catzooks.com

Sanding the boards, building the spreaders

Did a few things over the past ten days. First was to put three layers of black on the rudder and centreboard. Each layer takes nine squirts, three at a time. The third batch is a bit too much, and so I spread the extra over the entire surface of both boards. The third layer only took 7 squirts — for the four-squirt cup I just dumped in 1/3rd of a teaspoon. Close enough.

To support the boards vertically while I was painting, I took two wood boards and drove screws through the end: the screws press into the blue foam and ensure the board doesn't slide around. Both the boards have a number of ridges in them which are hard to see, until the board is reflective with wet epoxy; the rudder has a couple quite noticeable ones. Not quite sure how they got there.

Once everything cured, it was time to longboard them. Both rudder and centreboard have a ridge in the middle, where the two pieces of foam joined together. This is partly due to an actual bump being there (I'm sure), and partly because the upper section is flat, and the lower section tapers slowly to the tip - and this point is where the board transitions from being flat to tapering. I don't want to sand through the s-glass exterior either... needs another layer of black.

On both the rudder and centerboard I sanded the trailing edge all the way down to the kevlar - and even then there was a small depression right before the trailing edge. The curved trailing edge strikes again. With all that sanding complete, all I need to do is rough up the unsanded parts, and another layer of black can go on. Later.

Next up: build the spreaders. Per the list, when the spreaders are installed they should be 30.5 inches across tip to tip, and 6.25 inches to mast track from straight edge on the aft side of the shrouds. (Bill also describes the jig he used to bond the spreaders to the collar, they should be well re-enforced at the bonding point, and should have around a 7° negative angle of attack to account for the mast bend.)

The strategy with the spreaders is to build them in two steps: first, bond the inner part, then when it's just past tacky, put on the second part of uni-directional carbon. In retrospect I'm not sure splitting it up is necessary, and it does cause a few hassles...

First, I bonded the saddles to the wood backing with five-minute epoxy. I cut all the layers, each 24” x 5”. First step layers:

• foam core
• hybrid (scrap)
• s-glass (plain/coarse weave, from the centreboard/rudder construction)
• peel-ply
• muslin absorbent layer
• mylar

Then for the second step:

• 1 1/2” uni-directional graphite tape, three strands, one on either side and one along the leading edge
• 3” uni-directional graphite tape, along the leading edge
• s-glass (satin/smooth weave, from the deck/hull)
• peel-ply
• muslin absorbent layer
• mylar

Make sure everything is wide enough as the layers stack up - my second s-glass layer was just barely enough to cover the entire second spreader, and probably not enough to have a good bond at the trailing edge. 5 1/2” wide would have been better. Sigh.

For the first spreader, I cut the hybrid to shape the foam core. Then wet the first step out: foam, hybrid, s-glass, peel-ply, muslin, mylar, saddles, clamp, and wait.

When I came back three hours later, I pulled the saddles apart - and the saddles stayed stuck to the foam saddles. And ripped them off the wood. Wonderful. The mylar is supposed to act as a barrier for the epoxy, but a very thin layer (from my gloves) bonded the mylar to the saddles. And apparently, when the epoxy is still (just) tacky, the mylar sticks right to it, and it won't let free. Of course, I can't stop now. The first layer is still tacky, and I need a chemical bond. I bonded the pieces of the saddle back onto the support with 5-minute epoxy, and keep on going. I didn't destroy the spreader: it seemed to have turned out ok, although there could be more hybrid along the entire edge. Place the 1 1/2” carbon and the 3” carbon and the s-glass, this time wrap the saddles in plastic wrap, clamp, and wait.

This led to a pretty good result — the finished spreader looks fine. And it popped right out of the mylar, no problems at all.

Armed with all this knowledge from the first spreader, I built the next one. This time, I didn't trim the hybrid. Paint the foam, hybrid, s-glass. Peel-ply. Muslin and mylar. Saddles are wrapped in (the same) plastic wrap. Clamp. Wait four hours for the epoxy to set up.

When I returned and pulled off the mylar, it was clear that the epoxy was beyond tacky - still malleable, but not rock hard. The mylar was stuck tightly to the epoxy. It seems that once the epoxy gets to this beyond-liquid-not-rock-solid stage, it's still very stuck to the mylar. I had to destroy the mylar wrapping when I took it off — it came off in "jumps" like old masking tape does. More than that, it wanted to pull the peel ply and s-glass off with it. I gave it a blast with the heater for 20 minutes to cure it a bit more, then removed the peel ply.

Once everything was off, I used a trick Greg & Roger have mentioned on the list: You can cut right through aramid cloths if the epoxy is tack free but not rock solid. And it worked like a charm.

Next up, apply 1 1/2” uni-carbon tape to the trailing edges and leading edge, add in the 3” tape on the leading edge, the s-glass layer, peel-ply, absorbent muslin, and a new piece of mylar, clamp, and let cure overnight. (Note that the mylar doesn't have to be right up against the layup — doesn't seem to matter, you just get this line of epoxy along the leading edge outside the peel ply.)

And of course, the cured spreader just popped right out of the supports.

(50 images)

Worklist for the winter

My goal is to be able to sail by summer. (Well, hopefully. A friend of mine asked when I was going to be done. I said "You're asking a software engineer to accurately schedule something?")

I figure I have a fighting chance, anyway. Things I know I have to complete:

• Complete cutting the holes for the centreboard insert blocks.
• Complete the insert blocks
• Build the rudder box
• Build the rudder
• Build the spreaders
• Sand down the mast (c.f. here and here)
• Re-apply the sail track (c.f. here)
• Rig the mast
• Rig the boom
• Wetsand the deck
• Varnish the deck and hull
• Apply non-skid

And I figure I have about 8 to 12 weeks before the weather turns nice enough that I can begin the wetsanding of the hull again. Given that I completed sanding the hull in late October and it was starting to get a bit nippy, I figure the earliest I could start wetsanding the deck is late March. So I've got at least ten weeks to get the above list done (gulp).

(3 images)

Cleaning up the rudder

This was a brief week - I had to deal with certain household (mis)adventures in lieu of boat building.

I did sand off the trailing edge of the rudder. This lead to the conclusion that All the precautions I took to ensure the trailing edge was straight didn't seem to help much - I still had to sand through a bunch of layers of hybrid on only one side of the foil. I'll have to clean up the kevlar fuzzies.

I wet out the trailing edge with a brush, clamped it down with a strip of folded over mylar, clamped with strips, and let set overnight. Repeat with the centreboard. Once I had sanded lightly it looked quite good - much less fuzziness, and the strands that are in there seem more tacked down.

Bill asked on the list on how to clean up the trailing edge, but there wasn't any reply. In mail, he said:

When the extra was cut off of the trailing edge, I ended up with the fuzzies. I followed Bram's suggestion to lay up a fine edge with mylar and filled the taper with long carbon strands running the length of the trailing edge, with high-density filler and graphite powder in the epoxy (for strength). Some of the strands ran up on the surface of the board where there was fill needed anyway. Once this cured, a long board sanded the edge and surface true. My foils ended up with an extra layer (s-glass) and is a tight fit in the trunk. It remains to be seen whether the CB endures my learning to right this thing. Bram prefers to do the layup in one cure. Mine was in two. I don't recall that he ever replied specifically to my question #3.

With that in mind, it would have been helpful to keep all the strands I pulled out of the carbon as cut-lines. (sigh; I even thought to myself "what a waste of money!" as I pulled them out...)

(6 images)

Building the rudder

Actually skinned the rudder this week. Bonded the board on the large side, and screwed and bonded the small end, to try and avoid it falling off and make it more capable of holding all the weight of the saddles. Wrapped the saddles in plastic wrap (entirely this time - no little blue areas showing through). Cut some scrap strips to clamp on the trailing edge. Pulled eight cups of 5 squirts each (and fished the pump out of the resin container). Then started at 8pm and finished laying everything up just after 9 - way faster than the centreboard. (The centreboard took about 2 1/2 hours to lay up; the rudder took just one.)

Next was to get the saddles on and aligned. I forgot to duct tape them together, so that was first - leaving only about 1/8” of space between the leading edges. Then, apply clamps. On the top edge I put my bar level along the side opposite the wood strip, and used all the bar clamps; the bottom edge I used all the quick clamps, which seemed to work out well. On the trailing edge I clamped the cedar strips in place.

Once again, clamping wasn't perfect with about 1/8” offset between the top edges of the saddles. This time, this caused some problems: there is an indentation in the leading edge of the rudder, and the trailing edge has a jog where the saddles were offset. The next one of these I build, I think when I get to this stage I will place the rig head end on the ground, clamp the saddles in place, then draw a line where either saddle meets up with the wood end supports. At that location, I will screw in a support. Then the saddle will rest on that support, and be able to slide in and out to get the right level of compression, and won't slip down. (Of course, the PDF already says to do this, but I didn't follow that bit. Silly me.)

The saddles pretty much fell off the board - I may even be able to use them again (although they got pretty dinged up in the garage prior to this and there are spots in the board which aren't flat due to the saddles not being tight — sigh). Then rip off the peel-ply, and pry & chisel off of trailing edge cedar strip. And it's time to shape and fair.

(25 images)

Preparing the rudder blank

I spent this week preparing to glass the rudder. First, there was cleaning up the mess from the centreboard. I decided that I wanted to put a strip of cedar down the middle of the board (in lieu of Greg's technique - maybe if this one fails I'll do his next). I used one set of strips the length of the rudder, and a second set around 2/3rds as long, and glued them all together - the short pieces don't need to be as long, as the tip of the rudder is quite thin. In retrospect, it would have been easier to bond the blue foam in the exact centre of two long pieces.

Then, cut the rudder, bond back together with 5-minute epoxy, plane smooth, and fill the gaps. The plan doesn't like to pull the blue foam - it makes ripples instead.

I had a few concerns. First was sawing the foam: I was concerned the saw would not cut perfectly vertically through the foam, although I seem to have avoided this fate. Another problem was ensuring the foam was bonded to the cedar strips in a perfectly straight line (don't want a curved rudder, either). This is why having two long strips would have been good, then it's easier to tell if it is going on straight. Finally, planing the cedar makes the rudder very slightly flatter in the middle than it would have been with the foam, but it doesn't seem too much of a problem.

(10 images)