Sanding, Cutting out Doors and Hatches, and Fillets Gone Wild

 The top panels have been sanded with 120 grit.  Time to do the sides.

 Here I have flipped the shell onto its starboard side so I can sand the port side without standing on my head to get at the lower panel.

 Was glad I had this little scaffold out in the shed.  It was purchased years ago for painting outbuildings, but I used it quite a bit on the PocketShip and now it's coming in useful here.

 Port side is sanded, now I'm getting ready to sand the starboard side. Next, the hatches and doors are cut out of the shell.

 Here you can see the galley hatch and ventilation hatch have been cut out.

 

And the doors.

 The book says to stop now and get inside the shell to enjoy it, which I did, but not after a thorough vacuuming first.  It's going to be pretty nice, and roomier than I had thought.

 But look what I found when I removed the galley hatch!  I knew the fillet material was too thin when I started it and would droop down, but not this much.  I thickened it and finished the fillet, but could not see this side (which was below at the time) until today.

So out came the grinder.  I ground off most of the excess fillet material, then finished up with the sander.  Needless to say I'll be painting the inside of the galley.  BTW, there will be a fillet added to the joint later, so those voids you can see will be filled in and covered by the new fillet.

Another lesson learned:  Do not use a saber saw to cut out parts of the galley hatch.  I had forgotten how close the bulkhead was to the edge to be cut.  This will be easy to fill, but once again I'm glad I decided to paint inside.  There are some places where I cut out doors and hatches that don't look so good--either I couldn't see the line well enough when I was plunging the cut in, or in one case the saw just slipped a little bit.  The slots that were pre-routed to mark where you cut were a bit wider than the saw kerf, so obviously there will be some sanding to do.  I hope that is enough to even things out.  If there is much filling to do, though, I'll probably have to paint instead of bright finish on the outside also.  Just have to wait and see.

Some Thoughts on Sanding Fiberglass

I'll be sanding the entire shell now to get it as smooth as I can.  This operation could wait, but I think it's best to do it before the doors and hatches are cut out.  Quite a bit easier too.

The first thing I did was to sand down the edge where that third layer of fabric laps over the roof.  This will be the only place where 80-grit sandpaper is used.  Right now you can clearly see the edge of the fabric and the sanded area to the left looks dull, but if I temporarily wet it down with denatured alcohol . . .

 voila!, it looks quite a bit shinier and that edge has almost disappeared.  If I could have taken the shot quicker, before the alcohol started to dry, it would be even glossier--closer to what it will look like with high-gloss varnish, but you get the idea.  The round-over will be sanded by hand.  A machine sander will cut through to bare wood very quickly.  The manual recommends a sanding block, but I will be using a sanding sponge wrapped with sandpaper, so it bends around the round-over better.

Here, I am starting to sand the flat portion of the rear deck.  There are some lines I had noticed before which appear to be creases in the fabric, even though I was very careful with it.  There are also shiny spots.  These areas are un-sanded and will be a bit rougher.  Also, varnish will not stick nearly as well.

So the object is to try to achieve a uniform matte surface.  This feels very smooth at this point, sanded with 120 grit.  Eventually I will go over the whole thing again with 220 grit.

And here's that approximate same area wet down with alcohol to give a better idea what it will look like after varnishing.  (You can see some shadows of my finger and head near the bottom of the picture.  I was working fast to get the shot before the alcohol could dry very much.)

 

Applying the Epoxy Fill Coat

Here's a couple shots showing the shell with its second (fill) coat.  Not much to see--maybe a little shinier than after the first (wetting out) coat.  I used a roller for the whole thing--seemed quicker and easier than spreading it with the squeegee first, and I think I got it very even.  This coat is kind of a balancing act.  You want to get it thick enough to fill the weave, but not so thick to cause runs and sags.  I spent a lot of time going over each section again and again and looking for runs or sags, then went over the whole thing again.  I didn't find any, but that doesn't mean they don't exist.  At least I kept it to a minimum as much as I could.

In an earlier post I said I thought it might be overkill to add masking to the quarter panels, but as you can see it was not.  Those runs would probably not have caused a huge sanding job, but one I am glad to have easily avoided them anyway.

I thought it was a good idea to follow the manual's advice and use any leftover epoxy to go ahead and seal the inside of the transom.  Turns out there was exactly enough--how did they do that?
But here is where I part company with CLC.  They say to follow this step with sanding and then to add the final (gloss) coat, but according to MAS you can add the next coat as soon as it the previous coat achieves "surface tack."  There are several reasons I have decided to do it their way, although I'm sure the guys at CLC know what they are doing.  But here are my reasons:
1. If the fill coat was not thick enough in spots, you could sand into the fiberglass fabric, weakening it.
2. Applying the gloss coat before the fill coat loses tack ensures a good chemical bond.
3. It's just a lot easier.  Even though CLC never says to sand the final (gloss) coat I'm willing to bet you will have to.  So that's an additional sanding plus you've got to clean up all the sanding dust before you can apply the gloss coat.
4. This will provide two coats of epoxy over the fiberglass fabric before any sanding is done, greatly decreasing the chance of sanding into the fabric (repeating myself here for emphasis).

Glassing the Port Side

I checked around 3:00 and the starboard side was no longer tacky, meaning I don't have to worry about fluff from cutting the fiberglass fabric sticking to it.  So I went ahead with the port side.  (I like using the nautical terms, "port" and "starboard" because it doesn't matter which way you are facing, they are always the same. If you are facing forward and then turn around, the left side becomes the right side.)

I learned quite a bit from doing the starboard side and applied what I learned to the port side.  For example, I put pushpins in all along the edge, about 2 inches from the joint, to guide my cut.  The edge of the cloth doesn't have to be exactly 2 inches from the joint for its entire length, but here at the very top it can't be more than 2" because at the bottom . . .

it barely reaches to the edge.  It can be a tiny bit short, because later on that bottom edge gets rounded off.

 As you can see here there is a LOT of cloth to be cut off at the front (you should be able to make out the push pins marking the 2-inch line) . . .

and quite a bit at the back.

Here are the scissors I use to cut fiberglass cloth.  I ground off the pointed ends so they are less likely to get snagged in the cloth as I go along.  It helps quite a bit.

 And here is the 'glassed port side shown from the front . . .

and back.  As soon as the port side achieves tack I can go ahead and start the second coat of epoxy on the whole thing.

Glassing the Starboard Side

I decided to only do one side today for a couple reasons: 1. You have to trim the fabric in places on the top, and this leaves a lot of fuzz floating about.  The only way to get rid of it is to blow it away, which would send it to the work I'm trying to do on the other side.  2. The book recommends tilting one side up to facilitate the epoxy.  It would be a real mess to try to let that side down with wet epoxy on it and lift the other side.

First, though, it is necessary to smooth (feather) out the edges of the layers of fiberglass that were laid on the top panels.  This is so the fabric "flows" from the fiberglassed area to the new area without any gaps under it.

This is a section of the fiberglass edges prior to sanding.  They are not selvedged as the manual states, which would have made them very thick.  Instead, a few horizontal strands have been removed, leaving edges that are ideal for feathering.  In this case, the lower edge is further out than the upper, which is good too.  If I'd thought of it I would have purposely done it this way everywhere.

Here's that same area after some initial sanding with 120 grit.

And here it is after final sanding with 220 grit.  The edges should disappear entirely under the new layer of fiberglass.

I did have some problem areas though:

There are some fairly deep pits here. I may have laid the epoxy on too thick, or this may be the place where it had started to kick and I had to discard it.

The tough thing to decide is how much to sand.  I think I have sanded as much as I dare--any more could begin to weaken the structure.  After adding more fabric followed by three coats of epoxy, I may be able to smooth it out.  If not, then there's always filling with Bondo and painting instead of "bright finishing."  Or, instead of painting the whole thing I could bright finish and add racing stripes over the edges.  Maybe white vinyl?  I don't think I'll race it though.

The dirty deed is done. It was quite a struggle and I'm really glad I didn't try to do the sides at same time as the top--what a mess that would have been!  There was a ton of extra cloth on the bottom panels at each end and I thought I would never get all the wrinkles out.  As usual I didn't have anybody to help me by pulling on the other end at the same time, but I found that by wetting out the fabric first I could manage it.  The epoxy held the cloth in place on one end while I tugged (hard) on the other (and back and forth).  I was happy to find that I could use push pins to hold the top of the cloth in place until I had some epoxy on.  I could push them in by hand but decided to use a plastic mallet to save my fingers.  They only had to go in a tiny bit to hold.  Any holes they left were immediately filled with epoxy.

I am going to raise the other side up quite a bit higher when I 'glass it.  For the bottom panel I had to get the epoxy on a surface that was leaning away from me--not fun.  Also there are small areas at the very bottom where ruffles in the cloth are pulling it away.  I'll go out and re-stick them every now and then until they stay.  If I can't get them all stuck down it should not be a structural problem that close to bottom.  I'll just sand the ripples off.

And here's a shot from the front.  As you can see I didn't bother to epoxy the part that will be cut out for the window.  You can also see the little stool I used to hold that side higher.  Going to go with something bigger on the other side.

First Outside Fiberglass Layers

First, a little more prep work.  Here you can see I laid some plastic on the floor to catch any epoxy drips that otherwise would have to be sanded off later.  You can also see I masked off the two quarter panels, although that may have been overkill.

After laying on the first layer of cloth, I was having a lot of problems with wrinkling where the cloth has to curve down on to the #2 panels.

The book says to tug gently on the corners, and when that didn't work I posted a question about it on the forum.  The answers were mostly a repeat of what the book said, but what I figured out is that I had to tug much, much harder than "gently."  It's starting to look a little better.

Here are the first two layers of cloth all laid out with very little to no rippling on the edges.

After laying down the first (wetting out) coat of epoxy I found that the ripples reappeared.  Most of them were easily smoothed out by more (hard) tugging on the corners.  In some cases I had to be sure and grab the lower layer of cloth because that turned out to be the one that was wrinkling and showing through on the top layer.  After all that I still ended up with some very minor ripples on the port side near the front.  After waiting a couple hours until the epoxy was starting to gel I was able to push them down with a plastic spatula. 

Here's a nice shot of the epoxied fiberglass from the rear of the shell.  There are some areas that sort of look like wrinkles, but that's just the wood grain.  Light is reflecting on the 'glass on the #2 panel that makes it look like it is not stuck down, but believe me, it is.  You may notice a bit of whitish area at the very top where the cloth is not adhering.  That is part of the ventilation opening that will be cut out and discarded, so no worries.  BTW, you can also see that the entire #2 panel has been covered with epoxy.  I found that impossible to avoid.  I have to sand down the edge of the fiberglass anyway, so I can give the panels a good scuff sanding at the same time.  I decided to do the top first and the sides later.  I'm glad I did--probably would have messed up all my nice smoothing trying to get the side layer on without someone to help, and how would I have done the re-tugging during epoxy application?  It would have been extremely hard.  Also I doubt if I could have smoothed out the remaining ripples after the fact. I'm going to try using push-pins through the previously epoxied layers to hold the side panels in place (even though the book says you can't).  If that doesn't work I'll have to use masking tape, but it is really hard to get off the cloth without distorting it and pulling out a bunch of threads.

Important Note to Builders:  I had forgotten the lesson I learned on the PocketShip, which was not to use the cheap blue nitrile gloves when epoxying a large area like this.  The roller handle is bound to get very sticky and start ripping off parts of your gloves.  In this case it ripped the entire palm of the right-hand glove off.  I got them both off as quickly as possible, cleaned my hands, and put on "Venom Steel" gloves.  They are heavy-duty black nitrile, 70% thicker than the blue gloves.  Please, unless you are somehow so fastidious that you don't get any epoxy on your tools, use these gloves when epoxying large areas.

Preparing Shell for Fiberglass

Here the camper shell's mold/cradle has been removed.  To get it up off the floor CLC recommends building mini-sawhorses.  I happened to have some plastic bed risers that I no longer use.  They are the perfect height and saved me some work and time.

 The next step is filling in all the joint seams with thickened epoxy.  Much of the fillet epoxy came through from the inside but there are gaps.  The masking tape is to remind me not to bother filling from that point to the rear of the shell.  You may recall that section will be cut out to form the galley hatch.  Part of the cut is right on the joints that were never filleted or taped, so no need to fill.

I have filled every groove, hole and ding (I hope!) and the shell is ready for final sanding and fiberglassing.  My only small regret is that I didn't get off my lazy butt yesterday afternoon and finish the seam-filling with thickened epoxy.  Now I have to wait 24 hours before I can do the final sanding.  I could have done the final sanding before filling the seams, but after filling them you wipe them down with alcohol.  It turns out that the alcohol raises the grain, so those areas have to be sanded again.  A quick and easy job, but necessary.  Still, I plan to go over the entire shell again, just to make sure I didn't miss any spots.

Securing Floor to Shell

The fillet and fiberglass tape securing the floor to the camper shell have been installed.

This job proved arduous.  The filleting was not so bad, but as the day wore on the garage temperature rose to about 79.  Not bad, but inside the shell it must have been around 90 what with the lights (even though CFLs) and my body heat.  After I got all the tape sections in place and was painting on the epoxy, I got about 1/3 to 1/2 way through and noticed the epoxy was starting to gel.  I felt the container and it was very hot.  So I had to discard that batch.  I set up a fan to cool down the interior of the shell and mixed up another batch--only half as much.  This was enough to finish the job, but I was trying to hurry so the fillet would still be slightly pliable and tacky.  The hardest part was trying to get the epoxy smooth while lying in the cramped entryway, which will be the footwell on the finished camper.  It is under the galley floor.  It was impossible to keep the inevitable epoxy drips off my arms (should have been wearing a hazmat suit, but it was too hot and I was trying to move too fast).  I finished applying the epoxy in time to crawl back in and smooth the fillets around the bulkhead.  Then I cleaned up my hands and arms with waterless cleaner followed by soap and water.  Pretty sure I got it all off in time, as it only started getting on me toward the end when I had to lie in the entryway.  NOTE: Never try to remove epoxy from your body with alcohol or other solvents--it just spreads it and makes things worse.  Use a waterless cleaner like GoJo or Fast Orange, followed by soap and water, and lots of rinsing.  Otherwise you could develop an allergy to it and never be able to work around it again.

The book says nothing about any subsequent coats to the epoxy/tape or any sanding.  This is not really necessary since the following two coats (fill coat and gloss coat) do nothing to increase the strength.  They are cosmetic only.  Normally I would get in there and apply at least one more coat anyway, but the joint is near the floor and after all I went through to get on the first coat I'm going to leave it as is.  Besides, I just checked and it has already passed tack and thumbprint stage, which means I would have to sand it--weakening the fiberglass.  It's fairly smooth and near enough to the floor that I'm not going to worry.  Besides I am going to add the "headliner" which it appears to me goes all the way to the floor and would cover most of the joint.

Tomorrow I get to remove the mold and start fiberglassing the exterior!

Finishing Cabin Floor and a New Milestone

 I filled in some spots where the glue didn't quite flow all the way through.  Not structural so I used Famowood filler--takes 15 minutes to dry, vs 24 hours for thickened epoxy.

 Here is the same area after filled-in spots have been sanded.

 The floor has been covered with epoxy, except for the outer couple inches.  That is so the glue fillets will stick better when the floor is attached to the cabin walls.  (I feel compelled to mention something here about the application of multiple coats of epoxy.  I don't know why this is not mentioned in the book, but it is very important to get a good chemical bond on subsequent coats.  Rather than write about it myself, I refer you to the MAS Epoxies website FAQs: MAS Epoxies FAQ
Please read through FAQ 11. What is the Recoat Time? and FAQ 23. Checking for Hardness - Using the Rule of "Thumb."   Summarizing, the best chemical bond is achieved while the surface is still tacky--and no sanding is required. But, if you wait too long, a good chemical and mechanical bond can be achieved after tack but while you can still leave a thumbprint in the epoxy--light scuff sanding required.  Wait until the epoxy hardens and you will have to sand it for a mechanical bond only.  This is especially bad after the first wetting-out coat because you are bound to cut into the glass fibers while sanding, weakening them.)    

Here the floor has been placed on the shell and is being held in place with temporary drywall screws.  Plywood scraps are used to keep the screwheads from damaging the surface of the cabin shell.  

NOTE: The manual explains this step on page 109.  On page 118, they inform us that they had covered the underside of the screw-pads with packing tape so epoxy would not stick to them.  That would have been nice to know back on 109!  One of the advantages to reading the blog is, now YOU know.

Milestone:  The shell is rightside-up at last!

The only entrance for now is under the galley.  Once the doors have been sawed out, that particular opening will be covered with a transom.  For now, I have to go inside and fillet the floor to the walls, and a few other jobs. Then the mold cradle comes off for good.