## How accurately do I need to measure powder?

I am often asked which scale I recommend, and whether the A&D FX is precise enough for long range shooting. This article aims to answer these questions:

Once you have +/- 0.02 grains, any increase in precision will not make a difference to the practical performance of your handloads. Why? The short answer is that the effect is negligible, and you may be surprised how much so. For the long answer, read on.

- Is it really worth my time to perfect every charge weight to the kernel?
- Should I be cutting kernels into little pieces?
- What tolerance should I accept: +/- 0.02 grains, the nearest milligram, or better?
- Are the extra decimal places of more expensive scales worth the money?

Once you have +/- 0.02 grains, any increase in precision will not make a difference to the practical performance of your handloads. Why? The short answer is that the effect is negligible, and you may be surprised how much so. For the long answer, read on.

**First, the scale itself.**The A&D FX scale reports to the nearest 1 milligram. Here is a video demonstrating how it really works with kernels of Varget weighing 1.5 milligrams. It honestly is reliable to the milligram, if you close the doors to your room, use the side panels, and let the scale correct itself for zero after each charge.The Sartorius scale is technically much more accurate than the A&D, and more expensive. The Sartorius Entris 64-1S reports to the nearest 1/10 milligram. That is an extra decimal place. So in the above video when I drop one kernel, the Entris would have shown us the extra digit instead of rounding to the nearest milligram.

Note that if you operate the A&D scale in units of grains, it reports in divisions of 0.02 grains. That is actually 1.3 milligrams, so it is slightly less precise. However, this is still less than the weight of a single kernel of Varget.

Let's relate

That equates to

Now we can relate

If the measurement of the powder charge was the only factor that goes into the muzzle velocity variation of a load, then we would all be firing one-hole groups at 1000 yards. There are other random errors at play, most notably the combustion process itself.

We have to choose a value for "everything else", and we have to work in terms of standard deviation (not extreme spread) because we are talking about normally distributed random error. The SD of a well-developed load might be around 4, so let's use this as the example.

This is the point of confusion for most.

To illustrate the concept, I like to use a chronograph as an example. Suppose your chronograph has an inherent error SD of 0.1%, which is 2.75 fps. Now you would like to measure your ammo which actually has a true SD of 4 fps. Well, whether you like it or not, if you fired enough shots and measured the SD of what you get, it will not be 4. It will not be 6.75 either. You will get an SD of (drumroll please...) sqrt(4 x 4 + 2.75 x 2.75) = 4.85. Sometimes your chrono reads high when your bullet is slow. Sometimes your bullet is fast but the chrono reads low. When all is said and done, you get the square root of the sum of the squares.

Mr. Speedy is okay with +/- 0.02 grains which means an SD of 0.01 grains. This is an SD of 0.5 fps. Combining that with our 'everything else' SD of 4 produces a load with an

Mr. Careful applies a little extra effort to weigh to an SD of 0.25 milligrams, which is 0.19 fps. This produces a load with an

Mr. Perfect magically produces ammo with an

Now how does this relate to elevation on target? Well, for each of these loads, fired perfectly from a machine rest in completely still conditions:

The difference between these spreads is

This article should help you feel confident that points lost to elevation are not because you were slack in perfecting your powder charges. I believe +/- 0.02 grains is exactly the right amount of precision. With a capable scale and the Autotrickler, you can achieve this level of accuracy with less time and effort than other methods.

Please email me with any questions or concerns about this analysis: adamjmac@gmail.com

Note that if you operate the A&D scale in units of grains, it reports in divisions of 0.02 grains. That is actually 1.3 milligrams, so it is slightly less precise. However, this is still less than the weight of a single kernel of Varget.

Let's relate

**powder weight to muzzle velocity.**I will use my competition load as a reference: 308 caliber, 185 grain bullet, 30 inch barrel, with 44 grains of Varget. The muzzle velocity is 2750 fps. The relationship between powder and muzzle velocity is 50 fps / grain. It may vary from 35 to 100 for larger or small cases but 50 is a good working estimate.That equates to

**0.77 fps per milligram.**This means if I add one more kernel of Varget it will increase my muzzle velocity by 1.16 fps.Now we can relate

**muzzle velocity to elevation**at long range. For my load at 1000 yards, ballistics calculators tell us the relationship between muzzle velocity and elevation is**0.03 moa / fps**. That means that one extra kernel of Varget would put that bullet 0.035 moa higher on paper.**Now lets set up three scenarios for comparison:**(using fictional characters that bear no resemblance to actual people!)- Mr. Speedy would have us accept any charge within +/- 0.02 grains 95% of the time. This is how I recommend to use the autotrickler, because it will finish within this range most of the time and requires little effort from the user. Working in the unit system you are familiar with means you are less likely to make a mistake.
- Mr. Careful adds or removes a kernel by hand until the scale reports exactly the right number in milligrams. This is the best possible with the A&D scale, and what I'm sure many of us do to prepare for a national championship.
- Mr. Perfect settles for nothing less than absolute perfection. He reloads in an underground clean room and his powder charges have a weight error of exactly zero.

If the measurement of the powder charge was the only factor that goes into the muzzle velocity variation of a load, then we would all be firing one-hole groups at 1000 yards. There are other random errors at play, most notably the combustion process itself.

We have to choose a value for "everything else", and we have to work in terms of standard deviation (not extreme spread) because we are talking about normally distributed random error. The SD of a well-developed load might be around 4, so let's use this as the example.

This is the point of confusion for most.

**How do you combine two independent error sources?**__They do not just add together.__This is a key point to understand when applying statistics to shooting.To illustrate the concept, I like to use a chronograph as an example. Suppose your chronograph has an inherent error SD of 0.1%, which is 2.75 fps. Now you would like to measure your ammo which actually has a true SD of 4 fps. Well, whether you like it or not, if you fired enough shots and measured the SD of what you get, it will not be 4. It will not be 6.75 either. You will get an SD of (drumroll please...) sqrt(4 x 4 + 2.75 x 2.75) = 4.85. Sometimes your chrono reads high when your bullet is slow. Sometimes your bullet is fast but the chrono reads low. When all is said and done, you get the square root of the sum of the squares.

**So now back to our three scenarios.**For the purpose of converting our weighing strategies into SD, we divide the total range by 4. Because for a normal random variable, 95% of samples are within +/- 2 SD.Mr. Speedy is okay with +/- 0.02 grains which means an SD of 0.01 grains. This is an SD of 0.5 fps. Combining that with our 'everything else' SD of 4 produces a load with an

**SD of 4.031 fps**.Mr. Careful applies a little extra effort to weigh to an SD of 0.25 milligrams, which is 0.19 fps. This produces a load with an

**SD of 4.005 fps.**Mr. Perfect magically produces ammo with an

**SD of 4.000.**Now how does this relate to elevation on target? Well, for each of these loads, fired perfectly from a machine rest in completely still conditions:

- Mr. Speedy: 95% of shots will fall within 0.4837 moa.
- Mr. Careful: 95% of shots will fall within 0.4806 moa.
- Mr. Perfect: 95% of shots will fall within 0.4800 moa.

The difference between these spreads is

**less than 1 millimeter at 1000 yards.**Consider how much we struggle to keep the bullets within a 1 moa circle in match conditions. If you add all the other error sources like aiming, mirage, wind, and the gremlins, you can work out mathematically how little that last kernel really means.This article should help you feel confident that points lost to elevation are not because you were slack in perfecting your powder charges. I believe +/- 0.02 grains is exactly the right amount of precision. With a capable scale and the Autotrickler, you can achieve this level of accuracy with less time and effort than other methods.

Please email me with any questions or concerns about this analysis: adamjmac@gmail.com