Building a High-Performance Composite Kayak

Hi everyone, I'm new to the forum, but after searching through hundreds of older posts and finding answers to half the questions I had, I figured you guys seemed like the people to go to with a variety of questions I have regarding a new kayak project....

So, I'm in the process of building a 16' sea kayak that's meant to be nimble and light (ideally, under 40 lbs). The kayak design has been completed and CAD drawings are in the works. Once these CAD drawings are finished, the hull plug will be machined with a CNC router specifically designed to cut surfboard foam. I expect little trouble in the plug and subsequent female mold construction process, but have some questions about the process after this point.

I would like to vacuum infuse the boat (using West 107 epoxy, likely with 206 hardener) with an eye towards removing as much unnecessary weight as possible while maintaining stiffness and durability. To this end, I've been considering various lamination schedules but have yet to settle on anything. Short of testing each possibility, I'm not sure how to go about calculating the lamination needs, so I've pretty much guessed as to a lay up that should be relatively strong, stiff, and light.

I'm thinking I will start with a surface coat of epoxy mixed with graphite powder to provide some UV and abrasion resistance. I'm not exactly sure as to how I'll apply this considering the vacuum infusion. One possibility would be to coat the mold with the epoxy/graphite and allow this to gel before continuing with vacuum infusion under the normal process (i.e. lay up the various layers on the gelled epoxy). The other possibility would be to apply the surface coat after the boat is pulled from the mold, but I'd like to avoid this if at all possible so as to limit the number of hours spend sanding. The final possibility would be to allow the coat to dry significantly in the mold, thus providing a tack-free surface on which to lay the lamination, but I'm concerned about the bonding between the surface coat and subsequent layers if I do this. Are any of these methods preferable, or should I scrap this idea completely and go with a different surface coat? Ideally, I'd like the carbon cloth to be visible through the surface coat.

After the surface coat, I'd lay down a light layer of carbon cloth. I'm not exactly sure which cloth I should chose for this purpose, but I'm thinking either a 2x2 twill or two layers of unidirectional laid down either running at 0/90 degrees (i.e. bow to stern and port to starboard) or 45/45 (i.e. starboard bow to port stern, port bow to starboard stern). Can you recommend a particular cloth style/weight/or method for laying down the carbon? The goal is obviously to maximize stiffness in both directions.

Following the carbon, I'd like to layup a layer of Kevlar so as to provide some puncture resistance. Once again, I'm unsure what variety of cloth would be best for this application.

After the Kevlar, I would likely lay down some sort of core that would double as my infusion media. I'm drawn to Soric, but was wondering if this is the right choice (over infusionmat, divinycell, etc.) for core/infusion media. Once again, the goal is to minimize weight and maximize stiffness. My plan is to lay a core down in the entire boat, but I'm thinking it may be better in terms of weight to limit core use to structural reinforcements. If I do this, where would be the best places to reinforce? I'm thinking the entire cockpit area should receive a core and that a longitudinal stringer of core material should be run down the middle to add extra stiffness. I'm also thinking small strips of coring up the sheer would work well as stiffening ribs. I'm completely lost as to whether or not I'm going totally overkill in my construction or not. If I chose to use the coring in select areas, how do I promote good resin flow over those areas that do not have a core? Do I also add a flow medium over the entire hull after laying down the peel ply?
Would it be lighter to instead use a thin layer of fiberglass as the center of the sandwich?
Also, would it be better to selectively use carbon tape and no core to help stiffen the boat?

After the core material, I'd lay down one more layer of carbon (to take full advantage of the stiffening provided by separating the skins) or possibly Kevlar. Should I prefer one over the other for this inside skin? Lastly, I'd consider putting a very very thin fiberglass (S or S2) cloth on the inside to provide for easy sanding/abrasion resistance. Is this recommended and if so what style/weight glass should I use?

For the deck, I'm thinking one layer of Carbon on the outside followed by a layer of Kevlar and some reinforcements along the center, either in the form of a core strip or some carbon or kevlar tape down the middle. In the area immediately in front of the cockpit and around hatches, I'd add a layer of Soric followed by another very thin sheet of Kevlar to prevent buckling or oilcanning under the loads of a rescue.

I cannot thank you guys enough for your help! Hopefully after this project i'll be able to contribute some valuable know-how around here.

 

Whatever you do it is worth the effort to set up a trial process, write down all you do and then test the result. Make adjustments if necessary.

I think your process is incredibly complex for what you want. A 16ft kayak is not going to be particularly fast. The length is already a serious compromise on performance potential.

You will find you could build a much simpler hard chine hull that would probably perform better.

I used to go to similar effort you are going to until I deterimned hard chine hulls generally perform better than rounded chine and are much, much easier to build one-offs. A different matter if you want to go into production.

Hard chine hulls can be built from flat panels. The great feature of flat panels is the consistency that can be achieved with finish and strength in the layup.

My latest hull uses 200gsm biaxial CF cloth either side of 3mm Klegecell foam. It has some internal polystyrene bulkheads. It is designed to be fast for my power output. Hence it is a sit on but could be a sit in with a bit more width. Would be slower but could be made stable without the outriggers. The main hull weighs 11kg. The entire boat with the whole drive system weighs 21kg.
http://www.boatdesign.net/gallery/showphoto.php/photo/15138/size/big/cat//ppuser/18624

From all the things I have tried it is hard to beat flat-pack for the ease of building one-off designs. You can make quite complex multi-chine hulls but no real point as a single chine is likely to be the best performer for a slender hull.

Rick W

 

Hey Rick, I definitely do plan on doing many many tests before infusing an entire boat, but I'd like to have an idea to use as a starting point and also for purposes of calculating expected material costs.

As to your comment regarding hard chines... the boat we've got designed actually has pronounced chines and probably would not be difficult to build by hand, but we've decided to go with the CNC shaping because of the amount of time it will save us and the peace of mind we'll gain knowing that no mistakes were made in the making of the initial plug. Besides, we've found a guy who's willing to CNC a plug out of foam for only 50% more than we would pay in raw materials if building the plug.

It's good to know that I'm probably overbuilding, that's exactly what I want to hear! Maybe a layup of (from outside to inside) biaxial CF/Kevlar/Soric/biaxial CF would be preferrable? I guess I could always add a light fiberglass cloth in the cockpit area to protect the CF.

 

Also forgot to mention that I'm not looking for top end speed. Like you send, I can't expect top end speed out of a 16' hull. What I'm looking for is fast acceleration and a kayak that takes little effort to get cruising at a respectable speed.

 

There are some good photos here of a flatpack boat construction including layup of the panels:
http://picasaweb.google.com/adventuresofgreg/ExpeditionBoatBuilding#

I did my hull shown in the earlier post in about 40 hours effort over 4 weekends.

I think infusing will be a challenge to get right first go if you have not done something similar before.

There are some good threads here on infusing. You could lose a lot of expensive material if it goes wrong.

You can get fair results vacuum bagging a single layup over a male plug. You shrink polyethylene plastic over the plug and it comes off OK. Can be reused if you want.

Rick W

 

Luckily, I'm certain I've got the infusion thing down, that's not the variable I'm worried about. I have quite a bit of experience with wet layups, vacuum bagging, and infusion and am confident I can pull the infusion off. What I have never done with regards to infusion is used a flow medium inside the laminate, but I imagine after a few test samples I'll have that down.

I'm also dead-set on the female mold simply because of the resulting time saved in preparing the surface of the hull. This is the same reason I'd like to avoid adding the surface coat post-pulling.

I'll definitely take a look at those photos, looks like a pretty awesome project.
Thanks again for your help!

 

I build skin-on-frame kayaks out of wood with nylon fabric and polyurethane sealant. they weigh in at 16 to 20 lbs and they are fast to build. The next I build I want to actually engineer the wood frame for an optimized structure to get it down to about 12 lbs or so finshed weight.

So why would you got to all that trouble and expence to make a 40 lb molded kayak? Most Stitch-and-glue and cedar strip built kayaks are in the 30-35 lb range, for a lot less material costs than you are proposing. For people that build kayaks, 40 lbs is only considered light if it is a factory production kayak, most custom built kayaks are under 40lbs already. Buy a kit and have at it. Redfish kayaks of Port Townsend WA makes nice light strip build kits, or complete kayaks, all well under 40 lbs.

The problem with all hard shell kayaks is the failure mode is buckling, which means you have to have a real tough and heavy skin (to prevent buckling). The reason the skin-on-frame is so light is the frame members are free to flex without buckling, reducing stress and absorbing impacts, so the frame members can be much lighter.

Also, I do not think mixing kevlar and CF is a wise idea, they may eventually delaminate. Kevlar has very different thermal properties than CF, they behave different with temp change (expansion/contraction ratio) and they may eventually laminate. The years ago when I worked at Boeing, they have gotten away from kevlar panels because of the dlamination problems on mixed lay-ups, they are all carbon fiber and/or fiberglass.

To build stiffness to prevent buckling you need to separate the inner and outer layers of the skin. A thick but light weight core material is what you need to make it light, honeycomb core or very low density foam perhaps. And the CF is somewhat fragile, it will not take repeated impacts without getting damaged, a layer of fiberglass over the bottom should help that (but also adds weight). Also, the skin material does not have to be as stiff at the ends as compared to the center because of the smaller dia. The center section has the max loads and also has a large hole in the middle (the cockpit), so if you make the skin heavier around the coaming area, and directly below the cockpit, and make the bow and stern skins lighter it would save a lot of weight (and material costs too). The area of max stress, and where composite kayaks always start to fail first, is just behind the cock pit right at the bulk head.

Also, what do you know about kayak hull design? If you have experience with various hull shapes and know a lot about sea kayak hull behavior you will be alright. But even so, I would find some way to test paddle your hull design before you go to the trouble of making the plug and hull, espcailly if you intend to manufacture these for sale. Perhaps you can buy a small cheap used sea kayak and glue on foam to carve it to your shape, or build a skin on frame version, cover with foam, carve it out, and put on a layer of glass. I have built 8 or 9 kayaks, I have paddles perhaps 100 different kinds of kayaks and canoes, and know a fair bit about hull behavior, and I do not think I would make a plug and build a hull (even out of fiberglass) unless I got to test paddle something pretty close to what I want to build first. BTW the Mariner kayaks web site has a free kayak design program for download. Gives drag and stability calculations too.

Good luck.

 

Hey Petros, I am in total agreement that SOF and many of the other home-building techniques will yield boats that in many ways surpass composite boats, but we will likely be selling these boatsso those building methods are out of the question. As to the design... I know little about hull design, but luckily I'm not the designer. The boat is designed by a well-known designer (think Wilderness Systems) and has been extensively prototyped already. The design will definitely not be an issue.

As to the problem of buckling, I too am concerned about this. I figure the Soric and Kevlar layer should give some separation between the carbon skins to significantly increase stiffness. I'm hesitant to use honeycomb because of the problems inherent in using a real honeycomb in an infusion and ending up with cells full of epoxy. That's why I figured the soric was better, though i'm not certain it'll really provide much stiffness.

As to what you said about failure near the rear bulkhead, do you mean the bulkhead itself fails or the hull and deck around the bulkhead fail? I'll make sure to keep an eye out for that.

I'm not too worried about delamination between the kevlar and carbon since I'm using epoxy and plan on post-curing the boat to increase crosslinking. Also, since the layup will be done in one shot in the vacuum bag, I think it should be very consolidated. The amount of the heat the kayak will see won't be too significant so I don't think the difference in thermal expansion would be that problematic. Kevlar/Carbon laminates are super-common in Kayaks and I haven't heard of significant delamination issues yet.

 

I compared the rigidity of my CF on 3mm Klegecell with a glass layup I had previously used. The linked image shows a 400mm x 75mm sample of 200gsm CF cloth on the core. Deflection with 5kg load was 15mm. By comparison 320gsm glass deflected 35mm under the same load. Picture here:
http://www.boatdesign.net/forums/at...46-carbon-fibre-fiber-investment-img_4836.jpg

My hulls typically have quite close support with maximum width of panel about 300mm but they are flat so rely entirely on the panel stiffness. The CF layup I am using is quite suitable for this. With some surface curvature there will be some inherent stiffness in the shape so a plus for what you are doing.

Anyhow if you want to compare your layup with what works you can do a similar test.

Most of my hull structures are fully enclosed so they have good load paths for gross loading due to longitudinal bending and twisting. You need to work out how to accommodate these forces around your cockpit.

With these thin layups they usually fail on the compression side at around half the tensile strength due to buckling. It is surprising how much strength is gained by fairing the compression surface as the extra thickness increases the stability of the core. You can get something like 30% increase in the strength of the panel by just adding more epoxy on the compression side. It is also important to ensure the CF fibres are not kinked in the layup as these will reduce the strength. This is why I like the flat panels because all this is easy to control. You can also take test pieces from the offcuts.

Rick W

 

It is not the bulk head that usually fails, it is the skin around the bulk head just aft of the coaming. The combination of flexing and the stress concentration in this area is usually too much.

In one story I read some guys were surfing several miles off the coast of Australia and one of their hard shells broke in half right at the bulk head after going over a wave. He had to abandon his kayak out at sea. The guy had to ride back to shore holding on to his mates waist while laying on the aft deck of the kayak. Since reading about that I have inspected a number of hard used kayaks in that area, and sure enough signs of cracking and laminating was showing right in this area.

 

Indeed skin buckling is the worst part of the layup. Factors that improve things:

-foam with higher compression strength (supports skin to keep it in column)
-thicker laminate
-stiffer laminate

Using the materials you suggested, I would test the following laminates:

-CF200 woven (you stated you wanted a nice looking boat with CF visible...)
-CK170 woven (I do second to problems using aramide, but pucture loads can be a pig, and aramid at least keeps things together)
-3mm PVC foam 80 kg/m3 or even higher (100 kg/m3) or preferably balsa 100 kg/m3 in 3mm. (harder to get, thin and selected lightweight stuff)
-CF200 woven

Alternatively you could use Soric, but remember it eats 600 gr/m2/mm so by using 2mm thick material, you lose 1200 gr/m2 in the layup. Resin consumption of foam is less.

You could minimise stuff by skipping the CK layer, and you might beef up the outer laminate a but by using 280 gr/m2 carbon. Also, UD carbon or non-crimp-fabrics eat up less resin. But they do not look that nice.

I would not use a gelcoat. Too heavy. Either spray or roll the boat in a UV blocking PU paint, or apply the PU paint in the mould. (I have seen it being done). Then please apply it with a roller, and make a nice orange peel finish. When infusing, you will see the strange effect that you will have much less pinholes on a textured surface.

 

I'm building a 20-foot tri and the center hull is canoe like the lay-up as designed by my NA is basically 3/8 divicell foam with 2 layers of 200 gram 90 degree unidirectional carbon cloth laid longitudinally. The stress areas have 4 layers of Carbon cloth and 1/8 holes drilled through the foam filled with bog 1 inch on center. The outside will be covered with 4 oz E-glass for scuff resistance. West System epoxy with a slight amount of 407 powder mixed with the epoxy for a thin bog for the first layer of cloth against the foam, subsequent layers have clear epoxy all vacuum bagged out. The test I’ve done shows the 407 adds little weight but makes the panels much stiffer

 

There you are - I am very intrigued with this method. I have asked some questions on another thread that you have answered here, except 2

Those strips you are assembling, were they cut out of the Divyncell, or did you put a layer of cloth on the foam sheey before cutting the strips up ?

Whats the best thing to glue them along the edge of the strips ?

Very interesting idea.