What is a glue-up distaster?

[Darius Ferlas]

I am building a dining table, 72" by 38". The top is coming out nice and flat but it doesn;t mean it will remain that way forever, unless wood is prevented from having its way.

I am considering a bread-board at each end of the table, and probably two pieces of wood attached underneath the table top. The plan is to keep the surface flat for the foreseeable future.

Having seen various projects and comments on the subject, there is an abundance of warnings against gluing cross grain or else a disaster will take place. Since I'm still lacking in all the woodworking experience I would like to have, for now this disaster sounds to me like a boogie man or the big bad wolf. Especially that I have seen table tops of similar size of glued up underneath. After 20 years they show no signs of any disatsers or big bad wolves.

So what would that "disaster" consist of if it actually did happen? Bulged table top? Split members of the glue-up?

[Michael MacDonald]

don't glue the cross grain breadboard or the underside supports. Fix them in the middle with a dowel or screw. For the end fasteners, elongate the hole to allow for wood movement... for the breadboard, this would be the hole through the tongue of the table top... for the underside support pieces, this would probably be a the pilot hole.

Then when the table shrinks, it won't split, and when it expands, it won't warp.

[Rod Sheridan]

Hi Darius, the issue is wood movement due to changes in humidity.

If you have a wide board that's flat sawn, the wood will expand and contract with changes in humidity.

If you attach a board at right angles to the other board, it won't change in length by any appreciable amount, while the original board will.

If the connections are strong, the wide board will buckle or split depending upon which way the humidity is changing compared to when they were attached together.

For a breadboard end, I glue the middle section of the breadboard (maybe 2 or 3 inches) then leave the remainder floating.

To keep the ends of the breadboard tight to the table, use a sliding dovetail, or vertical dowel pins that are in fixed holes in the breadboard, and slots in the tongue of the table.

This allows the table to expand and contract regardless of the fixed length breadboard end.

Note that at different times of the year the breadboard will be shorter than the table is wide. That's why i generally avoid breadboard ends on tables.

Regards, Rod.

P.S. Google wood shrinkage calculators, their are some free applications that will allow you to calculate how much your table top will change in size.

http://www.woodweb.com/cgi-bin/calculators/calc.pl?calculator=shrinkage

The above is one example of one calculator. try inputting the example below.

If you have a 30" wide red oak table, and in the winter you have 8% moisture in the wood and in the summer 14%, the table will be 9/16" wider in the summer.

[Jamie Buxton]

I know a couple with a dining table about that size. It had legs at the corners, and the usual aprons. The boards in the top ran the short direction, and they were glued to the aprons. In the winter, the top would be nearly straight. In the summer, the middle would rise about an inch.

[Stephen Cherry]

Are there long overhangs? If not, then the tables apron should hold things flat, using a fixed center attatchment with others floating.

One thing I've done that seems to work OK is to use a pocket hole jig to make pockets on the inside of the apron, with the center pocket normal width, and the side pockets pilot holes elongated to allow for movement.

[Bill Davis]

First thing I would do if it's not too late is to try to use quarter sawn wood because radial dimensional change is about half of that for plainsawn. then make sure the wood has reached the average EMC (equilibrium moisture content) for the location where you will have the table. You need to know the average yearly RH. There are charts to determine what it should be for your area. That will assure minimum dimensional change over time. Then I would follow Michael MacDonald's advice on attaching the breadboard. If you glue it along the full length or use fasterers to try to hold and restrict movement you will pay a big price. Expanding/contracting wood has tremendous forces. When you get to the point of finishing choose a finish that is a good moisture barrier to limit the effect of relative humidity changes to add/remove moisture in the finished table. It is the ongoing process of moisture moving in and out of the wood that causes the problems. No finish will make the wood impervious to moisture content changes - just limit it to a more reasonable reduced amount. Finish all surfaces equally. The unseen surfaces need to be sealed with the same number of coats as the visible top that needs to look nice. Remember wood will be constantly adding and giving up moisture as your RH changes and therefore the wood will expand and contract. You 'simply' have to minimize it and not fight it.

Hope it goes well, and it can if you avoid some of the no-no's. And be sure to put up a picture of the finished product. We'll look forward to seing it.

[David Cefai]

I had a table where the top was firmly attached to the stretchers with pocket screws. One day the top split with an almighty crack. Fortunately the glue failed, not the wood and it was repaired as good as new - in fact better 'cos the carpenter allowed for wood movement.

[Aaron Berk]

It's on the news stand now. an article titled "attach tabletops with dovetail cleats"

Here are two pics from my issue, maybe this helps?

[Cody Colston]

As the others have said, a cross-grain glue-up on a wide panel like a tabletop CAN cause it to either warp or crack. It's not a given but there is a high possiblity.

In a modern climate controlled home, if the table were built with the wood acclimated to the home's environment and it was finished on both sides, you could probably get away with gluing cross-grain. But, why take the chance of it splitting later?

Those pics from the current Fine Woodworking Magazine illustrate a neat way to secure a tabletop and still allow for movement. I made my dining table out of Mesquite and I used two grooved crossmembers held at the aprons with sliding dovetails and secured to the top with shop-built wooden clips that ride in the groove. It also has breadboard ends with the center glued and the ends pegged with a slot in the tenon...no glue.

Note that breadboard ends do not hold a panel flat. It will still move as it wants with moisture changes from humidity. Rather, a breadboard end covers the end grain and reduces the moisture absorption, thus reducing the movement.

It's just good practice to design and build to accommodate wood movement. If not, sooner or later you will experience that disaster you mentioned.

[Steve Schoene]

The key question is how is the base of the table constructed? (And, in general the base should be build first, the top last.) On ordinary tables, with legs in the corners joined by rails, the base keeps the top flat. The top is attached with screws with elongated holes, or with figure eight devices, etc. to allow the wood to change width's with the seasons.

As far as the disaster that might arise by firmly gluing cross braces or breadboard ends in place is that the table will almost certainly split. The attachment of any cross grain pieces has got to allow for the wood movement.

[Henry Ambrose]

I am building a dining table, 72" by 38". The top is coming out nice and flat but it doesn;t mean it will remain that way forever, unless wood is prevented from having its way.

snipped.....




Your opening sentences show an assumption that is incorrect. If you built the top well it will stay flat on its own. When you try to restrain it is where the problem crops up. You're not stopping it from moving - it'll have its way and you can't stop it. Others have noted ways to attach the top to the aprons or how to attach breadboard ends that will work.

[Darius Ferlas]

Thank you for all the great responses. I read them all carefully, and all seem to point in the same direction, no matter how bad I'd like to go for shortcuts.

I followed some of the suggestions and I settled for partial gluing of the breadboard to the rest of the table top. The shrinkage/expansion calculator Rod linked me with show that I am looking at 3/16" of wood movement". I used glue for 4" on both ends of the table top's center, for the total of 8" of glued surface on each end. I also used 3 dowels within that glued area and these dowels are glued solid. My assumption is that 8 inches of lumber will not cause any problems with the minimal wood movement. If I'm wrong then ... oh well.

The remainder of the breadboards are only doweled, one dowel at about 4 inches from the edge of the table on each of the edges. These dowels will be able to travel within the table top's tongue, within slightly elongated holes (banana shaped). In addition, to help the breadboard stay snug at the edges of the table top I concaved it 1/32" in the center. That bow allowed me to use just one pipe clamp for the glue-up. I don't have more clamps of that length. Overall the nicely snug fit of the breadboard felt almost as if I didn't need any adhesives or fasteners.

The dowels do not go all the way through the breadboard and the table top. I drilled the holes from the bottom up, stopping about 3/16" below the table's surface. That allowed me to create a strong and yet invisible pin. The fit was what I'd consider very good. Not too tight, not slacky at all.

Here are some of the pictures I took - not the entire process though.
Again, thank you all for the great suggestions and advice.

Darius

[David Nelson1]

This thread is a keeper! I have read quite a few articles and threads about this subject ,but the light never really came on then. It's starting glow now!