Steam bending/laminating maple--might it work?

[GarF]

I'm making a small staked chair for one of my girls, and to extend my skills I'd quite like to steam bend the crest rail rather than carving it out of a large blank. Since this is my first attempt at a chair I'm using up some maple which has been sitting on a shelf for several years since I decided against the project I planned to use it for.

The cross section of the rail will be 2½" x ¾" which I very much doubt I could bend. However, I thought it might be possible to resaw/plane down to ¼" and then steam and laminate a stack of three.

Firstly is there any reason this approach would be doomed to failure (before I commit to prepping the stock)? And secondly any recommendation for best glue for this operation? I've got titebond 1, titebond3, some titebond liquid hide (out of date) or resintite (cascamite?) on the shelf, have already discounted gorilla PU as inappropriate. Oh, there's some West epoxy up there too.
Cheers
G

 

[Mike Jordan]

I can't see any advantage in combining the two methods of bending, but laminating the desired shape would work well. It would be better to make up the thickness with at least 6 layers of thinner laminate and preferably more. The thinner and more numerous the laminates the better. Cascamite is ideal for the job. You need to allow for some degree of spring back when the piece is removed from the formers after the glue has set.

 

[Phil Pascoe]

You might find the wood a little old and dry to bend as it's been on your shelf for years.

 

[custard]

In your favour is the fact that your crest rail doesn't have to conform to a precise curve, the beauty of windsor/welsh/staked chairs is that you've got a lot of wiggle room. So a bit of spring back, or a former that isn't curved exactly as you might wish, aren't too serious.

Against you is the fact that even fairly rough and ready curved lamination work is still quite taxing, and in truth it's quite a machinery intensive operation.

First issue is that 1/4" lamina is really pushing it, I normally work in the range 1mm-3mm. Here's a crest rail I laminated that's at about 1.5mm thick lamina, if the radius of your crest rail looks similar then that's what you should be thinking about too.

You could cut these on a table saw, but that will be extremely prone to kick back. So almost every one I know cuts these on a bandsaw at about 2.5mm thick initially. Here's my set up for doing just that,

It looks easy, and indeed it is...provided every detail is dead right. You need a bone dry donor board that's quarter sawn with arrow straight grain, a bandsaw that's perfectly set up in every respect, a tall fence with zero drift, a very sharp blade, an ultra slow feed speed, and you also need a bit of luck! Get all that sorted and your band sawn lamina will roll off without a problem. Miss even a single detail and it'll almost certainly end in tears.

In between each bandsaw cut you need to plane the previously band sawn face of the donor board flat. But even then your lamina will have one planed face and one band sawn face. You can't laminate up bandsawn lamina because the rough bandsawn face would leave shockingly ugly glue lines. So you need a plan on how to smooth off the lamina and also bring it down to finished thickness. The easiest solution is a drum sander, like this,

Without a drum sander you could feed the lamina on a sled through a thicknesser. Although if the lamina are curling, and they often are, then they'll just get shredded. And to pull this off you also really need a thicknesser with rubber feed rollers rather than toothed metal cogs. For just a single component like a crest rail it's viable to plane and scrape all twelve or so lamina by hand, but then again you need the hand tool skills to get each piece accurately flat, if you don't you'll end up with ugly, gappy glue lines.

After getting this far you have to think about making a former. The best glue is a UF adhesive like Cascamite. But that needs a great deal of pressure to force as much squeeze out as possible to keep the glue lines thin and regular. You might get away with a single former, especially in a vac press, but for the very best results you'll need precisely cut male and female formers like these,

You can understand from all this why generations of craftsmen preferred steam bending!

 

[GarF]

I can't see any advantage in combining the two methods of bending, but laminating the desired shape would work well. It would be better to make up the thickness with at least 6 layers of thinner laminate and preferably more. The thinner and more numerous the laminates the better. Cascamite is ideal for the job. You need to allow for some degree of spring back when the piece is removed from the formers after the glue has set.

All good points. Didn't fancy my chances of steaming the maple in one. Since I'm working with hand tools, I figured a quarter inch would be a reasonable compromise between thinness and labour. My gut instinct was that that is still too thick to push into shape as Custard has said and illustrated, and hence I was hoping that by steaming the lamina I would improve the odds of them taking to the curve.

For the former I have a stack of three quarter inch chipboard from an old GPlan wardrobe. I wasn't exactly sure whether the 'male' and 'female' surfaces should be shaped to the same radius or be offset by the thickness of the rail. It ought to look like Custard's when complete. While I can see the difficulties inherent in this approach I think it will be interesting to see how it goes. If it's not great then plan B is to carve the rail instead. I've set myself a goal to use up the timber I've got on the shelf already before buying any more as a means of freeing up some space to organise the garage better- otherwise the next best option would be to buy in construction veneer and laminate that.
Thanks all
G

 

[Beau]

I cant see a problem in combining the two techniques. Done the same myself for making a very tight radius curve. I would reconsider your take on PU glue though as it's the most excepting of higher moisture contents which you may get post steaming unless you wait a fair time for it fully dry out again. Also PU like West Sytem and Cascamite give you long open times which is a bonus with awkward glue ups.

This is a support for the neck of a guitar with 1mm veneers

 

[custard]

I wasn't exactly sure whether the 'male' and 'female' surfaces should be shaped to the same radius or be offset by the thickness of the rail.

For the best results you want uniform pressure with no pinch points, that means offsetting the radius of the formers by the lamination thickness. If you can arrange for the lamination thickness to exactly match a copy router diameter then there’s a neat trick to doing this.

 

[sunnybob]

Personally, I dont see what the fuss is about...
https://www.youtube.com/watch?v=R_cyx7t ... sCoachShop


:shock: :lol: 8) (hammer) #-o

 

[Yojevol]

I've just had the steaming or not conundrum myself with my lipping on my elliptical table. I am using 6off x 5mm laminae in American Walnut. I was afraid the they would fail around the sharp end tight radius. I therefore attempted to steam the first one to soften it a bit. Unfortunately it curled across the width making it impossible to clamp flat on the MDF substrate. I may well have left it in the steamer too long.
With the next attempt (unsteamed) I did a trial clamping and it proved OK, so off I went and it all worked out well. See photo below:-

I hope to post the complete job next week
Brian

 

[thick_mike]

I just lost a couple of hours watching those coachbuilding videos. Great stuff thanks

 

[sunnybob]

I wish I could find the utube I started out looking for. I saw it a couple years ago but cant remember what it was called.
Some art students had a baulk of timber must have been 6" x 8" over a bench and they bent that thing round into a horsehoe, just with steam and a dozen or so willing helpers.

I have bent small pieces of wood using dry heat (electric paint stripper gun) but dont think that would work for a chair back.

 

[GarF]

I wasn't exactly sure whether the 'male' and 'female' surfaces should be shaped to the same radius or be offset by the thickness of the rail.

For the best results you want uniform pressure with no pinch points, that means offsetting the radius of the formers by the lamination thickness. If you can arrange for the lamination thickness to exactly match a copy router diameter then there’s a neat trick to doing this.

Thanks Custard, I'm assuming that's to run the guide bearing along a template to create both complimentary curves in the workpieces below simultaneously, starting with the pieces separated by slightly less than the width of the cutter?
G

 

[custard]

I'm assuming that's to run the guide bearing along a template to create both complimentary curves

Yes, that's right. You start by gluing up a stack of MDF to the required thickness, usually 100-200mm. You then get a short, bottom cutting, bearing guided router bit and run it around a template exactly as you figured out. So then you're left with a curved channel, 10 or 15mm deep, in a 100-200mm glued up stack of MDF. So you bandsaw along the centre of that channel, and then use the clean section of the curve with either another bearing guided router bit, or more likely a spindle moulder with a ring fence, to smooth off all the bandsawn mess to create two perfectly complimentary curves. Yet more evidence of what I said earlier, good quality lamination work tends to be fairly machine intensive.

It's interesting to compare steam bending with lamination in the context of how you'd go about solving the practical problems that each method throws up. For steam bending you have to figure out how to source fairly green, ie wet, timber, how to raise steam, how to make a solid timber former, and how to apply the necessary pressure to form the curve. A hundred years ago none of that would be particularly difficult. But today sourcing green, or even air dried timber, is in itself an insurmountable problem for most woodworkers.

Contrast that with laminating. You need very specific adhesives, a fair amount of well set up machinery, plus a stable material like MDF is pretty useful when you need both male & female formers. A hundred years ago that would rule out laminating for almost everyone, but today any professional workshop and a large number of hobbyists could realistically tackle these challenges.

It's interesting how two alternative techniques, with similar end results, can be ruled in or out by social and technological change.