Have you ever air-dried a stack of pine boards and wondered why some bowed and some didn’t? Or noted that perfectly constructed joints swelled to a not-so-perfect fit in the summer? Or found that your dining room chairs wobbled more in the winter? If so, you’ve begun to puzzle over the dynamic relationship between wood and water.
Trees, of course, need water for their life processes. When they’re felled and brought to a sawmill, they’re still full of water and very heavy. At the mill, the boards are stickered – stacked with spacers – so that air can flow between them and dry them out, although, as we shall see, “dry” is a relative term.
Moisture content varies from species to species. In some softwoods, green boards can be three times heavier than oven-dried ones. That’s a moisture content of 200 percent, which seems counterintuitive. I’ll explain that number in a bit. But contrast it with white ash, which has a moisture content of about 45 percent, and it’s easy to see why white ash is the wood of choice if you need to burn green wood in your stove.
When wood dries, interesting and sometimes frustrating things happen. To understand why, we’ll look at what’s going on at the cellular level. Water is found in two distinct places in the cells: in the lumens (cavities) and in the water that is bound in the cell walls. As wood dries, the water in the lumens, sometimes referred to as free water, is the first to leave.
When all the free water in the lumens is gone and only the water bound in the cell walls remains, wood has reached what is called its fiber saturation point. The moisture content at this point varies a bit, both between and within species, but averages about 30 percent. There is no change in the wood’s size and shape as long as it is above the fiber saturation point. It is only when the cell walls start to dry that we begin to see changes in the size and shape of the wood.
The amount wood can dry is influenced by the temperature and relative humidity of the surrounding air. There comes a point where a piece of wood will neither gain nor lose moisture. This is called the equilibrium moisture content, or EMC. In general, the higher the humidity at any given temperature, the higher the EMC, and the higher the temperature at any given relative humidity, the lower the EMC.
We all know, however, that the relative humidity and temperature are always changing, both through the day and seasonally, so we speak of a year-round average EMC. In the Northeast, this is about 8 percent inside buildings and 12 percent outside, but the variation over time can be considerable. For instance, a wood floor laid over a radiant heat system can have an EMC in winter of 2 to 3 percent, whereas the EMC on a hot, muggy summer day can be 20 percent or more. One of the purposes of wood finishes is to slow down the transfer of moisture between wood and the air, thereby slowing down changes in the moisture content of the wood as the temperature and humidity of the surrounding air changes.
Machining properties also change as wood dries. Most hardwoods machine quite well with a moisture content in the range of 6 to 8 percent. Since this is the same as the average EMC for inside buildings, hardwoods are typically planed and otherwise shaped at this moisture content. Softwoods, however, generally machine poorly at a low moisture content – there is lots of grain tearing and other machining defects – but they machine quite well with a moisture content in the range of 12 percent. Therefore, when destined for interior finish uses, softwoods are typically dried to about 12 percent moisture content.
Size and Shape Change
When you go to the lumber yard to get some two-by-fours, you find that some are straight while others are bowed or twisted. If you’re a woodworker, you are familiar with boards that are cupped (bowed across the width of the board) and otherwise warped. What causes boards that were straight and flat when they came off the sawmill to change shape when they dry? It’s because wood is anisotropic, meaning that it has different properties in different directions.
We can think of tree trunks having three directions: length (up and down the tree; parallel to the grain), tangential (parallel to the growth rings), and radial (from the center of the trunk to the outside; perpendicular to the growth rings). As the moisture content of normal wood dips below the fiber saturation point, it shrinks differently in these three directions.
Shrinking that occurs along the length of the board is called longitudinal shrinkage, and it’s generally quite small. If a board dries from its fiber saturation point (about 30 percent moisture content) to oven dry, it’ll only lose 0.1 to 0.2 percent of its length. For most uses, the moisture content will not vary by more than about 3 percent over the seasons, yielding about one-tenth of this shrinkage and expansion through the seasons. An 8-foot board (96 inches) would change in length by only 0.01 to 0.02 inches, which would be imperceptible in almost all uses.
Tangential shrinkage is a different story. Wood shrinks markedly in this direction as it dries from the fiber saturation point to oven dry, in the range of 7 to 12 percent for most of our northeastern hardwoods and 6 to 8 percent for our softwoods. Thus, a 10-inch-wide board, with a seasonal moisture content variation of 3 percent, could be expected to change in width by up to 1/8 of an inch. This will certainly be noticeable in many applications.
Wood typically shrinks roughly half as much in the radial direction as in the tangential direction: in the range of 5 to 7 percent for our northeastern hardwoods, and 2 to 4 percent for our softwoods. This explains why quarter-sawn lumber (boards where the width is in the radial direction) is often considered more stable than flat-sawn lumber, and is sometimes specified for premium products.
Note that the above information applies only to normal wood. There are four types of wood that do not behave normally: compression wood, tension wood, juvenile wood, and wood with sloped grain. Compression wood and tension wood are referred to as “reaction wood.” They form in the trunks of trees that are not vertical, with compression wood grown in conifers under the lean (as if it is trying to push the tree upright) and tension wood in deciduous trees on the upper side of the lean (as if it were trying to pull the tree upright.) Reaction wood also forms on branches, and at the junction of branches with the trunk. Juvenile wood is the wood that is within five or ten growth rings from the pith. And wood with sloped grain is wood in which the direction of the grain is not parallel to the length of the board. It occurs near the ends of boards where there was a large butt flare in the log, and in trees with spiral grain, where the tree’s cells are not oriented straight up and down the tree, but rather spiral around it.
All these types of wood shrink markedly in length upon drying below the fiber saturation point. Note that reaction wood and juvenile wood are likely to be present in only part of any given board. This means that the board will try to shrink in length in some parts and not in others, which leads to bowed boards. This is quite common in preservative-treated boards at lumberyards, which are often made from plantation-grown southern pine. These trees have a lot of juvenile wood because they grew so rapidly: each growth ring is large. While the wood is excellent for pressure treating (it absorbs the preservatives readily), the large amount of juvenile wood causes problems. When the moisture content is above the fiber saturation point (and the wood comes out of the pressure-treatment cylinders fully saturated, and is rarely dried before shipment to lumberyards), the boards are nice and straight. Upon drying, however, the longitudinal shrinkage of the juvenile wood can result in dramatic bends. Boards with sloping grain can have a significant hook at the end (where the sloping grain is caused by a butt flare) or twist (where the tree had spiral grain.) In some cases, there can be so much twist that the boards look like propellers.
Even normal wood changes shape as it dries. Round crosssections don’t stay round, squares distort, and boards that aren’t quarter-sawn cup. The drawing on page 46 shows the shrinkage and distortion that can be expected in normal wood.
The difference between radial and tangential shrinkage also causes problems. Since the radial shrinkage is about half the tangential shrinkage, the surface is stretched as drying proceeds. When the stress exceeds the strength of the wood, it breaks, creating cracks (called checks) parallel to the grain. Occasionally, a single large check will develop, but more commonly there will be many smaller checks. On square timbers with the pith in the center, it is common to have a significant check roughly centered on each face.
The owners of new log cabins and timber frame homes often notice an interesting side effect of the checking in logs and timbers: sometimes a check forms with a big bang. This is most likely to occur in winter, when the low humidity causes the wood to dry relatively quickly.
Compounding these stresses is the fact that wood dries from the surface inward. The surface has a lower moisture content than the inside. The faster the drying, the steeper the gradient in moisture content and the more checking. This gradient has consequences for lumber, especially thick lumber, as well as for poles and timbers. It can cause problems including surface checks (which sometimes close as the lumber fully dries, only to show up again when finish is applied), honeycombing (where the interior of the board collapses), and case hardening (where boards pinch saws and cup when being re-sawn into thinner material).
People who work with wood, whether constructing buildings, making furniture, or laying floors, can do a better job if they consider how the wood may change size and shape as its moisture content changes. Understanding how moisture content is measured in wood is the first step.
For lumber, plywood, and similar wood products, the moisture content of wood is calculated using this formula:
Moisture content = (wet weight – oven dry weight) x 100%
oven dry weight
This formula is counterintuitive: if more than half the weight of a piece of wood consists of water, you end up with a moisture content over 100 percent, and how can the water in a piece of wood be more than 100 percent? The beauty of this formula, however, is that it uses a constant for any piece of wood, the oven dry weight, as the basis for comparison to determine moisture content.
The most accurate method for determining the moisture content in wood is to cut a sample, weigh it, dry it in an oven at 218°F until the weight stops decreasing, and then use the above formula. Obviously this method has some severe disadvantages: samples must be cut, ruining boards or finished products, and it takes a fair bit of time for the drying, so the results are anything but fast.
To overcome these problems, moisture meters have been developed that measure the electrical or electro-magnetic properties of wood. Since these properties change with changes in moisture content, these meters can calculate the moisture content. When used properly, they can quickly and accurately determine the moisture content of dry wood, though they don’t work well at moisture contents above 30 percent.
You can build with freshly sawn lumber; some people certainly do. (The squirt of sap when a nail is pounded home is sure evidence that the lumens contain plenty of free water.) But using green lumber has some disadvantages. For one, wet lumber is much heavier than dry lumber. More substantively, it can be difficult to tell which pieces contain reaction wood or spiral grain, and upon drying these will try to bend or twist, potentially causing aesthetic or structural problems. Another problem is that as the wood shrinks, cracks may open between pieces and the load will then be carried in unanticipated ways. Drying lumber first can reduce these problems: in addition to being lighter, any pieces of lumber that behave badly by bending or twisting can be discarded or used where their defects don’t matter.
Wood is a wonderful material. It can be renewable, local, recyclable, and has far less embedded energy than most other materials. And it is beautiful. Many of the challenges of using wood are caused by the changes in properties, size, and shape that occur as the moisture content changes. Understanding the effects of moisture in wood makes it possible to factor them into the design of everything from buildings to furniture to toys.
Irwin Post is a forest engineer living in Chester, Vermont. He has extensive experience working with wood, including sawmilling, air and kiln drying, building, and woodworking.