Home Forums Reloading Big Game Rifles typical variation in powder throw

Weighing each charge certainly does no harm, and if people want to do it, I make no objection.

However, consider the following bit of math:

Random variation adds by the square root of the sum of the squares, not linearly. That's a mouthful, and may not mean much to most folks. But it has profound implications for shooters.

The 223 has a small case and is therefore more sensitive to powder variation than say the 270 or 30-06. In round numbers, a grain of powder is 100 FPS with 55 grain bullets. Typical commercial ammo has about a 25 FPS standard deviation.

So, for example, assume that you have a lab grade scale capable of measuring to the milligram. Also assume that your ammo is a bit better than most commercial stuff, and has a standard deviation of 20 FPS. Based on my experience, that's in the ballpark.

Now assume that for the sake of convenience, you want to start using the inexpensive Lee Perfect Powder Measure, and that you are using ball powder. In that case, I have measured the standard deviation of the powder measure, and it is right around .04 grains. .04 grains is about 100 FPS X .04 = 4 FPS standard deviation in muzzle velocity.

Now do the sum of the squares math: (20^2 + 4^2)^.5 = (400 +16)^.5 = 20.39 FPS.

So in that case, moving from an extremely good lab scale to an inexpensive powder measure increases that standard deviation of muzzle velocity from 20 FPS to 20.39 FPS. You'd have to shoot a really, really large sample to even show that the change had happened.

I also tested the variability of H4831SC. Of course, that is not a suitable powder for the 223, but I expect that Varget is at least close in measurement variation. In that case, the standard deviation of the powder charge is .1 grain. So the standard deviation of the muzzle velocity due to powder charge is .1 x 100 FPS = 10 FPS. Doing the math, (20^2 + 10^2)^.5 = 22.36 FPS, from a process that was formerly 20 FPS.

The day I did that math, I quit individually hand weighing charges.

The result of the math is:

1. If one source of variation is much larger (say 2X) than the others, then it alone almost completely determines total variation.

2. You will never succeed in improving a process by working on several of the weak variables. You have to find the strong contributors, and work on those. In most practical situations, the main source of variation is the shooter. So there is more truth than poetry in the coffee comment.

Weighing each charge certainly does no harm, and if people want to do it, I make no objection.

However, consider the following bit of math:

Random variation adds by the square root of the sum of the squares, not linearly. That's a mouthful, and may not mean much to most folks. But it has profound implications for shooters.

The 223 has a small case and is therefore more sensitive to powder variation than say the 270 or 30-06. In round numbers, a grain of powder is 100 FPS with 55 grain bullets. Typical commercial ammo has about a 25 FPS standard deviation.

So, for example, assume that you have a lab grade scale capable of measuring to the milligram. Also assume that your ammo is a bit better than most commercial stuff, and has a standard deviation of 20 FPS. Based on my experience, that's in the ballpark.

Now assume that for the sake of convenience, you want to start using the inexpensive Lee Perfect Powder Measure, and that you are using ball powder. In that case, I have measured the standard deviation of the powder measure, and it is right around .04 grains. .04 grains is about 100 FPS X .04 = 4 FPS standard deviation in muzzle velocity.

Now do the sum of the squares math: (20^2 + 4^2)^.5 = (400 +16)^.5 = 20.39 FPS.

So in that case, moving from an extremely good lab scale to an inexpensive powder measure increases that standard deviation of muzzle velocity from 20 FPS to 20.39 FPS. You'd have to shoot a really, really large sample to even show that the change had happened.

I also tested the variability of H4831SC. Of course, that is not a suitable powder for the 223, but I expect that Varget is at least close in measurement variation. In that case, the standard deviation of the powder charge is .1 grain. So the standard deviation of the muzzle velocity due to powder charge is .1 x 100 FPS = 10 FPS. Doing the math, (20^2 + 10^2)^.5 = 22.36 FPS, from a process that was formerly 20 FPS.

The day I did that math, I quit individually hand weighing charges.

The result of the math is:

1. If one source of variation is much larger (say 2X) than the others, then it alone almost completely determines total variation.

2. You will never succeed in improving a process by working on several of the weak variables. You have to find the strong contributors, and work on those. In most practical situations, the main source of variation is the shooter. So there is more truth than poetry in the coffee comment.

as the OP...... I wasn't asking out of concerns for accuracy. I get good groups from most of my loads.

I was mostly curious if something might be wrong/dirty/jacked up with my powder measure. doesn't sound like there is, since others report similar variations.

I fearful that if something wasn't working right, if it might cause an unsafe condition. since the variations seem normal, and I'm staying in the recommended load ranges, it looks like I'm not causing an unsafe condition.

so many thanks to all respondents.

Very nice indeed.

If you like individually weighing charges, then I say go for it. A warm fuzzy feeling is reason enough.

When I see big changes in velocity when changing lots of powder, I see much smaller changes in volume than in weight.

Do you have anything to add about that, Denton?

When I see big changes in velocity when changing lots of powder, I see much smaller changes in volume than in weight.

Do you have anything to add about that, Denton?