Advice on building tall doors for a closet

[Daniel Bejarano]

I have to build 4 doors for a closet that are 105" x 22" x 1.5". The person I'm building it for already purchased 1/2" mdf and poplar.

The idea is to build them like a torsion box, meaning gluing 1/2" strips as wide or as narrow as it's ideal in the middle of the two panels of 1/2 mdf thus producing the final thickness of 1.5".

I don't personally own a vacuum and I know putting the doors inside one would help make straighter doors than clamping. My only resort as of now is to do the latter and kind of hope for the best, but of course, I also know this could be recipe for disaster if something goes wrong.

I am planning to use heavy cauls and sandwich the doors between them and let them to dry (after gluing the strips as I mentioned)

Am I working like a total rookie by doing this? Something here tells me that I could be wrong but also that I only need to try, press them with weight and cauls on a straight surface, put just enough glue and really hope for the best.

My worst fear is getting bowed panels that won't really work as doors that are this tall.

Thank you all for your suggestions

[Phillip Mitchell]

Do you have a totally flat reference surface that’s ~9’ long? If so, I would say you have a good chance of success using cauls and clamps with some care.

Do you have 10’ long 1/2” mdf? If not, how are you going to handle the length of 105”?

[Maurice Mcmurry]

With those dimensions and materials I would make a faux shaker panel door with MDF for the core and poplar styles and rails glued to both sides of the MDF. Standard door thicknesses are 1 3/8 and 1 3/4. If the jambs would accommodate a 2 inch thick door, full thickness poplar, 105 inches long could be used. The joint in the MDF could be hidden under the top and bottom rails. Always make your lamination balanced (equal material on both sides of the core).
For a torsion box slab door I would go for a 1 1/4 inch tall core with two 1/4 inch thick skins, for a 1 3/4 inch thick door. The book Ultralight Boatbuilding shows methods for scarf joints with thin plywood (to get the 105 inch length).

[Daniel Bejarano]

Yes the mdf sheets are 10' long!

[Brian Tymchak]

1/2" mdf for torsion skins in that size is going to result in a very heavy door. You can calculate the weight using 1.9 lbs/sq ft for 1/2" mdf. 1/2" is also overkill for a torsion box that size that is not going to support any load. Like Maurice recommends, go for 1/4" and save some wear and tear on your back and and your friends wallet.

I also have a concern about your friend buying the poplar. Did he take care to buy straight and straight-grained boards? Will go a long way to building doors that don't warp.

[Richard Coers]

A vacuum bag does not guarantee a flat panel, you need a flat platen to make it flat. When I made a 4'x12' conference table, I made a 15" high grid on the shop floor to bridge across any irregularities in the concrete. Then built up the sandwich on top of the grid. I used traditional clamps around the outside and used 1x4s wedged between the ceiling and the top of the table top. It worked out really well.478812

[Brian Tymchak]

478812

Wow! Beautiful work Richard!

[Walter Plummer]

I made replacement bifold door panels that are 12" x 88" tall by glueing 1/4" plywood to 1" styrofoam for a core. Since the ply is closer to 3/16" they came out at 1 3/8" thick like a normal interior door. It has been a while but I think I used 1 1/4 x 1" thick pine for the tops,bottoms and sides and filled the center with the styrofoam. Just white glue. A scrap of 3/4" ply for a caul and a bunch of what ever was around for weight on top. Trimmed and cut to length after glue up. Still hanging and working fine. Good luck.

[Dave Sabo]

I think 9ft doors 1- 3/8” thick is not to scale. Those need to be 2” + in thickness.

I, for one , will be surprised if you’ll find enough (any really) 10’ long poplar boards that are dead nuts straight. Specially since the customer chose the material.

[John TenEyck]

Build one and find out. It's not an expensive risk. Personally, if you start with flat stock and have a flat reference surface I think they will come out fine.

John

[Maurice Mcmurry]

A 1/2 inch thick core is by no means a torsion box.

[Richard Coers]

A 1/2 inch thick core is by no means a torsion box.
Sure it is. Anything with skins and a grid core is a torsion box.

[Dave Sabo]

Sure it is. Anything with skins and a grid core is a torsion box.

I suppose we can debate this, but Daniel’s proposal isn’t using the 1/2” poplar as a frame as much as it is a shim. The fact that he’ll use the poplar flat instead of on edge means his “grid core” is losing most of its strength since it’s not oriented on its edge.

I also think one of the main design tenents of a TB is that the “skins” are thinner than the core. Daniel’s design doesn’t adhere to this either.

At any rate ; whatever one chooses to call this design , I think it’s destined for disappointment.

And no one has brought up the nightmare the painter is going to have with the edges of this proposed design.

[Phillip Mitchell]

I’m still having a hard time visualizing exactly what the original plan is.

Have you considered stave core construction?

[Richard Coers]

I suppose we can debate this, but Daniel’s proposal isn’t using the 1/2” poplar as a frame as much as it is a shim. The fact that he’ll use the poplar flat instead of on edge means his “grid core” is losing most of its strength since it’s not oriented on its edge.

I also think one of the main design tenents of a TB is that the “skins” are thinner than the core. Daniel’s design doesn’t adhere to this either.

At any rate ; whatever one chooses to call this design , I think it’s destined for disappointment.

And no one has brought up the nightmare the painter is going to have with the edges of this proposed design.
The major strength of the core is not in the direction of the grid wood, it's in all the glue area on the grid and the two skins are reinforcing each other.