Weight Bearing Shelves

[Jan Bianchi]

I am making a cherry Shaker hutch that is detailed in Fine Woodworking Jan-Feb. 2007. I have asked several questions about it in the past from this community and have always received very helpful replies. I am attaching a picture of the lower cabinet which is finished and "in service".

My upper cabinet is different from the article's in that my version's upper shelves will be 15 inches deep rather than the 9 3/4, still 44 inches long and 3/4 thick. The shelves will be attached to the solid wood sides by tapered sliding dovetails ending in face frame. I am concerned about the shelves sagging. The upper part of the article's cabinet was designed for display, not storage. Do you think I will need stiffeners under the shelves to keep them from sagging?

I don't want to do that because they will show below the muntins in the glass doors. I have been using my great grandmother's walnut china cabinet as a model. It's 40" x 15" by 3/4" shelves are attached to the sides by cleats. They do not sag when fully loaded with china and glassware. While I could put stiffeners only below the lower shelf and store only light items on the top two shelves, I'd rather not.

I have checked the sagulator calculator online, assuming 10 lb load per foot (ten large plates being 6 lbs). It says that creates a visual sag of .003 which is not visually noticeable and is acceptable. I have to decide this now since I won't be able to add stiffeners after this is assembled. Doesn't the additional depth of the shelf mean less potential for sag, not more?

[David Eisenhauer]

Will the back of the shelves be attached to the cabinet back? Can You add a "stiffener" to the front of the 3/4" shelves (1"?, 1-1/8"?, 1-1/4" wide?) without affecting the look? In my experience, anything over 36" long for a 3/4" shelf may cause problems if the shelf load is not a known, controlled load. The plates aren't too bad, so it will probably be OK. I turn out bookcases from time to time, books get heavy quick, so perhaps I am overly gun shy about unsupported shelf lengths. The "display" style cabinets I turn out do allow me to attach the back of the shelf to the back of the case and I believe that helps a bunch.

[Jan Bianchi]

Will the back of the shelves be attached to the cabinet back? Can You add a "stiffener" to the front of the 3/4" shelves (1"?, 1-1/8"?, 1-1/4" wide?) without affecting the look? In my experience, anything over 36" long for a 3/4" shelf may cause problems if the shelf load is not a known, controlled load. The plates aren't too bad, so it will probably be OK. I turn out bookcases from time to time, books get heavy quick, so perhaps I am overly gun shy about unsupported shelf lengths. The "display" style cabinets I turn out do allow me to attach the back of the shelf to the back of the case and I believe that helps a bunch.

I probably can't attach the shelves to the back. The back is a large framed panel of 1/2"shiplapped or T&G cherry. I'd have to go through the 1/2" material which doesn't sound good to me. Any lip under the front of the shelf will show since the muntins are only 3/4" high.

[Jim Matthews]

Make the shelves from a quality plywood with lots of layers
and trim out the front with solid wood.

Plywood is amazingly stiff across a span this size.

[Guy Belleman]

A couple stacks of plates can get real heavy, especially over a 44" span. My experience is that any span over 30" will sag over time, even plywood, without some rigidity built into the shelf or cabinet. Also, depends on your vision of how long the cabinet may be used. Rigidity can be built into the cabinet frame, which will reduce the amount the interior shelf can deflect. Flexing, or sag, on a shelf in a cabinet means that the rest of the cabinet has also flexed, or twisted, some too. If you make the top and bottom very rigid, and your sliding dovetails are tight, then the interior shelf can't really flex, or sag. With your wider shelf, I think you might look at doubling up the top and bottom part of the case, or at least reinforcing them in some manner.

[Richard Wagner]

Why not build in some rigidity and quit worrying about it.

[Andrew Pitonyak]

I built a very large desk with a hutch. The hutch has two very large shelves. I used shelf pins so that the shelf is adjustable in height. I included two shelf pins along the back so that the shelf has some support on the back as well as the sides. You could do something similar. Since the shelf will be set in place, you could attach a small support piece of wood attached near the center along the back.

The next question becomes, how thick can you make the wood? Thicker is probably better.

[James Hansen]

When I make ply shelves, I use 1 1/4" X 3/4 end caps on the front and back with what ever material the rest of the project was made with. This not only makes them very strong, but gives it a nice looking decoritive cap.

[Wade Lippman]

I probably can't attach the shelves to the back. The back is a large framed panel of 1/2"shiplapped or T&G cherry. I'd have to go through the 1/2" material which doesn't sound good to me. Any lip under the front of the shelf will show since the muntins are only 3/4" high.

I don't follow you. Why is going though 1/2" material not good?

[Doug Garson]

I built a very large desk with a hutch. The hutch has two very large shelves. I used shelf pins so that the shelf is adjustable in height. I included two shelf pins along the back so that the shelf has some support on the back as well as the sides. You could do something similar. Since the shelf will be set in place, you could attach a small support piece of wood attached near the center along the back.

Just a thought, but if you support the shelf on pins as suggested and finish the shelf top and bottom so that it is reversible. If it starts to sag you could periodically flip it over to take the sag out. In a worst case scenario, if it sagged too much it could be replaced since it is not glued in.

[Jan Bianchi]

One-half inch thickness just doesn't seem like enough bulk to really help hold up a shelf. I may be wrong about that.

I can't use pins to hold up the shelves from the sides. The sliding dovetails are what keep the sides parallel and rigid. There is no "bottom" to the upper half of the hutch. It just sits in a groove along the back of the lower part of the hutch. I could put pins in the back. As much as possible I am trying to stay with woodworking techniques that would have been used by the Shakers.

The muntins in the glass doors are only 3/4 high so any shelf thicker than that will show. Not a disaster, but not the design either.

[David Eisenhauer]

I use 1/4" plywood cabinet backs to hold up my shelves. I am not up on all the Shaker habits, but how about a 1/4" - 3/8" dowel through the center (to minimize the effect of wood movement) of a vertical back panel board and into the shelf back every now and again?

[Sam Murdoch]

I don't want to do that because they will show below the muntins in the glass doors. I have been using my great grandmother's walnut china cabinet as a model. It's 40" x 15" by 3/4" shelves are attached to the sides by cleats. They do not sag when fully loaded with china and glassware. While I could put stiffeners only below the lower shelf and store only light items on the top two shelves, I'd rather not.

I have checked the sagulator calculator online, assuming 10 lb load per foot (ten large plates being 6 lbs). It says that creates a visual sag of .003 which is not visually noticeable and is acceptable. I have to decide this now since I won't be able to add stiffeners after this is assembled. Doesn't the additional depth of the shelf mean less potential for sag, not more?

While I would bet against this your Grandmother's cabinet is proof that I would be wrong - though you are adding 4" to the length :rolleyes:. My general rule for shelving, other than linen shelves, is no more than 3' span for 3/4" material (ply or solid) without added support. Making the shelves 7/8" or 1" thick would certainly add to the stiffness - your call as to how much that would change the esthetics.

Another solution would be to add a vertical support in the center. It would be hidden by the center stiles. Might need to be let into the shelves in order to place it properly within the depth of the cabinet, or make the shelves a bit less deep.

[Peter Quinn]

I'd go with a 1" solid shelf, or at least 15/16"' and taper the first few back on the bottom side like a raised panel, so I'd fades to the back, but shows 3/4" at the leading edge for the visual. Or dovetail in a support post that sits behind the stiles on the doors, doesn't have to be too wide, maybe 1" to 1 1/4", holds up the shelves in an inconspicuous way. Loading a 40+ inch 3/4" shelf with granny's china sounds like a problem to me....unless you use jatoba for the shelves, or glue a strip of cherry to a stronger species so you get the cherry nose, intensely strong back. And definitely use the back as a support, 1/2" solid paneled back will add great support.

[Frank Drew]

The next question becomes, how thick can you make the wood? Thicker is probably better.

I agree, both in terms of stiffness and also, past a certain length, appearance; 3/4" can look a bit dinky as a shelf gets longer; why not try 4/4, or even 5/4?

If you totally object to thicker "looking" material, you could use 4/- or 5/4 but bevel the underside of the shelves' front edges, leaving a 3/4 show edge for the look you want but with added strength (the bevel should be rather wide so that the true thickness of the shelf doesn't show to the casual observer.)

[Kent A Bathurst]

One-half inch thickness just doesn't seem like enough bulk to really help hold up a shelf..............

The sliding dovetails are what keep the sides parallel and rigid..............

The muntins in the glass doors are only 3/4 high so any shelf thicker than that will show. Not a disaster, but not the design either.

Reverse order:

3. You are right. Keep the shelves at same thickness as muntins, and aligned with them. Don't ruin a classic look. But - you need to deal with potential sag, so.....

2. The shelves and the case sides are the same material, right? Then, they will expand and contract at the same rate, in the same direction. No need for "sliding" dovetails, in terms of allowing them to slide after assembly. Glue the shelves in place is what I would do. Not sure what effect this has on rigidity....may not help to glue them - just saying that is what I do.

1. The half-inch back will do just fine. Put one screw [or peg] through the middle of each shiplap/T&G board into the shelf. "The middle" so that each board will expand and contract from that mid-point out in each direction. This secure attachment will make a significant difference in the "sag". You won't be perching everything at the very front edge of the shelf - the closer things sit toward the back, the more weight the back connections will carry.

You will not have any problems long term. All ducky.

[Jan Bianchi]

I like the idea of a thicker panel beveled to 3/4 toward the front. That seems like a good compromise. Would screws through the 1/2 back panel into the shelves or a cleat under the back of the shelves be preferable?

[Brian Holcombe]

A recessed center support would help out a lot. I avoid making 3/4" shelves wider than 24" without support.

[Sam Murdoch]

I like the idea of a thicker panel beveled to 3/4 toward the front. That seems like a good compromise. Would screws through the 1/2 back panel into the shelves or a cleat under the back of the shelves be preferable?

I didn't suggest the beveling as it seemed not appropriate to the "Shaker" look you are intending but I agree that the 3/4" muntins create a challenge. I would not object to see an 1/8" above and/or below the munitions behind the door. I think the eye would read that as part of the door.

As for the back cleat idea that would solve lots of problems - better than screwing into the shelves, but you might have some issues with locking the wood movement of the back if it is a solid glue up of boards rather than a panel in frame construction. In the latter case the cleats could be attached to the stiles of the back panel with no issues. I still favor the center support at the front and even add vertical support at the back. This could just be a 1x2 set between each shelf. These could be fastened to the back with no wood movement issues.

[Jan Bianchi]

I thought the purpose of the sliding dovetails here was not so much wood movement as a mechanical way to anchor the two sides together and keep them parallel. While I would love to not use tapered sliding dovetails, just gluing the shelves into a dado doesn't seem strong enough to keep the sides from splaying out when weight is added to the shelves. The two sides are dovetailed to the top. There is no bottom. The sides just sit on the top of the lower cabinet. The back of the upper cabinet is rabbeted into a groove along the back of the lower cabinet.

[Sam Murdoch]

I thought the purpose of the sliding dovetails here was not so much wood movement as a mechanical way to anchor the two sides together and keep them parallel. While I would love to not use tapered sliding dovetails, just gluing the shelves into a dado doesn't seem strong enough to keep the sides from splaying out when weight is added to the shelves. The two sides are dovetailed to the top. There is no bottom. The sides just sit on the top of the lower cabinet. The back of the upper cabinet is rabbeted into a groove along the back of the lower cabinet.

Your construction method will be very good and satisfy the intent. No need to 2nd guess yourself in this regard. Keep in mind that a cabinet like this with the separate unsecured top needs to be kept against a wall and not moved when loaded. You should think of some way to mechanically fasten the upper to the lower to prevent any accidental tipping. The upper may be deep enough that this will not be a concern but you should at least be aware of the potential. Opening doors and pulling things down from a top shelf changes the dynamics.

A face frame changes the dynamics - it helps to hold the cabinet together and so mitigates the need for the sliding dovetails. A full bottom would help too but the bottom rail of the face frame (in this case more like a mid-rail) will add lots of structure to help keep the cabinet together.

[David Eisenhauer]

Going back to your issue concerning the sliding dovetails for the sides. Simply gluing the shelves into dados in the sides will tie the two sides together and not allow any splaying when weight is added. If you intend to use the same wood for the sides and shelves, plus finish everything all the same all the way around (all surfaces of the sides and shelves), wood movement should be minimized. I too thought that the sliding dovetail was to prevent movement issues with the shelf/side connection and thought, "could happen, not sure where the poster lives and what kind of inside environment the house will have". As for potential strength, the glued shelf into a 1/4" dado will work just fine.

[Kent A Bathurst]

I thought the purpose of the sliding dovetails here was not so much wood movement as a mechanical way to anchor the two sides together and keep them parallel. While I would love to not use tapered sliding dovetails, just gluing the shelves into a dado doesn't seem strong enough ............

As other comments noted:

Nah. You are hard-pressed to come up with a "mechanical" joint that will, overall, outperform glue in terms of overall performance. Certainly makes sense that the dovetailed keyways will do better, but not shure how much v the time invested.

On casework like this, I generally use 3/16 - 3/8 - 3/16 dimensions, but I have no scientific reference to tell me that is better than 1/4 - 1/4 - 1/4. It is simply what I do.

Never had any issues.

YMMV.

[Jan Bianchi]

Thank you all. I am delighted to get out of tapered sliding dovetails. It's all the same wood and the same finish. I'm in Seattle. I don't typically experience serious problems with wood movement in my house despite all the water out here. It usually isn't in the house in which case I'd have bigger problems.

On locking the top to the bottom, this was the subject of an earlier thread on this. I got a lot of good advice about ways to secure it.

[Jan Bianchi]

Your construction method will be very good and satisfy the intent. No need to 2nd guess yourself in this regard. Keep in mind that a cabinet like this with the separate unsecured top needs to be kept against a wall and not moved when loaded. You should think of some way to mechanically fasten the upper to the lower to prevent any accidental tipping. The upper may be deep enough that this will not be a concern but you should at least be aware of the potential. Opening doors and pulling things down from a top shelf changes the dynamics.

A face frame changes the dynamics - it helps to hold the cabinet together and so mitigates the need for the sliding dovetails. A full bottom would help too but the bottom rail of the face frame (in this case more like a mid-rail) will add lots of structure to help keep the cabinet together.

I like the idea of the face frame changing things, but just to be clear, there is no bottom rail on this face frame. The top of the lower cabinet serves as the bottom of and the first shelf of the upper cabinet. So is a three-sided face frame enough to dump the dovetail idea?

[Jan Bianchi]

After much thought and cardboard model building, I have redesigned the upper part of the hutch reducing the shelves to 12 " rather than 15". I'm doing mostly for design purposes. I think the 16 inch top will be too big for the room and I don't need the extra width for storage. In looking at other hutch designs it seems that the upper parts were mostly meant for display not for storage and that it probably not an ideal design for storage but that decision has already set sail.

As for the shelf sag, I will dado and glue the shelves into the sides. I haven't decided on whether to just pin the shelves with dowels or screws at the back through the middle of the tongue and groove boards that make up the back panel as some of you suggested or whether to add three rails to the back frame and panel and dado grooves in them for the back of the shelves to sit in. I favor the former in terms of design, but would do the latter if rails behind the shelves would be substantially stronger.

[Sam Murdoch]

I like the idea of the face frame changing things, but just to be clear, there is no bottom rail on this face frame. The top of the lower cabinet serves as the bottom of and the first shelf of the upper cabinet. So is a three-sided face frame enough to dump the dovetail idea?

Sorry not to have replied - I somehow missed this last post of yours. In any case you have redesigned. The narrower shelves will be a good improvement - visually and functionally. As for how to support at the back - a few shelf pins would be a good solution but 1 screw in each shelf at or near center will be adequate unless you really intend to load these with dinner dishes - then I would use 2 screws spaced at 3rds. You could actually just use a machine screw set into a tight hole in the shelf as its purpose is to support rather than to hold together. The full thickness of the machine screw body will support more. Use a fine thread and keep the hole below the centerline of the shelf thickness to maximize the amount of wood above the screw.

[Kent A Bathurst]

I haven't decided on whether to just pin the shelves with dowels or screws at the back through the middle of the tongue and groove boards that make up the back panel as some of you suggested or whether to add three rails to the back frame and panel and dado grooves in them for the back of the shelves to sit in. I favor the former in terms of design, but would do the latter if rails behind the shelves would be substantially stronger.

Now you are, in my opinion, over-driving your headlights. ;)

I don't recall if you mentioned the width of the T&G boards in the back, but if they were, say 6"+/- you will have 7 or 8 across the width of the 44" shelves. If you put one screw through the ctr of each board, that is 7 screws holding up the shelf. [glued dowels are cool - except they will be forever hidden and you will have no bragging rights. use screws]

Unless you plan to store all your lead surf-casting fishing weights, gold bullion, and blocks from small engines on the shelf, you will hit what - max 40# per lineal foot max? Books don't weigh that much. I can't even get to that number, realistically. Even so - it would be max dead shear load of 15# - 20# per screw. Well within capabilities of any decent wood screw.

It is good that you are thinking through the loading question, and design options to address it. You are in the "ain't that dramatic" dimensional range in terms of extra work required to get home.