Interior Door Construction Questions

Here's a long and detailed discussion of lumber choices and construction methods for thick custom interior doors. March 20, 2012

I build contemporary interior doors and have had a lot of success with techniques I learned from WOODWEB. I'm now getting ready to build a small group of more classic three panel interior doors much larger and heavier than what I normally make. I've read a lot in the archives but hope I can clarify a few issues.

First, these doors will be 2-1/4" thick with most being 32"w x 82"h. I'm planning on using stave core construction and making the staves in house. It seems there's a lot of contradictory information on wood stability so I want to see if I can get a clearer picture. Some guys say poplar is great for stiles and rails, and others say it moves too much. Does it matter for stave core material? I can buy soft maple for the same price as poplar but it would make a heavy door even heavier! My feeling is once glued together the stability will be greatly enhanced, though my first concern is the doors stay flat over their service life so?

Second question - I'm going have a set of corrugated cutters made to run the cope and stick as it's only 13 doors, not really worth investing in spendy tooling. I'll be using loose tenon joinery as that's what I'm set up for. However the tenons I usually use are too small for this size door. The max my slot mortiser can handle is 5/8", but I'm wondering if I'm better off using a pair of 1/2" thick tenons at each joint instead of a single 5/8"?

Last for now - the panels are going to be MDF in order to keep the cost reasonable and also because I think it's a better surface for a painted door. I'll have to glue up to get 2-1/4" thick panels but is there anything I can use in the middle that will cut the weight and not be destroyed in the vacuum bag? I haven't yet figured out how I'm going to machine the raised panels yet either. Would it make sense to buy two RP cutters, one each direction, then stack them back to back to get it in one pass?

Forum Responses
(Architectural Woodworking Forum)
From Gene Wengert, forum technical advisor:
First, you need to indicate what poplar you mean. Both yellow poplar and aspen poplar have been used for door cores with good success but the two are not related and are quite different in processing. YP is heavier than soft maple. Aspen is going to give you the flattest door if the moisture is correct.

From the original questioner:
I'd have to guess aspen poplar as all I can say is it's lighter and softer than soft maple. Up here I haven't heard it referred to as either, so I'll have to call and ask.

From contributor Y:
I use two cutters to raise panels top and bottom at the same time. It works well and the cut is consistent thickness wise.

From contributor D:
Heavy interior doors 32" wide aren't such a bad thing. That said, I don't think I've ever had an order for 2 1/4" interior doors. Was this specified, or do you just have freedom to make what you want? You can save a few pounds by making two 7/8" thick panels separately, then assemble them with a 1/2" foam sheet between them.

Stacked rp cutters are can be pretty intimidating when doing small radius panels. I'm all for doing one side at a time myself, even then in several passes. Large rectangles can be power fed, but taking it all in one whack requires a bigger, stronger shaper too. Thirteen doors, three panels each, doesn't sound like a high production situation and I'd just as soon be safe than sorry.

From the original questioner:
I'm expecting a shaper upgrade very soon. Not sure I would attempt the double raised panel cut with my current "little" 5hp Powermatic 27. Yes I would certainly have my powerfeed set up and ready to go. These are all rectangular panels so keeps it pretty simple. I'm just concerned that if I run two passes I'll have trouble keeping the thickness consistent.

The 2-1/4" thickness is owner specified. Just for a clearer view these are going into a 1920's era building in the heart of Boston, so they want everything to have that "original" look....even if it's not actually original.

It's not a lot of doors, and luckily I'm not trying to do them on a shoestring budget. So overall I'd rather go the safer and better quality method throughout construction. That's not to say there's no budget either, just that I have a little flexibility to not have to sacrifice quality.

From contributor D:
Your 5hp Powermatic can easily do this, especially if it's just MDF you'll be raising. RP cutters are kind of a mainstay for a door builder, so I'd shop around for a good set, balanced to G2 standards, with polished carbide inserts that seat into a positive location easily with little fuss. You can also change profiles to a certain extent. Buying two sets of braze-on cutters, one backup for when another is out being sharpened, is not really a good way to go if you're concerned about spacing between the top and bottom, as with each sharpening the tongue gets a little bigger.

Really one side at a time I guess you just either like or hate. Not a big deal to keep some accurate calipers in your pocket and raise the cutter ever so slightly, like a sixth of a turn at a time until your last pass is set up to run, but to each his own. Someday you'll get an order for a different hip on one side, or flat or whatever, and then being able to be flexible is nice.

From the original questioner:
To contributor D: I don't like or dislike either way - yet! I haven't done double sided raised panels before so this is a first. I'm just trying to think the whole process through before I start ordering the materials and cutters I'll need. On a normal raised panel I cut the raise and the backcut at the same time. Doesn't necessarily mean it's right for this application though.

I already have a raised panel cutterhead but it uses corrugated knives, so I may need to move to a carbide insert head anyway. Don't think the HSS knives will last long in the MDF.

Lastly the poplar sold here is yellow poplar. Aspen poplar is sold as aspen. Not sure if the wood has different properties depending on location, but poplar here is definitely lighter than soft maple. I don't mind a heavy door, but the soft maple is going to be quite a bit heavier and my back wouldn't mind the lighter material. I may be able to get aspen if it's significantly better, but it's going to cost more money.

From contributor O:
I do not think (tulip) poplar would be out of line or risky for this, but I would prefer soft maple. 3 ply stiles from 4/4, I would forego the stave core construction. Face and edge and plane, then laminate to over 2-1/4, then face and edge again and then final plane will give you the most control on the straightness of the stiles. If your lumber is decent, you will have no problems.

We do 2-1/4 thick interior doors all the time, and have a series of 22 now in the shop at 10' tall, in alder. Why do you have the stave core concern? If these are interior the environment will be stable. Unless the stiles are inordinately wide, or if these will be hinged together as four panels, the movement across width will be manageable as single or paired doors. I think the stave core will just add labor for no real gain.

If you must use MDF you might consider making them of 1/2" MDF faces with a torsion box type construction to keep the weight down. Miter the raised (wood) molding around the panel, or raise it after assembly. Turn down the vacuum on the bag press, and use yellow glue and the torsion boxes will come out fine (+/- 6" x 6" voids). They don't need full pressure, just enough to keep the surfaces fully in contact. Do a dry run to find the best pressure settings.

The panels can also be done with all solid wood, with two separate, conventionally raised panels with either a framed spacer floating between the two panels or two plows for the panels (one plow for each panel) with a center section left to separate the panels. It requires about nine pairs of hands for assembly, but keeps the excitement level up there.

Tenons are to be about 1/3 the thickness of the members, so shoot for 3/4". This may require two passes on your mortiser, but if everything is planed to the exact same thickness, flipping stock for wider mortises will just keep everything centered. I would shoot for 2" tenon length as a minimum. Double tenons would be OK, but more work. I think you would still need the length even though you are doubling glue surface.

Being a contrary purist, I would look at the existing doors and try to copy them. They obviously lasted and were valued enough to be reproduced, so your product should be able to stand next to them in quality and value.

From contributor D:
I concur that a 3-ply stile is plenty sufficient for interior use. It doesn't cross my mind these days about this, as we have lots of core made ahead of time, and just rip what width we need, add edge bands and veneers, and put it in the rf press. If you're set up for it, you can actually come out less expensive labor-wise than going all solid. One of my obsessions also is bookmatching the veneers, kind of a game I play that gives me great satisfaction.

From the original questioner:
To contributor D: Having read through pages of threads on the archives it seemed like stave core was the way to go for doors, but if your method works well then I'll have to reconsider. It would save me considerable time!

I'll be alone for most of the assembly so I'm going to keep it to three panels per door for ease. I like the MDF as there's no real concern for seasonal movement (makes a nicer paint job). I agree that machining it is less than fun.

I can't copy the existing as there is no existing, at least not wood ones. The entire floor has been gutted to brick and steel and is being built from scratch. The unit was in pretty rough shape and had a different style door than what I'm making judging by the remaining elevator and staircase doors. They were long gone by the time I was called in, so no chance at inspecting the originals, (well except for the steel clad elevator doors). There’s no chance of duplicating as they don't like the existing anyway.

I want to say people these days want things to look like they may be old, but aren't really interested in sticking to what might be considered correct. They just kind of mix and match moldings and door styles or whatever to what looks good to them. So the six panel original doors may have been original to the building but don't fit their taste, so they pick a new style and off we go.

You would probably appreciate the bookcases I'm building for them. A straight 14' length, and then a half circle cabinet with curved glass shelves that wraps around the 3' diameter chimney, followed by another 8' straight section. This is the best part - the anticipation of a really cool project is the best.

From contributor H:
Three ply same species solid works good for 2 1/4” thick plus doors if the MC is all the same. Not so good for 1 3/4” thick. Veneered stave core is better for these. Vertical and rift grain will make a better door in three ply construction but will be hard to find in poplar. That is the advantage of the typical veneered stave core construction. If the staves are ripped from flat sawn material they will automatically be rift or quartered in the finished stave. As mentioned making veneered staves is labor intensive if you don't have the right equipment.

We stay away from poplar because all we get out west is the fast growth wide ring sap wood material. It is more expansive than what the wood calculator shows. I think it is a regional thing with the east coast shops getting better stock. Tight growth rings will make a better core in any species. Interior doors have to be stable in environments that are highly heated or cooled. In Colorado it can get down to 3 or 4 % inside in the winter months. During construction the contractors crank up the heat to dry out gypcrete, flooring, and plaster. If the doors survive that they will usually make it.

I question the need for a 2 1/4” thick panel. I would push for 1 3/4” thick. The 2 1/4 is going to cost a lot for what you might gain in design appeal. Double panel raising is a question of math. For a few doors single side is not too bad. 20 to 60 plus interior doors can get very time consuming with a single head and this is where the double setup pays off. We do all the exteriors back to back single anyway. Plank doors are more popular in Colorado so we are set better for that. We use a variety of old three wing braised panel raisers and a 250mm Garniga insert multi profile panel head for most of our work.

If the right job came along I would buy the mate for the Garniga head. We just did a giant 4” thick entry pair with 3” thick custom deep profile panels. We had corrugated knives made to work at a 30 degree tilt for safer knife projection. In hindsight I would have gone for a custom three wing braised cutter if doing it over. Insert panel cutters make the best cut with three wing braised not far behind. Keep in mind also if you get into a lot of small size door raised panels it is safer to do them single side, cutter on top. For us set up is no problem for double or single raising. The key thing is to have your panel material flat and accurate thickness.

From the original questioner:
So it seems the like it's pretty much agreed that three ply lamination is fine for 2-1/4" thick doors. This is great news as it will save me a lot of labor not making the staves. I'm going to avoid poplar and build them from soft maple as that's what I would have used for the faces anyway. Unless someone has a better alternative that's reasonably inexpensive, stable, and takes paint well? I'll build them in the same manner I build my solid doors, letting the wood rest in between milling operations, and picking the straightest pieces for the stiles.

I have a head for the raised panels. If I remember correctly it's called the Innovator head. It uses three corrugated steel knives which register against a round head machine screw on the bottom. I'm not sure if this would work well though as the MDF will kill the HSS pretty quickly. I could get the laminated carbide steel but I’m thinking maybe a brazed cutter will just be a better choice. I don't mind spending money on tooling. I budgeted in a good amount for this job. On the other hand I do maybe one raised panel job a year so not sure investing in an insert head is a good use of money?

As far as the panels go that's my only concern doing them in two passes. I'll use the vacuum bag to glue up so they should be reasonable accurate. I just always worry about deviation between the sheets of MDF. I'll give that some more thought before I make my decision.

From contributor N:
If I could find 10/4 or 12/4 stock I would make them out of solid stock. It would be less work and you would avoid the chance of the three ply lamination warping from the moisture introduced from the glue. This kind of lamination is very hit or miss and really is no more stable than solid stock anyway. If you can't find solid stock the stave core would be better than the three ply lamination.

From contributor F:
I have constructed well over 1,000 doors in my career so far, most of them being interior paint grade. Most of them have been made of solid poplar rail and stile construction. This is the most economical way to build them. I also have built a lot of stave core paint grade as well. My rule of thumb is any door that is 7' 6" and taller is stave core or I do not build it.

In response to your questions, I would use solid poplar if there is a budget to keep in check with. My next choice would be stave core. The multiple piece glue-up is time consuming and I have seen it move just as much as solid stock. I will say that poplar is not the most stable wood, but if you dress the material the right way you will have very good luck using it. As for joint construction, I use a hardwood floating mortise pin construction at all joints. The pins are 1/2" thick by 2 1/2" wide by 3" in length. I have never had a door joint fail, interior or exterior or 1 3/4 to 2 1/2 thick.

Keep in mind all the glue surface area of your cope and stick along with the surface area of the mortise pin. Panel construction on paint grade is best done in MDF, in your case I would use Ultra Lite. It saves on weight and is a little easier on the cutters. Also you can use two pieces of 1 1/16", profile with one cutter glue together if you choose to and your good to go. Just don't forget to use space balls to allow the panel to float. I also send all my doors out primed, this helps to insure stability until they are finished painted, on all six sides. I have a lot more info that I could share with you, (cutters, stave core construction, etc).