Efficient Door Construction
1. Rough crosscut
Am I wasting my time with stock prep, or am I saving callback hassles? Should I be pushing this issue as a quality thing with clients: "I build doors the old-fashioned way", or is it a non-issue?
From contributor R:
I started out 18 years ago doing it precisely as you described. Then I tried a couple jobs by slowly modifying steps I thought might be unnecessary and modifying machinery until I got the results I wanted. I ended up with this process which gives me top quality results and negligible defects. All of this also depends on the machinery you have for building doors. I have four dedicated shapers set-up all the time to do these steps for me, which I customized for my own needs. One big difference is that I use shapers with outboard fences for step numbers 6 and 7:
1) purchase lumber hit or miss @ 15/16"
All of the steps you indicate survey the purpose of solving a condition in working the wood that's always there. And so you do have to do those steps as you describe, or find a way for the machinery to automatically do them for you. I find for example, I can simply select out of my SLR2E stock the straightest and flattest stock for the longest pieces I need (by the way if you don't already do this, cut part lengths from longest to shortest).
I then work my way down to the smallest pieces where a little bow or bend in the piece will automatically mill straight because of the way I have modified my fences, thus eliminating the one step of pre-straightening, and so on. If you don't have a wide-belt or drum sander, flattening and straightening as you do is a must - simply for a good surface quality when it is time to sand. You should always push the issue of quality when you have some over the other guy, but the customer doesn't care how many steps it takes to build a door, but how does it look?
From contributor F:
I pretty much do the same as Contributor R, except after step 6 I plane the edges of the stock (usually to 2.25"). I am curious how lumber can be face jointed with a power feed? When I face-flatten or edge-straighten material on a jointer, I do it by hand and I am careful to refrain from pushing the bow out of the boards.
From contributor R:
To the original questioner: I myself used to take all those steps trying to build a quality door, but this is the best way I have come up with in 20 years. I rip 13/16 lumber to 2 1/2'' wide, and cut the rails and stiles to length (factoring in the fact that I will be burning an 1/8'' the shaper). I run insert tooling - running every thing face up, so any thickness difference is on the face side of the door. I stack the cutters on a super 27, with a sled built too exact thickness so the spindle never moves up or down and I don’t remove the cutters. The in-feed and out-feed fences are set to .001 and never move.
I don’t need or use rub bearings, I burn an 1/8'' on the stile cut so the door stock is 2 3/8''. The panels are then dry and I cut them to size and run them on another shaper stacked with a panel raiser and edge profiler. Once the are on the assembly I glue my rails, slip them into the stiles, flush them up, and pin nail them. It is a lot easier to flush them up and tack them than to flush them up with clamps and then pin them.
The clamps override the pin nails and pull the joints tight. I lay the clamps flat on the table because that eliminates cup problems, and since the rails and stiles are 2 3/8'' wide, I have 1/8 to play with squaring them up on the radial arm saw. If a sile gets trimmed down to 2 1/8'' wide in the middle because it was bowed, it's on the inside edge, not on the outside where it shows up. I can square up a set of doors in just a minute on the RAS – a lot faster than I can run everything through the joiner. I then edge profile and run them through the drum sander, and then sand the face side first. Running them 2-3 light passes gets them very flat. I then sand out any thickness difference from the stiles and rails, and hit them with the orbital and I’m done.
From the original questioner:
Thanks for the suggestions. First, I should clarify that I do a range of doors with frame widths up to 4", not just 2-1/4". Wider stock tends to be more susceptible to movement, no? I also do all solid panels, and reverse raise them for shaker doors. My customers pay the extra cost over plywood panels. Door factories seem to get higher cost when you get into wider frames and solid panels, not to mention unusual woods
Contributor R - Good advice to start eliminating steps. I'll start by going to 15/16" H&M on all of my lumber. I already crosscut all longer parts first with the straightest stock.
Contributor F - For face (and edge) jointing with the feeder, I run all wheels on the out-feed side, so that they reference off of the already jointed surface. This ensures that the wood is really flat. I am also using an outboard fence for sticking cuts. I edge sand all frame parts by running them between the table saw fence and a special sanding disk. This straightens the outside edge as well, if it has bowed when ripped. The sticking cut makes the inside parallel to the outside, and brings it to final width. I use various plastic spacers to set the outboard fence, and my shaper fence never moves. I cut all parts to finished size, so assembled doors just need a light hit at the ends of the stiles to remove any glue squeeze out.
From contributor M:
When I started building cabinets, I chose to build my own doors because I thought it would be cost effective. After I built about 20 doors and noticed that it took me about 10 hours to do so, I felt like this was a waste of my time. Doing all of this work and putting myself in more danger to pay myself about $5 an hour was not worth it. I like the person that orders them and I can purchase most doors from $18 to $25 unless they are oversized. Buy the doors and build more cabinets.
Would you like to add information to this article?
Interested in writing or submitting an article?
Have a question about this article?
Have you reviewed the related Knowledge Base areas below?