Thursday, February 5, 2009

Building the Amas

This is the first entry for 2009 and it corresponds with another feat in my building process. Yesterday, I popped another ama (float) half out of the form frames, so I now have two port halves.

It is now time for me to rotate each of the 11 stations 180 degrees to build the starboard halves. Once each starboard half is ready, I can take one of the port sides and join the two. I can't wait for this moment because it will be the first time I get a feel for the size of the amas (volume, flotation, weight, etc).

For other Farrier builders out there: I have been making a somewhat high density putty for the bulkhead fillets. In this putty, I am mixing epoxy resin with milled glass fibers (40%), colloidal silica (40%)and microspheres (20%). I originally used chopped glass fibers (1/4"), but I would advise against the chopped glass because it is very difficult (if not impossible) to get them to a smooth texture.

Other hints I've discovered:

1) The Fein multimaster tool does an incredible job of trimming the keel foam to the same height as the keel batten. I don't think there is a better tool out there for this job.

2) Use packing tape instead of mold release wax whenever possible. Wax on a fiberglass surface is just about the worst contamination imaginable and it is very difficult to remove thoroughly.

3) If you extend the foam at the bow past the bow bulkhead (at least to the length of the bow template), you will not have to fair this foam when the two halves are joined. You just have to fill the gap with pieces of foam, which in my opinion, is much easier.

Finally the pictures:

This is an image of the stiffeners underneath the ama deck. Ian's plans call for 3/4" thick stiffeners, but mine are 1" thick because I am using 1/2" foam, rather than 3/8". Thicker is stiffer in this case.

My vacuum bagging setup: One pump drives a manifold with 10 different ports. As the director of METC says "it could suck the chrome off a trailer hitch" :)

The second port half under vacuum bag.

Bulkheads filleted and taped - bow view

Stern view of the ama with raked transom. You can also see that the access hole in the aft beam bulkhead has been dug out and putty filled to a depth of 1/2".

The first float half I built is displayed in the top-right portion of this picture. It is resting on the air conditioning ducts.

Sunday, November 23, 2008

Glassing the Inner Skin

Oh boy, this was a fun job. Seriously!

On Monday, we completed a hand lamination of the inner skin, complete with reinforcements around the beam bulkheads and the bagged the entire half of the ama. I am quite pleased with the result and feel like having a party to celebrate. Mai tais anyone?

For all time intensive laminations, I am using Proset 125/229 epoxy, which is produced by Gougeon Brothers (makers of WEST epoxy). This resin/hardener combination has a very low viscosity (450 cps mixed) and makes wetting out the fiberglass easier. The greatest part about this epoxy is its incredibly long pot life (77 min at 72F) and working time (thin film) of 2-3 hours. It might sound strange, but I also find the smell of the 125 resin quite pleasant.

Anyway, in order to vacuum bag the entire hull half, I needed to create a flange to bed the mastic sealant tape and run the excess fiberglass, peel ply, release and breather onto. I left the keel foam proud about 2" above the keel batten and pre-glassed one layer of the deck flange 2" wider than need be before the bagging operation. In addition, I added excess foam at the bow and transom to define the bagging flange there. The entire outline for the mastic sealing tape was defined by a putty mixture to seal the foam and deck flange, in addition to providing a solid bonding surface for the mastic tape.

I cut all the dry fabric (fiberglass, peel ply, release material and breather) to shape before laminating and I left the fiberglass on the foam to be wet out. I did not pre-wet the foam, or, in Ian's building terms, I used the "Dry Method" to laminate the fiberglass. This worked out well - the epoxy has low enough viscosity to wet out the glass and the foam (I checked by lifting up the wet fiberglass) and the vacuum bag pushes the glass tight to the foam.

The results of vacuum bagging are terrific. In my opinion, there is just no way to attain vacuum bagging results with an open or hand lamination process.

Time to Make the Bulkheads

So it's been quite a while since my last post for two main reasons: one is that I had to wait for the fiberglass, vacuum bagging and filler shipment from Fiberglass Supply to come across the Pacific Ocean and two is that I have been working hard and not writing much.

Anyway, the bulkheads are now complete. I made them out of 12mm Corecell foam with single skins of 12oz BD (0/90) fiberglass on the transom, center and bow bulkheads and double skins on both the forward and aft beam bulkheads.

I did the bagging on a waxed sheet of 3/4" Melamine, which has a beautifully finished surface. The laminate schedule for each bulkhead was (from Melamine up) peel ply, skin (or two) of fiberglass, perforated Corecell (with holes 3 to 4 inches apart), skin (or two) of fiberglass, peel ply, release material (with holes 3 to 4 inches apart), bleeder/breather material and the vacuum bag plastic itself. I would recommend a vacuum pressure between 15 to 20 in Hg for these flat panels.

Some suggestions for other builders:

  1. Make sure none of the material (other than the first layer of peel ply) comes in contact with the waxed surface as contamination will certainly cause delamination.

  2. Wet out the bottom skin(s) of fiberglass on top of the foam (which is resting on the peel ply) and then flip them over onto the peel ply on the table. Then put the top skin(s) on the foam and wet them out. It is quite helpful to coat the foam with resin prior to applying the fiberglass.

  3. Cut the bulkheads to shape _after_ laminating them. It is a lot more work to deal with trimming excess fiberglass from pre-cut bulkheads than to just do the initial cut from a flat laminated panel. In short: glass and bag rectangular panels, then cut to shape with the full size patterns. The aft beam bulkhead is particularly nasty if it is pre-cut.

  4. Happy bagging!

Monday, October 27, 2008

CNC Routers, Vacuum Bag Test and a Surf Canoe

This past week was a hodgepodge of work. The F22 building process is somewhat on hold because I'm waiting for materials to be shipped from the mainland. Hawai'i is a wonderful place unless you need to find unusual materials. Biaxial fibergass is one of these. I'm currently waiting for a large shipment of fiberglass, fillers, vacuum bagging material and laminating tools from Fiberglass Supply in Washington state. It will probably be another two weeks before it arrives. Bummer.

Until then, I have been keeping busy with several mini-projects. The first one is a test of pulling a vacuum bag on the foam surface. When I was laying the foam strips down for the core of this boat, I extended the strips proud of the keel join line and further forward and aft of the ama's overall lines. On these extensions, I covered the foam with a putty mixture of epoxy, microspheres and colloidal silica. To test the vacuum that could be achieved, I re-used a piece of bagging material from another project, recycled breather material and a single vacuum port.

The picture below illustrates this better than I can in words, but the net effect is that I was able to obtain a vacuum pressure of 12 in Hg (roughly 6psi). This is not ideal (I'd prefer around 15-20 in Hg), but it will do. I think the leaks are due to the fact that 1) air is being drawn from the deck flange, especially at the bow and 2) I am reusing bagging material with holes that have been patched with masking tape (the blue tape). I'm hoping that the actual bag will be a bit better.

Also, the Marine Education Training Center now has a CNC router, which takes CAD drawings and converts them into computer code (G-code) to cut out patterns. We have a three-axis router and I hope to use it to cut out bulkheads, form frames, the deck and foils (daggerboard + rudder). It is so cool to watch the router process, so I have attached a video below. The sharp pitched squeaking noise is from the spindle RPM being too low. We increased the RPMs from 12,000 to 18,000 during the cut to eliminate them.

Finally, I spent a large part of the week filling and fairing a 4-man outrigger surf canoe, which I built from April to August and now just needs to be painted. I built this boat because it has a similar construction to the F22, foam core and knitted double bias (+-45/45) fiberglass with epoxy resin. It is reinforced with carbon fiber stringers for the hull and around the waes in case the canoe will be sailed (to anchor traveler, etc).

Sunday, October 19, 2008

Gluing the Foam Together

This week, I did not have much time to work on the F22, as I was busy fairing a 4-man surf canoe to be primed and painted in the upcoming weeks. In addition, I also got to play with METC's newest toy, a CNC router! (I'll post pictures soon).

However, I did get to glue the foam pieces together this week with a putty consisting of epoxy, microspheres (easy to fair) and colloidal silica (for thixotropy to cling on the vertical surface).

In the pictures below, you'll notice that the foam extends proud of the keel join line a couple inches in addition to extending past the first form frame at the bow. I am doing this because I want to create an auxiliary flange to support vacuum bagging on the part itself. I really don't know if this is going to work because it depends on several variables: 1) Is the 12mm closed cell foam airtight? 2) Are the putty joints between the foam pieces airtight? 3) Will the foam want to torque after the bag is applied?

I'm going to try to pull a "test bag" on the foam to see if I can get a good seal. I'm looking to get between 12-20 in Hg in pressure.

Anyway, here is the foam bonded together (before fairing).

Monday, October 13, 2008


Thermoforming is the process of heating a piece of foam to a point at which the normally rigid material becomes flaccid and conformable. It is a necessary process in building the F22 because the flat sheets of foam will not bend to fit the hull curves without breaking. In order to get the foam to bend, it needs to be heated between 200 and 220 degrees Fahrenheit (93-104C). The least expensive way to accomplish this is by using a hand held heat gun (think turbo-powered hair dryer) to bring the foam up to this temparature.

Thermoforming a 20+ foot ama in an 85F loft with a heat gun is a hot and grueling task. In addition, the heat gun creates localized heat where one spot in the foam is hot and conformable, while other spots are still cool and rigid. Heat guns will scorch and burn the foam if held in a location too long or the gun is too close to the material. In addition, due to the curved shape of the ama, the heat from the gun is redirected back from the foam to the person operating the gun. It is a hot and sweaty job, so much so that I drink 64oz of water before the day begins and I never have to use the restroom all day.

In the never ending quest to build a better mousetrap, we assembled a makeshift oven to heat the foam up to thermoforming temperature without holding the heat gun or battling the heat. We used a large foam insulated case, with plywood wedges partially opening the case and holes cut out for the heat guns (pictures below).

After heating the foam to its thermoforming temperature, the builder has about 10 seconds to get the foam from the oven onto the mold surface and conform it to shape. Using the thermoform oven was such a blessing to accomplish this task. Below are pictures of the thermoforming oven, the foam bent to the curve of the hull, the join lines of the foam covered in packing tape and the fully planked hull (4 hours total)

Thursday, October 9, 2008

And Planking Begins...

Today was the first day of planking the ama with foam. To get to this point, we first needed to lay battens over the form frames to define the shape of the hull. Rather than butt battens together, which I believe creates a hard, unfair spot, I scarfed three lengths of lumber together with a 12:1 scope. I used clear Douglas Fir, 1"x4"x8' (2) and 1"x4"x"10', with the 10' section in the middle and the 8' sections on the end.

I glued the scarfed wood with WEST system epoxy mixed with 403 Micro Fibers and 406 Colloidal Silica. After the glue set (2 days), I sanded the excess off and trimmed each 24' length of lumber into 4, 3/4" battens. All in all, I had 3, 24' lengths of lumber, which produced 12 battens. I only used 10 on the ama and you can see from the pictures below how the battens are oriented.

That was the fun part. I'll get to the thermoforming in the next post. What a grueling job.