Concentricity Effects of Sizing Dies

[GermanS1]

A few days ago I began to write a response to a post about concentricity and resizing dies. My post (which I ended up not posting) made me ask myself a few questions and soon became a big project. Now it's all written down and posted here:

http://riflemansjournal.blogspot.com/2010/04/reloading-two-step-sizing-and.html

I hope it is of interest or value to some of you.

 

[SS427]

Thanks for the link. For someone still learning, I found it of interest and value.

 

[BoydAllen]

German,
I always enjoy your articles. I have played with two step sizing quite a bit, and was therefore not surprised by your results, although I do remember being somewhat puzzled the first time that discovered it for myself. I have a couple of other tests for you to try. Instead of using a bushing neck sizer, on unturned necks, try a Lee Collet die. I know that they are more crudely made, and not nearly as pretty, but humor me. Heck I'll even send you one to play with.
The other option is more expensive. Try a custom one piece die that is made with a neck ID that only has the expander doing maybe a couple of thousandths of work. I am sure you have probably figured out the the reason that the common sort of one piece dies induce runout is that their necks are so small inside that the amount of pull when the expander ball is pulled through exceeds the yield strength of the shoulder, and it gives a little, unevenly, cocking the neck. The point is that with only a minor mismatch in diameters, there isn't enough pull to cause the brass to give, or the neck to become more crooked. If you have some old brass, and you just want to do a test of sized brass straightness, pull the whole decapping assembly out of a standard FL die and size a few cases, and check the runout. You can even neck turn a couple and then size them. I have a 6PPC with a .262 diameter neck and the regular, cheapest, one piece Hornady FL die has a neck diameter that is about right for proper neck tension without an expander, and if you use the expander only a little drag occurs. Runout on the end of the case necks is less than a half thousandth. in my experience. nothing with a bushing can do that well. Another interesting test is to use a bushing FL die like a body die, by removing the bushing. When I do that with my PPC, the cases, that are only slightly sized and bumped, are about as straight as they come out of the chamber after firing.

 

[GermanS1]

Thanks, Boyd, I appreciate the kind thoughts.

I like doing these projects and writing them up because it forces me to think about many of the small details that we otherwise take for granted. Your post does exactly that for me as well. I'm going to have to think about what you wrote and what to do next.

The main thought in my mind, especially with respect to unturned brass is that we're always fighting to get something aligned with the case body axis: either the outside diameter of the neck (with a bushing die) or the inside diameter of the neck (with an expander type die). Since the ID and OD of an unturned neck are not concentric, we can't have it all.

If we were really being smart about it, we would say that the ID is what matters most since we want to align the bullet with the throat. Unfortunately, it's easier to measure the OD's concentricity to the case axis so that's what we focus on. We might be barking up the wrong tree, don't you think?

Of course, once we turn the necks, we have mostly eliminated that problem and eventually, most precision reloaders go that route.

I think what I'm going to do is to fire that unturned Norma brass again, then size it with a final bushing of 0.330" and an expander in the die, that will give about 0.001" pull-through tension on the expander. Just a little bit, as you suggest.

I agree that a bushing die has an inherent concentricity problem, after all, the cavity for the bushing has to have some degree of clearance for the bushing to enter and exit and we can't alwways count on the bushing being drilled/bored dead center. Then there's the question of the cavity alignment with the body portion of the die since those cuts are made from opposite ends of the die. Actually, it's a wonder they work as well as they do!

I think I'm going to crack open a fresh bottle of Diet Dr. Pepper and give this some more thought. Thanks!

 

[BoydAllen]

One more thing...
Evaluating unturned brass should probably be done on a seated bullet, just in front of the neck. While not perfect, it will get you closer to seeing the alignment of the ID, and BTs are less seater/neck tension sensitive (easier to seat straight). With FB bullets, the more neck tension, the more critical the seater selection.

 

[Rust]

I occasionally think of custom spec'ing two reamers, one to chamber the barrel and one to ream a sizing die. The die would just size the brass enough for the chamber and no more.

One of my current rifles was chambered with a custom spec reamer that a Redding FL die came pretty close to the same concept. But resizes the neck just a touch more than absolutely required.

Of course with such a system work hardening of the brass would eventually require annealing due to more spring back with work hardened brass, which would change bullet pull which could affect accuracy by way of less repeatability.

Carried to an extreme, it would be like the chambers cut in some short range bench rest rifles which are so tight that resizing isn't needed. Of course it requires brass prep like turning the necks to within a couple of ten thousandths because loaded round clearance at the neck can typically be a single thousandth. A little more fuss than I care to deal with.

One thing I've not done yet is hone the neck of a FL sizing die, simply because I have some concerns over maintaining cocentricity. Which is why I like the thought of a reamer specifically for a die, if the reamer is good, cocentricity is assured.

 

[camac]

Great topic guys. I am new to the forums group but have played a fair bit with making my own dies and neck bushings over the years in an effort to achieve as close to zero runout as possible (call it an OCD obsession).
I tried desparately for a long time to try and make my own neck bushings with much less tolerance than off the shelf bushings (Most have several thou play for "self alignment"). I tried to reduce free play to under 1 thou and ensure bore of bushing was concentric with outer. This was all very good except then you have to ensure the shell is a perfect fit in the collet as well. Few are unless you use the same chamber reamer and create your own collet with it. (or for a full length die with 2 thou undersize reamer like you suggested Rust). This is all a bit much for the average reloader.

A neat, simple trick which ended up working just as great was to stick the bushing in a lathe (hopefully precision one with little runout but not as critical as it might sound) and use a die grinder to do a slow taper into the bush on one end only (That way you can still use the other end of the bushing like normal). Finish with some 600 and 1200 wet and dry. I take the start of the taper out to the fired neck size of my chamber so that it is a neat lead in. If you have the taper length about half the neck length then the "self alignment" of neck bushing and shell works almost perfectly. You can then do a one or two step neck size (one if you are happy with only half the neck sized). I personally do a two step neck size, one with the bushing upside down (tapered end down) and then return it to the initial position for another run. Almost zero runout as the half neck seems to help align very well instead of trying to jump straight on like most neck dies. Then you do not need a collet that fits perfectly or neck bushings that are in perfect concentricity with outside. Lets a less than perfect machinist get away with it.

Rust, Similarly I have found grinding necks out on full length dies not too bad or hard either. But requires more time and patience (as you have to work a long way into the die and grind off the correct amount in the correct places). Certainly both techniques are well within the range of an average machinist. I like using the non carbide or non hardened bushings and they are very cheap.

ALSO armed with a FL die honed out in the neck and an oversized expander it is possible to "resurrect" the concentricity of some pretty awful shells without fireforming. Quite often these shells are the ones with uneven wall thickness so I send down a neck expander, turn down the necks, then put them through the FL die (sometimes with a few quarter rotations and several times through). But of course all that working may require annealing.
Cam

 

[camac]

German,
After reading your great article a second time I have a few suggestions as possible reasons for some of the variations. Firstly the neck dies tend to work better on fired rather than FL sized brass. This is because the collet chambers are a little oversized. Therefore your shell fits and lines up better (less room for movement). Once FL sized, there is more "play". This is particularly apparent when variation in wall thickness might allow the shell and or neck bushing to be pushed to one side. I have also found the same effect with the redding and forster bullet seating dies. The shells align better in fireformed state rather than FL sized or new. Also align better if you leave a little bit of unsized neck at base of neck to sit better in collet. The closer the collet matches the chamber, the better. ideally a neck die will work well either way if necks are nicely uniform as they will self align better anyway. This hopefully explains most of the combinations you saw in your experiment.
Cam

 

[JCummings]

German:

Once again you have provided us with a nice thoughtful article which asks us to think. Thank You.

The only conclusions I have been able to reach ( even before reading this article and discussion) in this matter are:

1) that the less you do to the brass the less you can mess it up
2) you want any runout on the outside of the neck.

Mininal case sizing/headspace setback and minimal neck oversizing with an expander that just kisses out the neck has always seemed to work best and make the most sense (to me).

How to best achieve these goals is highly depended upon the equipment used and how it is adjusted. I bet I cause more runout than my equipment. One thing I recently (re)proved is that if you don't use the THE exact same shellholder with the die the results will be different. I haven't but probably should have a designated shellholder in the die box for every die I use. (I'm starting to work on this for my important precision cartridges at least and do for my benchrest guns). You have certainly shown that sequence matters, but best is not necessarily the same for each cartridge/brass being used. On top of this the bushing and probably the actual die set up that works best needs to be adjust for different brands of brass due to thickness and elastic variations.

In the end my head starts to swim and I begin to think it might not all be worth it. In fact, for most of my shooting it isn't. Highpower across the course with a service rifle is not the place for any of this stuff. Mid range prone maybe some of it. Benchrest and long range prone/F Class are the places where I think it matters.

I believe I may make a move to more strictly standardize the components (particularly brass) I use for long range prone/F Class to remove this additional variable. I already have done so for my benchrest equipment.

Once again thanks for the food for thought
Useful discussion fore me folks.
James Cummings

 

[GermanS1]

Boyd, Rust, Cam, James - I really want to thank you for your interest in the article and all the great ideas. This is exactly the kind of thing forums are best for and I really like it when we all pool our experience like this.

Cam, the idea that the fired case fits the neck die's sleeve/collet better than a full-length sized case is one of those things that causes me to slap my forehead and say "of course - why hadn't I thought of that!" I think that is reason enough to always use the neck die first. I am no machinist, but I want to make sure I understand your comments about modifying the bushings. As I read your comments, you're saying the bushings should have a taper for about half their length instead of just the little edge break they seem to have from the factory - is that right? And then, after using the half-tapered bushing, you flip it over and that allows you to finish the whole neck while maintaining good concentricity. I hope I've understood that correctly. What about bushing clearance in the die, or how tightly it's held by the stem? I nornally back the stem off a tiny bit to avoid binding the bushing; is something like that what you do also?

James, I certainly agree with both of your initial points: do as little to the brass as possible and keep runout on the OD if that's what it takes to get the ID concentric. Boyd mentioned measuring ID runout on the bullet and that's possible, but also introduced the runout effect of the seating die (which isn't much, so we can almost ignore that factor). I'm going to see if I can configure my NECO/Audette tool to read inside the case neck for a direct ID runout measurement. Not sure it can be done, but I'll try.

Like you, I spend my range time principally on the Highpower range, and a lot of these details don't translate directly to improved results there, but I enjoy these reloading problems for their own sake. Working on these things and writing and thinking about them beats the heck out of watching TV in the evenings!

Rust, as you and Boyd mentioned, a one-piece die with the neck correctly sized would be the best setup. I picked up a set of standard Redding .30-06 dies for $5 the other day at a match and I think I know someone who can hone the neck out on the sizing die. I turn my brass to 0.125" thick and finish with a 0.331" bushing, so I will see if he can hone it to 0.330" so that the expander just barely brushes the neck on the way out. I have a Redding carbide expander in my tool box that might be perfect for that.

Well, I can see there will have to be a follow-up article relatively soon! Thanks again for all the ideas. I can't think of a better way to spend my birthday than pondering all of these things and working on some brass - unfortunately, I have to go to work!

 

[cfrazier77]

German,
I thank you for your article and your insight. I have been reloading for 3 years and have been trying to make better and better end products. You, and the others on this thread, will help me with run out big time. I just ordered my first concentricity gauge, Sinclair's, this weekend and will try your ideas out very soon.

Have a happy birthday and shoot well!

 

[camac]

German and everyone, I AGREE TOTALLY and everyone here is right on track with real life experiences

QUOTE " I want to make sure I understand your comments about modifying the bushings. As I read your comments, you're saying the bushings should have a taper for about half their length instead of just the little edge break they seem to have from the factory - is that right?"
Yes, German, I believe they should have a relatively long taper. this allows the bushing to line up on the neck properly before trying to squeeze it in - I believe the taper should be around half (or slightly more) the length of the NECK - not the bushing though - probably around 1/6- 1/4 of an inch

"And then, after using the half-tapered bushing, you flip it over and that allows you to finish the whole neck while maintaining good concentricity. I hope I've understood that correctly. (YES but I would recommend leaving a little neck unsized at base - this allows shell to line up better in some collet seating dies and rifle chamber)
What about bushing clearance in the die, or how tightly it's held by the stem? I nornally back the stem off a tiny bit to avoid binding the bushing; is something like that what you do also? (YES -in fact it is important to have some freefloat so that the bushing can align itself correctly with shell. In almost all cases unless you have a machined collet that matches your shell exactly and a bushing that is aligned perfectly, and is perfectly concentric and is tightly held etc etc you will not acheive accurate neck sizing this way (Trust me I have tried and tried and tried - some machinists could do this for you but it is a hard job). It is also why all the die manufacturers have a bit of freeplay in their neck bushing dies.

This is why the full length die gives little runout (without neck expansion ore with minimum neck expansion)as it hold s the shell tightly in line with the neck sizing. Only drama with the FL sizer is it also reduces the shell back to SAAMI specs (unless you have a custom one or used a slightly undersized roughing reamer to make one like RUST suggested which is the ideal). This in turn leads to the next problem -it is not as good to have a bullet with zero runout but then does not line up with barrel in your chamber. Just like using new brass again and not as generally accurate as fireformed.

Another simple trick for measuring internal runout is to measure external runout and then use your neck wall thickness gauge or mic. You can put a few index marks on the shell, measure runout and then have a look at the neck wall thickness at each point. You will find the "thick" part of the wall drags the bushing over that way mostly but also pushes the bullet out the other way in seating. A few calculations will give you an idea of what is happening internally. Of course if you neck turn and have no real differenc in neck wall thickness then it is easier to measure and understand what is going on.
The more you think about it the more there is to cover...


I look forward to more articles German and the real question I keep posing myself but can't answer is "how much real world difference does all this make. I would suggest that it is more important with short bearing surface length projectiles than the long VLD's etc. BUT a very interesting challenge. I have always felt my accuracy improved significantly with low runout (e.g. latest 6.5x284 project - my dies were giving me 7-10 thou runout off the shelf. I neck turned, tapered the bushings etc and then got runout down to under 0.5 thou and groups tightened up from around 1 moa to .25 BUT is this the barrel just getting shot in? or other work I had performed - bedding, crowning etc..???? - Some guys shoot very well without doing all this work!! - I suppose I too thou would prefer to do this than watch television. There is as much enjoyment for me in this as there is in the shooting. A bit like a RC aircraft build. just as much fun to build as fly...(That might be just my old age talking)

The great thing about these forums is you can learn several guys many years of experience in a few paragraphs.
GREAT WORK GUYS _ KEEP IT UP.
Cheers, CAM ;D

 

[Rust]

German - I read an interesting post from JerryHM, the gist of which I have been thinking about for awhile now.

I have a DIFFERENT view on runout. My loads have very little runout, but it is not something I worry about.

One very important item left out of this discussion is neck clearance! I have found from MANY years of testing run out that : the tighter the neck clearance the more perfect the loaded round needs to be. I use tight throats (0.0002" bigger than the bullet) and a neck clearance of at least 0.004", up to 0.012". Testing 308 loads with a tight throat and 0.010" of neck clearance; comparing loads with 0.001" or less and loads with a runout of 0.012" there was NO difference in group sizes. The high run out loads were also shot with the high side at 12.3.6.and 9 o'clock in the chamber, also no noticeable change in group size as compared to the low runout loads.
I believe one of the reasons that improvements in accuracy is seen when neck turning the brass is the increase of neck clearance more than an improvement in the brass....

I have NEVER found in my testing that increasing neck clearance had any ell effects on accuracy.

Another post in the same thread noted that a round chambered in a tight throated chamber showed less runout after having been chambered.

It would seem to me that running a bullet into a throat with .0002" clearance would tend force the bullet to align with the centerline of the bore. This effect would be more pronounced with the long VLDs seated long to begin with as all that bearing surface and the tight throat clearance would have increased mechanical advantage. Say for instance a 40 gr vs an 80 gr bullet in .223.

But I'll still run a no turn tight clearance neck if for no other reason than it works the brass less on my match brass. My loaded rounds have minimal runout and very uniform neck thickness so the .003" to .004" neck clearance shouldn't be an issue that can negatively affect accuracy. Since my best rifle shoots in the teens at 100 yards I don't think the accuracy has been negatively affected. ;D

Certainly food for thought though.

 

[wapiti25]

I first want thank German for his hard work and sharing it with us. I also want to thank everyone else for their posts. This has been one of the best exchanges of information on a subject that I have read in a long time. I agree with each post as contributing to the original. Having been a modelmaker I would like to say a few things in maybe a different way. I refer to points that are measured as datums, I look at how datums on the caridge case affect the other datums that we measure. Awhile back the question of runout was posted and I came around to the thought as how the ammo sit's in the chamber before firing. My thoughts for runout measurements came to using a sleave such as the Wilson trim tools use that has the body taper cut concentric and straight to a cylinder that would roll in "V" blocks so you can measure necks, bullets at points that you are interested in. This would be inrelation to an"average of the case body" as it is being held in the sleave. I look at surfaces as not being perfectly flat, round or other geometric shapes but having small imperfections. All of you guy's are on the right track, thank you for sharing your thoughts.

 

[wapiti25]

I hope this not out of line but there is something I feel that may contribute to runout. I look at brass in the old shop terms as "dead soft", 1/4 hard, 1/2 hard. I anneal brass softer for some applications than others. When doing a lot of change in shape I usually use about dead soft brass. I bought an XP-100 in 7mmBR before I could find the old Rem (308) BR brass. I made brass out of military 308 by fully annealing, opening to nearly cylinder shape, inside reaming to uniform wall and capacity, then forming into 7 Br. I also started to prepare the necks by boring instead of outside turning. I think single point boring is the best way to give the bullet a cylinder to seat into. Inside boring is not a easy task. I don't ream for accuracy as a reamer will usually follow a crooked hole. Sorry, I got off on the forming but I have found that brass level of hardness can change things more than you think.

 

[GermanS1]

Rust, a VLD has a very short bearing surface because of the long nose and a tight throat will not contact anything but the bearing surface, so the leverage effect might not be too great. Still, it's a good thought but I prefer more throat clearance, typically 0.0005", because the conditions here are extremely dusty and my ammo is not spotlessly clean when I chamber it. Lying on the ground in blowing dust is not good for cleanliness! Others might have a better approach and I never claim to load for maximum possible accuracy under perfect conditions - I load for darn good accuracy under any conditions and there's a slight but important difference there.

Waipiti, I like the idea of checking in the Wilson case holder as a way of checking the neck versus the theoretical centerline of the case body, I'd been wondering how to do that. I'll give it a try tonight on the old Sinclair tool which uses a v-block. I'm preparing a little article comparing a few concentricity tools, that will be an interesting addition to it.

 

[camac]

I'm preparing a little article comparing a few concentricity tools, that will be an interesting addition to it.

Great German, Your articles and the corresponding threads are fantastic fodder for discussion. The Hornady "straightener" and measuring runout is going on in an interesting discussion at the moment. Wapiti, your view on "datum points" along the shell and bullet would be greatly appreciated there as well. I do something similar in using the V blocks. By moving V blocks to case head and shoulder, then firstly measuring "roundness" of shoulder and base. If they are round then you have effectively created a theoretical "centre line" as you rotate. Datum points can then be taken at critical sections such as along the neck, projectile and indeed down the body. Sometimes with surprising results. I like to do this with all my chambers (ie fire a couple of shells and then check all over). Rather than Plug casting, I like to see them pretty good before perservering too far - I have sold off more than a couple of barrels that didn't measure up.

 

[Rust]

German - I have some components on hand because I'm going to order yet another .223 barrel and reamer. I really don't know what the fascination with .223 is but it will probably end up like it did a few years back when I tested more than 750 loads for .223 with bullets 55 gr and under.

Anyhow, checking what I have on hand with the admittedly less than perfect caliper (the Starrett stuff is at home) the Berger 90 gr .224 VLDs have a bearing surface length of about .380" to .390". That being a bit of a guesstimate, I actually think it's a little longer but can't measure it accurately with what I have on hand. Anyhow, for a .224 bullet that is quite a bit of bearing surface even given the extreme length of the bullet at 1.241".

Loading as I do for single feed, for a chamber which allow the VLDs to be seated out as far as reasonable, and using at most .002" neck tension it seems a reasonable theory. Maybe once I get home I can load some dummies and see what works.

The Wilson case holder idea was excellent. Works for me because I have pretty much every one they make. Of course for the really picky, if they have their reamer, they can get a local machinist to whip a similar case holder up. One recommendation though, check the runout on the case holder before checking a case. Could prevent a high blood pressure incident.

 

[Viperdoc]

I am new to this, so perhaps I'm missing something: Running necks through a die without an expander (whether once or in two stages) should of course decrease the runout, but only in the OD. If there are neck wall inconsistencies, this process simply transfers the lack of concentricity to the ID (without using an expander or mandrel).

Does it not make more sense to run a mandrel through the neck first, to insure the ID is consistent, then neck turn as necessary in order to get the OD and wall thickness the same all the way around, followed by a final resize?

This would ensure that both the ID, OD, and wall thickness were the consistent, resulting in maximum concentricity and the least runout? Otherwise, just working on the outside of the case is just going to push any concentricity differences to the ID, which is probably where we least want it?

I admit to being a beginner, so please don't flame too hard- just trying to learn to make the best ammo and become the best shooter possible.

 

[GermanS1]

You're exactly right about all that, which is a good part of the reason why I turn necks. Somewhere in this thread, or perhaps the parallel thread on concentricity, we began to discuss mesuring runout on the ID (before seating a bullet). I have an idea on how to do that but haven't tried it out yet. However, your basic premise is correct: runout matters most on the ID. We measure on the OD as a convenient proxy; unfortunately, while convenient, it is is dependent on the uniformity of neck wall thickness.