1. This is a very interesting and simple design. The use of the tenons on the cross pieces really increases the strength of the cabinet without a great deal of complexity. Other online instructors constantly add complex joints that are not always necessary. In this case they would be using sliding dovetails all over the place. The rabbet for the hinges is also very clever. The simplicity and elegance of these choices make these videos very useful to me. I may not need a medicine cabinet right now but these design details can be used in other projects I have. Thanks

  2. Paul,
    I love that you showed the source of your wood in this project. I’ve sometimes felt, naively, that I need to find perfect wood for projects. You show me here that it’s what I do with the wood that makes the difference, not where it comes from. Question on the groove that is going to be cut into the sides: Can I groove the frame pieces after I’ve fit them together and before the tenoned pieces or do I need to wait for the tenons to determine the length of the groove? Thanks.


    1. I don’t see an issue. Either way the grooves have to match. Probably it’s best to wait for the back rail to be fitted and the entire back side planed flush that way you will have a simpler registration for your plough plane.

    2. Hello Peter, do you mean the additional groove for the back frame that is yet to be cut? That is an additional piece of stock that is fitted to the exact gap that is left once the tenoned pieces are fitted.
      Thanks, Phil

  3. Perfect timing. I need to replace two bathroom cabinets that I bought from one of the big box stores. Very cheap cabinets and I’ve wanted to replace them for a while.

  4. Paul, I know different plow planes are different, but FYI I’ve found the opposite with my Veritas combination plane — it’s hard for me to do the short stabbing strokes, and the long strokes don’t get clogged (full disclosure: I’m still very new to woodworking :- ). Thanks for another great project.

    1. I checked with Paul and he said it results in there being a greater width of wood at the end of the piece of stock before the dado starts which increases strength. It also a means of developing a higher level of accuracy and sensitivity.
      Thanks, Phil

  5. You must be a mind reader Paul! I have been wishing for you to do a bathroom cabinet for some time now. My bathroom is small, and just has a mirror on the wall where a cabinet should be. I’m thinking Cherry would be nice wood for this project or maybe Mahogany.

    1. It would be if you could make it, but it also would be a very difficult task with hand tools. Obviously, gravity is your friend for the top so nothing else is needed there. The bottom isn’t supporting anything other than it’s own weight plus whatever is sitting on it so a stopped housing dado joint is more than adequate strength wise. The rear top and bottom rails are going to be mortised into the sides of the cabinet and they are what will support the weight of the cabinet and it’s contents when it’s hung.

        1. It’s easy with a router. Even with a table saw if the slot can go all the way through. I don’t know a good way to do that easily with hand tools if the slot needs to be stopped like this would need to be unless of course you totally change the design. How do you do it?

          1. On this piece, Paul has the inner face of the cabinet side sunk into the housing bare. The outer face has a shoulder. He chopped both with a chisel. To make a sliding DT for this piece, you would keep the inner face of the housing exactly as it is (vertical). For the outer face, he used a width of 1/2. You would lay out 1/2″ at the edge and then, say, 3/8″ at the other side, which gives the taper. When you chop this side, you lean the chisel rather than keeping it vertical. This gives you the socket side of the DT. On the side of the cabinet, he cut the notch for the stop and then cut a rebate. I’m leaving out a lot of details, but you would do something similar except you need to make the depth change over the width of the side. This gives the tape that matches the socket. You also need to make this component be at an angle to give you the tail. The easiest way to do this is with a dovetail plane. You take passes to sneak up on the taper and the width. When things are fat, the side won’t slide all the way home. As you get to the final fit, it will go together with zero resistance except for the last maybe 3/8″ to 1/2″ when the tapers go home squeezing everything together and tightening the joint. Depending upon your skill level, you can leave some extra width so that, when it slides home, the side is proud of the back a touch, which you then plane down to perfection. A nice thing about this joint is that, if the wood shrinks in a hundred years and the joint becomes loose, give it a whack with a mallet and you’re good for another hundred. A regular housing would be loose forever as would a machined sliding dovetail that had parallel sides. If you don’t have a dovetail plane (ECE makes a wooden one), you can do it with a guide block. See Paul’s coat hanger project, although in that project he tapers both sides or puts tails on both sides, if I recall correctly, which is a little different from what I’m describing. If you want to see step by step photos of what I’m describing, take a look at Tage Frid’s book. He shows the use of a “chair saw.” That seems attractive because you can lean it on a guide block to guarantee the socket angle, but my experience with stair saws is that they are close to useless. You could use a tenon saw, though. If you saw, you’d chop the end of the stop so you have a place to saw into.

            As you say, though, this piece is fine as built. Sliding DT aren’t needed, but they wouldn’t be wrong or particularly difficult. I think Frid said something like, the sliding DT is not a hard joint to cut, especially after you screw it up a dozen times.

          2. By the way, when I’ve done these (and I’ve not done many, honestly), I’ve used the saw and guide block method in order to get the socket angle. I need to review Paul’s method…I don’t recall if he used a chisel and guide block. I should watch that again. He might have used this for the shaker stool, too, but I don’t recall.

          3. Not to loose sight of my initial point that you really don’t need to do a dovetail joint for this, I was hoping that you knew a trick for making them that I missed. But you gave a pretty good description of the process. On a good day, sliding dovetails are tedious and take much more time than a simple housing dado. When you can saw all the way through for the slot, it’s a bit easier but it still takes a lot of fitting to get them tight. And if you mess one up, you hope it’s the first one on that particular board so you’re only out some time and the material. But I’ll tell you how my luck runs. When I make a mistake, it’s almost always on the other end of a board with a joint that was already fitted. So not only do I get to redo the one I messed up – I have to re-cut one that was already done! So back to the original question. Is a sliding dovetail stronger than a housing dado – absolutely! Do you need one here – I don’t think so.

          4. Although a sliding, tapered dovetail wouldn’t be hard to execute, looking at the broader project, it may not work well here: I’m not sure you would be able to assemble the M&T’d nailing rails and get the back panel in it’s groove and still slide the dovetails. I _think_ you could do the assembly (top and bottom will go on last), but as I said earlier, they aren’t needed for strength, and it would be a complexity. I’ll be surprised if Paul doesn’t glue the nailing rails to the top and bottom, which will give even more strength.

          5. Ed, I think we’ll just need to agree to disagree as to the ease of making a sliding dovetail – especially for this application. It might be easy for you, but I’m quite sure I would struggle with it.

  6. Hi,

    I’m having trouble with the edges of the wood splitting out when removing the side piece from the housing dado. Is this likely to be due to the type of wood – I’m making it from old Ikea bed slats which are quite soft – or the joint being to tight? I suspect it could be a combination of the two, but any tips on how to avoid the edges of the wood splitting out would be much appreciated.


    1. It sounds like a combination. If it’s flat sawn soft wood, the end grain can try to split like you’re describing. It will even happen with some hardwood if the joint is too tight. You can try applying some CA glue to the edges of the dados to strengthen and repair them. If you’re going to paint the cabinet, that won’t hurt anything, but if you want a natural or stained finish, CA glue might show so be careful not to use too much. You don’t want the joint loose, but as long as it’s self supporting, that’s good enough. You may want to take a couple of swipes with your plane on the inside face of the sidewalls. That joint should seat with nothing more than a light tap with your hammer. If you’re having to drive them in, they’re too tight. It also doesn’t hurt to put a slight chamfer on the top and bottom edges of the side pieces to help them ease into the dado. Hope that helps.

Leave a Reply