I've been thinking about neck sizing and about a better way to do things -- I'd appreciate if some of you would be willing to read over what I've been working over in my head, and encourage me and/or poke holes in the theory as appropriate...
So, if you don't turn necks, you are going to have a degree of irregularity in neck thickness, the degree of which depends on luck, how anal you are about culling, and the quality of your brass...
If you use bushing neck dies, or have a custom honed sizing die, you'll be sizing the outside of the neck, but bumps and irregularities will be pushed to the inside of the neck, and also the inside of the neck will likely not be of uniform diameter due to the varying neck thickness of your unturned necks -- and in my understanding, this can't be good for consistant neck tension, and possibly runout.
One alternative is the Lee Collett neck sizing die, which squeezes down on a mandrel inside the neck -- since you're shaping based on the mandrel on the inside of the neck, after sizing with this die the inside of the neck should be consistent diameter, and all of the irregularities pushed to the outside. This is all good, but my brain got churning on an alternative.
So lets say you have a full length sizing die with the expander ball removed, but has been honed such that after sizing, the neck has been sized down just enough such that the thinnest neck-thickness brass that you allow (after culling) will have its neck resized after springback such that it would create just a bit tighter neck tension than you actually desire. This ensures that even the thinnest walled brass will be sized to have neck tension slightly higher than that which you actually desire (if the bullet were seated at this point).
So, then you run the above sized brass through something like a Sinclair expander body with a mandrel of appropriate diameter (my guess is that the oversized neck turning mandrel, which has bullet diameter minus .001", should spring back just enough to give you .002" of neck tension) to uniform and appropriately size the inside of the case neck. The slightly tighter neck sizing that occurred in the previous step insures that even the thinnest-walled brass gets sized up to the appropriate inside diameter/seated neck tension in this step.
I've seen lots of references to using the expander body/mandrel BEFORE using something like a neck bushing die to size the neck -- but it seems to me that sizing the inside of the neck last should make for the most uniform neck tension, especially if you don't turn necks, as you are explicitly sizing and uniforming the inside, as opposed to the outside of the neck and hoping the inside turns out okay. This should get you the positives of an expander button, without the negatives (ie, hopefully you're working the brass less due to your honed die, and you avoid the run-out problems associated with expander buttons).
I'm hoping that this will prove a relatively quick way of getting low runout with consistant neck tension by avoiding some of the pitfalls of other methods, and avoiding neck turning
(Note: Assume reloading for a no-turn match chamber)
Any gaping holes in my understanding or in the described method? I just wanted to bounce this off you guys before I spend money testing the theory...
Edit: Just a note, I realize I'm not just neck sizing here (assuming that I'm using a honed full sizing die in the initial sizing step) -- Intention is to still to full length resize every time for feeding reliability (though with die adjusted to only slightly bump the shoulder back) -- I'm just focusing in on how the neck gets sized here.