Grain direction for steam bending

[Prashun Patel]

I am perusing Peter Galbert's Fine Woodworking plans for a modern Windsor Rocker.

The spindles are 5/8" thick and steam bent in one dimension.

I understand the danger of grain run-out along the bent direction. However, is side-to-side (i.e., the non-bent dimension) grain run out also a structural issue?

I have a bunch of thin stock in air dried beech that is wonderfully quartersawn, but the flat sawn face has wavy grain run out.

[Jim Koepke]

I am perusing Peter Galbert's Fine Woodworking plans for a modern Windsor Rocker.

The spindles are 5/8" thick and steam bent in one dimension.

I understand the danger of grain run-out along the bent direction. However, is side-to-side (i.e., the non-bent dimension) grain run out also a structural issue?

I have a bunch of thin stock in air dried beech that is wonderfully quartersawn, but the flat sawn face has wavy grain run out.

With zero steam bending experience my thoughts are just that, thoughts. From some reading and other sources my understanding on this is for parts used in multiples, like spindles, it is always a good idea to make a few extras just in case.

Grain runout is also one of the reasons spindle stock is often riven from a log instead of sawn. Though chairs like those made by Sam Maloof often had flat bent spindles so there are always exceptions to the rule.

It may just be my imagination, but it seems in most steam bending videos the piece is flexed in hand as it comes out of the steam box. My question would be is this done to determine the best way to work any bends?

jtk

[Pat Barry]

Watch this video at about the 5:30 mark

https://www.youtube.com/watch?v=SW6UOXqJi_8

[Prashun Patel]

Perfect. Thanks! It's perfectly obvious now: When you place wood into compression, it's going to look for the path of least resistance to expand. If the grain runs out side/side, it stands to reason it may split that way. Thanks, guys. I am going to pick better stock. I will try to steam the beech slats as a test anyway. For me, I rarely understand what not to do until I've done it myself :(

[Brian Holcombe]

Prashun, Take a look at my posts from the Tea house chair series, I straighten the grain using rift sections for spindles. Same approach applies for bending.

[William Fretwell]

When you bend a piece of wood there is NO non-bent direction or plane just one plane more bent than the others. Lovely long straight grain works very well.

[Prashun Patel]

Thanks, William. Not sure why I didn't see that. I've found appropriate stock.

Thanks, All.

[Mike Holbrook]

There is a reason Galbert uses 5/8” white oak stock, riven from logs. Continuous Arm Windsors require long pieces that are twisted in multiple directions on the one piece of wood. In the class I took, 9 of 10 continous arms came out good. Pete had extras and we just reheated another to get 10. It was surprising to see how successful the teams of two people were at making these complicated bends. I believe the crucial issue was prior planing and prep to get the wood in a good form and bent within 40 seconds, before it cooled, without rushing.

I believe the one that broke had a little run out on the outside of the curve, where a crack occured. I believe the benders got into time issues and were attempting to force the wood too quickly. I would make sure you have a reliable, tested bending form, know how to use it and have a second person to help. More hands may be able to aply pressure more evenly during the bend.

[Sam Murdoch]

There is a reason Galbert uses 5/8” white oak stock, riven from logs. Continuous Arm Windsors require long pieces that are twisted in multiple directions on the one piece of wood. In the class I took, 9 of 10 continous arms came out good. Pete had extras and we just reheated another to get 10. It was surprising to see how successful the teams of two people were at making these complicated bends. I believe the crucial issue was prior planing and prep to get the wood in a good form and bent within 40 seconds, before it cooled, without rushing.

I believe the one that broke had a little run out on the outside of the curve, where a crack occured. I believe the benders got into time issues and were attempting to force the wood too quickly. I would make sure you have a reliable, tested bending form, know how to use it and have a second person to help. More hands may be able to aply pressure more evenly during the bend.

Excellent advice here. Do a trial run so that you can work out any glitches. Lots of clamps & pads - ideally a full length cover - at the ready. The first clamp should be a one hander. Better still, have a permanent lock block so that your bend piece can be set under and held tightly as you then bend to the form. Good hold down - screws or more - to hold your form down, is essential too. Bending can be nerve racking but also great fun. :cool:

Oh yeah, did I say - were gloves? ;)