Carpenters/woodworkers - reverse engineer a chair.

[lefty]

I have two wooden twig chairs approx 100 years old that I want to copy and recreate. I don't want to take them apart but need to determine what the angles and curves are of each piece. I am a neophyte with this kind of thing so how to proceed? As simple as a protractor and a pencil?

lefty

 

[Thomas]

I hesitate to post, because I fear making chairs more than anything else due to their angled joinery. That said I think a sliding bevel would be a good tool to have for measuring angles - in particular the back pitch tends to make or break a chair's comfort. Also consider that the back - in some cases - has a gentle curve to it, so you'd be looking at two different angles between the back-seat and the back-top rail. Good luck.

 

[venessian]

Are these like bentwood chairs, etc?

Depending on the complexity of the originals, you can produce templates for each section:
a) transfer the more easily measured straight lines (life-size) and angles to sheets of large paper or matte board by using rulers, adjustable angles (can be two straight-edges taped together to mimic the angle) or real adjustable angles, based on registration points marked on the original and on the sheets of paper/board

b) for curved areas you can either trace sections onto sheets of cardboard or matte board or transfer radii by using adjustable curves (the ones we architects used to use before ACAD), either way using the same registration points

c) for tight or difficult to access sections, make templates by hot-gluing, taping, stapling, whatever thin strips of cardboard or scrap materials, the ends of which follow the lines of the curves (like a fan template)

I agree with what Thomas wrote.

 

[Thomas]

^^ good points. Also, I didn't consider this until just now, but I'd keep a fair supply of poplar and plywood on hand for template / jig making. If I were to do this - and to be honest I've wanted to make a pair of Morris Chairs for years and years now - I'd probably do a mock-up out of poplar first to make sure the angles are just right, and then use that as your basis for construction. For a Morris Chair it's much less of an issue because they feature flat, adjustable backs.

 

[lefty]

Thanks, fellows.

There are three curves that will take some finessing. The rear uprights, the arms which curve around the entire chair and the back rest which is a solid piece of curved bark and has two angles as mentioned.

If I can put together a reasonable template and facsimile I'd eventually like to cast one in bronze.

lefty

 

[venessian]



If I can put together a reasonable template and facsimile I'd eventually like to cast one in bronze.

lefty

Interesting. In that case, what about the possibility of making 2-piece rubber or fiberglass molds directly from the chair? Do you have a photo of the chair?

 

[lefty]

Not sure if that's doable. I asked a props guy to look into casting a mold for me, but there may be too many small pieces for a bronze version unless I simplify it. I did find this so maybe it is possible.

The chair.



I think I'll start with building a few out of wood and go from there. Now to figure out what kind of wood they used.

lefty

 

[Thomas]



(...)The chair.



I think I'll start with building a few out of wood and go from there. Now to figure out what kind of wood they used.

lefty

oh, that's a horse of a different color. From the looks of it (and my design/production guy is out of town, otherwise I'd ask him) the top rail is bent rattan, which isn't an awful undertaking. The back and seat look like rattan/cane weave, which means you're making a box frame for the seat and looping the woven back through. Much less joinery than I was thinking. It appears your only mortise-and-tenon joinery is at the main box (front/arms/back) and even the stretchers are tied/nailed on. Much greater margin for error than I had imagined.

 

[lefty]

Rattan? No. The chair was made in Northern Ontario by natives. The seat back and seat are bark, not sure about the frame. I think I can replicate it easy enough and will be heading up to ON in a few weeks to cut some wood, but I need to determine the best material to use. I have made willow furniture before though I don't think this is willow. I'll see if I can take some better pics.

You have a design/production guy?

lefty

 

[Thomas]



Rattan? No. The chair was made in Northern Ontario by natives. The seat back and seat are bark, not sure about the frame. I think I can replicate it easy enough and will be heading up to ON in a few weeks to cut some wood, but I need to determine the best material to use. I have made willow furniture before though I don't think this is willow. I'll see if I can take some better pics.

You have a design/production guy?

lefty

Well, I say rattan because it's the first thing that comes to mind that's easily bendable (relatively speaking). Could be any number of products, but (generally speaking) the harder the wood you use, the more work it is to get it bent without splintering. Well...I say that, but green woods are also a bit more accommodating depending on the species. My first guess, though, is willow - if you get some long straight branches you're good to go. Actually, willow bark isn't out of the question for the weave, and now that I think a little bit more, I recall an article about a guy who made a chair from a single tree: it was a willow tree. Let me check my garage for the issue (iirc, Fine Woodworking or maybe Wood). Our design/production guy...he's more of a go-between for the designers and factories. So he knows design, and he knows some construction, but the details in getting from design to product aren't his forte - he trusts the factory to deal with that. But other thoughts come to mind, I can ping him and see what shakes out.

 

[venessian]

The photo really helps lefty. Not what I imagined at all.

I agree that, as Thomas says, since this chair is essentially composed of pretty simple individual pieces with very little complex joinery, molding the entire chair would be counter-intuitive and not very efficient.

Better to compile the dims, etc. of all the individual pieces (this will be pretty simple, since most elements are straight), fabricate each piece and then assemble the mock-up. The mock-up pieces can then serve as the mother pieces for the casting mold (again, all individual pieces). It looks quite doable.

I'm not familiar with Ontario, but wouldn't another possibility be that the wood is birch, cedar or spruce, probably steamed originally?

Info like this might help you, as the construction technology would appear to be quite similar.

Interesting project.

 

[Thomas]

okay, showed the picture to my design/production guy and he thought it could be elm, willow, hickory, or maybe birch. He's leaning towards steam-bent hickory, though, and has a little experience with it. Hickory is pretty bendy and it sounds like it suits this application pretty well. Good luck with it.

BTW - I checked my garage cabinets for the willow-tree issue and came up empty. Years of Fine Woodworking, British GQ, and other magazines that I keep around for no reason, even old Playboy calendars, but no Willow Tree article. dammit