I install automatic pool covers for a living. The tracks that our cover rides on are composed of extruded anodized aluminum. These tracks come in 22' lengths. Usually there is more than one section of track, set end to end, on a given side of the pool. We drill and anchor these sections to concrete decks with NO gap between. Several years down the road, we notice that there is now a 1/8" - 1/2" gap between the sections. These systems are installed in the heat of the Summer and we usually come back to the jobs in the same heat. If that's the case, does this mean that there is some sort of loss happening during the expansion/contraction process that happens between the seasons and hot days/cold nights? I know that there's potential for deck heaving and expansion and contracting of the concrete, but this has been seen across the board and there's no way that so many different stones and concrete mixes have the same expansion/contraction ratio. I guess my end-all questions are "Is there Net size-loss throughout the expansion/contraction process? If the aluminum is 10 units long at 100º, then 9 units long at 50º, but only returns to 9.9 units long when brought back to 100º, where did that .1 unit of length go? Is it gas loss that allows the aluminum atoms to get closer together?"