Lamination of 3D shapes???

[ian maybury]

Just wondering if there are practical techniques about that enable the lamination of 3D shapes of reasonable depth from veneer - that is shapes with surfaces that bend in two or more directions at once - e.g. half of a solid sphere.

The problem with forming shapes like this from sheet material is that they normally require it to absorb length in some directions by compressing and stretching, as well as by bending in the normal way.

Plastic sheet for example when thermoformed to take up shapes like this tends to show quite marked variations in the final thickness - some parts stretch, while others thicken.

A long shot, but if you don't ask the question....

ian

[Russell Sansom]

You can't deform a plane into a sphere (or vise-versa )without stretching or compressing. I think all the world-map makers have a similar problem in the other direction. Thus all the famous globe-to-plane projections.

It's a bear of a problem in computer graphics as well, where my favorite solution is a bunch of wedges with faceted circumferences. Which leads one to think of coopering and the backs of lutes as a solution.

[Bret Duffin]

Interesting that you would pose this question. I've been pondering the same problem. What I came up with was some pizza slice shaped thin laminations pressed and glued into the proper radius. I don't think you could go much bigger than 1/16 of a sphere with each piece. Then you would have to figure out how to precisely cut all the pieces to the exact dimensions to create a sphere when it's all glued together. No small task.

[Alan Schwabacher]

If you look up the technique of making woodstrip canoes, you'll find the simplest way to generate compound curves. Overlapping veneer pieces could do it, but would require much more of a form. A very nice feature of strips is that you can make cross sectional forms, and run the strips roughly perpendicular to the forms to get fair curves in all directions.

Here is a link to a good article on laminating veneer into interesting shapes: http://www.woodwork-magazine.com/index.php/archives/86

But for large gently compound curved surfaces, the woodstrip method is probably more appropriate. (There are many books and websites describing the process.)

[ian maybury]

Thanks guys. There's an interesting discussion here too: http://www.woodweb.com/knowledge_base/Compound_Curved_Veneered_Panels.html Signs are that there isn't a straightforward method of veneering 3D/compound curves - short that is of labour intensive coopering/stripping/splicing of veneer pieces. Which has advantages/disadvantages in terms of the appearance too.

Thinking it through it seems to me that the piece about fineply may point to some possible answers - in that the key requirement is probably a veneer that's been processed to form a very fine mesh. One that as a result has some ability to stretch and compress in all directions - much like in the case of a 'high drape' woven glass cloth. The openings/slits would fill with epoxy during lamination. It's not clear what this might look like at the surface, but if the slitting is very fine it might not be very evident once sanded and sealed.

Wonder if anybody has tried anything like this?

[Jamie Buxton]

Are you just trying to form a 3D surface from wood, or are you concerned with the surface appearance after it is formed? I'd presumed the latter, but from your last post, it might be the former, and it would be easier.

[ian maybury]

Preferably both Jamie, although core laminations of veneer could be different to the surface. Another way of looking at it is that (while it might present edge appearance issues) the core laminations wouldn't necessarily have to be wood. I don't have a definite application in mind, other than that I was playing with the possibility of laminating a fairly deeply contoured seat base for a chair.

My thought was just that if the veneer could have fine enough slits in it (would CNC water jet or laser be options?) then they might disappear with sanding and finishing - much like in the way that you can fill the grain on wood by e.g. wet sanding the finish. Soaking to fill any spaces and bonding with epoxy or the like would eliminate any structural issues.

When I say slits I'm thinking maybe closely spaced but alternately pitched rows of maybe 1mm long slits in the grain direction. Then stretch the stuff across the grain (after steaming if needed) to open up some spaces/create a mesh with lateral and longtitudinal 'give'.

This is more by way of a thought experiment/brain storming session than a request for information - unless that is it turns out that somebody has already figured out how to get around these issues...

ian

[Jamie Buxton]

You can get veneer to stretch across the grain a little bit. For instance, if you're making a dome that is 3 feet in diameter and an inch tall, you might be able to make it work. You can help by tapering the ends of each leaf of veneer before you tape it together. When you make the dishing too much, the veneer either splits or crumples.

There are glycerine-based liquids that are used to help flatten potato-chipped veneer. You might try it on normal veneer as an aid to getting it to 3D. Veneer suppliers offer it, but you can easily make your own.

[ian maybury]

Thanks Jamie, I think I've seen a glycerine based solution like that specified and used in a FWW piece on venering....

ian