Sorting cases by weight ??

[ireload2]

A PHD materials engineer I knew once worked on a problem like that. He wound up with an 11X11 matrix for his designed experiment. He could not solve the partial differential equation that resulted. No one in the company had that background so they hired a PHD mathematician who took about a month to solve it. Once solved, the top 4 variables in the system were adjusted to permit sintering a cobalt chrome alloy to a titanium alloy. Then consensus of industry was the process was not possible. The result was if you spent enough time, money and brain power you could make it work. The process was so valuable that it was never patented. It remained a secret proprietary process known only to about 4 people.

Yep, at the heart of the issue is the problem of stacked variability. If you have a variable (case volume) that in isolation could cause a deviation of 3-4 fps, and you add that variability to a broader data set that already has 8-9 fps of deviation from all other factors (primer, powder, bullet, neck tension, etc), you don't get a resulting SD of 11-13 fps. You only see an increase in SD of maybe 0.5 fps. That's just the fundamental math behind stacked variables.

You'd need a very large data set and very methodical procedures to demonstrate that 0.5fps SD improvement, and in the real world that just isn't readily observable or measurable.

 

[Sheldon N]

Here's what an actual scatter plot of precision measured case weight against resulting case velocity looks like.

 

[Das Capitolin]

I'm thinking that it would be a more accurate operation to standardize outside case size by "fire forming" the "prepped cases" and trim them to common OAL first, before undertaking the water weight tedium?

Dan

Most people consider fire-forming, sizing, and trimming necks to be part of case prep, so my suggestion is that you do all these things prior to measuring volume in water weight.

 

[Sheldon N]

This is a scatter plot showing the resulting velocity from volume measured cases. Cases were fire formed, sized, trimmed and then H20 volume measured on an an analytical balance capable of .02g resolution.

 

[Pumba]

This is a scatter plot showing the resulting velocity from volume measured cases. Cases were fire formed, sized, trimmed and then H20 volume measured on an an analytical balance capable of .02g resolution.

Thank you,

My findings on weight vs velocity correlate your findings.

Your post regarding volume vs velocity save me a great deal of time and testing.

I could never get myself to go trough the trouble of measuring volume and testing it against velocity.

There are more important things to consider when it comes to low ES

 

[waldo1979]

Here's what an actual scatter plot of precision measured case weight against resulting case velocity looks like.

Nice. How did you generate that trend line? Is it a linear regression or an excel charting feature?

I think people are better off sorting their primers by weight, but I don't personally have any hard data to prove that.

I've thought about applying a bar or QR code to brass to track them. You could scan them and log data for each phase of your reloading/firing process (sort of like engineering QC). But more importantly you can isolate pieces of brass that lead to flyers and set them aside for sighters. The bar/QR codes could be accomplished with a laser engraver (albeit a beefy one).

 

[Sheldon N]

Nice. How did you generate that trend line? Is it a linear regression or an excel charting feature?

It's an excel charting feature. Once you have a scatter plot, you can right click it and choose to add a trendline.

I've thought about applying a bar or QR code to brass to track them. You could scan them and log data for each phase of your reloading/firing process (sort of like engineering QC). But more importantly you can isolate pieces of brass that lead to flyers and set them aside for sighters. The bar/QR codes could be accomplished with a laser engraver (albeit a beefy one).

I did a little bit of looking into that, basically trying to find out if a case that produced a "flier" velocity would be likely to produce the same velocity deviation from the average again. Also looking to see if velocity from firing was a predictor of velocity for firing the subsequent time, which would let you "velocity sort" cases. Unfortunately I wasn't able to find any correlation, although the data was not complete enough for that to be a firm conclusion.

 

[dmoran]

Sheldon -
How can accurate qualification or correlation be determined from such large ES?
In my opinion, one needs a rifle-n-load capable of low ES, to be able to accurately pin-point culprits.
Then if there is a culprit extreme seen both in the velocity as well as on the target, it would truly be a culprit.
Myself, I consider ES much over 15 to 20-fps to be excessive.

Thanks for sharing....
Donovan

 

[Sheldon N]

Sheldon -
How can accurate qualification or correlation be determined from such large ES?
In my opinion, one needs a rifle-n-load capable of low ES, to be able to accurately pin-point culprits. Then if there is a culprit extreme seen both in the velocity as well as on the target, it would truly be a culprit.
Myself, I consider ES much over 15 to 20-fps to be excessive.

Thanks for sharing....
Donovan

Yes, that's one of the unfortunate flaws of the data available. While the loading practices were generally sound, the test load just wasn't in a good node and miscellaneous factors conspired to give an SD and ES that were only average at best.

And I agree that when you're looking for a small variable that it's even more tough to find it in data that has a lot of "noise". If the case volume or brass weight has a measurable effect, the only place you're likely to be able to observe and document it is in a load that already has very low ES and SD. That's supported not only by common sense, but also by the math underlying stacked variables.

If I had a counterpoint to the data above it would be to point out that while it's hard to prove the impact of doing just one thing, if you are able to reduce the impact of ALL of your variable factors through your reloading practices, that can make a difference.

 

[Lenard]

Most of you have far more experience than I with weighing and sorting cases. Having progressed from Rem or Win to LC cases, I have discovered that the variation in velocity has reduced to the 35-40 level. What is the best news of all of this, is that group sizes are smaller than previously. Now mind you this is not benchrest tiny groups, but under 1/4" and down into the .100"s. I do occasionally get a flier which can be more capacity or less, which I pitch immediately. All cases are weighed into 1 grain segments.

I am shooting Rock Chucks where sometimes, all we can see is a head at 200 plus yards. So I attempt to load ammo that will make some of those shots possible

 

[Just Dave]

After full match prep.. Annealed, sized, trimmed, necks turned and any donuts removed..
I weight sort bullets and brass, and match same-weight bullets with same-weight brass in large batches.
Along with all of the steps above, I am chasing any doubts out of my head that I did everything reasonable to produce IDENTICAL rounds..
Often, Im driving over 1000 miles one-way to compete, spending hundreds of dollars in food and lodging, not to mention entry fees
I will spend the extra time, however foolish it is deemed, to try and silence those gremlins.
Because when they announce to display 5 rounds only for recorded score, I want EVERY one of them to be as physically identical as possible,, the rest of the problems are with the nut behind the trigger.....

 

[Hengehold]

Here's what an actual scatter plot of precision measured case weight against resulting case velocity looks like.

Thanks for sharing. Great graph.

 

[Chrisjan]

I have always sorted my brass, it is only a few minutes to do.

I have tested my heaviest vs lightest brass at 1000m on my Dasher. It was both five shot groups. Both groups were around 5" in size. But they were close to 10" appart in elivation. This was on a weight sort spread of 1.2gr between lightest and heaviest.

I will for ever sort my cases. Even if it just gives me peace of mind.

 

[fguffey]

A reloader that gets to the point they have so much experience reloading they assume nothing can go wrong; they better hope nothing goes wrong.

Again; I was at a range to shoot 45ACP ammo reloaded by a local reloader. He suggested I did not know how to load for 2 of my 45 ACPs. Both pistols like factory and surplus ammo but did not like cases with bullet lines. Before we could get started another reloader was doing everything he could to pull the trigger on his 66 S&W 357. He could not pull the trigger, he could not rotate the cylinder, he could not swing the cylinder out nor could he pull the hammer back. We explained to him one of his reloades did not have powder; when he pulled the trigger the primer drove the case into the forcing cone and that locked his pistol. We drove the bullet back into the case and cleared his cylinder. He then immediately started loading his cylinder; we suggested it was a bad habit to continuing shooting; we suggested if he did not know one case had no powder how was he going to know if the next case had a double charge.

We offered him all the ammo he could shoot, we offered to check his ammo, and we offered to help him with his reloading. All we managed to do was make him mad, he left.

I suggested he know the weight of his components; after loading a quick check before firing would be to weight each round, I offered to loan him a scale etc.. Problem; if he did not sort his cases by weight he would never know if the difference in gross eight was caused by the weight of the case or powder. I suggested he could ignore the weight of the case after he formed good habits when reloading.

The one thing he made very clear to us was he had an Rl550B Dillon, I explained to him I also had the same press but I will not use it woth straight wall cases without a lock out die.

My 45ACPs did not like the reloaders reloaded ammo. His ammo flew through his 4 45ACPs.

To remove the bullet line in the case I ran the ammo half way through a RCBS carbide sizer die. The carbide sizer die returned the cases to the same diameter as factory and surplus ammo.

F. Guffey

 

[fguffey]

One reason case volume may not correlate well with case weight is because of variation in the extractor groove.

No one measure case head thickness; I have 30/06 cases with a case head thickness of .200”, I also have 30/06 cases with a case head thickness of .260”. The .200” is surplus; the .260” is R-P. The assumption is military surplus is thicker because it is heavier; that can be a half truth if the case head is thinner and or thinner.

Then there is the assumption volume is volume; if the case head is thinner and the case is heavier the case body is thicker/heavier.

Then someone has to decide if the powder column measurements make a different. The powder column in the surplus case will be longer and smaller in diameter than the powder column in the R-P case. The powder column measurements makes a difference in WSM and WSSM cases.

F. Guffey

 

[Dcaven]

I have always sorted my brass, it is only a few minutes to do.

I have tested my heaviest vs lightest brass at 1000m on my Dasher. It was both five shot groups. Both groups were around 5" in size. But they were close to 10" appart in elivation. This was on a weight sort spread of 1.2gr between lightest and heaviest.

I will for ever sort my cases. Even if it just gives me peace of mind.

How much variation do you accept in each group of casings?

 

[Chrisjan]

How much variation do you accept in each group of casings?

I do not group my brass. I shoot from low to high through all my cases. This means that my following shot will always be in the same weight class as the previous.

 

[Endyo]

Sheldon -
How can accurate qualification or correlation be determined from such large ES?
In my opinion, one needs a rifle-n-load capable of low ES, to be able to accurately pin-point culprits. Then if there is a culprit extreme seen both in the velocity as well as on the target, it would truly be a culprit.
Myself, I consider ES much over 15 to 20-fps to be excessive.

Thanks for sharing....
Donovan

This was the very reason why I decided not to publish this data. There were too many variables that did not appear to be under control. I did not feel that we could definitively answer YES or NO to weight sorting either volume or by brass weight and many others were far more qualified to answer that question.

I can't blame Sheldon for sharing the charts as he put in a lot of work processing the raw data, graphing, and providing input into the tests.

I plan on performing a future test when I'm capable of handloading to single digit ES/SD and share results with the community.

I have decided for myself weight sorting by brass weight or volume is not worth the time or effort. Neck turning provided the largest benefit in reducing ES/SD in our testing.

 

[Webster]

This was the very reason why I decided not to publish this data. There were too many variables that did not appear to be under control. I did not feel that we could definitively answer YES or NO to weight sorting either volume or by brass weight and many others were far more qualified to answer that question.

I can't blame Sheldon for sharing the charts as he put in a lot of work processing the raw data, graphing, and providing input into the tests.

I plan on performing a future test when I'm capable of handloading to single digit ES/SD and share results with the community.

I have decided for myself weight sorting by brass weight or volume is not worth the time or effort. Neck turning provided the largest benefit in reducing ES/SD in our testing.

If you look around on this website you will find at least one example of people setting 1000 yrd. records without weighing , measuring volume or annealing cases. I am not saying it isn't usefull. I understand you are only talking about making the best ammo but I think the biggest variable is the person doing the shooting. How do you put his/her skills into a formula. If there are 150 competitors at a match you can put together a short list of about 15 shooters that will probably be in the top five regardless of the distance. The rest never make the top 10. In Tony Boyers book he said that 90% of the shooters have a rifle that's good enough to win but they usually don't finish well.

 

[Endyo]

If you look around on this website you will find at least one example of people setting 1000 yrd. records without weighing , measuring volume or annealing cases. I am not saying it isn't usefull. I understand you are only talking about making the best ammo but I think the biggest variable is the person doing the shooting. How do you put his/her skills into a formula. If there are 150 competitors at a match you can put together a short list of about 15 shooters that will probably be in the top five regardless of the distance. The rest never make the top 10. In Tony Boyers book he said that 90% of the shooters have a rifle that's good enough to win but they usually don't finish well.

Well spoken. I was unsure how well received findings would be that were contrary to common knowledge. What's interesting is many in this thread are voicing similar results.

Science says that there should be a difference between a piece of brass with less internal volume than one with more. How much that difference is or makes is I think what matters when we determine if it's worth the amount of time and effort.

I think we agree there are much bigger factors that are worth time and effort!