Expanding vs Turning Mandrels

Yeah fair enough, you're basically right were I'm at in 6.5mm for bushing/mandrel sizes, I'm thinking that's a good place for 6-6.5mm in general since it seems to work for a lot of guys. I haven't messed with neck length on it yet. It made one hole with five shots, I said "done!".
 
I appears that obsessive annealing is putting folks in a tail chasing loop.
You over anneal to soft brass, and then size more to compensate, which leads back to a need for annealing..
It's similar to an obsessive sizing loop, where you size more only to end up needing to size more.

Pretend for a moment that there is no such thing as annealing.
You want good consistent tension, and you want your brass to last.
Here, your plan would center on minimal sizing. That's it, no loop, the brass lasts as many sizing cycles as clearances allow -before it hardens to the point of substantial difference. If your clearances were minimum, your pockets may open before ever reaching a neck hardness issue.

I do this with most guns, and this is how it goes for me;
By ~20 reloads, I've set aside 1or2 cases departing in measured seating forces. By ~30-40 reload cycles I may have enough cases set aside that I decide it's time to reset the batch. I'll dip anneal at that point, NOT to full anneal, but to stress relieve(process anneal) back to a standard that does not disrupt my load. Good to go
The only end to brass I've had with this, so far, has been because of opening pockets, and only with Norma brass (which I no longer use).

I do turn necks for most chambers, and all that I have control over. The only time I don't turn is when it would lead to excess clearances.
I've been mandrel pre-seating necks for ~30yrs now. I do this to reduce seated runout, and to measure and match pre-seating forces, as I'm correlating indirectly to tension. I can do this because I've installed a force sensor in my mandrel die.
With turned necks and my minimal sizing plan, I don't actually need a mandrel. But I still use it for measurement.

IMO, the ideal mandrel would be at cal diameter. Necks would spring back from this as they would after bullets were seated and pulled.
The force of that spring back times the area applied to seated bullet bearing, is your tension (in PSI).
If your sizing length is no more than seated bearing(which is proper), then this spring back force from cal is really what you have, regardless of excess downsizing. A bullet will expand a neck just the same as a mandrel at cal.

To picture it in distance(interference fit), necks spring back 1/2 to 1thou depending on cal diameter. The force within that spring back is what's gripping your bullets. If you downsize to leave 5thou interference for ~half neck length, it's just 4thou excess. The bullet will upsize that to cal, with 1/2 to 1 thou spring back force. All you've really done is work harden your necks.
Now if you need more force, work hardened necks can provide it for sure. That's on you to manage with a sizing plan or annealing.
I'm just suggesting that it's easy to manage with a sizing plan. Just avoid excesses, and adjust tension with LENGTH of minimal sizing.
That's X-pounds (of normal spring back force) times Y-inches applied (to bullet bearing area gripped).
 
My process is the same as you said - size down, mandrel up, use the same size bushing/die and mandrel every time, working towards minimal movement all around the case. The initial fireforming takes a more than one firing, that's the only reason I do anything differently at the beginning - only sizing the case web and neck until the shoulders fill out, anneal once, then start the consistent loading process cycle until the necks don't size correctly anymore.

The problem I'm facing is there aren't a lot of cycles before cases start getting hard because my necks move so much in several chambers. My goal is to get a consistent and repeatable anneal so that I can use the same bushing and mandrel regardless of how far the neck expands each shot, since neck clearance can't be controlled until I start buying a lot more reamers (and letting people I shoot with use them). I think that's what you're saying with sizing plan or annealing later on - you control via sizing, but I'm saying I can't control via sizing because of excessive neck expansion, so I'm using annealing to try to compensate. And don't want to get caught in a loop while I'm trying.

Now if you need more force, work hardened necks can provide it for sure. That's on you to manage with a sizing plan or annealing.
I'm just suggesting that it's easy to manage with a sizing plan. Just avoid excesses, and adjust tension with LENGTH of minimal sizing.

That makes perfect sense, and I agree with the sizing plan logic. I'm looking at an annealing plan as part of a process where I can move a neck 0.004-0.008" without excessively hardening the case and getting consistent, minimal spring back.

I'm thinking that since an induction annealer can be set very precisely in terms of time and energy delivery, I can find a repeatable anneal setting that can be used more frequently (potentially before each) sizing cycle - more akin to your stress relief anneal and not full anneal - in cases where the neck movement when fired is excessive. Salt annealing could probably do the same thing, but with the induction machine I can hit a few buttons to change the setting and not have to try to change the pot temp. Flame may work great for some people but not me, I'm not consistent enough to be successful with it with what I'm trying to do here.

IMO, the ideal mandrel would be at cal diameter. Necks would spring back from this as they would after bullets were seated and pulled.
The force of that spring back times the area applied to seated bullet bearing, is your tension (in PSI).
If your sizing length is no more than seated bearing(which is proper), then this spring back force from cal is really what you have, regardless of excess downsizing. A bullet will expand a neck just the same as a mandrel at cal.

And now that you put it in words, I guess my ultimate goal is exactly what you said - use a sizing mandrel at caliber size in a case sized such that running the mandrel doesn't move the neck, so the only spring back from the mandrel is exactly what a seated bullet would be subject to. That's got to be the ideal amount of spring back for any given caliber because seating the bullet wouldn't change anything in the case. At that point the length of neck sizing is the last variable, but as long as it's held constant at some setting that would get you very consistent and repeatable seating.

Now this might be a fool's errand, but I have a pin gauge set to measure actual spring back at different anneal settings to see if there's any truth to my thought that I can find a setting where a full-caliber mandrel goes in and out without the neck dimension changing after sizing down from whatever a chamber gives me.

Watch it that the AZTEC code does this the first time, and all I did was confirm their research. I wouldn't even post it if I did, I'd need an asbestos suit to survive the flaming. 🤣
 
AZTEC is the analysis mode where an AMP annealer nukes the ever loving life out of a case until it melts 🤣
IMG_7900.jpg

From left to right - Peterson 30-06 brand new, unfired 30-06 case from the same lot, then two Lapua 6.5CMs from the same lot/same firings. The three cases on the right have been run through the AMP for analysis. The 6.5CMs are at 90* from each other, one showing a "melted" 90* section neck, the other showing a non-melted section.

The AMP AZTEC "analysis" mode heats a sacrificial case to the point it's destroyed using the induction coil, but there has to be some kind of feedback mechanism in place that measures output current - once it starts spiking (because the case is absorbing less heat or something very technical?) it stops, picks a point in the cycle, calls that the correct voltage/current/time setting to achieve AMP's targeted anneal, then presents it as a code that can be programmed back in later for consistent repeatability. AMP also has a default mode where you can program in a code from a list provided on their website that is based on analysis of similar cases, but isn't based on feedback from a case in your actual lot.

Since timing and output are regulated by something faster and more precise than human hands and eyes, induction annealing has to be more precise (not necessarily more accurate) than flame annealing. And since heat input to the case is based on electrical input over a very short time cycle, it's faster to recalibrate than salt bath annealing because there's no physical mass of salt that has to change temperature. Time factor with either flame or salt is also manual, the induction annealer turns that over to a digital timer. There's a 100+ page thread on Accurate Shooter that goes into the details of homebuilt induction annealers - it was more than sufficient to get me to just buy an AMP because I don't have a masters degree in engineering, and the AMP is cheaper than going back to school for long enough to understand it all.

My thought is that since the AMP can anneal more precisely (again more precisely, not inherently more accurately) than I can by hand with flame or salt, then I can essentially tune annealing using the default code mode (make annealing more accurate using the precision the tool offers) to any given set of matched cases to the point I can anneal them every firing, then size them down however far they need to be sized down to pass a caliber sized mandrel without affecting the resting state of the neck - passing in and out and the neck measures the same after. Meaning for a .2640 caliber neck, a .2640 go (plus) pin gauge wouldn't pass, a .2639 no-go (minus) gauge would pass, then run a mandrel at the diameter of the bullet in and out, and then gauges both still confirm the original size. Meaning seating a bullet will not change the state of the neck, meaning this is the most consistent possible grip on the bullet that you can obtain from a neck. Just like you thought.

I might have go/no-go and the target spec backwards or wrong. Since we're measuring spring back and targeting a maximum size and not a minimum, then I'd want the minimum go to fail, and the maximum no-go at the next step down to pass right?

I have Z-gauges to the ten-thou right now to work with, but a pair of X-gauges is only about $40 if I can define the correct pair to get.

The AZTEC comment at the end of my previous post was basically saying just watch for it, I'll get to the point where I can size down 0.005" and a caliber mandrel passes in and out without moving the neck at all because that's what AMPs research on optimal hardness actually gives, and I end up sounding like an AMP shill saying what it did. I kind of doubt that happens because there's no way for AMP to account for differences in the work hardening that happens based on how far and in how many increments I size down before I run the mandrel. But maybe it returns it to some soft base state where that happens. IDK. That's why I have the AMP machine, a calibrated bushing, and a hundred bucks worth of pin gauges to find out.

🦫🌰🥜
 
Last edited:
No problem, it is very interesting to read all the different stuff people try and everything that is so different. I can tell from personal experience using the amp annealer that it definitely takes out a lot of spring back. I've noticed it with a 270 and 300 WSM where on a regular full-length die the bullets could be pushed in by hand. Which leads me too believe that a bullet diameter mandrel would produce the same results on any of the brass in my situation. Which is why I'm leaning towards .003-4 under and then back up to .002-.0025 under mandrel.
 
I agree with your underlying rationale in terms of practicality - anneal, neck too loose, so size down more, neck back up, then you can seat consistently in that brass lot at that time because you re-hardened the brass to some extent. Works fine if the outside neck dimension can't be controlled for.

My long winded meander was basically taking your logic and saying if I have to size down X.xxx amount each time, what do I need to anneal to to get repeatable spring at exactly caliber size back without a mandrel.
 
I really haven't played around with the codes other than what Aztec gives me. After reading all of their testing and the science behind it, I am pretty sure I cannot do a better job of fine-tuning. But I can see how going up or down with the codes could get you the results you desire.
 
The more that you work the brass, the faster it will work harden. That's an unavoidable trait of the metal. Working it less means that you'll need to anneal less often.

What I've put together from reading this an other threads on the topic, see if I've distilled all of this correctly:
It is important to first figure out how much smaller the neck needs to be for any particular bullet. Maybe it is consistent across all of the bullets you choose to use in a particular case and rifle and maybe each has it's own ideal neck size. I don't know enough to know much about that.

Once you have the ideal neck size, then you can work out what the bushing size should be. The bushing needs to make the neck ID smaller than the expander so that the expander sets the final ID of the neck. How much smaller I've no idea, but I'll guess that less is more so long as it results in a consistent neck ID.

[ ] As an aside, I wonder if large over-bore cartridges with steep shoulder angles have a tendency to need annealing less often than those with shallower shoulder angles due to the heat of combustion partly annealing the neck on every firing?
 

Recent Posts

Top