CNC Versus Conventional Equipment for Entry Door Construction

Woodworkers wax poetic as they discuss the practical considerations, the historical ramifications, and the spiritual significance of evolving technology in the field of door construction. January 23, 2014

Regarding entry door production with CNC, has anyone devised methods for cutting the cope and stick joinery on solid wood entry doors? I expect you would need some jigs to hold your rails at 90 degrees to your cutter for the copes and more jigs to allow another cutter to run profiles on stiles and rails?

Forum Responses
(Architectural Woodworking Forum)
From contributor D:
I will start by saying that I am staking out the far edge of opinion on the subject. Second, I'm not going to answer your question.

CNC may be great for making passage doors, hundreds of 'em. In fact, the makers of the equipment are counting upon it. At every major industry show, there are several behemoths buzzing out little stubby tenoned cope and stick doors. "The ultimate woodworking challenge," bellows the sales rep. Crowds stare agape. Most think this is where it is at.

I have no desire to convert or to use one to make my door parts. Give me one, and I'd sell it. I love the process of making doors. I have worked at it and refined it for close to 40 years, and find them still the most rewarding and useful things I make. I designed the tooling and the process and have everything calibrated just so. Yet, I have retained full versatility so I can alter things when and how I want. Even on the fly, right in front of the machine, if I see a better grain pattern or just because I want to. And no damn silly little stub tenons either; but real ones, inches long, fitting into precise mortises with square corners. How do you crown a stile for tensioned latching with one?

To my way of thinking, and I may be the only one so warped, making doors is the heart and soul of my life as a professional woodworker. Handling solid woods, beautiful woods, and fitting real joints that suck themselves together is a joy that is indescribable to those that have never experienced it. Real heart and real soul.

The damn CNC machine is a soul eater. Yes, it may make money for you, but you will trade your soul in the bargain. It removes the woodworker from the process and relegates him/her to operator, soul not required. No emotion necessary. The operator must become as the machine - cold, sterile, predictable and intellectually dead. No crown required.

There are probably shops that can use such a beast and not lose their soul in the process. Shops that learn to use the machine as a tool, a means to achieve an end, not the end itself. Those mass marketers would have us all believe that it is the machine that is the key, not the creative and inventive uses of it. Buzzing out countless parts all day is neither creative nor inventive, nor even productive.

My doors make me sing, or dance, or cry or cheer. Sometimes all at once. I can take real pride in what I - and my coworkers - produce. The doors keep out the winter winds, the bad guys, the raccoons and the mice. They protect and enhance the quality of life of those that live within. They set the bar high - too high for most, but are not to be trifled with. They scare away my competitors. They feed me and my co-workers.

For those that care to look, they can see what goes into my doors - the depth shows. And continues to show over time. Those CNC doors are as lifeless and sterile as the sleep inducing process. Heartless little things, grubbing up resources and hours of our lives and returning us nothing.

Yes, that CNC may make you some money - if you don't become enslaved by it. If it's money you want, go be a lawyer or insurance salesman. Woodworking - that is the real kind of woodworking - is not for money, it is for the heart and soul, and a nice livelihood.

From contributor L:
Regarding a recent video… I thought it took forever to get that door done. It would have taken about 3 minutes to run all those parts on a shaper. Probably less. I couldn't believe how many times the CNC repeated operations for whatever reason. But in the end the door got made.

11/9 #5: entry doors on CNC ...


From contributor E:
Elegantly stated. Although I'll bet our predecessors had a similar lament when their competitors started using power tools.

From contributor D:
Contriubtor L, that is what bugs me about most of the CNC work I have been able to observe - the process is terrible! Any decent woodworker would find 8 different ways to do it faster, with no sacrifice in quality or yield. If it is ever to be relevant, CNC work needs to be done by woodworkers first, programmers/engineers second - or last.

Contributor E, I agree, it is all relative. Each new thing has its place, and I'm not sure CNC has found it yet, since it is still being pushed as the answer to any problem, be it production, complexity, scheduling, labor, or constipation.

I love insert tooling, as well as power feeders. The shop I trained in (circa 1974) laughed at both as unnecessary and perhaps even girly-like. Both are an integral part of this shop that I would not like to do without.

From contributor J:
All emotions and "art of woodworking" sentiments put aside, I don't believe there is anything to gain either in speed or quality. More importantly you would never produce hundreds of doors on that thing - way too slow. It might gain you some time on a 5-10 door order.

However there is a happy medium in there somewhere. I have a CNC router I use in conjunction with all the standard goodies and I wouldn't live without it. Dead nuts accurate templates, radius forms, jigs, arched panels, on and on. I appreciate the art of woodworking as much as anyone, but the glory of spinning patterns on a bandsaw or router are long gone. I will also add that even though the technology is advertised as plug and play, a good router guy is as skilled and savvy as a good shaper man. It's still woodworking and any factor that comes into play on a shaper or moulder is there on a CNC router, if not more by nature of the tooling.

From contributor Y:
Pretty sure I could run those parts faster on a shaper at less cost than on a router. No issues with programs, or precision placement.

From contributor F:
I will agree that the process could be more efficient, which would cut down the cycle time. I timed it at around 7 1/2 minutes, which I think could be around 4 after reprograming. That said I would like to see that done on 8 machines faster and as accurate, as stated above. Or are they going to change tooling for each operation on their one shaper in 3 minutes, which is not possible?

From contributor L:
$100,000 CNC versus a few $3000 shapers. For the money you could have an army of shapers all setup differently and still be cheaper and more productive.

From contributor D:
With several machines, and typical workers running them, a door will get made in less clock time than one precious CNC. The individual/several machine approach also allows crowning stiles, matching grain, adjustable sizing and part reclamation, fitting on the fly and all the basic things that help make a quality product.

When running a batch of doors, the stiles and rails are all profiled as an endless board butted through the shapers, as are panels, rails, muntins and glass stops. This happens very quickly and often takes less time than a changeover.

The CNC process comes from engineers that are detached and unfamiliar with woodworking as a process. It will take real woodworkers, working with these things every day, to bring together programming and process and ease of work to make these things truly useful.

From contributor B:
Wow, I didn't expect to see so much negative sentiment toward technology on a web site...

I understand what you guys are getting at about the CNC being cold and sterile. But remember, the machine has to be programmed by a human, who spends hours upon hours refining and perfecting the craft. It could also be said that a shaper is cold and sterile, but when set up by skilled hands can sing a beautiful song.

I program a CNC at a place where we create high-end, custom millwork. I have 18 years of woodworking experience working in some the most exclusive homes in the country, and now I use my knowledge and skills to make a CNC sing a beautiful song.

From contributor D:
The subjective part of my response, is as I stated, far from what might be the normal opinion of professional woodworkers. Also, the discussion is about passage doors, at which the machines I have seen do not excel. I do not think doors translate to the CNC machine well, whereas cabinet sides or curved pieces of MDF translate very well. Finally, it will take a good woodworker with real process awareness to make one dance or sing, and that is apparently where you have succeeded. Do you guys use your CNC to machine every part of a passage door? With my work and my attitude, there is no place for one in my shop.

From contributor B:
We can use the CNC for all parts of passage and entry doors, but we typically use a combination of CNC and shaper to make things more cost effective. Our products are custom and specific, so more often than not we can't re-use our programs. However, if we did mass produce doors in large quantities, we have the capability to do all parts (stiles, rails, panels) of passage/entry doors on the CNC.

We have tried to integrate the use of technology into old school products. It takes some creativity and a little more time on the programming end, but it can still be a fun thing that keeps you in touch with the wood.

From contributor C:
They are just a tool and like any tool, in the hands of a good craftsman, superior work can be produced. The only difference is you also need computer skills to go along with woodworking skills. As far as crowning stiles or grain matching, it is no different than any other machine. Parts need to be marked and placed correctly on the machine.

Even though it appears to be moving slow, a well optimized router can do all the operations on a part with no further machining required, saving considerably on material handling.

One example would be machining a passage door using a pod and rail machine. With the right tooling the normal stile and rail panel cuts could be made along with the joinery, be it dowel or M&T. In addition the lock hardware, hinge mortising, beveling and jamb cuts could be made in the same setting. All this should take about 20 minutes and the operator could be assembling the last door while this one is being machined.

A lot of shops have been successful with these for door work. It is a costly piece of technology to set up and I think care is called for. Depending on what you are doing the tool and software cost can be as much as the machine.

You should post this on the CNC Forum for some positive answers. Holding the work piece on a pod and rail machine is pretty simple, but some shops are using flatbed routers for doors with success. In Europe, for door and window work, this is the emerging technology, especially for shops that need flexibility.

From contributor F:
Funny - no one's mentioned that when that CNC is doing its door routine, there's nobody needed to make it happen. It may be slower, but the person who's making it happen a little faster on a shaper could be doing something else while the CNC is crankin' 'em out.

Not sure about the resistance expressed here... automation as a bad thing. For those that believe that's the case, you should be throwing away all your jigs, your routines that increase efficiency, and just make every item as your mood dictates each day.

Reminds me of those who badmouthed the Iron Horse.

From contributor L:
You aren't supposed to leave a CNC unattended, ever. Fire is a worry.

From contributor D:
My comments are mine only, and admittedly a minority opinion. I'm not looking for converts, just stating buyer beware. I don't pretend to have the definitive answer, although I am inclined to agree that good work with a CNC that is well integrated into a production process as an answer to whatever the question is. For a larger door shop, the CNC will definitely have an important place in the production of good quality products - especially curves. It is not an answer for a smaller producer.

I am at the age - and attitude - that if I want to do it a certain way, that is the way I'm gonna do it. Life is short, and I'm trying to make the most of it. Fortunately, business is good and I can spend some extra time doing something I enjoy as opposed to slamming it through yet another step. I can also spend some more time teaching an employee several ways to do something - showing - rather than talking. I can play if I want, which is very important to my ability to stay on top of everything and preserve sanity.

As for the Iron Horse, I like the story of the French workers that made wooden shoes - the peasants' affordable footwear - in the late 19th century. They resisted the introduction of shoe making machinery, so they removed their shoes - sabots in French - and threw them into the gears, jamming the machines in a futile attempt to preserve their jobs. We get the word "sabotage" as a result.

From contributor J:
The CNC is as much a part of my door building as is my collection of shapers. I look at my operation and believe I have merged the best of both worlds. There may be many success stories of companies building a door business around their CNC routers, but I personally I don't know of any, and I've been around a long time. However I know of four shops of close proximity to me that went that route and are gone. One Masterwood, two Busellatos and a Komo all purchased with the sole purpose of building doors. The Masterwood crashed the first day and the damage was in excess of thirty grand to the tooling and the machine. In my neck of the woods, that's a lot of freakin' doors to make up thirty grand! I could have bought the last Buselatto for less than fifty grand from the lease company and still couldn't justify it to myself - it was 250K new and only three years old. Good news is they were all competitors.

Lastly and most importantly, I don't think they do a good job. The thought of pre-beveling and pre-mortising for hinges and hardware sounds great. The reality at least for me is the smallest amount of tearout doesn't work, undersized doesn't work, raised or rough sticking doesn't work. Every door I have seen at the shows and/or from my former competitors showed those traits.

From contributor D:
I think you have it right - both in the drawbacks and the future role of the CNC. Doors simply require a certain level of craftsmanship that the machine can't have. Unless you make doors, you just don't know this.

From contributor C:
I am not beating the drum for using CNC routers in door production. I looked hard a couple times at buying one for our production but could not make the numbers work. Mainly because I would have to repurchase all my tooling. However I just want to point out it is a viable method of door and especially window production. Not many shops can justify one and fewer have the skills to implement them properly.

We have had a lot of shop owners to our window building workshops that are using routers. Some have had the nightmares you describe, others use them for part of the process and others use them as I described for the whole process. From what I have seen the most successful users are the highly custom midsized shops with 5 to 10 employees with either the owner or operator skilled in both woodworking and computer. The larger custom door shops seem to prefer the CNC for curve work but like the higher production of track fed single end tenoners and prehang machines for straight door production.

I think the router has been more successful in Europe just because the trade schools there focus on CAD and using CNC machines. I would suggest to anyone interested in learning more about CNC for doors and windows to take one of the tech tours that Weinig, Homag, SCM and others offer in Europe to visit shops using them. It will open your eyes.

Contributor J, can you explain what this means: “undersized doesn't work, raised or rough sticking doesn't work.”

There is a lot to making these work right. I have a sash in my shop that was made by a now defunct manufacturer. The doweling was off a noticeable amount just because they were doing the drilling after coping and profiling. You cannot always blame the machine.

From contributor J:
The folks I saw using routers for 100 percent of machining were not getting good results. Lots of tear out in the copes and sticking. I just don't see the easy adaptability to all the different wood types that are needed in the custom door business. You know as well as I do to compete in that market segment that the CNC is suited for, you face a lot of wood varieties and challenges. I can run that hickory part on my shaper and if I get a little tear out, nudge the fence, run it again. Hit go on the CNC and you are committed! If you are running a 30-40 door custom order in some high priced wood, a handful of ruined parts is devastating. The thought of gluing up doors that are pre-beveled and pre-machined for hinges and hardware is cool, but if that door isn't exact after sanding, once again the target clients I need to hit are not happy.

I feel there is a misconception as to who the target market is for these machines, and don't know that there is one. You feel the 5-10 man shop; well, I am sure we can all agree Simpson is not cranking out box store commodity doors on a CNC router. So the small shop now competing in the upper semi-custom market starts out by sinking 150k plus another 30k in tooling to build 60 doors a month doesn't seem like a good ROI. Sure it's cool that this machine can do all this in a spiffy self-contained unit, but based on who you have to sell to, they better be perfect. You still are limited by what the market will bear as to what you can charge for doors. So a 20k investment for a handful of shapers and a doweling machine could easily accomplish the same thing for the small shop.

Window production may be better suited. The parts are smaller, more consistent species, less tooling and the machines can be a lot smaller/less expensive.

From contributor C:
True, at 60 doors per month you don't want to consider one of these unless they are very complex, expensive doors. The path we are on, building doors with numeric tenoners, profilers and shapers will end up costing as much as a router in the end. It can be done in small steps, though. Going router requires a commitment to jump in fast.

We did the dedicated shaper method for several years but that is not so good if you need to offer a lot of variety. Unless you have a lot of floor space to keep adding shapers.

I still do not see the issues with machining if using the right tooling and technique. Since there is no backup on coping, the blanks need to be wider than for normal machining to be able to clean up with the sticking cut. I believe they have entry and exit methods that help with this. We went all insert tooling a few years ago and have a few issues with tear out even in difficult material. I think it would be the same or better on a router as long as the work is held securely.

From contributor S:
Do this on a shaper. Started out as a square piece, was sized round stick profile was done all the way around drilled for four 5/8" x 5" dowels, then did the carving. The CNC is the best and by far the safest tool for doing shapes.

Click here for higher quality, full size image

From contributor L:
That's not a door.

From contributor S:
No, this is the door.

Click here for higher quality, full size image

From contributor J:
Very nice. So now the million dollar question. You clearly have a pod and rail machine of some type and you clearly cut door parts on it. Would you or did you buy that CNC machine as the principle piece of equipment for your door manufacturing?

From contributor S:
The CNC is our main machine for making doors. Stiles, rails, panels, jambs can all be done with the pod and rail. Cut quality and blow out are not an issue regardless of wood species. We do own a 7.5 hp shaper that we started with 8 years ago, but don't use it much anymore. I have been planning to purchase a Martin shaper that can use the same HSK tooling right out of our router to machine small parts. Would I do it again? 100%. That said, these machines are not for everyone, but I do love incorporating design elements that I know my competitors can't do easily, like weird shapes, carvings, complicated mortise, multipoint hardware, etc. Once a program is done, you have it forever.