Building a Structural Box Beam for Paikea’s Hull

There’s a section of hull in the aft cabin that used to lean on a bit of luck. Years ago someone built a partial bulkhead in there for an ensuite that no longer exists, and it never actually ran down to the floor. It’s an easy mistake to make in a boatyard — you glance at a bulkhead-shaped piece of joinery, assume it’s doing a bulkhead’s job, and put your stands and chocks against it without checking. We did the same thing ourselves before we understood what we were looking at, and it’s part of why the hull cracked at the bottom in that area before we caught it.

That panel — what we call the paddle, the side shell of the boat — takes a lot of different loads. Wave slap, the pressure of water moving past as the hull works, the boat settling and rising in a seaway, fenders shoving against it on the dock. A hull panel like this is really just there to keep the water out and the boat floating; it’s the surrounding structure — bulkheads, floors, the geometry changes at the hull-to-deck join — that decides how much that panel is allowed to flex under load. Paikea is built with relatively light panels and a fair bit of internal structure doing the work of keeping them stiff, which meant that half-finished partial bulkhead was quietly underperforming for years.

So the job this week was straightforward on paper: replace it with a proper box beam. Something that ties into the frame below the floor and blends up into the deck, restoring the support that panel actually needs.

Why Not Just Run a Longitudinal

The obvious alternative was a longitudinal stringer instead of a ring-frame style box beam, and we did seriously consider it. The problem is that a longitudinal can’t just stop at a bulkhead — when the panel deflects inward under load, an unsupported end wants to lever the frame it’s tied to right out of the boat. You solve that by running the longitudinal straight through, or by tapering a wedge into the panel on the far side to spread the load out gradually. In this case that would have meant carrying the longitudinal the full length of the boat to keep it looking intentional, which is a lot of extra weight and work for a fix that’s really only needed locally. A partial ring frame does the same structural job in this one spot, for less material and less weight — even if it costs us a small aesthetic compromise with a vertical protrusion instead of a horizontal one where the cabin steps down.

Choosing the Foam

The core for this box is a 10mm rigid PVC foam — but the recycled kind. Manufacturers end up with a pile of offcuts and second-quality panels that aren’t worth selling as-is, so they glue the strips and bun-cuts back together into full sheets and sell them on for low-load work like furniture. You can see the joins running through it if you look closely.

This isn’t foam we’d use anywhere structurally critical on the boat. If this were a bulkhead, it’d be the wrong material entirely. But the actual job this foam is doing here is forming a shape for the fiberglass to wrap around — the box itself carries almost no load on its own. The real strength comes from the unidirectional glass running along the hull and up the flange, and the double bias skin carrying shear into the wall. Once you’re clear on what a component actually needs to do structurally, using a cheap, light, recycled core in the right spot isn’t cutting corners — it’s matching the material to the job, the same way you wouldn’t spec the same door for an engine room as you would for a berth. Cost, weight, availability, and whether you can actually work with the material all factor in, and a five-star material that nobody on the job can process properly is worse than a modest one used well.

Putting the Box Together

Getting the shape right took a couple of days — templating it out of cardboard against the curve of the hull, cutting small blocks of the 10mm foam to hold spacing while the main panels were glued up, and gluing the whole thing together into the box profile. Before it went anywhere near the boat, we laminated a light 200g skin over the inside faces. That skin’s job was twofold: a bit of rigidity to handle the box while it was being built and fitted, and — more importantly — to make the whole cavity airtight ahead of infusion.

Airtight really was the operative word. With a hollow section like this, any leak into that cavity during vacuum bagging means the air trapped inside gets drawn out along with everything else, and a fully evacuated cavity under a closed skin has one full atmosphere of pressure sitting on it from the outside. That’s not a small load.

Infusion Day

With the box glued into position and the unidirectional laid down the hull and along the flange, it was time to infuse. Resin flowing well, vacuum holding, everything tracking normally — and then, partway through, the pull on the resin sped up sharply and the whole thing collapsed in on itself.

A week’s worth of templating, cutting, and fitting, gone in the time it takes to notice something’s wrong. The short version of why: somewhere in that recycled foam, at one of the many glued seams, there was a leak small enough that we hadn’t caught it in the setup. Once the vacuum found it, the air inside the cavity had nowhere to go but out — and the panel had nowhere to go but in.

What we do about that is a proper post on its own, because there’s a real diagnosis in it — not just why it failed, but exactly where, and why the fix is stronger than what we started with. That’s next.

In the Members Library

For more composite engineering deep dives like this one, visit our members page.

Read on: Fixing a Vacuum Infusion Implosion: The Repair — how we diagnosed the leak, rebuilt the box, and got it back under vacuum.