Oak coffee table design

[dh7892]

Hi,

I'm about to make myself a coffee table out of oak and I'd welcome a critique of the design before I start butchering any wood.

I bit of Googling for inspiration found me this:

Which I rather liked and decided to base my design on.

After a bit of work in sketchup, here's my design:

And another view:

And a detail of the joints in the leg:

Although I haven't shown it, I'm planning for them to be draw-bored M&T joints.

So, what do you think? A few key questions I have are:

How should I join the frame to the top? There's a decent horizontal runner which will provide some non-endgrain for glue but I think I should either dowel or M&T the legs to the top?

Also, how should I make up the top? The simplest approach would be to just joint a few boards together and I think this would look OK as I don't want to make it too fancy.

Any comments on the leg joints?

Thanks for looking.

Dave

 

[jasonB]

Is your table a lot smaller overall as it looks a lot taller to me, I would make a coffee table about 350mm high.

You need to allow the top to move so M&T is not an ideal fixing, some form of buttons or slotted holes will be best.

Edge jointing boards together will be fine but the design you are basing it on has breadboard ends(sides actually) which do help keep the top flat.


Your tennons are quite short so draw boring may not have much effect as it could just pull the end grain out of the tennon.
J

 

[dh7892]

Thanks for the comments.

I'm not sure of the dimensions of the original but my version stands at 450mm high. It's 1.5m long and 500mm wide. It's based on the rough proportions of my current coffee table which is about right for me.

Yes, I thought about bread boards but I don't actually like the way the original is done. I might have a go at doing a more accurate rendering of the top with different options to see if I can visualise it better.

How long would a tenon need to be in order to be good with a draw bore?

 

[jasonB]

I'd like to see double the tennon thickness beyond the hole so 28mm.

You don't really need to draw bore if you are glueing as well, just glue & clamp it up tight then drill the holes and drive in some dowel.

J

 

[dh7892]

Ok so they're ok as they are if I just dowel them? It was more the look of the dowels I was after rather than the strength of the draw bore anyway to that's fine.

 

[barkwindjammer]

The two 'upside down' truss brackets could be made of one piece and half lapped with the legs ?
just my tuppence worth

 

[xy mosian]

Hi, dh7892. I like the overall look of that table, and I appreciate it is only a concept sketch. However if I may be so bold:-
I took the liberty of altering the look of the area outlined in blue. I think it works better, less interupted if the curve flows across the leg, shown in green. If these pieces are made as one, with a cross-halving joint as suggested by barkwindjammer then it would be easier to achieve. The lower side of the bracket/leg assembly looks fine.

xy

 

[Mike Wingate]

Look at bridge design for shape and clues on joining. Don't weld the wood. Nice design.

 

[dh7892]

Thanks for the comments. That's why I posted it; so people would say things like this.

I agree about the flow of the trusses. It does look better if the line of the curve flows across the two pieces. I'll make the necessary changes.

I'm not so sure about doing a half lap though. For one thing, that would mean that I'd have to make the whole section out of a much wider board. That does seem a bit wasteful. Also, I'm not massively happy about the fact that the end bits would then be orientated a 45 degrees to the grain. Wouldn't that make them a bit weak? Not that strength is really going to be an issue here but it seems like bad practice.

Also, I don't think it would look as good but that's just me.

I agree about looking at other architecture for inspiration. In fact, the website from which I found that original table take their inspiration from old French barns so you can see why the trusses are quite architectural.

I'm not sure I agree about bridge design for clues about joining (unless you are talking about wooden bridges). Whilst the structural elements might share some similarity, the grain structure of wood means (IMHO) that it must be treated a little differently. You don't see many natural wooden I-beams or wooden planks being riveted together. But the general proportions of something like the Forth rail bridge does give a nice clue to the curvature for the trusses.

 

[SketchUp Guru]

You are correct about the grain on those diagonals. It should run the long way on the part. You'll have enough short grain as it is but those areas won't be structural. Have you thought about moving the end stretchers that are up tight against the table down to the height of the front/back stretchers? It might be too much joinery in one place but I think it would be interesting to look at. Besides, I can't help but wonder what will happen when someone inadvertently kicks one of the legs. On the original the legs look shorter and a bit beefier than in your sketches. I would be worried about it collapsing if someone kicked the table from the front.

 

[dh7892]

If I understand what you're suggesting correctly, then I agree on both counts: yes it would be interesting; and, yes, I think that would be too much joinery in one place.

The legs on the original were beefier. I think I like the look of mine better but I strength is something to worry about. I've currently got the legs as 40mm x 40mm thick. That seems like it should be strong enough as a leg but I think that the problem is any force pushing the leg from front to back. As the only cross-member actively resisting that force would be the arch at just under the table-top, I might want a re-think about the to make sure it would be strong enough. Perhaps if I make the sides of the arch taller, the resulting extra width of the tenon would be enough to add the necessary strength.

Any more ideas about attaching the top? Care to elaborate on the "buttons or slotted holes" and how I'd go about implementing them?

 

[xy mosian]

Any more ideas about attaching the top? Care to elaborate on the "buttons or slotted holes" and how I'd go about implementing them?

I wonder about a 'batten' along each side tennoned onto the tops of the four, two legs and two brackets, leg frame members? On a coffee table it's not going to be seen. Slots could then be made, across the grain of the battens, and screws used through these to fix the top.

Just a thought.
Xy

 

[dh7892]

I'm not exactly sure what you're describing.

I did think about just screwing up from the arches but that would, at best, only give me 4 screws.

I also thought about making the tops of the arches into a sliding dovetail to slide into the top but I think that might be a bit beyond my skills.

 

[xy mosian]

Ok Let's hope these quick, 'fag packet' sketches show my thoughts. Traditionally a table would have a frame around the top of the legs. This gives some method of fastening the top to the legs, either screws, stretcher plates or buttons. I can see that a conventional frame would spoil the appearance of this table. The solution I am suggesting, the lengthwise batten, is shown in the attached sketch. The slots for the screws would need to be organised according to the grain of the top. I think this rail need not be very thick, and would probably not be seen in normal use.
I have included an idea for the rail/leg/rail joint, basically a half-lap between the two tenons, that should not only strengthen the joints but may give somewhere for pegs to work.
Sorry about the crude sketches, I haven't got my sketchup head on this morning.

xy

 

[dh7892]

Thanks for going to the effort of sketching that up.

I like the leg joints. I think I'll go with that (or something very similar).

I get your idea about the rail at the top now. But I have a question. If the point of attaching the top with screws etc is to prevent problems caused by the wood moving, what's to prevent the same problems occuring due to that rail expanding? I can't really see why M&T joining to the rail would be any different to M&T to the top.

 

[SketchUp Guru]

Those rails won't expand and contract as much as the top will. Well, they will as a percentage of their width but with the mortises being centered in the rails and the two rails being separate from each other, their expansion and contraction will be a non issue. The slots for the screws into the top will allow the top to do what it wants without affecting the base.

 

[dh7892]

Silly me. I wasn't thinking clearly. I was thinking that the wood would expand and contract most in the direction of the grain but, of course, it's perpendicular to the grain that most movement happens.

I suppose that I could rebate the entire rail with an over-sized rebate (to allow for a little movement) which will hide it from view even more.

I'll try fiddling around with sketchup a bit more and see how it looks.

 

[dh7892]

OK,

I've had a fiddle and here's what I'm proposing to do. The above suggestion will work as a fall-back but I think this will look a little better:

Shown with the top hidden so you can see the screw holes. The centre ones are fixed to keep the top centred and the side ones have a slot to allow movement.

 

[SketchUp Guru]

I think it'll work. Give it a go.

 

[dh7892]

Cool, thanks.

Quick question while you're here Dave: When I was just messing around with SU I noticed something that bugged me. When I was editing a component, if I used the middle mouse button to spin around, sometimes it centred the orbit around the click and sometimes it seemed to centre it around a far away point. Seemed to be like it was taking the centre of rotation around a point on the surface where I clicked the middle mouse. If the mouse wasn't over a surface, it took a point in the distance.

Does it always do that? I don't remember it being like that but I may be wrong.

Cheers,

Dave