Self-tensioning Workbench Clamps

I saw an article where a very innovative woodworker made some clamps for his workshop that worked like a cam cleat on a sail boat. ( ). They are very useful for holding a work piece on the bench for sanding, filing, or other shaping. A piece of wood is cut with a decreasing radius curve in its outer surface. The theory is that if you place two of them on your workbench and put a work piece between them, the force of the wood against the curved surfaces increases the pressure on the sides of the work piece holding it in place. The outer surface of the clamp has a piece of sandpaper to increase the holding power. The clamp is movable, easy to build, and self adjusting. It works great on the modular side of my work bench.

The clamps are unidirectional. They tighten when you push against them, and release when you pull the workpiece back.  So they are effective when you are pushing against the clamps. So you need to orient your work process to the clamps.

There is no magic to these shapes. Ideally it would be a gradually increasing radius curve. Mine is a Bernoulli curve, which is an approximation that I could find a nice clean JPG online. It can be adapted to any size piece of wood, I had a scrap piece of 12 x 12 seven ply spruce laying in the recycle bin. Use what you have and adjust the size of the clamp to fit.

Author's note - A later commenter suggested using string tied around a nail (or small disk) to provide a true increasing radius curve. This is a better solution  and one that in no way infringes on any patents filed y the above mentioned video. To see how tis process works, look at a cam cleat on a sailboat. This adaptation has been around for decades and has never been patented, since it is an obvious example of an applied physics principle. I disavow all knowledge of Bernoulli, or his principles, or the application to fill scuba tanks using his principles.
Remove these adsRemove these ads by Signing Up
shannonlove2 years ago
The best shape for clam clamps is a Fibonacci Spiral aka Nautilus spiral, because the pressure increases smoothly against the workpiece as the cam rotates. This prevents damage to the workpiece and prevents the cam from jamming owing to a sudden increase in pressure caused by an irregular surface.

You can layout a Fibonacci Spiral of arbitrary size using ole' Fib's geometric technique or you can print out a spiral from an image and trace or transfer it to wood. Cut outside the lines and then carefully sand down to a smooth curve. 

Even though they are not as durable, clamps made from a softwood (like white pine) are best because (1) they do not mar other woods and (2) the soft wood deforms slightly and grips the workpiece. When they get lumpy, sand them down or throw them away and make new ones.
Far more work than is needed. For a smooth cam action you could simply cut circles and put the pivot off centre (closer to the edge would give more movement in the cam, closer to the centre less movement).

I do see the advantage to the Fibonacci Spiral, being that it give the maximum variation in work piece width, but it can only be used in one direction.

By using a circular piece you can work the clamp in either direction (could be useful depending on the project and work space available), and you can still get 0 or near 0 clamping by having the circles cammed so when both are facing to the centre they touch (or nearly touch).

The same cutting process could be used, draw the circle, rough cut and sand down, or you could use the centre of a hole saw cut, or pin your router to the surface with a string and use it as a compass.

I would also argue for a rubber face on the clamps to give longer life and better grip.
First, This is an awesome Idea... I really like this, though, you will find, that the plywood clamps you use will dig into softwood and leave marks on your project...

The glue between the layers of plywood are not soft and will not give at all.. but the layers of wood are softer and will press in a bit leaving the glue to damage the piece...this will happen over time. Maybe MDF would be a better option if available.. if not I would use a clamp block to keep the clamps from damaging the piece..

If you make it large enough you could use a ring of rubber cut from a bicycle inner-tube for extra friction.
ilpug2 years ago
You could put some kind of rubber facing on the clamping surface to increase friction.
Palmetto_x2 years ago
Good Idea. I will try that.

Another way to get that arc is by using a nail, string and a pencil. Fasten the string to the nail and wind it up. Connect the pencil and start to unwind it you will get that gradual arc.
vectorges (author)  Palmetto_x2 years ago
ANd if you used a slightly larger disk you could increase the differance. ANother great idea.
abstracted2 years ago
i think the same can be done if you just cut wood discs and drill your axis hole off center?
vectorges (author)  abstracted2 years ago
That sounds far too easy! You are probably right. The object is a varying radius surface. I think a parabola is optimal, but I have been playing with excel for a while trying to make a good shape. Unsuccessful so far.
drloki2 years ago
You could cover the face surface of the cams with an old bike inner tube if you were worried about marring a fine surface
bknow132 years ago
If you are concerned about marring you can use some scrap between the work and the clamps.
Truehart2 years ago
Nicely done, but I have some questions:
1) Do the clamps mar the workpiece?
2) Do they come unclamped while sanding (or other similar action) back and forth from the clamps? I see how they would clamp, but that clamp looks like it only holds tight if the forces are in one direction.
vectorges (author)  Truehart2 years ago
2. The clamps are uni-directional. They tighten as you push against them, but release if you pull the workpiece back. I'll update the ible to make that clear.

1. That being said, the clamps don't seem to mar the workpiece. Ideally they will grip and stay in that position.