Annealing: Harbor Freight propane torch 3 burner + hose

[hounddawg]

@RC for grain structure changes even at 850F the time is in minutes not seconds. I have no idea much softening occurs in seconds, for sure no grain changes occur. Even the MATLAB charts give the data in 1 hour format not in seconds

and define what you want in an anneal, are you reducing the tensile strength of the metal at 750 but the grain structure is remaining the same. Stress fractures in the necks (neck splits) occur becasue of stress.

I can see softening the neck/shoulder area for case reforming but for the purpose of extending case life I would still be sceptical

 

[BBarn]

As has been pointed out earlier in the thread, the heating affect approximately follows the Arrhenius equation where the chemical reaction occurs twice as fast for every 10 deg. C increase in temperature. So an increase of about 160 deg. F in temperature would reduce the time required from 1 hour to about 7 seconds (as an example).

 

[hounddawg]

As has been pointed out earlier in the thread, the heating affect approximately follows the Arrhenius equation where the chemical reaction occurs twice as fast for every 10 deg. C increase in temperature. So an increase of about 160 deg. F in temperature would reduce the time required from 1 hour to about 7 seconds (as an example).

and has has been pointed out even at over 800 - 1300 degrees the time on all the charts is measured in minutes not seconds

I can see where the heat would soften the brass but I think it would be very hard to effect and grain structure changes without a computer and some high quality sensing hardware and software. ie an AMP

 

[BBarn]

Temperature/time required to change all the characteristics isn't the same. It's dependent on the condition and alloy of the brass, and whether you want to change mechanical properties or chemical properties.

 

[hounddawg]

Temperature/time required to change all the characteristics isn't the same. It's dependent on the condition and alloy of the brass, and whether you want to change mechanical properties or chemical properties.

All the references were studies performed on 70/30 cartridge brass and in one study actual .308 Lapua cases, anything else you missed

 

[Unclenick]

and has has been pointed out even at over 800 - 1300 degrees the time on all the charts is measured in minutes not seconds

The 15-second number did come from the 1945 study for the 43% CW brass, so they are not all in minutes. I suspect the more common tables of constant longer exposure numbers that you find on most plots are done to make it easier on the investigator. If you try to measure things in seconds, the time required to bring the sample up to that temperature becomes a more significant part of the total heat exposure time, compromising the precision of the result. If, for example, I want to observe the effect of one second at 1000°F, I might need three seconds to bring the sample to that temperature with really good thermal transfer contact or by induction and some of the recrystallization would then occur during the heating, making the effect of the 1000°F exposure difficult to isolate.

The practical reason we know the short times at higher temperatures work is that if they did not, the very brief annealing time of the AMP unit would not work, and those guys custom measure the results on a Vickers microhardness tester for each size case to determine what is needed. Not to mention all the folks using the molten metal and similar approaches who've stopped their necks from splitting over the years, Herrett and Milek's experience being and example.

 

[RC20]

and has has been pointed out even at over 800 - 1300 degrees the time on all the charts is measured in minutes not seconds

I can see where the heat would soften the brass but I think it would be very hard to effect and grain structure changes without a computer and some high quality sensing hardware and software. ie an AMP

First I am not a metallurgical engineer, I am a practical applications guy. If it works that is good enough for me.

You keep going back to one data set that looks to be out of context. As noted, what works is what counts. For a reasonable anneal for the neck and shoulder that is in the 750 deg area.

Amp does not do anything sensing wise as far as I can determine. Its all fixed output and time is the variable.

If you have oddball brass they need a sample to setup the machine for it.

That is why they put in brass mfg specific rather than just time, their machine can control the time down to (1/100?) I would guess if its that critical.

They may get the perfect anneal or as close as you can come. I don't but I make up for my lower control end (manual and to a tench) by doing it every 5 or so.

If AMP has output variability I assume the program that in for that machine.

So they may be able to spot it tighter and they clearly are faster than I am.

You should also keep in mind theorist is not the same as reality, not that the theory is wrong but the variables can change it from academic to reality.

That is why we test. I believe Jeephammer takes slices and adulterer them with chemical to see the grain change not just the vickers hardness.

Out of my expertise a long way. But it goes back to Unclenick, you have repeatability in a process and you can take that to the bank. As long as it meets the goal you really don't care about the rest.

JH and I both found that the paint on Templiaq does not work for seeing what happens with Induction. It interferes with it and you get wrong data.

Not the end of the world, I used Crayon, I forget how he did it. Mine obviously has noise with the crayon but its closer. Keeping in mind all I want is close.

The paint on can be used to ensure lower down the case does not heat up.

I layer my situation with a dark room and watching in the neck of the cases to ensure there is no glow.

ANNIE warns you the unit changes as its used so you run 10 rounds through before settling in. I do that and cross check for glow.

While it sounds cumbersome, its not, its fast and I have the base data in time written down so I have a starting point each run. I do my test and adjust up or down as it goes along. I also carefully watch the color blush change. If I have none that is fine. If I have a bit I am ok but watch it closely. A lot and I back off.

I don't use it as the sole indicator, that is the wrong way as judging color and changes is not that good with people in a fast process. If I get blow before color change I back off. The brass is variable, glow is not, once it glows its too much.

The glow come first inside. So it may not be too much overall but I simply keep away from glow and it works as I likely get 70% and with a repeat 5 rounds fired latter I get it back again and never loose it.

My brother found that the times changed quite a bit if he did a Wet case clean. Makes sense as there is no carbon inside (I believe his takes longer )

An added check is does it polish off by polish point 3? (the slight or significant shift to silver). If so its another indicator.

I guess I need to test a for sure orange glow one, I believe it never polishes off to shiny brass once you go to far. That is experimental but I do know that factory that does not polish off the neck and shoulder anneal does polish off in 2 to 3 tumbling's.

Much like the chemistry of a Nuke plant, you constantly check it and adjust it if its not doing what it should.

 

[hounddawg]

Ok several points here

I am not going with a single study that is taken out of context, there are three or four separate studies posted in this thread and I am sure I could find more. All studies I have posted from the MATLAB number to the doctoral dissertations use minutes and hours list treatment times in minutes and all use 70/30 cartridge brass for the studies and in some studies actual fired cases

second point is that it is obvious that the longer you heat the object the more change that is affected on the metal. I may look at the data later and see if it is linear or logarithmic but time seems to be just as much a factor as the temp, the hotter the metal becomes the faster the anneal occurs

@RC in regards to the AMP, the original operated off set parameters of time and temp, the AMP II has AZTEC software where you use one or two sacrificial cases and the machine determines time and temp for the anneal

https://www.ampannealing.com/aztec-settings/

The case will have heated right up to melting point. Depending on the case shape, occasionally they can stick a little in the pilot. Note: ANALYSE mode will not harm your annealer. Make sure to remove the case after the red light goes out and eject it from the shell holder. Otherwise heat will transfer to the brass shell holder grip, which may become uncomfortably hot.

Once the red Start light goes out the display will show a four-digit Code number, and "USE” will be pulsing.

WRITE THIS NUMBER DOWN
Note: the annealing code generated is specific to the brand, lot number and neck wall dimensions of the sacrificial case. If you wish to anneal a different brand or lot number of the same cartridge, you must analyse another case for that batch i.e. Lapua 308W will be different from Remington 308W.

since this is simply a software not a hardware update that tells me that even the original AMP must have sensors that can analyze the brass

Now using some common sense I believe the AMP takes the case far in excess of 750 or even 1000F for very precise amounts of time to get the crystals to realign in just a second or two of treatment time. Something impossible to do with any degree of effectiveness using torches, repurposed induction heaters, molten salts etc. You might as well be holding the case neck in a candle flame till it gets too hot to hold like the bench rest crowd did in the 80's and early 90's as to use anything less than AMP style technology in my opinion.

To change my mind I would need to see some hard data with laboratory tested data not opinions.

Just my 2 cents worth but if I thought it was cost effective to anneal I would buy a AMP but since my cases only cost me a dime or so per firing it would take me a lifetime or so to recoup my investment in a AMP. If I used some esoteric case like .257 Roberts or even a wildcat where I would need to do some case reforming I would buy one. Edit - or maybe I would go back to the drill and socket routine and just heat the necks to a dull red in a darkroom, at least that way I would be softening the neck/shoulder a bit. At 700 - 750F for a second or so I doubt any significant changes are occuring

edit 2 - on using color

https://www.hearth.com/talk/wiki/know-temperature-when-metal-glows-red/
It doesn't really matter what the emitter is...stainless steel, cast iron, tungsten in your light bulb, the temps are about the same for a given color. Generally accepted colors/temps are:


::C ----- ::F ---- Color

400 -- 752 -- Red heat, visible in the dark
474 -- 885 -- Red heat, visible in the twilight
525 -- 975 -- Red heat, visible in the daylight
581 -- 1077 - Red heat, visible in the sunlight
700 -- 1292 - Dark red
800 -- 1472 - Dull cherry-red
900 -- 1652 - Cherry-red
1000 - 1832 - Bright cherry-red
1100 - 2012 - Orange-red

edit three - AMP has added a few more articles on thier website including articles on wildcatting, case weight sorting and salt bath annealing. Here is a quote from the salt bath article

https://www.ampannealing.com/articles/52/salt-bath-annealing--does-it-work-/

1. At best, all case necks were only partially annealed. Even using 550°C for 8 seconds, no case necks tested annealed softer than 120 HV, which is much harder than desirable.

2. Even when an insertion time of 20 seconds @ 550°C was used, necks were not correctly annealed. The case bodies, however, were over-annealed.

550C = 1022F

if you want to discuss annealing with the AMP people they are posting and replying to the article on this thread

https://www.longrangehunting.com/threads/salt-bath-annealing-doesnt-work-by-amp.220436/

Like I have said many times already I just don't seen any value in annealing, benchrest shooters got fantastic groups both short and long range and wildcatting cartridges was being done long before home annealing was ever being done. If people want to dump money and time into doing it , no problem with me but I don't like seeing people being conned into wasting their money on snake oil and unicorn poop. You may get a few more firings from a case using an AMP but otherwise basement amateur annealing using most methods is worthless and/or potentially dangerous from what I have learned from reading every scientific study I have found on the subject

 

[Unclenick]

Ambient light is what makes glow an unreliable quantity. The human eye's cone cells stop function in low visible light. Low visible light is detected by the rod cells only, and their spectral sensitivity peaks at shorter wavelengths than the cone cells do, so low light looks bluer than it would if it were brighter.

Board member Mete, who is a retired metallurgist, pointed out in a past post that grain growth takes time to occur. So, theoretically, if you could keep it brief enough, you could go up just short of the melting point and not cause grain growth if you could raise and drop the temperature fast enough. The problem is, there is no way to do that in a case neck, as heat retained by its own mass would extend the exposure, at least at the core, for too long.

I can't recall why, but something in the information about the AMP unit gave me the impression it has an infra-red temperature sensor watching the surface of the brass. If that is true, then, given these detectors have their exact reading values affected by the emissivity of the surface, it follows that a highly polished piece of brass (lower emissivity) might receive different time exposure than a dull, carbonized piece of the same brass does due to the temperature measuring error the difference in emissivity introduces. The actual temperature reached will go up with increased exposure, regardless of what the sensor says, so surface color and polish would be a potential source of error in the unit if I am right about the IR thermometer. Would love to take one apart.

The criticism of the AMP machine will depend on the definition of "properly annealed". A .308 case splitting in a chamber is being elongated perhaps 3% between the narrowest diameter given to it by the die before the expander goes through and the diameter it ends up with after expanding into a chamber. A 75% CW neck may only tolerate 1–2% elongation before breaking, but you only have to increase that enough to stop splitting and remain that way through ten or so resizing and firing cycles. That does not require anything like complete annealing. So, again, everything hinges on how "properly" is defined; 5% anneal, 20% or whatever.

 

[hounddawg]

Any thought on the temperatures needed to get any changes in hardness Nick ? To me it does not look like 750F would be of any benefit. My guess is the AMP takes temps well above 1000 F to get the changes needed to change the hardness of the necks

The AMP article on case weight affecting the anneal also has some interesting bits of info

 

[Unclenick]

The Kline study has a plot for time required to initiate recrystallization vs. temperature for both 50% and 75% CW brass. The time threshold varies exponentially, roughly halving every 33K for the 75% CW, for which the function made a nice straight line when the time axis is on a logarithmic scale, but not being such a simple function for the 50% CW brass. It gives about 6 seconds (0.1 minutes on the graph) for about 833K, or about 535°C, which is 995°F.

However, other charts he has show hardness starting to change before recrystallization actually commences. Even 100°C cause some change in the yield point over a period of a week. And, get this, the hardness and yield bottom in some plots at about 30 minutes exposure, and goes up again when it is drawn out to 60 minutes. This is explained by an accumulation along the enlarged grain boundaries of, IIRC, nucleation points. I'll have to reread that.

In any event, this is not nice neat all-else-being-equal parallel predictable behavior.

 

[hounddawg]

Nick the 550C number for 8 seconds is what I was referring to

At best, all case necks were only partially annealed. Even using 550°C for 8 seconds, no case necks tested annealed softer than 120 HV, which is much harder than desirable.

bear in mind that these tests were performed by METLAB, an independent lab service not the AMP company.

Do you still think that at 750F for a second or two will do anything at all in regard to case life or accuracy? That of course is the elephant in the room, it appears that unless you are willing to drop $1500 or whatever to get the AMP that anyone using a Annie/torch and 750 Templaq on the neck area is just wasting their time. I would suggest instead that people still wanting to use a torch or Annie use 450F Templaq 1/4 inch below the shoulder on the case body for safety reasons and not worry if they get above 750 or even 1000 F in the neck/shoulder area. Indeed it looks like the old plumbers torch till the necks glow dull red might just give the most cost effective anneal of all

 

[Unclenick]

METLAB's claim that 120 HV is way too high tells me they just aren't used to the cartridge case field. About this time two years ago I sent samples of 10 different headstamps of new bulk cases to board member Bobcat45 who worked professionally in making hardness measurements and metallographs for a major mining company, including having access to a micro Vickers hardness tester. The results for sectioned necks ranged from HV 96 (RP) to HV 124 (Hornady) with the average at VH 107 with an SD of HV 7.7. So the AMP result is just under 1.7 standard deviations above the mean and is not at all outside the normal range for new commercial brass at all.

Even if we think the commercial brass makers have got it all wrong for most of the last century, from the ASTM we have that HV 120 is about equal to Rockwell B scale 67. From Klein's paper, at 425°C, page 14, Table 4.11, we can see that a 50% CW sample drops from RB87 to RB67 in about 4 minutes. From Page 11, Figure 4.2, we see the elongation before break of a 50% CW sample in 425°C grows to about 30% in that same time. Therefore, it is easy to see it provides 10 times the elongation necessary to someone reloading .308 Win in conventional steps to prevent neck splitting.


Tested virgin brass

Federal
Hornady
IMI
Lapua
LC
Norma
RP
SIG
Starline
Winchester

From Bobcat45's images, the top is Hornady and the bottom is RP. RP is close to being 80:20 brass, which accounts for some of the appearance difference.

From Bobcat45:

"Note that the Vickers hardness is denoted HV500 for a 500 gram load, and cand also be called DPH for Diamond Penetrator Hardness.

The value is sensitive to measurement of the diagonals of the indent.

The formula is: HV = (1854.4 * P) / d^2 where P is the load, in grams, and d is the average of the diagonals in micrometers."​

For a time, aerodynamics were famously able to prove bumblebees couldn't fly. I don't think we stopped believing they could fly because of that. We accepted that they could fly and that we just weren't understanding something and finally some graduate students solved it, IIRC. We also know lower temperature stress relief works by observation (Herrett and Milek again and this board's members stopping splits, even with the candle method) so the question is why, and not if. What are we missing?

One of Klein's tables shows elongation at break increasing occurring at just 100°C over a period of a week. There is no recrystallization happening at that low temperature so there is some other mechanism for releasing some of the stress before recrystallization is clearly established. I don't see an explanation or discussion of that anywhere. It would be an interesting thing for a grad student to undertake and quantify.

 

[RC20]

@RC in regards to the AMP, the original operated off set parameters of time and temp, the AMP II has AZTEC software where you use one or two sacrificial cases and the machine determines time and temp for the anneal

I had not seen the latest, if I had to guess there is cross check with reality then in the sensor such that if it exceeds certain parameters it cuts it off.

It gets pretty deep into safe and a variables (ask Boeing what a bad sensor can do)

Agreed its the best, disagree its the only one that works per inductive.

 

[hounddawg]

RC my only question about the Annie is how do you know what the hardness of the necks are after your operation. From every thing I am reading and there are pages of links now supporting this but heating your case neck to 750 for a couple of seconds accomplishes nothing except a color change

For that matter according to METLAB heating to 1050F for 8 seconds accomplishes nothing except a color change

@Nick As far as what is acceptable hardness I could care less. It's a simple question - does keeping a case at 750F for 7 seconds accomplish anything as far as reducing brass hardness?

No sources that I can find says it does so if the report you keep referencing does I would like to read it. So if you can provide a link to the Kline report I would appreciate it. I tried googling it with no results and if I have ever looked at it the name does not jump out at me.

edit nevermind Nick found it behind a paywall but if it does say that appreciable change can occur that soon is everyone else wrong ? many reputable sources now state that 750 - 1000 degrees takes minutes to make any significant changes

 

[Unclenick]

Doesn't much matter what the sources say if they aren't checking what we are, which is do neck splits stop? They do. It's the explanation we are seeking and not the authoritative data that fails to explain it.

 

[hounddawg]

well Nick you may want to consider a new career as politician, lol

BTW for years now I have been lobbying for some real tests conducted under a controlled scientific study on case life and velocity for annealed vs non annealed brass. No annealing machine manufacturer has taken me up on this yet or if they did conduct a real study were too embarrassed to publish it. Most likely the latter since no one would want to say here buy our $300/$500/$1000 machine it will make a pretty blue ring on the necks of your cases

I have linked several studies that show that at the temperatures and times that have we have been using no grain structure changes occur. Without the grain structure changes stress fracture will continue to occur. I have no idea why you will not give a direct answer on this since it is the only logical conclusion

If I were really concerned with case life I would go to the 450 templaq on the case body 1/4 or 3/8 inch below the shoulder method. Heating the case neck to dull red is not going to harm that brass in the least and might even get it to the temps where some changes might actually occur. Or I would drop 1500 on a AMP that does get the brass hot enough and does it in a very controlled way and actually does anneal the brass. Expensive option but it seems to do what it claims which is get the grain structure back to the annealed state. That is more than I can say for the other options

 

[RC20]

RC my only question about the Annie is how do you know what the hardness of the necks are after your operation. From every thing I am reading and there are pages of links now supporting this but heating your case neck to 750 for a couple of seconds accomplishes nothing except a color change

HD: I can only conclude that you simply do not want to get it or let go of a notion that has lodged in your head that its impossible and then do the confirmation bias thing.

I have repeatedly laid out my methodology - I think anyone would agree is comprehensive as can be done by an individual without the unlimited resources.

It does not depend on a single source and it has rubber meets the road real world cross check.

1. No split necks or shoulders
2. Bullets seat with minimal force when they got harder and harder before (duh, its work hardened brass!!!!)

If it Quacks like a Duck, has Duck DNA, has Duck Feet, Swims like a Duck, flies like a Duck and produces Duck offspring we can conclude its a Duck.

Clearly you want to deny that ducks exist.

In one last effort I will present the following.

A number of times in my career, I came across a circuit that quit working and had worked fine before.

When I looked at the logic, I could not determine why it worked in the first place.

I do know how those circuits are SUPPOSED to be wired and they were not. That standard wiring had a logic that worked every time.

So I re-wired it to a standard logic circuit and never had another problem.

I don't have to understand why it used to work, I just need to wire it to where
1. It worked
2. Anyone who followed me could find a failure in that circuit because it worked the way it should.

There are a bazillion things in the world none of us understands. They are still arguing about electron flow.

Does it matter? No, all you need to know is IT WORKS and how to make it work consistently.

People have done so with various methods for a long time. Hell I can heat the damned case to dull red (I have done so) what I have is a messed up piece of brass as it nor longer is smooth its mucked up skin mottled that will not work in a die right.

I DO NOT CARE IF IT MAKES BETTER ACCURACY!!!!~ Cases cost 20 cents each and if I quadruple the life of the case the Anneal will pay for itself and I don't have to keep buying cases all the time.

When the Sun rises in the East and not the West and your facts say it rises in the West then you need to check your facts or accept you don't understand it.

 

[hounddawg]

HD: I can only conclude that you simply do not want to get it or let go of a notion that has lodged in your head that its impossible and then do the confirmation bias thing.

I have repeatedly laid out my methodology - I think anyone would agree is comprehensive as can be done by an individual without the unlimited resources.

RC you seem to be the one having problems with accepting that 7 seconds at 750F is doing little or nothing as far as reducing the brass hardness. I have posted numerous scientific studies showing that anneal times at that temperature are in minutes nots seconds and you have posted absolutely no studies showing otherwise. If anyone is in denial here it is you. I am simply looking for scientific proof other than gun blog/forum wisdom from guys with zero metallurgical or thermodynamic education. Call me the odd duck but I prefer scientific testing over hearsay and old wives tales

Your proof seems to be that you don't split case necks..ok I have never split a case neck either does that prove anything? I don't think so.

As far as my bias I am going to do one last experiment. I have just sized and trimmed 50 Lapua .260 rem cases on their 5th reload. I am going to "old school" anneal 15 of them ie the drill and torch routine, the Annealeze needs new wheels. Then I will take those 15 and 15 which have never been annealed and load them exactly the same. I will seat the bullets on the arbor press using a bit of graphite lube and shoot 3 each groups at 600 yards across a chrono with the annealed and non annealed and see if there is any difference in SD which is my main concern. Like I have said thousands of cases and no neck splits. If I do see a difference however I will get some new wheels for the Anealeeze, or just keep using the drill and socket method. I don't expect much difference though the attachment are is the same cases I am cleaning now and this was the last time they were shot. Most of the reason for the test test is just a 600 yard sight in and seeing if I need to back off a grain due to higher temps. It will be 90 - 100 on the range and the attachment was shot when it was in the 40's

Another interesting test would be if a annealing manufacturer or someone with the resources would take 100 or so .308's and divide them into three groups. Shoot 1/3 through a military rifle, 1/3rd through a off the shelf hunting rifle, and 1/3 through a match barrel. I shoot match barrels and like I said, I have never had a neck split and it could be like Nick said, that is due to the tighter tolerances of my chambers or the fact I use premium brass. Military rifles and Walmart rifles might see more neck splitting than I get. Winchester or LC cases may be more prone to neck splitting, I have no clue but annealing for case life is not a concern for me

On a different note I know why the salt bath annealing did little or nothing.

https://www.ampannealing.com/articles/52/salt-bath-annealing--does-it-work

Our additional findings confirm the initial report tests. Even at 550°C x 8 seconds or more, insufficient temperature was applied to the necks of all bottle-necked cases we tested by salt bath heat treatment for correct annealing to take place. In most cases, there was very little or no annealing to the necks. With the 308 cases tested in Chart 6, there was also virtually no annealing at all of the shoulders and bodies inserted for 5 or 8 seconds @550°C. This is confirmed by both hardness testing and examining grain structure.

The author was on the right track when he suspected that the case body was acting like a heat sink. One of the laws of thermodynamics is that the heat always flows from hot to cold and the greater the differential temperature between the two the faster the flow. As the temperatures approach equilibrium the transfer slows. A propane flame burns at about 3500 degrees or appx 3.5 X as hot as the salt bath so the neck shoulder area will be heating faster than the case body draws the heat away and the flame s temperature is much higher than the target temperature so equilibrium is never reached.

A induction unit operates entirely differently since the heat is generated by the metal itself which is the most efficient of the three methods and would be the least prone to heat migration assuming the magnetic field is "tight" enough to generate sufficient heat in the neck shoulder area without overheating the case body.

 

[BBarn]

First the obvious...

The "simple" methods of annealing yield an increase in ductility.


And a few things are fairly clear from the studies...

Increases in ductility begin at lower temperatures than recrystallization, and recrystallization begins at lower temperatures than grain growth.

To reduce case cracks an increase in ductility is needed and recrystallization is desired, but grain growth is undesirable since the brass is more likely to develop cracks when grain size increases.

The temperature versus ductility curves are relatively steep and getting a specific ductility is challenging and probably requires testing the results and knowing the history (%CW) of the brass.

Excessive annealing (too hot for too long) will produce brass that's very ductile but grain size will be excessive.


What's not clear is how (without expensive equipment or testing) to apply minutes/hours long anneal study results to a process that lasts only seconds (to allow annealing the neck and retaining body strength). It obviously is being done successfully by the case manufacturers.


As one who only anneals to reduce neck splits, I'll continue with the drill/socket/torch method since it works for that purpose. I'll probably consider reducing the flame and time somewhat so as to avoid grain growth. At the other end, I can sorta tell if the ductility is not sufficiently high when the neck expander is drawn through the neck during sizing.