Last week I worked on the legs. This time it's on to other parts.
To make the blanks for the posts, arm supports and stretchers, I used the same basic technique as for the leg blanks. That is, split the board to find how the grain runs. Then cut parallel to the split, wide enough for the needed workpiece. Finally, look at the edge grain where the board was split and find where the workpiece would fit along the staightest grain.
Using a template to copy a side stretcher shape on a blank whose grain was slanted |
Notice in the above pic that the outline of the workpiece is drawn catty-cornered to the piece of wood. I had to do this to follow the grain. That was really limiting - if I didn't care about wasting a lot of wood (and if I had more wood), I'd split out a blank large enough to make it easier to get straight-grained pieces.
I made life easier in the shaping stage by sawing off a lot of the waste. You can see a chunk that was sawn off in the above pic.
With the posts, there was a bit of a wrinkle. The board didn't split straight.
Not what I was hoping for |
Used a small hatchet and scrub plane to get them to rough dimension, but they're still curved |
I thought for a few days about what to do. In the end I decided to just live with it and allow the posts to bend backwards in the chair. I think the tops of the posts will be about 1-1.5" further back from where they should be. I might adjust the angles at which they are mortised into the seat, and I think I can do that. But curved workpieces present more complications - I use conical tenons in a conical mortise, and the tool I use to make the conical tenons works best when the pieces are straight. I normally turn the workpiece into the cutter and if it's curved, I won't be able to get a consistent, symmetric, conical tenon. This time I'll have to turn the cutter around the workpiece or maybe just cut the tenons freehand.
I soaked the workpieces in water for 3-5 days to soften the wood before shaping. This worked great - what a joy it is to use the drawknife with wet wood. When shaping, I'd draw the shape of the workpiece on the tangental faces so that I could shave to those lines on the radial faces first. It's much easier cutting away material from the radial surfaces than the tangental.
Here, I've already shaved this radial face, then re-drew the shape before I rotated the part to shave the tangental face. |
Shaving the tangental face, angling at a corner to get down to the line (I'm leaning to my right so I can see the lines) |
After both sides were shaved to the lines, remove the midsection |
I took the parts first to square cross-section, then octagonal |
Here's the two posts completed with the pile of shavings |
For making the parts octagonal, at first I drew lines to guide my drawknife work. But I found that I could get close enough without marking. This chair is a bit rustic, so the parts don't have to be perfectly octagonal.
When I posted last week about making the legs, I mentioned that the tube used to soak the parts could only fit one leg at a time. Realizing that was going to be horribly inefficient, I got something a bit larger. I really wanted to find something about 6" diameter, but had to settle for a 4" PVC pipe. And this one is about 5 feet tall, whereas my first one was 3" diameter and about 2 feet tall.
The original 3" diameter pipe could fit one leg at a time for 3-5 days! This pic was from last week. One part soaking, three others waiting. |
The new setup is much more efficient, but ... |
... I have to add labels to remember which parts started soaking when! |
After shaping, there are two challenges. The first is drying the soaked and shaped parts. In a comment on last week's blog, Sylvain reminded me about a Follansbee post where he said he sometimes uses his car dashboard (presumably parked in the sun) as a drying "kiln". I did that this week. I don't have a scale to see when parts no longer lose weight, but they sure seem dryer after a few days in the hot car.
The second challenge is tougher: bending the crest rail. I don't have a steambox and am not sure how to tackle this one. The crest rail will be about 3/4" octagonal, so I don't think soaking it alone will be enough. I need heat. I've researched this a bit and found some creative solutions, but most involve things I don't have. And I really like accomplishing things like this with what I've got on hand. We'll see ...
Your stool looks like a Paul Sellers stool (leg shaping). I made one a few month ago. I should rework the seating board which I couldn't shape nicely. Have you a post about it?
ReplyDeleteGood eye, Sylvain. It is a Sellers design, but I must have made it earlier than 2016, before I started witing this blog.
DeleteI have not done it, but I've read that you can steam parts in an oven with a roasting pan. Guess you're limited by pan and oven size.
ReplyDeleteI found a post about it at blog.lostartpress called "the $2 steambox".
I've thought many times about that post from The Schwarz. I don't have a roasting pan big enough, but I think if I put the wood in the oven diagonally it might fit. I'll experiment with it in the next week. The part (and another experimental part) is soaking now and I'm thinking of wrapping it in wet cloth, then aluminum foil and baking it for a couple hours at about 215 degrees F.
DeleteOne Guy adventures in steam bending:
ReplyDeletehttp://www.stochasticgeometry.ie/2016/10/23/for-science/
http://www.stochasticgeometry.ie/2016/11/27/steaming-cracks/
http://www.stochasticgeometry.ie/2016/12/02/platform-and-salvage/
http://www.stochasticgeometry.ie/2016/12/03/properly-bent/
http://www.stochasticgeometry.ie/2016/12/12/depressing-weekend-results/
Thanks for the links. I've tried to find similar plastic tube material, but on eBay I can only find huge rolls of the stuff. I'm also going to try the bend without a strap. My bend is not as tight as Mark did in those blog posts - hoping I can get away with it. I think the key will be if I can shave the part so that long wood fibers are intact on the tension side of the bend.
Delete