Straightening a bowed board

[Adam Cruea]

I debated posting this for a while, but then decided what the heck. . .gotta break the ice somehow.

Anyway, down to brass tacks. . .

I'm trying to build a workbench out of hickory. Much like my father (Dale Cruea), I don't like to do easy things first. :D

So I've got the tressel/sled feet made, they're nice and square. I was working on the stretchers, trying to get those straightened and flat. I finally realized that one of the stretchers had a nasty twist to it, so I got out the twist sticks, figured out how bad it was (I could actually feel my #7 plane rock from side to side over the length of the 48" board) and I ended up fixing that.

However, I failed (somehow) to notice the ungodly bow it had. To make a long story short, the middle of this board bows up about 1/4". When I try to plane it out, it doesn't seem to work, and being a fairly new guy at this, I figured I'd ask some folks who have been around the proverbial wood block.

So. . .any ideas? Should I just toss the board to the side and make a new stretcher, or is my warped, twisted stretcher able to be saved?

TIA

[lowell holmes]

I had the same problem a number of years ago.

I ripped the board down the center. It was a 8/4 piece of walnut, 7' long. I jointed the two ripped sides, then using a dowling jig, drilled dowell holes on 6" centers. I turned the two halves so the bow's were opposite in direction of the bow, doweled and glued the halves together. they compensated each other and made a straight board with no bow.

I did this on two boards and made a 7' door for a bank boardroom I was working on. the door has stood for 40 years.

[george wilson]

Try soaking the CONVEX side with wet,hot rags. Sounds wrong,but it makes the wood spring WORSE,but when it dries,it sucks the wood back further,eliminating,or removing most of the bow. We used this method at Williamsburg to repair bowed 18th.C. table tops,and the like. Sometimes water was applied for days or weeks UNDER plexiglass so it soaked way in before being uncovered and allowed to dry.

[Andrae Covington]

I debated posting this for a while, but then decided what the heck. . .gotta break the ice somehow.

Anyway, down to brass tacks. . .

I'm trying to build a workbench out of hickory. Much like my father (Dale Cruea), I don't like to do easy things first. :D

So I've got the tressel/sled feet made, they're nice and square. I was working on the stretchers, trying to get those straightened and flat. I finally realized that one of the stretchers had a nasty twist to it, so I got out the twist sticks, figured out how bad it was (I could actually feel my #7 plane rock from side to side over the length of the 48" board) and I ended up fixing that.

However, I failed (somehow) to notice the ungodly bow it had. To make a long story short, the middle of this board bows up about 1/4". When I try to plane it out, it doesn't seem to work, and being a fairly new guy at this, I figured I'd ask some folks who have been around the proverbial wood block.

So. . .any ideas? Should I just toss the board to the side and make a new stretcher, or is my warped, twisted stretcher able to be saved?

TIA

First, welcome to SMC.:)

Second, I would ask... so the stretcher bows... other than appearance, does it matter? You could just leave it alone.

Third... (the bad news), this piece of wood could have so much stress in it that you will never be able to remove the bow. You might plane and plane and it will keep bowing and bowing, until the board is down to such a small cross-section that you can bend it back with hand pressure. But then it would no longer be useful as a stretcher.:( Not every tree grows straight; it may have bent or twisted towards the available sun, or because of prevailing wind direction. It can be very difficult or impossible to overcome that kind of distortion that has "set" over the life of the tree.

Fourth... in attempting to remove the bow... because the natural tendency when planing is to remove more at the beginning and end of the stroke, I would first attack the concave side of the stretcher, using short strokes at each end, working down to the low spot which is in the middle of the span. After getting that side straight, I would use that as a reference face for a marking gauge to scribe a parallel line for the convex side. On the convex side I would start with short strokes in the middle, which is the high spot, and slowly begin to start and end further away from the middle, taking longer strokes, as you remove the bow.

[Matthew Hills]

I'd probably work across the grain with a coarse cut if I was trying to remove material. But I've never been really successful with badly warped wood -- it usually moves a bit more an hour after I thought I'd flattened it. The water treatment above might work better.

Matt

[george wilson]

Andrea is correct: The bowed wood may have too much stress in it to ever come straight. Try my suggestion first. If no luck,get better wood.

We had to have a 6 foot piece of beech wood to make a cooper's jointer in Williamsburg. We took a 6"x6" and planned it true and square. It twisted like a propeller in several days. We let it acclimate for months,and re planed it true. twisted again. Things were getting interesting. We waited months,re planed it. Same deal. The billet was getting smaller!! After SEVERAL months and several planings,we got it down to the size we had to have,a 4x4". For some reason,it magically finally stopped twisting and we made the cooper's jointer. We had had the wood in the shop for over a year by the time it was done. No telling when wood will behave.

[Adam Cruea]

I probably should have mentioned, this piece was originally ~8 inch wide 4/4 hickory I ripped down to 4 inch wide, then glued the pieces up to make an 8/4 piece. I guess I missed the bow and the chance to straighten it out then.

I'll try your suggestion, George. The worst that happens is I'm out a 48" x 4" piece of wood. But, at least I learn something. :) Question though. . .do I want to put thoroughly soaked rags (dripping wet), or wring them out until they're not dripping wet and just heavily moist?

Andrae> That's what I've tried, just in the opposite order. (Convex first, lower middle, then concave). But now that I think about it, that might weaken the convex side to allow the concave side to bend more, yes? Also, I had thought about the "meh, so it's bowed option" since when I do the mortise and tenon joint for it, I'll try and draw the tenon into the mortise with tapered dowels. I got to thinking, though, that it may not either pull the bow out, or it may cause some sort of shifting.

Thanks for the suggestions, y'all. After work today, I'll probably see if SWMBO has some old rags I can soak in hot water and slap those on the lumber.

[george wilson]

Just try to keep the concave side dry.

[Bob Jones]

Make sure the board is fully supported with shims. Otherwise it will flatten with the pressure of planning and not get flat.

[Dale Cruea]

Welcome to the creek son. You have just gotten solid advise from a few of the best here.
BTW... if George says do it. Get er done.

[Pat Barry]

My 2 cents = find another board. Georges idea sounds a bit off to me. I think in the case he was citing, the boards were straight at one time and became bowed due to use conditions and maybe because they were improperly finished. I suspect his method was able to get the boards straight again. Yours on the other hand sounds like a lost cause, although you could try to just rip it to a bit narrower width and hope for the best. Trim a little from boh edges.

[Jessica Pierce-LaRose]

Unless I have a piece of irreplaceable lumber, or a pressing need to use it in a project, like I need this board because it matches perfectly or whatever, I always tend to let the lumber win and use another piece. My recipe for bowed boards is to use them where I need shorter pieces - less bow to remove if I'm dealing with a 4 foot piece of a two foot piece rather than an 8 foot piece.

However, if your stretcher is straight enough to do the job it's asked to do, I'd think about just having a bowed stretcher, particularly if you only tend to work on one side of your bench (if it's against a wall or something) - the front stretcher is nice to have flat and true if you use it as a clamping surface, but I can't even really get to the rear stretcher on my bench, let alone clamp to it or support work against it.

[Harlan Barnhart]

I have seen several videos of woodworkers in Japan straightening bowed wood with heat and pressure. On both occasions they used the piece immediately after the heating and bending. One carpenter used an electric heater, a break to secure the board and hand pressure. The other example was a commercial "wood press" that heated, possibly steamed, and pressed at the same time. It held the board for about a minute or so and then the board was ready to use. Does this work only on Japanese wood or are we so wasteful with our abundant wood resources that we can buy thick and plane off the waste?

[David Keller NC]

My 2 cents = find another board. Georges idea sounds a bit off to me. I think in the case he was citing, the boards were straight at one time and became bowed due to use conditions and maybe because they were improperly finished. I suspect his method was able to get the boards straight again. Yours on the other hand sounds like a lost cause, although you could try to just rip it to a bit narrower width and hope for the best. Trim a little from boh edges.

George is correct on this, with some qualifications. One can straighten a bowed panel or board that is going to be captured in a frame by wetting the concave side, warming (and thus drying) the convex side, and when the bow is removed, immediately capturing the board or panel in a construction that will hold it flat until throughly dried. One can also use this technique and immediately place the formerly bowed piece under pressure (such as stacking and stickering it under weight) and wait for it to thoroughly dry. This is essentially the principle behind steam bending.

However, what George is saying is a valid technique in the case where one cannot immediately capture the bowed piece in a frame or under weight. It's based on the fact that the binding between the wood fibers (i.e., lignin and ray cells) isn't nearly as strong as the fibers themselves, and that the fibers will actually crush and not fully return to their originally diameter. When you wet a side of a board, the wood fibers swell (which takes up more room). That will make the bow worse. But when the fibers dry back out again, they shrink, and if the board is held under tension, the binding between the wood fibers will rupture. In an extreme case, the rupture will make itself evident as a crack.

This effect is why almost all antique tables that saw a good deal of use and cleaning with water will be dished (that is, the edges will be curled upward). When the table top was wetted, the fibers on that side expanded. Since the table top was held firmly to the rails, the fibers actually crushed when they were wetted instead of expanding in width. When the table dried again, the fibers again shrunk, but since they were crushed in the process, their new dried dimension is actually less than what it was originally. Thus, the edges of the table curl upward (or the table top cracks).

[Steve Elliott]

The original post referred to a board that was bowed rather than cupped. (Bow is curvature along the length of the board; cupping is curvature from edge to edge.) George's method of wetting the convex side could be effective for correcting the bowed condition.

Most lumber ends up with internal tension due to the drying process. Typically the outside of the board is in compression and the center is in tension, that is the wood fibers near the surface are being pushed together lengthwise and the fibers in the center are being stretched. This is why ripping a board down the middle often causes the wood to pinch the sawblade. Each resulting piece of the board has a slight curve toward the (former) middle of the board.

When George's technique is used, prolonged soaking of the wood fibers on the convex side raises the moisture content above the fiber saturation point and the fibers are able to slide past each other the way they can in a flexible tree branch. (Compare this to a dry branch which is brittle and snaps easily instead of bending.) This releases the compression on that side only and the board bends toward a straighter shape. When the fibers dry out they lock together in the new position and the board is both straighter and has less overall tension.

Kiln dried wood is sometimes given an extra "wet" cycle after it is dry to release the compression from the outside surfaces. When it is then re-dried it has almost no internal tension.

[Mel Fulks]

The method does indeed work; but before George's post I had only heard of it used on cups. Mostly table tops. It came to my atension in an article written by Bob Flexner. Your explanation of it is more complicated than Flexner's. I don't enough about it to say if it is essentially the same or not.The condition,( before the fix) is commonly called " compression ring set" for anyone trying to find more information.

[Jack Curtis]

...Kiln dried wood is sometimes given an extra "wet" cycle after it is dry to release the compression from the outside surfaces. When it is then re-dried it has almost no internal tension.

Steve, is this what steaming is all about?

And thanks to you, Dave Keller, and George, who've increased my knowledge appreciably.

[Steve Elliott]

Steve, is this what steaming is all about?


I've seen steamed cherry, which I believe is done to even out the color between the heartwood and sapwood. Other than that there's steam bending, which relies on the principle of increasing the moisture content to the point where the fibers will slide, then holding them in the desired position until they dry out again.

[george wilson]

Wetting the convex side was applied to a mahogany, VARNISHED table that had not been wetted with wiping over its life. The varnish was original and intact.

I missed where you said the wood had been GLUED UP. This technique might not work if you glued up the wood. It might also not work anyway. Depends upon conditions in the wood that caused it to bow. It was standard procedure on 18th. C. antiques in the furniture conservation lab in the museum,which was pretty much World class.

[Jack Curtis]

I've seen steamed cherry, which I believe is done to even out the color between the heartwood and sapwood. Other than that there's steam bending, which relies on the principle of increasing the moisture content to the point where the fibers will slide, then holding them in the desired position until they dry out again.

Yes, sure on the bending; but I wondered if steamed drying, even if it is to even out color and the like, also has some effect on keeping the boards straighter than plain kiln drying.

[Adam Cruea]

Well, I took Andrae's advice and flipped the board over and planed the bowed up ends. After about and hour of planing and checking for flat and square, I finally saw what was giving me so much grief. . .there was an abscess in the wood and a knot. I noticed right before I'd gotten to it that the grain was starting to go every which way and that my plane seemed to be getting stuck there near the end, or at least heavily dragging.

So, off to the trash pile that puppy went. I worked the second stretcher flat and pretty square (one end falls off, but since it's getting sacrificed as a tenon I don't really care much) and that took all of about 3, maybe 4 hours.

*sigh* Lesson learned. . .pay more attention to the grain and any possible knots. The guy at the lumber store even pointed it out to me. *shaking head* I just forgot it.

[Steve Elliott]

I wondered if steamed drying, even if it is to even out color and the like, also has some effect on keeping the boards straighter than plain kiln drying.

Most of my knowledge of kiln drying comes from Bruce Hoadley's book, Understanding Wood. On page 98 he describes a typical drying schedule and mentions the use of increased moisture near the end of the process to reduce casehardening, which is the type of tension I've described.

I don't have personal experience with a kiln but do think Hoadley's book is very good.

[Mel Fulks]

Different culture ,in Japan some practices are encouraged and even subsidized that would have you fired (or committed ) for suggesting them here. They have a commitment to continuing traditions that just does not exist in the West. We wait for old ways to disappear completely then pay for research to bring them back . Niether culture is going to change anytime soon.

[Jack Curtis]

Most of my knowledge of kiln drying comes from Bruce Hoadley's book, Understanding Wood. On page 98 he describes a typical drying schedule and mentions the use of increased moisture near the end of the process to reduce casehardening, which is the type of tension I've described.

I don't have personal experience with a kiln but do think Hoadley's book is very good.

Thanks, Steve, I've only scanned Hoadley; but I'll definitely study the drying schedule part.

[David Keller NC]

The original post referred to a board that was bowed rather than cupped. (Bow is curvature along the length of the board; cupping is curvature from edge to edge.) George's method of wetting the convex side could be effective for correcting the bowed condition.

Oops - maybe I should read the OP a bit more carefully next time. Your explanation makes more sense, though I've been told that fibers in wood won't actually slide past each other - the lignin binding and ray-cell structure prevents that. Instead, I've been told by a Forestry professor at my local University that branch-bending is more about stretching the fibers, which they will do when wet in a tree.

One interesting note - "Comp Wood" is a product that by nature of its processing has apparently broken the ray cell structure and lignin binding between the fibers, so it is highly flexible until dried out. I've never tried it, but it looks intriguing (and expensive).