Steaming wood

[ohowson]

Ok, so I've never tried this before. My missus wants a hammock stand for her birthday in a few weeks. I thought I might try and make one - seen a lovely one that looks like 2 pieces of 75mm square birch, bent to a gentle radius, for the arms, then another two at the bottom the same with the arms sandwhiched between them. I've build at steam box out of an 8x4 sheet of 50mm insulation (so it's roughly 12" square), connected it to a pot and heater to generate steam, and it certainly is: I have steam going in one end and coming out of the other - win.

Only the piece of 1/2" by 1" scrap wood has been in there for about 2 hours and isn't wet through yet!

do I need to keep the steam in there longer by letting less out the other end? Is it too big for the wood and should I be trying to reduce airspace inside (won't help with the bigger wood of course). Am I cruising for a bruising trying to bend 75mm square wood in the first place? Or should I just keep going and give it longer and longer and longer....?

 

[Keith 66]

75mm square is about 3" so its a big ask to steam if its square. Steaming always goes better if timber is steamed on the flat ie wider than its thick. Square stuff this large will likely need a compression strap on the outer side.

 

[Cabinetman]

I think you’ll find you’ll get on a lot better if you use green wood as well, that was the way it was traditionally done. Only ever done it once successfully and that was with green oak, just one recommendation and that is to plane the surface that is going to be the inside edge of the curve before you bend it, it will make life so much easier later. Good luck. Ian
Edit you will know when it’s ready as it will be floppy like a bit of wet spaghetti-ish

 

[Droogs]

It isn't the amount of steam that matters but the temperature that the wood gets to. this is what softens the lignum and allows you to bend the wood. I have a 5 injector common rail that I plug into my steam box (made from plumbing pipe) so that I get an even amount going in the entire length of the box. I also use 3 wallpaper strippers if doing anything over 2 1/2" thick. You also want lots of insulation to keep the heat in as long as possible. In the past I have had to use two layers of celotex around the box. You may have to leave it in there for around 4 hours to get the temp up long enough to penetrate that amount of wood.

 

[Cabinetman]

It isn't the amount of steam that matters but the temperature that the wood gets to. this is what softens the lignum and allows you to bend the wood. I have a 5 injector common rail that I plug into my steam box (made from plumbing pipe) so that I get an even amount going in the entire length of the box. I also use 3 wallpaper strippers if doing anything over 2 1/2" thick. You also want lots of insulation to keep the heat in as long as possible. In the past I have had to use two layers of celotex around the box. You may have to leave it in there for around 4 hours to get the temp up long enough to penetrate that amount of wood.

Hi Droogs, it sounds like you have been successful in your steaming and bending could I persuade you to do a bit of a sketch of your machine please? I have a complex project in mind and would prefer to get it right first time. Many thanks Ian

 

[Droogs]

It's in bits for storage but when i go down the wksp this evening I'll take pics of the bits of the rail and pipe so you get the idea and do a sketch as well

 

[GarF]

Bent lamination as an alternative option?

 

[stuckinthemud]

Submerging the timber a week or two prior to steaming really makes a big difference.

 

[Daniel2]

Just a plus one for the timber being green.
I don't have a steam outlet, as such.
I have the steam box on a slight tilt, where the condensate runs
back down into the steam generator (wall paper stripper).
It runs for hours without a top up.
Almost, but not quite, perpetual motion

 

[yetloh]

Agree about the need for insulation. The one I use is a piece of smooth stainless flue liner about 6" dia. packed around with loft insulation in a 12" square ply box with doors and decent rubber seals at each end, connected at the top door to a wall paper stripper. Also a 6mm drain tube through the box and liner close to the lower end. A really useful refinement is a small tube through the top face of the box and liner to accommodate a digital thermometer probe (less than £10). With this you know what is going on in the box - wait until the probe gets close to 100 deg c before starting steaming.

Green or at least air dried timber is best and you need a ring porous species (oak or ash have been best for me) with straight grain parallel to the wood - this is much more easily obtained if the wood is riven with an axe rather than sawn. If you want a thick component, this is much easier if you laminate from thinner stock - glue up with Polyurethane glue which uses the water in the wood to cure it - most PVAs and aliphatic glues (eg Titebond ) won't make a reliable bond if the moisture content is much more than 15%. Also glue-lam will have little or no spring back, unlike solid steam bent which can be significant and inconsistent.

Jim

 

[ohowson]

I think I have to accept the steaming isn’t going to work for this - just getting the wood was proving 3x the cost of buying one and as my landrover is broken I have limited ways of transporting something that heavy. I’m now considering using laminated glued ply and trying to figure out how to build a big enough jig.... and what size I can bend!

 

[No1retired]

Agree about the need for insulation. The one I use is a piece of smooth stainless flue liner about 6" dia. packed around with loft insulation in a 12" square ply box with doors and decent rubber seals at each end, connected at the top door to a wall paper stripper. Also a 6mm drain tube through the box and liner close to the lower end. A really useful refinement is a small tube through the top face of the box and liner to accommodate a digital thermometer probe (less than £10). With this you know what is going on in the box - wait until the probe gets close to 100 deg c before starting steaming.

Green or at least air dried timber is best and you need a ring porous species (oak or ash have been best for me) with straight grain parallel to the wood - this is much more easily obtained if the wood is riven with an axe rather than sawn. If you want a thick component, this is much easier if you laminate from thinner stock - glue up with Polyurethane glue which uses the water in the wood to cure it - most PVAs and aliphatic glues (eg Titebond ) won't make a reliable bond if the moisture content is much more than 15%. Also glue-lam will have little or no spring back, unlike solid steam bent which can be significant and inconsistent.

Jim

This sounds ideal for a project I have in mind. Any chance of a photo of your setup?

 

[Trainee neophyte]

I made one last year, and it was so much fun I made another one to go with it. 10mm laminations, and make sure you use a foaming polyurethane glue rather than pva because of glue creep. I know all about glue creep from some bookshelves that were perfectly square for the first couple of years but are now parallelograms.

You will need some more clamps.

 

[baldkev]

For those with experience.....
If you were gluing up say 10 strips at 10mm thick, would you be best to alternate the annular rings or pay attention to the grain directions? Or is it not important? Im thinking in terms of reducing the chances of movement
Thanks

 

[Droogs]

Mostly for me, if the part will be visible then appearance of the grain and minimising visible joint lines is uppermost and reliance on the strength of the glue is high, the chances of failure due to movement in modern resin glues is slim.

 

[Cabinetman]

For those with experience.....
If you were gluing up say 10 strips at 10mm thick, would you be best to alternate the annular rings or pay attention to the grain directions? Or is it not important? Im thinking in terms of reducing the chances of movement
Thanks

I draw a line across before I saw the wood up and then glue it up so that the line is still there my reasoning is that there is more chance of it looking as if it was originally one piece of wood just bent.

 

[Adam W.]

It's a shame you've given up on the idea of steam bending.

I was interested in doing a bit for the mahogany canoe I have as a restoration project and came across this method, which I thought was very clever.



The guy has a series of videos which are really good sources of tricks for doing curved work, which is unsurprising really as boats are just one mass of curves all bundled together to form a floating object that you can sort of steer sometimes.

 

[Adam W.]

An even bigger bit...

 

[stuckinthemud]

Loved that video, brought back lots of memories, "boil in the bag" was what the shipwight I used to work with always used, even on huge timbers, now I use a much smaller version on longbows, its very effective.

 

[Adam W.]

He's got a great presentation style, very deadpan.