Perhaps I should try FL sizing for better results with new brass? I have a Forster FL sizing die. My normal fired brass routine sizes the neck with a Lee collet die followed by full length (no neck) shoulder bump using LE Wilson bushing dies minus the bushing.
I've had the best results with virgin brass using an oversized expander mandrel (~.001" over bullet diameter) first, to open up all the necks to some minimum diameter greater than the bushing I intend to use. The cases are first sprayed
liberally inside and outside the neck/body with Hornady One-Shot lube. I use the following mandrels/Gen II die setup:
https://www.sinclairintl.com/reload...inless-steel-expander-mandrels-prod33134.aspx https://www.sinclairintl.com/reload...inless-steel-expander-mandrels-prod33134.aspx).
Next, I size all the cases back down with the appropriate FL bushing die. I typically select a bushing to give ~.002" neck tension (interference fit). [Note - I remove the expander ball from all my sizing dies, due to the potential for the exact issue you're describing]. The cases are then cleaned
I have found that the neck diameter of virgin brass can vary quite a bit straight out of the box. The expander mandrel opens up all the necks to some minimum diameter that is larger than the bushing, then the FL/bushing die re-sizing process brings them back down to a more uniform diameter. This process does not move the shoulder detectably. Finally, it is not necessary to use the OS expander mandrel step once the brass has been fire-formed, as the necks will open up noticeably to match the chamber upon the first firing.
The main reason for this process is to get better and more consistent performance out of virgin brass. I don't like the idea of putting a bunch of rounds into the dirt just to fire-form brass, getting little else in return. I have no qualms about using virgin brass prepped as described above in local F-TR club matches, and it has performed quite well in that regard. Whether this application would be appropriate for the most demanding accuracy/precision requirements in a discipline such as BR is questionable. Nonetheless, it works for my purposes. Once the all brass has been fire-formed, the load will have to be tweaked to bring it back into tune. Nonetheless, I get good use out of virgin brass with this approach, and gain some useful information about what a specific rifle "wants" during the process.
The most important point about this procedure is that I don't typically ever see runout of more than .001" to .0015". Cases prepped in this way have very straight necks. In order to correct a runout problem, it's easier to identify the source of the problem, or particular step in which it occurs, as a first step. Then you can correct it. Otherwise, you are guessing, then possibly changing a bunch of things and hoping the problem goes away. It sounds like you are doing the right things. Checking runout at the case neck after each step, and then again once the bullets are seated should allow you to identify the point at which the runout has been introduced. As I mentioned previously, the use of the expander ball in a sizing die has often been associated with runout. That is not to say that the expander ball must be a problem, many use them without issue. Nonetheless, I routinely remove them as soon as I get a new die because I haven't found them to add any benefit to my brass prep procedure, and I don't want the potential for any runout issues. YMMV.
One other thing worth mention is that when you use a mandrel and/or a bushing die for neck prep, it is critical that both the mandrel and bushing are adjusted to be "free-floating" as indicated in the manufacturer's instructions. If either one is too tight, or "locked down" in the die in a manner that is not perfectly aligned, runout will be the likely result. Finally, although it wouldn't typically be my first call, it
is possible to get a bad or misaligned die. For example, if you were to identify a seating die as the suspect point the reloading process [hypothetically] through stepwise runout analysis, and have removed the expander ball and also checked to make certain that the bushing has been properly free-floated, and the die mounted correctly, then I would probably give the die manufacturer a call to see what they recommend, whether that be replacement, or possibly an adjustment or doing some step in a particular way.