13

I recently made a tabletop for a rolling tool cart using 2x4's jointed and pocket-screwed. When I set the table upon the cart I noticed a very serious bow, almost like I was starting to make a barrel!

What methods are there to prevent this from happening during a fastening or glue-up scenario?

|improve this question
  • 1
    I think look at the grain direction and alternate it every other board (as mentioned by rob) is a vital part of any of the answers provided. – FreeMan Mar 19 '15 at 14:55
  • With properly jointed edges you shouldn't have any major bowing. First check I would do is confirm your jointed edges are at a perfect 90°. If they're dead-on then look at the pocket screwing, that could be introducing a slight bend into each joint which over the width of the top adds up to a large bow. – Graphus Apr 5 '15 at 9:06
15

A common solution is to alternate your clamps between top and bottom:

alternating clamps for a panel glue-up

Another solution is to clamp boards down across the panel as Drew suggested, although an even better variation on this solution is to use cauls, which start out curved but evenly distribute the pressure across the surface as they are flexed flat against both sides of the panel.

panel glue-up with cauls

Read more about cauls: http://www.familyhandyman.com/woodworking/how-to-clamp/view-all#step2

But even if you use both of the above solutions, your panel can still bow and warp with changes in humidity, so it's important to let it acclimate first, as Matt suggested, by letting it sit for at least a week or two, assuming the lumber was already kiln-dried or adequately air-dried. If possible, let it acclimate to the environment in which it will ultimately be used--for example, if it will be in your dining room, let it acclimate inside your house.

You can also minimize movement by using quartersawn lumber and cutting out the pith. If you're stuck using the typically low-quality lumber from a home center, the best you can do may be to look at the grain direction and alternate it every other board, so one board will cup downward, the next will cup upward, the next downward, etc.

|improve this answer
  • 1
    I am not sure if its implied and I don't think this would stand out as an answer but it is worth mentioning the wood should be allow a period of time to normalize, in terms of moisture content, in an environment. I cant find the link but someone making a workbench out of beech left the wood for 6 months in his shop before gluing it. – Matt Apr 27 '15 at 15:10
  • 1
    @Matt thanks, I incorporated your comment to make it more explicit. – rob♦ Apr 27 '15 at 15:27
11

I always clamp jointed boards with some good, straight hardwood boards across their faces. This way, you'll use long clamps across the face of the jointed boards to keep your joints together, and shorter clamps to clamp your truing boards to the face. Use a pair of boards with the work piece between them. When it sets up, it should not be bowed at all.

here is a maple table top I clamped this way

|improve this answer
  • 3
    In that case, it's a good idea to put an "insulating layer" (a sheet of paper or some felt or any such thing) between the hardwood boards and the work pieces. It sucks big time if you accidentially glue the hardwood onto the work piece. Doesn't happen too often, but when it happens it really, really sucks. – Damon Mar 19 '15 at 9:46
  • @Damon It happen to me once, it does sucks. I now use steel square tubing with a film to protect my work. – Maxime Morin Mar 19 '15 at 13:59
  • I would also recommend turning your 2x4 up. You want the tallest profile possible to prevent deflection in the caul board. you can see a slight bit of deflection, because when your clamps are off the glue piece, as in this picture, you are squeezing the ends of the caul boards together where nothing is between, and in turn, it will bow the caul up in the middle, off the intended piece to be clamped. The secondary way to combat this is to actually used slightly curved cauls, as mentioned above, with the belly of the caul in the middle toward the glue piece. – Jacob Edmond Sep 27 '16 at 12:14
  • Wax paper works well as the "insulating layer". Normal carpenter's glue won't stick to it very well, so it should be easy to peel off if some squeeze out does cause sticking. – FreeMan Sep 7 '18 at 14:03
3

No one has mentioned your jointer technique. If the jointer fence is not perfectly 90° then any small deviation get multiplied by the number of boards you have and you get a curve. The solution to this is NOT to try to get your jointer fence exactly 90° (although close is good). Instead, lay out your boards how you want them to be glued up, and draw a giant "V" across the table top with a pencil, marking every board. The V should span the entire top. Now flip every other board over. Now label the top face of each board with an "F". Now joint each board with the face labeled "F" towards the fence of the jointer. Now lay out your boards again, flipping every other board so that you reconstruct the original "V" you drew. The result is that the deviations from 90° alternate direction, and so cancel one another out, and viola! Flat top, not a barrel. If the deviation from 90° is extreme, the clamping pressure might tend to make the boards slip out of plane with one another, and so cauls, as others suggest, are still a good idea. But honestly, if you're anywhere close to 90° and have true joints you should require minimal clamping pressure and slipping shouldn't be a problem. YMMV.

|improve this answer
  • Properly jointed edges are obviously a prerequisite here and should have been mentioned in previous Answers, either perfect 90s or complementary angles (89 + 91 or whatever). One thing though re. clamping pressure, with the principal glue used in woodworking today very high clamp pressure at the joint face is vital for the strongest joints. – Graphus Aug 21 '18 at 18:36
0

An alternative is to make T-braces (my name). Screw two inch boards into a T shape. If the board edges are straight and the screws long enough, the brace will stay straight under very great pressure. Its the same principle as a torsion box. Then clamp 2 of them face to face with the table between them.

|improve this answer
  • 3
    Would you mind adding some pictures? – rob♦ Jun 10 '15 at 1:14

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.