Hull-deck joint - update

After the long north country winter, I began work on the hull-deck joint recently. In my earlier post, I said that I planned to grind off the outside surface of the aluminum H-channel, grind the hull and deck to a 12:1 taper, then laminate the joint to bond the hull and deck permanently. When I actually tried to do this, I discovered that the aluminum channel is a lot thicker and stronger than it looks. Grinding it off would require specialized equipment and would take a long, long time. Time to rethink.

Here's what actually works. First, I used an angle grinder and 40-grit flap discs to remove the paint and oxidized crap from the outside of the channel. I got it as smooth as I could, but there are many pitted areas, so the end result is far from smooth.

Next, I used a reciprocating multi-tool:

and a plunge cut blade like this: 
 

 

Holding it at a 45 degree angle, I sawed open a gap between the H-channel and the outside of the hull. There was a fillet of epoxy (or maybe gel coat) along the edge, and after cutting through that I was able to haul out the old mostly dead caulking material. The rivets get in the way, but I dug down with the tool as deeply as I could, and with a circular motion hauled out the crud. The resulting gap was about 3/32 wide. I then used compressed air to blow out the dust and dead caulk crumbles. I performed this feat along the top and bottom edges of the channel. All told, it took about 16 hours of work to grind off the paint and clean out the old caulk.

I then had to choose a material to re-seal the freshly-opened gap between the channel and the hull. I first tried Life-Calk liquid. My hope was that it would be runny enough to seep into the gap. It was too thick: it just lay along the top of the gap without penetrating at all. I then tried low-viscosity epoxy (Total Boat Penetrating Epoxy). I was thin enough to run down into the gap! I don't know exactly how far, but it certainly went down as far as I had been able to clean out - between 0.5 - 1.0 inches. A nice plus with this material is that even when fully cured, it remains just slightly flexible. My hope is that whatever minute flexing there may be along the H-D joint won't crack the epoxy.

The low-viscosity epoxy would only work along the top of the channel, because there's no way to force it up against gravity along the lower edge. For that, I used regular epoxy with thickened with silica cabosil to the consistency of peanut butter. I used a wooden tongue depressor to push the epoxy up into the crack, then ran the curved end along to produce a fillet. I know this will not be as solid a seal as the top edge, but my main leakage problem related to rainfall -- and waves breaking over the bow and washing over the deck. I don't often have my lee rail under water, so it is very rare to have water forced upwards onto the H-channel.

To finish things off, I used fairing compound (similar to Bondo, but designed for marine use). I coated the outside of the H-channel to seal the aluminum, and continued it onto the hull above and below the channel. I used a tongue depressor to produce a broad fillet between the channel and the hull.

When I tried out the fairing compound, I realized I had to do additional prep on the upper (deck) side of the joint. Here the PO had painted over the gel coat, and the several coats of blue paint was loose in many spots. I therefore used my angle grinder and 40-grit flap discs to remove the old paint between the top of the channel and the turn of the deck. This is a stripe about 1 1/4" wide all along the top of the channel. I ground down into the gel coat (and in some places all the way to the laminate). I covered this all with the fairing compound to produce a nice smooth surface. All this will take a lot of sanding and several additional thin coats of fairing compound, but the end result, when painted, should look very nice.

The main problem with this technique was the runny epoxy. I applied it with large syringes to try to confine it to the gap, but it always overflowed and ran down the outside of the channel. The stuff also ran out of the screw holes (where the decorative strips had been attached). It's very runny, and drips down the channel -- and onto my lovely AwlGrip hull! Waaaaah. If you wait till the stuff cures, you can use a razor blade to carefully lift the epoxy runs off the hull, but it's painstaking work. I needed a way to avoid getting the stuff on the hull.

DUH! you say. Just cover the hull with a wide poly drop-cloth taped to the channel at the top. Tried that. The damn poly would detach from the masking tape and fall away about half the time -- just as I was applying the epoxy to that section, of course. So I needed a different method.

My (ahem) brilliant solution was to attach lengths of 3/4" (inside diameter) PVC tubing right along the middle of the channel. I held the tubing in place with pieces of masking tape about every three feet or so. The tubing is quite light, so it held just fine. I then ran 2-inch masking tape along the channel and over the top of the tubing, forming a sort of shed. As you can see, the drips of epoxy run out over the tape, down the tube, and drip off the bottom of the tube -- 1/2" clear of the hull. Not a drop hit the hull! After doing the starboard side this way, I hit upon an improvement: the final word in using low-viscosity epoxy to seal an H-channel hull-deck joint. Before attaching the tubing, run 2-inch masking tape along the top of the channel, with half the width on the channel and half on the surface of the hull (actually, it's the deck piece above the channel, but I'll just call it the hull). Then attach the tubing as before.

The upper gap is now sealed with the tape, so you need to pry it open. Do this carefully, and you create a trough between the tape and the hull, as shown here. You can now use the syringe to apply the liquid epoxy, and it seeps down nicely into the gap -- and the overflow just stays in the trough! This is doubly wonderful: first, very few drips, and second, the excess forms a reservoir -- like a long ditch full of liquid epoxy. As the epoxy works its way into the gap and seeps down into wherever it can, the excess in the trough provides additional material. Yes, when the epoxy hardens, you have a long strip of material that has to be removed, but the angle grinder and flap disc make short work of that.

What's especially nice about this trough is that you can use fast-set hardener and mix up enough material to do the entire side of the boat in a single batch. Use a big fat syringe (I got the biggest one Jamestown sells). Mix up about 8 oz of resin and 4 oz of hardener (2:1 ratio), and work fast. Fill and refill the syringe, running it along the trough as fast as you can crawl along on your knees. You can do this so fast, the big tub of material doesn't have a chance to cook off -- which it would do very quickly due to exothermic processes. I did the entire port side of the boat in about 5 minutes (after many hours of prep work, of course).

After I applied the epoxy, I used a tongue depressor to push the epoxy down into the gap, just in case trapped air might be keeping the material from seeping in.

I started work on Saturday morning, and by Monday afternoon I finished the top edges of the channel on both sides of the boat, and about 5 feet of the fairing work done. That includes all the grinding and other prep work. I didn't have time to do the bottom edge of the channel before a big line of thunderstorms rolled in. It rained TORRENTS for a couple of hours, then moderate to heavy rain all night long. In the morning, I checked inside the boat. NOT A DROP OF WATER ANYWHERE! I'll do some more testing with a garden hose after I finish the bottom edge of the channel and the fairing.

I figure I still have about 20 hours of work to finish completely. When I have the fairing all nice and smooth, I will paint the channel blue to match the original. I will not re-attach the stainless trim strips, because the screws just provide more ways for water to get in. I think just the blue paint will look fine. I can always add a rubbing strake sometime in the future if I want to really dress things up. But I'll see how I like the plain paint first.

I am so glad to have this hull-deck joint problem solved: it means I can complete the interior refitting without worrying about everything being ruined by sheets of water running down the inside of the hull, which is what was happening before.

Of course, the real test will be how long the fix will last. Unless the epoxy cracks, it should be pretty much permanent. We shall see.

More comments on CruisersForum

Several people have given me useful suggestions on Cruiser's Forum. Rather than duplicate them all here, please have a look at:

http://www.cruisersforum.com/forums/f55/major-refit-of-a-columbia-43-a-154520.html

Your comments are welcome there or here.

Shower Enclosure

Here's a detail drawing of the new head:

The shower stall has fixed walls, behind the green panels marked A and B. There is a fixed fiberglass shower tray that I will fabricate to fit around the bottom of the entire head compartment (except for the vanity). It will have a drain leading to a gray water tank immediately below.

The green panels marked A and B will be made from frosted Acrylic (or similar) panels. When the shower is not in use, they will be positioned as shown by the solid green lines in the drawing. Each panel has a piano hinge. Panel A is hinged at the corner closest to the vanity, and B is hinged at the forward port corner. To use the shower, you pull panel A out in front of the vanity and clip it. Then you pull panel B out. It slides into a  clip fastened to the back of panel A. Here's a detail showing this:

This creates a waterproof shower enclosure about 26" square (inside dimensions). When not in use, the panels swing back against the aft and inboard bulkheads. Panel B could have a towel ring glued on its outside face; the towel would stay dry during a shower.

Panel B might have to be split vertically in the middle, with a piano hinge. Otherwise, it might be impossible to open that panel while standing inside the shower (unless you're built like Twiggy). The hinge would solve that.

The drawback of all this is that the shower becomes a rigid enclosure 26" square. That's about 6" smaller each way than a typical residential shower stall. Even at this size, however, I think there will be enough room to bend down and pick up the soap if it falls. I'll have to verify this with a full-size mock-up, of course.

I have never seen this design, and I'm curious to hear what others think of it. The good news is that if the whole thing is just too nutty, I can simply remove the two panels and install a shower curtain. But I hate those pull-around sailboat shower curtains.

Interior Layout - My Ideas

My goals for changing the interior are:

1. Remove the nav station and storage units added by the PO.
2. Move the head to the base of the companionway.
3. Re-arrange the galley to make it safer and make better use of the space.
4. Get rid of the Pullman berths in the main saloon and install pilot berths. Move the settees inboard a few inches so that they remain wide enough to use as additional berths.
5. Install a new table.
6. Remove the old chart table (now child's desk). Remove the old head. Remove one of the hanging lockers. Out of all of this, create a stateroom with double berth, dresser, etc.

Here's what it would look like:

Starting aft:

1. Holding tank. 45-gallon (maybe larger) holding tank to go in bottom of starboard lazarette. I'll install one of those Groco Sweettank units.
2. Head. The shower area is about 26" square. There's enough room in the bilge right below it for an enclosed gray water tank (with automatic pump). I have a novel idea for enclosing the shower (see separate posting). The sink faucet is at the aft end of the vanity. This gives easier access to the behind-the-mirror shelves (not shown). The toilet and the head sit just at the edge of the deck house. The shower and entry door have 6' 4" headroom, but immediately above the center of the toilet, there is only about 5' head room. Ditto for the sink. This will work, because when you get off the potty, you can't stand straight up! You have to lean forward as you get up, putting you in the taller area.
3. Nav station moved to just forward of the head. The seat will be a permanent mini-bench with storage under.
4. Galley. The big change is to remove the existing dish locker area (see the original layout), allowing the stove and countertop to be moved outboard about 11 inches. The new sink (single basin, thank you) is on a peninsula. This will reduce the passageway from galley to main saloon by about 2 inches to a new width of 23" (plenty). The result is a compact standing area between the outer counter top and stove to the sink peninsula. It's 21" wide, enough to open the oven door (you have to move out of the way, of course) and work comfortably. The counter next to the sink is 21" wide, with a drawer stack underneath. On the aft face of the peninsula there is a flip-up counter extension. It will have a sliding bold to secure it to the forward edge of the fridge/hanging locker. This gives the cook another 21" of counter space while preparing meals. It does impinge slightly on the companionway ladder (which is not as wide as shown in the drawing. With the flip-up counter up, there will still be 19" of clearance for people using the ladder. On one of my Victoria to Maui trips, the boat I was skippering had this flip-up thing. It was great: you were completely enclosed, making it super safe to work in the galley in any conditions.
5. Main  saloon. The old layout has storage shelves outboard of the Pullman berths. Impossible to reach with the berths up, and too small to be useful in any case. By removing them, I can install pilot berths as shown. These are 24" x 6' 5". The settees will move inboard a couple of inches so that they are 24" deep. The saloon table will have drop leaves, and can be lowered to coffee table height (as shown in the drawing), or raised with the drop-leaves up for dining.
6. New Stateroom. On the starboard side, the head is removed. The V-berth area is modified by removing the hanging lockers. The bulkhead at the aft end of the new V-berth will extend across the full beam to preserve hull and deck strength. The double berth will extend into a cutout in that bulkhead. On the port side, the child's desk is removed, as is the hanging locker from the old V-berth. This provides room for a built-in chair with a night stand next to it (with a lamp on it and a drawer stack below). At the forward port side of the stateroom is a large built-in dresser.
7. V-berth. There's room for an angled double berth as shown. On the starboard side there will be a small settee with storage behind it (extending to the deckhead). The V-berth cabin shrinks because it loses its hanging lockers, but there will still be plenty of room for a pair of kids at anchor. For extended cruises, I would remove the mattress and install a work bench with vise, etc.

Up next, a semi-nutty idea for the shower enclosure...

Rebuild the Interior - the Original Layout

Here's the original layout of a Columbia 43.

The galley/dinette space opened into the main saloon with a half-height bulkhead between. This gave a very open look to the interior. The main saloon was rather plain; two settees whose backs lift up to form two more berths. The nav station was opposite the head. The V-berth cabin included a pair of hanging lockers and plenty of space.

The galley/dinette is under the coach roof (indicated by the curved line on the drawing. There is a 6-inch step down from there into the main saloon.

The PO changed the interior to make it more suitable for living aboard.

The dinette was changed to a nav station. The former nav station became a child's desk/toy box (no structural change - just a change in use).

There are a few things I don't like about the current layout. The nav station/storage area doesn't make very good use of space. The storage units are about 46" tall, with counter space on top and drawers and cupboards below. The extra storage is always nice, but a lot of the space is just air.

The main saloon is still too plain. The Pullman berths are a problem, because when an upper is pulled out, you can't sit on the settee. There is a table that folds up against the starboard forward bulkhead, and when dropped down is suspended by a pair of ropes clipped to the deckhead. It works well, but isn't as attractive as I would like.

The old nav station, now child's desk, would really only be used if a young family were living aboard. Yes, there's more storage there, but the desk surface is simply redundant, given the nav station added aft.

The galley is rather dangerous on a port tack, especially if you get knocked down. You would be thrown eight feet across the cabin, to crash into the storage units. The double sinks are fine in theory, but they are too small to put a plate in flat, wash a pot, etc. The corner outboard of the sink is just an empty space that used to be an ice box. It's not being used at all.

The head is fine -- just old. It has a shower, but the drain leads to an open gray water tray in the bilge below the main saloon. It smells. Worse, the holding tank is a small, flexible bag that holds about 5 gallons. With a couple of kids on board, it fills quickly. And it smells. Constantly, despite all the holding tank treatments known to man.

The V-berth cabin is huge -- but it's, well, a V-berth. The kids like to sleep there at anchor. I'm just not a big fan of V-berths, especially at sea. They're fine for sail bags and sundry, but not to sleep in. Actually, there's no need to use them for sail bags, because the lazarettes on the boat are enormous.

In the port lazarette, there's a shelf where a propane bottle sits. It has a shut-off solenoid controlled by a switch in the galley, but no sniffer, and no airtight enclosure for the propane bottle. I want to install an ABYC-compliant locker.

One final note on the layout. The cockpit is rather large for blue water sailing. The cockpit well is almost 8' long and 2' 6" wide. On the plus side, an 8-person liferaft could easily fit in there, and if suitably anchored and protected, might reduce the cockpit volume enough to make it safe.

Next up... my ideas for changes.

Hull-deck joint

The Columbia 43 used an H-channel hull deck joint ("HDJ"). This consists of a long aluminum extrusion shaped like an H in cross-section. The hull fits into the bottom part of the H, and the deck piece fits into the top. Here's a page that illustrates it from Boating Magazine, September 1971. The diagram there is:

 Mine is slightly different:

There is no rubber trim strip, and two SS strips cover the top and bottom of the joint. Here's how mine is built:

I suspect that a PO removed the rubber strip and ground off the lip that held it. Two long stainless steel strips run along the upper and lower edges of the H-channel, screwed into the channel with SS screws. They cover the pop rivet heads and extend just over the edge of the channel. There's a fillet of epoxy along the top and bottom of the channel, which I suppose was meant to seal it. It failed.

This photo shows the SS strips removed:

I cleaned off the surface oxidation with a wire wheel:

You can see the pop rivet heads and the epoxy fillet along the top. It looks like a PO also smeared epoxy over the whole H-channel, then painted it blue.

There are many places where the epoxy fillet failed. I dug around one:

The brown stuff is failed caulk from the Pre-Cambrian era. I could dig it out with a knife.

Here's a simplified sketch:

The epoxy fillet has cracked in many places, and water has worked through the cracks and through the ancient caulk. But I notice water coming down the inside of the hull from the lower part of the aluminum extrusion. It's not obvious where this is coming from, but it may be that water running down the outside of the hull seeps up into the lower part of the H, up around the top of the hull, and down the inside. Or maybe water gets in around the pop rivets or the screws that fasten the SS strips. The water fills the extrusion, then leaks out the lower pop rivets, and down the inside of the hull.

And no, the surveyor didn't catch any of this. The POs had carefully cleaned the interior water stains, and didn't mention the leaks at all. Mind you, the boat is 40 years old, so I was hardly surprised that it leaked.

So here's my two-part plan.

Plan A

I have a combination tool with a 1/16th wheel that I will use to grind out the ancient caulk without going down far enough to decapitate the pop rivets. I'll do this top and bottom, then fill with the thin version of Life-Calk (the one that penetrates into cracks more easily). To prevent the material from running out from the lower part of the H-channel, I'll run a strip of outdoor blue tape along right after my caulk gun to seal it off. I hope.

Next, I'll test with a garden hose and see if the seal works. If it does, terrific. I just replace the SS strips (with caulked screws). If it doesn't, then on to plan B.

Plan B

I will rout a groove into the fiberglass immediately above and below the H-channel. The groove will be about 1/4" wide and 1/8" deep. I'll lay wet strips of FG cloth over the H-channel, and tuck the edges into this groove. When I have three or four layers of cloth, I'll fill whatever is left of the grooves with epoxy putty.

I won't be able to fair this glass overlay to the hull, because that would be just way too much work -- and would ruin my beautiful Awl-grip paint job. So to cover the whole thing, I'll build a solid teak rubbing strake wide enough to cover from just above the upper groove to just below the lower one. I'll mill a recess in the back of the wood to accommodate the thickness of the aluminum H-channel plus my glass layup. I'll then fill the back of the rubbing strake with caulk, screw it in \with SS screws (caulked), and cover with plugs. It would look something like this:

I'm wondering, though. When I dig out the old caulk, if I avoid the rivets I will only be digging out about half of the caulk. Maybe I should drill out each pop rivet, then grind out the caulk most of the way down the H (my grinding tool isn't wide enough to reach the cross-bar of the H, so I won't cut through it). Then insert the new runny caulk, then immediately insert a new (caulked) pop rivet. I would work in sections of about 6 feet at a time, so I won't actually separate the entire deck from the hull all at once! But I don't know how long the pop rivet would have to be. I guess I could drill one out, then feel around with a wire to find out.

Does that sequence make sense? Or should I apply the runny caulk, let it set for a day or two, then insert the new rivets? Or not chop off the old rivets at all? I figure if I'm going to this much work, I may as well do as thorough a job as I can. And I really want to avoid plan B if possible (although that teak rubbing strake would look really nice!)

The Adventure Begins

In July, 2015 I bought a 1975 Columbia 43. Here she is, stripped down, ready to truck from Florida to Bayfield, Wisconsin:

I am about to begin a major project to improve the boat, fix various problems, and to redesign and rebuild her interior. I'll use this blog to document my progress and describe how I solved the numerous challenges I encounter. And because this is a blog, I welcome comments and suggestions from anyone -- especially those with contrary opinions.

Here is my high-level project list:

1. Hull-deck joint. It leaks. Worse, it is the dreaded H-Channel construction, the hardest of all to repair. And you can't tell which part of the joint leaks, because the inboard side of the joint is covered with a flattened aluminum tunnel thing that simply transports water from wherever it enters the boat to wherever there is a gap in the tunnel. Oh -- and I have 86 feet of hull deck joint to repair.
2. Through-hulls. Some are frozen solid. Others need servicing. No big deal. But others are located under parts of the cabin sole that are non-removable. If they ever leaked, I will own a submarine, not a sailboat. But that's OK, because the cabin sole is ugly, and I plan on ripping it all out anyway.
3. Electrical. Rewire the whole boat. Existing wires are not labeled, and don't have watertight connecters. And I have about a dozen switches scattered about that don't seem to do anything, and are not labeled. But rewiring won't be so bad, given the next item...
4. Gut and rebuild the interior. The original Columbia 43 was designed as a racer-cruiser, but was laid out more like a racer. There is plenty of interior space, but it's not well used. So I plan on redoing most of it. Just not all in one go, because I want to use the boat during the summer, and I won't have time to rebuild the entire interior in one winter. Redesigning the interior has been a lot of fun, and I hope I get some comments on what I have in mind.
5. NMEA 2000. I bought a fairly robust set of NMEA instruments (B and G Zeus2 plotter, Lowrance 4G radar, wind instrument, DST sensor, etc. Plus a nifty component that lets you connect the engine instrument panel to the network. The boat has an ancient Cetek autopilot that doesn't work - but I hope to salvage the hydraulic ram, rudder position indicator, and maybe the electronic compass. We'll see.
6. Numerous miscellany. A new 40 - 50 gallon holding tank designed so that it will not ever, under any circumstances, ever, ever smell. Ditto for the gray water tank. Build a new companionway door, beef up the fresh water system, install a diesel cabin heater, sew new sail covers, dodger, and bimini, reinstall four large solar panels that came with the boat (but were located on a huge arch that I disposed of), and lots more.

Now the good news. I am a pretty good woodworker. I have built lots of cabinets, from plain to fancy, and have some experience fitting out a boat's interior. I know from past experience how long things take in real life, and have no illusions about the scope of all the things I want to do. I also have a lot of sailing experience, including years of dinghy racing as a kid, coastal cruising in larger boats (32 - 55 feet), and blue water (two trips across the Pacific, one as skipper). And I have a pretty good idea of how I will use this boat: cruising the Great Lakes for a couple of years with my two young kids, and probably an Atlantic one-year circle four or five years from now. I know those are somewhat contradictory, and will require some compromises in the design. Again, I hope to learn from the comments and suggestions of others.

The adventure begins.