I am sure that rarely is the finish itself the cause of cracking. In almost all cases, the basic cause of finish cracking with wood products is related to the wood substrate’s behavior. Let’s look more closely at what we call finish cracking.
Elastic and rigid
Wood is a dynamic material, even after kiln drying; that is, wood will swell when its moisture content increases and will shrink when its moisture content decreases. Further, changes in moisture occur whenever the relative humidity changes. A typical yearly cycle of average humidity can range from 25 percent RH in the wintertime when the heat is on to 50 percent RH or higher in the summertime when the windows are open. As a rough rule of thumb, a 5 percent RH change over a long timer period of a week or more can cause up to 0.25 percent size change in wood in width or thickness. That is, a 2.000-inch wide piece will change 0.005 inches when the humidity changes by 5 percent RH.
Most modern finishes used on wood are tough and durable, but are also quite thin and brittle. In the past, shellac, spar varnishes and other coatings were flexible even after curing. Also, today’s finishes, when they develop cracks, will develop a white line at the cracks.
Most, if not all coatings used on wood (the exception is wax) may slow the uptake and release of moisture, but none stop moisture changes and the resulting moisture movement. In other words, a finish does not seal the wood from moisture vapor changes (in spite of what you may read on the Internet or in advertisements).
Analogy
When we finish a piece of wood, what we really have is a moving substrate covered by a rigid coating. As an analogy, consider putting a rigid coating on a rubber band. Let the coating cure and harden. Then pull a little bit on the rubber band. What do you think is going to happen? The rubber band will move or stretch a bit, but he rigid finish cannot and so will crack, developing a multitude of cracks.
Now, consider the same rubber band that is stretched a bit and then coated. Now, after the coating dries, let the rubber band contract or shrink back to its original size. The band surface is now a bit smaller, but the finish is still the same size. To accommodate this stressful situation, the finish can develop a bunch of small cracks and at each crack, the finish can lift up just a bit, This lifting right at the crack will allow the larger finish to stay fastened (mostly) to the shrinking substrate.
Veener cracks
When veneer is cut from a log or cant of wood, the outer surface at the time of cutting will be pushed together slightly while the opposite surface that is in contact with the knife will actually have to bend a bit to allow the knife to pass. This bending will create a small amount of tension stress; should the tension be too large, a very small crack (called a knife check) will develop in the wood on this one side of the veneer. The check is often so small that it cannot be seen even under low magnification; nevertheless, there is a small crack in the wood’s surface. Veneer manufacturers can the compressed outer side the tight side and the knife side the loose side.
With properly manufactured veneer, the log is heated to reduce the risk of knife checks, knives are super-sharp, and a pressure bar may also be used to push on the wood to offset the tension.
During drying, if conditions are not proper, it is possible that the normal shrinkage can accentuate these knife checks. However, most drying is done properly to avoid this. Further, after drying, light sanding can remove these small surface imperfections; larger checks would be too large to remove. It could also be arranged, when gluing the veneer, that this loose side is glued so that the tight side is out.
It is the rule that all these checks are so small and so tightly closed at moderate humidity levels that they cannot be easily detected visually. The first time they are noticed is when the finish cracks.
Glue joint cracks and similar
In some cases, there are cracks in the wood prior to coating, such as from poor glue joints that opened up slightly. There are other causes of checking in wood. In my experience it is virtually impossible to detect these small checks prior to finishing. It is not uncommon to find that when such checked wood is exposed to drier RHs, the wood will shrink and open the pre-existing checks. Of course, the finish cannot handle this opening or movement, so the finish cracks.
Finish issues
If the checks follow the grain of the wood, certainly the substrate is highly suspect. If the checks are more haphazard, then consider a finishing issue. I am sure that in some cases, the extent of checking is accentuated by some finishing issues. A finishing practitioner friend of mine said, "From my experience with CV (which includes many pre-cat lacquers which are just CV hybrids), you can cause or aggravate checking 3 different ways:
1. Cold checking caused when the finish drops below 65 degrees anytime within 48 hours of spraying;
2. Over-catalyzation, especially likely when some finishers double the catalyst because it makes the CV dry faster especially in dusty environments; and
3. Exceeding the maximum dry film thickness limits."