The main thing to remember about keeping things straight is to have all the contributing factors balanced evenly. I cannot count the number of times I have heard that the answer to a straight blade is to eliminate the “stresses” or to get the blade “stress-free”, but this is fundamentally incorrect. First, one can never totally eliminate “stress” and second one must have “stress” in the blade in order for it to be hard. That is how hardening works, with the quench we induce massive amounts of strain on the entire internal lattice of the steel. I use the word “stress” in quotes because there are some technical semantics involved in these terms that is not totally accurate but it is just easier for the most folks to understand if I go with the commonly used words, even if they are not technically accurate.
Anyhow, since it is not only impossible, or even undesirable, to eliminate all stress, the real trick is to make it balanced and even in the blade. When you have more strain energy down one side of the blade than the other it is going to warp. This is why the major contributing factors to distortion in a quench are things like how the vapor jacket collapses. If it is minimal and collapses evenly and quickly things work. But if it collapses in a patchy way, one side of the blade will get more exposure to the quenchant than the other and you will get distortion.
The same is true of how the inside of the steel is setup to go into the quench, any uneven strain effects will react to the heating and cooling… well … unevenly. Also heavy segregation of carbide or phases in the steel will also be a major culprit in uneven strain. Put both of these together and one can see that the single most important way to help with warping is very good normalizing practices, followed by careful annealing and stress relieving. Every time I hear a smith point out things in the forging that causes warping in the quench, I want to set them down and explain proper normalizing to them, because get the feeling nobody ever has.
Damascus should[U/] be no different than any other steel, however it can have some quirks if certain factors are not taken into consideration. The most common issue one could encounter is a mismatch of alloys in the mix, made as common as it is by folks not even being aware of the importance of properly matching steels according to heat treat needs. The majority of folks I have talked to started out making damascus with a prime consideration of pattern contrast only, with not really considering if the two steels/metals should be put together even if they do look pretty together. One of the best examples of this that I used in a demonstration at Ashokan years ago was 160 layers of 1095 and L6 in a twist pattern square bar that I passed around the group to let them see it was perfectly straight. I then normalized the bar once and passed it around to let them see the 1/4 rotation the bar distorted into because of the two radically different behaviors of those alloys in heat treating.
Another factor that has to be considered with damascus is the evenness of the pattern distribution. The best example I have for this is a pattern I like to use for fittings, it is a double (chevron) twist welded to a random so the face of the butt-cap or guard gives you the chevron but the side butting up to the handle has nice straight lines down the sides. This looks nice, but it is a nightmare to keep straight, every single time you heat a piece of that stock it warps due to the differential expansion of the two patterns. Another stellar example is a Saxon/Migration Period sword that has multiple twists alternating with straight bars that are offset from each other on either side of the blade. Anybody who has had the “pleasure” of heat treating one of those things knows that they will come out of the quench looking like an accordion.