Steam Bending, Part IV

This the fourth part of a very long explanation and description of steam bending.  I cannot run the whole part at once, and too much goes on around here to run it over consecutive weeks.  Therefore, I am posting it as I can.  If you are only starting to read my blog, you may want to search for Parts I, II, and III and start there.   Mike Dunbar. 

If the steam box did not have some relief, the test caps on either end would be blown off as  pressure developed in the tube.   We make a 1/2 inch relief hole in the bottom of each end of the tube.  The steam enters the tube in the middle and travels in both directions, escaping out the vent holes.  This ensures an even distribution in the box.   When our boxes are running at full steam, a plume of water vapor blows down from the ends of the tubes  to the ground.  We can look out the classroom window and tell how things are going in the bending area.  

Because heat rises, the parts do not sit on the bottom of the box.  There, they would also be bathed in cooler condensed water.  Instead, they  rest on a rack made of stainless steel bolts that pierce the tube.  Regular steel bolts would leave dark purple stains on the parts.   

As little as 15 minutes is all that is  required to plasticize  red oak chair parts.  This time can vary somewhat according to the circumstances.  The  wood we use is freshly cut before each class.  That means our  bendings  already  have 25% moisture content and that all we need to do is heat them.  If your wood has been stored for a while and allowed to air dry, it may be around 14%.  You should steam a bit longer, perhaps 20-25 minutes.     

Chairmakers who live in the Rocky Mountain states or at other high elevations,  have another problem.   Water boils at a lower temperature the  higher you are.  I ran into this problem in 1980 and 1981 when I taught  chairmaking classes at BYU in Provo, Utah.   We managed to bend successfully by steaming our parts even longer than usual.  We left them in the box 40- 45 minutes.  Because the wood was  noticeably cooler than at lower elevations, we worked even more quickly.   

Adding fabric softener to the water in the boiler has been suggested in some  woodworking magazines as a way to soften the wood’s fibers and make it bend more easily.  I have tried that trick and have not found that it makes any difference. 

Before removing a piece of wood from the steam box, be sure that  every thing else is prepared and ready.  You only have about 45 seconds to complete the bends before the wood becomes too cool, and you do not want to waste any of that time fumbling.  Always check the form to make sure it is securely clamped to the bench.  Be sure   that the bending strap is easily accessible.  Check that  you have the required number of wedges and pins,  as well as a hammer.  I always go through this mental checklist. For beginners, I recommend  refreshing the bending process in their minds  by first pantomiming it.   

Remove the wood from the box with a pair of tongs.  We use the ones sold in a super market for picking ears of corn out of boiling water.   Before you open the steam box remember what  you learned in  your high school physics class.  Released steam will rise.    To avoid a burn, always open the box  and approach it with the tongs from below.  

Remove the part from the box and moving quickly,  place it in the bending strap and secure it to the bending form.   While you want to move quickly when setting up the bend, remember that as you bend,  speed is your enemy.   You must give the wood time to compress.  If you move too quickly, you shift the outside edge of the bend from compression into tension.  While wood compresses very well, it has a very limited ability to stretch.  Tension will cause failures.  Avoid it by bending slowly and deliberately.  

“Do you wear gloves?” is a common question asked by students.  I advise against them.  As you gain experience bending you will discover that you can sometimes feel problems in time to correct them.   We will sometimes feel a piece beginning to weaken and turning the part around, bend  it successfully in the other direction.  Roll up too, can be felt by a pair of experienced hands.  You will never develop this “feel” for the wood if you wear gloves.   

While steamed wood is quite hot, you can juggle it back and forth while carrying it to the form,  and can switch hands while bending.  The wood needs to be this hot and if it is not, you will experience more failures. I always tell students, “If you’re not swearing,  it’s not hot enough.”  

You  cannot tell if the wood has been sufficiently heated when wearing gloves.   Furthermore,  bending requires  dexterity.  Gloves make your hands too clumsy.   While we have few bending failures at The Institute, they do occur.  Failures are a fact of life that a chairmaker has to accept, as bending wood  is an art, not a science.  Four types of failure can occur when bending. They are:  delamination, tension shear, compression failure, and roll up.  

Delamination is by far the most common failure.  In this case, a layer of wood peels off the outside edge of the bending.   A  tension shear occurs when the wood fibers rupture, tearing like cloth across their width.  In a compression failure, the wood on the inside of the bend fails to compress evenly, and kinks up like ribbon candy.   Roll up occurs most commonly in pieces with a rectangular section, such as the sack back arm rail.  In this case, the part  does not remain in a plane as it is bent.  Instead, the rear edge rolls upward.   

A failure is not necessarily catastrophic and can frequently be fixed.  The difference between it  going into the scrap pile or into a chair is usually a matter of degree.  If it is not too large,  a delamination can be tacked back into place with glue.  It is best to wait a day or two for the wood to dry a bit. 

A small  tension shear can be consolidated with cyanoacrylate glue, which with a sufficient number of applications will actually fill the void.  We saturate small compression failures with cyanoacrylate and when it is dry,  smooth the crinkled, ribbon candy effect  with a plane or spoke shave.  We remove roll up by clamping the bending flat between two boards and setting it aside to dry.  

The first three failures can frequently be prevented by bending in a slow, steady motion.  As I said above, speed is the enemy.   Roll up results from one of two causes.  If the part’s edge against the bending block was not made at a right angle to the upper and lower surfaces, the part’s section  is a parallelogram rather than a rectangle. In this case, the rear edge will lift as the front edge is pulled tight.

If you would like to receive periodic updates, tips, tool reviews, and new sources, that are outside the scope of this blog, join our mailing list by emailing me at  Help us spread the word about this blog. Tell others.