Neck tension and ES

[thelaststraw]

I've recently made the switch to temp stable extruded powders. In the past I have only ran 5 rounds or so over the chrono and get pretty decent numbers...single digit SDs and teens for the ES. I started annealing after I shot through my first lot of brass and it made a very noticeable difference in seating pressure (not measured, just by feel). I decided to run 20 across the Magnetospeed both with and without my can to get better info on both velocities.

I wet tumble with pins (about 30 min) and I'm pretty sure it dings up my necks enough to where I have to do a light chamfer/deburr. If I don't chamfer good enough and it leaves a little lip, would that be adding enough bite to alter my neck tension? I was getting 20 ish SD and 60+ ES over 20 rounds. When seating the bullets, about 5 or so had a touch of resistance at first and then dropped in like normal.

I full length size all the way to SAAMI since I have reliability issues when I only bump 7K. Chamber would get dirty after about 100 rounds and bolt wouldn't fully lock up. I set neck tension with a .241 mandrel for a 6ARC gasser. My powder is dropped from an RCBS Lite and I actually have double checked them on a Dillon beam since I just got it and want to vet it for trust.

 

[kthomas]

How much neck "tension" (it's a misnomer) are you using?

I've found that inconsistent neck tension can absolutely effective the consistency in velocity of a load. I once had an incident where I had inconsistent neck tension (I believe I over-annealed some brass, was using the wrong pilot on an AMP annealer), and my SD and ES went to shit.

I'm not sure how much of an effect dings from wet tumbling may have. I don't tumble with that method, no experience with it or any perspective on how bad the "dings" may be.

I think the annealing process could have something to do with it. How do you anneal?

 

[thelaststraw]

I'm sorry, I tried to include as much pertinent info I missed that. 0.002 neck tension. 6mm with .241 mandrel. Annealing is done with an Annealez and propane. I'd love an AMP but I'm not deep enough in yet to justify. I'm barely a consistent sub MOA shooter at the moment.

 

[Feniks Technologies]

Keep in mind, spring back happens the opposite way you sized.

When you use a mandrel, the brass springs back inside.

So, using a .241 mandrel will almost never yield a .241 inside diameter. It will likely be .240 or .2405.

That’s likely not an issue, just understand using the same size mandrel you want for the diameter likely won’t size it to that diameter consistently.

Also, the spring back will be different depending on how much the bushing or non bushing die sizes the neck below your intended inside diameter.

For example, if you size neck down to .238 and use a .241 mandrel, you will likely get a different final inside diameter than sizing down to .240 and using a .241 mandrel.

I’ll make another post here regarding the other factors besides inside neck diameter/interference fit that make up the total sum of “neck tension.”

 

[Feniks Technologies]

Wet tumbling will always peen the necks. Even if you don’t use pins. The brass rolling around hits the necks.

So, if you wet time you will almost always have to chamfer again as the last chamfer will be peened smooth.

The inside finish of the neck matters for friction. A perfectly clean neck will have more friction than a polished or lubed neck. There’s also the work hardening that affect spring back which annealing changes. And there’s also the thickness of the neck which plays a part in elasticity.

But, for simplicity sake, let’s look at interference fit and the condition of the inside of the neck.

Less interference fit = less resistance to bullet seating

Lubed and/or polished = leas friction and thus less resistance.

And the two are on a sliding scales of such.

If you decrease the friction via polish or lube, you *may* need to increase your interference fit to find a happy spot.


Typically when “neck tension” is the culprit of high SD, that means your totality of neck tension (interference fit, lube, state of brass hardness, neck wall thickness, etc etc) is not consistent.

So, it’s releasing the bullet in a more inconsistent time during the ignition process. More neck tension = more resistance thus more pressing and higher velocity. Less neck tension is the opposite.

So, you likely need to experiment with the sliding scale of neck tension to find the combination that yields the most consistency.

 

[Feniks Technologies]

The above post assumes it’s a neck tension consistency issue.

The other possibility is you don’t have enough total neck tension to prohibit the bullet from moving during chambering or handling of the rifle.

ARs are known for dislodging the bullet. So, take measurements of the OAL or BTO, chamber the round. Remove it and check it again to see if it moved.

Measure some, put them in magazine and beat it around a little. Measure again.

In either of those scenarios, if the bullet is dislodging, you’ll need to increase the interference fit or friction.

 

[thelaststraw]

I appreciate all the information and you taking the time. I don't use bushing dies currently, just a full length Hornady sizing die. Most all the carbon stays inside my cases, I never tumbled really long. I guess I should break down and get a dust bowl vibratory.

I was aware of spring back and actually worried about not getting enough once I started annealing. I'll check a few rounds again, I have not had issues in the past when I measured. I measured rounds that the bolt didn't lock up thinking they were hard jamming lands but it ended up being a head space issue with carbon from the can.

Once I went to a cam over, full length size I had no problems as suggested by Lowlight at a recent class. The only thing I've changed recently was the sample size on the Chrono. Instead of 5-8 rounds I ran 20. Does that just mean I'm getting a more accurate number of my load's performance?

 

[Feniks Technologies]

I appreciate all the information and you taking the time. I don't use bushing dies currently, just a full length Hornady sizing die. Most all the carbon stays inside my cases, I never tumbled really long. I guess I should break down and get a dust bowl vibratory.

I was aware of spring back and actually worried about not getting enough once I started annealing. I'll check a few rounds again, I have not had issues in the past when I measured. I measured rounds that the bolt didn't lock up thinking they were hard jamming lands but it ended up being a head space issue with carbon from the can.

Once I went to a cam over, full length size I had no problems as suggested by Lowlight at a recent class. The only thing I've changed recently was the sample size on the Chrono. Instead of 5-8 rounds I ran 20. Does that just mean I'm getting a more accurate number of my load's performance?

Yes. You could have definitely been seeing statistical variance in your earlier tests and now a more accurate representation of your ammo.

 

[thelaststraw]

I guess the moral of the story is I have a lot of learning to do. I thought I was doing good with ES in the teens and single digit SD. I wanted to verify with a larger lot. My brass prep game needs to improve. Thank you again sir.

 

[Feniks Technologies]

I guess the moral of the story is I have a lot of learning to do. I thought I was doing good with ES in the teens and single digit SD. I wanted to verify with a larger lot. My brass prep game needs to improve. Thank you again sir.

No prob. Reach out anytime if you need some help.

 

[BuildingConceptsllc]

I appreciate all the information and you taking the time. I don't use bushing dies currently, just a full length Hornady sizing die. Most all the carbon stays inside my cases, I never tumbled really long. I guess I should break down and get a dust bowl vibratory.

I was aware of spring back and actually worried about not getting enough once I started annealing. I'll check a few rounds again, I have not had issues in the past when I measured. I measured rounds that the bolt didn't lock up thinking they were hard jamming lands but it ended up being a head space issue with carbon from the can.

Once I went to a cam over, full length size I had no problems as suggested by Lowlight at a recent class. The only thing I've changed recently was the sample size on the Chrono. Instead of 5-8 rounds I ran 20. Does that just mean I'm getting a more accurate number of my load's performance?

Yeah I started doing much longer strings for that reason.

I just ordered a redding bushing die set to test this out and see if it will improve my SD and ES.

 

[kthomas]

I guess the moral of the story is I have a lot of learning to do. I thought I was doing good with ES in the teens and single digit SD. I wanted to verify with a larger lot. My brass prep game needs to improve. Thank you again sir.

People put too much stock in small sample sizes. What you see in 1,3 or 5 shots may not be indicative of what your reloads will do over 10, 20, 30+ shots.

I've personally gotten burned on small sample sizes a few times. The first was with Nosler RDF bullets, testing bullet depth seating with 3 shot groups. Shot some tiny groups. Loaded up more ammo with my promising bullet seating depth, and starting shooting 5 round groups - lots of fliers. In fact, couldn't manage a single 5-round group without a flier.

The second time I got burned was on the "Satterlee" load development method. If you are not familiar with that, it's loading up ~10 rounds or so of progressive charge weights, looking for a so-called "velocity node" in which the velocity seems to stay consistent over a couple of different powder ranges. Seemed quick, and based off of who was promoting it, supposedly effective. Perfect for someone who wants to spend less time reloading. Well, eventually I started to load up two identical ladders, because more data is always better. I would shoot these ladders back to back, and guess what? That so called "node" is almost never repeatable. It's because they don't exist, they are an artifact of small sample sizes.

Personally, I find the best way to produce ammo is by using good quality components (lapua brass, berger bullets), quality reloading equipment (I would argue a proper powder trickler and scale is the best gear to splurge on), and using a repeatable reloading procedure every time. Keep things simple and consistent.

 

[thelaststraw]

Numbers are numbers to me. If I'm getting good results downrange I'm happy. This just put a wrench in my confidence gears...lol

 

[Doom]

Let the target give you information. Way too much emphasis is put on too little data.

If you were to analyze your 5 round standard deviation and compare it to your 20 round standard deviation using typical analysis of. Variance techniques you would likely find that they are not statistically different.

As for the amount of data a 5 shot test single digit standard deviation of 7 would yield a 95% confidence that the true standard deviation of a large population would be between 4.2 and 20.1 fps. However, a 20 shot sample with a SD of 14 would yield 95% confidence that the true standard deviation of a large population would be between 10.6 and 20.4 fps.

 

[thelaststraw]

Wow, that's pretty detailed. Had to read that a few times.

So as sample size goes up the spread gets smaller vs small sample size with a wider potential spread? So a 40 shot group would be about around 5-10 fps spread?

Thank you.

 

[acudaowner]

after trying a 3 thou neck tension and a .5 thou neck tension I my self wonder why people add the extra steps as I can get just as good results either way it's almost the same with length short or long . but as I have already paid for the mandrels I am going to use them till they are no longer any good . the powder shape gives me the biggest question load 20 round of the same load and almost every one will be different is it the way the powder ignites or how it shape allowes timed release of burning that makes the biggest difference . I tried to see if better primers helped but even bench rest primers I never got the es or sd numbers lower than a 3.9 and maybe an 8 so cheap primers did the exact same thing just the br primers had a larger price tag while and were much harder to get while the standard primers ( federal or Winchester ) both were cheaper sadly just as hard to find now a days provided the same results as the cci bench rest lrp or the match versions of the federal or Winchester cases also make me wonder a lot while it's nice to have good cases I am still shooting my factory hornady cases 4 th time through and get just as good or bad results depending more on other factors than the cases than I have with both my Peterson brass or my new alpha brass and it was even annealed before I got it . while It's definatly the prettiest of my brass I could not tell you one from the other just bases on anything the brass did that or I am missing it . except one cost 105 per 100 + shipping for the alpha , and the other way 134. per 200 rounds loaded hornady factory ammo at the pime of purchase , and the peterson I think was 184 per 100 at the time of purchase maybe a little more from prime . I even tried for a little while to weigh 100 pieces of the different brasses but got way to board to keep doing it but they all held about the same amount or so it seemed . maybe someone else that knows more could tell I could not .

 

[Feniks Technologies]

Let the target give you information. Way too much emphasis is put on too little data.

If you were to analyze your 5 round standard deviation and compare it to your 20 round standard deviation using typical analysis of. Variance techniques you would likely find that they are not statistically different.

As for the amount of data a 5 shot test single digit standard deviation of 7 would yield a 95% confidence that the true standard deviation of a large population would be between 4.2 and 20.1 fps. However, a 20 shot sample with a SD of 14 would yield 95% confidence that the true standard deviation of a large population would be between 10.6 and 20.4 fps.

The example you just provided has a ~40% difference……

Also, “letting the target give you information” requires larger sample sizes as well.


You can try all you want, you can’t outrun the numbers.

 

[Doom]

a 40 shot sample with a SD of 14 would yield 95% confidence that the true standard deviation of a large population would be between 11.5 and 18 fps.

In statistics extreme spread means little to nothing. But that number will increase as the number of shots increase, but those two numbers have a decreasing effect on SD. In reality, The two ES values should be >+/-3xSD of the mean (average).

 

[thelaststraw]

"In statistics extreme spread means little to nothing."

So standard deviation is a better representation of velocity consistency? And how big of a sample size would be required to get a moderately accurate representation of a lot of ammo.

For example, Chrono 20 rounds of a 250 round lot of ammo loaded all at the same time to be used for a match or a class. I understand everyone's requirements vary, just a ballpark area here I'm asking so I can understand a little better.

 

[Doom]

The example you just provided has a ~40% difference……

Also, “letting the target give you information” requires larger sample sizes as well.


You can try all you want, you can’t outrun the numbers.

I’m sorry but could you explain your 40% reference?

I didn’t say anything about the number of shots on target. Obviously, we should be evaluating every shot we take.

For anyone interested in trying to understand statistics and probability as it applies to shooting there is a series of 3 articles on the Precision Rifle Blog that was published starting last December that is well written and and covers the subject with examples.

 

[thelaststraw]

Thanks I'll check the articles out. I have a lot of work to do still as the shooter. Any variables I can eliminate can only help on the firing line.

 

[Doom]

"In statistics extreme spread means little to nothing."

So standard deviation is a better representation of velocity consistency? And how big of a sample size would be required to get a moderately accurate representation of a lot of ammo.

For example, Chrono 20 rounds of a 250 round lot of ammo loaded all at the same time to be used for a match or a class. I understand everyone's requirements vary, just a ballpark area here I'm asking so I can understand a little better.

There is no exact number. Since we are dealing in probably, it depends on the level of confidence we want or need. If you shoot 60 rounds any more is not like to improve your confidence interval. Practically 20 to 30 is a good goal but may be impractical. If I were to pick a minimum number it would be 10 assuming I was comfortable with the confidence interval it would provide.

I have a post below that you might find of interest.

You are being redirected...

 

[thelaststraw]

Thanks, I'll check that out also. Should keep me busy for a while.