I was fascinated by the revenue cutters, blockade runners, and countless fishing vessels from 1770 that sported freestanding wooden masts. Granted, the masts were massive and heavy, but the simplicity of the arrangement appealed to me. If it’s not on the boat it can’t break. There’s also an old saying in the art world that says “less is more.”
When the first Freedoms came out in the 1970s, I was thunderstruck. The implications for the future of freestanding masts were obvious. From that point on I knew that someday I would build one.
Copernicus, the boat that Carey, my girlfriend, and I live on is a Spencer 42, a big sister to Hal Roth’s original Whisper. The boat was designed with a sail areadisplacement ratio of about 14, which doesn’t speak well for the boat’s medium- and light-air performance unless one hoists a 150% genoa. We have tried to stay away from large headsails for various reasons. That, coupled with the need for new standing rigging, compression post, chain plates, and a complete refit and paint on the existing 20-year-old tube, not to mention replacing sails of close to the same age, was all the excuse I needed to dive into the deep end.
I learned about Eric Sponberg, a yacht designer operating out of Newport, R.I. It struck me that he was a methodical engineer with a strong background working with carbon who had an openness that I felt the project needed. In other words, someone who wouldn’t laugh at the idea of sticking a freestanding mast on a 20-year-old CCA Rule voyaging boat. I contacted Sponberg with a design brief for a mast, carbon schedule, and sail plan. Sponberg felt that a change from an existing stayed rig to a freestanding one had not been done to his knowledge, but he thought that it would be possible. Sponberg’s first request to me was to supply him with a set of righting-moment calculations based on the actual boat. I provided Grant Brandlmeyer, a marine engineer and son of the designer of the Spencer 42, with the basic numbers. Brandlmeyer, in turn, supplied Sponberg with the righting moments and stability curves.
The next step was to track down carbon fiber. Dry fiber would be fairly easy to obtain. But in this case I felt that wet lay-upthat is, using dry fabric and adding epoxy to the layupwould be too difficult over a length of 60 feet. I didn’t relish the prospect of handling all that epoxy.
Opting for pre-preg cloth
Even with the slow-curing epoxies available, I had visions of the panic and pressure of trying to lay down cloth, apply glue, and then vacuum bag. All of this before the epoxy kicks off! No thanks. I knew that for me pre-pregs were the answer. This type of material is carbon fiber pre-impregnated with epoxy. In the factory, carbon is run through a bath of epoxy designed to cure only when cooked at a temperature of 200° F or more. The material is then shipped frozen to the end user. Using pre-pregs would mean building an oven, something I had no idea how to do. But I felt the long open working time would be worth it.
Unfortunately, there was no pre-preg carbon fiber available.
After several months of fruitless phone calls I despaired of ever finding anything. I had been pouring my woes into the willing ear of Sponberg when Rich O’Mera’s name came up. From that point on my material-supply problems were over. O’Mera was able to make arrangements with Gary Patz and the people at Y.L.A. Advanced Composites to provide me with what I needed. O’Mera didn’t stop there. He was an unlimited source of practical information on the whole process of building in carbon-fiber pre-preg, including the voodoo bits. His great good humor at being pestered by Carey and me at all hours of the day and night went beyond the pale. I’m sure the project could not have reached its natural conclusion without his help.
Arrangements were made with one of the theater scenic shops in which I work to use the space for the summer. We were only going to use the shop for a month, but as it turned out the project took three months, well into the busy scene-building season. For more than a month the carpenters building the first show of the season had to live with a 60-foot thing that they had to either duck under or walk around to get to the tool room or office. The production manager, Michael Wood, and staff showed great goodwill and kindness that all the inconvenience a 60-foot mast in the middle of a shop entailed. Their willingness to accommodate us will not be forgotten.
Sponberg’s design called for a plywood cratelike structure that consisted of an athwartships three-millimeter-ply vertical sheer web the length of the mast. Three-millimeter-plywood stations every two feet defined the elliptical shape of the mast. The trailing and leading edges of the ellipse consisted of two thicknesses of 3/4-inch marine ply. The skin or surface of the mast was made of three-millimeter okoume ply. Sponberg’s specifications called for 1/16-inch veneer, but this was another material I could not obtain. The ply stations were cut by a CNC router, which saved me several days of tedious cutting and shaping. Wherever there was a high load point in the mast, such as the jib fitting, Sponberg specified 3/4-inch ply blocking doubled or trebled in thickness, as the case may be. The ends of this blocking inside the mast were then tapered on a 12-to-one scarf. I might add that all ply joints, of which there were approximately 48, had 12-to-one scarf joints as well.
The inside of the mast was treated to three coats of ProSet epoxy with high-density filleting where necessary in way of the stations. With three coats of epoxy, plus fairing on the exterior, the wooden part of the mast equation was finished. ProSet epoxy was chosen because of its ability to withstand the temperatures needed to cook the carbon.
I had built various wooden masts and booms, both hollow and solid, in the past and had thought that this structure would be a similar and straightforward process. However, it was more akin to building a 60-foot rowing shell with almost all its attendant frames and stringers.
The next step on this voyage of discovery was to pre-cook the wooden part to a higher temperature than what the carbon would be heated to. This was to promote out-gassing and stabilize the wood. The mast-cooking oven consisted of a long tube built in halves lengthwise and in eight-foot sections.
The oven tube was lightly framed, skinned with 1/8-inch doorskin and insulated. The heating system, designed by an oven engineer, proved unable to heat evenly and sustain the minimum heat ramp rates needed. Forty hours of tests bore this out.
Around this time I noticed a pattern that had existed since the project began. People whom we had never met would appear magically and unbidden with the answer to the problem of the moment. Such was the case here. A client of Sponberg’s who happened to be in town on business came by the shop with a friend who turned out to be a mining engineer. His sound solution to the problem of even heat distribution got us past the pre-cook stage and on to the actual laminating.
Running overtime
We were now about a month and a half into the project. Two weeks overdue and we hadn’t started laminating yet! As a result, we were feeling the pressure to finish and were not a little anxious. This may, in part, explain the results of our first four layers, which we ruined in our haste. Chalking it up to a learning curve, we removed the badly laid-up carbon. Armed with fresh advice from O’Mera, Sponberg, and Jeff Kent of Composite Solutions (who in a 10-minute phone conversation very kindly cleared up a set of nagging problems regarding handling the material), we were able to see an significant increase in both speed and quality. The laminating schedule consisted of four layers of alternating diagonals. After that came zero-axis unidirectional tapes one foot wide and in sets of diminishing lengths. We finished up with four more layers of 45° axis tapes.
A typical procedure cycle would consist of pulling the carbon off the roll, paper backing face-up, and placing it directly onto the mast. To keep the strip from slipping off, nylon string would be loosely tied around it at about four- to six-foot intervals. Since each strip has a specific compass point location along the axis of the mast, the centerline of the one-foot-wide tape was marked. With a wire stretched tightly along the mast at the compass point, say 90°, several of us would both wiggle and stretch the fabric until the centerline and wire were lined up. With all the nylon string snugged up, and checking for kinks in the fabric, we would hand-pat from the centerline out to the edges. After removing the paper backing, the entire surface was hard rolled, followed by a tight wrap with packing plastic.
Every four strips, not including overlaps, meant one layer, so eight strips meant two layers. At that point we would vacuum-bag for about one hour to debulk. Two people could manage, but three or four was easier, especially on the longer pieces. As a result we ended up hiring some of our friends to help speed up the process.
O’Mera’s horror stories about parts that weren’t kept clean was enough to keep us in a state of perpetual alarm. In spite of the excitement of working with the material and the laughter from having the help of friends, both Carey and I were very tense. We had no idea whether the quality was up to snuff, despite verbal assurances from O’Mera and Sponberg as we sent them photos. Since we had never physically seen it done, it was hard to take their confirmation to heart. Added to this was the pressure of nearing the safe shelf life of the RS1 resin that was in the carbon. Once out of the freezer and on the mast the clock starts ticking on the stuff. It stops ticking about a month and a half later, and that was pushing it. Sooner would have been better.
According to O’Mera, the resin has an inhibitor in it that prevents the hardener and resin from cross-linking at room temperature, but only for a set period of time. In our case the ambient temperature of the shop hovered around 50° to 55° Fahrenheit, which lengthened the life of the material and our luck.
With a final push we managed to get the last layers on by Saturday, Oct. 11, Canadian Thanksgiving weekend. Sunday was spent setting up the vacuum bag and thermocouples. These were spaced every six feet and exited out of the oven. They would enable us to track the temperature along the length of the mast throughout the heat-curing process. The final step was to finish assembling the oveninternal ducting, fans, and propane space heaters.
We started the cook at 1:00 P.M. Monday. Initially we took thermocouple readings every 15 minutes as the ramp-up started. When graphed, these readings indicated that uneven temperature increases were occurring along the length of the mast, due in part to differences in mass that cause something called heat lag. Also the propane/ducting arrangement was not contributing to the level temperature increase. Something had to be done, and quickly, as the temperature was rising too rapidly in some areas and not fast enough in others.
The temperature increase could not be faster than 7° a minute nor slower than 2° a minute. Four hours’ sleep in a 36-hour period was not helping my problem-solving abilities. I was genuinely up against a wall and couldn’t see a solution to the problem. As I stood there flapping my arms in the air, two friends, Jeremy Aikin and Sebastion Riedl, came in and spotted the solution almost immediately.
When they explained it, my fuzzy brain couldn’t grasp their logic. I felt that since I wasn’t coming up with anything brilliant myself I’d best go with their idea So they patted me on the head and, to keep me out of the way, gave me a harmless job to do. By redirecting the internal ducting and moving one of the propane heaters they were able to save the cook.
As the temperature approached 170° F things started to even out. At that point we took the temperature up to around 190° to 210° F. Once we were staying at temperature properly we found that the temperature could be fine-tuned by opening or closing two large doors, one at each end of the shop, a fraction of an inch. Careful monitoring throughout the night and early morning brought us to 6:30 A.M., at which point we shut down the oven and went home for some much-needed sleep.
Coin for void testing
Tuesday evening saw us back at the shop, and with several friends we stripped the oven and vacuum-bagging material off the mast. O’Mera and Sponberg had both suggested a U.S. 50-cent piece as a tool for testing voids. It being a theater scene shop complete with a props department, the coin was produced. Apart from one tiny void that was easily repaired, the mast produced a clear, bell-like tone when tapped with the silver. A sawhorse at each end and two people with a total weight of around 500 pounds sitting on the mast in the middle caused very little deflection.
We weighed the mast with a digital scale, and it came in at five pounds shy of Sponberg’s calculation of 590 pounds. At this point we had to put the project aside due to a backlog of work obligations, so, using theater-style technology, we flew the mast up to the ceiling at two pickup points along its length.
Winter passed and spring was well into May when I was able to clear time and shop space to continue working on the mast. It was as strong as it was ever going to be, so we were able to move it at will, flying it up out of the way as shop needs dictated.
The mast fittings were first on the agenda. The sheave box was laminated, vacuum-bagged, cooked, and assembled separate from the mast. The masthead was then notched to accept the sheave box, which was glued and laminated in place with several layers of 11-ounce biaxial glass. The rest of the fittingsjib, gooseneck, etc.were to be laminated directly onto the mast.
The gooseneck was very similar in construction to the jib fitting. First, two plywood fairing pieces (in the case of the gooseneck to take the vertical pin that would hold the boom) were glued in place one above the other. Carbon was laminated to approximately 3/8-inch thick around the mast about three to four inches wide, much like a belt. These straps essentially held the fairing pieces to the mast. With minor additions and differences to each, both parts were bagged and cooked onto the spar.
The oven was built as a rectangular plywood box with an open bottom. The small ends were cut out so that the oven would span the mast without touching the part to be cooked. After the horrors of trying to make an 60-foot oven work properly, this tiny oven was so easy, I kept thinking I was forgetting something. For heat I used an ordinary propane camp stove that packed enough BTUs to ramp and cook to the temperatures I needed without any insulation.
The winch pads glued and faired into the mast had 5/8-inch aluminum plates glued and glassed into place. The plates were then tapped to take the machine bolts for the winches. Fairing and painting came next, at which point we were able to start installing hardware such as track, sheaves, etc.
While all this was going on I was also preparing the boat so that it could handle the different kinds of loads a freestanding mast would impose on a hull. The partners had a 3/4-inch-thick solid fiberglass (knytex) and epoxy ring four inches deep that was larger than the diameter of the mast by about a inch and a half. This ring was glued and filleted into the deck opening with two inches below the deck and two inches above the deck.
Under the deck, biaxial, 24-oz. cloth and carbon tape were used as reinforcement, tagged from the sides of the ring outwards to the cabin sides and bulkheads, then down the cabin sides and under the deck to large radius foam hanging knees. The glasswork continued down the foam knees and down the hull sides. The knees spanned from the main bulkhead just aft of the mast to another structural bulkhead approximately three and a half feet forward. On deck the outer skin was cut out in a five-foot-wide circle and a 12-to-one taper was shaped in the balsa core to within five inches of the deck ring.
The mast step
Biaxial cloth was then laid up in increasingly larger concentric rings until the original thickness of the deck was achieved. At the mast step a similar ring to the deck ring was laid up, but in this case it was snug fitting. Two-and-a-half-inch-by-six-inch plywood beams on edge would radiate out from the mast step diagonally, fore and aft and athwartships. Filleting and gluing these in place with two alternate layers of biaxial and two alternate layers of 24-oz. cloth would complete the structural work on the hull.
Although the mast step ring was prefabricated, none of the mast step structural work was installed until after the mast was actually installed. As the mast was lowered into the boat the mast step ring was slipped over the butt of the mast just before the mast touched bottom. At that point a series of wedges, beams, and temporary brackets held the mast securely in place. This enabled us to plumb the mast as well as adjust the rake at our leisure. Only when I was happy with the plumb and rake (about a week of measuring and squinting) did we install the mast step permanently.
It was raining the day we installed the mast. All of us who had ignored the weather to assist and witness the event laughed at the wet, so pleased were we that everything was going well. When the crane left and everything was secure (the boat was on the hard) we retired to the local pub, sat by a fire to dry off and ate enormous amounts of food. We toasted our friends who had helped, each other, and the success of this seemingly endless project to date.
Clouds of a different kind were looming in the shape of a sailmaker. We had been in discussions with a reputable sailmaker regarding a new suit of sails. Unfortunately, he sold the business, and unbeknownst to us the new owners were out of their league with this project. They built sails that failed after 10 hours of sailing. This sailmaker then went out of business, sending us off in search of another sailmaker, who built us a new set out of Spectra sails.
So we were temporarily stuck with awful, badly cut sails and were about to take our new mast out for its first ride. How did the boat do? We expected some improvements in some areas and none in others. I realize now that we were far too conservative in our expectations. It is a completely different boat. We went out to English Bay on the coast of British Columbia on a hot summer day with SW winds at about 12 to 15 knots true.
As mentioned earlier our hearts were in our throats from the time we left the dock to when we started loading up the mast for the first time. Not a creak, not a groan from the mast or hull structure. We sailed under main alone for awhile and were amazed at how well the boat sailed. Five to five and a half knots and pointing well to boot! The old bucket was doing better than it ever had. When we unfurled the jib things just got better and betterpointing higher and sailing deeper. At one point on a deep broad reach we were coming up to a Beneteau 37 sailing with a 150% genoa. In order to stay apace and to leeward we had to furl the jib. You have to realize that previously we would have been thrilled with just staying up with a boat of this type with our main and jib up.
We also noticed when we sailed downwind, the boat was easier to keep on course. In the past, whiteknuckling the helm was the order of the day. It remains to be seen how the boat steers in big winds and seas, but it looks like the downwind steering problems may have become manageable. The other improvement was faster tacking.
I’ve been asked why I would go to such trouble to build a carbon fiber mast and a freestanding one at that. Before I started I probably would have listed all the financial and performance reasons. And now looking back I realize that the best reason for doing this thing was that we couldn’t do it alone. So many friends, and strangers who became friends, came forward just when we needed them. That was one of the best rewards of doing this project.
Bryan R. Pollock is a visual artist who lives in Vancouver.