Winston Chang
05-16-2019, 8:23 PM
I need cauls for a project. I've read Mike Henderson's article (http://www.mikes-woodwork.com/Cauls.htm) about making cauls, as well as Tipton Lum's post (https://sawmillcreek.org/showthread.php?209901-Make-Your-Own-Curved-Clamping-Cauls) on this forum, and I settled on a slightly different technique -- one that I think has some advantages in terms of simplicity and consistency.
The basic idea is that you bend the wood with a certain amount of force, then, while the force is applied, you cut the concave side so that it's straight. Then, when the force is removed, the side you made straight will become convex, and with a shape that applies even pressure when the ends are clamped together.
I started with a 2x3, cut it in half to make two 4-foot long pieces, and jointed the narrow edges (that is, made the edges very straight). There are many different methods you can use to joint the edge: you could use a table saw with a jointing sled, a jointer, a track saw, a router with a straightedge, or hand planes. I used a track saw and cleaned it up with a hand plane.
Next, I stacked the two pieces of wood on top of each other and put some spacers on each end. Then I pressed down in the middle to see how much force it took to get them to contact. I adjusted the thickness of the spacers so that it took a bit more than my body weight (so about 200 pounds total) to get them to touch in the middle. The cauls are 4 feet long, so when they're finished, this means (I think) that the clamping force will be about 50 pounds per foot of length. Hopefully this is enough.
409991
(Note that I used spacers that are really long and stick out past the sides of the wood. The wood will be jointed again later with these spacers in place, and depending on the method you use, you may need to shift them so they're flush with one side, or even use narrow ones that don't stick out at all.)
I used to clamps near the middle to bring the pieces of wood together so that they touch.
409992
Now the outer edges are bowed inward and we want to make those edges straight. It's not possible to do that with the clamps in the way, so I used screws to attach a block of wood to the middle of the two pieces, then removed the clamps. Notice how the block doesn't protrude past the top of the cauls-to-be.
409993
Now it's time to straighten the edges. Again, there are many ways of doing this. You can use a table saw with a jointing sled, a jointer, a track saw, a router with a straightedge, or hand planes. I used hand planes. (Some methods of jointing may require that you don't have the spacers sticking out the sides, so you can either cut them shorter, or use shorter ones to begin with.)
409994
Next, I unscrewed the block, with the wood clamped so that it doesn't suddenly spring back when screws are removed. Done.
409995
You can see that they bow a little bit.
409996
The curve is more visible with the convex sides facing each other.
409997
To test, I put some wood shavings between the cauls and clamped the ends. (If you don't have wood shavings handy, you can use paper instead.) When I pulled on the wood shavings, I wasn't able to get them to move at all, so the clamping pressure appears to be pretty even.
409998
Finally, mark them so you know they are a matched pair, and wax them or put packing tape on them so that glue doesn't stick.
I like the method here because, like Tipton's method, it doesn't require any manual fine tuning of the shape -- the pressure on the wood during construction makes it automatically take a shape that applies even pressure. I think it also is an improvement in that the amount of curvature is determined by the clamping force (in my case, 200 pounds). Woods vary greatly in their stiffness, and the width of the caul can vary, so if you make curve of a particular size, it can result in greatly varying amounts of clamping force in the middle when the ends are brought together. With this method, you decide on the clamping force, then let that force determine the curvature. I used 2x3's, but if I had used 2x4's, I would have ended up with less curvature for the same clamping force.
This also results in a pair of cauls that are matched for clamping force. If you make individual cauls with the same shape, but the pieces of wood happen to have different stiffness, then when clamped together, the line where they meet could be curved. With this method, if you have pieces with different stiffness, the curvature when unloaded may be different, but when clamped, they should meet in a straight line.
Hopefully this is useful. I like that the process is simple, requires no jigs (unless you count the spacers and wood block), and will work with many different jointing methods.
The basic idea is that you bend the wood with a certain amount of force, then, while the force is applied, you cut the concave side so that it's straight. Then, when the force is removed, the side you made straight will become convex, and with a shape that applies even pressure when the ends are clamped together.
I started with a 2x3, cut it in half to make two 4-foot long pieces, and jointed the narrow edges (that is, made the edges very straight). There are many different methods you can use to joint the edge: you could use a table saw with a jointing sled, a jointer, a track saw, a router with a straightedge, or hand planes. I used a track saw and cleaned it up with a hand plane.
Next, I stacked the two pieces of wood on top of each other and put some spacers on each end. Then I pressed down in the middle to see how much force it took to get them to contact. I adjusted the thickness of the spacers so that it took a bit more than my body weight (so about 200 pounds total) to get them to touch in the middle. The cauls are 4 feet long, so when they're finished, this means (I think) that the clamping force will be about 50 pounds per foot of length. Hopefully this is enough.
409991
(Note that I used spacers that are really long and stick out past the sides of the wood. The wood will be jointed again later with these spacers in place, and depending on the method you use, you may need to shift them so they're flush with one side, or even use narrow ones that don't stick out at all.)
I used to clamps near the middle to bring the pieces of wood together so that they touch.
409992
Now the outer edges are bowed inward and we want to make those edges straight. It's not possible to do that with the clamps in the way, so I used screws to attach a block of wood to the middle of the two pieces, then removed the clamps. Notice how the block doesn't protrude past the top of the cauls-to-be.
409993
Now it's time to straighten the edges. Again, there are many ways of doing this. You can use a table saw with a jointing sled, a jointer, a track saw, a router with a straightedge, or hand planes. I used hand planes. (Some methods of jointing may require that you don't have the spacers sticking out the sides, so you can either cut them shorter, or use shorter ones to begin with.)
409994
Next, I unscrewed the block, with the wood clamped so that it doesn't suddenly spring back when screws are removed. Done.
409995
You can see that they bow a little bit.
409996
The curve is more visible with the convex sides facing each other.
409997
To test, I put some wood shavings between the cauls and clamped the ends. (If you don't have wood shavings handy, you can use paper instead.) When I pulled on the wood shavings, I wasn't able to get them to move at all, so the clamping pressure appears to be pretty even.
409998
Finally, mark them so you know they are a matched pair, and wax them or put packing tape on them so that glue doesn't stick.
I like the method here because, like Tipton's method, it doesn't require any manual fine tuning of the shape -- the pressure on the wood during construction makes it automatically take a shape that applies even pressure. I think it also is an improvement in that the amount of curvature is determined by the clamping force (in my case, 200 pounds). Woods vary greatly in their stiffness, and the width of the caul can vary, so if you make curve of a particular size, it can result in greatly varying amounts of clamping force in the middle when the ends are brought together. With this method, you decide on the clamping force, then let that force determine the curvature. I used 2x3's, but if I had used 2x4's, I would have ended up with less curvature for the same clamping force.
This also results in a pair of cauls that are matched for clamping force. If you make individual cauls with the same shape, but the pieces of wood happen to have different stiffness, then when clamped together, the line where they meet could be curved. With this method, if you have pieces with different stiffness, the curvature when unloaded may be different, but when clamped, they should meet in a straight line.
Hopefully this is useful. I like that the process is simple, requires no jigs (unless you count the spacers and wood block), and will work with many different jointing methods.