The Concept
The Boat
Bringing Her Home
Progress Log
The Boat Barn
Other Sites
Email Tim





Progress Report:  2004 Archives

Archives Menu     Main Project Menu

January  February  March  April  May  June  July  August  September  October  November  December

Reports from December, 2004
12/5/04  12/12/04  12/19/04  12/26/04

Log for the Week Ending December 5, 2004

With things finally settling down after thanksgiving and our vacation before, it was great to get back to the shop for some real work.  My immediate goal was to complete the short list of tasks that needed to be done before the next major step on the process:  fiberglassing the deck.

Click here to read more about the final steps leading up to glassing the deck.

I also worked on the interior, roughing in the plywood platforms that formed the tops of the settees and v-berths, and installing access hatches in the new platforms.  As of this writing, the new platforms were awaiting permanent installation, which you can read about next week.

To see the construction of these platforms, click here.

Log for the Week Ending December 12, 2004

It was a busy week in the shop as I made preparations for one of the larger projects on my list:  fiberglassing the deck surface.  I had lined up some help for Friday, and spent some time over a couple days measuring and preparing fiberglass cloth for application, and obtaining other materials needed for the job.  Finally, on Friday morning spanning a period of 3.5 hours, we applied both final layers of fiberglass to the deck:  a 15 oz. layer of biaxial structural fabric beneath, covered with a layer of lighter 7.5 oz. cloth over the top, with staggered seams.  When the fiberglass had cured, I spent Saturday sanding the new surfaces to prepare for fairing and later construction steps.

Click here to read more about the fiberglassing.

Although fiberglassing and the necessary preparations consumed most of the time I had available for shop work, I also completed additional work on the new settees and V-berth, as I closed in on their ultimate completion.

Click here for more.

I must be getting serious about finishing this boat, because on Friday (during about the busiest portion of the fiberglassing process, of course) I received a large shipment of marine plywood and mahogany needed to build the cockpit, cabin trunk, remaining interior structures, and trim work.  A full wood rack is always a good sign!

Included with the shipment were some very long, wide boards that I ordered to make the cockpit coamings and portions of the cabin trunk.  These boards are about 12" wide and over 17 feet long; planing promises to be a challenge.

Looks like there just might be some more action in the weeks ahead.  See you next Sunday!


Log for the Week Ending December 19, 2004

Monday, I dove right into work with an unsavory task:  planing out some of the rough mahogany boards I had received Friday.  However, I simply couldn't justify putting off the job:  I needed the wood to continue with work on the boat.

As unpleasant as this chore always is, this time seemed even worse.  I selected about 10 of the boards for planing at this time--enough to build the cockpit support frame structure with some left over for the inevitable unplanned needs.  I set up the equipment, donned my safety gear and respirator, and plunged in.  Immediately, I was annoyed, as the boards, all ostensibly 4/4 rough stock, were of a wide disparity of thickness, meaning that with the planer set at a given height, some of the boards would not pass through the opening, while others slipped through and didn't even hit the knives.  This greatly extended the whole process, as it took numerous passes through before all the boards began to be planed on even one side.  I also had to cut a small portion off one end of a particularly wide board after it became stuck in the planer, as it was over 12" wide in that one spot.

Whining aside, I managed, over a period of about 2.5 hours of constant shuffling, to plane all 10 boards down to a consistent 7/8" thickness, and smooth on both sides.  Since two or three of the original boards were nearly 1-1/4" in the rough, and the others about 1-1/8", this process left sheer mountains of planer chips and clouds of dust behind.  After filling three empty trash cans to the brim with the chips, I cleaned the whole shop and blew the dust off the walls, light fixtures, boats, and other areas it had invaded.  Then, I set up to cut one straight edge on each board, using a long straightedge made from a piece of plywood and my circular saw.

With the prepwork out of the way, I could finally begin milling some of the pieces into the sizes needed for the cockpit structure.  I needed two longitudinal laminated supports a foot or two off the centerline, plus two additional laminated supports--curved to match the shape of the deck and carlins above--for the outer edges of the cockpit sole.    Since I didn't wish to install supports and bracing to the hull beneath the cockpit, I planned on some rather deep laminated beams for the center supports--5.5", with 2, 7/8" laminations--and 2-3" deep boards to make up the laminated and curved outer supports.  I milled the various pieces needed and set the remaining planed boards aside for later use.  Then, just before the end of the day, I laminated the two beam blanks needed for the center cockpit beams.

Click here for more detail on the cockpit structure construction.

Tuesday, I continued with the cockpit.  Over the course of the morning, I cut, fit, and installed the two new longitudinal beams.  Then, I moved on to the edge beams and, by the end of the day, had glued up the two beams necessary.

Click here for more.

Wednesday I removed the two curved edge beams from the clamps and repeated the preparation and installation process that I did the day before with the centerline beams, though the cuts on the after ends of the edge beams were somewhat more difficult.  By the end of the day, I had epoxied the beams in place and left them to cure overnight.  Thursday was a more abbreviated day in the shop, as I had some other obligations, but I managed to spend a few hours tabbing all the beams in place with biaxial tape and epoxy--a step I really wanted to make so that I could grind the new tabbing Friday and be ready for installation of the cockpit sole.

Click here to see the edge beams detail.

Friday, after an hour's worth of grinding on the new tabbing, I prepared to cut and fit the various pieces of plywood needed for the cabin sole.  This took a surprisingly long time, not the least of which was because the plywood was so heavy it seemed to have been made of lead.

Click here to see the cockpit sole.

I also completed a couple smaller tasks during the week.  Once I had the first of the cockpit support beams installed, I had no further need for the small bit of the original cockpit platform that had been fiberglassed to the rudder tube.  This small bit of the original boat had remained in place; I left it simply to provide me with a useful reference as to the original cockpit sole height.

But with the beams installed permanently, I could finally cut off the excess tube.  Later, I'd glass on a new length of tube to extend it to the new level required.

I also decided to go ahead and enlarge the companionway opening in the amidships bulkhead.  Much earlier, I had cut a temporary opening that was a bit smaller than the one I eventually planned.  Since I had my big router out for part of the cockpit sole job, I decided to install some cleats alongside the rough opening and use the router to cut a new, clean opening.

Thus endeth the week.

Log for the Week Ending December 26, 2004

Merry Christmas!

Now that the three pieces comprising the cockpit sole were in place and cut to fit, my next step was to prepare them--and the surrounding boat--for their final installation.  I moved a little slowly at this stage, as I didn't want to forge ahead and regret some lack of access or another in the coming days and weeks, so I carefully thought through each stage of the process that I was anticipating.  However, I needed to get the pieces installed, as several deck-related projects hung in the balance until I could get the cockpit sides installed.

Before I could think about installing the cockpit side pieces, I had to permanently install the cockpit sole.  Before I could do that, I needed to be sure that I would have adequate access to the bilge area beneath the cockpit for all future tasks.  This area would also be where the engine is installed, and the access needed to be adequate enough for its installation and maintenance.  Therefore, my first step was to concentrate on how and where to locate the access hatches.  Obviously, I didn't care to make a mistake at this point and either box myself in (or, more specifically, out), or make some future job more difficult or, worse, make an incorrect cut on the plywood sole that I might need to repair later.

Click here for more about the cockpit sole and hatches.

After the hatches were cut, I decided to go ahead and install a platform to secure the eventual fuel tank and batteries.  I researched the tank options in a catalog, and found one of appropriate size and shape.  With these measurements, as well as the basic measurements of two battery boxes, I could cut and install a plywood platform for the tank and batteries.

Click here for more on the tank platform.

I had anticipated getting significant work done during the week, but on Monday and Tuesday I found myself waylaid with a miserable cold/flu virus type thing.  I rarely get sick to the point that I cannot function, but this was one of those times.  After a full day of misery and rest on Monday, I had high hopes for Tuesday...but, unfortunately, I simply couldn't face working in the shop on Tuesday, either; despite feeling better early in the morning, my condition deteriorated and I felt miserable once more.  What a waste of a couple good days.

Wednesday, although I was still far from the picture of perfect health, I could finally get back to work.  I began by working on the rudder tube, which needed to be extended from its present height.  Some fiberglass tubing I had ordered arrived late Tuesday, so I set right to it.

Click here to see more about the rudder tube.

With the rudder tube glassed in place, I continued by beginning to paint out the inside of the hull and the remaining structural members, including the bulkheads, bottoms of the deck beams, and the cockpit support beams.  The process was more challenging than I had anticipated, partly  because at the time I was armed with only a puny 3/16" nap paint roller--far too fine for adequate rolling onto the fiberglass hull--so before I was even  halfway done--having completed only the sections aft of the aftermost bulkheads--the paint fumes were so strong that I had to call it quits for a while.  I returned the next day with a fluffier roller cover (3/4" nap) and finished off the painting (wearing a respirator this time) that needed to be done before I could install the cockpit sole, which was the main issue at hand.  Obviously I had only thought I was feeling better on Wednesday, but my poor performance at work that day would seem to indicate otherwise.

For now, I left some areas unpainted, most notably the center portion of the bilge in way of the eventual engine foundations, where additional fiberglass work would be required.  However, all of these areas would be able to be easily reached even after the cockpit sole was permanently attached.

As it turned out, the nasty virus dogged me for the remainder of the week, but fortunately to a lesser extent than Monday and Tuesday.

Thursday marked the 2nd anniversary of the boat's trip home to Maine.  She's come a long way since then.  And just last year at this time she was only an empty hull, with nothing inside whatsoever.  Now the major structural work is nearly complete, though ultimate completion of the boat is still much work ahead.

2005, here we come! 

Continue to January, 2005>


All photos and text on this site 2002-2009 by Timothy C. Lackey and Lackey Sailing, LLC
All rights reserved.