Table Design Considerations

Design, like most other things, is not as simple as it might appear at first glance. Aside from the actual look of the piece, it is quite easy to design a piece that looks right, then find it impossible to make, or even if made, it will not last because the basic "engineering" is wrong.

As an example, take a simple table design with four upright legs, connected with rails and braced nearer the bottom with stretchers. Looking at Diagram 1, it becomes clear that a sideways force in example "A" will have more chance of racking the piece than in example "B" simply because of the proportions and hence the multiplied stresses working on the frame.

Diagram 1.

Tables must resist side forces if they are to last, to be safe and to withstand daily use and accidents. Most tables will have long rails and shorter rails and the trick is to ensure that they can withstand the forces by making the joints appropriately strong.

The main factor in a frame strength is the joints and commonly mortice and tenon joinery is the most favored for a number of reasons. They are fairly straightforward to produce, not too difficult to make accurately and - provided that the mortice is made first, not too difficult to adjust or "fine tune."

However, whereas most would opt for a square section leg, if you look at the plan of a square section leg with mortices cut out, it will appear as in example "C" - quite a bit of the strength of the upper part of the leg has been sacrificed to mortices. OK, so they're going to be filled with their tenon parts and glue, but the grain is weakened. Now look at example "D" - a leg with a rectangular section. You can see that more strength is maintained and the option of a longer tenon for the longer rail is introduced.

Dealing with mortices and tenons a little more, the next question is "Should the rail be central within the leg, or can it finish flush with the outside face?"

Well, it can do both, as in Diagram 2. The main deciding factor here is the denseness of grain of the material - in other words, softwood or hardwood. The denser (harder) the wood, the less chance of a collapse of joint if using flush-with-outside tenoning. That's not to say you cannot do so with softer woods, just that the risk of joint failure is higher.

Diagram 2.

Another factor to consider is that the closer the mortice gets to the top of the leg, the more chance there is of the joint failing under a side force, as the resulting small shoulder left will not resist the racking and will pop out. Yet many will worry more about the fit of the tenon in the mortice than whether the amount of wood left above the mortice will resist a sideways strain on the tenon without popping out.

That said, simple mortice and tenon joints are still the way to go for the majority of tables with the above configuration. Modern phenolic or polyurethane glues are so good now that they will not only almost match grain strenth - or even overcome it - but the polyurethane glues will even fill any voids within the mortice produced by a too sloppy fit. Beware, however, of fitting a tenon so tightly that it needs hammering in. This will not only strain the joint, but will cause problems with the glue as well. Either there will be a pressure build-up as the tenon is driven down, resulting in a situation where it is impossible to close the joint, or all the glue will be pushed downwards, leaving little, if any, on the shoulders. Tenons should be a push-fit into the mortice and the shoulder of the tenon is as important a glue surface as the rest of the joint.

Diagram 3

If that is your chosen route, spend careful time constructing the frame and get that totally right before even bothering about gluing up the top. The top is mostly incidental, decorative of course and essential to make use of the frame, but it is the frame that takes the skill.

There are alternatives, but the degree of difficulty also increases as different design concepts are considered. The style of two slab legs, braced with a central stretcher may look simpler - but it is far more difficult to achieve without appreciating the differing forces acting on the structure.

Diagram 4.

Looking at Diagram 4, a force applied as in example "F" will much more easily cause racking as there is no rectangularity to the base, just two perpendicular slabs with a bracing stretcher. To stabilise this design, the central stretcher becomes vitally important, as do the methods of attachment to the top. The top, in this example, becomes much more a part of the structure than in the first design and yet, being made from a glued-up panel, will need to expand and contract just as much.

The most practical way is to use a through tenon in a square mortice, locking the shoulders against the inside of the leg by passing a wedge through from above, as in this diagram. Making a square mortice is easy, making a square tenon to fit not difficult, but the proportions of the mortice within the tenon itself (where the wedge will pass through) is a danger area all of its own. Too small and the pin will be puny and fail. Too large and the mortice will weaken the tongue of the tenon sufficiently for it to fail. This stretcher is a job for dense-grained wood if the table is likely to be hard-used.

Diagram 5.

It is possible to use more than one bracing stretcher, even one above the other, which will give almost as strong a base as the traditional table designs. Fixing tops to conventional rails is simple - run a dado round the inside of the rail and affix the top via right-angled clips - I expect you've seen Norm do it many times.

The grain is likely to run in the same direction as the top if using slab legs, so that's one consideration in our favor. The other is that as the legs are fairly long boards, they'll probably move equally as much as the top - but there are NO guarantees - wood is organic!

If the base is braced sufficiently, then a simple frame of cleats fixed (screws only) to the underside of the top, within which the legs locate, will be OK. The only problem you've got left is - will the top take sufficient downloads, because there is nothing supporting it between the uprights.

Whilst none of the above in any way forms part of a plan for a table, I hope that it has given you a few things to think about before either following someone else's plan, or thinking about designing your own.