Building a door

[StephenM]

Pardon the repost... it's been 24 hours and it still has not shown up:

I am wrapping up a bedroom remodel and have yet to build the passage and two closet doors. I am building rather than buying because I would like them to match the rest of the (1860 vintage) house.

These are four (flat) panel doors with a simple shaker-like bevel surrounding each panel. The center and lower rails are about 8-inches wide. The center rail meets the stiles with a twin through tennon rather than a single tennon.

I have always designed wide rails with a single tennon, pinned in the center with unglued "stubs" that sit in the panel groove to mitigate any potential cup or twist.

While I suspect that the wider stance of twin tennons would do a better job of controlling overall wracking of the door frame, I would think that the crossgrain implications would not fare much better than a single full width tennon. That is, shrinkage could cause the center rail to split.

Is there some other compelling reason to use a twin-tennon design? And does the availability of modern glues change what makes sense today vs. the original design?

Segue into question 2: Should I avoid regular yellow glue because of "creep"? The door, being constructed of ash will be fairly heavy and enjoy significant racking forces from gravity. The through tennons will be pinned, so perhaps it just doesn't matter.

Thanks,

Steve

** Posted from **

[Denis M]

I have worked a lot with ash. On a solid ash door of 18 inches wide it will expand 1/4"+ between winter and summer. So allowing for expansion is a must. As for glues I have used in the past yellow carpenters type with fair results. Now I only use the exterior glues made by Lepage and Titebond lll or better for ref. see

?prodcat=1These glues are good but when subject to stress they have to be used with mechanical fastening devices like screws or joineries. Good luck. Denis M, Rothesay NB

[Swingman]

"StephenM" wrote

Actually, what you want are "double" tenons, not "twin" tenons, in joinery parlance.

(twin tenons being side by side within the thickness of the rail).

Glue was never the reason for using double tenons, particularly on the lock and kick rails of a large door frame, it's the mechanical strength of the joinery itself that has stood the test of time in door making, irrespective of the glue used. So, yes', there is a compelling reason to use a "double tenon" design, at the very least on the lock rail.

While you're at it, a good practice is to also use a haunch between the double tenons (in rail width: 1/3rd tenon-1/3rd haunch-1/3rd tenon), which will help prevent the rail(s) from warping.

[StephenM]

I never really thought about the distinction. Good point.

If I understand correctly, make the point that the mechanics of the double tennon joint is what provides strength. Would a single-tennon joint (of equivalent aggregate tennon size) not be nearly as strong without the "captured" cross-grain geometry which might be prone to splitting?

** Posted from **

[dpb]

StephenM wrote: ...

quibble..."tenon" is only two "n's" total, not three...end quibble

If a single tenon of the same aggregate size were used instead, the distance from the top to the bottom would be less than that same distance from the top of the top to the bottom of the bottom double tenon. Hence the moment arm of the double resisting wracking is larger than the equivalently-sized single. This, combined w/ the noted grain orientation is advantage of the double.

--

[N Hurst]

Does anyone have any book recommendations or links to share for door building? One of my dream projects is to build new doors for my house. Nothing fancy, just frame and panel type based on another house from the same timeframe (~1960s). I want to do a good job, but I can't find much to help me out.

Plus, it just wouldn't be right for me to go visit my grandmother in law and ask to break apart one of her doors to see how it was built. I don't think the family would like that at all. :-)

Thanks, Nathan

[dpb]

N Hurst wrote: ...

...

Didn't look for the specific link but there have been two or three articles within last few years at the outside in FWW and/or FHB.

My favorite "how to" was a reprint Delta used to include w/ their spindle shapers but it, unfortunately, is now out of print and no longer available. My copy got smudged enough as to make it an unsuitable candidate for scanning or I'd post it somewhere.

There's a book on architectural woodworking also published by Taunton press I know--I don't recall who was the author but I've never seen anything by them that wasn't at least adequate. If you have a decent public library, I'd think there would be a section there that would have something--I don't have anything specific in my library, sorry; I rely on the Delta sheet referred to above so don't have a specific reference. I remember a text from a college shop course I took as an elective that I was to cheap/broke to keep but don't even remember the author.

The biggest difference between current practice and older is the tendency towards the stub tenon and dowels in commercial doors and windows as opposed to the older full through or tenon. To cope the matching mold on the door required a stub spindle cutter which is very difficult to find any more (in fact, I don't know of any generally available any longer since Delta quit making theirs outside the commercial manufacturers) other than the stub router bit set from CMT. To make the cut now is typically done by mitering the sticking instead outside the factory setting.

[Lew Hodgett]

Take a look at the NYW project Norm did a few years ago.

Lots of good info.

Lew

[Swingman]

"StephenM" wrote

Mr. Marks has professed an ignorance of traditional joinery in his younger days in the past, and in the joinery he used on this particular project, IIRC, was loose tenon joinery and I don't recall him using double tenons on the lock or kick rail ... I could be wrong about that, but it would be worth checking out if you can get a copy of that episode.

While I'm a big fan of loose tenon joinery (I own a Multi-router that is used strictly for that purpose), I'm not too sure I would be completely comfortable using single loose tenons on a large door, particularly on the two traditionally wider rails (lock and kick), and without a haunch ... probably wouldn't be all that concerned with strength double loose tenons as much as the lack of the haunch in between which logically, and in practice, seems a bit of extra help in mitigating warpage of the wider rails over the long haul.

Don't want to sound preachy, but nothing wrong with a belt and suspenders approach on doors as warp/bow in any door component, regardless of the size, is eventual hell to pay, so do what you can to mitigate as much as you can with your joinery, as well as taking extra care in selection, milling and acclimating your stock before assembly ... IME, the latter is just as critical as the former.

[dpb]

Swingman wrote: ...

What he said... :)

Only add that in selection, species is important particularly for exterior doors; you can get away w/ most anything interior if going exotic.

And to add, in selection quarter- or near-quarter-sawn stock is preferable as it will provide the most stable dimensional material you can obtain of a given species outside composites.

I'd not feel uncomfortable at all w/ the loose tenons on interior doors although I certainly don't disagree w/ the logic on the haunch. On exterior, I'd feel far more "iffy" and wouldn't dispense I don't believe; certainly not on anything over a 2-0.

In the end, despite the heft (a 7-ft 2x8 lock stile blank is pretty good chunk), door and window making is a very enjoyable task imo(*). One thing almost required imo to do it with any speed is a minimum 60+" jointer table--anything less is a real hassle.

Enjoy!!! :)

(*) If I were in a more populated place where there were enough demand, my "retirement" job would be to open a small, custom architectural millworks...