Custom door window and shutter maker here. I don't love it when rails wider than 10" are spec'd on a door or shutter. Doors are almost always domino'ed unless they are massive doors over 3-6 wide or 8-0 tall. In that case they would get stub tenons. Any rails wider than 10" I'll usually engineer with a composite core like extira to mitigate the movement. I make shutters with pegged through tenons. Can't really engineer a shutter rail since they're usually only 1 5/16 thick and require tenons which would be wasted on a composite product like extira or exterior ply.

Recently I've had a set of large shutters spec'd with am 11" wide bottom rail. I've seen some historic shutters with rails this wide fork tenoned (or double tenoned, if you call it that), but wont that encourage a crack in the event of contraction? I've also seen them tenoned the whole width besides a 1 1/2" haunch at the bottom, but either way, gluing that tenon will trap it. We use QS sapele (8-10% MC from the distributor) and titebond III.

I suppose the other option is a narrower through tenon maybe 4 1/2" width, and do a long haunch like 1/2" deep for the rest of the width and let it float with no glue to allow for the contraction. With this method, any contraction would be apparent because the bottom of the stiles would stick out past the bottom of the rail. This is probably the route I go with on this set, but would like some input.

Any wisdom would be appreciated.

Can't you use your idea but glue the bottom so the stile to rail joint at the bottom stays consistent and the rail will grow/shrink into field of the shutter?

Josh
Not sure why I didn't think of that... I guess I am always used to haunches being towards the ends vs towards the panel.

Titebond 3 has been problematic on exterior applications when exposed to high heat. If you do a Knowledge Base search some threads should show up. One of WW's most respected door guys David Sochar has posted extensively about the problems.

On double tenons I leave room between the two tenons. It seems to be working, I have been doing it for 35 years. Wide rails like that as a bottom rail tend to grow down beyond the stiles when they get wet. I don't put them in wet locations. Wide lock rails seem to do fine. This is all through mortise and tenon, coped and pegged.

I spent the first few years of my career asking why we forked the tenons on wide rails. None of my colleagues had an answer. The closest they came was “to keep 'em from cracking”. How and when and why they would not answer, and I felt I was annoying them. The best I came away with was that is what they learned - we always done it that way.

Whether a single wide tenon, or two shorter tenons, they are all glued and, as you say, trapped. That sets up for something to crack. Hence your good logic on a composite core.

I would make the shutter rails a five ply affair, with 3/8” to1/2” thick center core for tenons, then 1/8” faces, and the remainder will be vertical grain. This will give you a stable rail, plywood if you will. You get the strength of tenons, and grain orientation needed, but the stability of a composite core.

As to TBIII, yes their tech people can give you the info, but it looses a good amount of strength when heated. It is not unusual for dark wood to heat to 180 or even 200 degrees. This can cause the glue in question to loose 40% to 80% of its strength.

This tidy little fact once cost our shop to loose about $12,000 in free repairs to repair or replace panels with open joints.

I highly recommend epoxy for all exterior joinery.

The reason for “forked tenons” on wide rails is that a long mortise slot will weaken the stile. Especially in the case of early 1900s mass produced doors that had through mortises. I dislike 12” to 20” wide bottom rails found on some old doors because they usually are cracked due to the cross grain construction. When we have to reproduce these I leave the top tenons loose for width and don’t glue all the way up. All this haunching is a bit of work.

I would think you could use the same technique for coped joints with Dominos.

Never any issues with TB3 here but my doors don’t live in a hot climate. Also never edge glue exterior panels without doing any kind of joint reinforcement such as T&G or glue joint. I know the argument for well machined edge joints don’t need reinforcement but for exterior I prefer something in case the joint ever separates.
I will say that TB3 is inconsistent for viscosity and have had a few suspect pails that needed mixed a little. Lately been trying Jowat PVA. I like it so far but as it comes it’s a D3 rating. To get D4 a additive is needed then the shelf life is short.

View higher quality, full size image (1362 X 1703)

View higher quality, full size image (960 X 1280)

Joe Those are some pretty m&t joints there. When did you change over from dowels?

I would love to see your work.

David Sochar
I use both dowel construction and tenons depending on the work I’m doing. Tenons mostly for historic work as my tool set is designed for that. I build my 68mm and thicker Euro doors using window tooling and usually dowel those because the haunching can get complicated on those. Dominos are great for joining curve work to straight. Domino is a good choice for the small shop that builds just a few doors. Good dowel systems and efficient M&T equipment is expensive to set up. For small shops I am an advocate of using what you have on hand and can afford for the corner joints. Loose tenon of some sort usually fits this bill.

The first 20 years of my door building career I used loose tenon joinery and sometimes through tenons for historic work. A lot of doweling after acquiring a Hofmann frame drill. From a manufacturing standpoint it is the most efficient. AKA making a profit and meeting payroll!

View higher quality, full size image (960 X 1280)

View higher quality, full size image (1280 X 960)