Case construction methods

Discussion of the strongest, most efficient methods of joinery. July 29, 2003

Question
I am setting up a shop and want to develop a system that works for kitchen cabinets, entertainment centers, etc. What's the best way to construct the cases? My thinking is to use either dado construction with staples, or some type of biscuit setup for all case goods. My plan is to use pre-finished sheet goods and hope to get a vertical panel saw in the near future, as I think this will minimize potential scratching of the plywood. The scratching is something I see as a likely problem if using just my cabinet saw to cut all the sheets. Problem with biscuits is clamp time, I guess... but nothing to fill in after. If I were to choose today, I would use a 3/8" dado and then toe nail through the joint from underneath the shelves/bottom.

I have seen pocket screws, but I do not think the pocket screw system makes a high-end joint. I am not saying it isn't effective, but I plan to cater to the high-end market and you can get pocket screwed everything from Home Depot, etc. I am not competing on price, but rather quality. Any experiences or suggestions would be appreciated.

Forum Responses
(Cabinetmaking Forum)
You might search the Knowledge Base on this site. My recollection is that Jon Elvrum wrote a very informative article on this subject. I think most would say that a doweled system is the industry standard.



Another way to look at a pocket screw is as a threaded steel dowel. We dowelled everything for a couple of decades, then got smart and bought a castle drill. We use that drill for everything.

In marketing this approach, we are right up front about it. We show it to the customer and tell them that the reason we use it is because it is so strong and fast. We then reallocate those resources we saved to a custom pressed veneer or something else the customer might actually appreciate.

When everything is said and done, what we build is a box. Some look a little better but they are all just a box. Put your focus on the parts that make them look a little better.



I'm not sure a pocket screw is any more a threaded steel dowel than just a regular drywall screw. I wouldn't consider either dowel construction.

I prefer Confirmat type screws, which I feel really are threaded steel dowels because of their larger diameter.

I think all three methods are acceptable. I certainly don't think dados and toenailing are high-end compared to any of these methods.



A high-end joint means mostly strength, right? Five pocket screws along a 24" carcass side give you that. Confirmats are very good too, and the Zip R screws won't let you down. Dados are great for the New Yankee Workshop, who isn't trying to make a living. For those that are, screws. Pocket, confirmat or Zip R. If you have the equipment, though, doweling is probably fastest. And strong.


I use plywood for casework and biscuit-join them. Face-frames and door frames are joined using mortise and loose tenon. I fasten face frames to cases with biscuits. I use pocket screws for some things like fastening stretchers and hangers to cases, but use them sparingly. Clamping is a bit of a bottleneck sometimes but I try to keep the flow going by just working on the next thing. Whenever I glue something up and clamp it, I put a post-it on the item with the time it was glued written on it. An hour after that, it's okay to take off the clamps and move it. My machine for slot-mortising is shop-built and works really well. It's fast and accurate. I make lengths of tenon material ahead of time and cut off tenons as needed.

I've been building cabinets for 30 years or so but just starting in the actual business. I'm like you in that I'm aiming for quality rather than price. I enjoy working with wood and won't be building anything from melamine. I just don't like the stuff.



From contributor D:
I believe the best joint to be the strongest joint that you can make the fastest. Unless your typical client states that money is no object (and means it), coupled by you being one of the finest cabinetmakers, finishers and installers on the planet, you would do well to consider the most efficient method of cabinet construction possible, given your current equipment and experience. Remember too, that dados are not commonly offered by the factories, mainly because they are sloooow to make and require glue and clamps to assemble – not DIY friendly. Also, dados are the most common joints used by anyone that sets up shop in their garage, equipped with a table saw. Nothing wrong with dados, but you’ll be competing with a lot of part-timers and hobbyists who don’t have a clue how to properly price their goods. The people who buy from dime stores and home centers are unlikely to be interested in your custom cabinetry to begin with - unless you can beat their prices! In my area, custom frameless cabinetry is not common, being offered generally by the more established and well equipped cabinet shops. These are the shops that command the highest prices for their cabinetry.

I've been in business for only 13 years and started with dado/glue/clamps, building face frame cabinets. Clamps took too long so I eventually substituted 2" finish nails in place of the clamps, and the assembly process went much faster - still dado and glue, all 3/4" veneer core ply. I'd sometimes use a clamp (here and there) in order to press the parts tight before shooting the nails.

A few years ago, I read an article about using industrial hot melt glue as a method of fast and strong joint bonds. I tried that on case construction but the glue was faster than I was.

For the past couple of years, I have been researching frameless design and construction, (among many other parameters unique to frameless cabinetry) and have tried but rejected the following methods of case construction: dowels, pocket screws, biscuits, conformats, and staple/screw. All butt joints, of course.

I’ve been building frameless cabinets for about a year now and as before, strength and speed are a primary concern to me. Pocket screws and staple/screws are the strongest method of construction, in my opinion (with the possible exception of dowel construction), especially with PB core materials. That does not mean that all the other methods mentioned are not sufficiently strong for cabinet case construction. Dowels and biscuits provide for indexing of parts, and more than sufficient joint strength. Dowels are fastest but require a healthy investment in equipment. Biscuits also provide indexing of parts but can be tedious during actual assembly, and also require some form of clamping.

Pocket screws require a relatively small investment in equipment. They are strong, but provide no indexing of parts and since drilling must be done from underneath, can be tedious to get the parts aligned properly, and difficult to alter the alignment during assembly if the parts don't index perfectly.

Conformat screws require some form of specialty pre-drilling and I believe they don't provide as strong a joint as I get with #6 screws. If you are manually pre-drilling conformats, part alignment can be tricky and blowouts are a real possibility. There is little ability to adjust parts alignment during assembly.

Staple and screws provide all the benefits I'm looking for with very low cost in equipment. Armed with a dead level and flat assembly table, parts can be held in position while a few staples are shot, for the sole purpose of "fixing" the parts. The entire cabinet is assembled using only staples to begin with. 1-1/4", 1/4" crown staples perform this job quite effectively - quick and no clamps required. The staples act as non-protruding clamps and more importantly, they allow slight adjustment of parts in order to get them perfectly aligned before the screws are driven. A few light taps with a rubber dead blow mallet accomplishes the alignment and is done just prior to driving screws into each joint, successively.

The screws I use are specialty screws designed specifically for PB case construction, but work just as effectively for veneer core ply. I pre-bore (just through the end panels) with a 1/8" (3mm) bit and drive 1-3/4", #6 screws into the joints. The head of the Zip-R screws are nibbed and easily countersink themselves. Some use #8 screws, but I'll explain why I believe that #6 screws provide the strongest joint, and certainly as strong as any other assembly method I've mentioned. First of all, you can't break a #6 Zip-R screw; therefore I promise that the particleboard will break before the screw does. This is the case in every method mentioned - the PB is the weak link in the joint. Dowels, if executed properly, I reason may actually strengthen the joint since the wood dowel can become part and parcel of the core material as it is glued tightly in place. That said, I have witnessed may failed dowel joints because of PB failure surrounding the dowels. Following this rationale, the reason I believe that #6 screws provide a stronger joint than possibly all other methods is that they remove or displace less material than even a #8 screw, and certainly less than a conformat screw, dowel, or even a biscuit. This leaves me with the belief that the #6 screw or the pocket hole assembly (also a #6 screw) method provides the strongest joint.

Having said all this, I'd use dowels, if I were equipped to do so, since they provide fast and perfect parts alignment and invisible fastening.



The main reason for dowels is that they provide a square box with a blind joint and are conceded to be the strongest, easiest to achieve wood joint. That said, the money required to do dowels well at speed is a terrifying thing for most garage and other small shop graduates, and it is frequently a long time before they come to terms with the simplicity of the dowel system. Most people who get there rarely turn back, and when they do, turn back away to a previous form. The reasons frequently have nothing to do with the systematic approach to cabinet construction but relate more to some other issue, mostly insufficient profitability.

Even the commentary here, focusing on the pocket screw system as being less expensive, fails to recognize that every connecting screw is handled manually, and pre-drilled and fiddled with, clamped or otherwise handled, and is more time consuming and labor-intensive than any mature dowel system. But no question, the dollars needed are way, way less with pocket joinery.

Confirmats and their mix of brands - I personally like the Titus 0180 - are equally manual systems, with lots of preparation and special tooling, but no where near as speedy as a case clamp. Most screw systems increase product cost because finish ends will be required: extra labor and extra material, but perceived to be less than the cost of a construction drill - unless you keep tabs on the money spent. Machines are always paid for someday, and from the day they are paid off their produce is profit. Whatever they eliminate from your process is future profitability and contributes to the pay down by building less for more (even if you charge the same).

Speaking as a manufacturer of pocket joinery (Ritter Mfg. had the patent on pocket joinery for more than 25 years), I am in complete agreement that it is a very useful and flexible method. But it is not the fastest or strongest or cleanest. That title belongs to a fully developed dowel system. Sorry, but that's the truth. Don't believe me - ask the people who have decided to do dowelling, and who have then moved up to the necessary basic machinery.

Jon Elvrum, forum technical advisor



From contributor K:
I guess we’re one of those who switched construction methods because it was producing insufficient profits as well as inferior quality.

We were set up as efficiently as anyone to dowel cabinets, and did so for 9 years. We had the construction drill, dowel inserter, glue inserter and case clamp.

In 1997, we switched over from dowel construction to assembly screws and haven’t looked back. We have engineering proof that the modulus of rupture is stronger with the Zip-R assembly screw method we used than the way we were previously doweling. When you factor that a higher percentage of those cabinets were seeing joint failure due to glue drying out, or surrounding fibers giving way in humid areas (bathrooms, laundry rooms, kitchen sink cabinets), and that the screw holds the joint together from the outside with compression, whereas the dowel holds more as suspension with surface tension, the screws over time have proven to be definitely more profitable, and a better end product for our marketplace. You are correct in stating that they are not a concealed fastener. We started out pre-drilling pilot holes for the assembly screws and quit after the first year. The Zip-R black oxide screw has a tip sharp enough to “bite” into the melamine (note these are square drive screws) without wobble and self start with the Type 17 auger tip to easily penetrate and drive into the melamine.

We still have the tools if we ever decided to go back, but it would take a lot of convincing at this point. The last 5 years have proven that the 38 (!) processes that we exchanged for 16 processes (average for typical base cabinet) are a better joint and value for our customers. We simply cannot afford to provide a doweled cabinet in a competitive commercial environment, especially when we factor in that the screw holding strength is producing a tighter box and is holding up better over time than our previous dowel construction methods have. When we factor in machinery costs, additional labor for the extra 22 processes, additional floor space, and cost difference of each fastener we find that dowel construction was costing us over $10,000 more per year than assembly screws. That fact alone was substantial enough for us to reconsider and change.

Note that we are a 7 man shop, 9000 sf with 4 guys on the production floor. We are a typical commercial shop. As you stated, when you are much larger you can get the economy of scale with reduced labor force and repeatability setups with dowel construction. I can see us growing another three-fold easily and still use our assembly screw method, profitable and with continued success.

I just wanted to add a differing opinion, based on actual real life conditions, not hypothesis. We have millions of dollars of product in the marketplace produced with both dowels and assembly screws and we can see which are holding up better: the assembly screws win, hands down.



From contributor D:

Jon, your insight is valuable and appreciated. I don't doubt that dowel construction is the fastest and cleanest method of case joinery. I have (since using Zip-R screws) had my suspicions that these screws provided a stronger connection of parts, than do glued dowels. My theory, based on my own limited testing, demonstrated to me that case parts broke apart under less stress than did my cases fastened with #6 Zip-R screws. In all my examples, the PB broke, rather than the fastener. I, however, yield to the masses of manufacturers that do use dowels because I obviously don't have the skill or resources to thoroughly and exhaustively test or evaluate the various case fastening methods. I suspect dowels were adopted and developed as a standard joinery method, more for their mechanical application efficiency than probably anything else. In most of the manufacturing world, speed and quality are continually at odds with each other. Unfortunately speed often takes precedence over quality, as seconds and pennies add up considerably, relative to production volume.

Contributor K is in a market where function rules and aesthetics come second (if even slightly so) to strength and longevity of product. I appreciate that and I appreciate customers who appreciate that. I am often asked (by prospects) what makes my cabinetry different (better or special) than the cabinetry of my competition. With the approval of Jon and contributor K, I'll use much of what I have read on this thread to provide (at least part of) that answer.



I like the idea of pocket hole assembly with 1 1/4" staples to aid in alignment. In that type of assembly, is glue a factor?


From the original questioner:
I have the same question - is glue a factor with the confirmat screws, and what are you doing for the backs with this system? If the side is visible, I am typically just running a rabbet on the side panels to conceal the back. Of course, this consumes little time since I am currently already setting up for dado construction. I usually pre-finish the backs, then pin them to the back, running a little caulk or liquid nails on the back of the cabinet to lend some extra strength to the brads or staples.


From contributor D:
I use no glue with my frameless cabinet construction. I do know of one local cabinetmaker that uses screws for frameless assembly and a small bead of melamine glue is added to joints. I personally don't believe this is necessary.

No glue is typically used with conformat type screws. At least part of the appeal of conformat type screws is that they provide reversibility. In fact, they are commonly used as knock down (KD) fasteners in ready-to-assemble furniture and cabinets.

The backs of our frameless cabinets are slipped into a groove plowed into the ends and bottom of base cabs and the ends, tops and bottoms of wall cabs. Room is left behind the groove (and back) for nailers - in my case 3/4" (19mm). To be more precise, I leave exactly 19.5mm behind the back to accommodate the 19mm nailers. The extra half mm ensures that the nailers do not run proud of the back edge of the end panels in the case of panel stock being slightly off thickness, due to manufacturing tolerances or environmental conditions which might cause slight expansion of the substrate.

As a side note, I've built many hundreds of cabinets with 1/4" backs stapled to rabbets on the end panels. Never glued or caulked a one. At the time I was building 3/4" veneer core ply face frame cabinets with glued and clamped dado joints.



I don't make kitchen cabinets, but do a lot of large wall units, display cases, desks, bookcases, etc. I recently replaced my panel saw with one that is larger and has a router carriage. By using a combination of "stops" I can cut and dado a sheet of plywood without moving the sheet off the saw bed. I like the dados for indexing, strength, and ease of assembly, especially when working alone. I think the choice of connectors depends on the project and the design, and often use several on the same project. Pocket screws (at the proper depth) into a dado make a strong joint. You can drive a screw straight through a side panel if it will be covered by trim. If I were mass producing cases, I'd probably stick to one system. But for one-off projects I like to keep my options open to suit the design.


The comments below were added after this Forum discussion was archived as a Knowledge Base article (add your comment).

Comment from contributor A:
Some of the writings here show unique ways to justify one way or another of putting boxes together. The most common answer boils down to money, either making it or working around not having enough.

I prefer a frameless box. They can be made to work for 90% of the cabinets installed. They are assembled on the job site using biscuits to align and reinforce the joint. I "clamp" them with pocket screws. This works for base cabinets with a counter. For upper cabinets I cut the tops and bottoms to the width of the box and the sides but them. With finished end panels, no screws show, the customer gets to see me sweat and I'm not carting around a lot of air inside empty boxes, a real consideration when your jobs may be an hour or more away.

Besides, most job sites have more space for me to work in than my shop does. They provide the heat and cooling and there are actual people to talk with. This method works while working alone or if you can leave another person in the shop cutting the next job up that would effectively double the capacity of your shop.

It works for me. Any reasonable system can work if you stick to the system and don't try to reinvent the wheel every week, as many of us try to do.



Comment from contributor B:
My family has been doing this for three generations, from the introduction of the slotted screw to the now popular screw-lux and star-drive screw.

A simple box can be constructed from a butt-joint, fastened with a few pins or staples and secured by simple screws.

The finished ends can be either applied wood or laminated plastic. Either applied moulding, flutes or whatever your imagination desires can be added... Just like adding salt and pepper to a recipe. Fast, simple, tested through many years. Our company has had no call-backs and many referrals.



Comment from contributor C:

I own a high-end cabinet shop in Houston, Texas, where we work with exotics (teak veneer, cherry, walnut, mesquite, etc.) and in my opinion, the difference between the good stuff and the junk is that you use good cabinet grade veneer core plywoods. The type of fastener you use on the boxes is not that critical. It is just to hold the thing together until the glue sets.

I plunge cut the face frames by rolling the cabinet saw blade up through the veneered ply, then use a bayonet saw to clear the corners. The drop then becomes the door blank (depending on the doors ordered) and the grain is a perfect match. I use counter sunk, coarse thread drywall screws of appropriate length on non-visible areas just as a belt and suspenders approach. One piece, plunge cut face frames is the main difference between the "furniture grade cabinet" and the other stuff, because as has been stated here, a cabinet is a box with a fancy face. Liberal glue use, good material and one piece face frames have been the mainstays in my shop.