table top problem

[Ryan Basler]

Hi everyone. I am new the the woodworking arena. Started about a year ago. I make things for a living out of composite(Fiberglass, Carbon fiber) and metal but new to wood. A buddy asked me to build this desk for him. Photos attached Table 1-3 are undamaged, upon delivery. The table was build this summer with my limited tools (TS, RAS, Router, ROB sander). I live in Georgia so hot and humid until October. Buddy called me a last week stating the desk top has cupped downward from front to back about 1/4 mainly in the center of the desk. He noticed it when he left a pin on the desk and it rolled the middle. Before all the flaming on how I constructed it. Yes I have learned my lesson on what i think is expansion and contraction, and yes i will build it differently next time.

Construction:
2 inch strips of 4/4 Walnut, hard maple, wormy maple, cherry, and Red oak. All cut to about 24 inch length. Strips were arranged on the 1/2 Birch plywood then glued in place. 4/4 walnut mitered and glued to the edge like face framing with about 1/4 over hang below. I Fabricated the steel legs with angle iron stretchers front and back. I installed threaded receivers into the bottom the table. 3 front and 3 back into the angle iron. 3 on both sides by the w shaped steel legs. Steel hole are about 2 sizes big with bolts and washers.

I told him to remove the front and side bolts to maybe release the pressure. This made the cupping worse about 3/8 possibly more. Seen in image: Table flex This scared and he just forced it back and re-bolted. I believe it is the top strips shrinking and pulling a cup due to the plywood not changing. Im not sure how i can solve this but i was thinking of cutting some 1/2 deep cuts from left to right on the underside to give the hardwood top room to move front to back without fighting the plywood. My only concern is the miter face frame. i could put a small kerf in rear face to allow for it to move but it is still glued to the plywood.

On a funny note, maybe have him just add a big humidifer to that room. Maybe that will solve it!

[Jacques Gagnon]

Hello Ryan,

The table looks good!

In terms of fixing the problem, I think your proposed approach should work. I have the same impression as you do; the glued solid wood strips expand and contract while the plywood substrate does not.

"Disconnecting" the solid wood top from the plywood substrate should solve the problem. I would think that one cut every third strip (ie 1/8 inch expansion every 6 inches) would be sufficient. To be safe, I would make sure to slightly "dig" into the solid wood top to completely "disconnect" the plywood. You will likely also need to either elongate your hole in the steel (difficult) or make some slots in the plywood (should be easier) to allow the movement across the width to take place throughout the year. I am thinking along the lines of "T-nuts" for fastening the top to the legs.

People who deal with this type of challenge on a regular basis will likely pitch in with other solutions.

I hope this helps!

Jacques

[Jim Becker]

The situation comes down to the nature of wood products...plywood moves very little seasonally and solid wood can move a lot seasonally. And that movement is across the grain. Laminating the two is problematic in that respect unless you float the solid stock on the plywood with mechanical fasteners from below that still allow the wood to move. But that then complicates the banding you put around your table top since it also is installed on the end-grain sides and tries to constrain the wood movement. You seem to understand that. The kerf you mention isn't going to help a lot, however, since the frame is fasted to the plywood and again, that plywood isn't going to materially change size over time.

Note that a thin or slightly thicker veneer on plywood doesn't generally cause wood movement, but I'm guessing that wood panel you made has some level of thickness to it.

I'm not sure what the best solution here is outside of remaking the table top such that it's either completely from solid wood with floating caps on the ends or remaking it with veneer plywood, either pre-laid or with shop made "thick" veneer (1/16" or so) to avoid the problem you have now.

[Robert Engel]

This is one of THE most common problems with moving furniture indoors (even aside from the incorrect construction method sorry).

A cup can be fixed. The easiest solution is to re-rip, re-edge an re-glue. I've done this successfully many times. However in this case it is problem not the most desireable solution.

You can also use relief cuts. I think this is a good option to try. Route 1/4" slots lengthwise full depth of plywood. Leave the end caps in place. If you can flex the cup out, fasten in such a way to allow for further movement (clips, elongated holes, etc.).

If you can't flex the cup out, you'll have to re-rip & re-glue the top.

[Jacques Gagnon]

Jim:

Oups! Your comment made me realize that I had forgotten about the banding on the sides. This complicates things a little!

[Keith Weber]

What a beautiful table! It's a shame about the plywood being glued underneath it.

You know now that plywood and hardwood do not mix where they have to fight each other for expansion. The other thing with your table that you may or may not know is that there could be issues with the mitered walnut skirt on the edges. Wood doesn't expand much along it's length, but it does across the grain. Therefore, the end skirt pieces will not expand and contract as much as the main table boards will. Right now, the plywood is trying to maintain it's width, and the contracting top is causing cupping. If you rip lengthwise slots in the plywood, it will probably help ease the cupping, but it may very well transfer stresses to the mitered corners, causing them to squeeze together and crack where the walnut is glued to the top.

Solid wood, mitered banding works best with veneered panels over solid substrates. With a solid table like this, if you want the skirt/banding on the ends, it's probably best in the form of breadboard ends that allow for expansion/contraction.

If you knew what the humidity level was in your shop when you built it, you could give it to your buddy to use to set his humidifier!

[Prashun Patel]

Here's what I would do:

1) Flip the table over, and rip kerfs ever 3-4", lengthwise through the plywood, but not into the solid wood.
2) Dock the mitered skirt on the ends. You can either leave it off completely, or attach a matched molding "breadboard style" to the ends. Don't miter it; just butt joint it.
3) Add battens to the underside of the table. You can do this in a couple ways. If you don't want them to project downward, you can recess angle iron into the bottom. But this has to be sized very carefully to resist flex without weakening the top too much.

[Andrew Hughes]

I think you should take the table back and bring him one built properly. Esp if he paid for a piece built with integrity.
It will also add a good vibe to your friendship.

[Ryan Basler]

I think i will try by ripping off the side rails cut kerfs into the bottom and re attach the rails in a butt joint fashion. If all else fails, ill re build the whole top.

PS: my buddy knew the risk when building this. He knew I had never made anything like this before and is a very understanding person. He is only in this the materials.

[Jacques Gagnon]

Ryan,

Good approach. Please refer back to Jim's point about the ends. Ideally, they would "float" and be able to absorb the expansion. Something along the lines of a sliding dovetail, with only the front end being "fixed" (glue on a short distance). Breadboard assembly would be another approach.

[John TenEyck]

It might be less work to just make a new top. You could probably even re-use the solid wood off the existing top and add a little new wood to make up for any lost width.

Your mistake is oft repeated, so don't feel too bad. I was at a customer's house yesterday and noticed the wood countertops were warped much the same as your table. I went over to look at them and, sure enough, there was a perimeter frame of wood glued and screwed underneath to double the thickness. When the top inevitably cupped this Winter the HO called the maker (who's been in business for more than 20 years) and he sent someone out to add more screws underneath to try to hold it flat to the tops of the cabinets. Shortly after the tops cracked in two or three places. Wonder what his next move will be.

John