The powder volume mass changes with the same weight. There is a post on here somewhere that showed volume changes with the same case weight. Case weight has little to no connection to case volume. The powder volume to case volume is the critical measurement.
This is simply not correct. People post this sort of thing all the time and it's simply not true. I'm convinced that in most instances, this misleading opinion results largely from user error in accurately determining water volume. The simple fact is that the only places in which brass that has been
fully and uniformly expanded by firing at 50K to 60K+ psi in a rifle chamber can change in thickness without affecting internal volume are in the extractor groove and the primer pocket. In fact, it is only the
variance in the volume of the extractor groove and/or primer pocket between different cases that really matters. The total variance between the extractor groove and primer pocket volumes of different cases as a percentage of the total internal case volume simply isn't that large. It just isn't. Although you will always find some "outliers" where the data point of case weight (y-axis) and case volume (x-axis) does not lie directly on the scatter plot trend line, determination of correlation coefficients supports the conclusion that case internal volume generally exhibits a good linear correlation with respect to case weight.
Further, powder mass doesn't "change", and I'm guessing "powder volume mass" is a term that you made up for the purpose of this discussion.
Assuming that a given mass of powder has not been artificially compressed or unevenly packed in such a way as to give a dramatically different packing density, the major change we need to be concerned with is the volume of the pressure cell in which a given mass powder is ignited. As internal case volume decreases, uniform ignition of a given mass of powder will create higher pressure due to the expansion of the same amount of gas in the smaller internal case volume (i.e. pressure cell). If a specified mass of powder is unevenly packed, resulting in non-uniform density from case to case, minor differences in internal case volume are the least of your worries; the pressure will not be uniform even if all the cases have exactly the same internal volume. Although it is possible to alter the packing density for a given mass of powder in a case, for example by using a longer drop tube or a vibrating table to settle or compress the powder, it will generally be relatively uniform from case to case, even between those of slightly different internal volume, as long as the reloader's technique is fairly consistent and the load is not highly compressed once a bullet is seated.
The whole point of this thread is the use of a much simpler and more expedient method of weighing cases as a surrogate for determination of their actual water volume. Is it perfect? Of course not. As I mentioned earlier, you will always find a certain number of outliers in a scatter plot of case weight versus case volume. I have never observed ALL the plotted values to lie directly on the trend line. However, most are not "gross" outliers, and plot reasonably close to the trend line. Close enough, in fact, that that the correlation coefficients indicate a strong linear correlation between case weight (x) and case volume (y) as I mentioned above. As a result, sorting cases by weight will typically result in more uniform internal volume than doing nothing at all. Whether the time and effort of doing so is actually justified by the relatively small benefit probably depends on a number of factors, including the brass brand/manufacturer, inherent accuracy/precision of the rifle/load setup, and the relative skill level of the shooter, themselves. In other words, everyone need to do their own testing and decide for themselves whether it is worth the effort.