More Glass Pains

The last couple of days we have managed to apply the two longest pieces of fiberglass yet.  Two different 12 foot long (52 inch wide) sections were laid across the width of the top to complete the second layer of glass on the top of the top.  I'd love to say everything went well, but I would be lying.

Dry fitting a 12 foot long piece of fiberglass cloth.

We were once again working mostly at night because of timing and temperature issues for the first piece.  We did a process similar to our first large piece of fabric.  My wife would make up 16 ounce batches of resin, I would roll part of it out on the top, then roll the fabric over the wet area and finally use the remaining resin to wet out the top of the fabric.  We would then roll and work the fabric to make sure the resin was penetrating the thick fabric.  When the next batch of resin was mixed, we would roll the fabric back a bit so I could wet the bottom of the next section and continue the process.

Trying to do the best work we can, it takes us about 4 hours to apply, wet out, and roll a 12 foot long swath of this fabric. I think things would go faster with a 3rd person so one would mix up resin, another could apply the resin and wet out the fabric, and the third would work the fabric with the fiberglass roller and squeegee. I tried getting the dogs to help out, but they keep objecting due to the lack of opposable thumbs. I think they are just smart enough to know better.

Doesn't it look pretty at night.

As you might guess, the problem with working at night is the light.  We have a couple work lights (a 500 watt construction work light and a clip-on LED light that we can attach to the canopy right above the top) but, as it turns out, this just isn't enough light.  By having only one or two point sources of light, it is nearly impossible to see bubbles and dry spots inside the fabric layer.  We did the best we could and everything looked great as best we could tell when we were done (do I know how to spend a Friday night or not...heh).  It wasn't until the next morning that we discovered some issues in the layup.

The first issue was that there were dry spots in the layup.  The dry spots seemed to occur in a line, and I believe it was caused by the dry roll of fiberglass.  When we would roll the fiberglass back so I could apply more resin to the top, the roll would soak up some resin while sitting on the wet edge.  Then once we rolled it on, we didn't see that some resin was sucked out the area.  The second issue were a few bubbles that we both swore weren't there the night before.  Fortunately, repairing fiberglass isn't difficult.

When we did the second long sheet of fiberglass, we started a bit earlier and so we had the advantage of daylight for much of the process.  I paid particular attention to the spots where we stopped the roll, making sure to use the fiberglass roller to rebed the glass in those areas.  The result was much better that time.

Grinding down bad spots in the second layer layup.

After things cured some, I went back to the first piece and ground down the bad spots with a die grinder fitted with an abrasive sanding wheel.  After smoothing the edges of the ground down spots, I cut patches of cloth that overlapped the ground down areas and applied them to the spots.  As of this morning, they are looking much better.  Unfortunately, it does make a bit more work to smooth out the top surface.

Cut out repair patches.

So, the top now has a minimum of two layers of fiberglass across the top side, with three layers along the back edge and the mounting flange, and up to four layers where the sheets of fiberglass overlap.  The next task is to smooth things out as best we can on the top side.  Since we will be applying nonskid across most of this surface and the thickened gel coat we will use will easily hide most of the minor smoothness imperfections (like wall texture does in a house), it shouldn't be too difficult.  The hardest part will be reaching the center of the top (until we get a layer or two of fiberglass on the other side of the foam as it still can't support a lot of weight the way it is now) to smooth things out.  I'm thinking a drywall pole sander may be my best bet.  I might use a thin layer of resin to help smooth things out as well.

Top side patched and ready to be smoothed out.

I am not looking forward to sanding this stuff down as fiberglass dust is the equivalent of itching powder. Then I need to come up with a plan to disassemble the canopy and tarps and figure out how to remove the top from the mold and flip it over so we can start applying glass to the other side. There is always something new to figure out with this project.  But it is starting to really look like a bimini top for the boat and even gets compliments from those who stop by to see the project. Hopefully that will be the case when this project is done.

A Pain In The Glass

As every sailor knows, your life is dictated by mother nature. We rely on the wind for propulsion, the sun for power (or I will once I finally get the solar installed on this bimini), and the seas for a smooth highway or bumpy ride, and we try to hide from things like thunderstorms and hurricanes. Those further south than I are currently bracing for Erika (a storm that formed right behind Danny and looks like it will be heading to Florida). But even when sitting in a marina and working on projects far from the storms, the weather still dictates my life.

Tropical Storm Erika, National Weather Service Hurricane Center Image.

If I had been smarter, I probably would have held off on this project for a month or two, but the timing just didn't work out that way. So instead, I continuously wait on weather windows in order to continue the fiberglass work. Water contamination is a concern, so it can't rain. Dew that collects on the shade canopy most mornings can drip down and make a mess of things. By using polyester resin I can adjust the speed of the cure by the amount of catalyst used...but only to a point. Using the minimum amount of catalyst, I can get about 25 to 30 minutes of working time out of a batch when it is 90 degrees Fahrenheit. The only weather problem I haven't encountered thus far is the minimum temperature of 60 degrees...they say you shouldn't try laying up fiberglass below that temperature either.

Like in many days past, we have been somewhat cool overnight, at least to the point that the temperature and dew point meet. About the time it dries, the temperature is shooting past 90 degrees in the yard. So that only leaves us with the evenings. And that is the time slot we have been using. Around 4 or 5 PM the temperature drops back below 90 F, and if there aren't any storms, we can finally start applying fiberglass to the big foam sheet that makes up the core of the hard top.

Front curve of the top now cut.

Unfortunately, as the light fades, it gets more difficult to see what we are doing. The 500 watt work light helps a lot, but it is no replacement for the ambient light during the day. Not wanting to spend the rest of our lives at the boatyard, we press on. Over the last couple days, we have managed to cover the remainder of the top side of the bimini foam with one layer of fiberglass and with two layers along the rear edge where it will mount to the arch.

On the first new panel we started, we wanted to do the whole section with a single 9 foot long by 52 inch wide piece of cloth after laying up the 18 inch wide strip at the back/mounting flange. Wetting and applying that much fiberglass at one time with two people simply cannot be done in the time it takes for the resin to cure (at least not for us novices). In order to put down a sheet of glass that long, we need to do it in stages.

Positioning the rear mount reinforcing strip.  Note the black alignment marks.

First we positioned the dry cloth on the top, rough trimmed it so there was only an inch or two of overhang, and made marks using a permanent marker to act as guides for placement. We rolled the fiberglass onto a section of 2 inch diameter PVC pipe. Then we applied a coat of resin to the foam and allowed it to begin to gel. For the rest of the process my wife would mix up 16 ounce (1 pint or approximately 473 ml) batches of resin. I would apply some to the resin to the pre-wetted foam, we would position the end of the cloth on the roll using the marks, and roll out a couple of feet of the cloth. Using more resin I would wet out the top of the cloth.  My wife would make more resin, and I would roll the cloth back up to the wet area, apply more to the foam, and then roll more of the cloth out and wet the top side of it.  Then whoever wasn't busy at the time would use the fiberglass roller and squeegee to embed the cloth in the resin, work out bubbles, and move any excess resin toward the dry roll of cloth. This process of mixing some resin, wetting the foam, rolling out some glass, wetting the top of the glass, and working the glass was repeated until the entire piece was done. By the time we completed the last section, the polyester was starting to gel in the first sections.

Initial dry-fit of the long piece of fabric. Note PVC "roll".

The key to this approach is to stay well ahead of the resin that is starting to gel. You also need to make sure that the section of cloth is well wet out and in its final position while the resin is still wet. This process is a bit nerve-racking the first time you try it...or at least it was for us. Making sure to stay ahead of the curing resin, getting the cloth to wet out well (particularly on more vertical surfaces) and making sure the resin isn't too thick means you have to move fairly fast. Oh, and remember we are doing this by the light of a single 500 watt bulb, outside, where the light is attracting every insect within 5 miles to our little oasis in the boatyard. I'm sure if anyone had been watching, they might have found the whole ordeal quite entertaining. It took us from around 6 PM to almost midnight to complete the mounting flange band of fiberglass and the 9 foot long sheet (and clean and pack up the work area for the night).

The second side panel was done the following night and went about like the first, except that we ran out of material on the roll before we reached the end of the panel. So, in this case we had to pre-cut another piece of cloth and during the layup process it had to be positioned and integrated into the first piece as we went along. Having done the process once, the second time it did go a bit faster, and I think we made it back to the boat just a little after 11 PM.

The morning after each layup, we were finally able to see our prior night's work in the light of day.  For the most part, it looked good. There were a couple small spots that had bubbles or were not wet out well and we will go back and repair them before the next layer, but overall I think it went well. At least we are making progress towards the goal.

The first layer of fiberglass on the top.

As a side note, a few bugs (mosquitoes or similar) did manage to make it into the wet resin. They actually look just like the fossilized "fly in amber" you might have seen in grade school. Maybe a 1000 years from now some grade school kid will get to see a 2015 mosquito perfectly preserved in glass fiber reinforced polyester resin...wonder how the teacher would explain that. But hey, at least those are a few mosquitoes that won't be biting us again.

While The Weather Cooperates

Hurricane...or is it tropical storm now...Danny doesn't look like it will be headed our direction. (I do hope all my friends and fellow cruisers in its path are OK).  Meanwhile, the weather has been at least somewhat cooperative here so we've been putting in a lot of time on the hardtop project. So, sorry that posts have been a bit sporadic, but if there is one thing that this project has taught me, it is that we need to work when the weather allows.  By the end of the day today we are ready to lay more glass, but there isn't enough time so I'll finally see if I can catch up on the blog.

Work space or refugee camp? Attempting to keep things dry.

When we left off there were a couple days of unsettled weather that arrived.  After the storm that hit at the end of the last post, we mopped up the mess and tried to get everything dried out. They called for 70% chance of rain most of the following day and that usually means we will get soaked (water and laying fiberglass just don't mix).  So we did some shopping and other preparation while we waited on the weather...that never came.  The ominous looking clouds finally just fizzled out about 3 PM. We cut the two smaller pieces of foam that need to be glued to the ends of the 8 foot panels so we end up with 9.5 foot panels and punched holes in them, but that was it.

The next day the forecast was similar, but we decided we wouldn't be fooled again.  This project was already taking a lot longer than I had hoped, so we didn't want to sit around waiting for weather that may not arrive. We pulled out the two remaining foam panels and spent the day punching holes in them.  The 9 nails in the tool make this a much longer process than I would like, but it does work.  Well, until a nail breaks off. You see, you tend to bend the nails after pushing them into the foam so you have to constantly straighten them.  Eventually the bending weakens the nails and...well...you end up digging one of them out of the foam panel with pliers.  Good thing I had the remaining piece of the original longer board full of nails so I could just cut another 9 out of it and finish up. Again, thankfully the rain never came.

Gluing the extensions on the large panels.

Then it was time to start gluing the remaining panels together.  Since I don't have a full mold and only stringers, I had to attach the two side panels to the mold so the end would lay over one of the stringers and then glue the small pieces to them.  This would keep them properly aligned and holding the curve. Using polyester resin that I reinforced with some chopped glass fiber and fumed silica, I created what I could only describe as "hairy polyester mud" to glue them together.  I smeared the concoction into the joint, then pushed the smaller piece into position and held them in place with sandbags covered in plastic.  I used a plastic spreader to remove the excess that squeezed out in order to minimize later sanding.  It is much easier to scrape off wet goo than sand down hardened polyester.

After the two side panels cured, it was time to adhere them to the center panel.  Now one might think that the edges of a 4 foot by 8 foot foam panel would be straight, but they did not line up with the center panel very well.  I discovered that the center panel had pulled up from the mold a bit, I assume from the shrinkage of the polyester.  It seems that the 30 lb. test fishing line I was using stretched or the knots slipped a bit.  I was able to pull it back down with some wedges between the frame and the line. This shrinkage may have given the center panel a slight curve along the edge as well. so I then had to sand the panels a bit to get a better fit.

Panels tied to the frame and glued together.

Once the fit was better, I used a similar concoction to glue the side panels to the middle panel.  We tied the edges that are closer to the seam to the mold so we could slide the panel back a couple of inches.  I also threaded ties for the other side through the panel but left them untied so we could move it.  By sliding the panels apart a couple inches, I was able to apply the hairy mud from underneath the mold.  We slid the panel into position, checked the alignment, and tied the free end down to the frame.  Since there was one spot on each panel that didn't line up flush, I used boards wedged underneath the mold to hold them even.

All the panels are attached. This thing is getting big.

When all of that cured, we had a foam panel roughly the size of the final top with a fiberglass strip down the center. All that was left was to finish shaping the foam construct so it would be the final shape of the top and ready for glass.  Using a razor knife, I roughly carved the back edge of the two new panels and then sanded them so they matched the needed curve for the mounting flange. In order to protect the back edge of the mold, I slipped a sacrificial piece of heavy plastic between the foam and the mold and adjusted its position when sanding would start cutting through it.  This worked well and allowed me to get the curve to match up very well with the mold.

Rough cutting and then shaping the back edge.

The final step was to cut the curve at the front.  I used the trick I originally figured out for creating the curves on the stringers.  Taking a boat stand, some line, and a pencil, I created a crude compass to draw the arc.  In this case I didn't have a radius (for some reason, SketchUp wouldn't tell me the radius) but I had the intersection points on the edges and the length at the center, so I figured it out where to place the boat stand (a.k.a. the compass pivot) using trial and error.  I then cut the curve using my jig saw fitted with a fine tooth (32 teeth per inch) metal cutting blade. Where the foam intersected one of the mold stringers, I switched from the saw to the razor knife to keep from damaging the form.

Scribing the arc with a boat stand and string again.

A little sanding and it is now ready for applying the remaining fiberglass on the top side.  If the weather continues to cooperate, that will likely be what we will start tomorrow.

It Begins

Finally.  Working on the actual hardtop.  Can't believe how much work has been put into this project just to get to the starting point.  Building a workshop, a work table, a frame mold, a fiberglass roll holder/table, trying to make the workshop a bit more weatherproof and other improvements took a long time. But now we are making a little progress toward actually creating the top.

It started with dragging two panels of foam to the work site,  One was the piece I cut the test panel from and I used it to cut another 15 inch section off of the end. Then we started poking holes in the face of that piece and the full 4 foot by 8 foot section.  The hole poking tool that we started with had 36 nails in it and we quickly decided it was difficult to work.  Those 36 nails, when pushed into the foam, hold on pretty well and you really have to pry it back off.  So, the tool was reduced to just 9 nails and that worked better.  Another problem with the tool was that the nails were pushing back out of the back of the board, so I mixed up a bit of polyester resin and glued the nails in with it.  After that it was just a matter of using a 6 inch square took and slowly poking holes in all that foam.  Did I mention it takes a while?

The new 6 inch square foam hole poking device.

Once the holes were poked, the next step was to position the foam on the frame and tie it down with some fishing line so it would retain the curve.  It needed to be positioned over a stringer so we could tackle the next step, joining the small piece of foam to the end of the large one.  With some fiddling, we got the large piece positioned and figured that we could use some light sandbags covered in plastic to hold the smaller piece in position.  I mixed up some resin, thickened with fumed silica and a little chopped glass fiber (in my spare time I've taken some of the small pieces of clean scrap cloth that were cut away from the test piece and cut the fibers into 1/4 inch or so strands), and we glued the two pieces together.  After a little curing time, we had a 4 foot by 9.5 foot slightly curved panel.

Hand sanding the curve at the top back edge.

After that, the panel had to be re positioned to it's final location on the mold and re-secured to the frame. Then the rear edge needed to be shaped so the fiberglass would curve over the edge of the foam and continue down the back of the 60 degree angle that makes up the mounting surface.  Since it is a compound curve I hand-carved the foam with a razor knife and then hand-sanded it into the correct contour.  Another labor intensive and time consuming process, but it did turn out well.

There was much debate about how we were going to fiberglass the middle of a 9 foot by 12 foot panel. We decided that we would start by putting a band of fiberglass down the middle section before we attached the side panels.  This way we can be up close and inspect the initial layer's bonding between the glass and the foam.  So, the next step was to actually lay up some glass.

Naturally, the...ahem...cooler dry days we had the last weekend were no longer in sight.  Now it is supposed to be nearing 100 degrees Fahrenheit with chances of rain in the afternoon.  Not exactly the best conditions for doing fiberglass work.  We got up yesterday ready to do some fiberglass before it got hot, we just needed to run to the store to pick up a couple of last minute things.  By the time we got back to the yard the temperature was soaring over 95 degrees. Too late to get it done until it cools off in the evening.  So, we did some other things in preparation for later steps like punching more holes in more foam panels. And we bought a work light. Remember how I lamented doing fiberglass work in the dark? Well, it seems like that might be the only time conditions are favorable for doing the layup.

It finally got back down to around 90 F after 5 P.M. and we were able to start work.  With the panel in place, I started by putting a strip of glass along the frame that will become the mounting flange along the arch of the boat.  I had previously laminated that surface of the mold with cellophane tape and sprayed it with PVA so it wouldn't stick. I coated that with some thickened resin, let that get tacky and then applied a little more regular resin and the glass.  Applying resin on a surface that is 30 degrees past vertical is a real pain in the butt.  I think I ended up wearing about as much as the fiberglass mat was.  But I was able to get it laid up with a little work.

Yep, doing fiberglass at night again.

Before the strip we applied had cured, we applied the first layer over the back part of the arch.  This was a square of the mat that was 52 inches long and 46 inches wide. It was applied to the top rear of the the foam sheet and wrapped around to the mounting flange at the back.  This left the sides of the foam clear so we can easily bond the next sheets of foam without any interference from the fiberglass mat.  Of course it was getting dark at that point, but we were a bit more prepared than just a hand-held flashlight.  This time we had a 500 watt worklight and a LED clip light to help light the work space. By the time we got that piece done, it was getting late. We packed everything up and headed back to the boat...getting there about 10 P.M.

The next morning we awoke to a good coat of dew on the inside of the work canopy.  Since I didn't like the idea of it potentially raining on me while working, we setup a fan and waited for the canopy to dry.  It did...finally...around 10 A.M. or so.  The same time it was approaching the 90 F temperature mark again.  But we did lay up the two final pieces of glass to complete the covering of the center panel of the top.  And it wasn't even dark yet.

The top of the center section with the first layer of glass.

Feeling like we might get more accomplished today, we stopped for a late lunch and were rushed back to the tent with the sound of thunder.  We get back and start buttoning things up and as we are getting things closed, the downpour starts.  And it pours.  Trying to tie things up, I'm getting soaked and the panel does get a little wet.  We quickly dry it off (the glass had cured by that point, so blotting up with towels did seem to work) and my wife re-positioned a tarp to block the rain.  Hopefully it won't cause any issues with the surface of the laminating resin so we can continue the layup when the weather improves.

It feels good to finally see something that resembles a top starting to form.

Con-struction and De-struction

After laying up the fiberglass structure on my test panel, the next thing was to come up with options for the finish. The goal of this test was to come up with a nonskid treatment that utilized gel coat and not paint (most non-skid renewal options are paint-based) and be something that can be done relatively easily given my construction methods and limited resources.

The factory nonskid used on our boat is integrated into the negative mold of the outer hull and produces a nice, uniform diamond pattern.  Of course, I don't have a negative mold and I doubt the pattern material is inexpensive.  I could create a mold from the existing surface of my boat, but that fails on the "relatively easy" test.  A molded-in solution just wasn't likely to work.

Searching the internet (in addition to all the people who have used Awlgrip and Kiwigrip), I found someone who mentioned that a few production boat builders run a roller over a curing gelcoat layer to produce a nonskid pattern.  This sounded like an interesting approach and, if we could figure out how, might be an easy solution.  So we gave several variations on this theme a try.

The first attempt consisted of applying a layer of gelcoat, waiting for it to tack, and then using a fine nap roller to see what would happen. In the same coat, we also used a chip brush to stipple the gelcoat.  At first it wasn't tacky enough, and the gelcoat would flow out relatively smooth.  As it continued to cure, we were able to get a pattern from both the roller and the brush.  The roller ended up producing what I would equate to a very fine orange peel type of texture, and the brush version was a little more pronounced version of the same.  Not too bad, but we didn't think this would be nonskid enough.

Since we liked the brush stippling result better (but figured doing that by hand would prove far too tedious for a larger surface), we went back to the local big-box hardware store and looked at various brushes and rollers used for wall textures.  We found a special texture roller from Wooster that we thought would provide a more brush-like stippling effect but with the speed of a roller. The roller has small, coarse plastic loops of material that should do the trick.

With our new weapon, we tried another test.  This time I thickened some gel coat with fumed silica. It was thicker than normal but would still flow fairly well.  Once it started to set up, we used the new roller as well as a brush for two new test cases.  This time the pattern didn't flow out as much as created a more aggressive nonskid pattern.  The brush was still more aggressive of a texture compared to the roller, but not by much.  With the speed that we should be able to accomplish this texture, we think we found a winner. I sprayed a coat of PVA on the gel-coat so it would fully cure.  Once cured, we broke out the hose and bare feet and tested the options.  The thickened roller approach was the winner.  It provided good traction while not feeling uncomfortable on bare feet (the brush stippled version we found to be a bit more "pokey").

Nonskid tests.  Thickened gelcoat on left, earlier attempts on right.
(Purple hue is the PVA during cure)

The next morning, I did a little destructive testing of the test panel. We did the layups in about the worst conditions we figured we would encounter during the actual layup, and I wanted to know how well it did. After jumping up and down on the panel a bit more, I drilled and cut a hole in the panel with a jigsaw and it looked good. Since I left the raw edges of fiberglass on the panel, I then attempted to pull apart the layers with pliers. This is where things didn't go as well.  I wasn't able to completely separate the plies, but I was able to pull a significant piece of one layer of the biaxial cloth apart.  Despite the look of a pretty complete wet out, it seems that the layup was dry in the middle of one of the heavy layers of fiberglass cloth.

I know this #1708 fiberglass cloth can be difficult to wet out. I thought I had taken extra steps to help saturate it with resin, and it did look good when I laid it up (nice and clear), but I guess it wasn't good enough. I'll have to increase the amount of resin applied before I apply the cloth. Glad I found it now and not after building the whole top.

The day ended with final preparations to the mold (making sure things won't stick to anything it shouldn't) and my poking those holes in the foam. I ended up cutting down the size of the hole poking tool because the larger version was difficult to pull out of the foam panel, and then I bedded the nails in resin after finding they were pushing out the back of the board. Hopefully tomorrow we will start assembly of the actual top.

Fiberglassing in the Dark

It is kind of funny.  It seems most projects that my wife and I start, at some point along the way, result in performing work late at night in the dark. Be it planting trees in our old backyard, re-tiling our old bathroom, or a myriad of other projects, we end up misjudging the amount of time needed.  And now we can add fiberglass layup to the list.

After getting the mold built, I wanted to do a test fiberglass layup.  I've never created a large composite (foam cored) panel and have never used polyester resin, only epoxy. I've read about as much as I can on the subject and feel comfortable with performing much of the task, but I think it is prudent to try out several techniques on a small sample before I go about making a 9 foot by 12 foot mess with $2000 (U.S.) worth of materials.  This way I can practice as well as try out a few different construction ideas.

I started my test piece by cutting a one foot section off one of the 4 foot by 8 foot, 5 pound, 1 inch thick Divinycell panels. The stuff cuts with a razor knife, but at an inch thick, it isn't a single pass cut. Oh, and cutting the stuff kinda sounds like running fingernails down a chalk board to boot. I then cut my 1 foot chunk in half, so my test panel will be a 2 foot square and I can practice gluing two pieces of foam together and testing the joint.

I think it was someone who commented on a prior post on the blog who suggested I poke small holes through the panels.  I think the theory is that this will help with bonding to the foam and, since that is critical in a sandwich structure, I figured I would give it a try. So, using a scrap piece of plywood (I seem to have several to choose from) and some finishing nails, I created a board with nails on a 2 inch grid.  I then used the board to poke holes into one of the two panels.

I used 30 pound test fishing line to tie the foam panels to the mold.  This would allow them to slide just a bit along the frame and still hold the curve.  Using some fumed silica, I thickened some polyester resin (to a peanut butter consistency) so it would stay put on a vertical surface. I applied it to one edge of a panel and then the other panel was pressed up against it.  Once that cured I had my 2 foot square test panel of foam.

Next was the fiberglass layup itself.  To start, I mixed up some resin and applied a thin coat to the foam surface.The main goal of this step is to limit the amount of resin uptake when applying the fiberglass mat.  I'm told if this is not done, the foam might suck up too much resin when the glass is applied and leave the glass resin starved (not good). After letting that initial coat gel, I applied and wet out the first layer of glass.  The 17 ounce biaxial cloth with the 3/4 ounce chopped strand mat backing is thick stuff and was a bit difficult to wet out.  To resolve this issue, in subsequent layers I've put down a bed of resin and then worked the cloth into it, adding resin to complete the wet out as needed, and this seems to work better. Since I was using laminating resin and wanted to test the bonding characteristics overnight, the second layer was applied the following morning. Yeah, that's the story...it had nothing to do with the fact we finished the first layup at 9:30 P.M. using a crappy LED flashlight we happened to have in the car.  By the way, I don't recommend doing fiberglass in a tent in a boatyard at night.

The first layer wasn't my best work, but it wasn't too bad given the conditions.  I did a little destructive testing to confirm that there was a good bond with the foam and there was.  I took this chance to round the corner on part of one edge so I could try molding the mounting flange. I then thickened some resin a bit and coated the mounting flange on the mold to hold the cloth for the next layup.  After mixing up some more resin, I applied the second layer of fabric. Since it was daylight and relatively cool that day, the layup went well.

Second layer of fiberglass and practice with the mounting flange.

After the top side cured, it was time to break the test piece from the mold, flip it over, and apply glass to the other side.  The part came free of the mold (which I had previously covered with cellophane tape so the panel wouldn't stick) after cutting the fishing line holding it down to the frame and giving it a little push away from the flange part of the mold.  Flipping it over, it became pretty obvious what the next step was.  Resin had indeed seeped through the holes that were made in the core material as well as drips that ran down the edges, so they had to be sanded until the foam surface was back to flat.

Sanding the resin nubs off the bottom side.

Two more layers of resin and cloth and the test panel was complete...at least structurally.  Being a somewhat impatient person, I didn't want to wait until I had the finish applied to test the structure.  If you recall, I'm using a laminating resin so the surface doesn't fully cure (to allow for more layup layers or better bonding of the gelcoat), so I temporarily wrapped the panel in plastic and suspended it between two wood blocks.  With only an inch of the ends of the panel on the blocks, I hopped up on the panel.  It supported my weight without any problem.  I bounced up and down on it and didn't even see it flex.  Whew...glad to know the plan is working and the less-than-perfect test sample was performing well.

The composite panel seems strong.

After playing with the panel for a bit, the next step was to apply some gelcoat.  I wouldn't normally worry about this, but one of the things we are trying to figure out is an easy way to make a nonskid coating in the gelcoat.  I had read that a few boat builders wait for the gelcoat to start to gel and then run rollers over it.  The roller supposedly will pull small peaks up in the sticky gelcoat and cure with a rougher surface.  We tried that technique as well as using a brush to stipple the gelcoat as it was starting to cure.  We found that the brush seemed to work better...but rain was threatening so we had to close up the tent and will check on the success or failure in the morning. I guess the best part of the storm now raging outside is that it is giving me a chance to update the blog...if the internet at the marina will cooperate.

So work continues, and thus far the tests seem to be going well. Still a bit nervous about how I'm going to apply these techniques to a 9 foot by 12 foot panel though.

Going Topless

Whew.  Been a busy...oh my, has it been 5 days? Well, at least I've been making some progress on the hardtop in between the rains. Oh, and if you were doing an internet search and came across the title, sorry to disappoint but this has nothing to do with drunk college girls on spring break.

The frame, or mold if you want to call it that, (the recent joke around here is it and my table look like a really nice roof frame and floor for a shed) is finally done. The last bit was to come up with the mounting surface at the back where the top will bolt on to the side of the arch. Of course, this means we finally had to figure out the exact angle we wanted the new top to have as it extends from the arch forward over the cockpit and the salon roof. The old soft top was multi-faceted due to the frame and came down far enough in front that it felt closed in and restricted visibility from the helm. Since the new top is straight, we needed to figure out an angle that would: improve visibility, look good with the lines of the boat, and not interfere with the boom.

The first step was to remove the old soft bimini.  It should be a simple matter of unzipping the cover from the frame supports, undoing a couple of ties, and sliding the back edge out of the bolt rope slot. Of course, this is a boat so it wasn't quite that easy. This is the first time we've tried taking the top down and quickly realized that this may be the first time it has been down since it was installed.  The zippers worked ok, but the nylon webbing ties had become, for lack of a better word, petrified in the Florida sun. Whoever tied the top on, in addition to using the D rings to hold it, tied knots with the webbing. It took prying the knots and then the D rings apart with a screwdriver to finally get them to release their grip.

Rover Topless.  Can you see the string?

The boat does look different without the top. Almost a bit sporty...like a convertible roadster. Once the boat was topless, we tied a string to the arch and took it forward to the mast so we would get an idea of the line of the top. We tried several placements and finally settled on one. We even had a boat neighbor who was passing by come give us his opinion, and we all agreed it would be a good look for the boat. It allows for plenty of room for the boom and adds about 2~3 inches more opening to look forward from the helm, and the line seemed to work well with the various lines and angles of the boat. So I took out my recently purchased adjustable angle square and measured the angle from the face of the arc to the line (this time the Harbor Freight tool did not disappoint).  As it turned out, it was exactly 60 degrees.  Interesting that our eyeball based determination resulted in a nice round number like that but I figured it was meant to be.

So, with that information in hand, I constructed a way of holding a board in position at that angle at the back of my frame/mold.  I took two pieces of plywood and attached them together so they would cover where it looked like the arch would intersect the 60 degree angle and then scribed the arch with a 4 foot long straight edge (another handy Harbor Freight purchase) using the existing arches of the frame as a guide.  I first tried using a small laser level to project the shape of the arch, but the levels I found wouldn't sit flush against the stringers and projected lines instead of points so the intersection was not accurate (and I couldn't find a simple laser pointer anywhere in town...I thought these things were sold out of buckets at cash registers just about everywhere a few years ago). I did use the laser to mark every couple of inches on the arches so I could keep my scribe perpendicular to them. After scribing the arch onto the angled piece of wood, it was cut at the appropriate angle with the jigsaw and sanded until the contour was identical with the rest of the frame. Sanding was a time consuming process of checking the area with the 4 foot straight edge ruler, hand sanding the high spots, and repeating until correct.

The newly installed angled mounting surface board.

The reason for needing this board to be accurate is that I will be using that surface to lay up the mounting tabs for the back of the arch against it.  Since this is a rather critical area in the "lets not have the top come crashing down on our heads" concept, I want to make sure it is right. I'm probably being a bit of a perfectionist on this construction, but I figure the more accurate the mold is, the better it will look and the less "corrective" work I'll have to do later.

Finally, the frame was built.  I took some cellophane packaging tape and "laminated" the tops of the stringers and the newly finished angled mounting surface board so that fiberglass resin would not stick to it.  I'm finally ready to start work on the actual top.  Well, almost.  Since this is my first time working with polyester resin and Divinycell foam, I want to build a small test piece to practice my techniques with these materials and try a few layup concepts before I dive in with $2000 worth of materials.

More on the test next time.  But I feel I'm finally making some real progress on this project.  At least more than just building a shed anyway.

A Puzzle With All Beige Pieces

Still working on the top.  Ok, technically I'm still working on the frame on which I will build the top.  Where we left off, I had created the curved stringers on which I will attach the foam core so I can start adding fiberglass.  Of course, a stack of 5 of these stringers won't get me very far by themselves. I need them to be spaced at about 2 foot intervals so they will hold up enough foam to create a 12 foot by 9 foot top.

I thought the hard part would be cutting the arc for the stringers, but it has taken me quite a while to figure out how to connect them all together.  Just a pile of lumber...or beige puzzle pieces.  The trick is to assemble the beige pieces together to create a sturdy, yet lightweight, frame.  Ideally, I need the frame to be stiff and not flex when moved about or when tilted or stood mostly on end (I figure this might help with applying the glass...since I don't have 6 foot long arms).

Stringers all in a row

It didn't take me too long to figure out how to attach the stringers to a couple of 1 x 3 rails. Actually attaching them was a bit more complex.  Having to line everything up and keep it level and square can be a time consuming process. Fortunately I heard of a trick of using superglue to help hold things in place and it made the assembly a bit easier.

After the first two rails were installed and holding the stringers straight up and lined up, I turned my attention to attempting to straighten the warped pieces of wood.  I could rant about the quality of the lumber at the big box hardware stores again, but I won't. Another puzzle...how do you keep wood straight that really doesn't want to be straight.  When assembling things, I tried to orient various pieces so the warping would provide opposing forces that would cancel out...but that only went so far. Then I had to figure out how to keep the frame somewhat light and still be able to hold the wood where it belongs.

Bracing needed to keep the stringers straight and square.

A lot of thinking and a few cycles of trial and error and I think I have solutions to many of the issues. So, now I just need to finish the assembly and figure out the best way to transfer the complex curve where the top will mount to the arch.  I have a couple ideas to try there, so hopefully it won't take too much more thinking about the issue.

The "shop" closed for the evening.

With any luck, I can actually start fabricating the top soon.  I just hope all of this goes well and the result looks decent.

Bug Spray Update

A while ago I wrote a post about homemade bug spray.  Since that time I had a chance to test my original formula against noseeums and found that it wasn't as effective as I had hoped.  So, I continued playing with concentrations and ingredients and researching what others have tried or claimed would work.

It has taken some time, but I think I have an improved formula. The last several days at the boatyard I've been working well into the evening. And the area has plenty of mosquitoes, noseeums, and some species of horse-fly that are quite annoying. So, the past few days should have been a very good experiment. Although not 100% bite free, probably due to my application, I have to say I'm pretty impressed with its effectiveness.

My issue with application is the spray bottle I'm using. Somewhere in the move out to the boat, the old hairspray bottle I was using in Colorado was lost.  I had a small plastic spray bottle lying around the boat (I think it came with the installation kit for the boat name graphic), so I have been using it.  The problem is that it puts out a very fine mist and even in light winds I find it difficult to apply without most of it blowing away.  Since my wife isn't going through much hairspray these days, I haven't had a free replacement so I've been waiting for this bottle to die before I get another one.  Although the spray head leaks a bit (just like every bug spray bottle I've ever had), it has lasted over 6 months. Essential oils are known for eating plastics, so I didn't think the bottle would have lasted this long. I guess one of these days I'll have to just break down and buy some hairspray at the dollar store and use that bottle, hoping it has a coarse spray pattern.

Anyway, here are the changes I've made to the old formula. For the carrier, I've swapped out the witch hazel for a generic version of Listerine (original flavor). I heard someone used it in a noseeum spray. It makes a cheaper carrier than the witch hazel and is easier to find.  I've also added tea tree oil at about 1/2 the concentration of the other two oils.

So, the current recipe, using the 4oz bottle I am using is:

• 35 drops of Oil of Lemon Eucalyptus
• 35 drops of Citronella Essential Oil
• 15 drops of Tea Tree Oil
• 2 ounces of Original Listerine (or store brand equivalent)
• 2 ounces of water

Place the mix in a dark opaque container (or use something like duct tape wrapped around the bottle to "darken" it.. Shake well before applying as the water will separate.  I prefer a sprayer (just not my current one) but you can also just rub it on. I find I have to apply it every couple of hours or so when I'm working in the yard (or at least just a little before dusk if I haven't been sweating). Obviously if you go swimming it will probably wash off, so you would need to reapply.

The best part of this stuff is that it is rather pleasant smelling, unlike most bug sprays.  My wife jokes that I "smell lemony-fresh". Well, that and it is less expensive and I don't have a reaction to it like I do with the high concentration DEET stuff.

Obviously I cannot guarantee it will work for you, all I can say is a version of it has worked for me in Colorado as well as several places along the eastern U.S. coast. If you do give this a try, drop me a note and let me know if it works for you.  If it doesn't seem as effective, try higher concentrations of the oils.  All I can say is that it seems to work well for the bugs in Virginia.

Work Continues on the Top

I continue to make progress on the fiberglass replacement bimini top for the boat.  Sometimes it doesn't seem like much progress though.  It kind of feels like this: You want to build a house, so first you have to go into the forest to fell some trees, then saw them into lumber, then mine for ore to make iron to forge nails.  Ok, it isn't that bad, but it isn't too far behind.

I finally completed the table after giving the drill a chance to recharge for several hours.  With the screws in place, I then used boat stands to hold it up. Using five stands, one on each corner and one in the middle, worked out really well.  Since the height of boat stands are adjustable, I was able to level the table...at least as far as the warped lumber from Lowe's would allow.  It is now a good height for a workbench and will work fine for this project.

The work space all set up, not too bad for a boatyard.

After getting the table squared away, I turned my attention to the stringers.  When I left off, I had drawn the arc, cut it out, and did a little sanding.  I also took it over to the boat to confirm that it indeed would match the curve of the arch when standing vertical and, as best we could tell, it did (remember, the arch is at an angle so we had to guess the position while holding a board over our heads).  I did just a bit more sanding on it, and then used it as a template for creating 9 more stringer pieces.

Once the stringer sections were cut out, I clamped all of them together and started sanding them down so they were all the same shape.  This is where I ran into a problem.  I didn't get very far with the sanding before the random orbital sander I was using had a problem. (Yes, there would be better tools to use for this...but I live on a boat and don't have a full workshop of tools). The almost new sander got about 10% of the job done when the head stopped "orbiting". I turned off the power and lifted up the sander. The motor hadn't stopped spinning yet and the head of the sander went sailing across the boatyard. I know Harbor Freight tools aren't high quality (the reason I call them the cheap Chinese tool store), but this is the first time I've had one fail so quickly.  Apparently the screws that hold the head on the sander got hot and melted the plastic to which they were once attached.  The sander didn't even last through its first piece of sandpaper.

Stringers clamped together and sanded.

So, a little time was wasted going back to Newport News to return the sander (it was purchased there less than a month ago) and then going elsewhere to buy a new one. Since I was impressed with the Ryobi jigsaw I bought for this project while in Palm Coast, I decided to go with their random orbital sander (it was only about 50% more than the HF one and still a pretty good price).  The next day I resumed my sanding and was able to get each of the stringer pieces virtually identical in shape.  I had purchased some 40 grit sandpaper with this second sander and, while I can't say it made the job quick, I would probably still be sanding those things if I had gone with a less aggressive sandpaper.

Attaching the two stringer sections together.

As I previously mentioned, the stringers need to be 12 feet long, and the pieces I just finished sanding were only 8 feet wide.  This is why I needed 10 of them. I took two of the pieces, flipped one around to be the mirror image of the first, lined them up so they would overlap and result in one continuous 12 foot curve.  The next task was getting them all aligned properly and attached together. I drew lines at two and four feet at the wide end of each board, checked the level of the table, attached a stop block to work against, and lined everything up.  I then used some clamps to hold them together while I pre-drilled holes for screws and temporarily installed them.  From there I moved the clamps and half the screws, applied some glue, and reattached them.  Repeated on the other side, and the result will hopefully be a solid arched stringer for the project.

Stringers done and clamped to keep/make them a bit straight.

I'm trying to figure out how best to attach the stringers together at two foot intervals to create a straight frame for attaching the foam sheets. I thought I had a good idea for that but ran into a problem while trying to implement it yesterday. Now I am having to rethink the solution - the fun of this sort of project.

Once In A Blue Moon

Seems like a long time since I broke out the good camera.  So, after a hard days work I decided to see if I could get a couple pictures of the blue moon...

Blue Moon over the marina.

Moon still up early in the morning

And the sun rising on the Severn river.