I’d like to hear more about the weighing the powder charges. I probably won’t understand the math, but I’m interested in the effects.
The math isn't very painful.
The basic concept is that variation does not simply add. If you're shooting 6" groups offhand at 50 yards with a perfect rifle, switching to one that does 2" groups at 50 yards (an SKS!!) will NOT make your groups 8". It will make them more like 6.3". If there is one large source of variation in the chain, it will almost completely determine the total variation. That is why fiddling with small sources of variation is pointless. You have to find the big sources if you want to make any progress.
One good measure of variation is standard deviation. The higher your standard deviation, the more spread out your data are (and the higher your "extreme spread" generally will be). Standard deviations add by the square root of the sum of the squares. It sounds forbidding, but it's really not so bad if you work through it step by step.
Take the case of the 5.56/223. In a small case like that, small changes in the powder charge are more important than they are in a large case like the 30-06. So this small cartridge is sort of an "acid test". The changes in larger cartridges will be less important.
With a stick powder like Varget, my powder measure throws charges with a standard deviation of .11 grains. In the 223, near normal loads, a grain of powder is about 100 FPS in MV. So a standard deviation of .11 grains in charge produces .11 x 100 = 11 FPS standard deviation in muzzle velocity.
It's not too hard to get the standard deviation of 5.56/223 handloads down into single digits, but commercial ammunition tends to run at about 30 FPS standard deviation of muzzle velocity.
So for purposes of illustration, assume that a handloader is making 5.56/223 ammunition with a standard deviation of 30 FPS in muzzle velocity. As part of the process, the handloader is using a lab grade scale, and is creating powder charges down to the last 1/10 of a granule of powder, essentially perfect loads.
OK... starting from perfectly measured loads, and a 30 FPS standard deviation in MV, what would be the effect of switching to my Lee Perfect Powder Measure that has a standard deviation of .11 grains?
First, we square the two standard deviations involved:
30^2 = 900
11^2 = 121
Now we add the two squared numbers: 900 + 121 = 1021.
Now we take the square root of the sum of the two squared numbers: square root 1021 = 31.95.
So going from a perfect measurement of powder to my $27 powder measure increases the standard deviation of muzzle velocity from 30 FPS to 31.95 FPS.
QED
I hope you're not sorry you asked!
