4-Leg Stool – Design

A few months back I started working on a design for a 4-leg stool. Coming up with a design that I was excited about was a struggle, partly because a 4-leg stool is fairly basic…what can you do with 4-legs and some stretchers to make it interesting that hasn’t already been done? The reality is there are plenty of possibilities, but it definitely takes time working through different ideas/concepts before you start finding something that you’re excited about. For this 4-leg stool design I also had some specific criteria I wanted to accommodate, so it was coming up with a design that I was satisfied with that fit those criteria that was the big challenge.

The first criteria I had for this stool was that I wanted to keep it in a reasonable price range for the market I’m in. Here in Tokyo that’s in the 40,000-60,000 yen range (roughly $400-600 if the yen and dollar are equal). At that price range I know I have to design a stool that is pretty straightforward to make. For me that meant no overly complicated joinery, and no bent parts. The second criteria was to incorporate a Danish cord woven seat. A woven seat is comfortable, and I love the warm contrast between the paper cord and wood. So with those criteria in mind, the design of the stool became fairly limited quickly. Essentially I ended up with a limited number of parts to play with. At a minimum I knew I needed 4 legs and 4 cross pieces/stretchers for the seat, plus I wanted additional cross pieces for leg support. That’s it for parts. In some ways super simple, but working out an interesting design with those parts was a big challenge.

I started the design process using a stool that a friend designed and made as a jumping off point.

In a way this stool is as simple as it gets. The design is really all in the sizing and location of the parts, plus some subtle detailing in the chamfering of the wood parts. I also know for a fact that this stool is really straightforward to make, since I helped assemble several of these. My wife and I have had a couple of these stools in our apartment now for the past 4 years or so, and we use them daily. My wife sits on one all day long while she works from home. The Danish cord seat has been really comfortable, and though it’s started to wear from daily use, it still has plenty of life left before it needs reweaving. The construction is rock solid with double tenon joints connecting the legs and stretchers, and the location of the lower stretchers being relatively low, add a ton of stability and strength to the legs. All in all I think this stool is about as simple and rock solid as it gets, and it was a big challenge to try and see what I could change about the design without compromising strength or the ease of fabrication.

So using my friend’s stool as a starting point, here’s one of the initial stools I modelled in Blender. The only real change from my friend’s stool at this point was the addition of round legs. I really like the look of round legs with rectangular stretchers intersecting, but that is a joinery detail that really complicates fabrication. So, I pretty quickly abandoned the idea of using round legs, but kept using them in various models just for the sake of exploration.

From there I started to explore different ways to deal with the lower stretchers. The basic design is fantastic for it’s stability and strength. The stretchers are relatively low giving a ton of support to the legs, while also serving as a place to set your foot. But in an effort to explore other possibilities I started to play around with some other options. And one idea I had was to have the lower stretcher connect diagonally to the legs.

This design looks kind of cool, but it’s definitely lacking in strength in few areas. For one the lower end of the legs are only supported from one direction, meaning there’s a lot more room for flexing and movement. Also by having the stretchers criss-cross one another so that they are flush requires using a lap joint which further weakens the stretchers. It would be possible to offset the stretchers to avoid a lap joint, but the overall poor leg support led me to move on from this design.

In another design I explored the idea of incorporating a series of slats into the lower stretchers. I think this design has better potential for a coffee or side table (minus the woven seat), since the slats definitely turn that lower space into a shelf of sorts. Structurally I think this is a pretty solid design, but with the addition of the slats the stool starts to look a bit bulky to my eye.

Below is another attempt to explore a different stretcher arrangement by bringing one set of stretchers down to the very ends of the legs. (For comparison sake the stool on the right is the original concept.) I pondered this design for quite a while. Structurally it would be really solid, especially if you joined the ends of the legs and side stretchers with bridle joints. One sticking point for me with this design was the added work of introducing another type of joinery (bridle joints) which would require an additional machine setup/jigs. But I’m also just not a big fan of the look of this style. It seems less graceful than the stool on the right; more clunky and weighted down. It also reminds me of institutional chairs like you might find at a school or library.

Next I started to explore some designs with the lower stretchers raised up closer to the seat. The downside to raising the stretchers is that the legs don’t have as much support and resistance to racking and twisting, but the upside is that it opens up some new possibilities for how to detail the legs. With this design by raising the stretchers I was able to add some taper to the legs. The taper starts right after the lower stretchers, which keeps the joinery simple by maintaining simple 90 degree connections between the stretchers and legs.

I think the overall look here is starting to get a lot lighter and graceful, but something about it still looks kind of plain to me.

Next, I started to move in the direction of splaying the legs a bit. I started by splaying the legs in one direction only. Meaning if you’re facing the wide side of the stool the legs would appear parallel, but when facing the narrow side, the legs will appear to be splayed out slightly.

And here’s the same stool but with the legs splayed out equally in two directions. (3 degree splay and rake)

Between the two of these, I prefer the design with equal splay and rake. Overall, I think the visual balance is better. One thing that particularly bothers me when the legs are splayed only in one direction is how the lines of two diagonal legs appear to converge towards the foot of each leg. Whereas the same perspective on the design with legs with equal splay and rake is nice and parallel.

I also briefly experimented with some slightly more decorative options for dealing with the stretchers. Although this was a concept that I set aside, I think it’d be interesting to pursue different options along this line in the future.

In the end I kept circling back to a design that was very reminiscent of my friend’s stool with the addition of splayed legs. The splayed legs look nice to me, and to my eye somehow give the impression of both movement and stability. I also really like the idea of keeping the lower stretchers closer to the ground. For one, having the stretchers low makes the construction really strong, but also I like having a foot rest. (Watching my wife sit on her stool at home everyday it’s clear the stretchers are a comfortable place to prop up your feet.)

So that’s where things started to come together for me with this stool design. I made a few more cosmetic changes to the design once I started building the first one, but I’ll share more of those details next time. Ultimately I didn’t stray too far from my friend’s design, which I’m glad he is okay with… I think it’s a really smart and simple design and it’s hard to improve on! The addition of the splayed legs does complicate fabrication, but after spending a lot of time pondering all the possibilites, I decided to just go for it and see how much slower it would actually be to make.

Lastly, here is a line-up of the various models I played around with throughout this process, some of which didn’t make it into this post. I love 3D modelling for the ability to do comparisons like this. It’s super helpful to be able to see different designs side-by-side.

Thanks for following along if you made it this far. Stay tuned for some more on the 4-leg stool build next week.