Annealing Verification

[dcali]

All I’m saying is that the numbers tested out by AMP correspond to a fully annealed state, red hot or not. That also jibes with the state of necks that were tested from the factory in an old PS article.

 

[dcali]

By "fully annealed" do you mean dead soft? I do want some elasticity in the brass. I don't think AMP settings are designed to produce dead soft brass per my observations with a few different brands/chamberings. I did not sleep in a Holiday Inn Express lately so my statement is not qualified.

Just to make sure I wasn't mis-remembering things, I went and looked it up again. It looks to me like the AMP machine takes the cases down to around 100-100 vickers hardness, which corresponds to roughly a quarter-hard state. So let me correct my earlier statement - their machines don't fully anneal, but they get close. What they're able to achieve is actually pretty impressive, if you ask me.

It's my opinion based on the data I've seen that what a lot of people do with torches is basically a full anneal, and that fears of over annealing are overblown. The fly in the ointment is that data on thin sections is very hard to find and collect in a meaningful way. But it's hard to argue with the before and after tests done by AMP. Their machines appear to do what they say.

I haven't a clue what number is best. I don't anneal at all, personally.

 

[rardoin]

I haven't a clue what number is best. I don't anneal at all, personally.

Oh My GOD...say it ain't so.

Thanks for the reply.

 

[Mozella]

....... snip............My curiosity aimed toward the question of why a certain Brinnel number is aimed for and what experimentation, on targets, determined that that is the ideal level of annealing to shoot for.

Here's what I think I know. The AMP guys have a precise machine and also apparently have the wherewithal to do precise hardness testing of the resultant product. The AMP machine also costs a big wad of cash. Your AMP machine can duplicate the results of another AMP machine. That's not true with my DIY "Skip Design" automatic annealer. Although I can get very repeatable and reliable results on MY brass, expecting you to duplicate those results on YOUR DIY machine with your brass is unreasonable.

Does it matter? I don't think so because we can both win matches using DIY machines. I'm not convinced the AMP machine offers an advantage; even a tiny advantage. In fact, for me, the process is too tedious so I would never buy an AMP machine even though I acknowledge that they do a good job when it comes to precise and repeatable annealing even when using different machines.

Add to this, the fact that Bryan Litz did some careful testing and concluded that he couldn't detect any advantage in annealed-every-time brass when compared with brass fired and sized 10 times without annealing. That strongly hints that the difference in precision at the target must be very small if it exists at all.

I anneal every time and I know for a fact that it keeps my necks from splitting until the primer pockets wear out. That's a good thing and a big improvement from my pre-annealing days. I also hope that it might help my score, but I'm not offering any guarantees that it does. On the other hand, I'm pretty sure it doesn't hurt.

I'm quite happy with my annealing process, my results, and my confidence level when I go to a match. I don't know my exact Brinnel number, nor do I intend to find out.

I have a pretty good uncalibrated tire gauge with which I can measure my tire pressure within about 1/2 psi plus whatever error is built into the gauge. It has served me well for many years both on my street vehicles and back in my days of sports car racing too. If someone offered an expensive calibrated tire gauge which has a guaranteed accuracy of .001 psi would I buy it? Of course not. Do I need it. No, I don't. Do YOU need it? I'll bet you don't. Would some guys buy it? Yes, some would. You get the point.

 

[mikecr]

Pretty good read on this: http://www.6mmbr.com/annealing.html

 

[rijndael]

Add to this, the fact that Bryan Litz did some careful testing and concluded that he couldn't detect any advantage in annealed-every-time brass when compared with brass fired and sized 10 times without annealing.

Was this analysis documented somewhere online? I'd like to read it.

 

[dcali]

Was this analysis documented somewhere online? I'd like to read it.

It's in his 2nd "modern advancements in long range shooting" book. It's a simple test. Using two sets of cases (.308 palma), he shot and fired each set several times, only annealing one set. The velocity measurements of the annealed cases were not significantly different from those that hadn't been annealed. That's the gist of it. It's worth buying the book. Lots of interesting stuff in there.

 

[Fred Bohl]

When I was trying the molten bath techniques I settled on the following test process:

For 6mmBR case (both Norma and Lapua) the lots used as received had neck IDs of 0.2395 to 0.2405. To establish a reference, I ran an expander mandrel thru these necks (actual max size of 0.2416) and found new cases gave a finished ID of 0.2408 to 0.2412. After quite a bit of experimenting with time and temperature I decided arbitrarily that testing the cases would need neck sizing for the ID of 0.2395 to 0.2405 followed by running the expander mandrel thru the neck. Again arbitrarily I decided that a post expander ID of more than 0.2415 was too soft and less than 0.2405 was too hard. This proved satisfactory for evaluating the molten bath techniques and for the cases used, necks outlasted the primer pockets and the cases lasted longer.

After acquiring an AMP unit I repeated the process for a test lot of each type and found that all gave post expander neck ID of 0.2398 to 0.2402 after annealing. I have no means of correlating these or the earlier results to actual hardness or grain structure but I'm gratified that the AMP results are within my arbitrary limits. I'm comfortable with using this snap-back data to asses an annealing process but doing the ID measurements is tedious using a friends 0.0001 Deltronic pin set.

 

[Fred Bohl]

Process annealing ( sub-critical annealing) is a thermal treatment to remove residual stress of cold working without a significant reduction in cold worked strength and stiffness. Dislocations are not affected and the fracture resistance is also not affected. If your brass is process annealed only seating stiffness would not be reduced.

My approach to annealing has been intended to un-do only the work hardening done by firing the case. From your definition above it would appear that I'm attempting to achieve Process Annealing. Would I be correct to assume that to achieve "Process Annealing" on the neck-shoulder area of a specific starting caliber, alloy, and factory Process Anneal, I should expect my annealing process to require a very specific combination of temperature and time?

 

[Fred Bohl]

I know I'm over thinking this (it is 3 degrees F out side and not yet game time) but bear with me. Say I have two rifles chambered for 6mmBR one with a 0.272 neck and one with a 0.271 neck. Further, I start with the same lot of brass and bullets and load for both the same. Assume that the loaded neck diameter is 0.269 this will give 0.003 clearance in the first and 0.002 clearance in the second. Therefore, the first will potentially work harden the neck 1.5 times as much the second.

Should the annealing process time and temp (or AMP program) be different for the two conditions?

 

[riflewoman]

I know I'm over thinking this (it is 3 degrees F out side and not yet game time) but bear with me. Say I have two rifles chambered for 6mmBR one with a 0.272 neck and one with a 0.271 neck. Further, I start with the same lot of brass and bullets and load for both the same. Assume that the loaded neck diameter is 0.269 this will give 0.003 clearance in the first and 0.002 clearance in the second. Therefore, the first will potentially work harden the neck 1.5 times as much the second.

Should the annealing process time and temp (or AMP program) be different for the two conditions?

You really have to account for the whole dimension change through a whole firing cycle.

Let’s use a 223 case as an example.

A max chamber neck dia is .257. A min case diameter is .251. A sizing die Likely has a .245 neck diameter and the expander gives a finished neck diameter of .249. Seating brings it back to .251.

Each CW is (large diameter-small diameter/starting diameter)x100.

So firing gives (.257-.251/.251)100 = 2.4%
Sizing gives 4.7%
Expanding gives 1.6%
Seating gives .8%

Add it all up and you get 9.5%. Per firing cycle.

A side note is that you get a low critical temperature at between 35 and 45% cold work. You also get grain refinement during rechrystallization.

As always YMMV.

 

[mikecr]

Should the annealing process time and temp (or AMP program) be different for the two conditions?

This is the rocket science behind an induction system. Change anything, and you better be familiar enough with that. Same with flame heating.. You can very quickly go to a wrong temperature.

The beauty of dipping is you don't have to know anything other than the medium is at 750-850degF.
No timing to it, regardless of all possible situations, because the time it takes to dip is more than sufficient, and you can't dip too long at the right temperature.

 

[Fred Bohl]

The beauty of dipping is you don't have to know anything other than the medium is at 750-850degF.
No timing to it, regardless of all possible situations, because the time it takes to dip is more than sufficient, and you can't dip too long at the right temperature.

I did not find your assertion to be true in my testing but then again YMMV.

Joe,

I read Damon's fine paper back in May of 2013 and occasionally re-read it as a reminder.

Riflewoman,

Your example is a reminder to me of why I've done only neck sizing and tried for neck IDs of 0.001 to 0.002 under bullet diameter in my benchrest rifles for the last 59 years. During my 6PPC days I even turned necks for tight-neck chambers. In my old age, I've switched to 6mmBR with no-turn but still fairly tight chambers and routinely get 30+ reloads and as many as 44 reloads without annealing. My dabbling with annealing has been one of many things tried in my pursuit of the elusive one-hole-group.

 

[mikecr]

Name an attribute that anyone could possibly screw up with salt bath dipping.
I just can't see it

 

[dcali]

Depending on how long the transition takes, you could overshoot a half-hard state with any method. If you don’t care about that then any method will work. If you do, timing is important and will be temperature (and method) dependent. The transition is pretty sudden whether you vary temperature or time. Most charts show temperature vs hardness, keeping time constant. But they have a similar shape when you plot time vs hardness while holding temperature constant.

Lead and salt dipping are nice because they provide a very consistent temperature, but they don’t change the inherent behavior of Annealing.

 

[mikecr]

So then it comes down to this;
How long does it take for proper recrystallization(process anneal) with brass that's immediately taken to ~800degF +/- 50deg (inside and out)?
And how long under this condition does it take to go beyond recrystallization, into grain growth(full anneal)?

I may be wrong, but I think it takes but a few seconds, and no amount of additional time matters under that condition.

 

[dcali]

It all really depends on how long it takes to get there. You have to keep in mind that most of the materials data out there is for one hour annealing time and (if memory serves) 1/2 inch thick test pieces. Temperatures and times are going to change for case necks to be done in a few seconds. How fast the transition from full hard to half hard to annealed happens is not known to me for every temperature. That data is tough to find and hard to verify. AMP’s tests (on their website) are the best I’ve seen but it’s very specific to their machine and they don’t mention temperature. Generally, the data I have seen has not been linear with time- there is a fairly dramatic drop in hardness over a relatively small time period.

 

[mikecr]

AMP's system is not based on temperature because they don't know the temperatures, and neither do any of the flamed timers.
These systems do not apply steady state correct temperatures, but transient temperature timing, as their process dictates.
And then they look at outer surface grain(under a microscope) and press against the outer surface for hardness, neither showing inner, nor middle, or showing hoop forces, -because they have no means of measuring this (neither do we).

It's my contention that your best bet is to go right to a known good temperature, for any amount of time, and see for yourself that it works just fine. No rocket science or 2nd mortgage needed.

 

[Fred Bohl]

The beauty of dipping is you don't have to know anything other than the medium is at 750-850degF.
No timing to it, regardless of all possible situations, because the time it takes to dip is more than sufficient, and you can't dip too long at the right temperature.

I did not find your assertion to be true in my testing but then again YMMV.

Mike, I was not intending to be sarcastic in my reply. To clarify, my data for the salt bath testing included some trials with the bath temperature at 750ºF that required a dip time of 4.5 to 5 sec. to meet my arbitrary test criterion (see post #28 above) and some trials with the bath temperature at 800ºF that required a dip time of 4 to 4.5 sec. to meet the same criterion. In both cases, longer dip times resulted in less spring back (too soft) and shorter dip times resulted in more spring back (too hard) per my arbitrary criterion.

Also thank you for posting the Metals Handbook graphics. That helps clear the cobwebs in the brain of this old retired engineer.

 

[Webster]

So then it comes down to this;
How long does it take for proper recrystallization(process anneal) with brass that's immediately taken to ~800degF +/- 50deg (inside and out)?
And how long under this condition does it take to go beyond recrystallization, into grain growth(full anneal)?

I may be wrong, but I think it takes but a few seconds, and no amount of additional time matters under that condition.
View attachment 1032659 View attachment 1032660

Chart 10-31 looks like a 1 hour at temp chart.