Deck Core Repair, Chain Locker

The completed deck core repair job, as seen from inside the chain locker
There are numerous articles on this website that concern the construction of new items for the Ericson 25. Let's see . . . there are the multi-part articles on the construction of a new centerboard and on the construction of a new companionway hatch, and there are shorter articles on the construction of various items such as bookcases and tables. All of those projects were time-consuming and, at times, tedious, but they were always pleasurable, because they all involved some thought and some imagination. Yes, all of those projects concerned the construction of new items, not the repair of old ones. Repair work on a boat (or on a house, or on anything) can, at times, be pleasurable, simply because it generates some satisfaction that a problem was identified and addressed, but repair work in general can never rival the feeling one gets by the creation of something entirely new. Repair work can also, at times, be less than pleasurable, and even irksome. I must say, that of all the irksome repair work that I have completed on a boat, or on a house, or on anything else for that matter, none has been more irksome than the repair of the rotten deck core in the chain locker of my Ericson 25. May Poseidon be kind to me and grant that I never be required to give myself over to such a damnable enterprise, ever again.

The story of this hellish endeavor began simply enough with the desire on my part to install an anchor roller on the bow of the boat. I'll address the installation of that anchor roller and its related hardware in another article, but for now, we'll focus on the subject at hand.
In order to use the anchor roller I knew that I would need to install a new chain pipe, one that was situated several inches aft of the existing chain pipe.
I also knew that I would need to install a piece of mahogany forward of the chain pipe so that I could provide a fair lead for the chain, and so that I could install a chain stopper between the anchor roller and the chain pipe. Here, I have a piece of pine in place as a test piece, just to get a sense of how far back I will need to install the chain pipe.
If you've read my article, "Deck Hardware Removal," you'll recall that soon after I transported Oystercatcher from Oriental, North Carolina to Charleston, South Carolina, I removed almost all of the deck hardware. I tented the boat soon thereafter, but to make sure that no water entered the deck hardware holes in the event that the tent collapsed or leaked, I filled most of the holes with caulk. Here you see clumps of dried caulk and a deck covered with lots of sandy dirt from the yard and driveway.
My plan was to remove the old deck plate and fill the hole with layers of epoxy-soaked cloth.
My first indication that there was trouble on the horizon was the shoddy hardware used to install this plastic chain pipe. There had been two prior owners, and one of them had chosen to cut corners with the hardware. Instead of spending a few dimes back in the 1970s for some consistent stainless steel hardware, this guy had used whatever he could find in his toolbox or junk drawer - wood screws, machine screws, and left-over electrical hardware. My guess was that if he cut corners with the hardware, then he probably didn't bother to excavate the balsa deck core around the perimeter of the hole and fill it with epoxy in order to prevent water intrusion.
My hunch was confirmed, as soon as I removed the plastic ring. The deck core was not white, as it should have been, but black.
It was black in all directions around the perimeter.
Observing the problem, I decided to experiment. I grabbed a putty knife and shoved it between the upper layer of fiberglass on the deck, and the lower layer of fiberglass in the chain locker. There was absolutely no resistance as I shoved the knife forward. The blackened balsa core that I had observed was absolutely rotten. It was so rotten that it was little more than powder.
The sight of this putty knife fully inserted into this slot filled me with gloom, but I was heartened to some degree when I shoved the knife in the opposite direction and met resistance after about an inch or two. This gave me some hope that the rot was limited to the area immediately surrounding the hole.
Now let's take a look at the chain locker. One thing I found strange about this chain locker from the very start was the fact that it had been painted at some point with a yellowish paint. No other part of the interior of this boat had ever been painted. Why, therefore, I thought, would anyone bother to paint such a remote and difficult-to-access space that was always concealed by a mahogany panel that was bolted into place? I could only guess, but my guess was this: the prior owner knew that this area of the boat leaked, and he, for some reason, thought that by painting the underside of the chain locker he could stop the leak. Why he thought this, I do not know, but there were other screwy patch jobs here and there on this boat that I had already had to deal with, so it was not at all surprising.
Another reason that I suspected the prior owner was conscious of this problem was the fact that he had hovered around the bow of the boat when the surveyor and I were standing on the foredeck, back at the time of the survey, when I was trying to decide whether or not I should buy this boat. At that time, the surveyor walked around all areas of the deck with his moisture meter in hand, trying to determine if there had been any water that had entered the balsa deck core. I had a crystal clear memory of the surveyor and me standing on the foredeck talking about the condition of the deck. The surveyor was quite pleased, because his moisture meter showed that the core was dry, which was not always the case with old boats such as this one. He told me to hold out my hand. In it he placed the moisture meter. The meter indicated that my hand, which was perfectly dry by my reckoning, was wet, at least when compared to the deck core, which was incredibly dry.

So how was it, then, that the surveyor had not detected the rot at the bow? I puzzled over this for a while, and then one day during this core-repair project it suddenly came to me. I remembered that the owner had gotten down on his knees and fiddled around with something at the bow while the surveyor and I were talking about the moisture meter. I had forgotten that the owner had been there at that time, and I think that he had been there for a reason. In keeping with boat-buying etiquette, the owner should have given the surveyor and me plenty of distance, but we were on his property and his dock, and given his age and his health, I wasn't up for telling the old timer to get the hell away from his own boat while we scrutinized the hell out of it. Maybe I should have, but I had the peace of mind then, as I have the peace of mind now, that I got a great boat at a great price, as a result of that survey, and I doubt that I would have walked away from this boat, even if I had learned about the problem at the bow.
In the picture below, you can see the problem area in greater detail. Notice how much more paint is missing in the problem area than anywhere else.
Forward of the chain pipe hole there was a hole for a 12 volt receptacle. I guess the prior owner had used the receptacle, at some point to power a spot light or something of this nature. At any rate, it had long since become non-functional, and I had removed it when I removed the rest of the deck hardware, soon after the purchase of the boat. The three holes on either side are, of course, for the pulpit hardware.

Knowing that this entire area appeared to possess a rotten core, I would need to remove this layer of fiberglass, so that I could dig out the rot, insert some new core material, and create a new fiberglass layer to hold it in place.

Some people choose to do core-repair work not from the underside of the deck, but from the topside. This is certainly easier, but it really wasn't an option, given the location of the rot on this boat. Removing the fiberglass bow and replacing it with a new fiberglass bow of my own creation would have been a daunting task.
This problem area was almost an arm's length distance away from the V-berth, and I figured out that the only way I could cut out the layer of fiberglass would be to use a Dremel with an attachment made for cutting fiberglass. I had used to attachment elsewhere on the boat, and it had been very helpful. You can also see in this picture a cutting wheel sitting nearby. The reason why I couldn't use the wheel was because it did not provide the proper angle of attack.
The first two cuts that I made went very easily. I started by placing the cutter on the lip of the large hole and pushing it forward toward the small, 12 volt receptacle hole. The core was powdery and, for the most part, non-existent in this area, so I didn't meet much resistance. This operation, however, did require careful attention and some degree of finesse. I took great care not to press the cutter too hard against the work surface or push it too hastily through the fiberglass layer. One careless flub, and I could have damaged, or at least weakened, the upper layer of fiberglass, i.e., the deck.
Fortunately, the deck was considerably thicker than this lower fiberglass layer, so it was tolerant of some gentle grazes with the cutter every now and then.
As the removal of the fiberglass layer proceeded, the task became more painstaking, because my arm and shoulder grew weary of this work. I had to sit in a crouched position in the V-berth with my arm extended into the chain locker.
The farther I moved outward from the area between the large hole and the small one, the more core material I encountered and the less powdery dust. That was encouraging, because it appeared to indicate that most of the rot was in the area between the two holes.
It was especially encouraging when I moved the cutter to the starboard side of the large hole. It was then that I began to hit good core material. I knew it was good, because it was white and it was not easy to remove.
Even though I had hit core material that was perfectly fine, I still needed to remove it on account of the bad core material that existed in the area. You can see the bad core material just forward of the good stuff. This dark brown balsa was not powdery, but it was not strong. I could poke at it with a putty knife and it flake off.
If you look closely in the picture above, you'll see some stainless steel nuts and washers at the forward end of the chain locker. These were securing the stemhead to the bow. I didn't want to do this, but it appeared to me that I needed to remove this hardware so that I could fully excavate that area with the Dremel and the fiberglass cutter. This area of the bow was important, and I had to do this repair job properly, so off the stemhead came.
The stemhead had been installed by Ericson, and I don't believe that any water had ever entered the bow on account of the deck penetrations of this hardware. I believe the rot in this area was simply the result of the water creeping into it from the other nearby areas.
On either side of the bow there had been navigational lights. These I had removed back when I had removed the deck hardware, soon after purchase. I dug the caulk out of these holes and inspected them. It appeared to me that water had also entered the bow through these areas.
From what I could tell, this beefy area at the front of the bow consisted of a very thick layer of fiberglass over a thick section of core material which itself sat on a thick layer of fiberglass at the deck level. I couldn't see this thick core material from inside the chain locker, because it was encased in fiberglass. I could, though, feel it with my finger when I felt around inside of the hole. I poked around with a Phillip's head screwdriver inside of the hole, and this caused a little bit of the material to flake. Fortunately, however, it appeared that the core as a whole was solid and that the water that had entered through these holes had been led downward into the chain locker or into the regular deck core material in this area. My guess is that the wires that ran downward through these areas caused the water to trickle downward, away from the very thick core material at the front of the bow.
Just to be certain that there would be no water or moisture inside this very thick core at the front of the bow, I poured acetone, in small amounts, into all these holes for several days in a row. I had read in reputable sources that this was one way to extract moisture from deck core material.
The picture below shows the area beneath the stemhead after I had excavated the core material. You can also see that I had, at this point, also attacked the area to the left of the large hole.
Before going any farther in the excavation process, I thought it would be smart to pause, so that I could figure out where exactly I wanted to cut the new hole for the new chain pipe. Wasn't this what got me started on this project? At any rate, I ended up deciding to orient the chain pipe in a forward-to-aft direction rather than port-to-starboard, as I had originally planned. This forward-to-aft direction made much better sense when I took the non-skid surfaces into consideration. The chain pipe, oriented in this fashion, would fit well between the non-skid.
After double-checking and triple-checking my measurements, I drilled a test hole in the center of the oval. I was really worried that this area might be infected with core rot, but I was pleased to discover that this first hole generated a lot of white core material remnants.
Then I drilled a pilot hole along the edge of the oval, so that I could insert a saw blade.
Again I was pleased to see what appeared to be nice, white core material in this hole.

I then took the plunge. I grabbed the jig saw, plugged it in, and slowly made a cut along the penciled oval.
Presto. One new chain pipe hole in the deck. There was no turning back. I was fully committed.
Yes, I was overjoyed to see solid core material sandwiched between the two layers of fiberglass.
I was, though, a little puzzled by the darkness of the core in some spots. This could not have been from the saw. I had cut another area of this boat with this same saw, and I had not seen any friction burn marks on the core.
I probed all of these dark areas with the putty knife. All of them were solid. Even the area between the forward end of the oval and the old circular hole was solid. All I could figure was that at some point some water had seeped into this area, but not enough water to cause core rot or delamination of the fiberglass skins.
The new chain pipe looked pretty good when I dry-fitted it in the new cutout. Now, though, I needed to get back to business fixing that core rot.
Here's how the new hole looked from down below in the chain locker. At this point, I was hoping that there wouldn't be much more work to do in terms of removing the fiberglass and core material. I was hitting good core material on either side of the round hole, so it looked like I would be able to get away with making a diamond-shaped pattern out of the cut-out area.
Below you see the fully-excavated diamond-shaped area.
The whitish spots you see within the cut-out are areas were I had to remove good core material. It clung tenaciously to the deck above, and parts of it were still clinging here and there. Based on the patterns of rot that I could see, my best guess as to the nature of the rot was this: water entered in the small 12 volt DC receptacle hole and water also entered around the circular plastic chain pipe hole. Since the core area between the 12 volt hole and the circular hole was so rotten that it was non-existent, the water from the small hole appears to have regularly drained downward toward the large circular hole. That water, together with the water that the circular hole itself admitted, probably dripped down into the chain locker at this point. The plastic chain pipe, after all, had a plastic flange that projected downward into the chain locker. This probably directed the water downward at this point. This was why the rot was limited to this area.
The last task I faced at this juncture, was the digging out of any remaining core material along the edges of the cut-out, and also roughing up the fiberglass with sandpaper, in order to prepare it for the epoxy work that was coming.
This picture should provide some testimony to the hellishness of this repair project. About halfway through the core-removal stage of this project, I figured out how I could get my torso into this chain locker so that I would work with at least one of my hands overhead. Do you see that board? I would lean back into this hole with one arm outstretched. This would be my working arm. If I didn't put my arm in the hole first, then I wouldn't be able to put it in there later, after I had leaned my torso into that space. Once I had gotten my arm in there (holding any tool that I wanted to work with), then I would lean my torso inward at an angle (so my shoulders would fit through the opening). Then, when my shoulders were inside I would lean my torso backwards onto the board. Of course I was wearing this full shield face mask to protect my eyes and face from the dust and falling debris. I was, of course, also wearing an organic vapor protective mask to protect my lungs. Oh, and let's not forget the long sleeve shirt to protect my arms from the fiberglass. Last, but not least, I should mention that I did all of this in August in the Lowcountry of South Carolina. High temperatures, high humidity, and lots of of mosquitoes were part of the mix.
Up until this point I had avoided removing the two large cleats on either side of the bow. The core rot did not extend into these areas, but my epoxy work would likely come into contact with the nuts and bolts that held these cleats in place. Therefore, I had no choice but to remove this hardware. It was either now, or never.
It wasn't easy getting to those nuts on the starboard side. They were very close to the hull.
The nuts on the port side were a little more accessible.

It was impossible to remove the stainless steel screws from these aluminum cleats.
Fortunately, I was able to simply lift them out of the deck after the nuts had been removed.

The next task was to cut a piece of cardboard that would fit into the diamond shaped area that I had excavated. My plan was to use the piece of cardboard as a pattern for the piece of plywood that I would use as the replacement core material. It took many trips back and forth from the chain locker to the cockpit where I did the cutting of the cardboard. Eventually, I ended up with a piece of cardboard that fit just right into the cavity.

I had quite a bit of 1/2 inch exterior grade plywood around the house, but it was too thick. I had a little bit of 3/8 inch exterior grade plywood, but, of course, not quite enough for this job. Therefore, I had to buy an entire sheet of plywood for this one small repair job. Fortunately, I later found a few different ways to put the rest of it to good use.
I had considered using balsa wood as the core replacement material, but some workers at a local boatyard said it would easier and better to use plywood. Therefore, I followed their advice.

A little more sanding, and the cavity was ready for a dry-fit.
Do you see the black stuff up there at the tip of the chain locker? That's burned rubber. That area was too far away for me to reach with the sandpaper. Therefore, I used the Dremel with a sanding wheel. I went through several of these, and eventually I burned up the rubber drum beneath the sanding wheel.
The plywood fit well inside of the cavity.
When I pressed the plywood upward with my hand, it sat either flush with the cavity or slightly recessed within the cavity, depending upon the slight variation in the cavity itself. This was as good as I could get, since this would allow room for the layers of cloth and epoxy that were one the way.
Another thing that the boatyard workers advised me to do was to cut kerfs throughout the length of the plywood so that it could more easily conform to the curvature of the deck above.
When I was finished, I could easily flex the plywood. Nevertheless, it still seemed rigid and strong.
Next, I epoxy-coated the plywood so that it would be as protected as possible from any future incursions of water.
I used RAKA brand epoxy for this clear coating, just as I used RAKA for many other clear-coating projects. It's much more affordable than MAS epoxy and far, far, far more affordable than WEST Systems. Like they say on their website, the people at RAKA actually answer the phone when you call, and they give you lots of good advice. When I say they answer the phone, I mean they actually pick up the phone when it rings. You don't have to navigate your way through a stupid, automated phone tree.
I didn't clear-coat the top side of the plywood. Why? Because I wanted to clear-coat it at the same time that I applied the the thickened epoxy. This way I would form one single chemical bond through all the epoxy that I applied. If I had clear coated it in advance, then I would have had to sand it prior to applying the thickened epoxy, and I would only have achieved a mechanical bond.
When the day came to install the new plywood core, I called upon a buddy down the street to give me a hand. I needed someone to be on deck while I worked down below. The local boatyard workers had told me that I should approach the task in this way. They said that someone needed to be above in order to put screws through the deck and thus through the new plywood core material. They also said that someone needed to be down below to tighten the nuts on the screws. They said that without the screws this project stood a good chance of going bad. They added that I would likely end up with some voids between the deck and the plywood, and that if this happened, the strength of the deck, of course, would be compromised.
I wasn't able to take any pictures of the initial parts of the glue-up of the plywood core. This, though, was how I did it. I first mixed up about 6 ounces of MAS F.L.A.G. Resin and Medium Hardener and thickened it with colloidal silica until it was the consistency of peanut butter. I did all of this in the cockpit. Then I carefully carried the pot of epoxy to the V-berth, and with a chip-brush in hand, I dabbed as much epoxy as I could into all the cavities around the perimeter of the main cavity. All of the left-over epoxy and brushed onto the underside of the deck. I then mixed up 9-12 ounces of epoxy in the same fashion and brushed it all over the topside of the plywood as it held it in my hand in the V-berth. Then, with my face shield on. I leaned my torso up into the chain locker and pressed it upward into the cavity. Bracing it as firmly as possible with my hand, I told my buddy to start drilling the screws through the deck and plywood. Some of the screws he used were wood screws. These really helped to grab the plywood and pull it upwards. He also screwed in some of the machine screws, and as he did so, I began to install the nuts. Just like the boatyard workers had said, these screws made a big difference. I got good squeezes of epoxy out of the edges of the plywood, and my buddy reported the same thing with regard to some of the holes on deck.
I used the chip brush to brush the excess epoxy smoothly and uniformly around the perimeter of the plywood. It was impossible to apply anything else to the plywood at this time, on account of the screws. I had no choice but to let the epoxy fully cure, then sand it, and then begin the layup of cloth. In other words, I had no choice but to sacrifice the chemical bond in favor of the mechanical bond. The boatyard workers had told me, however, that this was an acceptable trade-off. Better to get the plywood well situated without any voids. That's what they had said, so I went with it.
My buddy had used some scrap pieces of wood to help support the heads of the screws that he had put through the deck.
He also had to grab some of the old chain locker hardware that was still sitting around on the deck. This took more screws than we had anticipated.
There were enough existing holes in the deck, so he didn't have to drill any new ones, thank goodness.
I left the screws in place and let the entire project sit idle for one week. The next weekend I got to work on the final phase of the project - laying up the cloth. First, I cut two pieces of bi-axial cloth, one a little bit larger than the other. The larger of the two would end up being the very last layer of cloth. The smaller of the two would cover the diamond-shaped layers that I would initially lay up. In other words, these two pieces of bi-axial cloth were the last two layers that would go on. I used the negative of the cardboard pattern to judge how large to cut these triangular pieces. After trying out these dry piece of cloth inside the chain-locker, I realized that this triangular pattern would not fit well, so I ended up trimming the corners to make these pieces more diamond-shaped.
For the diamond-shaped layers that would cover the plywood, I used four pieces of 10 ounce cloth.
I cut all four of them to the same diamond shape.
Just as was the case with the glue-up of the plywood one week earlier, I used MAS brand epoxy for this job. Specifically, I used F.L.A.G. (Filleting, Laminating, and Gluing) Resin and Medium Hardener. I also used colloidal silica as a thickening agent. I would not have been able to complete the lay-up, if I had used the RAKA Low Viscosity Resin and the 350 Non-Blush Hardener that I had used for clear-coating the plywood. I've used these same RAKA ingredients with colloidal silica for gluing-up things, and I think it does a fine job. On this particular job, however, I needed more time, and the MAS formula I describe above has a much longer open time. The RAKA 350 Hardener kicks a lot more quickly, and I'm certain that if I had used the RAKA formula on this job, I would not have been able to lay up all these layers of cloth without it all turning into a sticky, unworkable mess.
Having removed the screws and sanded the plywood, I laid up the first two pieces of the diamond-shaped 10 ounce cloth. I believe I used 6 ounces of epoxy to wet-out these two layers. 
I had to use the same technique (described elsewhere in this article) to do this work. With a face shield on, and with a long-sleeve shirt on, I lay on my back and worked overhead with one hand holding the epoxy-soaked chip brush. You'll recall from the other descriptions I have given, that I could only use one hand at a time inside of the chain locker. There just wasn't enough room for both arms. Therefore, while I dabbed on epoxy with the chip brush in one hand, I held onto the plastic cup of epoxy with the other hand by extending my arm aftward into the V-berth. I started out trying to wear a baseball cap, in order to prevent the occasional drips of epoxy from falling into my hair. The bill of the baseball cap kept getting in the way, so I had to take it off. Did I get epoxy in my hair? Yes. Did I have to cut out a chunk of my hair after this whole project was finished? Yes. Did I have some explaining to do to my wife, when I was getting ready for our wedding anniversary date later that evening? I'll let you be the judge of that.
After crawling out of the hole and mixing up another 6 ounces of epoxy, I climbed back inside and laid-up the third and fourth layers of 10 ounce cloth.

Having completed the lay-up of the four layers of 10 ounce cloth, I climbed back out, mixed up 6 more ounces of epoxy, grabbed the first layer of bi-axial cloth, and climbed back in.
Bi-axial cloth is stronger than regular fiberglass cloth, not only because of the orientation of its fibers, but also because it soaks up lots of epoxy and makes for a nice, thick layer of new fiberglass.
Satisfied with the first layer, I returned to the cockpit to mix-up six more ounces of epoxy and grab the last layer of bi-axial cloth.
This second layer did a good job of covering the smaller layer of bi-ax, and the four layers of 10 ounce cloth.
I could still see quite a bit of the fibers in the bi-axial cloth. It looked to me like it needed to have another 6 ounces of epoxy to fill in the weave.
I returned from the cockpit with that additional epoxy and began to apply it.
That additional 6 ounces of epoxy really did the trick. It filled in the weave and it created a smooth, epoxy surface over the entire work area. The streaks that are visible here and there are mostly just brush marks. I was finishing just in the nick of time, because the work area was becoming more and more sticky as the earlier applications of epoxy were beginning to kick.
When all was said and done, I was pleased with the results, despite the fact that there was nothing whatsoever that was pleasing about the steps I had to take to get to this end point. Yes, this core-repair job inside of the chain locker was now over, so at last I could return to the  original task at hand - installing the anchor roller. I will describe that project in a separate article. The End.