Ripping and Re-Gluing to Limit Cupping
The Wood Doc weighs in on the old-timer technique of ripping and re-gluing boards in order to lessen moisture-related cupping. February 19, 2008
I was taught as an apprentice that ripping a flat-sawn board down the middle and gluing its edges back together in the same sequence that it came apart helps to reduce the board's natural tendency to cup due to tangential movement and shrinkage. In my mind's eye I can't quite believe this, but that's what the old timers always claimed. It seems to me that the board will still react the same way as if it had not been ripped and re-glued at all... What do you think?
(Architectural Woodworking Forum)
From contributor B:
The reason that a ripped and re-glued wide board will cup somewhat less than the original board is because the two pieces will now work as individual boards. So, a 12" wide board prior to ripping might cup 1/4". However, if you rip it into two 6" wide boards, each of those would cup less than half of the 12" board's cup... so maybe 1/16"+ on each. Putting the two 6" boards back together doesn't change how they will work individually. The theoretical result is that instead of one "u" shaped board, you will end up with a "w" shaped board, with less cupping on the two sides of the joint.
From contributor J:
I was always told to flip the boards after ripping. This way one would cup up and the other down, therefore minimizing the amount of movement in either direction. I too would think that if you ripped a board and glued it back together, it is going to act as a single board again. Although I'm no expert, I'm not sure I see how the board will cup in two pieces into a "w" shape after being glued. Seems like the glue joint would have to fail to allow that to happen? Hopefully the Wood Doc will chime in on this one and shed some light.
From contributor O:
I'm with contributor J (and the questioner's suspicions) on this one. I don't buy the idea that cutting a board and gluing it back together with the pieces in the same orientation will have any noticeable effect on future cupping. It is, however, a repair strategy for a board that has already cupped. You can rip a U-shaped board into strips and then plane the cut edges so they go back together in a W (or triple-U, or...) shape that is, overall, much closer to flat than it had been.
From contributor B:
Even though I supply the theory above, I still go back and forth on this a issue bit. However, unless the edge of the individual boards angle upward with the cup, you will get a "w" shape. Imagine letting the two halves cup on the bench, and then jointing the edges (cupped away from the jointer fence so it is riding the two points). Then when you glue them together, it is clearly going to be "w" shaped.
Back to the original question of gluing the two halves together, though... If the edges remain perpendicular to the cupped face, then you will end up with a single "u" shaped cupped effect, as is easy to picture. If the edges don't change angle, though, you will get the "w" effect just like with the halves that were jointed after cupping.
Bottom line, though - no matter how the glue up is done, the two boards will act independently either side of the glue line. The cut interrupted the continuous ring configuration as it was cut from the tree.
Personally, I flip boards to get opposite grain orientation. There are many that call this a waste of effort, and it does make it more difficult to get grain continuity on the face of the panel. I also have seen many wide cabinet doors made with a single 12"+ panel out of one uncut wide board. Most have been flat.
From contributor A:
In my experience, ripping a board down into two strips and re-gluing does prevent warp and cupping. It obviously does not limit normal expansion and contraction. I like to think of woods like poplar that are prone to warp or woods like African mahogany that have a lot of reaction wood (saw binding?). When you take a 16" wide board of poplar or mahogany and rip it into four 4" strips and independently flatten them and then glue them back together, it will be much less prone to cup or twist versus the original board.
From contributor O:
I'm only aware of two general reasons a board will cup. Both involve moisture content changes. One reason is that, thanks to the structural restrictions imposed by ray cells, wood tends to shrink more tangentially than radially; this causes cupping in flatsawn boards, as well as other distortions in boards with other sorts of ring orientations. This sort of cupping would be moderated by the rip and flip technique.
The other general reason involves environmental factors; one side of the board is exposed to something (rain, direct sunlight, a heat source of some kind) that causes it to pick up or lose more moisture on one side of the board than the other, causing cupping similarly to the way a bimetal strip in a thermostat works. Ripping and re-gluing, in any orientation, won't help this at all. The rest sounds like old wives' tales to me.
From contributor B:
No one is saying that ripping/regluing will stop cupping. The discussion is on how to minimize its effect on the project.
From contributor O:
I get that, though the questioner asked about whether cutting and re-gluing "helps to reduce the board's natural tendency to cup." That's exactly what I'm responding to. Contributor A's suggestion that cutting and re-gluing can relieve internal stresses and perhaps avoid future warp makes some sense, but the idea that boards that have been cut and re-glued act independently of each other isn't clicking for me. Like others, I hope Gene might clarify some of this.
From contributor C:
The natural tendency for any given specie of wood to react to changes or season conditions will not change no matter how many times it is cut. The cutting does very much reduce the tension in any given cross section whether it is a 1 1/2" rip or a 4" rip and many of the previous posts do point out many good points about wood's reaction to atmospheric conditions after being killed and harvested. I have a strong desire to make my wood joints in furniture and millwork as relaxed as possible. Basic acclimatizing, double sealing all backs and undersides, superior and correct jointing for flatness, are all contributing factors as well as everything mentioned.
From Professor Gene Wengert, Sawing and Drying Forum technical advisor:
For some reason, the idea that contributor B expresses has been around and continues to surface. However, if an individual piece is then immediately re-glued back to the same position as when it first existed (minus the saw kerf), it will cup as much as it did without being ripped. That is, the glue rejoins the fibers so that the piece behaves the same as it did without the glue. Many years ago, we actually did a test and showed the above to be true. So, I can attest to the validity of my statements above 100%, both through theory and practice.
As mentioned, for cup to occur, there must be a moisture change (usually drying in practice).
I believe that the story originates because if you rip a piece and then dry it (or wet it) and then re-glue it, it will, if the glue jointed edges are prepared to be 90 degrees to the general shape of the each piece, turn into a "W" shape when it is glued back together (immediately). Note, however, that if you re-glue them after drying using the original ripped edges (maybe cleaning them up a bit), then you will have the same cup as without ripping. (Technically, because of drying set, if you do this with wet lumber, there will be slight differences.)
Now, you will have less cup to remove by planing or jointing if you flip the one piece end for end and then glue, as mentioned.
The discussion about tangential shrinkage being greater than radial and this causing cup is true. In fact, the bark side of a piece of lumber will shrink or swell more than the other side, so when drying, the piece will always cup toward the bark. It will cup away form the bark face if being wetted. (I am assuming the drying or wetting is done to both faces equally.)
In practice, when edge gluing pieces together to form a flat top that will stay flat, the adjacent strips that are re-glued may be oriented with the bark side of the pieces alternating up and down. But if there are no significant moisture changes, it will not make any difference.
In this ripping event, it is important to understand that there may be longitudinal stress in the unripped piece that will cause immediate warp lengthwise when ripping and make it difficult to have to flat parallel edges to re-glue. You can straighten the lumber pieces with clamps when gluing, but the pressure on the joint will be excessive in some places, leading to a weak joint in those places.