I'm designing a bench for the entryway of our home that will have a place to sit, have two drawers and a bottom shelf for shoes. I'm designing it in Blender, but would like some help confirming the soundness of my plan since I'm new to fine woodworking.

Let me know if more details are needed below. Thanks.

Bench preview

Basic Details

  • Overall dimensions: 52" W x 21.5" H x 17.5" D
  • Leg rail/stiles are 1.75" wide in two of the dimensions
  • All other panels are 3/4" thick except the drawer faces which will be 5/8" thick
  • Plan to use walnut and another pale hardwood for the drawer faces

Side & Back Panel views

I planned on using a double mortise-tenon on both sides of these panels to join to the leg stile. The tenons would roughly be 1" long, 3/8" thick and 1.25" wide. The panels are 5.25" in height.

I chose a double mortise-tenon joint based on the internet's "rule of thumb" size for this joinery; one mortise-tenon on each size would be too large if I wanted to avoid movement in future is my understanding. Please let me know if I'm wrong here though.

Side panel Back panel

Top & bottom pieces of drawer carcass

Joining these pieces are my main trouble point right now. My first thought is to use dowels. See the picture below as an example for joining the top to the side panels. I was thinking of doing similar for the bottom piece.

If I went with this plan, then I believe the only joints really holding the stress of someone sitting on the bench, and the contents of the drawers, is the mortise-tenons described above in the side panels section -- would these joints be enough to support say 250-300 lbs?

Dowel joinery

Bottom Shelf

I'm open to suggestions on how to join the bottom shelf to the legs. The shelf will not hold a significant amount of weight (several pairs of shoes). My first thought is to chisel out a notch in the leg stiles at the four points of contact and glue the shelf into the recesses. I'd like to avoid using more material to support the shelf if possible to keep the bench look minimal.

Bottom shelf points of contact

  • 2
    I'm not going to add another Answer here but wanted you to have some notes. I'd use plywood for the back panel, no movement issues and for an economy of materials. If you're firm on using solid wood for the top and shelf (personally I think ply would be preferable given the plan) I think it's a must to use QS stock, although this likely means you won't be able to pick for particularly nice grain. Even then you'll want to allow some gaps for movement. Speaking of which do not use those dowels for attaching the top! Doing it this way will constrain movement of the top, which is a Bad Thing.
    – Graphus
    May 27, 2019 at 6:56
  • Thanks for the tips! I'm interested in learning more. If dowels are the wrong way to go when using solid wood, what would you suggest instead? Lastly, what kind of movement would you expect to be constrained with the dowels and what would be the bad outcome?
    – Scott Lin
    May 27, 2019 at 7:12
  • I did a bunch of reading about wood movement in the past two hours since my last comment, and I now understand that constraining the movement of the wood with dowels in the front to back direction would lead to possible failure. However, I'm still at a loss for an acceptable way to join the top/bottom carcass pieces to the rest of the side/back panels. Any tips would be appreciated.
    – Scott Lin
    May 27, 2019 at 9:08
  • I'd also like to confirm why plywood/quartersawn stock must be used for the top and shelf. I believe I have the answer from my research, but would love a nod. Essentially, it boils down to the fact that these pieces will move in the front to back direction over the 17" distance and because the legs constrain this movement, it will lead to failure of some kind. Therefore, a stable stock must be used to stack the deck in my favor. Is this right?
    – Scott Lin
    May 27, 2019 at 9:20
  • 1
    Using plywood (or MDF or chipboard) for certain components in a design essentially removes all concerns about movement with those pieces because dimensional changes in manmade boards can basically be considered to be zero in all directions.
    – Graphus
    May 27, 2019 at 12:20

1 Answer 1


Fundamentally, I think you have a good plan.

The way I look at it is this: you have a 3/4" thick hardwood plank, approximately 4' span, that can be well supported on both ends. This should be strong enough to hold up 300#. I'd make the tenons 1/2" wide, rather than the typical rule of thirds, and as long as possible (like, 1-1/4"). It's not impossible with these tenon dimensions, but it would be easier and stronger if you pushed the side panels a little further outward from the corner of the legs (1/2" would be my pick.) If you don't feel like changing the design, offset the tenon by a bit so you have more substance at the inside corner of the leg. (Top tip: offset tenons are a pain.)

The back panel is what'll keep the span from deflecting too much. I think that it'll do the job, but you'll have to experiment while building. My best suggestion is to not build the drawers until you've built the carcase, so that you can add a stiffener to the junction of the top/back if needed. (Pocket screw or spline the center divider so that it can go in after the carcase is glued up.) Maybe you'll find that the front of the top deflects more than you'd like... then glue/screw a stiffener to the underside (behind the drawer faces) and change your drawer height to suit. Inset drawers are quite unforgiving, so you want this to be solid (or you want wide reveals around the drawers).

Speaking of the back panel, you have a problem with swiss cheesing the leg and having your tenons banging into one another. I don't know exactly how I'd approach this, as it's important to have good joinery at the back/leg junction to prevent racking of the entire structure. Maybe 3 long but not very tall tenons... maybe run the back tenons through the middle of the side tenons.

Dowels to attach the top and bottom planks to the sides/back will help make the drawer box stiffer, so that's good. Chiseling notches for the base will be fine, though I'd pin them with a dowel in the long direction. (It can't hurt, in terms of improving the resistance to racking.)

Edits, to clarify and correct errors. (Thanks, @graphus, for the comments.)

Graphus is absolutely correct that the sides shouldn't be joined to the hardwood top/bottom with dowels because of wood movement. (The back can still be joined like that.) Look to tabletop-apron joining methods for inspiration on how to allow for wood movement. (You'd also need a bit of a gap at the legs to prevent hardwood top/bottom expansion from breaking your leg assembly apart.)

If you opted for plywood/mdf, you'd do away with the expansion problem, but you'd have a weaker span, so would probably have to build stiffening under the front edge of the top. (That can easily be hidden behind the drawer fronts.) The center divider might help stiffen a tiny bit, but I don't think it'll help that much. (Strength won't be an issue on the lower shoe shelf.) If you used a ~3/4"x3/4" strip of hardwood to edge the front of the ply, that would stiffen it (and give you the option for a slight easing of the hard edge and make the edge more durable).

Racking: imagine a square cabinet. Push sideways on the top corner. If the cabinet is stiff (usually because it has a back panel), it'll stay square. If it isn't stiff, the top will move sideways and the square becomes a parallelogram. Do that enough times and your cabinet will fall apart.

Regarding the location of the side panel relative to the legs... in my opinion, design drives engineering solutions, not the other way around (as long as you can come up with good engineering solutions!) So I wouldn't suggest moving the side panel if that's not what you wanted to do with it. But if you did move the side panel to be nearly flush with the leg (never exactly flush -- always plan a tiny reveal!), you could always pad out the hidden interior space with material to provide somewhere to attach your drawer hardware.

  • Thanks for the information! Sorry for the simple question, but could you define "racking" for me? I tried searching the term with Google, but as you can imagine "wood" and "rack" gives mostly shopping results. :) Lastly, on the topic of deflection, do you suspect it will occur in the top plank even with the middle piece in place between the drawers as pictured? With this piece in place, there would be roughly two 2' spans where the drawers are that I can see maybe deflecting when sat on. I don't expect it to deflect in the middle at all with this piece in place.
    – Scott Lin
    May 27, 2019 at 6:36
  • Based on your suggestions, I will make the side tenons 1/2" thick now instead of 3/8". I planned on offsetting the back panel tenons to the furthest outside position to allow the side panel tenons to be as long as possible without colliding. However, you make a good point about allowing for more substance at the inner leg points; I'll also offset the side panel tenons as you suggested to accomplish this. Unfortunately, I cannot move the side panels as a whole outward because I need them flush with the inner part of the leg in order for my chosen drawer glides to work.
    – Scott Lin
    May 27, 2019 at 6:51
  • 1
    @ScottLin "Racking" is a term builders and craftspeople have borrowed that means, in general, when a system is deflected by some force. e.g., building systems have a racking strength so they are not deflected by wind, etc. Similarly, cabinet and furniture carcases need to be built to have strength against being deflected out of square (or whatever angle used in the joinery) by static and dynamic loads they encounter. For example, a cardboard box is very good at deflecting modest racking forces as long as it has at least 5 sides. With 4 sides it collapses easily. Carpentry has similar issues.
    – user5572
    May 28, 2019 at 14:59

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.