Thread: Attaching bench top to the legs. Why does everyone use mortise and tennons?

.

I suggest two reasons. The first is that the mortices provide additional lateral rigidity if you only have one set of stretchers in the design to bridge from one end frame to the other, and benches are built in this way in order to keep the undercarriage of the benchtop clear of obstructions. The other is that the front mortice registers the front edge of the benchtop in line the front of the legs....at least it will if you designed the bench that way. I built in the capacity for backwards expansion of the benchtop in my build but I don't see how the Roubo bench with massive legs dovetail- morticed in the benchtop is designed to accommodate the top expanding. I assume it just bends the end frames open as expansion happens. To me that is not a good design though.

Having said that there are many ways to skin a cat and if you, like me, want to use just what you have (i.e. timber) instead of using additional brackets then mortice and tenon joinery is pretty effective. I thought it was good practice for morticing with a chisel for me too. It didn't take too long to cut them either.

The main reason for me was because the bench has no stretchers under the top. Top sits directly on the legs. The mortises are blind as I didn't want the end grain showing through on the top and having the possibility of them sitting proud or dropping below the top as the wood moved. Probably wouldn't have been a problem but I wasn't willing to find out.

Here is the plan for my workbench. Given the stretchers run front to back right under the top, I can't see that mortising will add any benefit.
Thoughts?

Attachment 181370

My design calls for top stretchers similar to your design

but then I'm going for a split top where any expansion movement will be into the centre. I'm going to use mortise and tenon or large floating tenons (depends on salvageable timber), mainly to keep the top level (registered) with the frame.

I was not so much thinking that the mortice and tenons would be useful for front-to-back rigidity but more for left-right (side to side) rigidity. If the end frames are solid units (drawbored or suchlike) they will be very rigid but if you use single long stretchers in the longer span between the end frames then you may benefit from additional rigidity especially if you use knock down joints for the long stretchers running left-to-right. In your design with those twin stretchers at the back providing left-right rigidity, then morticing the legs into the rear of the benchtop is of little additional benefit.

I am not sure what the definition of a stretcher is..... I always assumed it's a horizontal of the. long axis of a piece of furniture not the short axis.... but maybe it's just any horizontal structural member?

With the way your design stands at the moment, you couldn't MT the top anyway because the risk is either popping the stretchers out of their mortices or splitting the top depending on which way the top moves from it's starting condition and with the seasons. You could fit 3 coach screws though the stretchers and into the top, with slotted holes for the front and back screws to move with the seasons or you could delete the top stretchers and MT the top to the legs as Roubo benches are done. I just cannot see buttons, no matter how large providing enough friction to stop the top sliding around when you apply (significant) lateral forces to the top. That's the idea of buttons, to allow movement. The coachscrews work because the centre screw is a zero-movement fastener.

Mr pampelmuse, I am not sure what the definition of a stretcher is either! If it runs horizontally between uprights, then it's a stretcher to me! )

mic-d - The idea of the coachscrews is a good one. I was planning on using coachscrews in the end caps with a sliding mortise/tennon (as opposed to sliding dovetail which is just way too much work for my money), so I might just do as you suggest to hold the top down as well.

It doesn't looks like anyone's going to recommend that my design could benefit from mortising the bench top, which works for me, so I guess I'll be sticking to the original plan with either buttons/brackets/ or coachscrews to hold the top down.

TN

A further suggestion from mic-d's, If you want to ensure all the front surfaces of the bench align over time I would fix the bench at the front edge with a coach screw/dowel/m&t to register it there and allow expansion/contraction toward the rear of the bench. Of course you may find it never moves enough to be problematic but you can never be sure

I find it odd that in Roubo style benches the problem of the top expansion pushing the legs apart appears to be ignored. I would have thought it a design flaw worth commenting on but I have not seen it discussed in the various blogs/forums

Originally Posted by pampelmuse

I find it odd that in Roubo style benches the problem of the top expansion pushing the legs apart appears to be ignored. I would have thought it a design flaw worth commenting on but I have not seen it discussed in the various blogs/forums

Schwartz discusses it in his book on workbenches.

Originally Posted by mic-d

Schwartz discusses it in his book on workbenches.

and he didn't seem concerned if the frame(?) became out of square

Seems some of the French style benches with massive tops had the rear legs already set at a slope, only the front legs were vertical, according to Schwarz and others. They also had huge dovetailed mortise joints from the legs into the tops.

A more practical method for modern construction that I have seen mentioned is to use 'loose tennons' - basically drill matching holes in the top cross bar and the underside of the top, then drop matching dowels/pins into the holes. If you are worried about the top lifting you can put a lag bolt in at each end as well, but you'd have to be using a fairly large sledge to pound things to get that much bounce.

Originally Posted by bsrlee

... but you'd have to be using a fairly large sledge to pound things to get that much bounce.

sorry
we keep our best sledging for the cricket field

Originally Posted by TimberNut

For my upcoming bench I plan on attaching the bench top to the stretchers between the legs using the equivalent of large table buttons (might even end up being recessed steel right angle brackets).

Most designs I see here have the legs mortised into the benchtop. I'm struggling to understand the benefit in doing this. It seems like an awful lot of work for no apparent benefit that I can determine.
One beneft of large table buttons (or similar) is that the top will be removable in the future if needed, should the bench need to be shipped somewhere other than my shed.

So, before I cut the legs to length and commit myself to one attachment method or the other, can someone enlighten me on the pros/cons of each approach?

TN

Hi TimberNut.

I've recently made a Roubo bench as demonstrated by Roy Underhill, and I made the mistake of buying green oak.
The wood moved between the cutting of one tenon and its mortise, and the cutting of the other.
It got a bit frantic getting the front two legs plumb and square with the top...
Here's the thing: Once I'd got the top on the four legs (the back legs slope away from the bench-top), the top has moved within itself ('crowning' on the top surface and a dip out of level at the 'top' end of the bench), but the bench has remained completely stable.

I'd say the fact that the joints are wood-on-wood is an important factor here as any movement is absorbed by its flexibility.

Maybe, since earlier woodworkers didn't have access to the kiln-dried, planed wood that we have today, let alone the tools...they made do with green wood and hand-made joints, among them the mortise and tenon.

It works, even in my clumsy hands...

The Pom.