GENERAL FIBERGLASS REPAIR
(referencing a Yamaha 2005 AR230HO where the hull skimmed over sunken concrete, resulting in engine bay damage)
FRP = Fiber Reinforced Plastic = fiberglass
I successfully made it through several seasons without any indication of failure of the repaired areas before eventually selling the boat. I put the boat through some rough swells and I never once worried about taking it easy due to my repairs. In the end, the engine bay actually looked better than it did when I first got it - from what I can tell, there was a late FRP modification from the factory.
BONDING SLOPE: With the most-aft area cleaned up, I ground the area surrounding the repair an additional 1.50" from the hole, at a bevel. I used an angle grinder with a coarse sanding disc for this. There should be at least a 10:1 ratio (12:1 is optimal) for the new FRP to adhere to old FRP. The factory glass is generally 1/8" to 3/16" thick, so that is roughly 1.50" out from the repair hole. Use 60-80grit sandpaper to ensure a good bonding surface.
SHAPING: This area needed a particular shape but there would be no way to remove the support following the layup. So I shaped and permanently filled the port and starboard voids with Darice (brand name) Foam Sheet. It was the most practical option I could come up with at the time even though it may not be the optimal solution due to moisture. However, this foam is far from a sponge and the bilge pump should keep the foam from ever sitting in water. Maybe using spray foam (which is moisture and mildew resistant) and shaping it would have been a better option as it should be solvent resistant once cured.
The bilge pump area was bridged with plexiglass. I precut the plexiglass with a 3.25" hole saw and taped the circle back in so I could easily remove the hole after gelcoat. I did this so when I cut the smaller 3" hole, the plexiglass edge would not be seen. I made sure to always keep the location of the screw visible that helped support the popsicle stick (which kept me from accidentally pulling the plug out) so I knew exactly where to place the 3" hole saw when the gelcoat was cured. I used epoxy to secure the bridge (the plexi is still there, hidden under the new FRP) and then added more foam to smooth out the transitions.
RESIN CHOICES: I considered the three common types of resin. The factory uses polyester because it is the most economical. Some recommend epoxy because it will make a very strong bond to anything, has a long shelf life, is fumeless, and will cure no matter the ambient temperature - I imagine these are the same people that think 3M 5200 should be used on everything. Epoxy is very expensive, has less flex than the others, and has minimal off gassing (if any). One problem with epoxy is that gelcoat doesn't always stick to it, so I try to avoid it for this reason. Note that Marine Tex is an epoxy. I chose vinyl ester because it is a good compromise between price and strength. Essentially, it is a stronger, more water-resistant version of polyester. Like poly, it requires MEKP (ketone peroxide - the catalyst) to cure - always. Anytime I mention a resin or gelcoat (gelcoats are poly based), there is always MEKP mixed in. The main difference with vinyl is that it has a fairly specific MEKP mix ratio (2%), whereas poly has a bit of range (1-1.5%) which can be utilized for different ambient temperatures and desired cure rates. Overall, I was extremely happy with the vinyl and used it in future repairs without any thought. Once I did some tests, I knew I did not need the higher bond-strength of the epoxy.
REMOVING DAMAGED FRP: There were a few deep cracks within the engine bay floor caused from the hull flexing into the it. These were bad enough to compromise the integrity of the FRP and would have likely continued to spread or the fibers would be infiltrated by water if not removed. I had to grind the gelcoat off to determine how far the damage had spread. To find the extent, remove the gelcoat around the crack and wipe the exposed FRP with acetone. The acetone will highlight the damaged areas of FRP - good FRP looks pink, while damaged FRP is lighter. Once the cracks are found, you'll need to completely grind/cut them out and then grind a bonding slope (1.50" in my case) for the new FRP.
To support the layup of new FRP, I took pieces of dry 1.5oz mat and wetted them out with resin then allowed them to cure flat. Then I used tin snips to cut the shape I needed, added a screw as a handle, applied some epoxy, fished the backer into the hole, and used the screw to pull pressure while the epoxy cured. For some of these areas with curves, I placed wax paper over the hole so I could lay the resin-wetted mat on top of it - this would then cure into the proper shape and I would again glue it in from the back of the hole.
PEANUT BUTTER: Before laying in my first piece of mat for the layup, I mixed up what is referred to as "peanut butter" due to its consistency and used this to level the offset between the backer and the backside of the old FRP. This provides a nice slope which prevents any bunching of the first layer that could result in air bubbles on subsequent layers which will be weak/brittle points. There were also surface-only gelcoat cracks in the bay area that I ground out with Dremel bur #117 which were not wide enough for glass mat. These cracks just weren't deep enough, in my opinion, to warrant the 1.50" slope. So I ground them out at least 1/2 the thickness of the glass, but never the full depth. I then filled these new grooves with peanut butter as well.
This mixture is is also useful for fillets (pronounced "fill it") when making the inside of corners using a tongue depressor. It is resin mixed with fumed silica to reduce sag and sometimes loose fiberglass is added for tensile strength. There is chopped glass (loose strands) and milled fiber (looks like balled-up newspaper for hamsters). I used both, depending on the situation and my mood. The chopped glass produces a much stronger result - a few small pinches to 2 oz of resin is a good start. You'll find that if you add too much chopped glass, it will become "hairy" and not spread well - that's the trade off. To give it a more uniform consistency (prevent it from being runny), add a decent amount of fumed silica. Be sure to mix the resin and MEKP thoroughly before ever adding fillers to it or you may have a curing issue. If something doesn't cure in the proper amount of time (fillers increase working time), there's a 99.9% chance it never will - it will just have to be removed and then redone.
FIBERGLASS LAYERS & LAYUP: I made patterns out of waxed paper and transferred them to the raw fiberglass. Always alternate strand mat and woven cloth. Start with thinner mat (1.5oz) that is the entire size of the hole & slope area, then a slightly smaller area of cloth (6-8oz). I tried to make the smaller pieces 1/8" smaller around the edges. then a layer of thick mat (I occasionally used the 3oz stuff every few layers to speed up the process). Alternate mat (1.5oz) with the cloth (6-8oz) as it builds up level with the old gelcoat. Multiple thin layers of fiberglass are stronger compared to thicker layers that move the job along quicker. Always try to finish with mat as the top layer, otherwise there's a slight risk that the cloth's weave may show through the gelcoat. Cut all the layers out first and check that they fit as planned before starting to wet any of them out. Even if a new batch of resin needs to be made before all the layers are done, there will be time to get the next layer on before the last layer cures - this will give it the strongest bond. If it cures, you will need to score the surface with 60-80grit again before adding more layers. There are calculators online (Fibreglast.com has a good one) to ballpark how many layers of raw fiberglass and amount of resin you will need for each patch. This varies on technique and I soon figured out which way to skew the sums for my methods (a little extra resin).
I originally tried to use a finned roller to spread the resin. It only caused grief as it pushed around my perfectly laid glass (I do keep one around during layup). I found that pre-wetting the glass on a board wrapped in foil, placing the glass into position, and then stippling/dabbing the glass with the tip of the bristles worked great to get it into corners, over edges, and remove bubbles quickly. Tape off the gelcoat surrounding the repair area to minimize cleanup. Any spilled, non-cured resin can be wiped-up easily with acetone. Otherwise, use a razor blade to later chip it off.
ROUGH FAIRING: This step is just like automobile bodywork. I basically made some filler putty with the resin, silica, and microballoons. This step is best done as the last layer of glass starts to kick to increase the bond. If it cures, it will need to be scuffed with coarse grit prior to application. In a lot of cases, it may just be easier to let it cure and maybe knock the new FRP down level before moving onto fairing. The fumed silica is a thickening agent and the microballoons are tiny glass beads that make the resin easier to sand. Again, add these agents after you have thoroughly mixed in the MEKP. The microballoons will absorb some of the catalyst heat, giving you a little more working time before it kicks (starts to congeal). I found that keeping it a consistency just shy of Jello pudding allowed me to spread it easier. It will take a lot of microballoons to get there. If it's too thick/dry (like actual peanut butter), it will ball up behind the plastic spreader. I usually started with a thicker-consistency "rough coat" of filler to get the basic shape tuned-in, sanded this out, and then 1-3 more thinner-consistency coats to help it flow into smaller spots and fair all my surfaces. Here are pics of some rough coats:
MAINTAINING CURING TEMPS: Poly and vinyl resins produce better (read: more resilient) bonds if they are allowed to cure overnight at a temperature above 70*F. I did most of my repairs in February and although I am in Florida , it was in the upper-50s to low-60s those nights. The solution I came up with was to use a couple space heaters set up to keep the area around 75*F. I placed the heaters inside the bay, being careful not to have them blowing directly on repairs, covered the engine bay with a giant piece of cardboard, then the entire boat with the mooring cover. I also placed a shop lamp under the keel to allow radiant heat to keep the hull warmer while the engine bay cured.
MORE FAIRING & PREPPING FOR GELCOAT: Paint is at least 90% prep and it holds true to gelcoat as well. Prep the area to be coated with 220grit for spraying, or 150grit for brushing (rolling is also a fantastic option I used on my deck repair). I used several styles of sanding blocks. The ones I liked the most were the Shopsmith velcro sanding blocks I found at Lowe's. They change out paper quickly and come in different sizes, which helped a lot. The replacement paper isn't cheap, but it is certainly convenient and seems to last longer than conventional papers. The object is to smooth the surface of course, but also very slightly dish out the repair area to allow for the proper thickness of gelcoat. The desired thickness of the gelcoat is between 0.010 and 0.020 inches. Too thin and the gelcoat may crack and/or the repair will be visible and more susceptible to UV damage; too thick and the gelcoat can crack over the years as it shrinks.
Note: Exterior gelcoat remains tacky for increased bond to the next coat, while interior comes pre-mixed with a surfacing agent to eliminate tackiness. I prefer to add an agent to exterior gelcoats for most repairs.
HVLP vs Standard/Conventional guns: from what I've read, High Volume/Low Pressure guns are generally more efficient but usually require more CFM's from a compressor. An HVLP produces less overspray and puts more materiel onto the desired surface. Standard/Conventional can use less CFM, but spray at a higher PSI - this will produce finer atomization for a smoother finish, but can lead to increased overspray. I used a Husky 30gal 175PSI compressor that is rated for 5.1 SCFM at 90PSI and 6.8 SCFM at 40PSI. The Husky gun only requires 4 CFM at 40PSI. The Vaper specs state it needs 7-9 CFM, but I had no issues with it on my compressor for this job.
CFM vs SCFM or Cubic Feet per Minute vs Standard Cubic Feet per Minute: CFM is a calculation of the volume of air and how fast it is delivered. SCFM is a standardized CFM rating at 68*F, sea level, and 0% humidity. If you go higher in temp, altitude, or humidity, the compressor will produce less CFM. So doing any work outdoors is generally going to require a higher CFM compressor to get the same result at the tool. It doesn't mean a high-CFM tool won't work because I used the Vaper gun with no problems.
Gravity Feed: Any gun where the paint is supplied from above the tool is a Gravity Feed, as apposed to a Suction/Pressure Feed that can be supplied from below. Seems the simpler Gravity Feed is the preferred option (likely for sheer simplicity), though the benefit of a Suction/Pressure Feed is that it can work at any angle. I did not encounter any real issues spraying horizontal surfaces with my gravity fed guns - I was able to keep the gelcoat going into the gun.
THE HULL REPAIR: The hull scraped across sunken concrete beams that were left near the channel years after a bridge repair. It was lowest tide and I skimmed across them while going about 45mph and making way for an oncoming vessel. The speed had me on plane with minimal draft, so the damage could've been worse. It was about a 15' scrape maybe 3" wide with four large areas of damage where I believe the "stringers" were resisting flexion. These areas had to be enlarged with a grinder until I found undamaged FRP and then added my bonding slope. Yamaha seems to use some sort of grey resin for the final layer of the mold process on the outer shell. This grey material is then sandwiched between the outer FRP/gelcoat we can see and an inner layer of FRP that could be seen when I replaced areas of the engine bay. From below the boat, I ground the old FRP just beyond this grey layer and used the inner FRP layer as the backing for my new FRP. Layup was no different than the engine bay - since it was a simple surface, it went very quickly by comparison. The gelcoat was brushed on.
BRUSHING GELCOAT: In my opinion, either way (spray or brush) is a trade off. I learned that with a decent quality brush and a technique of feathering the brush outward with very quick strokes, I could get good results. The benefit of brushing was eliminating the overspray that would be impossible to contain and the tedious spray gun cleaning. The downside is that it requires a build-up, sanding, then additional brushing to get a level surface due to the "inconsistent nature" of brushing, followed by hours of sanding. Because of repeating these steps to get a near-perfect surface, it took much longer to do.
Note: I later found that rolling gelcoat is very practical and can produced excellent results with less sanding.
Once the gelcoat's skin has been removed, start with 220grit, then 320, 400, 600, 800, 1200, and finish with 1500. It is not necessary to use all the grits, but this gives a good idea of what to expect if a great finish is desired - lots of sanding. Don't apply a lot of pressure with the homemade fairing boards - let the paper do the work. Since the brush can lay down some really thick peaks, an Ingersoll-Rand 4151 6" Orbital Sander (or similar) can be used to knock these spots down before continuing with the fairing boards. With an orbital or DA, try 220grit - if that's not cutting, try 180grit or a few quick swipes with some 80grit. Be extremely careful with this type of tool because it can get away from you very quickly and put you back in the gelcoat phase. Using the easy-to-control fairing board and going slow and steady will be tedious, but the control it allows pays off in the end. For large, flat surfaces/planes a rigid 4' aluminum ruler on hand as a straight edge can help find the high spots.
Doing tighter areas requires something smaller than the fairing board. For those circumstances, I also used the 3M Stikit Plastic Sanding Blocks (they are red) because they have a felt pad and quickly feed out new paper. Once I got into the finer grits, I would switch to a 3M Stikit Soft Hand Pad (3M 05530) because it provides a lighter-touch cut - these are also helpful for shaping corners or irregular surfaces at medium grits.
TERMINOLOGY AND ADDITIONAL NOTES:
CrankyGypsy (established 2001)