Replies

1. 
   JeffKurtz | Mar 14, 2003 12:43am | #1

   Sorry, but with the grain oriented vertically, there really is no way to avoid all that movement.  The only thing that can be done is to rebuild it with the grain running horizontally, or just live with it as it is.  (These sort of mistakes are costly lessons, but we never forget them.)

   Jeff

1. 
   Curtis | Mar 14, 2003 02:54am | #2

   : 

   GJ131

   I am sorry to hear about your problem, ITS THE NATURE BE THE BEAST!

   BUT if you know how the wood is going to react you can work with it, exp. use molding to cover up the woods movement, it will still move but noone will see it. or use bread board ends to hold the project together, STUFF LIKE THAT.

   plan on the wood to move and figure out how to work with it.

   sincerly C.A.G.

   P.S.

   also cut down your wood into widths of no more than 3"  and then glue them back together it will cut down the woods movement greatly.

   SEE YA!   C.A.G.

1. 
   StanleyNiemiec | Mar 14, 2003 07:08am | #5

   "also cut down your wood into widths of no more than 3"  and then glue them back together it will cut down the woods movement greatly."

   The individual widths of the pieces in a panel have absolutely no effect on the overall expansion and contraction across the width of a panel.  The only thing that individual piece width has on panel performance is to minimize warp.

   Also as I read what various individuals write, I am surprised that no one recognizes the fact that miter joints open and close with changing moisture content.  I might suggest that the "square" is now distorted to a parallelogram  --  that the total stress on the four corner joints was relieved by the weakest joint.

   My suggestion would be to convert the sides (by cutting them down) into floating panels that fit into an outside frame.

1. 
   HowardAcheson | Mar 14, 2003 07:22am | #6

   At this point I'm not sure any of us understand the construction. While you may be correct, I see nothing that that suggests to me that miter joints are part of his construction.

2. 
   Curtis | Mar 14, 2003 01:19pm | #7

   "The individual widths of the pieces in a panel have absolutely no effect on the overall expansion and contraction across the width of a panel. The only thing that individual piece width has on panel performance is to minimize warp."

   Sir. we are going to have to disagree on this one.

   wood expands more across the grain, then it does a long the length of it. Doesn't it make sense if cutting your wood down to 3 inch pieces and than gluing them back together, would not only help stop warpage, but to minimize as well the wood's ability or need to expand and contract across its grain?

   Perhaps Howie is correct

   "At this point I'm not sure any of us understand the construction"

   without actually seeing the project everything is just conjecture.

   I still stand by my origanal staments .PLAN on the wood to move and try to work with IT!

   Sncerly C.A.G.

1. 
   jjf | Mar 14, 2003 03:42pm | #9

   Given a set width and the same grain orientation a panel made of many pieces glued together will shrink the same as a panel made of one piece.  If the orientation is changed to all radial from a mixed or all tangential grain direction then yes the shrinkage will be less.  The idea of by cutting the board up into pieces releases the shrinkage is a falicy.  Please explain how it makes sene that shrinkage is less after you cut it up and reglue them together again.

   Edited 3/14/2003 9:01:19 AM ET by jim

2. 
   GJ131 | Mar 14, 2003 06:14pm | #10

   Whoa! Lots of interesting responses on this one.  I'll be more detailed on the construction.   The corners are mitered and they look o.k. so far.  The  lid is completely removable, no hinges or attachments.  No breadboard on the top of the sides.  The bottom:  If it were hanging on a wall it would look like a picture frame.  The cube sits on top of the "picture" frame.   The cube is glued and screwed to the bottom frame. I didn't allow for movement because....well, I'm a neophyte.   The more I think about it and the more I read your responses, the more I realize there's more to this than I thought. 

   The inside bottom is a piece of 1/4" plywood.  

   Be that as it may, it looks like one of the glue lines is what let go.

3. 
   HowardAcheson | Mar 15, 2003 08:11pm | #13

   Now, with this info we can hazard an opinion on the problem. It sounds like you have a cross grain situation with the bottom. In other words you have a rigidly attached bottom piece running across the grain of your side pieces. This prevents the side pieces from expand/contracting with humidity changes and something has to give. In this case, it was one of your glue joints.

   The easiest way to deal with it is to attach the cross pieces with slotted holes so the side boards can expand and contract without restriction.

   The above is based on assuming that the 1/4" plywood bottom just rests on the cross boards.

4. 
   StanleyNiemiec | Mar 14, 2003 06:56pm | #11

   Based on the information I obtained when I got my Master of Science degree in Wood Science and Utilization (Penn St. '79) and in all subsequent readings, it is apparent to me that the sum of expansion across 16" of width will be more or less the same and it does not make a whole lot of difference if the individual pieces of the glued up panel are 1", 2", 3", 4",  5" or whatever.  Grain orientation (ie whether the grain is radial or tangential) may make a difference because grain orientation provides differing coefficients of expansion but otherwise 16" of wood, changing dimension with changing moisture content is still 16".

   The formula that applies (obtained from the Wood Handbook [USDA Forest Service]) is as follows:

   Dimensional Change = Di [CT or R[(Mi - Mf)]

   Where

   Di is the intitial dimension

   CT or R  is either the tangential or radial coefiecient based on grain orientation

   Mi is the initial moisture content (percent)

   Mf  is the final moisture content (percent)

   The radial coefficient for Black Cherry is 0.00126 and the tangential coefficient for B:ack Cherry is 0.00248.

   If you are using all flat sawn pieces, the applicable coefficient will remain the same and it doesn't make any difference if you calculate the width of the individual pieces and sum them, or just calculate the total sum of widths.

5. 
   Curtis | Mar 15, 2003 03:40pm | #12

      NIEMIEC1

   After receiving your message, I went back to my library of books on this topic, trying to find the information necessary to prove you wrong. Unfortunately I was not successful, I stand corrected. (I was  the one who was wrong) that is as far as my statements on the cutting the of wood down into smaller elements to stop the woods expansion. (I guess is time for me to eat crow) the cutting of the wood into smaller elements does relieve internal stress make in the wood more stable, but it does nothing to stop the woods movement.

   Thank you for correcting my error.

   Sincerely C.A.G.

2. 
   HowardAcheson | Mar 14, 2003 03:48am | #3

   If I understand you correctly, it sounds like a edeg glue joint opened up. Is that what you are saying? Also, I understand that all the grain on the four sides is running up and down. Correct?

   There is nothing wrong with that type of construction. You have no cross grain situations as I understand it.

   So, if the glue joint let go, it was just probably not glued or clamped correctly. Without knowing more about what you did, I can't guess as to the cause.

3. 
   Clay | Mar 14, 2003 04:18am | #4

   It is a better construction to join the wood on the end grains for the four sides of the box.  The bottom (or top) should be set in a slot like a frame and panel system that allows some room for the expansion and contraction of the bottom (or top).  Nothing you can do about the existing problem except maybe salvage some of the wood for a smaller cube (of better design).  You can get away with making such a cube solidly joined at all corners if you use plywood.

4. 
   Froed | Mar 14, 2003 02:18pm | #8

   As I understand your description, the grain on the sides (and front and back) runs parallel to the 13-1/2" height dimension, correct?  Like Howie said, I don't see a cross-grain situation.  But how are the top and bottom attached?  If the top is just a flat piece attached with a couple hinges on a side with the grain parallel to the edge, you have a cross-grain situation, which will probably result in the glue joint at the top of the box failing, or a hinge failure or split back.  If the hinges are on the side perpendicular to the grain, it's oriented the same way as the back, and the hinges will move with the wood.

   Is the bottom a solid piece of wood?  If so, it should be rabbeted in with enough room to allow for expansion and contraction of the sides.  Over a 16" width, there could be over 1/8" reduction in width across a flat sawn piece as it loses moisture.  If the bottom is glued all around, there is no allowance for movement, and as the wood shrinks a crack will open up starting at the bottom of the box.

   Now, with the grain of the sides oriented horizontally, the only issue is the bottom and having a strong corner joint, so you can see why it's more typcal.  But there's no reason vertically oriented grain won't work, and it's more unconventional, which is reason enough to do it.  It just requires a bit more thought.

   As for gluing up smaller pieces, they will minimize warpage, not expansion across the grain.  If you have a 16" piece that'll move, say, 1/4", you can cut them into 4" pieces that'll move 1/16".  But when you glue them together, you'll have 4x1/16" = 1/4", which is why I don't like cutting wide pieces down, unless they have a significant cup in them to begin with.

   But then, with nothing to look at, this is all just a long-winded guess.