making tables

[Sgian Dubh]

Is that drawing available for me to copy and expand? It would certainly save me many hours marking out.

Help yourself to it, although bear in mind the dimensions I used are random. I just used numbers that seemed about right for the height and depth of a coffee table like structure. I could send you the .dwg file, but if you don't have access to AutoCAD I suspect it's not much use to you, and I could also supply you with the PDF 'printout', but that's set to print at A4 so you'd need to muck about with magnification and possibly multiple sheets to get the printed dimensions to match the actual dimensions.

Alternatively, take the information in the JPEG of the CAD drawing in my earlier post and replicate it as accurately as you can on a piece of board material, ideally something like 12 mm MDF which is easy to work on its edges, shape this by any means you can and use it as a template for use with a router and top or bottom bearing pattern bit.

Jacob is right that in your case it looks like you'll have to resort to that old standby of paper and pencil - not unless you know someone with a CNC machine that could run my CAD drawing for you. Slainte.

 

[sunnybob]

I'm certain there isnt a cnc machine on this whole island, maybe not even in a factory.

My main concern is the angles of the legs. I can do technical drawing, I'm just rubbish at any thing to do with maths.

I'll see if i can use your drawing as a base. Thanks very much.

 

[sunnybob]

Just had my annual brainwave.
I can buy all 9mm ply, and double up on the legs to give 18 mm, that will minimise wastage and still give the strength it needs. 2 sheets should be ample.

 

[Sgian Dubh]

My main concern is the angles of the legs. I can do technical drawing, I'm just rubbish at any thing to do with maths.

Well, if you're making an exact copy from the drawing I provided, there's little or no maths to consider. It's pretty obvious that the acute angle described by the legs from the floor is the complement of 105º, i.e., 75º. All the critical dimensions are identified to the nearest millimetre, meaning that here and there a sum or two don't quite add up because of the rounding up or down that AutoCAD executed. For greater precision I could have set the dimensioning tool to work to quite a selection of decimal places, but I generally prefer to keep drawings as uncluttered and clear as possible. Additional decimal places in dimensioning can sometimes make a drawing look rather overcrowded, and are frequently not necessary in woodworking - maybe useful in space technology and engineering (for example) but not for most wood butchery, ha, ha.

What I found really intriguing in your initial image (first post) is that the grain of the stacked legsets aren't consecutive or very near consecutive, and I wondered why. Is it because the grain really is consecutive if the tables are stacked in the 'right' order, and in your image they've been stacked randomly in both height and end to end? I can't imagine the makers would deliberately set out to make random 'non' grain matching, but maybe they did - who knows? Slainte.

 

[Jacob]

You can avoid maths entirely and almost avoid measurement if you start out by drawing your rectangle (300x500?) a parallel line for the top then the legs - guessing the angles from the picture but making sure that it's all symmetrical by marking out with dividers and a straight edge.

 

[MatthewRedStars]

My main concern is the angles of the legs. I can do technical drawing, I'm just rubbish at any thing to do with maths.

What I found really intriguing in your initial image (first post) is that the grain of the stacked legsets aren't consecutive or very near consecutive, and I wondered why. Is it because the grain really is consecutive if the tables are stacked in the 'right' order, and in your image they've been stacked randomly in both height and end to end? I can't imagine the makers would deliberately set out to make random 'non' grain matching, but maybe they did - who knows? Slainte.

The factory will be banging these out by the hundred. The legs will be CNC'd from full sheets nested as tightly as possible. Grabbed off the CNC bed as quickly as possible, piled on a trolley.... then over for edging, back onto another trolley (maybe putting a few to one side for mending/binning). Up for spraying, onto the drying racks. Then over to the maker. The maker will just grab a leg off the pile of legs and screw/glue/whatever them to the box. Grab another leg etc.

Ensuring grain matching will add another process (physically and mentally) - and increase waste as if one leg gets mauled by the CNC then another 5 have to go in to bin.

This type of work is margin, margin, margin. Saving seconds at every process. Not very edifying but such is life!

(edit: they may even be deliberately mixed to ensure an accidental match doesn't occur)

 

[woodbrains]

Key to making the thing as Woodbrains points out above is matching the inside and outside radius. That is not at all an easy thing.

Extremely easy if you draw it up (with a pencil) as a stack as I suggested above. (Or as per Richard's left drawing of a stack). One compass (or freehand) drawn line does both the inside and the outside of the curve - they can't fail to fit, just like a jigsaw, even if wiggly and erratic. You can then fair them in to suit, freehand, or with bendy curve rulers, french curves, compasses, etc etc. Difficult to get wrong, but not impossible of course!
You then make up a template and have every opportunity to refine this by further trimming if you really want to.
A basic hands-on craft process is "offering up" where you put one component against another (in this case a duplicate, or the outline drawn out) - see how they fit and adjust accordingly.
Making mountains out of molehills! - I sometimes think that should be the UK woodwork motto!

Hello,

Making the drawing is extremely easy, Jacob as you keep repeating over and over.......

But making the template accurate enough is not, which is the point that is being made. It is not impossible, but not easy without, let's say, a bobbin sander to fair the inside curves. And it will have to be tested against a trial, seen where any errors lie, fettled and tested again.

If Sunnybob will pay the postage, I will CNC him a master template, if he indeed decide to do the job. If he wants me to use Richard's dimensions, then I will, or I will draw up my own from a scale drawing of Bugbear's photo correction thing. That will put paid to Jacob's infernal ramblings of how easy it will be!

Mike.

 

[sunnybob]

Woodbrains, a very nice offer indeed, but that postage may well add 25% to the overall cost.

We are getting to the "it aint a spaceship" stage on this build. I dont need sub thousandth of an inch accuracy especially as it will be ply and not grained wood.
With a drawing, I have the skill to make a template. I have a bobbin sander and a big router table and a 350 bandsaw and I can use them quite well.

I cant believe this has gone to 5 pages, but as said at the start, I thought the legs were all different. That was what I started the thread for. The fact that you have all convinced me they are the same has greatly simplified the job.

I have slung in a 150 euro quote, which guarantees covering the wood costs, everything else is hobby time.

I'll let you know as and when (if) he accepts, and make a thread of the build if so.
Thanks to all for the help.

 

[bugbear]

There's something interesting in the design - you're not as free as you might think. The leg thickness, leg angle
and height of the "web" at the top of the leg are intimately interlinked.

Where 'L' is the thickness of the leg and X is the vertical height of the web, and α is how
far the leg is off vertical.

So:

sin(α) = L/X

Or, for a given angle and leg thickness:

X = L/sin(α)

BugBear

 

[woodbrains]

There's something interesting in the design - you're not as free as you might think. The leg thickness, leg angle
and height of the "web" at the top of the leg are intimately interlinked.

Where 'L' is the thickness of the leg and X is the vertical height of the web, and α is how
far the leg is off vertical.

So:

sin(α) = L/X

Or, for a given angle and leg thickness:

X = L/sin(α)

BugBear

That is one way of looking at it!

Simpler to just ensure the centre of the lower radius is vertically aligned with the centre of the top one. Provided the legs are parallel in width, you can do whatever you like with the leg angle, top 'rail' width, size of (congruous) curvature. It is cutting it out so the inside of one will mesh with the outside of the other that will be time consuming.

Mike.

 

[bugbear]

There's something interesting in the design - you're not as free as you might think. The leg thickness, leg angle and height of the "web" at the top of the leg are intimately interlinked.

Where 'L' is the thickness of the leg and X is the vertical height of the web, and α is how
far the leg is off vertical.

So:

sin(α) = L/X

Or, for a given angle and leg thickness:

X = L/sin(α)

BugBear

That is one way of looking at it!

Simpler to just ensure the centre of the lower radius is vertically aligned with the centre of the top one. Provided the legs are parallel in width, you can do whatever you like with the leg angle, top 'rail' width, size of (congruous) curvature. It is cutting it out so the inside of one will mesh with the outside of the other that will be time consuming.

Mike.

That you're describing is the issue of making the pieces tessellate (which I showed earlier with my inkscape sketches) ; I'm now talking about the size/shape of the pieces themselves.

BugBear

 

[Jacob]

......
Simpler to just ensure the centre of the lower radius is vertically aligned with the centre of the top one. Provided the legs are parallel in width, you can do whatever you like with the leg angle, top 'rail' width, size of (congruous) curvature. It is cutting it out so the inside of one will mesh with the outside of the other that will be time consuming.

Mike.

Thats about it. Or to put it another way - there is only one line to draw - the line delineating the inside leg shape is exactly the same as the line ditto the outside leg shape, except the latter is extended so that they both meet the floor.
You can't do whatever you like with the top rail width - it's a function of how far apart the two lines are - dictated by the desired thickness of the legs. You can't add material, but you could take it away and they'd still "tesselate" (less the missing bit!)

I like BBs attempt to make it really difficult - madly mystifying the craft keeps out unwanted competitors! "Tesselate" :lol:

Waiting for Pete M to zoom in on an egregious tautology!

 

[bugbear]

I like BBs attempt to make it really difficult - madly mystifying the craft keeps out unwanted competitors! "Tesselate" :lol:

Just basic craft geometry, Jacob, same as plotting the dovetails for a splay sided tray.

It's only Sir Oliver's Horse, if you remember learning trig at school.

BugBear

 

[Phil Pascoe]

I'd not heard of this one - but if you go into it it comes up 403 forbidden.

 

[sunnybob]

Oh man, last time I mention tables!
The only word i know that fits in here is esoteric. Two words would be "dark arts" but I dont swear in public.
I got lost about 7 posts back.
think I'll go back to my yin yang boxes.

 

[woodbrains]

Hello,

Oh FFS I drew this with no trig, just a compass, rule and a square. you can do it easily on a piece of mdf. Draw a half section through the centre line, cut out, fair smooth and use as a template for making the other half. repeat. double stick tape them together and make them the same. Test the nesting fit of each in the outside and inside of each other. if all is well, biscuit joint them together and this will form the master for making the other 6.

The dimensions are in inches, obviously. Dont sweat the leg width, it will look after itself when you construct the radii, just draw the inside line parallel to the outside one and touch the arcs, it will work fine.

If the overall width needs altering, just add or subtract material at the centre line.

Mike.

Oh, forgot to mention, the vertical line the centres of the circle are positioned on is 9 1/4 from the outside edge.

 

[Jacob]

That's it! Almost. No need for the radii to touch in the way shown. The profile of the top edge can be moved up or down to alter the thickness of the legs and the top rail, as you like.
If you haven't any compasses you can draw the radius around a bean tin or anything. Doesn't have to be round - an ellipse would be fine if you have french curves or something (a funny shaped tin?). Choose your shape. You don't need to know the centre of the circle you just line your tin lid up to fit between the straight lines neatly.
As long as the sides are mirror images and the line is repeated exactly for the inside and the outside, but with the outside line extended to the floor, the thing will 'tessellate'.
A straightedge, bean tin and pencil is a lot quicker and easier than trying to do it with Autocad - and you can do it straight on to the workpiece, which is impossible with Autocad which would leave you with problem unsolved - you'd probably have to borrow a pencil and start again :roll:

I think we are getting somewhere - last time a similar issue came up I was banned before we got this far!

 

[sunnybob]

Jacob, if you mention pencil one more time.......

 

[Brian Denton]

Hi,

They are half ugly and half attractive!

The only way I can see to make it worth doing would be to batch make 10 sets and sell the surplus ones online. It would be an interesting job if you are not too busy.

B

 

[sunnybob]

What? I've got a 6 page thread about making ONE set and you want me to make HOW MANY?????

(theres a (lol) in there somewhere)