I'm planning to build a dining table. Current plan is 8ft x 4ft with two pillars.

My goals are to avoid:

  1. People sitting down and banging their knee on some out-of-sight component,
  2. The table pillars breaking under the stress
  3. The table to be stable when someone leans on one of the sides.

With this in mind, are there any rules of thumb for:

  1. The distance away from the edge to keep the components (e.g. the pillar should be 16" from the end, or whatever),
  2. The thickness of the pillar (current thought is the pillars will be 2" slabs, with a semi-circle(ish) cut from the sides to make a kind of hour glass when viewed from the end of the table. What thickness should the thinnest part of the hourglass be? (And with reference to criteria 1, how wide and deep should the thickest part be?)
  3. How wide should the pedestal be to be stable? Again, any rules of thumb that woodworkers know? A back of the envelope suggests that it's more stable than you'd think. Even if the center of mass was at the top of the table (30" up) and your legs were only two feet wide, you'd need to tilt it by 20° before it was falling. I haven't calculated the torque needed yet, so I'll do that, but I seem to remember doing so before and being pleasantly surprised.

Anyway, I'd love to know how experienced woodworkers thing about this problem, and if there are rules of thumb that I ought to know.

Edit: I sketched out a rough shape in tinkercad. This is a rectangular approximation, but might make my question clearer. I'm going to try do some strength analysis to see where it needs to be strengthened, and the corners would not be as sharp as indicated here. There are still multiple variations I'm playing with, but hopefully, this makes it clear.

Side view: Side view

Length is 8'. Cross beams are 5' (so the overhang is 1.5' on either side) All wood is 2" thick.

Is there a rule of thumb for that 1.5' overhang?

End view: End view

Width is 4'. Pillar is 1'. Pedestal and support are 3'.

Is there a rule of thumb for the pedestal and support widths?

Is there a rule of thumb for the dimensions of the main support pillar?

Three quarter view threequarter because why not?

Here's the link to the tinkercad.

  • 1
    I realise this is all interrelated info but Qs should be reasonably scoped, and tightly focussed, so should as much as possible have just one question in them.
    – Graphus
    Feb 4 at 14:58
  • I didn't draw it in, but I'm planning two beams connecting the pillars. One at the top and one towards the bottom. The middle of the table will be 2 inches and then taper to the edges.
    – Dr Xorile
    Feb 4 at 16:11
  • This answer woodworking.stackexchange.com/a/1639/5783 have the rule of thumb of 4' span for a 1" board, which is useful.
    – Dr Xorile
    Feb 4 at 16:12
  • "The middle of the table will be 2 inches and then taper to the edges" Presumably you'll be using a slab or glued-up panel here, have you a plan to do these long tapers (chamfers)? The bigger the taper/wider the chamfer the more challenging they get to do. Plus consider that the positions for attachment of the 'legs' should probably remain unaffected, giving a built-in limitation for how far you can actually take this, although this isn't by any means set in stone.
    – Graphus
    Feb 5 at 8:32
  • 1
    Thanks for the added images. It occurs to me that you could ditch the top stretcher for this design. As long as the legs are attached securely. Building with solid wood there certainly isn't any need for that top stretcher to support the top to prevent sagging! A 2" thickness of even pine is remarkably stiff. And this could allow you to raise the bottom stretcher significantly, if you don't mind the aesthetics, to essentially remove the possibility of someone bumping a toe into it.
    – Graphus
    Feb 15 at 9:18

2 Answers 2


There are published numbers for some of the info you want here, but you're in a better position (haha, sort-of pun intended) to get the numbers you actually need by actually measuring, with your legs and those of other family members or friends, on your seats, with the various preferred sitting postures of everyone.

This way you won't fall foul of the problem of theoretical numbers versus real-world situations (which don't always line up the way they're supposed to).

Just in general on one aspect, it's almost impossible to completely avoid the possibility of building in a toe-stubber, since anyone slouching back and extending their legs under a table could potentially run into virtually every possible leg/central spindle arrangement if they do it carelessly in the wrong position.

How wide should the pedestal be to be stable?

The stability of the table is subject to many variables, not least of which is the sheer mass of the whole thing and thence its centre of gravity. Without complex mathematics I think the only way to find this out is empirically — so it's a built it and see. However, since the top is intended to be 8' x 4' (default size for a full sheet of almost all manmaded boards) you can testbed this quite well using a couple/three full sheets of ply, MDF etc.

  • Actually, the mathematics is not that complex. Basically torque calculations. I have also been measuring family members sitting, so appreciate the thought.
    – Dr Xorile
    Feb 4 at 16:16
  • While the formula for calculating CoG is (ΣD* W) / ΣW applying that to something like this would not be straightforward because there are complexities here. Assuming equal force exerted at the edge, the turning moments for positions X, Y or Z would not be equivalent because the mass is not equally distributed (like textbook examples seem to assume from what I can see). So I'm presuming higher-level mathematics would need to be involved. Much easier to find out empirically given the calculations would in any case need to be based on the finished parts and their actual masses anyway :-)
    – Graphus
    Feb 5 at 8:23

Based on your sketch, there are several things to consider.

  • The table-top is very thin and will need additional support to help keep it flat. You can see in the side view that the top can sag in the middle between the two pedestals. The table should be supported by a continuous apron (3"-4") near the perimeter and closer to the center each way.
  • The side view shows that the table-top will have a tendency to rack to the left and right. Think of the connection point of the top and pedestals like a hinge. There is not enough connection to resist the racking so the hinge is free to move. Since the pedestals are wider in the perpendicular direction, the hinge action is contained and racking in that direction will not occur. You can address that racking by creating a connecting frame between the pedestals just below the top. This will also serve to support the top from sag.
  • The end view illustrates the problem for resisting a tip over if someone leans heavily on the edge of the top. If the base of the pedestal is narrow, the table can easily tip. While the pedestals do not need to be wide over their entire height, they should have wide feet and be internally adequate structurally to resist tipping.
  • Your issue regarding knee banging may be a minor concern. If the pedestals are set back adequately at the long edges to accommodate seating there, the location of the sides should be easy to fit where seats actually are.
  • Another concern will be wood movement. If the top is plywood or other sheet goods, then connecting the top and base will not be a problem. If the top is solid wood, then you need to consider how the top and base connect more carefully. (There are lots of questions on this site to review dealing with wood movement.)

Like with many furniture design concerns, there are a lot of answers found in visiting a furniture store with a tape measure, pad and pencil. You can try out a number of options and if it works there, it will work for your design.

  • I didn't include the cross beams in my drawing, but the plan is to have two connecting the pillars so the base can stand independently. I'll connect the two so that they can be separated for transport. The middle of the table will be 2" thick. I made a desk that's 7 or 8 feet wide and 2" thick, supported right at the ends and takes a reasonable load without complaining. So the overhangs will be like floating shelves from this central pillar area.
    – Dr Xorile
    Feb 4 at 16:20
  • 1
    @DrXorile Sounds good. Feel free to come back with further questions as you develop the design further.
    – Ashlar
    Feb 5 at 1:42

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.