Do you test to pass or test to fail?

[old_heli_logger]

Do you look for passing data and stop or do you continue to see if things fall apart? This applies to all aspects of handloading...
My latest testing has been looking for low ES... consistent low ES. Do you shoot 3 or 5 shots and plug that into your ballistic calculator and never test again or do you check your ES with every shot? I have found that consistent low ES is elusive (including different days, different temperatures, different humidities, moon phase, star alignment).
Anyone else seeing this in the real world?
Thanks!

 

[nwmnbowhunter]

It's either getting better or getting worse!!! I turn away from worse.

 

[CF Junkie]

I test for accuracy but I load for a seating depth, trim length and powder charge that will yield a velocity and Optimum Bullet Time (exit time) for a particular powder.
MV changes based upon variations in powder charge around 12 fps per 0.1 grain change, seating depth and trim length about 1 fps per 0.001 change.
Even my very best hand loads (all of them loaded with near identical seating depths and trim lengths (within 0.001) and powder charges (within 0.1 grains) have a SDs of 5 to 7 fps.
Those numbers are not generated from ladders, the velocities were expected to all be the same.
Factory rounds have SDs around 15 fps at best for supposedly the same loads for each round.
ES for those SDs could be about 3 to 4 times those numbers.

A lot of the variation in velocity is a product of the natural variation in primer and powder ignition.

The worst temperature sensitive powders make a difference of 1.24 fps per deg F. change.
The best temperature insensitive powders make a difference of less than 0.1 fps per deg F change.
How much temperature change do you think you experience between the first shot and the last shot in each of your muzzle velocity tests.

It is really tough to make any conclusions without an enormous amount of test data.

 

[STS]

I test to learn. I can learn from a positive result, or a negative result.

 

[JEFFPPC]

Only one test counts. Your target. I test to evaluate what I am doing. Target shows the only data that counts. My hobby is precision shooting not spreadsheets.

 

[222Jim]

Do you look for passing data and stop or do you continue to see if things fall apart? This applies to all aspects of handloading...
My latest testing has been looking for low ES... consistent low ES. Do you shoot 3 or 5 shots and plug that into your ballistic calculator and never test again or do you check your ES with every shot? I have found that consistent low ES is elusive (including different days, different temperatures, different humidities, moon phase, star alignment).
Anyone else seeing this in the real world?
Thanks!

It depends.....
(1) For load development, I test to learn through statistically valid samples. That means 4 5-shot groups (including chrono).

(2) To confirm my scope zero hasn't shifted due to getting bumped or banged around, 3 to 5 shots (but I use the 4 5-shot groups to set zero).

 

[Doom]

My latest testing has been looking for low ES... consistent low ES. Do you shoot 3 or 5 shots and plug that into your ballistic calculator and never test again or do you check your ES with every shot? I have found that consistent low ES is elusive (including different days, different temperatures, different humidities, moon phase, star alignment

ES is not a statistic. It is descriptive of two rounds out of any sample or population that is the result of extreme statistical standard deviation or to possible errors in the loading process (or any other proscess). For large sample sizes it will likely lie somewhere around 6x the standard deviation of the true standard deviation of the population. Note that that is about the extreme of the 99.7% of population.

ES at small sample sizes it is less and in general grows as the sample size increases. Just how would you define low ES or consisten ES. Standard deviation is a statistic. It describes the amount of variation in the individual samples in a test from the mean (average) of the samples. Because of random probabilities associated with sampling small sample sizes usually result in lower standard deviations than larger sample sizes. This is the natural result of 68% of the population lies with 1 standard deviation. That leaves 32% lying beyond 1 standard deviation. that means there is a more than 2 to 1 chance you pick a 1 standard deviation round each time you shoot or load a round. Unlike the mean or average value which tends to the normal distribution or Student's T distribution the Standard Deviation follows a Chi-Squared distribution which is heavily weighted to the low side for low sample numbers. For instance a test that yields a 9 fps standard deviation will have a 95% confidence interval between 5.39 and 25.86 fps. meaning that if a large number of test were run 95% of the SD's would fall between those numbers. 5% would be outside.

 

[old_heli_logger]

Wow! Great answers, thank you very much everyone!!!

 

[357Mag]

Do you look for passing data and stop or do you continue to see if things fall apart? This applies to all aspects of handloading...
My latest testing has been looking for low ES... consistent low ES. Do you shoot 3 or 5 shots and plug that into your ballistic calculator and never test again or do you check your ES with every shot? I have found that consistent low ES is elusive (including different days, different temperatures, different humidities, moon phase, star alignment).
Anyone else seeing this in the real world?
Thanks!

Oldhelllogger -

Howdy !

There are many reloaders/shooters today, that do load refinement w/o constant use of a chronograph and/or functions of a ( digital ) "ballistics calculator" like you mentioned.
These could be shooters like some varminters, hunters, and others that ( typically ) are not full tilt competition shooters.

When I'm doing load refinement range work, I myself establish what the max allowable charge is for a select powder, shot under a select bullet wt: and for use w/ a particular primer. * I stop increasing charge weight, before there is component or gun damage.
Keep load development safe, and not frenetic.

I thereafter reduce test power charges incrementally downward, until best accuracy is shown/proven. In varmint calibres, I have found the accuracy load' powder charge wt to be < 1 gr bellow the established max charge wt. Oft times, accuracy load' charge wt has ended up being found within just a few tenths of a grain of the max pressure load's charge wt. After reaching that point, I may make incidental slight reduction in charge wt, IF/when I take the gun to some incidental competition; and weather makes such adjustment necessary. But hey, that's just me.

I don't find my load development approach to be all that uniformed. After all, I have found a load that operates w/ safe pressures, and that provides best accuracy.
In that sense.... obtained muzzle velocity " is what it is ". After reaching THIS point, use of a chronograph and ( example ) on-line ballistics calculator comes into play; to allow generation of a bullet " drop chart " based on the chosen load's Mv. For me, a load's extreme spread and standard deviation are things I don't obsess on. Sometimes, it's sort of a " don't ask.... don't tell " type of deal; as regards ES and SD. The target shows you what works.

YRMV.


With regards,
357Mag

 

[Doom]

Oldhelllogger -

Howdy !

There are many reloaders/shooters today, that do load refinement w/o constant use of a chronograph and/or functions of a ( digital ) "ballistics calculator" like you mentioned.
These could be shooters like some varminters, hunters, and others that ( typically ) are not full tilt competition shooters.

When I'm doing load refinement range work, I myself establish what the max allowable charge is for a select powder, shot under a select bullet wt: and for use w/ a particular primer. * I stop increasing charge weight, before there is component or gun damage.
Keep load development safe, and not frenetic.

I thereafter reduce test power charges incrementally downward, until best accuracy is shown/proven. In varmint calibres, I have found the accuracy load' powder charge wt to be < 1 gr bellow the established max charge wt. Oft times, accuracy load' charge wt has ended up being found within just a few tenths of a grain of the max pressure load's charge wt. After reaching that point, I may make incidental slight reduction in charge wt, IF/when I take the gun to some incidental competition; and weather makes such adjustment necessary. But hey, that's just me.

I don't find my load development approach to be all that uniformed. After all, I have found a load that operates w/ safe pressures, and that provides best accuracy.
In that sense.... obtained muzzle velocity " is what it is ". After reaching THIS point, use of a chronograph and ( example ) on-line ballistics calculator comes into play; to allow generation of a bullet " drop chart " based on the chosen load's Mv. For me, a load's extreme spread and standard deviation are things I don't obsess on. Sometimes, it's sort of a " don't ask.... don't tell " type of deal; as regards ES and SD. The target shows you what works.

YRMV.


With regards,
357Mag

Great reply! What I didn't say in my epistle on ES and SD is exactly what you said. Find a load on target that works, be it at 100, 300, 600yds. If it works at short range and you are using good quality known components it's likely a good load. IF it opens up at distance then the chronograph will likely show you the issue in SD. But not with 3 or 5 rounds but 20 or 30 at least.

Two many people assume that the powder charge is the source of low round count SD variations for small changes. The truth is that it is likely attributable to all of the other variables that are changing shot to shot including Bullet, case primer, barrel condition, and minor powder weight variances.