Fiberglass or Plumbing

One of the tasks I've been struggling with on this hardtop bimini build has been how I was going to apply the edge detail.  I wanted to use PVC for all the reasons I had previously mentioned (makes a good handhold and can integrate supports for the enclosure) but thus far my attempts to get it to follow the curve at the front of the hardtop have been foiled.

I decided the only real option I had if I wanted to use the split-in-half PVC pipe was to use heat to attempt to curve the pipe by hand.  No tools or jigs, just the heat gun and a flat piece of board.  I did trace the curve from the top onto the board to help use as a guide, but that was it.  I would heat up the pipe until it was somewhat flexible (but not so hot it would deform or flatten on its own...which it will do at the right temperature) and gently curve it along the line.  Every so often I would take it over to the top to test the fit.

The result was a curve that wasn't perfect, but using clamps I could get it to stay in the right position along the edge of the top.  I had heard that superglue could be used to hold it in place until it was glassed over, but the texture of the raw fiberglass made me wonder if it would stick well enough.  Not having any better ideas, I gave it a try.  The superglue did the trick and I was able to glue each section of tube down in 5 or 6 places using clamps.  After a couple minutes I could remove the clamps and the tube stayed put.

So, now I have all the PVC pipe attached to the top and it looks like it will work just fine.  So, I guess my stint playing with plumbing is over for now. Next, I need to create the finishing pieces for the ends of the PVC.

Rain Delay

The last few days we have been working on some of the details on the underside of the top.  I mentioned in the last post that we added the support beams and wire chases and started glassing them into the top.  That work was completed yesterday.  We used the 1708 fabric to create a good structural fiberglass skin and tabs to secure the new structure to the top and distribute weight. Where there are tight corners and other places where the 1708 wouldn't conform to the surfaces, multiple layers of chopped strand mat were used.  As a result, except for the edges, the structural glass is complete.

Next on the agenda was to work on the detail around the side and front edges of the top.  For this, the plan was to take 1 inch thin wall (schedule 20) PVC pipe, split it down the middle lengthwise, and glue it on the top and bottom of the edge to create an oval shape around the edge.  The PVC would then be glassed over to create a convenient hand-hold. A slit would be cut along the bottom to accept the bolt rope for panels of a dodger (enclosure) set.  Unfortunately, I found that the split PVC would not bend around the front curve as I had hoped it would.

Guided by the suggestion of a few blog readers, I tried using heat to bend the pipe and that allowed me to reshape the PVC.  I tried to create a template or jig to help make a uniform curve by hammering a bunch of nails into a board along a curve.  Unfortunately I found another issue with this approach.  While the curve appears fine to the eye, it is not a constant curve in two-dimensional space (remember it was drawn along the curve of the arch using my string-compass technique).  This means a jig that replicates a curve segment won't work for the entire curve.  Best guess now is that I may have to slowly curve it by hand until it is "close enough" and then use clamps to hold it in place while the glue dries.

And that brings us to today.  A storm that was predicted for this weekend seems to have arrived.  It includes high winds and rain and makes work in a tent made of tarps assembled in a boatyard almost impossible.  So, instead of working directly on the top, today was spent doing some shopping.  We went looking for a paint sprayer I can use to apply PVA once the gel coat non-skid texture is applied to the finish.  We also went looking for a new baking sheet for our propane oven after figuring out that a pizza we had wouldn't fit directly on the rack in the tiny oven (the oven box is only 11.75 inches deep by 17 inches wide). I also needed a new pair of sandals as my current ones were starting to fall apart.  So, getting some things done...but just not a lot of progress on the top right now.

The weather is supposed to be a bit touch and go for the upcoming week, so it will be interesting to see how much we can accomplish.  I have to say I'm more than just a bit anxious to get moving again, so I find all of these weather delays a bit annoying.  On the bright side, some friends are supposed to be coming for a visit soon and we have been forced to take a few breaks and be more social...which have been good things.

In an unrelated note, I saw this while shopping today.  Is it just me or is Kohl's jumping the gun a bit on the Christmas shopping season?

Well, the Weather Is Different

After my last post, the weather has continued to deteriorate.  The temperatures have been nice both for living and for getting fiberglass work done during the day, but it has been overcast with random rain showers, fog, and mist.

I've done my best to get some work done in between the worst of it, but it continues to be slow. After attaching the foam and channel to the top, I hand-sanded the foam so it matched the curve of the PVC post and would allow the heavy fiberglass fabric to curve around the edges. I'm using the heavy 1708 fabric to cover as much of the structure as I can since it creates a strong laminate without too much bulk.

On one day I cut a couple small pieces of cloth to glass in part of the structural beams and wire chases. Just as I added the catalyst to the resin and started mixing, the rain came.  I managed to get them done without getting anything wet, but cutting any more cloth or trying to do any more work would likely result in water contamination.

Today we got one more piece in place and then it started to rain again. So, at this point, we have the beams and chases about 70% glassed in place.

The shaped partially glassed beams hiding from the rain under the tent.

One evening after the rain let up, I tried shaping one of the cut PVC pipes to match the curve at the front of the top.  A few readers had recommended, either via blog comments or private messages, that heat would work to bend the stubborn PVC.  They were right, heat does make PVC soft enough to bend. I was able to curve the C-shaped pieces but it wasn't easy and I didn't get the curve quite right.  The heat gun makes the PVC very flexible, and if I wasn't careful, it would flatten or fold or warp. And did I mention that when it is in this state, it is a bit hot to touch? I'll need to come up with some sort of jig that can hold the C shape and form the proper curve in order for this to work...but I have high hopes that it will.

MacGyver-ing

I know, it's been a week.  But how many different ways can you say that we battled weather issues and laid up large pieces of fiberglass?  The weather has improved slightly the past week.  We've gone from way too hot to laying fiberglass during the day to it being too cold for fiberglass to cure at night (oh, and the battle with the dew on the canopy is a daily task, deploying heaters under the tent at night and in the morning). And then there is the occasional rainstorm that halts work from time to time. But we have been able to complete the two main layers of 1708 fabric that cover both sides of the foam as of yesterday.

In the process of completing this task, we've had to come up with some tools to allow us to reach the center of a 9 foot by 12 foot structure.  We've also had to figure out other tools we needed.  The usual tools like fiberglass layup rollers, plastic spreaders, and polyethylene mixing cups are always needed.  Paint rollers to spread resin are also needed, and trying to figure out what covers are compatible with polyester resin was fun.  In the case of the roller cover, we quickly gave up on the idea of cleaning the covers and determined the cheap foam rollers at the local Lowe's work well.

Fiberglass table, bucket on a stick, and more traditional tools.

To get a plastic spreader to the middle of the top, we found a cheap stain spreading pad that could be screwed to a painters pole and then taped, yes duct taped, a spreader to the end.  To pour resin at the appropriate spot in the middle of the top, I needed to create a bucket on a stick.  I found a large scrap hose clamp at the boatyard and screwed it onto the end of a scrap piece of wood.  I could then attach a quart size or smaller mixing cup and reach past the middle of the top from any side.

Tools aren't the only place where thinking outside the box is handy. I've probably mentioned it before, but one of the design features of the top is to have a couple of integrated wire chases (for lighting and solar wiring).  But how do you make a 2 inch by 4 inch hollow rounded rectangle on the underside of the top.  Wandering the local big-box hardware stores, we were asked several times if we needed help.  Unfortunately I've found that most people employed there aren't very good at thinking outside the box, and when I describe what I'm trying to do I usually get the "deer in headlights" blank stare.  So we wander the aisles and try to come up with inspiration.  What we found was a 4 inch square hollow plastic fence post.  If we cut it in half lengthwise, it should make two forms we can then fiberglass over to create the channels.

Jig for splitting PVC pipe into C shapes.

Another feature of the top is that I would like to make integrated hand-holds around the edges of the top.  I'd thought about taking some of that foam pipe insulation and sticking it over the edges of the top and then glassing over that, but discovered a better idea.  I was told of another blog, via a comment on one of my posts, where the couple built their own hard top.  They had the same idea for handholds, and they used PVC pipe split in half to create the form.  The neat thing about this approach was that they could then cut a slit in the PVC to create a bolt rope holder for installation of a dodger.  I quickly decided this was a great idea and incorporated it into the design.  But how do you split a piece of PVC in half lengthwise.  The other blogger I think used a hacksaw (and obviously a lot of time) to make the cuts.  I know I wouldn't be able to cut it straight enough by hand, so I came up with another option.  Using some scrap wood from the build of the form, I was able to create a jig I could clamp onto a tablesaw and it worked perfectly to split the tubes in half.

Of course, this presented a new problem I need to figure out.  I thought the thin wall PVC would be a bit more flexible once it was split, but I don't think I can bend it to match the compound curve along the front of the top.  I tried making some cuts in the C shaped PVC to see if I could get it to bend better...but if you think cutting a tube is hard, try cutting the sides of a C shaped bit of PVC.  I'm sure I'll figure out something, but it is another problem to deal with.

Meanwhile, I've got wire chases and other structure to install while the weather continues to somewhat cooperate.

Update: You are reading this fairly late because we were invited to dinner by new friends at the marina.  Yes, while the focus has been to get the top done, we have taken a few breaks. They made a multiple course Indian dinner that was absolutely awesome...thanks Stuart and Julia! The picture below shows the wiring chase and other structural foam just before I glued it to the top with polyester resin.  Hopefully tomorrow we can start glassing that in place.

The split "fencepost" and additional foam structure.

What Was I Thinking

I wonder how many times I've asked myself that question during this project.  Let's see...I probably average 3 times a day for...now how long have I been working on this project...never mind, I don't want to know. Anyway, work has been continuing in between oppressive heat and torrential rains.

After getting the top flipped over, we needed to start glassing the other side.  Of course, there is one little issue.  Fiberglass cloth doesn't really like to bend around sharp corners. If you did manage to bend it into a sharp corner you would break every glass fiber so the result wouldn't be structurally sound.  Now the mounting flange at the back of the top is a 60 degree acute angle, so this would be an issue for the glass. To resolve the problem I needed to fill the angle and make a radius the cloth could handle.

I took some of the scrap Divinycell foam board and cut little 1/2 inch isosceles-triangle-shaped filler strips using a table saw (borrowed from a friend at the marina) and a fine tooth blade.  I then mixed up some thickened polyester resin and glued the triangle strips along the edge of the flange where it meets the top.   Making some even thicker resin (near peanut butter consistency), I created little fillets at the edges of the triangle using Popsicle sticks so the cloth would have a more reasonable transition from the bottom of the top down onto the flange.

Triangle filler strip to curve fiberglass around the sharp corner.

That evening, after the triangle strips cured, we applied fiberglass along the mounting flange and around the corner onto the bottom of the top.  Due to the curves, and to make life a little easier on this step, the pieces of fiberglass were only long enough to cover the flange and about 18 inches of the bottom side of the top. As usual, we ended up working late into the night...I think it was about 1 AM by the time we got back to the boat.

In order to make sure the first layer along the bottom was one nice homogeneous piece of fiberglass, we wanted to get an early start so we could apply the two 12 foot long pieces of cloth that would complete the first layer. Apparently mother nature had other ideas .  This time, unlike the high temperatures that she had been throwing at us, we woke up to a thick fog.  Our hearts sank.  High humidity isn't good for fiberglass and water vapor, well it can ruin the stuff. Was all the work the prior night ruined?  I made my way out to the work area to inspect the project, bracing myself for the worst.

Fortunately, when I peeked into the tent, I found that the fiberglass had apparently cured well enough before the onset of the fog.  There were a few drips of water where it had condensed on the inside of the canopy and dripped down, but I was able to blot them up and everything dried out just fine.  Of course, thanks to the humidity levels, it took a LONG time to dry out and only the heat of the day ended up doing the trick.  By the time everything was dry enough we could continue, the temperatures at the work site were climbing past 95 degrees. We had to postpone the application of the second strip of fiberglass until it cooled off that evening.  We spent the evening applying the next strip, completing the task around 11 PM.  Oh, and the humidity levels the past day apparently hatched every sort of insect within a 5 county radius, so we were fighting off swarms of various insects attracted to our work light. I wonder what the acceptable number of insect carcasses is per square foot of fiberglass. Let's just say it was a trying evening.

After getting the second strip of fiberglass installed, we got up early the next morning in an attempt to get the final piece for the first layer on before the temperature got too high.  This time there was no fog, but there was a fair amount of dew on the boat.  I go back to the work area, and the canopy was covered in dew as well. A couple drips were again found on the fiberglass and foam and were blotted up with a rag. Then everything was left to dry for a bit.  Not wanting to wait until the evening, we started applying the fiberglass as soon as the foam was dry and the water spots on the fiberglass disappeared. It was a very different experience applying the glass during the day.  I thought that the added ambient light would help to find any issues during the layup, but it only helped right around the edges.  Everywhere else there was a fair amount of glare from across the top so it seemed visibility was about like working at night.

Yeah, a little warm for fiberglass.

There was another difference working during the day.  The higher temperature means the resin cures faster, even at the minimum catalyst amount recommended by the manufacturer.  When we started the layup, it was 84 degrees. The first couple feet worth of fabric went OK, then I noticed I was starting to sweat.  I checked the temperature and it had jumped to almost 90 degrees.  I could tell the resin was going to set up quickly so we had to work very fast.  In the rush to get the fabric wet out properly, we used a bit more resin and were unable to get it cleaned up or redistributed before it started setting up. As a result, the fiberglass has some excess resin floating on top of the fibers in a few spots.  Guess we will need to sand those spots down a bit so we don't end up with an overly resin rich layup. When we were done, I looked over at the thermometer and it was reading 104...no wonder the resin was setting up so quickly. That is why I was avoiding this process during the day but the lack of progress and forecast of lower temperatures convinced me to give it a try.  Guess we won't be doing that again for at least large layup areas.  The irony is that I think we actually do better work at night, probably due to the extended working time at those temperatures.

First layer applied...might need a little cleanup on the front section.

Making this top has been an interesting, if not incredibly slow, process.  I wish I had been able to find a place were I could do this work indoors as I think it would have alleviated many of the issues that slow down the process.  I should have chosen a different time of year as well, but the desire to head to the islands in the fall only left the summer (hurricane season) to get the work done. Of course, ideally I would have found a reasonably priced top I could have just had installed.  What was I thinking.

Turning Turtle

I came up with a design for the finished top and made some plans for how to get there.  But, as they say, the devil is in the details.  I've spent a lot of time at just about every step of the process thus far figuring out some detail related to the build. How do I create the desired arch, how do I calculate and lay up the complex angle that makes up the mounting flange at the arch, and even how to apply 9 to 12 foot long swaths of glass fabric (particularly when you can't really reach the middle of a large surface).

The last couple days haven't been any different.  With the main layers of fiberglass applied to the top of the top, it was time to flip the thing over so we could start applying the glass to the underside of the foam.  Now it sounds like a relatively simple task to flip the thing over, but there are some catches.  Once the foam core of the top has fiberglass on both sides, it becomes rigid and can support weight.  With the fiberglass on only a single side, the expectation is that it is still rather flexible.  And if it flexes too much, I could imagine that the bond between the foam and the fiberglass skin could be weakened (I don't know if this is true, but it seems like a possibility). When the top is upside down, I expect that it will not retain the curve without assistance too.

Then there is the frame to which the foam was attached.  It held the foam at the desired curve (mostly) with a number of loops of 30 pound test fishing line. While this was sufficient for holding the foam above the frame, what would happen if we flip it over?  Would the lines pull through the foam? Can we keep the frame attached or will we have to remove it before turning it over? And will we be able to control this huge piece of plastic and foam as we flip it end-over-end?

Oh, and did I mention the canopy top?  It is a 10 foot square (measured at the feet, so the top is actually only about 9.5 foot square), and the bimini top is 9.25 foot by 12 foot at its widest points.  Can we maneuver the top out from under the canopy or will I have to disassemble all of the tarps and move the canopy first?  All of these problems to work through on the flip are similar to what I've had to figure out at every step of the process.

We thought about it for a while and came up with a plan that seemed like it would work. We recruited a few friends at the marina (6 of us total) to help with the task.  Naturally, as we were getting ready to flip it over, storms started brewing on the horizon.  As the fiberglass is still rather raw and water soaking into the foam would not be a great idea, we had to button everything back up and wait for it to blow over.  After the rain, I was able to get everyone back together and we were ready to go.

The process was that we would first move the top and frame off of the boat stands that had been holding it up, maneuver it out the side of the canopy, and carefully set it aside.  Then we would place the tabletop I originally built back on the boat stands so we would again have a table.  Using a bedspread we bought at a local thrift store, we added a bit of padding to the tabletop and then covered it with plastic so no drips will glue it to the top. We would pick the top and frame back up, carefully set it on the rear edge (where the fiberglass wrapped around the foam) and then continue flipping it over supporting the top with the frame just sitting on top.  The next step would be maneuvering the upside down top back under the canopy and carefully setting it on the table.  We would then use the scrap curved pieces of plywood, padded with pipe insulation foam, to cradle the top on the table and help maintain the curve.

Surprisingly, the move went about like we planned.  At first we didn't get the boat stands supporting the tabletop well positioned, and the tabletop started to pull apart after we set the top down.  We quickly went around and lifted the top and table to reposition the stands.  After slipping the padded cradle wedges into place, the top has successfully become a turtle laying on its back. Our friends Stewart and Julia brought a bottle of champagne to celebrate the event, and then we adjourned to our boat for some drinks, snacks, and conversation to complete the evening.

Working on the support cradle.

The next morning I went out to adjust the position of the top on the table and tweak the position of the cradle wedges before attaching them to the table.  We also cut up some foam pool noodles to help add support and distribute weight.

One of the things I haven't really mentioned yet is an issue we discovered while doing the layup on the top side.  Polyester resin shrinks when it dries.  As a result (we think anyway), we noticed that the side edges of the top pulled away from the frame a bit, stretching the fishing line as it did.  The movement was about 3/4 of an inch.  Since I assume applying fiberglass to the other side of the top will result in it curling back the other way, I set up the cradle wedges so the ends of the top were held about the same distance from the end of the frame.  My hope is this will allow for any shrinkage of the fiberglass laid on the bottom of the foam...and if the bottom doesn't shrink that much it is OK as the only place the curve is critical is at the arch, and the laid up tab at that position seems to be holding the curve well.

Removing the mold/frame.

Up to this point the frame was still mostly attached to the top in order to make sure we got the cradle positioned well.  But it was time for that to change.  Using some wire cutters, I carefully clipped all the fishing line flush with the foam.  We then tried removing the frame.  Unfortunately, but not surprisingly, the frame was stuck along the back edge.  When I made the frame, I intentionally didn't glue the back piece to the rest of the frame, it was only held with screws.  I removed the screws so we could remove the rest of the frame...but it still wouldn't budge.

Don't let anyone tell you that polyester resin doesn't make a good glue.  The resin that dripped through the holes we punched in the foam managed to glue the back lip of the mold to the rest of the frame despite my best efforts to not do that.  In the end I had to cut the supports to move the rest of the frame.  Using flexible putty knifes we were able to work the last piece of the mold free.

Almost ready to glass.

Now with the top free of the mold, it is time to clean up (sand down) any remaining drips and bumps, clean the PVA off the back of the mounting flange so the next layers of fiberglass will stick, and then we will be ready to apply more fiberglass.