Steam Bending in Reverse?

[ErnieM]

So I'm in the process of bending a harpsichord bridge. The hard maple bridge is 60" long and tapers in height and width from 1/2" in the front to 11/16" in the rear. I thoroughly wet the bridge with water and then wrap it in wet towels for an overnight soaking. The next morning, I put it in the steam box for 45 minutes. After removing it from the steam box, I secure it to a drying jig and let it dry for a month.

A month later, when inspecting the bridge, I realized that I used the wrong drying jig. The jig I used was for a smaller instrument - one more instance of encroaching senility. However, since the bridge can be bent somewhat during installation, I hoped I could persuade the bridge to follow its intended path. No such luck! Facing the decision to make a new bridge or trying to save the old one, I decided I had little to lose by trying to save it rather than replace it. I set up my steambox and found that I could still get the front end of the bridge (the part with the severe curve) into it. So I repeated the process - wetting the bridge and soaking it in towels overnight. The next morning, I got the steam box up to the proper temperature and removed the bridge from the soaking towels. I was surprised to find that the bridge had almost completely straightened out from the overnight soak. It now fit all the way into the steam box. After a 45 minute baking, I put it in the correct drying jig and it looks like all will be well.

​

I'm amazed to realize that, after building these instruments for over 30 years, I still capable of finding new ways to make stupid mistakes. There appears to be no limits to my ability to find new ways to make life interesting. Still, I never thought that a steam bent piece would straighten out almost completely the way it did, but it's comforting to know that a mistake like this can be re-bent. I'm wondering if a thicker piece of wood would react the same way to water.

Ernie

 

[KenOfCary]

Amazing story Ernie. We all learn something from mistakes, don't we?

 

[ErnieM]

Amazing story Ernie. We all learn something from mistakes, don't we?

That's true. I'm just not sure what I learned.

 

[KenOfCary]

If nothing else, you learned the bending is reversible. So did I - didn't really think that was doable. Thanks for sharing.

 

[Jim Wallace]

Well, Ernie, if you want to talk about mistakes that are related to age (or maybe carelessness) here's one:

​

Check the 5th spindle from the left of the end panel. This was fitted, glued, clamped, checked for sanding, and being finished before I noticed it. Fortunately, the side panels will eventually be the head and foot boards of a full sized bed, and the end panels will not be used.

As far as steam bending a piece twice goes, most of my reading says that the second time is most likely to result in failure. Mostly the things I am researching have related to making chairs, however, so we're thinking of thicker pieces than you have, and often it is a situation where someone doesn't get the bend he wants and tries to heat the piece back up to get a tighter curve the second time. Even in a situation where this bend could be made in a previously unbent piece, the piece that is bent twice will fail much more often.
Also you are soaking your piece. Most of what I have read discounts the moisture content of the piece being bent: what is important is that the piece gets to 210 degrees or so all the way through at which point the lignin becomes plastic. In this case the piece only has to cool to keep its bend. I usually take mine out of the form the next day. By soaking the pieces, you're sort of combining steam bending with green wood bending. Many woods can be bent green and allowed to dry in the form after which they will retain the bent shape. With either of these methods the strength of the piece is virtually the same as if it had grown in the shape you wanted.

 

[Jeff]

I'm in the very beginning phase of steam bending and learning about it so I'm curious.

Hard maple is one of the absolute worst woods for steam bending so it flexed while bending and sprang back about 95% after you removed it from the form. Will it hold that form after rebending or spring back again?

Maybe the curve/offset/radius of the bend is mild enough that the maple will hold the curvature? You've done this before with "thin" hard maple strips?

 

[ErnieM]

I'm in the very beginning phase of steam bending and learning about it so I'm curious.

Hard maple is one of the absolute worst woods for steam bending so it flexed while bending and sprang back about 95% after you removed it from the form. Will it hold that form after rebending or spring back again?

Maybe the curve/offset/radius of the bend is mild enough that the maple will hold the curvature? You've done this before with "thin" hard maple strips?

The goal in making harpsichord bridges is to get the wood to bend and stay bent long enough to glue the bridge down onto the soundboard. There is a lot of spring back when using maple, and the risk of breakage is higher. That said, I can't remember the last time I had a failure. Trying to compensate for spring back by introducing a more severe curve in the form is futile and unnecessary. Even if I was capable (and I'm not) of determining the amount of spring back to compensate for, the severity of the bend would no doubt cause failure. As long as the curve in the bridge is close enough to be persuaded into position during the glue up, all is well. Although I do tend to allow for a really long drying period, I've often removed the bridge long before gluing it down, and I've never had a bridge fail to hold most of its curve - nothing even close to reverting to its unbent shape as happened when I re-soaked the bridge this time.

 

[Jeff]

Interesting observations from your long time experiences and thanks.

So the rebent piece can be removed from the form, glued and clamped to the soundboard without a springback?

 

[ErnieM]

Interesting observations from your long time experiences and thanks.

So the rebent piece can be removed from the form, glued and clamped to the soundboard without a springback?

Oh there is plenty of spring back, but not so much that the bridge can't be bent by hand into the right position. Once it's firmly glued down, it's not going anyplace. When I take this re-bent bridge off the form, I'll take a picture and show you how much spring back there is. Keep in mind that I'm dealing with a long, but relatively thin piece of wood. Except for the area of severe curve at its front end, the rest is quite easy to bend into position by hand.