Years ago, I stumbled into an old storefront in rural Virginia with some friends and encountered one of the thousands of rocking chairs that have been inspired by Sam Maloof. It was a dramatic example of the form with a huge sweeping back and long rockers. But most amazing was the seat itself.
It was a massive chunk of walnut that had been deeply scooped out – perhaps a 2"-deep saddle. And the pommel of the seat was tall – almost 2" high.
One of my friends said, "Do you know how come that's a lady's chair?" The rest of us shrugged.
"Because that seat 'lifts and separates.'"
Underwear jokes aside, the guy was right. Sitting in the chair felt like I was being prepped for a medical exam that few people enjoy. The seat looked gorgeous, but you don't sit on a seat with your eyeballs. So, you have to think hard about the human rump (and other body parts) when designing a chair.
Chair design is a topic that can fill an entire book. We don't have the space for that here, so I'm going to write about how I design the all-wood chairs that I build. Some of my guidelines are at odds with modern rules for chairs, but that's because my chairs aren't entirely modern. I take many cues from ancient chairs.
The seat of a Sam Maloof chair. The saddling is deep, but Maloof left some room for the legs to move. People who copy his work tend to make the saddling deep and without much room to move. Ouch.
Start with the Saddle
One of the most important principles in chair comfort is that "sitting" and "sitting still" are not the same thing. And we rarely sit still.
"One of the major difficulties in the design of seating is that sitting is…viewed as a static activity while, in actuality, it is a rather dynamic one."
— "Human Dimension & Interior Space," by Julius Panero and Martin Zelnik.
This is the problem with deeply saddled seats. We sit in them and they feel amazing at first – they support and cradle the bottom in a pleasant way. The only problem is that we can't often sit still. Here's why. In a typical chair, the sitter's weight is confined to about four square inches of buttocks. The pressure on that small area requires us to shift our weight, even just a little, to remain comfortable.
But a deeply saddled seat doesn't allow us to move much, if at all. So, these sorts of seats become agonizing in short order.
I have yet to see an ancient chair that is sculpted as dramatically as our modern Jell-O moulds with legs. I'm sure they are out there, but they've never been the dominant form. Instead, many old chairs had shallow saddling (maybe 1/4" to 1/2") or even no saddle whatsoever. A shallow saddle gives you some curve but also allows you to reposition yourself with ease. (Oh, and they are easier to make.)
I also suspect that many all-wood chairs would be draped with an animal skin or a small cushion. I've put sheepskins on all my chairs and can attest that even the minor cushioning they provide makes a world of difference in the department of butt comfort. (You'll see this cushion concept again when I sneak it into a discussion of seat height.)
For some reason, some modern chairmakers are masochists and seek to make a chair as comfortable as a La-Z-Boy recliner via the magic of curvy valleys. This strict attitude reminds me of people who insist that a single scrap of sandpaper in a shop is an abomination. Lighten up, Francis, and go fetch a cushion.
How far is the front of the chair from the floor? The typical modern chair height is 18" – that's almost inviolate. Sorry to say, I think that is too high to be a general rule.
Tall seats are punishing for shorter sitters. If their feet cannot rest flat on the floor, the front edge of the seat will constrict blood flow in the thighs and produce agony.
Slightly shorter seats, however, are just fine for tall sitters. Their feet can still sit on the floor and their thighs hover above the seat – allowing blood flow. The only downside to a tall person sitting in a shorter chair is the short chair is a little more difficult to dismount.
(Side note: This is true for a table's height as well. Standard table height is 30". A high table is a pain for a shorter people. But a slightly shorter table [29" or even 28"] is no problem for a tall sitter.)
So, if 18" is too high as an overall rule, what should the height be? The answer is not cut and dried with a custom chair. Here are the questions I ask to calculate the seat height:
1. What is the sitter's "popliteal height?" Some people call this "stool height." It's the distance from the bottom of the foot to the bottom of the thigh of a seated person. It ranges from 14" to 19-3/8" in the general population.
2. What sort of footwear will the sitter use? Work boots, 3"-high heels and moccasins all can change the equation.
3. What is the chair to be used for? If it's for dining or keyboarding, it should be a little higher so it is easy to mount and dismount. If it is for relaxing, it should be lower. How low? Seats can be as low as 12"-13" for lounging. Low seats allow you to stretch your legs – a luxury. Low chairs are harder to get out of – but that's the point.
4. Will there be a cushion or other seat cover? Cushions can add 2" or more to the seat height, so you should subtract that when making the chair's frame.
You might be wondering how to determine the seat height for the general populace instead of for a particular person. When I need to do that, I typically use 16-3/4" or 17" for a dining/working chair. And 15" to 16" for a lounging chair. These are on the low side, but they aren't radically low. Tall people will hardly notice. Short people definitely will, and they'll be grateful.
The seat of this Welsh chair is less than 13" deep. Yet it sits just as well as a chair with a 16"-deep seat. That was a revelation for me.
A typical seat depth for one of my chairs is 16". Once you get deeper than 17", you risk cutting off the sitter's blood flow behind the knees. Surprisingly, shallow seats work well. I have made seats as shallow as 12" and they sit just fine (unless you have an epic backside or the seat is too high. Having both is a disastrous combination). A shallow and low seat also prevents the blood in your thighs from being constricted.
In general, I don't mess with the seat depth too much. If it's between 14" and 16" I know it will work in most cases. This slight flexibility allows me to build using narrower boards. If I have to glue up my seat from two 7"-wide boards, I'll do that and call the 14"-deep seat done. I won't glue on an additional 2"-wide strip of wood to get to the magic 16" depth.
We're not done with the seat quite yet. But to understand the last bit of seat data, we need to first understand the chair's armrest (sometimes called the "armbow").
Biometric data suggests the top of the armrest should be about 7" to 10" from the seat, depending on the sitter. I usually shoot for 8" to 9" (or less). People have asked for 10" – this height makes some people shrug their shoulders, and you can feel it in the neck after a while. My rule of thumb is 8" for shorter people and 9" for taller ones.
Backrest & Seat Tilt
The small of the back – sometimes called the "lumbar" region – is where I do a lot of work to make a chair comfortable. If you build a chair that supports the lumbar spine, you will make friends – as well as chairs. The lumbar is about 7" to 9" above the seat. This is why I keep my armbows at 8" as much as I can and add a "doubler" above it (and sometimes below it) to increase the thickness of the armbow so I can support the lumbar.
Chairs that lack lumbar support are fatiguing to me. I squirm to push my lower back against the chair's back, but my shoulders and buttocks prevent it. I guess this is why we have low pillows. Nothing wrong with a pillow.
One of the oft-overlooked aspects of chair design – the seat's tilt – can help the lumbar region get to its destination, which is the armbow in my chairs or a lower slat in ladderback chairs.
Most chairs tilt a little toward the back. A seat that is flat to the floor can feel like you are being thrust forward and out of the seat. Adding some additional tilt can encourage the sitter to slide backward and put their lumbar directly on the armbow.
But how much tilt? I like Welsh chairmaker John Brown's method of using his fingers and a spirit level. Put a level on the seat's pommel so it runs from the front to the back of the chair. Raise the spirit level at the rear of the seat until it indicates it is level. If you can get one finger (plus a little more) under the level at the back, that's a chair that's for dining or other proper things – keyboarding etc. Two fingers and you have a chair that is good for lounging. Three fingers – alcohol consumption. Use that information to cut the legs down to get the tilt you want.
Here is the Irish Gibson chair I finished up in May. The back is angled at 25°. Surprisingly, it sits like a fairly normal side chair.
For me this is where the rules get blurry and surprisingly flexible. Most modern chairs have the chair's back tilted back about 5° to 7° or so. That's fine. But adding a couple degrees can also encourage the body to touch the chair's armbow and doubler.
So, I tend to tilt the back about 9° backward, but I will tilt it a little more at times. And I will continue to play with different angles as I think there are some discoveries to be made. Earlier this year I built a copy of an Irish chair that had its back tilted at 25°. That's about three times as much as normal. As I built the chair I imagined that sitting it would be like visiting the dentist. I was wrong.
It wasn't as different as expected. This chair – called a Gibson, hedge or famine chair – was historically used as a kitchen table chair. Even though it looked more like a chair for sunbathing. Your eyes and expectations can deceive you. So don't believe them, this book or your vicar. Work it out yourself.
The high headrest on this chair is comfortable but looks disproportionate to the chair's other elements.
Lowering the back by a few inches greatly improves the look of the chair without sacrificing comfort.
How high should the crest (aka headrest or comb) be? That depends. For customers who want to be able to pass out in their chairs I want to have a high comb to cradle the head during unconsciousness (after reading a good novel, of course). So, I measure to the base of their skull. I think these chairs look too tall, but some customers like the anthropomorphic appearance of these chairs.
What I prefer in a chair is to have a crest that supports the shoulder blades without the crest digging into them. This is about 22" above the seat.
But this location can be tricky depending on how beefy the sitter is. Broad-shouldered sitters can feel the ends of a significantly curved crest rail push into their shoulders. Thinner sitters of the same height cannot.
The easy way out is to simply raise the crest rail a couple inches, however this dramatically affects the way the chair looks. I prefer a compact chair. Another way out is to make the crest rail so it doesn't have as pronounced a curve. My crest rail typically has a 10" radius. That's tight, but it keeps me from over-bending my chairs' spindles. The crest can be curved less, such as a 14" or 16" radius (which is OK). Or it can be flat (which fixes one problem but causes another by reducing support for the curve of the shoulder blades).
Here's the good news on the crest rail: You can do it at the end of the construction process and experiment with different crest rails – different curves and stick spacing. Dry-fit a prototype crest rail and see how it feels to your back. Make some changes and see how they feel.
When all else fails with your chair's design, add a sheepskin. Or pillow.
When I talk about chair comfort with other chairmakers, it's inevitable that someone will say: Ah heck, just put a cushion on it and call it done.
Me, I like sheepskins – a traditional Welsh chair covering. They don't add much bulk to the seat, and they won't make the seat too high. But they do add some cushioning and warmth. (And they give me an excuse to go to IKEA without a disguise.)
So do your best to make your chairs comfortable. But know that it's never a bad idea to become a man (or a woman) of the cloth.
Christopher Schwarz is the editor at Lost Art Press and one of the founders of Crucible Tool. He works from a restored 1896 German barroom in Covington, Ky. You can see his furniture at christophermschwarz.com.