Thread: Three way lap joint

If it was two pieces you would be talking about a halving joint - right? so in that case each piece is only half thickness at the join. If you extend that to three pieces you will not have much material in each.

What if you only partially cut them in? just enough so that there is no gap down the center?

There's a fundamental principle of stress analysis that states (more or less), a structure cannot be weakened by adding material, nor strengthened by removing material. I think the best you could do would be to reinforce each leg with an offset stout steel bar, so that the bars pass each other at the joint; then cut whatever is needed from the timber. I'm still not sure what the crossing geometry of the timber should be; Google [soma cube] for some possibilities - maybe. In effect though, the timber would be surface decoration for the steel. Sorry, but this doesn't conform to my definition of woodwork.

Joe

I know that this has been discussed here before.

But I'll be damned if I can find the thread to point you in the right direction.

Skew, I think there was one for a corner lap, not in the middle of the pieces. Extension beyond the corners wouldn't be very robust, I think.

Joe

No, it was for a similar table. I remember that someone posted a possible joint but the general consensus was that it wasn't a structural feasibility.

I also recall someone suggesting using a half-lap for two pieces and dowelling on the halves of the "third" leg with a metal rod. 'Cos I gave serious thought to trying it for myself out of curiosity...

Ok then... had a play yesterday and this was the best I could come up with. I cheated to get the third bit in and as already stated, there are structural issues. Don't know if I will use it or not but the test bit looks nice so might have to paint it up and keep it for future reference.

I've been thinking about this a bit further and i think you could get away with it. if you used the halving joint as above, but then joined in the third leg mortise and tenon style, where each half of the leg makes up each half of the tenon. The final trick would be to place the tenons in the third legs so that the one which reaches up the the table top has its tenon in the top half of the mortise, and the half that will be the leg has its tenon in the bottom half of the joint. This way the two tenons will resist the bending forces on the joint.

As a general rule try and get material at the edges of a join because it is better placed to cope with the leverage placed on it.

Sorry for the verbal speal, if you like i will throw some sketches up tonight but i am at work right now....

The problem with the T&M is that the mortise cuts away the structure of the half-lap.


A half-lap inherently halves the structural integrity of the members, creating a mortise to add the third leg reduces that to 1/4. If the legs are 2" sq cross section, at the joint they'd only be 1" sq cross section, so you've effectively weakened them by 75%

Originally Posted by Skew ChiDAMN!!

The problem with the T&M is that the mortise cuts away the structure of the half-lap.


A half-lap inherently halves the structural integrity of the members, creating a mortise to add the third leg reduces that to 1/4. If the legs are 2" sq cross section, at the joint they'd only be 1" sq cross section, so you've effectively weakened them by 75%

Sort of.

Really not trying to open a can here but you can remove a lot of material from a solid section and it can still be strong as long as you leave solid material at the periphery. The material in the center of the solid section really only stops the structure from buckling. It's how 'I' beams survive.

I would have to look up the numbers but as an example I would argue that if you remove the inner 1/2 of thickness from a solid section (leaving you with two tenons 1/4 thickness each) you are left with more than 70% of you original strength.

Now I am feeling the urge to conduct some destructive tests.....

The table that I was thinking of is more of a coffee table than a dining table. How much load does it really need to be capable of handling. A couple of remotes, a few brews and a couple of magazines. I do agree that by removing that much from a single length will reduce the strength of that individual piece by 75% but once it is all put together, strength should be increased by the other two sections. I don't expect strength up at 100% but wouldn't think 50% of the original would be unreasonable?

An I beam is rated as an I-beam... it has problems when used as an H-beam. A table leg needs strength in both directions, to both hold the top up under load and survive when the table is dragged laterally along the floor.

But in essence I agree... which is why I was tempted to try the metal rod idea. 'Cos it's stronger than wood, you could make a smaller hole through the half-lap, thus leaving it stronger, and both halves of the 3rd leg could be slipped over the rod as a sort of floating tenon.

I'd imagine that small M&T's would be an asset on each side, just to key both halves and prevent 'em from rotating around the rod. It'd also give a bit more glue area so you're not entirely dependant on a wood-metal glue up.

AFACT it's the strongest possible method... although making it actually work is another thing.

One day I may get around to building it... I'm still curious. Then again, maybe not. Too many projects on the go, not enough tuits.

Have a search for a "three piece Burr" puzzle, or look here:
http://www.research.ibm.com/BurrPuzzles/Burr3.html

It's very possible!

Cheers,

P

Yeah i agree about dragging it across the floor and loading in general. Like i said, maybe get away with... it almost feels achievable but ultimately it would be fragile.

Corbs do you want it to be thin round sections? If you went for 3 ovals you could taper the tips off to keep it looking light but the center where they all meet could afford to be a bit thicker. The pieces would be interesting to mark out though because they are not all in the same orientation (or plane)

Honestly I am just playing with ideas at the moment. Will see what comes about from it if anything. I thought there was a way to do it without parting one of the three sections but the only way I have seen to do that is with a round section which doesn't help with the structural integrity of the piece (I am still sure I have seen something like it in a flat pack table though)