China hutch door warped. Help

[Thom Porterfield]

Hey fellas.
I just finished installing a built-in china hutch and in attaching the doors, I discovered one of them is NOT flat. Because the doors will have glass panels, this is going to be a major issue. Here is an image of the SketchUp model, that shows the size of the doors.
Dining Hutch_SketchUp Model.jpg

Here is a photo of the offending door (the one on the left) which is out of flat by 3/8" to 7/16".
20210323_132008.jpg
The material is QS White oak, 2 1/4" x 3/4". Joinery is bridal joints. Dowel pegged for appearance). Glued with epoxy. I intend to use a bullet catch at the bottom edge of the door. MAYBE the catch will hold the door sort of flat—it doesn't require much force to make the warp go away, but that doesn't remove the possibility of cracking the glass, if even I could get a sheet of glass that large to lay flat in the rabbet. The door frame is finished.

Short of rebuilding the door, is there some way of flattening the frame to remove the warpage?

Thanks in advance.
Thom

[Andrew Hughes]

Is it the door or the hinge. The top hinge is not showing the same as the middle and bottom.
I guess if you have taken the door off and layed it flat on your flat bench. There’s no disputing and I cannot think of a fix.

[Thom Porterfield]

Hey AJ.
Not the hinges. They are really nice Horton Brass, with 3/32 leafs. At the moment, I have just one screw in each, because I will have to demount them to install the glass.

I am surprised I didn't notice it when I was finishing them. Or when cutting the hinge mortises. But I have a leg vise on my bench and it doesn't necessarily clamp things as flimsy as the door flat to the bench front.

I suppose rebuilding the door is in my future. But maybe first, I'll sand off the finish and try using steam, heat, and a bunch of big weights. lol

[Gregory King]

Thom

Have to ask. Is the frame perfectly plumb top to bottom on both sides?. Can you shim the back at the top of one side and the bottom of the other to make up the difference? Should help reduce it quite a bit.

[Thom Porterfield]

Hey Greg.
Yeah, the frame is perfectly plumb (within the limits of a 24" level) and the opening is square. The door is square. It just isn't flat.
And no, I can't shim anything...it fits in the cavity without room for a sheet of gold leaf between the rough frame and the sheetrock.

Besides, if I was able to shim the carcase, the OTHER door wouldn't fit flush like it does now.

[John TenEyck]

If you have the depth for 1/4" glass I think that would pull the door back flat. Steaming, etc. is unlikely to be a permanent fix.

John

[Warren Lake]

was it clamped out of whind or did the style warp. Not fan of 3/4" material on larger or really any doors. Opened high end kitchen doors held by magnets that bend in my hand and are already getting trained to be out of shape. Bullet catches work nice and give a quality sound when they shut. One top and bottom would be nice and hold it in place but wont help you you depending on handle placement and how tight the ball catches are friction wise. If the style had warped you could run a dadoe top to bottom pull it to shape or past and glue a new strip of the same wood in and flush it all up and likely barely show. Are all the hinges in to the same depth as that has an affect as well, one in further will pop the opposite diagonal corner out

[Brice Rogers]

I'm guessing that rebuilding the door my be the only sure way to handle this.

But if the amount of force required to pull it into alignment is low, perhaps a magnetic catch may pull it flat.

[Bruce King]

Whatever you decide don’t forget that it must be tempered glass. Leave space for the wood to move without breaking the glass.

[John TenEyck]

Whatever you decide don’t forget that it must be tempered glass. Leave space for the wood to move without breaking the glass.

Safety (tempered or laminated) glass is only required in passage doors. It's certainly a good idea for large cabinet doors, but not a legal requirement.

John

[Mel Fulks]

It looks like the hinges are symmetrical in spacing . If so ,you could turn the doors upside down. With the bad corner at the bottom you
could hold it with a catch .

[Thom Porterfield]

Hey guys.
Thanks for all the advice. This is why I come to Sawmill Creek. Here's another image describing the dilemma:
Warp.jpg

John TenEyck:

If you have the depth for 1/4" glass I think that would pull the door back flat. Steaming, etc. is unlikely to be a permanent fix.

The glass rabbet is 1/2” deep. I had intended to use double strength glass (3.2mm thick) But accommodating 1/4" isn't that big of a deal, I suppose. See illustration:
glass rabbet.png
Still, I am leery of depending on the glass to hold the frame flat.


Warren Lake:

was it clamped out of whind or did the style warp.

Maybe it was clamped improperly, but I did the assembly on a sheet of plywood on top of my table saw and outfeed table, using risers of identical dimensions, and pipe clamps. Maybe in clamping the bridle joints, something went awry, but I made both doors in exactly the same fashion and one does not exhibit this problem.

Bullet catches work nice and give a quality sound when they shut. One top and bottom would be nice and hold it in place but wont help you you depending on handle placement and how tight the ball catches are friction wise.

I will use bullet catches and have enough to put them top and bottom. However, as you infer, they will only hold the door flat WHEN IT’S CLOSED. The handles will be installed a quarter of the way up from the bottom rail. See illustration in original post.

If the style had warped you could run a dadoe top to bottom pull it to shape or past and glue a new strip of the same wood in and flush it all up and likely barely show.

I don’t understand this. A dado? Where?

Are all the hinges in to the same depth as that has an affect as well, one in further will pop the opposite diagonal corner out

I took extreme care with the hinge mortises, ensuring they are of the same depth and the same distance back from the frame face. Both on the doors and the frame. The door is warped when it is NOT installed in the frame.


Brice Rogers:

I'm guessing that rebuilding the door my be the only sure way to handle this.

I tend to agree with you. Dammit.

But if the amount of force required to pull it into alignment is low, perhaps a magnetic catch may pull it flat.

So, probably, will the bullet catches, but that still doesn’t prevent the warp from straining the glass when the doors are open.


Mel Fulks:

You don’t say if it’s for you or a customer. If I made something like that for myself, I would probably just leave it.

The cabinet is in my dining room. Maybe you could convince me to leave it alone, but my aesthetic sense screams “NO WAY!”

No matter how good the material is or how good you are at dressing wood it’s best to get the rails out as long as the stiles, (or a little longer if two rails are shorter than a stile. Then you dress them all at the same time and use the straightest pieces for the stiles. Some of us who take great pride in accurately dressing material would have to be more humble without that option.

The material was in pretty good shape, albeit stored in an unheated warehouse, rough planed to about 7/8”. I selected stock for figure and lack of internal blemishes, near as I could tell, such as checks, splits, knots and inclusions. The wood spent some quality time in my heated shop before I planed the stock for the doors, maybe two months at 65°F – 70°. I had cut the plank chosen for the doors to rough length, at least a foot longer than required. The stock for the doors was planed at the same time, to 13/16” and then set aside for about two weeks. (This took place over the Christmas holidays, and I was doing neat things in the kitchen instead of the shop.)


Upon getting back to the project, I noticed one of the planks had a twist to it, and as there was sufficient material for the doors in the other one I discarded it. I then planed the stock to 25/32”. Then I ripped the plank to width, cut the pieces to length, and milled the bridle joints. Then I cut the rabbets for the glass. See my response to Warren above for my method of assembly.


Mel Fulks:

One thing that helps in getting out straight material is before each planer trip , sight down the length for straightness.

Then plane the side that is convex. If the stuff is straight after a pass then you should plane the other side.

None of the stock I used exhibited any cupping that would produce a convex or concave condition. There were, initially, some regions of boards that the rough planing hadn’t removed saw marks, and I addressed those sides before proceeding to thicknessing, where I flipped the boards top to bottom in each pass.


William Hodge:

If this is for your house, you could try laminating it.

Hollow out the stile, insert a 3/16" piece of wood in it with Titebond 3, and clamp it flat. Titebond 3 sets up a little slower than Titebond 2. It's like making a bent lamination, but the wood is being bent to flat.

I’m not sure what you mean by “hollow out the stile.” Do you mean something like extending the bridle joint mortise from top to bottom? Wouldn’t I also want to do that to the bottom rail? I fail to see how this would correct the deflection of the one corner. For a lamination to work, wouldn’t the “hollowing” need to extend entirely through the stile or rail?
My suspicion is that the bridle joint at the bottom rail to hinge stile is at fault, not being accurate and therefore producing the deflection. The idea of tearing into that with power tools is daunting. (Visions of ripping all the way through the assembled frame on the table saw with the blade raised 2”? Scary.)

If it's for a customer, it's a re-make.

It’s for my ego. It’s probably a re-make. I could probably rebuild the doors (I’d do ’em both, just because they should match) in only a week or so longer than it would take me to screw up the courage to “hollow out” the offending door assembly. Something I would likely screw up, and have to rebuild the door anyhow.

Brice:
That's a thought. The corners were INTENDED to be held by draw pins, but every time I drilled the holes, they were backwards. Yep. All eight times. So I just made the holes straight through, stuck in a dowel and glued it together with epoxy.

Mel, again:
The hinges are ALMOST symmetrically spaced. I'd have some fussing to do, which isn't much of an issue as the hinge leaves aren't visible (normally), but enlarged mortises would show from the front. A little. And it is the BOTTOM of the door that is sproinked out, not the top. And anyway, I would have to relocate BOTH doors, meaning adjusting all 6 mortises, not just three. The doors clear the frame opening by less than 1/8 inch all around.




Guys, I appreciate your advice. I’ll report on my progress.
Thanks
Thom

[Mel Fulks]

You don’t say if it’s for you or a customer. If I made something like that for myself ,I would probably just leave it. No matter how good the
material is or how good you are at dressing wood it’s best to get the rails out as long as the stiles ,(or a little longer if two rails are shorter than a
stile. Then you dress them all at the same time and use the straightest pieces for the stiles. Some of us who take great pride in accurately
dressing material would have to be more humble without that option.

[Ron Citerone]

You don’t say if it’s for you or a customer. If I made something like that for myself ,I would probably just leave it. No matter how good the
material is or how good you are at dressing wood it’s best to get the rails out as long as the stiles ,(or a little longer if two rails are shorter than a
stile. Then you dress them all at the same time and use the straightest pieces for the stiles. Some of us who take great pride in accurately
dressing material would have to be more humble without that option.

I just learned something that Will come in handy. Thanks Mel.

[Mel Fulks]

Ron, I’m glad I could help. One thing that helps in getting out straight material is before each planer trip , sight down the length for straightness.
Then plane the side that is convex. If the stuff is straight after a pass then you should plane the other side.