ANNEALING

Discussion in 'Reloading Forum (All Calibers)' started by Webster, Aug 6, 2019.

Tags:
  1. SGK

    SGK Silver $$ Contributor

    Joined:
    Mar 14, 2018
    Messages:
    609
    Curiosity got the better of me. I grabbed a few old cases and ran them for extreme times. I did not notice any drop off in current, even running things to the point of a very brightly glowing case in a sunlit room. 14.2A for 15 seconds. (The case stuck in my shelf as it partly melted the trap door.) Now, these cheap ammeter/voltmeters we use may not be sensitive enough but there was plenty of time for it to react. The case in question is totally toast and I could just squeeze the neck flat with my fingers.
     
  2. jthor

    jthor

    Joined:
    Jan 14, 2018
    Messages:
    187
    correct, it’s not measuring hardness or temp. They’ve said several times, that heat isn’t their main concern, hardness is. That’s why I giggle when folks say you can’t have a bit of red when annealing. I call bs on that just cause the higher the temp, the less time it takes to anneal. Their software is programmed to get your brass around a certain HV, 95-100 HV, and they do this from the HV tests they’ve done after backing off of the melting point.
     
    Mikemci likes this.
  3. SGK

    SGK Silver $$ Contributor

    Joined:
    Mar 14, 2018
    Messages:
    609
    AZTEC Mode is meant to be a self-contained analysis tool. Put in a case, it measures/monitors SOMETHING and then computes a time/power profile for that case. I doubt it has the wherewithal to measure hardness. How does it know - in your hypothesis - that the melting point has been attained? Temp? Possibly but it's not easy to measure temp that well. A change in the electrical load presented by the case in the work coil would be easier to monitor - the current would change and that can be measured/monitored relatively easily. Now I just realised I had my current limiter on and so if the natural load presented by a 308 case in my work coil regardless of state was always above 14.2A (@ c 48V) I wouldn't see a change on my ammeter when such load changed. (i.e. lower than 3.4 ohms) So my recent experiment is likely flawed as a result. Of course even if current peaked and we could record the precise time it did, we'd still need to know how much to back off. I doubt "about 10%" would cut it but who knows.

    Pic of the case mentioned above. Yeah, I think I softened the brass.
     

    Attached Files:

    Last edited: Aug 11, 2019
  4. Webster

    Webster

    Joined:
    Aug 29, 2009
    Messages:
    1,847
  5. jthor

    jthor

    Joined:
    Jan 14, 2018
    Messages:
    187
    To the best of my knowledge and my testing with the GinaErick annealer, it notices the moment the brass melts due to the current drop. There is a constant they then use to back off of the timing and power from the melting point with each given pilot. I’m assuming it is more complicated than that though, but that is the basics. If you know the timing that it took to melt the brass at a given power level and then anneal the brass until you get an HV around 95. Simply put, all you have to do past that is see if the delta off of the melting points time is consistent with other brasses melting point. If it is then you have something to fine tune/build off of.
     
  6. SGK

    SGK Silver $$ Contributor

    Joined:
    Mar 14, 2018
    Messages:
    609
    At some point such condescending responses will really start to annoy me...

    (Part 2 is better still. Note the discussion of cleaning cases and consistent neck tension which is why I raised the question above. But neither help someone looking to optimise time and power in an induction annealer such as a GinaErick.)
     
    Last edited: Aug 11, 2019
  7. SGK

    SGK Silver $$ Contributor

    Joined:
    Mar 14, 2018
    Messages:
    609
    Those running their GinaErick's via Arduino could potentially monitor and log/record current draw and its peak with very good accuracy. (With current limiting off of course.) It would be interesting to then correlate that with a simpler indicator like Tempilaq and distance from case shoulder junction which hits 750F. It would be crude in comparison to AZTEC but a step in the right direction.
     
    jthor likes this.
  8. Webster

    Webster

    Joined:
    Aug 29, 2009
    Messages:
    1,847
    You don't have to melt the neck to determine the hardness at various times.
     
  9. SGK

    SGK Silver $$ Contributor

    Joined:
    Mar 14, 2018
    Messages:
    609
    Of course not. It appears you aren't following what jthor is trying to say. The 'melt state' is a proxy. It avoids having to directly test hardness. In other words, if you aren't going to place inside the AMP the necessary equipment to directly measure brass hardness - something that likely simply isn't practical - you need another proxy for determining just when the brass reaches a state with known characteristics. If indeed there is a load change when brass hits a particular state you can measure the time this occurs by observing current and noting when it changes. (You can determine how much power was required to deliver the brass to a state with known hardness.) You don't have to measure directly brass hardness (which would involve special equipment and likely cutting the brass case in half). Having determined the time required to hit a known state you can use a formula developed alongside actual brass hardness testing to determine a power/time combo for the brass in question which ought to deliver it to a predetermined target hardness. Clearly they will have tested their formula against additional hardness measurement on new brass.
     
    Last edited: Aug 11, 2019
    damoncali, dgeesaman and jthor like this.
  10. damoncali

    damoncali Bullet Maker Site $$ Sponsor

    Joined:
    Mar 30, 2011
    Messages:
    2,331
    I bet this is exactly what they do. This makes me wonder if you could reverse engineer it without hardness testing.
     
  11. SGK

    SGK Silver $$ Contributor

    Joined:
    Mar 14, 2018
    Messages:
    609
    Ideally you still need the hardness testing to design the algorithm which is used to back off from this point.
     
  12. riflewoman

    riflewoman Gold $$ Contributor

    Joined:
    Feb 24, 2015
    Messages:
    1,111
    Hardness testers aren’t that expensive nor is the equipment for making metallurgical mounts. Getting the expertise...priceless.
     
  13. jthor

    jthor

    Joined:
    Jan 14, 2018
    Messages:
    187
    Ummm... Vickers Hardness Testers are pretty expensive mate.o_O
     
  14. Webster

    Webster

    Joined:
    Aug 29, 2009
    Messages:
    1,847
    I did metallography, mechanical testing and failure analysis for 45 years.

    1. The Leitz MiniHardness Tester (which I loved) cost $950. You would need to machine a fixture to hold unsectioned cases.
    2. For metallography you need a way to section cases.
    3. Metallographic polishing wheels are expensive. You need 3. You need high quality wet sandpaper, diamond abrasive, Al2O3 powder abrasive and expensive polishing cloths to put on the polishing wheels.
    All this work has been done and presented on this website.

    After determining the hardness you need to interpret the results. Probably just duplicate AMPS goal hardness. None of this work is necesary since many hardness, time and temp charts have been d/l to this website.
     
  15. jthor

    jthor

    Joined:
    Jan 14, 2018
    Messages:
    187
    Their ideal HV is 95-100 HV. Same as Peterson and Lapua.
     
  16. Nick Caprinolo

    Nick Caprinolo Gold $$ Contributor

    Joined:
    Jul 18, 2018
    Messages:
    637
    Do I size first and then anneal, anneal first and then size, or size first, anneal and then size again?
     
  17. SGK

    SGK Silver $$ Contributor

    Joined:
    Mar 14, 2018
    Messages:
    609
    I would speculate that the collapse in current draw might have something to do with some sort of saturation effect rather than the resistivity of the brass changing (which appears to change in rather linear form with temperature). It’s interesting that AMP appear to mount some sort of sensor on their c-shaped air-gapped ferrite core.
     
  18. jthor

    jthor

    Joined:
    Jan 14, 2018
    Messages:
    187
    Anneal and then resize. Annealing warps the brass to an extent. Which is why you do it before resizing.
     
    Last edited: Aug 12, 2019
    Bob L. likes this.
  19. Nick Caprinolo

    Nick Caprinolo Gold $$ Contributor

    Joined:
    Jul 18, 2018
    Messages:
    637
    Thank you.
    Nick
     
  20. David101

    David101

    Joined:
    Dec 16, 2018
    Messages:
    181
    I did some tests to have a look at the current draw Vs time / heat into the case. What i found was that the current gradually increased during the heating process. Then for a very brief period it increased more than the previous linear ramp. Then it took a quick decrease. The last two phases which would be the critical data points for the Aztec happened very quickly. A much better shunt would be required than what most of us that have built Home made machines also the sampling time would have to be very fast. All this before you even get into trying to work out the formula.

    I find it much more reliable to work from the other end with Templiaq. firstly increasing the time to where the colour changes and then continuing to different changes in colour of the case itself.
     

Share This Page