Friday, January 21, 2011

A Pair of Contemporary Chairs: Layout and Design

Never fear, you didn't miss any steps in the construction of the Mt. Lebanon Side Chair.  My efforts to improve my steam rig have backfired a bit, and caused a slight delay in the project.  After building an upgraded box, with options for expansion, and enlarging my fittings to an inside dimension of 1 1/2", and increasing water capacity by using a 2 gallon gas can, my first run was a bit of a damp squib.  It seems that the 1000 watt hotplate didn't have enough gumption to get the job done.  Maximum temp. in the box: 155 degress farenheit.  So, if Amazon cooperates, the mighty "Bayou Burner" should arrive today with 185,000 btu's of propane muscle that will allow me to power a small locomotive.

Meanwhile. . . I was approached about building a pair of upholstered chairs.  Modeled on the photo below, made of oak, and with a deadline.  So I thought I'd go over the way I lay out a project when I don't have a set of plans.

The first order of business is to say that I don't make direct copies of other people's work unless they put the plans out in the public domain.  I wouldn't, say, make a "Mira" chair and sell it -- not cool.  This piece is a fairly standard contemporary design, by an unknown manufacturer, and I'm using it only for inspiration.

I call this approach "furniture lofting" as it is inspired by my very limited knowledge of boat building.  By drawing the actual piece (templates, angles, and dimensions) I save myself heaps of frustration and define techniques and form a loose cut list.  I start with the most important element, the double curve of the back leg/post.

A couple of things are going on here at the same time.  First, I'm looking at the design -- how dramatic is the curve, what are acceptable chair dimensions, where can I anchor a known dimension to get the proportions right.  And second, I'm trying to nail down the technique.  In this case, will the profile be cut from stock, steam bent, or laminated.  Steam bending would be tough.  The stock in some points is more than 2 inches wide, and unless I felt lucky, I'd need to bend it first and then do the profiling.  Lamination makes a lot of sense, but again, I'd need to form it to the inside curve and then profile it -- running the risk of a bunch of lamination run-out along a very visible edge.  In this case, as the curves are not too dramatic and I have a lot of faith in the strength of oak, I think I will cut the profile from the stock.  The downside is that there will be some concern about short grain, but in this thick stock I believe it to be less of an issue.

Using a pair of splines, and starting with a 1 1/2" block glued to the plywood to mimic the width of the foot, I start to lay out the design.  I align the profile of the leg with the short edge of the plywood as the floor, and the long edge as the vertical axis.  This means that in order to simplify the joinery, I want a flat spot from 10" to 13" up the front face of the leg.  It is at this point that the side rail will meet the leg.  Adding six or so inches in height for the cushion and batting will net me a final seat height of around 19 inches.  I make this spot flat by glueing a 3 inch block.  The rest is by eye.  I first wanted a final height of 35 inches, then felt that 37 looked better.  I glued one more block, fiddled with some clamps, and was happy with the look.

I use a similar method to lay out the shorter front legs, but I raise the flat by 1/2 inch.  That will tilt the chair back slightly, without any compound joinery.

What did I learn.  First, I  have a template that looks good - in actual size.  Second, without even fiddling about, I determine that eight inch wide, 8/4 stock, will accomodate back leg pattern with space for the front leg.  If it were any wider, I'd be kicking in a price multiplier at my supplier.  As it stands, I can get a handle on my stock needs and begin calculating board feet.  For every two legs (approx. 3 feet x 2" x 8/12 wide) I'll need 4 board feet. 

I lay out the seat joinery by drawing an actual-sized overhead view.  Four 2" by 2" block stand in for the legs and I start with my final front and rear dimensions.  Like most seats, it will be slightly wider at the front than at the back.  I settle on 24 inches at the back, and 27 inches at the front.  I also decide on 27 inches for the final length of the combined seat posts and side rails.  Drawing these dimensions on my board, and taking advantage of the right angles that exist, I can quickly place the posts. I draw centerlines on the posts and connect the lines of the trapezoid.  The front and rear rails are easy -- 23 and 20 inches with right angle joinery.  The side rails are different -- but it doesn't matter, I'll be making a template from the drawing, and in this case, making a mock-up before I go live.  I do take one measure with the bevel gauge to get the Master Angle.  This angle, and its supplement, will determine how the side rails meet the posts.

I'm choosing to go with loose tenon joinery on this project, and that means that the net and gross dimensions of all parts are the same.  The bevel will determine the angle I place the posts in the jig later.

And that's where I'll pick up next week!

Thanks for reading.


  1. Bless you for making chairs. When I have enough time to slow down some, I may try a pair. Like you, I would have to give them some serious thought.

    What are you using to draw the subtle curves in the fourth photo? They seem to be too precise to be free handed.

  2. Jeff-
    Those are my 1/8" thick, 5' long poplar splines. I use them to make fair curves and can make infinite adjustments. In the places where I know I want the line to cross I glue down a block and clamp with spring clamps. I just trace these outlines in pencil, go to the bandsaw, cut, then sand with a flexible sanding stick.

  3. You could laminate the curve with tapered laminations or you could also make a non structual curved core and laminate to the finished dimensions to avoid lamination run out. If you are aiming for a non painted finish your plan A sounds the best. Aren't deadlines motivating and exciting!


  5. Pablo - Thanks for the link. The curved core idea is a good one. As this is progressing, I'm feeling fairly comfortable with the stock's ability to hold up -- I've even found some grain that curves in a convenient fashion around the leg.