Reuel Parker The dry fit of the Douglas-fir staves for schooner Reef Chief‘s 47’ 10″ (14.6m) bird’s-mouth construction mainmast is supported by temporary forms in an outdoor workshop on Stock Island, Florida.
Text and photographs by Reuel Parker
Allen Cody Taube sailed his wooden Block Island Cowhorn Island Belle from California to Florida nearly 50 years ago. Landing in Key West, he identified an opportunity to make a good living in the charter-boat trade there and purchased a three-sail bateau (also known as a skipjack) to pursue this new career. It was heavily built, 65′ (19.8m), plank-on-frame workboat, nearly new at the time. She was unusual for a skipjack in that she had a full keel instead of the traditional centerboard. She came from Deltaville, Virginia, designed by Joseph Gregory and built by Carl Pederson in 1972.
Renamed Reef Chief, the boat was certified as a subchapter-T passenger carrier, Taube put her to work hauling tourists out to the Florida Keys’ reefs for the next 40 years. It’s a tribute to her designer, builder, and Taube, who also worked as a marine surveyor and wrote a couple of books on wooden boatbuilding techniques, that Reef Chief survived those decades of hard service in Florida where wooden boats are particularly vulnerable to rot, teredo worms, hurricanes, and the vagaries of extreme heat and humidity.
In 2024, with retirement in mind, Taube considered selling the boat, but he was realistic about her tired condition. He knew a transaction could take months and involve expenditures for maintenance, storage, insurance, etc., so he approached a knowledgeable party, his longtime competitor in Key West’s dayboat charter fleet, Danger Charters.
Founded by Wayne Fox and sold in 2022, Danger has been in business for more than 30 years, sailing three schooners of my design—Danger, Sarah, and Danger’s Prize. Danger’s owners accepted Reef Chief as a donation knowing that an extensive restoration would be required, as well as recertification with the U.S. Coast Guard.
Despite having sold the company, Fox remains essential to Danger Charters as a skilled advisor, mentor, and hands-on technical expert capable of running a large, complex business and making repairs to everything from woodwork to rigging, machinery, plumbing, and electronics. He remains on call to solve the myriad problems that crop up for the company. No surprise then that Fox became the Reef Chief restoration’s hands-on manager and shipwright.
Reuel Parker I got involved when Fox invited me to help him with restoration work on his personal fleet of Herreshoff and Bahamian Regatta sloops and dinghies, but before we could do more than replace one keel on a Winer Malone Abaco dinghy, the Reef Chief project took over. The boat remained in the water at Buddy’s Marina and Boatyard on Stock Island.
Wooden boats are much safer afloat in the tropics than hauled out, where they dry out and deteriorate very rapidly. Most of Reef Chief’s repairs—replacing many structural components like masts, deck beams, plywood decking, hardware, rigging, and some superstructure including hatches and portlights—will be made while she is afloat.
Start with the Spars
Early investigation confirmed that the boat was going to need new masts. Fox asked me to help design and build them, using the bird’s-mouth octagonal laminated method I had applied to masts for Danger’s two Exuma 52 charter schooners. The new masts for Reef Chief would be similar in size, shape, materials, and construction methods. At 10″ in diameter, and around 50′ tall, they are huge.
Reuel Parker Specifications:
10″ (254mm) O.D.
6 1/16″ (155mm) I.D.
1 3/4″ (44mm) max. wall thickness.
1 3/16″ (30mm) 45° cut depth
4″ (101mm) stave width
Taube supplied copies of the original design blueprints from Joseph Gregory, including the mast-taper drawings and sail plan, from which I designed the new hollow masts (see Figure 1). The only modifications I made were to soften a “bump” I perceived when lofting the parabolic mast taper, and to slightly increase masthead diameter for new stainless-steel collars/caps. I made hounds to the shapes and locations shown on Gregory’s plans. In appearance and function, the new masts are virtually identical to the originals.
One note about the schooner rig, which most people define as having a taller mast aft (mainmast) and a shorter one forward (foremast): a more technical definition is that if the larger sail is aft, usually requiring a longer boom, the rig is classified as a schooner even if the aft mast is shorter. On traditional working vessels of Chesapeake Bay, this configuration is common. The three-sail bateau (skipjack) rig on Reef Chief is defined as a schooner. I mention this here because in the following text, I will refer to the shorter aft mast as the mainmast.
Shop and Materials
Antonio, the yard manager at Buddy’s, allowed us to set up a spar-building shop under a long roof-only structure used by local commercial fishermen to build lobster traps. The shelter has a concrete slab about 10′ x 40′ (3m x 12.2m), covered by a slightly pitched shed roof and no walls. In my experience, there is no better arrangement for working in the tropics—shelter from the sun and lots of fresh air. We installed a portable toolshed at one end, and Fox and I built a spar bench, utilizing six upright posts on the shady north side of the shelter. We made supports 27″ (.7m) high by 48″ (1.2m) deep, spaced about 7′ (2.1m) apart. We brought electricity from a source 80′ (24.4m) away and installed outlets in the work area. Because several of the tools we needed would draw 15 amps, we used 10/2 with ground Romex cable installed in ¾” PVC electrical conduit.
I called around trying to locate the lumber we would need for the masts—air-dried, clear, heart, vertical grain Douglas-fir. I found it at Americas’ Wood Co., (Washington, Maine,) and ordered about 10% more than my materials estimate total from the full-dimension 2×4 material they had in stock.
The wood arrived within two weeks by common carrier and exceeded expectations. Some of the 4″ (102mm) widths had 50 growth rings, indicating that the tree this material came from could have exceeded 1,000 years of age. A truly majestic tree, tall and straight, with no knots for probably the first 50′ (15.2m) of height.
For epoxy, I called Glue Products in West Palm Beach and ordered a 15-gallon kit of 2:1-mix epoxy: five gallons (18.9l) of thick resin; five gallons of thin resin; and five gallons of slow hardener. From RAKA in Fort Pierce, I ordered two gallons (7.5l) of Tropical Hardener (the slowest-curing hardener available). For the vital scarf joints, we bought WEST System epoxy locally, mainly because Fox was nervous about using an epoxy that he was unfamiliar with for that application. WEST is very expensive, and I find those little hand pumps difficult to work with, but it is reliably very strong. Appropriate to the conditions and geography, we used WEST’s Tropical Hardener.
Reuel Parker Five-gallon (18.9-liter) pails of fast and slow epoxy resin and hardener were fitted with PVC spigots for rapid, efficient decanting and mixing.
The Epon epoxy I sourced from Glue Products for the mast assembly is made by Shell Corp. and is about a third of the price of WEST. I used three-ounce paper cups for mixing small batches, and one-pint, one-quart, and one-gallon plastic containers for mixing large batches. During mast assembly, we mixed a quart of epoxy at a time, which meant one person working nonstop with an electric stirring device. During layup, each mast uses about 3 gallons (11.3l) of mixed epoxy in about 45 minutes of working time.
Our first step after unloading the Douglas-fir from the pallets it was shipped on was to organize the material by length, weight, grain, and quality. While all the lumber was excellent, it varied in density, amount of sapwood (very little), tightness and straightness of grain, and weight. Lengths varied between 17′ (5.2m) to 20′ (6.1m). We sorted them, reserving heavier stock for lower in the masts when possible, and any slight sapwood material to be used aloft. There was also some variation in thickness and width.
Because all the wood was rough-sawn, we had to plane it on all four sides to a uniform width and thickness. I settled on 1 ¾” (44mm) as the thickness. Planks of less than 4″ (102mm) finished width were located aloft where mast diameter was less. Our goal was an average plank width of 3 15/16″ (100mm). With three segments for each stave, eight staves for each mast, and two masts, this meant running 48 pieces, plus spares, through a power plane four times. That’s more than 200 passes through the power plane, much of it for 20′ planks, which comes to in excess of 3,600 (1,097m) lineal feet through the machine, with thicker pieces requiring more than one pass. We purchased a new Rigid power plane for the job and mounted it on a steel stand to achieve a below-waist working height. We used roller stands at each end of the machine. The work took days to complete and generated some 20 extra-large garbage bags full of wood shavings.
Construction Preparation
Bird’s mouth spar structure consists of eight panels, or “staves,” assembled octagonally in section, with one longitudinal edge of each stave cut to a 90° V-shaped notch, or bird’s mouth (see drawing above). We purchased a stripped-down 10″ 15-amp tablesaw online to cut the V-grooves. After the first few cuts, the saw choked on sawdust and stalled, so we removed the dust shields and collection devices. We set up roller stands about 8′ (2.4m) on each side of the saw to support the wood.
Each piece of wood (called a “segment”) must go through the saw twice, reversing it end-for-end, to make the two 45° cuts that yield the 90° V. For that sequence, I made and installed a wooden fence attachment because the saw blade would otherwise cut into the bottom of the metal fence. We also had to get a ripping blade, as the all-purpose blade that came with the saw was completely inadequate for the job. For the record, in my shop I use a 10″ Delta Rockwell Contractor’s table saw, with a 2-hp, 240V motor and reversed drive pulleys for more power at the blade. A hobbyist table saw will not do this job, even though the cuts are only 1 ¼” (32mm) deep. For two 50′ (15.2m) masts with eight staves each, plus spares, the V cuts came to over 1,800 (549m) lineal feet and required a few days to complete.
Reuel Parker Staves of clear, vertical-grain, air-dried Douglas-fir mast stock with bird’s-mouth edges and scarfs cut await assembly into a hollow mast. Scarfs will be staggered in adjacent staves.
While cutting, you must keep the material going through the table saw simultaneously tight to the fence and flat on the table. It therefore requires two (or even three) people to feed the stave segments through the saw. I always make a “finger joint” guide on the bandsaw and clamp it securely on the bed of the tablesaw to help hold material tight to the fence. Keeping the material flat on the table is more challenging, because any curvature of the stock on the flat will cause it to ride up at the blade, cutting away too much material at the edge of the plank. One purpose of the roller stands is to prevent this. With only the two of us working, we ended up with several bad V-grooves, causing us to reject or re-cut the segment.
When making these cuts, it’s essential to make them ever-so-slightly deeper than the center of the plank, to avoid leaving a small ridge in the bottom of the V that would stop the adjacent plank-edge from seating properly, leaving voids in the glue joint. If you can’t avoid it, the two solutions to this problem involve using a rabbet plane to remove any center ridges or a block plane to ease the corner of the mating edge of the adjacent plank thus accommodating the ridge.
Reuel Parker Wayne Fox applies a small grinder to the scarf joint surfaces for better glue adhesion.
Cutting Scarfs
Each of the eight mast staves consists of three “segments” scarfed together to yield the full height of the mast. To avoid the scarfs of neighboring staves ending up in the same place (radially), we planned on two different scarfing schedules to assure adjacent scarfs would be at least 2′ (610mm) apart. The first schedule is for four staves of (from the bottom) a 20′ segment, 17′ segment, and 18′ (5.5m) segment, for a total length of 55′ (16.8m). From this, the two scarf joints are subtracted, with each scarf comprising two cut and planed ends. Using a scarf ratio of 8:1, on the flat (as opposed to on the edge), with a finished plank thickness of 1 ¾”, the length of each scarf is 14″ (356mm). So, subtracting 30″ (762mm) for the scarfs, our finished stave length was a little over 52′ (15.8m). The second layout schedule for the alternate four staves would have an assembly order (from the bottom) of an 18′ segment, a 17′ segment, and a 20′ segment, similarly scarfed and totaling the same 52′ length. The finished masts would be 47′ 10″ (14.6m) and 50′ 9″ (15.5m), main and fore respectively. When the mast is assembled, all scarf joints should be 2′ apart, alternating stave to stave.
With the planning done, we got to work cutting scarfs using a circular saw with 7 ½” blade set to maximum depth. I made diagonal cuts on both sides of each plank end with the help of a simple plywood guide-form to scribe them. Fox then power-planed the sawn cuts smooth and true, after which he scuffed them up with a small body grinder with 40-grit sandpaper to achieve “tooth” for better glue adhesion.
The middle segment of each mast stave must be cut for scarfs at both ends, while the top and bottom segments are scarfed only where they meet the middle segment. That means there are 16 scarfs per mast, requiring 32 cut and planed ends, which took us a couple of days to complete the two masts.
We glued the scarfed mast segments together on the spar bench stacking four staves at a time to allow us to use the same clamps for all four staves (and to save time). We placed Visqueen plastic between the staves to keep them from adhering to one another.
It is critical to align the V-grooves while gluing scarf joints to assure longitudinal continuity. If a plank segment has a slight curve on the flat (side to side), it is more important to align the surfaces at the scarf than to achieve a straight line along the finished stave. A bump in surface continuity at a scarf joint will cause at least two large glue voids and cause an asymmetrical bulge in the finished mast; a long gentle curve on the flat of the stave will be straightened out during final layup.
Slight inconsistencies can be corrected with a rabbet plane or mini-grinder, and after curing, the glue joints are cleaned up using a body grinder.
Taper
Next, we laid out taper on the full-length scarfed staves. Working from my compressed drawing of mast taper (Figure 2), I multiplied mast diameter by 0.40 (40%) to find the width of each stave. For a 10″ diameter mast, the stave width is 4″ (102mm). Stave width decreased above the lower cylindrical part of each mast, going into the parabolic tapered upper sections. Working from Gregory’s mast-taper design drawings, I laid out the stave width at each design station.
The mainmast diameter is 10″ for the first 19′ 6″ (5.9m), after which parabolic taper is employed, with a design station every 7′ (2.1m). The taller foremast’s diameter is 10″ for the first 22′ 6″ (6.8m), with taper identical to the mainmast for design stations above that. I made a “compressed” drawing of mast diameter, reducing the 7′ station spacings to 2″. By using a flexible batten, I reproduced the mast taper lofting from 50′ to 15″. Multiplying each station by 0.40, I determined the stave width at each station. By extrapolating the spacing of the spar bench onto my compressed mast station taper drawing, I found the diameter for each spar bench location and multiplied it by 0.40 to determine stave width at each bench support, where the open-top forms would be located.
Reuel Parker After marking each station location on the squared edge of each stave (opposite the V-groove), I tapped in a brad at the calculated width for each point. Fox and I then laid a long, perfectly straight batten along the brads, and clamped or held it in place while I used a Sharpie pen to draw the cutline on the stave. This cut would start just after the full width 4″ bottom portion of each mast (20′ and 24′, respectively) and run to the mast head. The saw cut would be about 30′ (9.1m) long, through 1 ¾” old-growth Douglas-fir—not a job you want to trust to a hobbyist’s circular saw.
I ordered a vintage 7 ½” Skilsaw for the job. This saw has the blade on the left, so that it is in line with your arm and shoulder, and so that you can see your cut (you watch the “V” at the leading edge of the saw’s base—not the blade) without having to lean over the saw. It’s gear-driven, with an oil-bath reservoir, and draws 15 amps. We bought a new ripping blade for it. It is essential that the cut be perfectly straight and at exactly 90° to the plank surface (disparities can be corrected using a long-bed hand plane). The top edge of the stave must fit perfectly into the V-groove of the adjacent stave.
If the sharp edge splinters during mast layup, or if there is a slight ridge in the trough of the V-groove, the glue joint will be compromised. With that in mind, after making the cuts I eased the sharp corners using a block plane.
Assembly
Working from the mast-taper figures for the spacing of the spar bench locations, I designed open-top forms to lay up (assemble) the masts. The spar bench supports are, essentially, fixed rigid sawhorses (Figure 3). I made the forms from cheap ¾” plywood. We positioned them at the inner ends of the horizontal 2×6 supports. The forms have vertical guides drawn on both sides to align them.
Reuel Parker The bottoms of the forms were aligned in a straight line, so the after side of the masts would be straight parallel to the sail luff. While the sides of the spar taper symmetrically, all fore-and-aft taper is on the front side of the mast only. Thus, the first stave was laid out in the bottom of all the forms to be the straight back of the mast. On the first stave of each mast, we laid out and attached all wiring that would run internally. The mainmast has a 4 AWG stranded ground cable and a shielded VHF radio cable, and the foremast has the same ground cable, a wire for the mast-head anchor light, and a wire for the steaming light. (All wiring holes in the mast are drilled to angle downward to prevent water entering the hole.) For the wires for the lights, we used 14/2 AWG PVC-jacketed USCG-approved boat cable.
Reuel Parker The stave for the back of the mast is the first one laid in the bottom of the form with cable already attached for radio and grounding.
Next, I selected the subsequent staves to be laid in the forms and numbered them for the sequence in which the mast would be assembled. For example, if a stave had some curvature (on the flat), it would be used on the side where it would be easily straightened out during assembly. We did a dry fit of the assembly for the first mast (mainmast) to assure that the staves would lay up tightly without problems. We didn’t use the final stave, as it would be difficult to put in place and nearly impossible to remove as each stave is locked in place by the bird’s mouth joint when the next stave is installed.
Reuel Parker Trays, rollers, brushes, rags, and gloves ready for mast assembly.
For actual mast layup, Fox drafted a large crew from the employees of Danger Charters. (You need at least eight people to glue up a 50′ mast.) One person did nothing but mix large batches of epoxy as rapidly as possible, using a drill motor with a stirring attachment. Another person (frequently me) distributed the mixed epoxy to all the workers placing and gluing the staves.
During assembly, epoxy flies everywhere. Knowing that, we instructed everyone to wear disposable clothing and two pairs of disposable plastic gloves, so one pair could be stripped off if torn or hopelessly fouled with epoxy. The second pair saved the time of getting to the glove box and putting new gloves on.
To start, all the staves were laid out with their inside surfaces facing up. Six workers were given plastic trays and 9” roller frames with throwaway 3/8″ nap covers. The mixer (one of Danger’s managers) first mixed “thin” resin with the “tropical” (slowest) hardener, and the crew wet out the staves with the thin epoxy as rapidly as possible. Then the staves were turned with the V-grooves facing up, and the workers coated those using rollers and/or 2″ (50mm) disposable “chip” brushes. Because the mast-back stave was already placed in the forms, Fox and I used brushes to coat the same inner and V-groove surfaces where a roller wouldn’t work.
With the V-grooves still facing up, the workers coated them all again, using thickened epoxy glue with the same 2″ chip brushes. For glue, we used “thick” resin, further thickened with structural cotton fibers. After all the coating was finished, we installed the staves in the open-top forms on the spar bench in numbered sequence: The bottom was already in place, then the adjacent bottom pieces, side pieces, and top side pieces. These last two had to be held in place by four workers (two for each stave) to prevent them from falling inside the mast. The “shutter plank” was very carefully laid in by the last two workers, Fox and me. On first try, it dropped inside the mast, and we had an anxious few moments retrieving it. This thing was 50′ long, totally slick with wet epoxy, and six people’s fingers had to be gotten out of the way as it slipped into place.
Reuel Parker The mainmast laid up with staves held in place with rope clamps as the epoxy cures and Fox fills any gaps in the seams.
As soon as all the staves were in the form, everyone grabbed a length of ½” polypropylene rope and a precut stick about 21″ (.5m) long. We looped the ropes around the mast, tied a square knot in the rope above the mast, inserted the end (not the middle) of the stick under the knot, and twisted clockwise until the rope was digging into the exposed corners of each stave and glue squirted out of the joints.
Placing these tourniquets (Spanish windlasses) about 18″ (457mm) apart, they became the clamps of the spar bench with each subsequent clamp stick locked on to the end of the adjacent one. With compression pressure applied the length of the spar, Fox and I used 2-pound hammers and blocks of wood to jar closed any open joints. Then, armed with wide putty knives, we filled any glue joints that weren’t tight.
The entire process occupied about 45 minutes of nonstop action from mixing the first batch of epoxy to clamping the last Spanish windlass. Time was of the essence as we had to complete the full mast layup before the epoxy began to kick in the 80°F (26.7°C) Key West weather. Curing time was about 18 hours at that temperature, so we let the epoxy set overnight. While total cure time for epoxy is about 72 hours, the clamps can come off long before that. Our first step after removing them was to power plane all the raised corners off, plus any large globs of epoxy.
Reuel Parker The mainmast planed to an octagonal form. The gap in the back stave is to accommodate an exposed keel bolt on Reef Chief‘s mast step.
Shaping & Finishing
The initial goal was to create a perfectly octagonal spar, full length. Having achieved that, I went on to remove the octagonal corners from the mast, creating even-width surfaces for a 16-sided form. This must be done carefully, as progressively less material is removed as the mast tapers. Next, I planed those corners off, creating a 32-sided form. After that, I smoothed the spar to a rounded surface using a Makita 8″ low-speed grinder/polisher with an 8″ soft pad and 40-grit adhesive-disc sandpaper. My technique involves “rolling” the grinder toward and away from me, easing all the edges off the faceted form. Each pass along the length of the mast sands about one-quarter of its exposed surface, so it is necessary to roll the spar one quarter turn after each complete pass of its length. I sanded the mast in this same manner using progressively finer grades of paper, after which I switched to a 5″ orbital (dual axis) sander, using 100-grit paper.
Below the location of the boom shelf a few feet above the deck, we left the mast octagonal in section, as is traditional. In addition to being aesthetically pleasing, this practice provides conveniently flat surfaces to bear against the mast wedges that are driven at the partners (deck level) to secure the mast in place.
When both masts were at this stage, we cleaned them thoroughly with compressed air and tack rags before soaking them with penetrating epoxy. For this, we used Glue Products’ thin resin, thinning it further (about 10%) with Sherwin Williams Reducer No. 54 (made for reducing epoxy paint).
Referring to Reef Chief’s old masts, I made mast heels for the new spars. This involved cutting the correct bevels to match the existing mast steps in the boat and making the heel tenons to fit the mast step mortises. For these, I shaped five pieces of solid Douglas-fir blocks to fit 24″ (610mm) into the hollow butt of each mast and glued them in place with lots of thick epoxy. After curing, I cut and shaped the tenons, making them identical to the discarded originals.
Reuel Parker The foremast tenon shaped to fit the mast step.
The standing rigging for the masts was designed in the traditional manner, with both shrouds on each side of the mast formed by a single wire rope seized together aloft after looping around the mast over a thumb-cleat, or “hound,” on the opposite side of the mast. Material for the hounds was donated by our friend Tommy Mann, a master carpenter at the Safe Harbor Marina on Stock Island. The wood he gave us is called tiger wood because of its distinctive coloring. It is a dense, rot-resistant tropical hardwood. I sculpted the hounds to match the original architectural drawings by Gregory and routed shallow mortises into the mast sides at the attachment locations.
Reuel Parker Marking the location on the mast to rout out for installation of a hound made from tiger wood.
Using a battery-powered circular saw set to a shallow-depth cut, I made numerous circular cuts around each masthead deep enough to accommodate the stainless-steel mast collars, or caps. I used a chisel to chop out the roughcut material. Working from the top of the mast downward, I made the mast heads a smaller diameter to fit the collars, leaving rounded shoulders where the masts were full-size.
Reuel Parker Reduced-diameter masthead with protruding wiring awaits fitting of the metal collar.
To house the lightning grounds on both masts and the masthead anchor light on the foremast, I made filler blocks for the open top of the mast with appropriate holes for the wires. The finished collars, or caps, have corresponding holes.
The last project before painting the masts was to make the boom shelves. I removed one from its original mast to use as a template to make the new ones of air-dried, pressure-treated yellow pine with synthetic tops for the boom jaws to ride on. The boom shelves were glued and screwed to the masts using large stainless-steel fasteners. Each shelf had three knees supporting it.
Reuel Parker The boom shelf viewed from below where the mast remains octagonal in section.
After allowing the epoxy coatings to cure, we lightly hand sanded the masts with 180-grit sandpaper, cleaned them again, and applied Awlgrip 545 epoxy primer. The primer was then lightly sanded and followed by a second coat of primer. After lightly sanding this, the masts were coated with two coats of Awlgrip polyurethane finish—white for the mastheads, and Prairie Beige for the remainder.
Conclusion
As much as I would have liked to stay, I had commitments that required me to return home to Maine while work on Reef Chief continued. In my analysis of the mast project, I calculated that Fox and I had built the two spars in about 175 hours, with occasional help from Danger Charters’ crewmembers. We worked five days a week, averaging 6 hours a day, enjoying long lunches, mostly at El Mocho, an excellent local Cuban restaurant. The entire project lasted about six weeks.
About the Author: Reuel B. Parker is a yacht designer, custom boatbuilder, and author of three books on boat design, construction, history, and cruising. He has been building boats to his own designs since 1958 (at age 12) and is a lifelong cruising sailor under sail and power.
