6mmbr's annealing temperatures seem questionable...

[kombayotch]

I think most under anneal their brass. I've compared polished then annealed brass at 800F to new Lapua annealed brass for look. The Lapua stuff appears to have seen much higher temperature or long time than my few seconds at 800F.

I agree that most probably do under anneal. One thing I find fascinating is that people are undertaking an exercise that's meant to restore the ductility of the brass, but aren't doing any kind of test on the brass afterwards that actually verifies that their goal has been accomplished. If you're goal is to restore ductility, you need to test that in some way (ex: test above), and not just look at a color or follow some directions and hope or assume that you have.

It's similar to setting up a FL resizing die: if you follow the manufacturer's directions, sure it'll chamber and go bang. However, there's a good chance that you're over sizing the brass and setting the shoulder back too far. Under sizing, like over annealing, is easy to detect. But, over sizing and under-annealing probably aren't going to get noticed if you aren't measuring for it until its too late and the end result is less brass life.

You may be doing it just fine, but unless you actually measure the thing you're trying to affect, you really don't know for sure.

 

[Ron AKA]

The problem with the brass neck and shoulder is that it so thin, getting a legit hardness test is difficult. I'm sure a company that does NDT commercially will have equipment to do it, but it is likely to be expensive. You could also destructively examine the grain structure under a microscope, but again unless you know somebody that can do you a favour, likely very expensive. I've played around with various ways of denting or bending the brass on junk cases before and after annealing. While you get a feel for how soft it is, it is kind of hard to put a number on it.

My research on the subject has lead me to believe that 3-5 seconds with the temperature a bit north of 800 does the job, so the quality control is ensuring the temperature gets there for that amount of time. That is primarily how heat treating and stress relieving is done in industry - time temperature. Quality control documentation typically is a copy of the time temperature recording.

 

[kombayotch]

In industry, they are actually controlling the temperature of the torch precisely. Most people doing it by hand likely aren't, and on many of the torches, how it is adjusted can make significant difference the heating time. I noticed that a lot playing with the old set of torches, and a lot more after switching them to the swirl flame torches.

 

[Suputin]

Where did you find swirl flame torches?

 

[kombayotch]

I got them off of ebay, out of the US, two BenzOmatic JT539. These are really nice because they also have a regulator, and not just a valve. I have the fitting on the way to connect them to a 20 lbs. BBQ tank also.

I was also given this idea:

If you block the air holes on a regular torch, you don't get the sharp cone. You get the blue flame in more of a wash, which is more like the swirl flame. However, this won't be as hot as the swirl flame and there are still "layers" of blue color. The swirl flame is a uniform blue, consistent in temperature throughout and wider.

 

[Cocked&Locked]

why not get an IR thermometer? point and get temp accurate to +/- couple of degrees at that temp range? get a good one from an electrical instrument supplier,

 

[Suputin]

I have the fitting on the way to connect them to a 20 lbs. BBQ tank also.

Where did you find this and what is it called?

 

[kombayotch]

The IR thermometer is too slow to keeps up. Like with the Tempilaq, you've already gone past the temperature it is telling you.

I think I've settled on using the spring back of the brass itself as the indicator. Choosing a random exposure time or basing it on something like Tempilaq isn't reliable. The exposure time required is going to vary based on the torch setting, the type of torch and the number of torches being used. You are never holding the case at a fixed temperature. Your torch is hotter than the target temperature, so the case's temperature is always climbing when its exposed to the flame. The rate at which it climbs depends on the stuff just listed. The best indicator of successful annealing is the property you're trying to affect and that comes from the material itself.

I'm going to do some more work on setting up the flame temperature to be the same each time. Then, the exposure time will also be the same each time once I figure out what it needs to be using the spring back test. The IR thermometer will work for that.

I ordered the 'Y' from here:
http://www.bbqguys.com/item_name_Bayou-Classic-Propane-Y-Splitter_path_11428-9786_item_2708358.html

You can find them at RV dealerships in Canada, but the cheapest I found them for was $90.

The intermediate hoses are available from Home Depot for $16 each.

 

[BattleRife]

The number shown for annealing in your reference doesn't agree with data published by the American Society for Metals (ASM):

But it is worth noting that any published number on annealing temperatures is going to be based on a long soak, probably 1 hour, precisely to avoid the transient conditions you discuss in your opening post. Since brass cases are not annealed under those conditions both the ASM graph I show and the table you provided are not useful. I have been way too busy to finish my work intended to find, once and for all, some concrete evidence to support an ideal target temperature for case annealing, but I now know that it will in fact, be somewhere right around 400C.

But it really isn't that important. As Mysticplayer said, it really is all about consistency, especially if, like him, your goal in annealing is to improve accuracy. It then doesn't matter if your temperature is 450C on the neck OD but only 400C on the ID, so long as every case is done the same.

Somebody asked about what temperature brass starts to glow, all metals luminesce at about the same point, in the 700-750C range, much hotter than needed for annealing, but the idea that it is too hot is not proven, either.

Somebody else asked about the temperatures being achieved during annealing, based on the depth of colour tint on factory brass. Well, people might find this a useful reference (all temperatures in Celsius):

And one final note, for kombayotch, is that you shouldn't be surprised to find conflicting and inconsistent information on the topic of home annealing of brass cases. The one thing that I have emphasized over and over in these threads is that the understanding of what we are doing when we anneal and why we are doing it just isn't there. Virtually every loading manual, magazine article and internet blog on the topic is based on old wives' tales. Just how flawed the common knowledge is can be demonstrated with one sentence:

The whole notion that brass case failures are due to cold work embrittling the brass is a myth.

Yep, that's right. The explanation of the very basic "why" of annealing, one that has gone around shooting circles for decades before the internet was ever invented, is complete bulls**t. That's how completely messed up the common understanding of annealing is. I'm not saying it's useless, just misunderstood.

I'll finish off by showing some images of cracks in brass case necks. These are cases that cracked in regular use, in my AR-15s and M14s, and I prepped the cases (which is why the prep is frankly, pretty bad) and took the photomicrographs. The cause of these cracks is clearly not cold work, because the brass crystals are still plainly intact.

Case stamped FC 223 REM, unknown number of firings:

Winchester White Box 7.62x51, case fired 3 times.

 

[kombayotch]

Interesting... That particular chart came from http://www.matweb.com/, but was similar to a textbook I checked

I would agree wholeheartedly about a lot of shooting information out there being based on old wive's tales, and not just annealing. Which is why I like to see data, or concrete sources of info. Keep posting it!

What are your thoughts on testing the spring back of the case necks as a means of determining if the temperature has been applied hot enough, for long enough?

Can you post a snapshot of the textbook for that luminance number?

 

[kombayotch]

Also, curious about how the heat was applied to the cases in your photo, and how you concluded what temperatures were hit since there was that transient condition happening with the torches.

I do note that your 400C sample shows none of the color that people often look for. I have noted that it never appears on my cases either if I only anneal to the point where the spring back is reduced to the level of new cases unless I heat very slowly.

 

[kombayotch]

While your photomicrographs do show that brass that has not been work hardened can crack, they do not prove that cracking cannot occur from work hardening. I am curious about how you arrived at that conclusion.

That is easy enough to test though. Separate a bag of brass into two groups; one that never gets annealed (control group) and one that gets annealed after each firing. Load and shoot them the same way, in a rifle with a generous neck and note the mean life of each group. I've never cared to examine this because my cases either get discarded due to loose primer pockets or head separation signs. Neck cracking is quite rare. My annealing has been strictly to restore consistency in neck tension and to lower runouts, which I have observed it to do.

Incidentally, the source of BR's graph, for anyone interested:
http://web.nchu.edu.tw/~jillc/me/Ch10- Kinetics - Heat Treatment.pdf

.

 

[sunray]

"...glowing red hot..." That's too hot. Makes 'em too soft. Heat 'em with a regular propane torch until the brass changes colour and tip 'em.
Worrying about the actual temperature is over complicating it.

 

[Ron AKA]

Here is a reference to a metal colour vs temperature. I'm still convinced if you want to anneal in a few seconds then you need to get to 800 F, which is a dull glow in a darkened room.

http://hearth.com/econtent/index.php/wiki/Temperature_when_metal_glows_red

As far as annealing preventing cracking, I'm convinced it does, if done early enough. Yes, low cycle fatigue is the cause of case cracking, and it progresses around the grain boundaries, but I still think annealing before the fatigue damage starts can prevent it.

And on the question of how much to anneal, I'm convinced it is going to be extremely difficult to partially anneal brass on any kind of consistent basis. If one does this then, I would suggest the result will be very inconsistent neck tensions, compared to cases that have not been annealed at all.

Darn, all this discussion is going to force me to unpack my metallurgical text books...

 

[BattleRife]

Also, curious about how the heat was applied to the cases in your photo, and how you concluded what temperatures were hit since there was that transient condition happening with the torches.

I have done my annealing experiments using a bath of molten salt. The heat transfer rate from a liquid to a thin-section solid should be very high, and my guess is the metal reaches the bath temperature within 2-3 seconds. I soak for 5 seconds. The temperature of the bath itself is monitored with a thermocouple and I have very good control, which is why I do it that way.

While your photomicrographs do show that brass that has not been work hardened can crack, they do not prove that cracking cannot occur from work hardening. I am curious about how you arrived at that conclusion.

I've never cared to examine this because my cases either get discarded due to loose primer pockets or head separation signs. Neck cracking is quite rare.

I did not exactly conclude that brass cannot crack due to work hardening, what I have concluded is that it takes so long as to be an unrealistic mode for the vast majority of reloaders. It would require dozens, perhaps hundreds, of reloads of a case to crack it due to work hardening alone. This may be achievable in a .38 Special or .45 ACP, but any high-pressure, bottle necked round is going to fail in any one of a bunch of ways well in advance of that.

Your final comment sheds some light on why it is going to be so hard to narrow this stuff down: the reloader himself is a big variable. In 20 years of reloading, I have discarded maybe three cases due to expanded primer pockets. The vast majority (probably well over 90%) of my case failures are due to neck cracking. My cases, loaded on my equipment, shot in my guns and stored in my house, have a very different life cycle than yours. I would have to have a lot of metallography volunteers to characterize everything that's out there.

 

[BattleRife]

I'm still convinced if you want to anneal in a few seconds then you need to get to 800 F, which is a dull glow in a darkened room.

800 dinosaur degrees is 425C, so we are in rough agreement as to what temperature we are trying to obtain. My position, which I didn't explicitly state, is that we cannot necessarily judge what temperature we need based on recrystallization temperature of the brass, because deformation of the crystal structure is not the issue.

As far as annealing preventing cracking, I'm convinced it does, if done early enough. Yes, low cycle fatigue is the cause of case cracking, and it progresses around the grain boundaries, but I still think annealing before the fatigue damage starts can prevent it.

Oh, I think it does to, just not for the reasons that everyone thinks it does. For what it's worth, fatigue typically goes through grains, not around them, and it would be a pretty big stretch to classify 3 firings as fatigue, even the low-cycle variety. I have looked at maybe eight cracked cases from my guns over the last two years, and none had any features that I would call fatigue.

 

[kombayotch]

Your final comment sheds some light on why it is going to be so hard to narrow this stuff down: the reloader himself is a big variable. In 20 years of reloading, I have discarded maybe three cases due to expanded primer pockets. The vast majority (probably well over 90%) of my case failures are due to neck cracking. My cases, loaded on my equipment, shot in my guns and stored in my house, have a very different life cycle than yours. I would have to have a lot of metallography volunteers to characterize everything that's out there.

Agreed, several benchrest shooters have told me that they never anneal, and don't need to. However, the neck clearance of their chamber is nearly null. Far different from a factory or custom tactical rifle.

 

[kombayotch]

What affect does your potassium nitrate bath have on the brass? Guessing that was chosen since it would not react chemically with the brass...

Is it easy to clean off?

 

[Ron AKA]

For what it's worth, fatigue typically goes through grains, not around them, and it would be a pretty big stretch to classify 3 firings as fatigue, even the low-cycle variety. I have looked at maybe eight cracked cases from my guns over the last two years, and none had any features that I would call fatigue.

Just goes to show I should have done a bit more research before speaking. I have reviewed many failure reports over the years, and my memory is obviously not what it used to be... I guess I was thinking of stress corrosion cracking when I say inter-granular.

Brass SCC

So could cartridge brass be cracking from SCC? Ammonia is commonly used solvent for barrel cleaning and certainly will attack brass. There must also be chemicals left in the powder residue. The stress in SCC would seem to have to be long lasting -- not just a few cycles. So, that would make it the residual stress from the cold forming and firing. Annealing would eliminate that residual stress and control case cracking. However one thing does not add up. If this is the sole explanation then the cases that I have sitting in a box for 50 years, fired about 20 times without annealing should be cracking in the box, especially the loaded ones that are stressed further by the seated bullet.

So my thoughts are that there is more to it than just simple SCC. Another possible effect is corrosion fatigue. It can be trans or inter-granular and is a combination of low cycle fatigue and corrosion. It is like a combination of slow corrosion over time in the presence of residual stress (almost like SCC), with periodic high stress events(firing?) that open up the corrosion cracks, and set up the next slow corrosion cycle. Again if you eliminated the residual stress due to annealing this may interrupt the cycle.

Interesting subject...

 

[scout]

Never did learn to like three phase diagrams ... I've been lurking on this thread and find it informative. My take on annealing was to run the brass through the recovery and recrystallization temperatures, but keep out of the active grain growth, which seems to be the whole point of this thread - how do you monitor it?

Kombayotch - have you tried cycling a thermocouple through the flame to see temperature profile and max temp reached? Or perhaps insert one inside a tight fitting piece of brass?