Bending 3/16 Solid Wood

Other Versions
Spanish
Methods for bending wood to fit an arch. October 5, 2004

Question
If I soak a piece of 3/16" solid wood, will it wrap around this radius? Is there a better way to do this? I will have a piece that is 1" wide and 3/16" thick and I will plow out a dado in it so it caps the plywood in the radius.

Forum Responses
(Solid Wood Machining Forum)
From contributor R:
What is the actual radius and species of wood? Will the curved piece fit on the inside of the curve? I would prefer to glue bend a couple of thinner strips myself.



From the original questioner:
I had to fake the radius. The piece is 18" wide and I came in 2" on each side. From the top of this piece to the top of the arch is 3.25". I then drew a straight line down from there. I'm planning on running a dado in this so it will overlay the rough cutout. I will take it on the spindle sander to sand it to the line before I apply my piece of solid oak. I don't get into very much of the solid wood aspect - I do more cabinetry type items.


From contributor R:
Is there a reason you don't just edgeband this instead of trying to apply a thicker solid edge? Anyway, to answer your original question, I don't know if you can bend a wet piece of solid wood to this radius without steam bending. I have never done either method.

Instead of bandsawing the curve out, have you given any thought to using a trammel base on the router to cut the curve? And as I am writing this, it occurs to me that you may not need to soak the 3/16" strip. If you use a trammel with a 3/16" router bit, you can use the cutout to simply glue and clamp back together with the strip of wood in between the two routed pieces. I actually do this type of glue bending often.



From the original questioner:
I dadoed a slot down the length of the piece so it overlays onto the 3/4". The solid wood piece is 3/16" x 1" with the 3/4" by 3/32" dadoed out. This way there will be a lip of 1/8" on each side. I'm doing this as a detail to this unit.

I probably could get away with edge banding it, which is about what I'm going to do, because I'm getting really frustrated.



From contributor B:
If you use the cutout plywood center as a clamping form, then it should work just as contributor R suggested.


From the original questioner:
Here is what I ended up doing on the larger radius, but in the mean time I'm soaking the pieces of wood to go around the really tight radius like the picture above.

I did do the 3/4" x 3/32" dado.

Here is a closer view.




From contributor J:
Is the reason you are dadoing the solid wood to cover the possibly uneven edge? You run into problems when you dado before you bend - things tend to twist. If you wait until after, you run an increased risk of splitting out unless you climb cut in small passes.

Wetting solid wood does not really make it easier to bend. Green wood bends more easily than dry. Steamed wood bends much more easily. I would tend to glue a few pieces together to make the curve. It takes three pieces to hold a curve, by the way. Two will not do it. My foreman is a boatbuilder and he steams 1 1/2" oak. He would say I'm nuts, but I would say I don't have to guess at springback like he does.

By the way, species is important when you are bending. It looks like you are working in red oak, which is a good bending wood.

If you are looking to make a precision cut, make a template with a router trammel, then saw the straight parts with a carefully stopped table saw, then fettle it in by hand. Use a follow bearing or a template follower to transfer the cut. If you make a mistake that way, you don't screw up your grain match. It's also good if you are making multiples.



From contributor G:
Boil the wood for at least one hour (boiling time, not soaking time) and bend it to a slightly smaller radius than the final shape and let it cool/dry for several hours. That's all!


From contributor E:
Tack on end securely, then with a steam iron, iron and steam the piece as you bend it. You will feel the wood give as it is ready to bend.


From contributor I:
I don't know if you have access to a CNC router, but I would have it machined on a CNC and then possibly use a piece of bending plywood to wrap the radius with, and then laminate it with red oak veneer. The CNC will give you a really nice cut on the wood, and you won't have to cut it again, and bending ply will wrap around anything.


From contributor J:
You are getting lots of good tips here. You could work with any of them.

If you go the solid wood route, bend a longer piece than you need, and use a form with more curve to it than your part actually has. You will find that the ends do not bend as well as the middle does, so you need to make more middle and cut off the ends.




From contributor Y:
3/16" red oak is an easy bend with steam. I have steam bent to form up to 5/16" x 8" easily using some simple steam bending techniques for this type and reason for bend. As mentioned in an above post, species selection is important to bending, as some woods have a natural propensity to resist cracking in a tight radius. This is where steam bending shines.

I use a "steam tube" fabricated using a 8" x 72" metal tube resting (fixed to) on a 5 gallon metal can lid then placed atop the can with enough H20 in it to produce steam for 15 to 30 minutes. The tube and water can are placed carefully over a propane heater (the type used by spacklers and the like to warm a house, although any safe and suitable heat source will work). The heater causes the water to boil, at which time steam is released through holes drilled in the metal can's top. I sometimes put a loose fitting cap on top of the tube (also with holes) to regulate to some degree the water loss. Sometimes it is necessary to flip the banding to steam the piece more uniformly. I will do this after about ten minutes. This can be used in a production environment as well (with the correct safety measures and proper supervision). The eight inch tube will hold a generous amount of banding stock in this length. I use Titebond 1 or 2 as adhesive.

Another note: Producing steam in a sealed container can be extremely dangerous and caution should be exercised at all times if instructing employees to fabricate steam tubes. Failure to properly vent the tube and container can have disastrous results. There is little room in most woodworking shops for creating banding cannons. Always steam in an open area where ventilation is plentiful. Additionally, adding humidity to a controlled environment can have other consequences, such as short term additional electrical consumption (A/C units will run longer), moisture spotting of sensitive veneers (close proximity) and machine rust.



From contributor L:
Yes, it is possible to bend 3/16" thick hardwood with pretty good success. I made snowshoe frames for a class project in college. I steamed the 3/4" x 3/4" blanks for 1 hour prior to making the double bends and had about a 90% success rate when I used straight grained pieces. Another group used strips that had been soaked in a lake for a couple of weeks prior to bending with similar success.


From the original questioner::

Well, the process is done. I don't even want to begin to explain how I made the tight radius pieces work. I had a few words to shout out as it took forever and a few pieces with little success.

Next time, I'll certainly do things differently and charge a hell of a lot more.

The sad thing is that even though it overhangs the face 1/8", you can hardly see it. At this point, I should have just edge banded it and called it a day. All I can say is that it was quite a learning experience.

What really sucks the most is that when the customer looked at the job, he didn't even take note of the solid wood being bent. He looked at it as if they're just arches. Sometimes I wonder why I take extra steps making sure something looks really nice. I could have cut the arch out with a jigsaw, left that edge raw plywood, stained and finished it and he would never have known the difference. On a different note, you never know if one of his neighbors is an ex-cabinetmaker or carpenter who would be able to tell whether or not a good job was done.