Steam Bending kiln dried Gen. Mahogany

[sean m. titmas]

The back rest for my chair has a shallow bend @ 20"+/- radius. the stock measures 4/4 x 9" x 28".

i plan on steam bending the wood and would like to get some info about length of steam time,
how to calculate for spring back,
and weather or not i can use kiln dried lumber or if i need to use air dried because of the change that kiln drying has on the bending and setting of wood.

if any of you guys have experience with this i would like to hear from you about this project. thanks

[harry strasil]

Never worked with mahoganey, but tried to steam bend straight grained kiln dried ash once, all it did was break, try a small piece to see what happens, you may have to submerse it in boiling water to get it to bend without breaking if its kiln dried. I watched a show one time where they did it that way in the hills of Tenn or Arkansas. Full concave and convex dies front and back may make it bend easier and hold its shape better after removed from the dies

[Nate Carey]

Steam bending goes hand-in-hand with traditional boat building. Stock that is intended to be steam bent is often green and never more than air dried. Once "bound water" is removed from the cell structure by kiln drying, much of the natural flexibility of the timber is lost; exchanged for rigidity and strength.

[Frank Drew]

Sean,

For some curved work, I like bent laminations because they're fairly predictable (i.e. low to no springback); you can use either bought or shop-made veneers and do the pressing either in a vacuum bag with a one piece form, if you're set up for it, or between a pair of forms with a forest of clamps.

[george wilson]

Might be better to make a bunch of 1/8" strips and laminate. If you can make consecutive "flitch" cuts in mahogany,the glue lines would be hard to see. Don't use yellow glue,of course!! I think the hardest to see glue would be liquid hide glue. Soak the bottle in hot water for a while. The glue gets MUCH thinner,and easy to squeeze out of the laminations for a thin glue line. Try the glue on a sample first. Check the expiration date in the store. Old new stock hide glue may not dry.

[sean m. titmas]

bent laminations would work fine as long as when i start to shape the edges of the piece that the glue lines dont show up. the profile will cut across the glueline at a 45deg and im not sure how visible the glue will be. the SU drawing is a cross section view of the back rest profile.

[Jay Jolliffe]

If your going to do the bent lamination use unibond 800 with the dark catalyst so you don't see the lamination's.

[David DeCristoforo]

Oooo...bad idea. When steamed, kiln dried mahogany gets hot and wet but not flexible (OK, guys, just ignore the "double entendre"). And green mahogany is not going to be the easiest thing to find!

[sean m. titmas]

Oooo...bad idea. When steamed, kiln dried mahogany gets hot and wet but not flexible (OK, guys, just ignore the "double entendre"). And green mahogany is not going to be the easiest thing to find!

thanks dave, i was going to try bending it until i read you hot and wet comment. you saved me a lot of frustration.

when i was requesting quotes for the wood the only air dried Mahogany i found was sold in 1000bf minimum so kiln dried was the only option. and now i guess doing bent laminations will be the best option.

[george wilson]

I meant to make laminations and bend them to the shape all glued together. Are you talking about gluing up a wide board and sawing the bend out of it? That would expose lots of glue seams.

Correct me if I am wrong.

I have made many many guitars with kiln dried sides bent to shape around a hot pipe,and sometimes dipped in water first.They were only 1/8" thick,though,however,bent around much sharper curves than you are considering.

[David DeCristoforo]

"I have made many many guitars with kiln dried sides..."

Maybe not the same. Don't you bend guitar sides on a hot pipe? Thin wood, much more heat and not so much moisture don't you think? But my "reading" of the OP's question was that he was asking about much thicker wood (4/4) in a steam box. That's something I have had no success with using KD mahogany.

[Frank Drew]

George,

Perhaps I'm wrong, but I think Sean meant that he plans to cut a slightly rounded bevel on the top of the rail once it's curved (by whatever method) and is concerned that with a lamination that might really highlight the glue lines. Like when you put an acute bevel on plywood.

[David DeCristoforo]

"...with a lamination that might really highlight the glue lines..."

If the lamination is done with consecutive layers, all cut from the same board and glued up in the same order in which they were cut, and the glue lines are tight, the glue lines will almost disappear. This is much more of an issue with a denser wood like maple or cherry but with mahogany being as open pored as it is, the glue lines will not be nearly as prominent.

[george wilson]

Do you think I don't know that?;)

[sean m. titmas]

George,

Perhaps I'm wrong, but I think Sean meant that he plans to cut a slightly rounded bevel on the top of the rail once it's curved (by whatever method) and is concerned that with a lamination that might really highlight the glue lines. Like when you put an acute bevel on plywood.

you got it Frank, that is exactly what i planned to do.

to stack the deck in my favor i'll use a glue with a darker color and make sure i use plenty of clamping pressure at the perimeter of the lay up

[David DeCristoforo]

Sean... try this. Since the curve of your piece is going to be a "true" radius, cut your form template using a router compass with a 3/4" straight bit. That way, you will have a perfectly matching two part form. Then mill your layers so that the stack is exactly 3/4" in net thickness. When you clamp the layers between the two halves of the form, you will get full "clamping pressure" over the whole surface of the piece. With thin layers, it is very hard to achieve this with a "one piece" form.

[Frank Drew]

It's not a bad idea to line your forms with a thin layer of a slightly resilient material, to compensate for any irregularities of either form or laminates.

I've laid out two-part forms from full-sized drawings; with 3/4" material you might not notice much difference, but the two parts of the form won't nestle perfectly one within the other because the outer, convex face of your curved piece will be a section of a slightly larger circle than the inner, concave side. This really becomes apparent with thicker pieces: The male and female forms for say a two-inch curved leg will be noticeably different in curvature.

[David DeCristoforo]

"...but the two parts of the form won't nestle perfectly one within the other..."

If you use the technique I described above, the two parts of the form will have different curves. They will be "offset" by the diameter of the router bit (in this case, 3/4"). But if you mill six layers of material, all 1/8" thick and place them between the two form halves, the forms will "close" perfectly with no need for "resilient material, to compensate for any irregularities". But you are correct in stating that the two halves of the form will not "mate" if "closed" without the 3/4" thickness of the material being placed between them.

PS I just finished laminating some curved inlays for a '55 chevy dashboard. Each piece is made up of nine "consecutive" layers of purpleheart veneer, which has a grain structure very similar to that of mahogany. (These were pressed in a vacuum press on a "one piece" mold, not between two form halves. FWIW, I did not suggest vacuum pressing for the OP's chair backs because 1/8" layers will be too stiff to bend in a vacuum press.) The pieces "net" 1/4" thick and are rounded on all edges with a 3/16" radius. No glue lines are detectable, even on close inspection.

[Frank Drew]

"...but the two parts of the form won't nestle perfectly one within the other..."

If you use the technique I described above, the two parts of the form will have different curves. They will be "offset" by the diameter of the router bit (in this case, 3/4"). But if you mill six layers of material, all 1/8" thick and place them between the two form halves, the forms will "close" perfectly with no need for "resilient material, to compensate for any irregularities". But you are correct in stating that the two halves of the form will not "mate" if "closed" without the 3/4" thickness of the material being placed between them.



I probably didn't express myself clearly, but I wasn't arguing with your method, or saying it wouldn't achieve the same correctly curved forms (which won't nestle perfectly when empty, which is what I meant). Of course, unless you have a straight router bit for every possible part thickness you might be working, another method for laying out forms is also useful to know.

As for the form lining I suggested (e.g. 1/16" adhesive backed latex or other rubbery material), everybody should do as they wish, but if a form is over a certain thickness it will probably have to be freehand cut on the band saw; same for curves that aren't part of perfect circles. Shop-made veneers cut on the band saw might also have some "irregularities". Perhaps not necessary, but not a terrible idea in any case. That's all.

[David DeCristoforo]

Hey Frank... I'm not trying to contradict you. And you are correct in stating that the method I described will not work for an curve that is not part of a "true circle". It's just that the OP stated a "20"+/- radius" which implies that, in this case, it is.

As to forms over a certain thickness, I usually build those up a layer at a time using 3/4" thick plywood or MDF (one of the few things the stuff is good for). I make the first layer as "fair" as I can, then screw subsequent layers to it, flush trimming each one in turn. This avoids the need to bandsaw a blank that is several inches thick which will, as you stated, invariably involve some irregularities.