The splitting of wood that everyone is reporting is caused entirely by wood shrinkage. Wood shrinks as it dries, and swells again as it absorbs moisture from the air. I'll explain wood shrinkage, and people can then understand exactly what's happening and why, and can then formulate their own ideas on what to do to prevent further shrinkage.
When wood is alive, the wood cell cavities are full of water and the wood cell walls also have lots of water inside them. The difference is that the water inside the cell walls is "bound water", or water that's chemically bound to the cellulose that the cell walls are made of. As the wood dries, the water preferentially comes out of the wood cell interiors. Absolutely nothing happens to the wood while the cell interiors are drying out except that the wood gets lighter in weight because of the evaporation of that water.
Once all of the water has evaporated from the wood cell interiors, and at around a 28 to 31 percent water saturation (compared with the original water content) all the water is out of the wood cell interiors, and the bound water starts evaporating from the wood cell walls. This moisture content (where the cells are empty, but the cell walls are still saturated with water) is called the "fiber saturation point" cuz only wood fiber is saturated with water. When water evaporates from the wood below the fiber saturation point, the water comes out of the cell walls, and the cell walls get thinner and stiffer (just as most things do as they dry out). The result of the wood cell walls getting stiffer as the wood dries is that lumber actually becomes stronger as it dries below the fiber saturation point.
Now, wood cells are shaped like long drinking straws with a wedge shaped (kinda) cross section, and they're closed off at the ends. There are "doorways" along the wood cells where moisture can move from one cell interior to a neighboring cell interior. The wood cells are about 80 times longer than they are in diameter. As the wood dries, the wood cell walls shrink in thickness, but that shrinkage isn't uniform.
Because wood cells are very long, there are many more cell walls going across the grain of wood than along the grain, so the shrinkage along the grain is generally negligible (except that it causes truss uplift in houses with trusses in the attic). For just about everything else, longitudinal shrinkage in wood is presumed to be zero. The shrinkage in the radial direction from the center of the tree trunk is much greater, typically being 3 to 5 percent from the fiber saturation point to an oven dried condition (0% moisture content). However, it's the shrinkage tangential to the growth rings that's the greatest, and varies from about 5 to 7 percent from the fiber saturation point to an oven dried condition in most softwoods used for construction like fir, spruce, cedar and pine and redwood in North America. Imagine if you had a fir post 6 inches in diameter. It's circumference would be about 19 inches. However, if you put it in an oven and dried it, it's circumference would go down by 7 percent, and then it would be 17.5 inches in circumference. The result would be that there would be a 1 1/2 inch wide split in that post at a 0 percent moisture content.
Wood Shrinkage Table
Generally, hardwood shrink much more than softwoods because the wood cells in hardwoods are smaller, so there are more wood cell walls per inch to shrink as the wood dries, but shrinkage is highly species specific.
Dr. Stephen Smulski used to teach at the Building Materials and Wood Technology department at the University of Massachusetts at Amhurst. He has since left teaching and heads his own wood consulting company. One of the paper's he's best known for is called "Detailing for Wood Shrinkage" and you can read it here:
Minimizing Wood Shrinkage Problems
Anyhow, wood both absorbs water and water evaporates from wood 15 times faster at the end grain of the wood than across the grain of wood. So, if you're ever at a lumber yard and you notice that EVERY SINGLE DAM 2X12 in the whole F$*%$#ing lift is split for about 18 inches from both ends, that's the reason why. Because wood evaporates 15 times faster from the end grain of the wood than out of the sides of the wood, the ends of the 2X12 will dry out faster than the wood a foot or two from the ends. That means that there will be tension in the wood at the ends (where each wood cell wall is trying to shrink in thickness), but there won't be a foot or two from the ends where the evaporation rate is much slower and the moisture content in the wood has time to equalize across the cross section of the wood as it dries. It's that rapid drying at the ends of the 2X12's that causes them to split at their ends, and every one will be that way.
The way to stop that in a 2X12 would be to paint the ends of the 2X12 as soon as it's out of the drying kiln with a high gloss oil based paint. That paint would form a barrier to evaporation so that the wood wouldn't evaporate out the end grain at all, and therefore the joist would dry out much more uniformly all along it's length. That would prevent 2X12's from splitting at their ends. Lumber companies know this, but they pretend to be stupid. They teach their employees to test the moisture content of the wood in the kiln by pushing the pins of the tester into the end grain of the wood. And, of course, the end grain of the wood is ALWAYS gonna be dryer than the rest of the wood. But, good employees pretend not to realize that. If you're buying hardwood to make furniture or musical instruments, the wood has been stored for many years before even being put on sale to ensure it has fully come to an equilibrium moisture content with typical indoor air temperatures and humidities. Only then is it cut to shape.
So, if you ever buy a 2X12, cut the split ends off, and then paint the end of the joist with high gloss oil based paint to prevent further splitting.
When I was a kid, EVERY piece of 2X dimensional lumber was painted at it's ends, but now they don't do that anymore, and I can't understand why not.
And if it's larger dimensional lumber like a 4X6 or a 6X6, the same principles apply, but it's harder to put into practice simply because the shrinkage you're most concerned about is tangential around the center of the tree trunk. Certainly, I'd paint the ends of the post with a high gloss oil based paint, to prevent any moisture at all evaporating out the ends. But, I'd also paint the entire outside surface of the lumber as well, but with a latex primer instead this time. If it's cedar or any wood you don't want to paint, then I'd apply a light coat of linseed or tung oil. That will penetrate into the surface of the wood and help prevent moisture evaporating from the wood, too.
And, if it's pressure treated lumber, I'd paint the end grain with oil based paint immediately, but wait a year to paint the exterior with a latex primer. The reason why you're not supposed to paint pressure treated wood for the first year or two is because the pressure treating process involves pumping water treated with chemicals into the wood's surface. And, as that moisture evaporates from the wood, it'll lift off any paint. So, the recommendation to wait a year to paint pressure treated lumber is because the wood is still wet with water from the treatment process.
This site won't let me post anything longer than 10,000 characters, so I'll continue the response in the next post.
One split is about 1/4 inch wide, and I am quite cheesed off. Is there a repair to this, or do I need to pull all the posts out and replace them? I have a feeling I am going to replace them, but I am hoping that someone will have a better solution. I have custom cut each one for railings, and they are bolted and glued to the deck frame. Thanks.
