Sunday, February 8, 2009

Roubu Bench - part 4: Figuring out the Wagon Vise

So, I've got the boards for the top all sorted out, numbered and aligned for grain direction. Now my job is to do any work to them that I want to do before I glue them up. Cutting a groove for the sliding deadman in the bottom of the first board is a whole lot easier in a single board then trying to do it by flipping and manhandling a 140 lb. top. Yes, the top should weight just about exactly 140 lbs. I weighted one of the 14 boards for the top and it came out to exactly 10.0 pounds on my bathroom scale.

The main problem I need to solve is the wagon vise. After looking at all kinds of tail vises, reading what Chris Schwarz and lots of others have written, I decided that I wanted to try and build this rather simple-looking sliding bench dog. That's the approach I'm taking is to look at it like I'm creating a bench dog that slides.

There's not a whole lot out there on wagon vises. Chris's book shows his retro-fit to his workbench and how he did it. I like the look, but I'd like to see if I can make something a little more integral to the bench itself than the slots screwed to the bottom of the bench.

I also had another major limitation. When I went to buy a screw for my wagon vise, all I could find locally was at the Woodcraft they sold a small veneer or cider press vise screw.

I chose this because I didn't think I needed a massive screw, it looked similar to what Chris had in his bench, and it was cheap and readily available. As I began to work through the problem of how to install this in the bench, I did do another search and figured out that there's a real gap in the bench screw market. We need a solid 1/2" diameter bench screw version of the normally massive 1" one you find all over the place. Even screws for shoulder vises were just too beefy. I saw a place in England selling one that seemed just what I wanted made in the Czech Republic, but it didn't seem to be sold on this side of the Atlantic and looked like too much trouble to get. I'd just make do with what I had.

The two primary problems, at this point, to be solved were: how to make the chop and attach it to the vise, and how to attach the chop to the bench so that it stays fairly stable and yet slides easily. You want a fairly low tolerances of slop, but there will always be some. The first question was made even more interesting in that the end pad of the vise screw did not have holes to fasten it to the chop, and not a whole lot of meat to drill the holes through. I was also concerned about just using screws to hold the pad to the chop, especially if I was able to stabilize the rest of the setup enough to be able to use the vise to pull as well as push, such as disassembling parts.

I began by actually trying to solve the second problem first. How would I attached the chop to the bench, and how to let it slide? Chris had installed runners under his bench and had a free floating chop running through the slot cut into the bench top. I wanted mine to be a bit more integral so I started down the road of a tab in a slot.

My first "brilliant idea" was to get a pre-made dovetailed drawer slide from Rockler and use that as the slide for the chop. I thought about mounting it underneath like Chris, or embedding it in the sides with the male part of the dovetail slide (the pin?) embedded in the chop. I ordered one and it was fine for a drawer, but it didn't seem tight enough or beefy enough.

I then thought of Occam's razor and decided that if I was going to have to cut a groove to embed the slide, why not just use a simple groove in the interior face of the slot, and tabs on the sides of the chop. Voile!

I played around with possibly making the tab a half-dovetail shape so that as the chop tried to move upwards with clamping pressure the chop would become wedged tighter in the triangular space above. BTW, I abandoned the idea of a sliding dovetail at this point for the sides as I thought of the wonderful fun I'd have cleaning chips and dust out of the lower parts of the groove. The bottom of the groove has to be flat for easy clean out.

I ended up going with the simplest solution that works, and no simpler: a simple square groove cut into the sides of the slot in the bench, and tabs on the chop riding in that slot. One problem decided. Sort of.

OK, that whole process took a bit longer than it did to describe it. I actually came to the conclusion over the long process of building the chop with the general idea of tab and groove construction. I actually got to the point of marking out the half-dovetail-shaped tabs on the chop when I tried to actually cut the groove in a test piece of wood. It was significantly harder to do than a straight groove without a good stair saw or equivalent, especially as I wanted this to be a stopped groove. And I wasn't going to stop my bench building in order to finally make the stair saw I've been wanting to make for a while.

So now I had to figure out how to make the chop. The easier thing would be to take a couple of hunks of wood that fit the size of the slot in the bench. OK, how big will the slot be? My original drawings showed it to be the width of one of my boards, 1 3/4" wide. As I looked more closely at the screw, the guide plate, or whatever it's called that holds the screw up by the handle, is just over 1" wide at its widest. That would not leave much room for the screws on either side. I decided to double it and make it beefier. So, my slot is now 3.5" wide.

The next problem to think through was how to make the chop. (wasn't that the problem above? That just illustrates the constant process that in order to solve one problem you've got to solve three others) So, I wanted to think about grain orientation and stress on the tabs. It's much harder to snap off a piece of wood across the grain than it is going with the grain, so it made sense, at least to me, to have the grain orientation of the chop perpendicular to the orientation of the bench top. This way, the upward or downward pressure on the chop would be trying to snap off the tabs across the grain. It also allowed me to laminate the chop in such a way as to capture the pad of the vise within the wood to make a solid connection for either pulling or pushing.

Let me talk about the laminating for a second. I figured there were only a few ways of affixing this pad at the end of the vise screw to a wooden chop: drill holes in it to use screws or small bolts, or encase the pad in the wood and count on the strength of the glue on a large face-to-face glue joint to hold it in. I went with the second approach. I figured if it didn't work, then I could still try the first.

The chop is in three parts. The first part is just a piece of 4/4 maple that will eventually hold the hole for the bench dog. The second, middle piece has a shallow square cut out for the pad so that the pad sits below the level of the face that abuts the third part. The third part is the most complex part in that it must have a hole for the collar that captures the screw itself, and, since I wanted the ability to take the screw out, I needed a slot to access the screw at the bottom of the pad that held the screw into the collar. Perhaps we need some pictures.

I roughly cut to shape the three parts, marked out my final cuts and then glued them together just in the places where I'd eventually cut them out, so just at the corners. This allowed me to cut them all at the same time. And man, do I need a good carcass saw.

Here's the second part with the small square being cut out

Here's a picture of it all glued up. I left it long on the bottom since I was using some scrap 5" wide boards from the legs. After glue up I cut them to the width of the top. You can still see the layout lines on the tabs for my original idea of a half-dovetail shape.

Here's how the whole thing fits together

In the last picture you can see a couple of the reminders of just how painful the whole process was to create the chops. You'd think this was a fairly straight-forward piece of woodworking, and you'd be right, but it's where the piecemeal nature of my shop time can be deadly. I ended up cutting four sets of the chop pieces before I got it right. The end of the screw is sitting on a couple of the victims of my inattention.

The first set I cut out and realized the tabs were too small. The second set I glue up together before hitting myself over my head with my persuader because I hadn't cut any slots or holes or squares and encased the pad. The third set I mis-measured and the dimension to the outside the tabs was equal to the inside dimension of the slot, so too narrow. ("Measure 10 times, cut once. Measure 10 times, cut once...") I even cut two pieces of a fifth set to the dovetail shape before abandoning that idea, so it was almost five sets. Each mistake is almost directly tied to the fact that I had no real plan for the chop and since I was coming into the shop for only a couple of hours a week, I was always rushed and had to catch up to what I had been thinking a week before. Almost. I can also just be plain stupid at times.

Anyway, I have the chop, and now needed to test it out.

I'm all about doing this whole thing at least twice to learn the first time so that when I do it for real I have that much more of a chance of doing it right. So, I took a couple of off-cuts from the lumber for the top and set up a small model of that part of the bench with the slot, and marked out and started to cut the groove.

The groove for the tabs to ride in is approximately 1" wide, at least 1/2" deep and 10" long. I've thought about doing them as stopped only on one end, stopped grooves on both ends, or stopped grooves with The Schwarz's technique of drilling a hole in each end and using a moving fillister to cut them. Unfortunately, I don't have a moving fillister beyond my combo plane, and I didn't relish trying to use that to cut such a large groove. So I tried the chisel technique I learned at Woodworking in America.

It worked, and I now have a full-sized, stopped groove and it worked really well, but it is a lot like work, especially with the quality of my 1" bench chisels. I recently found a galoot who is selling me a 1/2" pigsticker mortise chisel which I hope to try out and use on the final cut. It should make much shorter work of the final groove and can help with cutting the other groove I need to cut for the sliding deadman. That groove will be much longer, closer to 36".

Well. We're finally caught up to where I am on the bench. It's been an adventure so far, and I'm sure I'll have plenty more to learn as I go along. And after all, that's kind of the whole reason to build my own bench, for the experience of it.

See you next time,


No comments:

Post a Comment

Since this is my blog, for my purposes, if you comment on here, I reserve the right to delete whatever I feel like. But I'm pretty friendly.