Replies

1. 
   jfsksa | Aug 07, 2020 05:04pm | #1

   I don't know about rules, but the only place I use screws on cabinet construction is to attach hardware. The wood goes together with mortise and tenons, the drawers with dovetails or box joints and the drawer bottoms sit in dados.

2. 
   lat_axe | Aug 07, 2020 05:21pm | #2

   The basic issue is that wood expands & contracts across its grain 3 - 4X the expansion & contraction along its grain, for a given change in humidity that gets from the surrounding air into the wood.

   This differential varies somewhat with the species of timber(s) involved. So, therefore, does the length of a cross to long grain join you can glue without issue.

   If you glue long grain to cross grain along a joint of any significant length (for example, a 3 ft breadboard-end glued across the whole width of a table top end) you risk differential expansion causing a crack if the piece will be subject to significant humidity changes that cause it to expand or contract to a certain degree.

   Its not always easy to know what humidity changes a piece might be subject to. It depends on your geographical area and to what degree the house or other location for the furniture is isolated from external atmosphere changes by central heating, air conditioning and so forth.

   But long grain to cross grain joins are always at some risk. And you can't know all the future life and location of a piece of furniture for certain.

   There are many joinery techniques for dealing with this issue. Using screws that will pivot or otherwise allow differential movement between two joined pieces is one method. There are many others. Which method you employ to avoid or reduce joint failure due to humidity changes depends on the style of the piece, the nature of the joins and many other factors.

   Fine furniture doesn't necessarily avoid the use of screws. But it is usual to employ other methods for many styles - sliding dovetails and other mechanically strong "dry" joints that can move without coming apart, for example.

   Screws alone won't solve the issue. They have to be used in a way that allows the joined parts to move differentially without tearing out the screw. One method, for example, is to have the head-end of the screw go through a too-large hole in the wood but via a washer that allows the head-end to slide or pivot as the wood part it retains moves differentially to the part it's joined to by the screw thread end.

   But what screw-joint did you have in mind yourself?

   Lataxe

3. 
   fourier_transform | Aug 07, 2020 05:34pm | #3

   jfsksa may have left you with the wrong impression about screws.

   Screws are used often when they can't be seen and/or to permit wood movement. One example is attaching drawer runners. Drawer runners typically run across the grain, so they can't be glued.

   Another example is, the back of solid wood cabinets. They are often attached with screws because of the wood movement issue. For example, a back often is made of slats that 'connect' to each other with a ship lap joint. Each slat is screwed to the back of the cabinet.

   By the way, the only rules in woodworking are: "Be sure to read, understand, and follow all the safety rules that come with your power tools. Knowing how to use your power tools properly will greatly reduce the risk of personal injury. And remember this, there is no more important safety rule than to wear these: safety glasses!"

4. 
   user-2368618 | Aug 07, 2020 06:06pm | #4

   Thanks for your replies.

   I will provide a little more detail on my approach. Curious to know if it sounds appropriate.

   I am using dovetails on ogee feet, with horizontal support piece glued in a dado in each foot. The front of the base to which feet are attached is mitered with M&T joints for sides and back. Feet and support are attached to base with screws (and glue?).

   Base is attached to carcase bottom with screws (and glue?).

   Drawer dividers are glued in sliding dovetails, with drawer runners and rear left floating.

   Mahogany "filler" covering carcase bottom which extends in front of carcase sides to form a shelf will be glued but not screwed (since screws are not to be visible). The filler will be about 7 inches wide, with the grain running in the same direction as the secondary carcase bottom. A drawer will pull out over the filler. I don't believe there will be a wood movement issue but I'm not sure.

   Thanks for bearing with me on this!

5. 
   lat_axe | Aug 08, 2020 03:37am | #5

   That description is a bit too complicated to easily visualise from the words alone. Do you have some kind of drawing or plan? Could you make a sketch or two?

   In some cabinets I've made, the body of the cabinet has been unattached to the base, with gravity and profiled upper edges of the base (raised a little above the base's inner platform) keeping the cabinet carcass in place on that platform. This can avoid some of the potential movement issues, such as having the cross grain of a cabinet side glued to the long grain of the base on which it stands.

   Lataxe

6. 
   rob_ss | Aug 08, 2020 06:21am | #6

   I think from your description, you will be fine, though it is hard to follow.

   Your issues with wood movement only become a problem where you have grain running in different directions. This is because the wood will move more or less the same amount relative to it's surrounding wood.

   There are well-described ways to handle this.

   Large panels should 'float' in a frame so that there is room for them to do their thing, drawers should have reveals to allow for movement. Table tops need to be fixed down with movement in mind.

   Provided you accept that where grain runs in a different direction, you will need to allow for the differential expansion across the grain compared with along the length. About 1/8 inch per foot is commonly quoted, though this will be less for quarter-sawn and more for flat-sawn timber.

   It will also depend on where you plan to live.

   In my location, the humidity does not change a great deal Exterior joinery binds because it gets wet, but if it is not rained on, wood stays nearly the same size. I can build stuff without much concern for grain direction, but I do allow for movement in case I sell the piece or move house.

   To an extent, it also comes down to 'look'

   If it 'looks right' then it will probably work. This is because your ideas of what 'looks right' are informed by you having seen hundreds of pieces of furniture, most of them made by people who did 'know the rules'

   As to when and where to glue and when to screw, you should glue where you are needing a permanent joint AND where the grain is in the same direction. Usually the only places to avoid glue where grain runs the same way is drawer bottoms and panels. A little dab of glue, a single nail or screw to retain a part yes, but not down the grooves.

   Screws can be used anywhere that glue is ok, but have the advantage that if you fit them in slotted holes, they can restrain a table top whilst allowing it to move. Many antique pieces of furniture used screws and nails where these were the most cost-effective way to get the job done.

7. 
   user-2368618 | Aug 08, 2020 06:48am | #7

   Thanks to all! I believe I have a better understanding on how to proceed. And if I use screws, I will be sure to consider slotted holes. I live in Maryland, so I will have to consider wood movement, given the changes in weather here. Thanks again!