Strength of Cabinet Joinery Using Pre-Finished Plywood
When you use pre-finished plywood, does the finish compromise the strength of glue joints? The question sets off an extended discussion of structural issues in cabinet carcass construction. September 23, 2006
Question
I am building a kitchen's worth of face-framed cabinets with 3/4" pre-finished ply carcasses. I've not worked with pre-finished ply and have a few of questions about joints. For background, I usually use butt joints with both biscuits and screws. I also use applied end panels, so covering screws on the outside of the case is not a problem. My back is a 1/4" ply with nailer strips.
1. How much strength will I lose by not having a good wood-to-wood glue joint at the butt joint because of the finish? I still plan on biscuiting and screwing the joints. Will this be enough?
2. Should I consider reinforcing the joint with a PU glue? I am really trying to avoid dadoing, as I usually use separate ladder bases and these are not conducive to dadoing the base cabinet floors (it would have to be a rabbet). If so, which PU glue would you recommend?
3. For optimum strength in the base cabinets, should I butt the edge of the cabinet side against the cabinet floor, or the edge of the floor against the cabinet side, screwing and dadoing in each case?
Thanks in advance for your help. I am very seriously considering switching to pre-finished ply and I am using this job as a test case.
Forum Responses
(Cabinetmaking Forum)
From contributor A:
All you have to do is butt and screw and it will be plenty strong.
From the original questioner:
To contributor A: Thanks. I kind of expected that (I'm a bit of an over-designer). But I have another question on a related matter for you. I usually butt the edge of the cabinet side against the cabinet floor, the idea being that the load of the countertop (typically granite) would be transferred from the edge of the side directly to the floor. But I've seen many books where the floor panel is butted against the side panel, and the load seems to be transferred to the screws. I can't understand why you would do it this way, but many experts with more experience than I seem to prefer it. Do you (or anybody else) have any idea why?
From contributor A:
I don’t fully understand your question, however I always run the side full length and floor, shelves, etc. are butted to them on the inside. This way all loads are directly from countertop to the floor they are sitting on, if this helps.
From the original questioner:
I’m sorry for not being clear. Let me say it another way. I install my cabs on separate bases, so I don't really have the option to run the sides to the floor. In other words, the intersection of floor and side of my base cabinet forms an "L" as opposed to the way you describe, which is essentially a "T" on its side. With this "L" connection, I have the choice between (1) butting the bottom edge of the cabinet side against the cabinet floor, or (2) butting the edge cabinet floor against the cabinet side. I believe the first way is better because the countertop load is transferred through the cabinet side to the cabinet floor, which is directly supported by the separate base. It seems the second way relies on the sheer strength of the screw alone to support the countertop, and the bottom edge of the cabinet side is not well supported. I do it the first way, but I see a lot of people do it the second way. Any idea why? I'm always looking to improve techniques.
From contributor B:
If I have the picture clear then this is just my opinion - by bringing your floor all the way across and butting side on top of it then you have the end grain of the floor exposed and in sight, which doesn’t look good and is a bear to finish. Even if it is painted the end grain will not look like the "long" grain" of the sides.
From contributor C:
Let me suggest this - the strongest line in a butt joint is the most direct one. So your strongest scenario is one where your side wall runs from your counter top to your deck with no interruption. That said, why would you not consider full or blind dados in your casework? Until recently I full dadoed all my casework in house. Freud makes a stacking dado set with a dial-in main blade that lets you calibrate in any dimension to a micron. I dado, glue and screw all my casework, all 3/4" stock. Sometimes I'll use 1/2" for my back walls, but mostly 3/4" all round. The difference in cost between 1/2" and 3/4" is only about 5 bucks and not worth the time to recalibrate the dado set. A few months ago I started using E-cabinets software. At the same time I found a guy who I can download all my casework to and he runs it all on his CNC with blind dados for me. I still glue and screw though. I do have the occasion to do some in-house casework and revert back to my full dado construction. The dado construction ensures a tight and solid product that you can park your car on, whether full or blind.
From contributor A:
You're plenty clear in your first explanation. I butt the bottom into the side you've seen in books. I've done it this way for about 10 years now and many times with 5/8" melamine. I've never had a problem with granite tops. The reason I like to do it this way is because most bases are made to hold drawer boxes and I can cut the base bottom the exact width of the drawer opening. If the bottom extends past the sides (your current way) then I'm always fighting with varying thicknesses of plywood.
From contributor D:
I build my boxes with both biscuits and screws. I have jigs set up so it goes fairly quickly. The glue in the biscuits is surprisingly good alone but the screws I feel give added holding power.
As for the sides, they should full height with the top and bottom sandwiched in between. Your idea of the weight coming down on the screws doesn't work in this situation. The weight of the c-top will bear on the vertical members. The horizontal surfaces (i.e. top and bottom) will not support any real weight. The ply will only bend under the weight. Picture a single box built with a top that is bowed upward. As you apply the weight of the c-top to the box the bow will be pushed down until the top rests on the sides. The weight is then transferred directly to the floor, ladder frame, etc.
If, however, you were to stack the pieces so the sides were between the horizontals, the weight of granite could compress the horizontal pieces. (Think of a couple hundred pounds being put on essentially a 3/4" wide piece of ply.) I would stick with the first method as I have never seen the second done. And sometimes it's better to not rediscover the wheel. Another thing I do with my boxes is use 1/2" backs rabbeted into the sides/top/bottom of the cabinet. Again, it’s very quick to do once you get the hang of it, and no need for nailing strips.
From contributor E:
Where we have an applied end panel, we typically have the toe kick inset just 3/4", so the weight transfers through the toe kick. That being said, I don't have any problem relying on our confirmats to hold if the toe kick was inset 3-1/2". The weight you're supporting is only the weight at the end of the cabinet run. Times when I was worried about it I added a 3/4" plywood sub-bottom to the cabinet.
However, we seem to run into issues all the time where we have to build a cabinet where the tops and bottoms run by. We do this pretty frequently in uppers, depending on the light valence detail, but also for sink bases (we do a full top, and it's no fun to try cutting through confirmats doing a sink cutout); and for engineering oven cabinets as well. So I've briefly contemplated making this our standard cabinet construction. Jim raises a good point about sizing drawer boxes, but you have the same issue with sizing doors and drawer fronts with variations in panel thickness in a cabinet where the sides run by.
From contributor A:
As cabinetmakers we some times overthink our designs. I have been using biscuits and screws for years as well as placing my long sides directly on a separate bottom (a plinth I believe is the correct term, but I could be wrong). Even with 2" cast stone counters I have yet to have a failure. Also, someone mentioned a jig for biscuit jointing the butt joints. You can save yourself the step of making and using a jig.
Usually what I do is to measure the distance between biscuit joints by lining up the outside edge of the biscuit jointer with the front edge of the panel you are cutting into. When you line up for the next slot you slide the biscuit cutter along the joint until the outside edge of the tool lines up with the end of the last biscuit slot you cut. Hopefully that makes sense, it would work better with a picture but I don't have one. I save most of the detail work for the stuff on the outside, face frames, doors, drawers, end panels, millwork, sanding and finish.
From contributor F:
The only time I make the deck (bottom) of the cabinet full width is on heavy pantry or double wall oven cabinets. Since you are applying end panels anyway, do what I do for the backs; 1/2"prefinished ply shot and screwed, and edgeband the bottom if you want. Most of my clients use full granite backsplash, so in my case the bottom of the 1/2"back doesn't show. If you want to inset back, then pockethole all around and screw to top/bottom/sides.
From contributor D:
To contributor A: I do my biscuit alignment the same way you described. Where the jig comes in handy is for the bottoms of base cabinets. It aligns the bottom of my biscuit jointer so that my cabinet bottom is 4-1/2" high (toe kick). Those using ladder frames underneath won't need it. But it saves a lot of time instead of measuring and marking every side, and takes little time to make.
From contributor G:
Screw it together with confirmats and use RU glue – it’s that simple, like frameless.
