Joinery for True Divided Light Doors

Woodworkers discuss the best way to attach muntins to door stiles and rails for true divided light cabinet door construction. June 2, 2009

With our new 1 3/4 ext/int doweled door line we've been experimenting with true divided lite muntin attachment. I'm curious to know how some accomplish this. For instance, for a typical 3-0 x 7-0, 8 lite exterior door (cope and stick, doweled), is doweling the vertical and horizontal muntins to the stiles and rails time and structurally proper?

Forum Responses
(Architectural Woodworking Forum)
From contributor C:
This is the stuff that separates the men from the boys when it comes to efficient making of these in a custom woodworking shop. I have found that with some simple fixturing and a small mortise for a loose tenon setup, you can produce several doors a day once standardized. How about tall trophy case doors, genuine 15 lites and mullions from 5/4 stiles and rails double beveled back to back? I typically cut the stock oversized, mortise thoroughly while square, then shape all my copes and pattern cuts. Being careful with alignment and chip out.

From contributor R:
The answer sort of depends on the profile involved. If you are using a profile that includes a stub tenon and then doweling your door at the main joints, then coping the same stub tenon onto the muntins is sufficient. If you need to use small dowels, that's not a problem, you'll just need well thought out and made jigs.

From contributor J:
We dowel our TDLs to the stiles and rails and cross lap at the middle joints. We usually put in 2 - 8 or10mm vertical dowels on chunky muntins or, in the case of a historic job we just did with 3/4Ē wide muntin bars, we used one 5mm dowel in the profile area. We have a Hoffmann Borhe machine for doing this. Itís not a high production machine but accurate and easy to set for one-off frame work. For the cross lap we use a Morso notching machine, only because we had one. Stegherr makes a nice machine for this and well worth the money if you do many TDLs and SDLs. They also have a doweling machine similar to the Hofmann and automated machines for all aspects of door construction. Some shops do the cross lap using short pieces coped into either the vertical or horizontal bars. Myself, I like the rigidity of the cross lap.

Starting out I would do what works best with your existing machines and try to improve the process as you go. We built a lot of doors for years doing loose tenon construction with a marginal slot mortiser and doweling muntins with a very marginal horizontal drill. We got the quality but with a lot of work compared to how we do it now.

For mortise construction with TDLs it is a little more difficult without the right equipment. I did a few of these with an import hollow chisel machine and wanted to throw it in the dumpster every time I used it. Contributor C's router jig would be better.

The ultimate for mortise and tenon work on doors is the Centauro Beta machine. It is CNC controlled, uses a Maka type swing chisel for the big mortises, has a separate hollow chisel for the muntins and also an optional router for louver and hardware routing. I didnít time it but would guess about one minute to do a very complicated stile with double mortises and TDL prep. It is ideal for a shop that does a lot of different type doors. The CNC router guys have some interesting techniques for TDL work as well.

The issue of not doing any connection at the coped TDL is another option and used by some low end makers. In theory the glass beads, when applied, should hold everything together, but I would be worried about the effects of wood shrinkage over time, especially for exterior application. Another issue with this is the difficulty of aligning everything at clamp time.

From the original questioner:
Great info. Thanks. We've been experimenting with small diameter dowels at muntin/stile/rail connection. We have a Hoffman PDS portable doweling machine for the stiles/rails and the Delta horiz and a jig for the muntins. We've had good results with accuracy and strength, but these methods are for experimenting only, and if we proceed we'll be looking at other equipment. The fitment we achieve with doweling and coping is impressive. I think this is worth pursuing.