LumberJocks Woodworking Forum banner

Chess Board Buckled

8K views 25 replies 17 participants last post by  Andybb 
#1 · (Edited by Moderator)
Hello Lumberjocks,
I recently built a chessboard table for a friend. Everything went great and I was very happy with it. Then something went wrong and the chess board portion of it has buckled. Here are the details:

The board is 2" squares of maple and walnut that are 1/4" thick because I was trying to save money on the wood and re-sawed the 3/4" thick wood to 1/4" thick. The chessboard portion was made using 2" strips of alternating wood, edge glued, then cutting those panels into 2" thick strips running crossways with the different woods alternating to make the squares. The chessboard was then glued onto a piece of 1/2" birch plywood. The plywood was cut 1/4" wider than the chessboard on all sides. This allowed me to fit the plywood into dadoes that were cut into the table top frame that is also 2" wide all the way around. Once the table top was built, I used 3 light coats of shellac as the finish. (Top Image)

My friend sent me a picture today showing that the center "strip" is buckling upwards from the table and the center glue joint for that strip has split apart. (Sorry about the sideways image at the bottom.)

I live in the high desert near L.A. where temps this summer were in the 90's to low 100's. Our normal humidity range is around 7%. My shop is outside in the garage. We then had about a week of temps near or over 100 and 30% humidity. My friend has a swamp cooler for the house, which I'm sure added humidity to the air around the table.

So, any ideas as to what I can do, aside from rebuild the chess table? I'm sure several of you see glaring mistakes in the way I built this and I would appreciate your kind advice as to what those mistakes may be as well.
Table Rectangle Chess Sports equipment Wood


Wood Water Wall Art Rectangle
 

Attachments

See less See more
4
#2 · (Edited by Moderator)
The mitered frame is preventing seasonal
expansion. You'll need to remove the
frame and the buckled section may relax.

A better way to go about it is using veneer
squares. Solid wood has greater need to
move while veneer is thin enough to
stay stable on a plywood substrate.

In my family there is a dining table with
solid wood panels set into frames. The panels
have beads around the edges which serve
to camouflage the fact that they are actually
raised panels set in grooves. Still, junk
does get down in the small gap adjacent to
the bead. The table is over 100 years old
and still in very functional condition.
 
#8 ·
Rustin congratulations on breaking so many woodworking no no s on one project.Dont beat yourself up just make a new one the right way.
Many of us have made woodworking mistakes including me.:)

- Aj2
And me. I made a humidor years ago. It was a thing of beauty, with dovetailed corners and Spanish cedar lining. Problem was, I laid the cedar base in without room for expansion and whenever it's humid enough inside for cigars, that darned base spreads the joints…lol

It sure does look nice dry, but it's not good for cigars at all.

Short of having a mentor, stuff like that is what's needed to learn.

Look on the light side. At least you're not a highway engineer with buckled pavement and miles of backed-up commuters hating you because you didn't make the expansion joints wide enough.
 
#9 ·
rustin,

Sorry about the buckling, it really is a nice board otherwise!

For my own knowledge, as I'm intending to build a board and/or chess table later this year, would taking Loren's suggestion of using veneer over plywood (but otherwise using the same building method rustin did) be sufficient to prevent any issues?
 
#11 ·
rustin,

I agree with Loren that the chess board expanded and buckled upward since the expansion was restrained by the mitred frame. Fixing the problem would be a lot of work but may be possible.

The idea would be to release the chess board from the table, provide for some expansion of the maple and walnut that is fixed to the birch plywood substrate, and then re-attach the chess board to the table so that the chessboard sets atop the mitred frame, rather than inset in the frame.

Releasing the chess board from the frame could be done with a router and a straight bit. A groove would be routed in the mitred frame along the edge of the field of the chess board deep enough and wide enough so that the birch plywood substrate can be lifted upward and freed from the table.

Once freed from the mitred frame, the birch plywood could be cut flush to the walnut/maple squares around the perimeter. Since the walnut/maple squares are glued to the plywood and the plywood may resist expansion, kerf cuts in the plywood substrate could be introduced. The kerf cuts would be deep enough to cut through the plywood but not so deep as to cut into the walnut/maple squares. Each kerf cut would run parallel to the grain of the walnut/maple squares ideally at the joint where one row of squares meet the adjacent row of squares.

Two edges of the chess board could then be edge-banded with a hardwood strip so that the hardwood strips run parallel to the wood grain in the walnut/maple squares. The edge-glued strip would be as wide as you like but wide enough to cover the grooves routed when releasing the chess board from the table and thick enough to conceal the plywood. The edge-banding would be just long enough to be flush with the ends of the chess board. However, the edges on the ends where the end grain of the walnut/maple appear would receive no edge-glued strips.

At this point the chess board is ready for re-installation into the table, but the table must be modified to support the chess board. The support could consist of three or four stretchers that run from one side of the mitered frame to the other. Some means is needed to secure the stretchers, but without a better understanding of how the table and frame were constructed, I am of little help in identifying options for this attachment. Perhaps rabbets on the ends of the stretchers could work. Once installed, the up-side surface of the stretchers should be flush with the upper surface of the mitered frame.

Before installing the stretchers, stopped through slots would be cut into the stretchers wide enough to accept screws. The slot will accept screws that will install into the plywood chess board's birch plywood substrate and hold the chess board in place while allowing for expansion. Pan-head screws could be used, but if the slots are chamfered, bugle head screws could be used and the screw heads would set flush with the underside of the stretchers.

At this point, the chess board can be installed in the table. The chess board would be position on the stretchers so the stretchers run perpendicular to the wood grain direction in the walnut/maple squares. One screw though the slot in the stretcher into the center of the plywood substrate between each kerf cut would provide the greatest support, but perhaps skipping every other space between the kerf cuts would be adequate.

The last step before applying finish would be to add trim strips to the mitred frame to cover the end-grain raw plywood edges. These strips could be the same width as the edge banding and long enough to be flush with the edge banding previously glued to the chess board. There should be no glue in the joint between the chess board and mitre frame installed trim strips.

The result would be an elevated chess board free to expand and contract.
 
#12 ·
JBrow,
Thank you for the advice. I plan to cut the chess board out of the frame, hoping to keep the frame in the same condition with the inlay so that I don't have to rebuild that.
From your suggestions, I'm getting that I need to raise the board above the frame to allow for expansion. I plan to create a new board that is glued together the same as the old board.
This leads me to some questions: First, is gluing the entire bottom of the board to the plywood base an issue? Or would I be better off putting a small dot of glue in the center of each square or perhaps 4-5 dime sized glue spots just to hold it to the plywood?
Second, how much space between the board and the frame do you think I should allow for this expansion? 1/16th all the way around? 1/8th? I can stain the plywood bottom to darken the space and make it blend with the frame, but if it's much more than 1/8th it will become an eyesore for the project. The squares are 2"x2" currently. I was thinking of setting the table saw at 2" and then add a piece of masking tape to the rip fence. This would shave a small amount off the strips, but when glued together as a board, could "shrink" the board enough for a 1/16th space. Your suggestions, again, would be appreciated.
Rustin
 
#14 · (Edited by Moderator)
With all the grain running in the same
direction a 16" square chess board may
move up to 5/16" in width. Alternating
grain direction would in theory bring
that movement down to 5/32" in both
directions. With grain going all one way
I still think the board is at risk of buckling
as it fights the stability of the substrate.

Another approach would be to use something
like waxed pasteboard shimming between
the tiles while gluing them down. The shims
would be removed and a fine gap perhaps
1/32" wide would remain between the squares,
allowing for movement.

I've never made a chess board but I'm
sure there are some articles and posts
on the web about solving movement
problems in this particular situation.
 
#15 · (Edited by Moderator)
The chessboard was then glued onto a piece of 1/2" birch plywood. The plywood was cut 1/4" wider than the chessboard on all sides. This allowed me to fit the plywood into dadoes that were cut into the table top frame that is also 2" wide all the way around.
Gluing the chessboard to the birch prevented expansion. Setting it loose on the ply would have been better, and allowing the board to run long and fit into dadoes (thus allowing for movement of the solid wood board) would likely have prevented the buckle. Similar to the way panels float inside framed doors.
 
#18 · (Edited by Moderator)
3/4" hardwood (Walnut and Puduk) glued to a piece of half inch PB, framed in.
I v had this board over 25 years, no problems whatsoever.
Don t ask me why, it seems to break all the rules everyone is speaking of ??

- jbay
Your grain alternates between woods. That not only halves the overall expansion on either axis, but causes it to expand evenly in both directions. Beautiful board, btw.

I started to say something about the OP's board being all the same grain orientation, but it was too late to do anything about it.
 
#19 · (Edited by Moderator)
I believe your huge differences in humidity are the main culprit.

I have built four versions of the chessboard detailed in Woodsmith Vol. 22 / No. 132 (Dec 2000). The field of the board consists of 1 7/8" squares of contrasting hardwood, 3/8" thick that are glued to a 1/4" plywood substrate. The instructions recommend using paper shims between the pieces, but I have never done this. I just butted mine against each other. None of the boards has ever buckled. My personal copy is over sixteen years old. Flat as a pancake. The grain of the individual blocks all runs in the same direction.

Here are three of the boards. The one with the chess pieces and homemade checkers is mine. Maple and walnut with bubinga veneer accents. I just took that picture a minute ago. Still flat after all these years.

Table Wood Rectangle Flooring Indoor games and sports


Wood Rectangle Floor Flooring Hardwood


Table Wood Rectangle Board game Indoor games and sports


Wood Rectangle Wood stain Flooring Hardwood


Wood Rectangle Wood stain Flooring Hardwood
 

Attachments

#21 ·
rustin,

I am not sure whether or what extent the walnut/maple squares glued to the plywood substrate contributed to the buckling; but remain convinced that the hardwood expanded with nowhere to go other than up and hence the buckle. If I am wrong and it was the plywood substrate that caused the chess board to buckle, then a dab of glue at the center of each square would solve the problem. However, if the expansion of the hardwood caused the buckle, then allowing each square to move on the plywood while trapping the squares in the mitred frame could end with the same result; a buckling chess board.

I am not sure how much of an expansion gap would be needed for the hardwood chess board. But when making frame and panels where the panel is captured and trapped in grooves in the frame, I like to leave 1/8" - ¼" or even a little more for expansion depending on the panel width. In this case, perhaps 1/8" is enough, but then maybe not. But since the chess board buckled in your friend's home, I would tend toward an overall expansion gap of ¼" on even 5/16" as suggested by Loren.

Since you wish to preserve the inlay strip (which is, by the way a nice accent), and assuming this is the last time you wish to fix this chess board table, my approach would be very conservative. Since you are re-making the chess board, I think Loren's initial idea of using veneer squares glued to particle board or MDF is the safest bet. His approach would allow the 2" x 2" squares to remain this size, allow the board to be remounted to the table as it was originally, and should be stable in your friend's home.
 
#22 ·
I have made chessboards like this before. Never had one buckle. Is it possible that gluing down to the plywood is the problem here?

- BenDupre
The boards I showed are all glued to the plywood substrate. That is what the Woodsmith plans called for. In my experience it works. Four boards constructed in my basement workshop and now residing in a variety of locales from Michigan to central New York, to St. Louis, and on to Palm Beach, Fla. No buckles, no cracks, no problems.
 
#24 · (Edited by Moderator)
Now I've got anxiety. I've made a few of these over the years. None have buckled yet and nobody has called me to say that they have. I think I'll make a few phone calls. Was I wrong in thinking that birch plywood would be less subject to movement than jointed hardwood?

Is the substrate warped or just the squares? Just curious how long you waited after taking the squares from 3/4' down to 1/4". Even thought they are small I'd imagine that the "if you resaw to less than half of the original thickness it will warp" rule still applies? That combined with the grain all running in the same direction kind of gave you a double whammy?

I'm thinking that the major difference between yours and jbay's board is that his are 3/4" thick and stable, realizing that if he had resawed/planed the 3/4" squares from 2" thick stock and glued them up right away they would have moved also? Also wondering if the squares were glued together first or if they were glued onto the substrate?
 
#25 ·
I let the wood "season" in my shop for 48 hours, as I always do. Then I resawed and let the 1/4" strips sit for 24 hours. I saw no warping at all. The squares were glued together first, then as one panel were glued to the substrate 1/2" pllywood.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top