Wood movement

[Chris Schoolland]

I'm a finish carpenter working on a house where we're having all the doors (and beam ceilings and cabinetry) made out of afromosia, an african teak-like wood. The doors will mostly be 2" or 2 1/4" thick, 3' to 5' wide, and 8' to 10' tall. They're a combination of pockets and pivots.

Do any of you have experience with afromosia? My concern is with the proposed door design.

The doors will all be horizontal slats of undetermined width (probably around 6") each separated by a 3/16" x 3/16" dado going all around the door (3 sides for the pockets). We talked the owner into going end grain on the door edge instead of mitering around so we have at least a remote chance of shaping these things. The concern I have is that in essence we're making a board up to 10' wide. I'm thinking we need to design the slats like horizontal siding, lapping each other to allow some movement.

Sorry this is getting so long, but a bit more info: the doors will likely be built around a metal frame with a plywood core. The afromosia cladding will likely be 3/8" thick. The top of the pocket doors are not allowed to have any trim, so the top section needs to be tricky. If I can attach my sketchup brainstorming, you'll see what I mean. Play with the layers.

[Larry Edgerton]

I have used some afromosia on boat work and as it is a relitive of teak it is oily, and so requires special care when gluing. I wash it with carbon tet or acetone just before gluing with epoxy. West System has a section in their manual that deals with these kind of woods.

Sounds like an awesome house you are working on, I'm envious! Have no work here in Michigan, thats why I'm bugging everyone on here.....

I've made a lot of custom doors, but that one sounds a bit different. A plywood core is a good idea, but really see no need for the metal unless I am not understanding and it is an exposed detail? I would not like the ends of the doors with the end grain showing, and in my opinion there needs to be room for expansion between the slats, even with a legume. That will make the end ugly in my opinion, even if they are tight together.

I would instead build then with a thin wood band all the way if for no other reason than to hide the ugly end grain connections. I would set them in a bed of West System to minimize wood movement. I have built many heavy outside doors with this method and have yet to have a failure.

Keep in mind that this wood is a bit toxic, so if you manage to make dust, try not to breath it. It is also an threatened species, but not by our hand as woodworkers but to loss of habitat.

Does this have something to do with an interior decorator? Sounds like one of their ideas....

PS, when I pulled up your sketchup I got a bunch of japanese characters? I'm sure I did something wrong on my end?

[Chris Schoolland]

Thanks for the reply, Larry.

Interior decorators are definitely involved, although the owner is equally involved. While not a speedy job, I very much appreciate their desire to achieve a certain look. More importantly, I am most definitely blessed to have work right now. Right place, right time kind of thing.

The reason for the metal is to help stabilize the doors. They will be up to 10 feet tall with only two hinges on the pivot doors; an upper and lower pivot inset around 3 inches from face of plaster wall. The plaster will be the jambs on the pivot and head sides of the door, with a minimal metal jamb only on the strike side. Controlling the plasterers becomes vital. We're having custom offset pivots made to accommodate the fact that our doors can't lift up onto a pivot. The pockets are even more tricky, as they too will be trimmed by plaster. That's right, a 10 foot tall pocket door going up into the ceiling and 1/4 inch off the floor. Non-removable unless the door top or bottom comes off. We're working on that. Basically, both styles of door have to stay straight or they stop working. We just don't have the adjustments we usually do with stop and jambs and whatnot.

I feel good about most of it, but the door edges and wood movement need to be resolved. Good call on the West System. One of the issues is the owner wants to see the dado (you mentioned "legume"--I'm unfamiliar with that term. Probably better term than dado) wrap from the face to the edge to the other face to the other edge, connecting all the way around the door. I agree that edge grain is not ideal, but the only other option we came up with was to miter the stuff all the way around which would look better, but eliminate our ability to shape the door edge at all. The owner is ok with the edge grain (we did a small mockup) but I'd love to hear better ideas.

Has anyone else been able to see my sketchup model? I'll try looking at it from a different computer. Thanks again for your input. This is a complex project, and I value any ideas/observations you all might have. I very much appreciate this forum.

Chris

[Brad Shipton]

This link maybe of interest to you. The door construction being discussed has some similarities. Well above my door building skills, so I have nothing else to add.

Go to woodweb.com: home>forums>architectural woodworking> 2 3/4" thick doors

Brad

[Chris Padilla]

I brought up the SU just fine, Chris. I hadn't learned about the "layers" feature and so you just taught me something new about SU! Kewl.... :)

I don't think I can add much to the discussion as this seems relatively complex and I certainly don't fully appreciate all the issue you are running into with the overall design and how they need to be installed. This is one of those, you have to see it in person to grasp it things, I think (well, for me anyway). :D

[John Schreiber]

I've never worked on a project anything like that, but after looking at your SU model, I've got some ideas. They may be worthless, but it's something.

Each board would not be attached to the one above or below it, but only to the plywood. The top of each board would be screwed to the plywood through an oval hole to allow for wood movement. That would require the tongue at the top of each board to be a little taller than in your model. The bottom of each board would be glued to the plywood (and maybe pin nailed or even screwed from the other side of the plywood where you can). That allows the top of each board to float and allow for wood expansion. The top of the top board could be pocket screwed or similar from the top surface of the door.

These are just some ideas. When I say board, I mean each piece of wood on the front, back and edge of the door. The screws in the oval holes at the top may not be necessary if you are able to attach the bottoms of the board securely.

Seems kind of risky to me to try something that challenging on a set of doors for a whole house. I hope you have an opportunity to mock up a couple of ideas to test.


Good luck. Very nice SU model by the way.

[Chris Schoolland]

Glad you could see the model. I love SU. I keep finding out more it can do, too. Layers were cool to discover, as were shortcuts and scenes. Some of the video tutorials are really helpful.

John, I've never worked on a project quite like this, either, but I really like the problem-solving aspect of this; which is one reason I threw it out there to see if any one could make sense out of my situation and see something I've missed. I have some great co-workers there with me, but at this brainstorming stage more heads can be helpful.

Whatever we try, we will definitely make a complete functioning mockup before going too far. That's how it works on this job. There are some unbelievably cool bamboo grids we've got coming down the pike as well. They approved our mockups and we ordered the material--I'll post photos of our mockup if the contractor thinks it's OK.

Nice to learn about the woodweb architectural site, Brad. Probably more helpful stuff on there, too.

I've been sick at home the last few days, so thanks for helping me pass the time mulling this over!

[Larry Edgerton]

The reason for the metal is to help stabilize the doors. They will be up to 10 feet tall with only two hinges on the pivot doors; an upper and lower pivot inset around 3 inches from face of plaster wall. The plaster will be the jambs on the pivot and head sides of the door, with a minimal metal jamb only on the strike side. Controlling the plasterers becomes vital. We're having custom offset pivots made to accommodate the fact that our doors can't lift up onto a pivot. The pockets are even more tricky, as they too will be trimmed by plaster. That's right, a 10 foot tall pocket door going up into the ceiling and 1/4 inch off the floor. Non-removable unless the door top or bottom comes off. We're working on that. Basically, both styles of door have to stay straight or they stop working. We just don't have the adjustments we usually do with stop and jambs and whatnot.

I feel good about most of it, but the door edges and wood movement need to be resolved. Good call on the West System. One of the issues is the owner wants to see the dado (you mentioned "legume"--I'm unfamiliar with that term. Probably better term than dado) wrap from the face to the edge to the other face to the other edge, connecting all the way around the door. I agree that edge grain is not ideal, but the only other option we came up with was to miter the stuff all the way around which would look better, but eliminate our ability to shape the door edge at all. The owner is ok with the edge grain (we did a small mockup) but I'd love to hear better ideas.

Chris

I have used hinges that work in this manner, and one of the bad points is that there can be no stop, so there is a visible gap around the door. The thicker the door the bigger the gap has to be. I did one that I made curved side jambs on a 9" thick wall that followed the curve that the swinging doors created. The owner did not want people to be able to see from one room to the other, but the door swung both ways so it was all I could figure out. There is still a gap for sound to fly through. Also you have to keep inertia in mind, as a door made in this manner is heavy! You have to figure out a good stop or they will rip themselves off of the hinges. I machined my own hinges out of aluminum using wheel bearings from a Toyota and the stop was part of the aluminum housing, interlocking aluminum steps that allowed 180 degree swing. The first ones failed when I did not make the stops big enough and the swinging 450# door sheared them right off. The next ones were beefier and I put a cushion in to absorb energy. I installed the bottom hinge from the floor below before that was finished. There is access to adjust height in the basement ceiling. Its been 20 years and I have not heard from them, so it still must be working?

Another thing to keep in mind is that a door this heavy is a danger to kids if it has no dampener. On a normal door if a kid gets his fingers caught in the door it will hurt, maybe even break something, but when the inertia goes up so does the danger to kids. For this reason I use a bigger stop on heavy doors and round over the sides of the door so it is not a shear.

I have been pondering your edge treatment, and I have not had a brain storm yet. I have had a few ideas, but they all have a drawback.

I just got a call this morning to finish a house, so I was only out of work for one week. Short job, maybe a month on the outside, but that is one month closer to summer. Its a lot more fun to be broke in the summer! :)

[Thomas Knapp]

The reason for the metal is to help stabilize the doors. They will be up to 10 feet tall with only two hinges on the pivot doors; an upper and lower pivot inset around 3 inches from face of plaster wall. The plaster will be the jambs on the pivot and head sides of the door, with a minimal metal jamb only on the strike side. Controlling the plasterers becomes vital. We're having custom offset pivots made to accommodate the fact that our doors can't lift up onto a pivot. The pockets are even more tricky, as they too will be trimmed by plaster. That's right, a 10 foot tall pocket door going up into the ceiling and 1/4 inch off the floor. Non-removable unless the door top or bottom comes off. We're working on that. Basically, both styles of door have to stay straight or they stop working. We just don't have the adjustments we usually do with stop and jambs and whatnot.


I am surprised Rixon Pivots won't work. You can tip the door back on the top to get it over the bottom pivot.The side and top pivots can be lowered down with an allen wrench. Then raised when the door is tipped up into place. You could make a Lapped rabbet joint at each querk and alow for some movement that way. You would be back to mitering the edges to hide the joints though. Arn't architects and interior designers fun? Often the know how they want things to look, but have no idea of how they work.

[Chris Schoolland]

Sounds like we're going over a lot of the same issues you did, Larry. Gaps around the door, inertia concerns. Informative to hear about your experiences. I like the basement adjustment idea. We will have a one-legged metal jamb on the strike side, far enough out to keep the strike from rubbing on the wall. Most of these pivots are in hall-like situations.
Great to hear you have some work coming up, Larry.

Thomas, we would like to use Rixsons. The owner doesn't want to see the plate up there on the ceiling, even if it only shows when the door is open. Also, in my experience, the door needs to be close to vertical before slipping onto the bottom pivot, and we only have 1/2- 9/16" to work with; 1/4-3/8" off the floor to bottom of door, and 1/8-3/16" top of door to the plaster, no head jamb. The offset pivot was decided on before I got to the job. I like the lapped rabbet idea--maybe it could be stopped short of the end or something. Thanks for the input, guys.