Armory Sponsor
|
Originally Posted By -Obsessed-: If you're only getting to 750 momentarily, you aren't fully annealing your cases. It takes more than that length of time to anneal cartridge brass at 750 (6 seconds to fully anneal). At 850 it's about a second and by the time it hits the draper point (infrared radiation) it is instant essentially. The goal is to optimally anneal and not fully anneal the brass. To fully anneal brass takes 1300F for 30min as shown in the graph below. Attached File https://vacaero.com/information-resources/metallography-with-george-vander-voort/1440-deformation-and-annealing-of-cartridge-brass.html The question becomes what's optimum? Over annealed (but still not fully annealed) the neck is too soft to obtain good neck tension and accuracy falls off. The below article describes an example of how this happened using the AMP, the results and how it was corrected. The manufacturer has found annealing time depends primarily on case neck and shoulder thickness, and my Nosler brass was from a batch more than a decade old. Back then Nosler purchased cases from more than one manufacturer, and mine differed from the Nosler .30-06 test cases listed by A.M.P. As a result, they got over-annealed, and groups were considerably larger than normal. After consulting with A.M.P., a different program for a shorter annealing time was tried with fine success. https://www.handloadermagazine.com/annealing-rifle-brass A white paper done by AMP suggests there is simply not enough time and retained heat for localized induction annealing to heat the case head or webbing enough to anneal it, as even 650 degrees takes 15 minutes to anneal. I could do more, tighter coils, like AMP does, which focuses the Eddy currents tighter. Doing so reaches anneal temp in about 3.5 seconds on a 223 case, which would be quicker than 300 BLK due to the latter having larger webbing. For repeatability and simplicity, I chose a single coil vs multi layered coil, but I may revisit it. At least for my needs though it's not necessary as there is still ample time for it to cool before the heating gets anywhere near the case head. It obviously travels down but by the time it reaches the bottom the entire case has cooled to just over boiling temp.l, which is far, far too cold to anneal. I was only concerned with melting the plastic cartridge specific insert if it's touching the base of the case. If the filament melts at 250F when printed, then it may melt when the case gets annealed. ETA just ran a 223 through. It took 6 seconds to get this result. 850 reached below the shoulder. This piece shows pretty good coloration as well. https://www.ar15.com/media/mediaFiles/467079/1000008982-3087275.jpg It takes about 9 seconds to get this on 300 BLK. I'm running longer times to get the shoulder to a red glow to ensure full annealing, which is why I'm running longer times. This afternoon I confirmed I can get a red glow to just below the shoulder on heaviest and lightest weight brass in my 100-piece sample at 13.0 seconds. 12.5 it basically approaches and just starts to glow at the bottom/outer radius at time on the heavier piece. I did 13 for a bit of safety factor. I think I could very safely cut this time back some as I reach 850 below the shoulder in 9 seconds. So it takes another four seconds to get that last 75 degrees or so. Obviously thermodynamics are at play here and naturally temp rise gets more gradual at the top. ETA 2. After give ng it some thought, if I know I hit 850 below the shoulder in 9 at the most...and annealing is 1 second at 850, I can safely assume at 9 + 1 seconds the case is fully annealed even discounting the cooling time as it falls back down to below 850. Can you provide the source where you've found 1sec at 850 is ideal for annealing? I haven't been able to find it yet. A graph on the AMP website shows they reach 1000F for 223 Nosler brass with AMP program 43. Attached File I think I'll just set 300 BLK to 10 seconds for this machine. I'd do more testing but I'm guessing 223 would be around 7 seconds. I really don't know the what the ideal temperature should be and from AMP's statement below appears they use a hardness test instead. It is extremely difficult to accurately measure the peak temperature of flash annealing. As noted above, the required temperature also varies with time. That is why AMP focuses on the end result – neck and shoulder hardness – rather than temperature. |
|
Could just approach it empirically, too. Since his objective is just longevity, do 5 cases at 5 seconds and load and shoot them. Rinse and repeat and take note of how many cycles until the necks start cracking. Then do another batch the same but with a 6 second anneal. Same process. Then another batch of 5 but with a 7 second anneal, etc... eta: Start with a control batch that you fire and reload without annealing. |
Never confuse faith that you will prevail in the end—which you can never afford to lose—with the discipline to confront the most brutal facts of your current reality, whatever they might be. - Adm James Stockdale
|
Originally Posted By Blowout: I really don't know the what the ideal temperature should be and from AMP's statement below appears they use a hardness test instead. It is extremely difficult to accurately measure the peak temperature of flash annealing. As noted above, the required temperature also varies with time. That is why AMP focuses on the end result – neck and shoulder hardness – rather than temperature. I'd have to look, I think it was the AMP white paper from their early testing, and I saw a discussion on I believe Accurate Shooter but it may have also been referencing the same source. I saw three sources so far saying you can't get too hot on the neck...is you can't over anneal, and thus heat transfer to the head is the primary concern. I'm no expert. Is what I'll probably do is a type of empirical testing. Not like Andy above says, since I can't shoot in my backyard, but I may anneal, test neck tension, repeat. In theory, if you are correct, over annealing should also result in neck tension falling off a cliff. Agree, or no? |
|
Originally Posted By 1Andy2: Could just approach it empirically, too. Since his objective is just longevity, do 5 cases at 5 seconds and load and shoot them. Rinse and repeat and take note of how many cycles until the necks start cracking. Then do another batch the same but with a 6 second anneal. Same process. Then another batch of 5 but with a 7 second anneal, etc... eta: Start with a control batch that you fire and reload without annealing. Based upon my prior testing of this type, you will never get there with this approach. Also, if longevity is your objective, it can be a very long road, indeed, only to find out you did it wrong. BTW, I think Obsessed is after uniform neck tension, not enhanced case life. |
|
and to everyone saying it takes X amount of time; That's a variable that changes for each machine. Coils and power and mass are variables that make X. You can dial it in by watching for how red the case gets, IIRC you don't want full red, but just before red when its a dim brown. This X will change for each round size of mass. |
|
Originally Posted By -Obsessed-: I'd have to look, I think it was the AMP white paper from their early testing, and I saw a discussion on I believe Accurate Shooter but it may have also been referencing the same source. I saw three sources so far saying you can't get too hot on the neck...is you can't over anneal, and thus heat transfer to the head is the primary concern. I'm no expert. Is what I'll probably do is a type of empirical testing. Not like Andy above says, since I can't shoot in my backyard, but I may anneal, test neck tension, repeat. In theory, if you are correct, over annealing should also result in neck tension falling off a cliff. Agree, or no? I'd guess a more gradual loss of neck tension from looking at the graph below. That data is calculated and not measured. Attached File The article states most people target 50% hardness. I wouldn't be too worried about hitting a specific temperature end point. Too many variables with case thickness between mfg and lots. AMP does an analysis on each type case to come up with a specific time setting. If shooting mixed brass, you'd drive yourself nuts trying to figure out the best setting for each type case. https://bisonballistics.com/articles/case-neck-tension-a-stress-analysis Using an AMP Press would be a way to measure the effect of annealing times on bullet seating. $1400 fun toy! |
|
Originally Posted By -Obsessed-: Did some CAD work on the front face. This will enable the mounting of all the components and test space. After that I'll start working on the hopper and chute/trapdoor. I'm going to redesign all this to be more flexible since I've received so much interest. I was originally figuring I'd just make my own adapter pieces as needed but if others are interested, I may as well try to make it as universal as possible. That may include redesigning the coil to be multi-layer wrapping to enable faster, more focused induction. It's just a lot harder to wrap them that way and designing a jig may not be possible. We'll see, but I'll do this for now. https://www.ar15.com/media/mediaFiles/467079/Front_Face-3088177.png Not many put their annealer design in CAD. You definitely have more skills than most!! Made some changes to my flame annealer to reduce the annealing time because I'll anneal 1K or more cases in a sitting. Automating the process was great, but sitting for 2 hours watching them process wasn't. Increasing the flame intensity with a higher pressure regulator reduced annealing time considerably. Sounds like you have a lot of brass to anneal. Coil redesign for more intense heat could save you hours spent in in front of the machine. |
|
Originally Posted By Millennial: what positive control ensures that the brass is right side up? If a piece of brass were to flip around on the way down or were accidentally loaded into the hopper backwards and the machine were to anneal the wrong end... that could be very bad. On the previous design the brass manipulation was such that it was physically impossible for it to invert. There simply wasn't room for it to flip. In the new design I'm hoping for the same to be true but I have to test it first. In theory the way the chute is designed it could technically flip if it's super short, like 300 BLK |
|
Originally Posted By Blowout: Using an AMP Press would be a way to measure the effect of annealing times on bullet seating. $1400 fun toy! That is the reloading item which most interests me, despite the relatively high price, because KNOWING is better than just "should be". |
Quick little update...just a video that shows the new hopper, drum, and chute design. This allows quick changeovers of sizing. Just quick and dirty. Note the hopper isn't full size, I truncated the part to only the components I need for testing.![]() Annealer Update 1/13/24 |
|
That's awesome! I don't have the design skills for it but I will chime in and say if you ever plan to use a dillon case feeder or something with this, you need to figure out some way so that a case that gets dropped from the feeder upside-down physically cannot get dropped into the annealing coil. On the amp-mate, if the dillon casefeeder drops an upside down case, it just doesn't get transferred to the shell holder. It'll just fall on the floor. I don't know how you'd do that with a drop through style induction annealer. |
Never confuse faith that you will prevail in the end—which you can never afford to lose—with the discipline to confront the most brutal facts of your current reality, whatever they might be. - Adm James Stockdale
|
Nice work op! Your design skills are impressive. I did something similar a couple years ago (during the covid work from home lockdown). Used a MGNZ(?) print for the case feeder. (had someone print it for me ($30)). Stepper motor control but converted from audruino to a cheap little plc. Been programming PLCs for 25years and couldn't find it in myself to want to learn aurdunio. I'd bought all the parts to do the water cooled annealer but changed my mind when I found people using inductive bolt heaters with a cheap high resolution timing relay from amazon. Just easier for me to alter and combine premade systems and they already have built in safeties. If you can use it, I'd send you the 1000W induction driver I never used. this one. Never got around to implementing the feeder onto the annealler. Need to redesign the entire box to make it more serviceable. Here's the feeder test video. Think i need tie down resisters on my inputs for buttons. I have to press them twice to register. video Here's a vid of the annealler. You can kinda see my adapter plate setup for different sized cases. I change the plates and coils based on case size. GAAMPG (Ghetto Ass Annealling Made Pretty Good) |
I'm not here because I make sound financial decisions.
|
Bravo Zulu, OP. A few non-engineer suggestions, all adding complexity: 1 - If you put a positive pin sensor at the bottom of the case holder, if it goes into an upside down case instead of being pushed down by a primer pocket, it can shut down the machine. 2 - Non inductive cooling loop for the base of the case. 3 - Leave induction coil on all the time & use a boiler to run steam through the pipe - constant 212 degree radiator. You can then capture the steam and cool it. 4 - Instead of a positive physical stop, used compressed air - or steam - to suspend the case in the inductive coil.
|
Death to quislings.
|
Very cool. Nicely done. I am working on my own right now. Played with an Arduino Uno for a while, then Pi Pico's. Decided to go with a Pi Zero on mine. |
Every cloud has a silver lining (except for the mushroom shaped ones… which have a lining of Iridium and Strontium 90)
|
Originally Posted By -Obsessed-: Quick little update...just a video that shows the new hopper, drum, and chute design. This allows quick changeovers of sizing. Just quick and dirty. Note the hopper isn't full size, I truncated the part to only the components I need for testing. Looks like the new hopper design is really laid back! The hopper on my flame annealer is 11 degrees off plumb. I notice when its filled up, the cases work forward and need to be pushed back every so often. Your design should prevent that from ever occurring and some testing should confirm it. |
|
Originally Posted By Blowout: Looks like the new hopper design is really laid back! The hopper on my flame annealer is 11 degrees off plumb. I notice when its filled up, the cases work forward and need to be pushed back every so often. Your design should prevent that from ever occurring and some testing should confirm it. My very first prototype was 3°. I learned extremely quick that was the wrong call. ![]() Then 10°, was very good, but not great. Final design was 15° and I never had issues. That was my old version of the annealer in the OP. This one is 40°. It makes it a lot easier to slip larger cartridges into the coil. |
|
Originally Posted By -Obsessed-: A lot more CAD work. I'm on the final stretch of getting everything inside the box and getting it finalized. Obviously the prototype has everything strewn about. The bottom piece of the trap door assembly works beautifully. I'm very happy with it and my custom gear mesh seemeds to be perfect for the task. https://www.ar15.com/media/mediaFiles/467079/1000009111-3099655.jpg https://www.ar15.com/media/mediaFiles/467079/1000009112-3099656.jpg I wondered how those trap doors functioned... thanks for providing the pics! The trap door assembly was one of the reasons I didn't even try to DIY an induction annealer. Another was the hopper/chute. Your taking all the guesswork out of it! Another advantage of an induction system is it's smaller footprint. My DIY flame annealer is probably about the same size, but add the BBQ tank and hose and the space requirement easily doubles. I'd be interested in the anticipated footprint and weight of your final system. |
|
Originally Posted By -Obsessed-: Finished with the front plate. It's printed. https://www.ar15.com/media/mediaFiles/467079/1000009128-3102914.jpg https://www.ar15.com/media/mediaFiles/467079/1000009127-3102915.jpg wrapping up the base plate now. I had to design a way to hold the Arduino securely but still give me access to the USB port, reset button, and pins. I'm happy with it. But now I need to figure out cooling tubing routing and power cord routing. I don't really plan on implementing built in water cooling. The external cooling is so much easier, and I don't have to worry about draining it, refilling it, or leaking. If someone really really wants me to design it in, I can, but someone is going to have to send me a pump cause I'm not planning on going out of pocket on this machine any further. The point was to use things I had on hand, and this allows some flexibility. But if the populous hates that idea, let me know. Merely a suggestion. As I said, all my suggestions added complexity. You're making me wonder if I could build something similar out of Lego, the gears are similar size. |
Death to quislings.
|
Originally Posted By -Obsessed-: Dimensions of the main "box" are: 9-3/8" wide, 5-7/8" deep, 7-3/8" tall. Wow, that's compact! Suppose someone could attach a removable vertical reservoir on the back if desired. What is the ID of the tubing your using? Half of my career was design, development and manufacturing of medical laboratory instruments. The rule was avoid fluidics over electronics whenever possible. If unavoidable, don't make a fluidic connection over electronics. If that's unavoidable, shield the electronics for when that connection leaks. Can't tell you how many visits and calls I've been on where leaks damaged electronics. The connection shown below is most concerning... when the fluid heats up that soft tubing will loosen up. A small zip tie on those connections will help. - Maybe with a next gen coil the copper tube could be extended to where the other end terminates outside of the electronics? - Another thought is to make the connections in the block that's clamping the soft tubing. - Another idea would be to modify the block and use panel mount QD fittings on that block. Then the reservoir would be removable for storage. Attached File Panel mount QD fittings: Attached File Attached File |
|
Originally Posted By Blowout: Wow, that's compact! Suppose someone could attach a removable vertical reservoir on the back if desired. What is the ID of the tubing your using? Half of my career was design, development and manufacturing of medical laboratory instruments. The rule was avoid fluidics over electronics whenever possible. If unavoidable, don't make a fluidic connection over electronics. If that's unavoidable, shield the electronics for when that connection leaks. Can't tell you how many visits and calls I've been on where leaks damaged electronics. The connection shown below is most concerning... when the fluid heats up that soft tubing will loosen up. A small zip tie on those connections will help. - Maybe with a next gen coil the copper tube could be extended to where the other end terminates outside of the electronics? - Another thought is to make the connections in the block that's clamping the soft tubing. - Another idea would be to modify the block and use panel mount QD fittings on that block. Then the reservoir would be removable for storage. https://www.ar15.com/media/mediaFiles/375214/Induction_fluidics_png-3103829.JPG Panel mount QD fittings: https://www.ar15.com/media/mediaFiles/375214/QD_fitting_png-3103904.JPG https://www.ar15.com/media/mediaFiles/375214/QD_male_fitting_png-3103905.JPG I completely agree with everything you're saying. Truly it's just a PITA to wind a new coil. I probably will anyway, but in my testing the working pressure of this pump is basically nil (imagine holding the feeding reservoir up about a foot and letting it be gravity fed. I'm talking barely more than a dribble), so it doesn't leak, and it doesn't get even warm to the touch, let alone hot. Granted, it's still not a good idea and if someone did want to enclose everything it would be necessary to remedy it. I would do so by making the connection past the electronics, to your point. I could explore non-conductive coolants as well but probably not worth it over just fixing the design. At any rate I'll ziptie it. ![]() I built this with what I had laying around. If I had the reservoir, pump, and radiator on hand I'd use them but that easily adds another $50-100 to the project and an open vat solves my needs for $0. That said it would be TIGHT quarters to implement cooling internally, but I could expand the size to accommodate. Quick connect fittings are a good idea. Do they make them such that I can screw them together on either side of the panel? The tube is about 5mm OD, 3.5ish ID, again, just what I had laying around. |
|
Originally Posted By Millennial: or solder some copper tubing extensions far enough to get solid tubing to the edge of the box. It would be pretty bad if you sprung a leak over the power supply. Originally Posted By Millennial: Originally Posted By backbencher: Maybe just mount the electronics above the water tubing? or solder some copper tubing extensions far enough to get solid tubing to the edge of the box. It would be pretty bad if you sprung a leak over the power supply. Not all electrical fires are bad...
|
Death to quislings.
|
Originally Posted By -Obsessed-: I completely agree with everything you're saying. Truly it's just a PITA to wind a new coil. Understood, just thinking if you were going to have files made available it would be a nice to have the design updated. I probably will anyway, but in my testing the working pressure of this pump is basically nil (imagine holding the feeding reservoir up about a foot and letting it be gravity fed. I'm talking barely more than a dribble), so it doesn't leak, and it doesn't get even warm to the touch, let alone hot. Granted, it's still not a good idea and if someone did want to enclose everything it would be necessary to remedy it. I would do so by making the connection past the electronics, to your point. I could explore non-conductive coolants as well but probably not worth it over just fixing the design. I wouldn't waste your time on this. Coolant only allow the fluid to get hotter before boiling. At any rate I'll ziptie it. ![]() Hahaha... I've heard all sorts of reasons why it won't leak, until it does. I built this with what I had laying around. If I had the reservoir, pump, and radiator on hand I'd use them but that easily adds another $50-100 to the project and an open vat solves my needs for $0. That said it would be TIGHT quarters to implement cooling internally, but I could expand the size to accommodate. I personally think an external reservoir would be preferable. They make couplings that have a built in valve so the fluids don't go all over the place when disconnected. They are about 2x the cost of unvalved couplings. Quick connect fittings are a good idea. Do they make them such that I can screw them together on either side of the panel? The tube is about 5mm OD, 3.5ish ID, again, just what I had laying around. Yes, the female side (top dimension photo) is a panel mount. Doesn't show the mounting nut. |
|
Originally Posted By -Obsessed-: You guys worry too much. ![]() When you make your annealer you can do whatever you want. Speaking of, I'll get a parts list together here soon. Hahahaha... I get voices of the past whispering in my head! ![]() ETA: Here are the couplers with a built in valve so fluids don't pee out when they're disconnected. Attached File Attached File |
|
Another update. I'm going to pick up some insulating material for the pipe, to try to wind a double coil. Should allow faster induction times, which will translate to more productivity and less chance of heat escaping to the case head. I'm also contemplating using compression fittings on the copper tubing/aquarium tubing junction. I'm just trying to keep this cheap. But for about $10 I can convert theme to barbed/compression fittings. Still doesn't guarantee it won't leak. Printing the right panel now. Top panel is good, but then I realized one of my dimensions on the back panel I fat fingered. So I need to reprint it. Oops. Once i get it assembled and know it's how i want it for a "final" design, I'll reinforce the joints with nuts. Right now, a lot of the screws just thread into plastic. It holds well, but if you're not well versed in screwing into PETG you can strip them. It takes some finesse, and I'd rather eliminate that. Thinking I'll put hardware kits together to allow people to get a simplified way of building one. PM me if this is something you'd be interested in. I'm realizing as I'm getting all the parts together that most people probably don't have M1.6, M2.5, M3, and M4 hardware just laying around, let alone MG90 servos or optical sensors, all of which are cheaper if bought in bulk. DF3, if this last paragraph isn't allowed, feel free to edit. |
Global Warming Hoax Skeptic before it was cool
Joined:
Jun 2007
Posts:
67978
EE: 100% (1)
|
Originally Posted By dryflash3: I gotta ask 8E7? Never mind I get it. About the other, direct comms to IM's a soon as possible, and we should be good. It was tongue in cheek. ![]() I don't know if I'll brand the final version. I thought about making the panels clear and calling it the Peek-a-boo annealer. I need something catchy and differentiating. So people know what someone is talking about about. |
Global Warming Hoax Skeptic before it was cool
Joined:
Jun 2007
Posts:
67980
EE: 100% (1)
Armory Sponsor

















