Correcting Warp in Cabinet Doors

A cabinetmaker collects many suggestions for working the warp out of a set of oversized cabinet doors. He tries one. It fails. January 13, 2015

Question
We have a few cabinet doors (soft maple, with non-true mullions for glass, unfinished, outsourced from door manufacturer) approximately 24w x 52" tall and one is particularly warped (3/8" on one stile), the others are slightly warped 1/8+/- on a stile. Because they are over the recommended limit without a middle rail (48") I can't ask the company to rebuild them. These are inset faceframe cabinets, so the doors want to be very flat. Before ordering new doors, I was thinking of clamping them overnight in direction opposite of the warp, perhaps adding some moisture to the situation. Has anyone had any success with this method? Any input would be appreciated.

Forum Responses
(Cabinetmaking Forum)
From contributor K:
Use the glass to help straighten the door in combination with the methods you described, 3/16 glass and a heavy bead of silicone. You might even need to use the mechanical fasteners that are made for attaching glass because the silicone may tear.



From contributor D:
Hafele has the Planofit door straightening system. #407.90.701. It comes in white or black. That should do the trick.


From Contributor O:
I am of the opinion that the 'over-bending' and 'wet one side...' and similar strategies will never work. Think about a warped passage door that is forced closed (for years) that never even hints at going back to flat. I do know I have tried, more than once, along with various incantations and exhortations. None of which worked. It is the trial of a desperate man. The Planofit is exposed on the backside and will raise questions as to why it is there. Replacing the door, replace the stiles are obvious. Another method, though with less predictability, is to rout out a 1/2"x1/2" recess in the back of the stile, near full length, and make a curved part that will fit into the plow. The curve should be about twice what the stile curve is. Dry clamp it in place to see if it makes the stile acceptable. If it does, then add the glue, clean up and refinish the area.


From the original questioner:
I think that Contributor O hits the nail on the head. The doors aren't going to de-warp. I like the Planofit system, but it would not be acceptable in this location. That's a great point about the glass taking some of the curve out. We'll test that. (We have the nylon retainer strip). We are also going to try Contributor O's method of rabbeting the back and inserting a curved piece (we are looking for another warped board). We are going to reverse the warp during the glue up (equal to the same amount of the curved door) and see what happens.


From Contributor O:
You can preview how much the stile will straighten by clamping the piece into position at either end, before gluing. I would saw out the curve on the bandsaw to be able to control the curve, and make it twice what the stile is. Nothing wrong with rolling the dice, but it gets expensive. Yes, the glass will help - up to a point. It can sit for a while and then decide it has had enough and then crack.


From contributor S:
Soft maple you say? Are these going to be painted? If so the easiest fix is to put a series of kerf cuts in the back of the door frames, clamp to the table with a very slight over correction, add epoxy like West Systems and add some filler to it when mixing presto, problem gone!


From contributor M:
I've had success with sawing the stiles on edge (kerfing out the middle thickness-wise) then filling in with a laminate: turning the stile into a three ply laminate. Make the cut shy of the profile detail, and clamp it up to counteract the twist or bow. In most cases, it's best to just rebuild the doors, but sometimes it helps to have a fix handy.


From Gene Wengert, forum technical advisor:
We can wet and then bend flat, but it is not a permanent fix, as a change in humidity can cause the warp to return sometimes. It is also difficult to know how much to bend the door beyond flay so that when the pressure is removed, the springback will result in a perfectly flat door.

From a technical point of view, this initial warp is caused by a moisture change, but the tendency to warp is within the wood itself, most often slope of grain. That is why pool cues, long handles, shingles and some other items that must be perfectly flat when the moisture changes are manufactured so that the grain is perfectly flat. A split follows the grain, so the wood is initially split and then the sawing follows the split. For large doors that require straightness, consider straight grain selection or else use a metal stiffener rod. Note that a moisture gradient can also cause warp use to different shrinkage, so coating all surfaces will usually avoid moisture gradients in smaller pieces of wood.



From Contributor C:
You should think about just replacing the door. How much did it cost? You are seriously thinking about cutting curved splines, and attempting to restrain the curve with the glass? What happens when the glass explodes on a child? Just replace the frame.


From the original questioner:
The only method that de-warped the door was the remove a portion out of the thickness of the door and then gluing up a new ply. We were not able to sufficiently control the curve. In the end, the naysayers were right. It's better to just get another one made (even though it was another $250 - we spent way more in labor, etc. for something that was not ever going to be right.) Thanks for all the advice! Now I know better.