Brass Hardnes and Annealing Test

Discussion in 'Rifles, Bullets, Barrels & Ballistics' started by 4mesh063, Apr 8, 2003.

Help Support Long Range Hunting by donating:

  1. 4mesh063

    4mesh063 Well-Known Member

    Dec 8, 2002
    I just finished building an annealer and got a chance to use it yesterday. After a conversation today about it's use actually helping the accuracy potential of a gun, I decided to undertake a test tonight.

    I have a gun whose case is based on the 416 Rigby. It is shortened considerably. The 416's body is formed down to become the shoulder and neck of my case. About .800 is cut off. The hardness of that brass down at the body is greater than the neck of the parent case. I wanted to know how hard MY necks were as compared to the factory necks and if there was any difference in the 2, what could be done to replicate the factory results. Also, My theory on annealing necks seems to be considerably different than the opinions of 99% of people, so I thought I'd share what MY findings are on the subject.

    For the most part, I have tried to make the test data/conditions consistant. For lack of forsight a year ago, I cannot use 100% consistant cases. An additional test some time in the future could complete the test with 100% accuracy, but, that's for later.

    I have brand new cases that have never been fired. I have a couple of the pieces that were cut off the necks of the brand new cases. I cut up a piece of brass that has been fired at least 10 times, perhaps more. I have not kept track of the firings because I figures that these brass were scrap anyhow, in hindsight, they may work well. I have no use for # of firings so when they die, I will throw them out.

    Now, in addition to those 2 cases I have a case from my old barrel that very closely resembles the new, It has been fired 45+ times (and was primed for another go). This also got cut up.

    I started by taking the standard that comes with the tester and checking it. It Rockwelled at B72.5 the markings on it say it should be 73 +- 1.0 OK, Im happy with the tester.

    I started with the neck that was cut off the new case, obviously never fired. (pulled out of the chip pan in the lathe) It rockwelled at 40.0 twice. The neck from my 10 times fired case that was just run through my annealer came up with readings of 35, 29.5, 30, 26, 30. Some of the error can be attributed to the shape of the part in the tester. Putting a potato chip on the anvil, getting it to sit flat and getting the penetrator to start properly is difficult. In any case, one could say My new necks are considerably softer than the factory ones.

    The case neck from the 45 times fired case gave readings of 46.5, 44, 50, 45.5. Now, I cannot say for sure when the last time was that that piece of brass was annealed, but, when it DID get annealed, it spent a LOT more time in the torch than my new ones. It is a good case for work hardening being a reality. This piece of brass was also from a different lot than the others tested and as I pointed out above, sorry I didn't get them all from the same lot. I would make an educated guess that this brass is VERY similar to the brass I have now, but, that's just a guess.

    Also, keep in mind that as I was testting, I would move the part so as to show the range as I went around the neck. I can only get readings about every .100 anyhow, since the metal is distorted after testing. In general, throughout the entire time, readings seemed to repeat in certain areas of the brass leading me to beleive that change is not only from my part being shaped in such a way as to be difficult to test, but also from very real changes in hardness. What the changes relate to, I cannot say at this time. I would guess, thickness of the brass being inconsistant, and causing changes in temperature when in front of a torch is a good place to start.

    The body's of the cases were as you would guess, much harder.

    The unfired body

    On to the bodys. Now these are a lot easier to get in the tester and get good readings off of. First, they're a lot bigger, so there's more to test.

    The Unfired case, measured 84, 84, then a different piece from the other side of the case measured 86, 86. Ok, I'll buy that its about 85.

    The 10 times fired case, measured 88, 87, 78 (possible a bad reading because of the set) 84, 85. Not much change if any. Pretty interesting. I would have thought that it would also work harden much like a piece of steel that got forged. Increase in density and decrease in volume. Not apparently so. Not at least in any great amount.

    The 45 times fired brass, admitedly, from another lot, measured 87, 87, 87, 87. Well, about 87.

    Now, I don't know how hard brass gets on the B scale. Marine stuff is MUCH harder than the softy crap we deal with.

    If I had to make a conclusion, I would say that through time, pretty darn little changes. Now, this is not a test of how mallable the material is, simply it's hardness. Perhaps some crazed metalurgist out here can add something that would be heplful or continue the testing.

    My current annealing process is done with a dual motor, microprocessor controled gizmo. A stepper motor moves another spin motor into a flame , spins the case and removes it after a period of time. I DO NOT quench the cases and the case spends LESS THAN 2 seconds in the flame. I'm not sure exactly ho long it is, if someone want's to know to the 1000th of a second, I can check the program in the simulator but suffice to say , the case gets NO WHERE NEAR red hot. And, it goes from roughly 85 rockwell, to 35. I take these cases out of the annealer with my fingers immediately after being torched and put the next one in. The total process takes less than 5 seconds. My skin on my 2 fingertips did get a little hard, but never burnt. That was after doing 30 cases. No big deal. I just grabbed down as near the base as I could. After the annealing process, there is absolutely no discoloration of the brass whatsoever. The only thing that can be seen is where fingerprints were not properly wiped off the case before heating it. They show up and need removed with some nevr-dull. Otherwise, they come out as pretty as they went in.

    Determining if this means anything is left up to the reader [​IMG]
  2. 4mesh063

    4mesh063 Well-Known Member

    Dec 8, 2002
    As an additional test of the annealing time, I just took the same piece of brass that had not been fired (the one from the body that measured 84 or 86, I can't remember which was which, and annealed it.

    I counted 1000 1, 1000 2 and removed it from the flame. Left it air cool till I could hold it in my fingers and tested it. It rockwelled 35 and 33. pretty soft for 2 seconds, huh. And, it measures .030 in thickness. So it's about 2.5 times as thick as anyones case necks, 3 times thicker than mine.
  3. Tailgunner

    Tailgunner Active Member

    Apr 30, 2002
    It's been a lot of years since I worked with Rockwell (used to calibrate them), but I seem to remember that the thickness needs to be around 5 times the diameter of the indent in-order to avoid reading errors caused by the anvil (punch through).
    You can drop to the 60Kg level "F" scale or increase the ball size to 1/8 "E" scale or both 60Kg and 1/8 ball "H" scale to eliminate the punch thru error.
    I'm assuming your using a 1/4 anvil, with the inside of the radius facing up.
    BTW, intresting test and thanks for shairing the results.
  4. Brent

    Brent Well-Known Member

    Jun 12, 2001
    RC 80? Isn't that harder than a typical reciever or bolt? Isn't that impossible, even though it does seem to give a relative reading?
  5. Steve Shelp

    Steve Shelp Well-Known Member

    May 3, 2001
    I beleive he stated he is using the Rockwell B scale as compared to the C scale which is used for the materials that actions and bolts are made out of.

    Could you describe possibly how intense your flame is and what your source of gas is for it? I find it interesting that 2 seconds or 1 1000, 2 1000 anneals that much. For my 338 Lapua cases with my setup using a Burnz-o-matic I normally count 1 1000 -> 10 1000 to get what I consider a good anneal and the "needle" of the blue flame would stop right at the centerline of the case when the torch's brass nozzle has about 1/8" clearance between it and the case body. My smaller 6.5 cases normaly take 6 secs.
    Don't get me wrong I'm not saying your wrong or anything. Simply trying to compare your results with what I've been doing time/temp wise and the only difference is the intensity of the flame used I would guess. I spin my cases also, so the heat is evenly distributed around the circumference.

    I've been thinking of "turning up the wick" sort of speaking during my annealing process to heat the neck/shoulder faster and not leave it in the flame for so long and you sort of backed up my thinking. Would appreciate your comments.

    Thanks for the writeup!

    [ 04-09-2003: Message edited by: Steve Shelp ]
  6. 4mesh063

    4mesh063 Well-Known Member

    Dec 8, 2002

    Well, you could write what I know about the calibration on the head of a pin, if you used a big marker.

    I will guess that what you meant was to have a ball size 5 times the thickness instead of the other way around. Knew what you meant though. I'm afraid that the Ames tester I have is only set up for B and C, Perhaps I could get another ball and maybe theres one in the case with the other anvils I missed. I'm not sure. I used a 3/8" anvil I believe, I didn't bring it home with me tonight so I can't say and yes, I had the IS radius facing up. I see you noticed I had some trouble getting a good "set". That was resolved by bending the piece with a pliers, just around the edges and then making sure that when the ball JUST touched, that the part could be turned freely (no 2 points touching) It seemed to work pretty well.

    In any case, I'm not going to vouch for the rockwell as being ANSI perfect. But, the numbers did stay consistant regardless of the thickness as long as I treated the piece the same. For what I wanted to know, it was ok data. I figured to share the results and thought it was possible that the test would be skewed somewhat because of the thickness of the material. I will say this, the back side of the thinnest material tested has a "punch throug" mark roughly .080 to my eye, that is flattened on the anvil. That's the back side of course. The thicker material does not have that mark with the exception of the one that I annealed and got a similar reading of 34-36. That one also has a mark and for what it's worth, the mark is indistinguishable in diameter from the thinner material's mark. It's not punched through as far, but the size is the same within my view.

    If you have access to a unit that would read on another scale, say E of F by all means, try the test yourself. You do loose a little blood cutting the cases up with a tinsnips though. You've been warned [​IMG]

    Brent, As tail pointed out, and I specified at the beginning of my post, that was RB80 not RC80. The test is performed the same but instead of pressing a diamond into the part, you use a carbide or HSS ball penetrator of a calibrated radius for softer materials. They even have scales for plastics and that sort of thing though you won't find a tester like that in many machine shops.

    Steve, I think you are getting the case red hot. Perhaps, white, at least bright yellow in 10 seconds Hehehe.

    I tried the red hot thing and I ruined a bunch of cases. They were burnt to a crisp. Keep in mind again, my necks used to be a case body, so they're hard. Not like the guy who necks a case down and trims off .050. I did this test because I weld aluminum wire at work every day. The weld is done in a presswelder, no heat or electricity. This weld must pass through a crimper and survive so I started annealing the welds about a year and a half ago. I anneal the .063 wire about 1 second, with a bernzomatic torch using regular propane. Now, that wire is 5 times as thick as a case neck. It'll melt off in 2 or 3 seconds. Now, granted, that's not brass but the change is obvious. It's a limp noodle when I hit it for 1 second, and it's hard before.

    My torch is the kind with the diffuser in the nozzle. The flame is not as nice as the old unit at work. That baby makes a real flame. Mine is roughly 5/8 to 3/4" diameter at it's greatest. The torch is a Bernz TS-4000 from Lowes about a year ago. I DO use Mapp gas instead of propane, but that's because it's what I've got for plumbing(Redid the whole house and didn't want to spend my life waiting for fittings to warm up). For this experiment, I placed the case neck into the flame about 1" from the orfice of the torch. My flame is about 2-1/2" in length if unobstructed. When doing this by hand with a cordless drill and fixture, I simply could not help myself making the time longer as I went along so I decided to make a gizmo that takes it out of my hands. I also spin the case (or the gizmo does now) so we are on the same page there. I think that I may have to move the case away from the flame some more because of the way I've attached the torch to the unit so I may be out near the tip of the flame like you are on the next try, unless I change the work I did tonight. I want less heat yet, so moving away may let me have more resolution with the timing. I will say this, the 1-1000 2-1000 is longer than the automated gizmo runs. I'd bet it does 1.5 seconds. I'm gonna have to run MPLab's simulator with the current code and see just how many machine cycles the case is left in place because I don't want to increase the time when I make some programming changes on it. Right now, it's clunky code and just has a loop in there to waste some time. Later, the user will be able to set the time with a switch and a manual run button. Then there will be a default setting for my run time since I have some friends who will most definitely use the thing till I make them one. I could also just keep a chip for me and one for them, whatever.

    Oh yea, before I finish, about not leaving in the flame so long. That's what I am trying to get away from. I don't want the burnt crap on my cases, inside or out. If I was one of the guys who shot cases 3 times and threw em out, I wouldn't care, but I want mine to live forever. I am near convinced that there is no falloff in accuracy if the brass is cared for properly. I'll find out this season.

    Just as another test, I may put a case in the oven at 500 and see how soft IT gets. I'd like to know really how hot you need to go till you get all your gonna get. Since you draw most tool steels at 300-700, I'd bet 1500 (red) is not necessary. I know for sure that if my gun doesn't shoot this year, it'll at least win the award for the prettiest ammo! No burn marks here!

    And, Thank You for readin.
  7. 4mesh063

    4mesh063 Well-Known Member

    Dec 8, 2002

    Also, I DO NOT quench em. According to all the manuals I've read (on annealing) that's a no no with brass. Slow cool. preferably in powdered lime. Well I'm not gonna use that so air is gonna have to do. My brother told me to do a quenching test also and I will do that, but, I'm ruining match brass at $1.85 a piece for every one I test so I'm just not gonna try everything. My sence of humor ran out yesterday when I had to close my eyes and cut up one of 23 matched cases to do this.
  8. CatShooter

    CatShooter Well-Known Member

    May 8, 2001

    I have always water quenched my cases... the ol' "pan and watter" technique.

    It was how I was taught years ago, and it seems to work fine.

    I would like to see your test of the differences between air cooled, and water quenched... you could do it on plain cases and not waste match cases.

    Some time back, I got a catalogue for a annealing machine that used air cooling, and didn't get it because I thought water was the only way to go, but I would re-consider it, if the air would work as well.

    Could you post photos of yoru annealer??


    [ 04-14-2003: Message edited by: CatShooter ]
  9. 4mesh063

    4mesh063 Well-Known Member

    Dec 8, 2002

    First, I like my cats! [​IMG]

    I did a test on the red hot and quench deal but since there didn't seem to be any interest, I just emailed the results to anyone who was interestes. As for pictures , I borrowed a camera and got a photo of it while it was unfinished and I'll send it to ya in the mail. I have no place to post the photo.

    I took the same body brass that rockwelled in the 80's can't remember now how high exactly, and got it red hot. Left 1 piece air cool and 1 I quenched. I also did a neck for the 2 seconds and quenched it. the pieces that were red hot were so soft they could not be read on the tester I have. They were like butter. Air cooled or quenched. The 2 second neck stopped at sixty something so that leads me to believe that quenching stops the process as well as makes the final hardness a random number. I don't like the dead soft stuff, so I have decided to continue doing the 1.5 second heat thing. If it's wrong, it won't affect anyone elses gun even a little bit.

    If quenching stopped at 65 lets say and air cooling stops at about 35, then the amount of time between heat to quench is very important for consistant neck tension and neck size. There's a big difference in how much rebound you get when sizing depending on how hard the brass is, as you probably know.
  10. Cybra

    Cybra Well-Known Member

    Dec 4, 2002

    I have been following this thread, and am very glad you've gone to the trouble of posting your findings, as it's something that's been rolling around in my own mind for some time. I've just recently been *pushed* into the annealing process by the necessity of forming a case design that I'm now making. It's a small case--quite small, actually, and so...I've been pondering exactly what sort of set-up I should develop for the process. Forgive the bland ignorance here, but ---- it, if I don't ask, no other foolish men will, so here is; do you have any idea how much work-hardening would/could occur on an overly soft case in a firing process developing a Max pressure of say...54-58000 psi? I've also wondered if the SPEED of the pressure rise would have certain...effects on the hardening process? I.e. a faster rise time might result in a slightly harder case? <shrugs his shoulders> All these technical questions, but I say if you wanna have fun, you've gotta get technical! Again, thanks for the very exciting post--I'll certainly look forward to future findings if you get the time.

  11. Darryl Cassel

    Darryl Cassel Well-Known Member

    May 7, 2001
    Catshooter ---(I like the name)

    I have to agree with you.

    I believe one should check Lapuas tecnique in annealing their brass cases.

    Lapua is probably the BEST brass on the market and I have been told, they use the water quenching tecnique. The brass I get from them is already annealed which saves time when preparing cases.

    I believe some try to put too much tecnical applications in their 1000 yard match shooting to NO avail.
    Some of the old timers and excellent shooters will tell you, just work up a load, do the "normal" things to a case that everyone else is doing, and try it out.
    Conditions will either allow a good group or tear it apart. The number one thing in match shooting is to get a REAL good barrel to begin with.

    Like John Hoover (one of the best) told me several times. He NEVER cleaned his rifle from the second match, to the end of the year in 2002. He uses a 6.5/284

    Do we over do things, I think so.