Roubo crochet design

[Steve Voigt]

I'm building a traveling bench that will have no vises (to keep down weight and speed assembly time), so I've decided to use a crochet to hold work for edge planing. In researching this a little bit, I ran into an interesting conundrum.

The first I heard of the crochet was in Chris Schwarz's 2005 PW article, which I think was the precursor to his first book, and which sparked the craze for Roubo benches. Chris's crochet looked like this:

Screen Shot 2015-10-06 at 2.39.55 PM.jpg
Notice that the crochet makes a very acute angle with the side of the bench, maybe 20°-25°. From what I can tell from searching Google images, just about everybody who built a crochet after that copied Chris's design. But interestingly, it's not what Roubo shows. Here's Roubo's diagram:

Screen Shot 2015-10-06 at 2.26.34 PM.png
Notice that the angle is much larger; I would guess 60°-70°. Here's another where the angle is almost 90° (a little hard to judge accurately from this view):

Screen Shot 2015-10-06 at 2.54.59 PM.jpg
The only modern example that I've seen that looks like this is from Mike Siemsen:

Screen Shot 2015-10-06 at 2.59.59 PM.jpg
So, I'm interested in hearing from anybody who's built a crochet. What angle did you use? How did it work?

Just want to be clear, I'm NOT trying to knock Chris Schwarz's design--nobody else was writing about this at the time, it was new territory for him, and I doubt he was focused on picayune little details like this. I just want to get some feedback on the ideal design, because I'd rather not have to make the thing twice.

[Pat Barry]

What is the thickest piece you can see edge jointing with this method? From an engineering point of view a simple angle of 60 to 75 degrees would be fine as a simple stop. The curvature detail does though help when you shove the board in to keep it compressed against the edge of the bench until you fix the other end.

[Warren Mickley]

Rob Tarule made a Roubo bench 30 years ago or more. His bench has a real single plank top (heart up) and a crochet. It was featured in a Fine Woodworking article in 1987 and in the Scott Landis Workbench Book of 1998. Tarule's crochet looks a little more like Roubo and Siemsen, not like Schwarz. Almost like a quarter circle. Tarule's crochet has a longer section that is flat against the bench than either Siemsen or Schwarz, again more like Roubo. I think the crochet is supposed to keep the the board from going forward, but not so jammed in that it does not need support at the front end.

Rob has experience using the crochet; if you call him up he might enjoy answering questions. A preview of the Landis Workbench book is available on line so you can look at the Tarule bench.

Moxon's bench has a crochet with a wooden screw going through the cheek. It might keep the board tight against the bench, but I don't think you could plane without support underneath.

[Warren Mickley]

Here is Tarule's bench with the crochet in use.
tarule crochet.jpg

[Steve Voigt]

Warren, I have Landis's book and I looked at the chapter on Tarule's bench last night, I just forgot to mention it in my post. Which was crappy of me, because Tarule has been pretty much forgotten in the Roubo craze of the last 10 years. So thanks for mentioning him. I'll go back and have another look, and I will try contacting Rob to see what he says.

Pat--I think that you need support beneath the work at two points, the leg and the deadman. So the function of the crochet, as Warren says, is more to pull the work tight against the bench. You're right though that it seems like a simple engineering problem; the question is what is the maximum angle (e.g. 70°, 80°) that will serve this wedging function? Because it seems to me that with the much more severe angle that Chris uses, the force on the fasteners that hold the crochet in place would be much greater.

P.S. Rob Tarule's website, in case anyone is interested.

[Mike Siemsen]

I have made them several different ways and I have simplified them to pretty much a straight line. The two I checked in the shop were 25 and 30 degrees. I could see going to less of an angle but the 30 degree angle seems to work pretty well for me. it keeps the board from tipping and holds it tight against the bench and gives me a range from a bit over 2 inches on down. There isn't enough load on the crochet to cause me to worry about the lag screws pulling out and I make them from soft wood about 1 1/2 inch wide.
It is important that the end of the board rest over the peg in the leg. With the angle of the crochet too far past the peg and the board rears up when you go off the end, too close and only the narrow boards will rest on the peg wider ones missing the peg and falling past. You need to figure the range of widths you will be planing. If you are planing 4 inch wide stock you would need the angle closer to 90 degrees to accommodate the narrower widths as well and still work over the peg. The nice thing is it is an easy part to change to suit your needs.

[Steve Voigt]

Hey Mike, thanks for the feedback, I appreciate it. I got lots of good ideas from your knockdown Nicholson video as well.

[steven c newman]

I tried to keep mine a bit on the simple side..
IMAG0081.jpg
Not sure how weel this shows up..
smooth plane.jpg
Bench isn't all that big, either. Angle between the two parts is about 25 degrees or so. Would have to go measure it, IF you want.

[Pat Barry]

Warren, I have Landis's book and I looked at the chapter on Tarule's bench last night, I just forgot to mention it in my post. Which was crappy of me, because Tarule has been pretty much forgotten in the Roubo craze of the last 10 years. So thanks for mentioning him. I'll go back and have another look, and I will try contacting Rob to see what he says.

Pat--I think that you need support beneath the work at two points, the leg and the deadman. So the function of the crochet, as Warren says, is more to pull the work tight against the bench. You're right though that it seems like a simple engineering problem; the question is what is the maximum angle (e.g. 70°, 80°) that will serve this wedging function? Because it seems to me that with the much more severe angle that Chris uses, the force on the fasteners that hold the crochet in place would be much greater.

P.S. Rob Tarule's website, in case anyone is interested.

To your point about the support from below - yes, you are correct on this - oversight on my part. The crochet doesn't do as well with the task of keeping your boards in position to resist the planing forces from above as it would in conjunction with a lower support of some type.

With regard to the curvature, there may be some benefit as compared to a straight angle. Lets say you have a relatively thin board to edge joint. It would not take much force to hold it up against the bench edge you are using as a reference for straight. A thicker board however would take significantly more lateral force to hold a board flat up against the edge of the bench. The force goes up as a function of the thickness although I don't recall if this is a linear function or not. Regardless the curvature serves a second purpose in that it reduces the overall length of the crochet as well as providing more force for thicker boards. Think of the interface of the curve and your square board end and the vector direction the force is applied. For a thicker board the vector force wold be higher toward the bench edge as a benefit of the curved feature.

I drew the following sketch to illustrate the curved crochet and the approximate force angle being applied to a board that is forced into the crochet. As you can see the angle becomes steeper (more directed toward the bench) for the thicker boards where the increased force might be beneficial. The force vector is perpendicular to the tangent line for the point in the curve that is contacted by the corner of the board

Crochet.JPG

[Graham Haydon]

Interesting Steve, when you look at the plates is does appear to be just a planing stop with very little lead in for gripping the board.

[Steve Voigt]

With regard to the curvature, there may be some benefit as compared to a straight angle. Lets say you have a relatively thin board to edge joint. It would not take much force to hold it up against the bench edge you are using as a reference for straight. A thicker board however would take significantly more lateral force to hold a board flat up against the edge of the bench. The force goes up as a function of the thickness although I don't recall if this is a linear function or not. Regardless the curvature serves a second purpose in that it reduces the overall length of the crochet as well as providing more force for thicker boards. Think of the interface of the curve and your square board end and the vector direction the force is applied. For a thicker board the vector force wold be higher toward the bench edge as a benefit of the curved feature.

I drew the following sketch to illustrate the curved crochet and the approximate force angle being applied to a board that is forced into the crochet. As you can see the angle becomes steeper (more directed toward the bench) for the thicker boards where the increased force might be beneficial. The force vector is perpendicular to the tangent line for the point in the curve that is contacted by the corner of the board

Pat, thanks for the diagram. I'm usually skeptical about trying to solve woodworking problems like this--the solutions usually crumble when they meet hard reality--but in this case the situation is pretty simple and your explanation seems sound.

Just curious, are the forces proportional to the length of the vector lines you've drawn? From what I remember of high school physics, that was the whole point of vectors, but I could be way wrong.

[Pat Barry]

To be precise, the 'vectors' I have drawn were meant to show the composite force directions. Each vector therefore is typically thought of as being composed of two vectors. One of these is the force vector pushing back toward your vise, for example as you clamp in the board, the other is the perpendicular force vector that is pushing the board inward toward the bench. So, If one of the 'vectors' that I drew, for example the one that would be for a thick board, the component of force is higher in pushing toward the bench than pushing toward the vise. So, the more perpendicular the line I indicated is, the more of the total force is in the perpendicular direction. These force vector magnitudes are proportional to the total force being applied by the vise.

Crochet.JPG

[Steve Voigt]

Right, if I'm once again remembering the HS physics correctly, you make a right triangle, with one of your "composite vectors" as the hypotenuse, and the relative lengths of the legs tells you the breakdown of the component forces. Or more succinctly, multiply the hypotenuse by either sin or cos of the appropriate angle. It's all coming back to me…thanks for the interesting discussion.

(and if anyone is still reading at this point: yes, we left actual woodworking relevance behind a while ago, but it's still interesting.)

[Ron Bontz]

Thanks for posting. Ron