Building the floating wall: the engineering.

Turns out, building a floating wall is a little more involved than framing it, standing it upright, and bolting it to a floor joist. Before I get into that, first we did in fact have to frame the thing. It went pretty quickly.

Based on my little sketch, we built it from 2x8s. At first I used nails. I ended up going back to all the butt joints and reinforcing them with 3-1/2" #10 screws.

Once it was done, K and I stood it up. That's when we realized, well, building a floating wall is a little more involved than framing it, standing it upright, and bolting it to a floor joist. 

First of all, it rocked side to side pretty significantly. That and it seemed a little top-heavy. So we carefully set it back on the floor, then I took to the interwebs while K took to a DIY Slack channel at her work to get more information.

Since it's not attached to anything but the floor, a floating wall takes a bit of engineering.  I discovered there's not a lot of information about them. Possibly because they take a bit of engineering.

Our first problem to solve: the lateral swaying. Think of that as along the x-axis.

The options to mitigate it included either sheething the whole frame with plywood and then screwing sheetrock to that, or blocking the crap out of it. Given the exorbitant cost of wood these days, two five-dollar 2x4s sounded a lot more appealing than four forty-dollar sheets of half-inch ply. So we went with 2x4s.

When we stood it up a second time, the thing was R-O-C-K (in the USA) solid. Okay, solved that.

Next, there was the back-and-forth movement along the y-axis.

I designed the wall to be 80-inches long by 88-inches high. At close to eight feet high, the thing has a lot of potential to move back and forth. My original idea of bolting it to the floor joist with hefty 1/2" x 7" lag bolts was laugh-out-loud funny. Weighing in at about three hundred pounds and standing eight feet tall, it made for an impressively large lever. It wouldn't take much to get the wall moving and just rip those bolts right out of the joist. Along with a bunch of our floor.

We had to be smarter than the wall.

So we turned to railing designs. Think of posts on exterior deck railings. They're bolted to the joists beneath the flooring of the deck. Somehow, we'd have to do something similar. But how? We really didn't want to rip open the floor. 

One day last week during lunch we hatched a plan. It involved cutting just-big-enough holes on either end of where the wall would go to get 2x4 posts attached to the joist. Which also meant big enough to get my impact driver into the spaces on both sides of the joist, battery and all.

Time to take a step back. 

At this point, all we had was a framed wall lying on our bedroom floor. Yes, I had ripped up one piece of the wood floor to locate the joist we'd be using, but it’s back in place and no one is the wiser. The next step required cutting into our floor. There wouldn't be any going back. Did we still want to attempt this floating wall idea? 

Yep, we did. We were confident we could problem-solve and come up with a solution that worked. So I taped off the cut line and broke out the circular saw.

No going back now. Next I hacked through the subfloor to take a peek.

Ahh, an engineered floor joist. Like an I-beam. 

In order to attach our intended 2x4 posts, I had to salvage some scrap pieces of plywood for blocking. The joists measure 11-7/8" tall. The web is 3/8" thick and the flanges are 2-5/8" wide by 1-1/2" thick. That meant we had a 15/16" gap on either side to fill in order to create a solid block around the posts. Luckily, I had just the right sizes of scraps. I ripped them down and glued them together.

Then shoved them through each of the subfloor holes to glue them to the web. I picked up some more PL400 subfloor glue and Simpson nailing plates for the job.

The blocking was now essentially part of the floor joist. I called in my inspector to sign off on the work.

He, umm, approved. I think. Then went back to playing with Hot Wheels. Now we had to test our 2x4 post plan with some scraps.

The plan was to screw these posts to the blocked floor joist using Simpson hurricane ties. The framed wall would slide in between them and I'd bolt the 8-foot tall posts to the bottom, middle, and top of the wall's end studs. That would effectively transfer the y-axis movement down either end of the wall, into the floor joist, and across the entire subfloor. Once we bolted the ties to the joist using the scraps as markers, it was time for the real deal.

I grabbed the 2x4s and hacked off the spacing for the joist with a combination of a table saw and a miter saw.

In order to get the battery for the impact driver through the hole, I had to cut out a slight shape for it. No worries because the baseboard trim will cover it. With K's help, we positioned the posts. She held them while I finagled the impact driver into place. It was tight quarters, but in the end everything was secured. We used the specified #9 1-1/2" Simpson screws.

After all of that, we had a pair of 8-foot tall posts standing in our bedroom. Also, a pretty good-sized hole in our floor.

Considering their height, they were surprisingly sturdy. As in, no y-axis movement. What x-axis sway they had I knew would disappear as soon as we bolted the wall studs to them. That meant it was time to stand up the wall again and position it between the new posts.

Boom.

Then bolt the posts to the wall and, what the heck, the sole plate to the subfloor and joist. I used 3/8" bolts throughout. For the plate, Grade 8 washers just for fun, alternating every pair of bolts on opposite sides of the web.

We gave the wall a tug. Zero lateral sway, a little y-axis give. But not much. Our wall was solid. I leaned on it. No problem. 

Our conclusion: we're good to go.

Next we have to hang and mud the sheetrock before adding the wood accents to the sides, top, and built-in shelf. All in all, I'm impressed with our engineering ingenuity and the steady progress we've made on this project. Good times.