Lean Process Shellcase Induction Annealing

The mission of Case X Case is to design a more reliable, faster, and cheaper annealing process that will simplify shell casing production flow while maintaining similar/better quality of current neck annealing operations.

Opportunity

During the manufacturing process, brass shell casings are work hardened into a brittle state that must be reversed to ensure a quality product. This is done by applying heat to the brass shell cases, thus reversing the effects of work hardening by annealing (softening) the cases. Vista currently uses a large batch oven process to achieve the required annealing. This current process requires batch sizes of 15,000 to 20,000 shells to be prepped and stored and utilizes a combination of gas and electric ovens to achieve annealing temperatures for a sustained period of 15 minutes. The failure of a proper anneal can mean losing the entire batch.

Solution

Induction annealing offers the ability to anneal each case individually as they pass though Vista’s manufacturing process. It achieves this by sending a high-frequency alternating current through a coil conductor (such as a copper coil) and generates a magnetic field. When a conductive material (such as a brass casing) is placed within this magnetic field, eddy currents are generated within the casing material, thus causing molecular friction and heating the conductive material.

The proposed induction process can achieve annealing temperatures within ½ second intervals with all required case sizes, matching current case production rates. This process is capable of a steady, single piece flow output. This process can be integrated into the current production line, eliminating the need for a large batch oven process that requires sitting inventory and added cost. This streamlined system is more efficient and cost effective, saving Vista money. By using induction technology, the case neck can be directly targeted while avoiding the case head. This process also allows for more fine tuning by adjusting the current used. These features allow for more controlled annealing and more efficient heating per case.

Background
Vista Outdoor is a global designer, manufacturer, and marketer of consumer products in the outdoor sports and recreation markets. Ammunition comprises about 60% of their business as the world’s leading producer of centerfire, rimfire, and shot case cartridges. They are a wholly and vertically integrated manufacturer, taking in raw materials and controlling the fabrication process thru to the retail shelf space as well as direct to consumer sales. This project will work directly with the Lewiston facility where they manufacture some of the most iconic brands in the industry, servicing civilian recreationalists and more than 80% of law enforcement nationwide.

Cartridge shell cases are formed from brass wire through backward extrusion draws and subsequent forming and cutting operations. This intensive work hardening shortens grains and increases latent energy, reducing the ductility needed in the finished part. Targeted ductility is required to allow for case mouth expansion during loading and firing while still retaining adequate head hardness characteristics.

Current anneal processes use natural gas‐powered, belt‐fed, batch ovens. This method forces a tradeoff between throughput and part‐to‐part consistency; As parts are ‘stacked’ on a conveyance belt, the depth of the stack determines the time needed ‘in soak’ at temperature to sufficiently reach the middle‐most parts. Defects introduced by this process can propagate through an entire batch or, if not, at least prove inseparable from good parts. Batching is currently done in 10,000 to 20,000‐piece bins, increasing in‐process inventory and material handling costs.

Case mouth anneal is of critical importance to shell case function and safe firearm operation. Cases that are too hard can split at the mouth upon firing. This is not detrimental assuming the split stays in the mouth and doesn’t propagate down the case body and vent propellant gas into the unsupported portions of the chamber. In some cases, however, a hard case mouth can split during seating of the projectile which can prevent the round from chambering properly in the firearm.

Conversely, cases that are too soft will not have enough hoop strength to retain the projectile during the combustion event. If the projectile is dislodged from the case mouth before sufficient pressure is generated, this can lead to an increased combustion volume and inability to reach the intended peak pressure. This can result in a “bullet‐in‐bore” defect where the projectile becomes lodged before exiting the muzzle end of the bore. This is the most critical of defects as a second round fired into the now‐obstructed bore can cause catastrophic failure of the firearm. Even notwithstanding a second firing, the firearm is now inoperable; Losing the ability to fire can be a life or death mishap in a lethal force encounter.

Deliverables
The goal of this project is to design and determine the feasibility of an induction annealer and to estimate the projected cost savings if this induction technology replaces the batch annealing process currently used.

This project’s success will hinge on the ability to properly anneal the case mouth. Equipment developed is not expected to run at the two part per second rate typically needed to match upstream operations but must demonstrate the ability to consistently anneal with a high rate of confidence as to prove that scaling of the device is both possible and a better alternative to current operations. The quality of annealing is verified in several ways. These methods include: Quality of anneal will be verified in a number of ways including: microscopy to look for appropriate grain growth, micro‐hardness evaluations, and function testing of sufficient quantity to establish statistical probability of defect rate. A one‐in‐a‐million defect rate is unacceptable as we load 18 million rounds each and every week of centerfire handgun alone!

This projects end goal is to produce the following...


 * One‐part flow case mouth annealing device applicable to 9mm, .40, and .45


 * A near zero defect rate


 * Analysis and flow diagrams documenting current and ideal state process


 * Cost saving analysis from reduction in human labor and inventory


 * A process that does not pose a hazard to employees

=Design Considerations=

As implementation of the design into Vista Outdoor’s manufacturing process is a crucial component of this project, the possible methods of implementation were carefully considered.

Design proposition 1:

The Annealer could be inserted as the shells come from the hopper, leading into the tapering machine. This is our primary design focus. This concept considers annealing the shell cases as they pass from the hopper into the inspection machine through a tube. This concept, while requiring some manufacturing process adjustment, would allow for easy customization of the system and would be easily accessible for necessary maintenance.

Design proposition 2:

Annealer could be inserted as the shells exit the tapering machine. This is an alternative approach consideration. This concept considers annealing the shell cases as they pass through the discharge chute. This concept would require little to no adjustment of current manufacturing process. But would be harder to implement due to space constraints.

=Test Bed Prototype Design=

This prototype was used to demonstrate the functionality of using induction to heat shell casings and to validate the math models for power use in the final design. This design consisted of an induction board and separate power supply that provided 30 V of alternating current at a maximum of 5 amps. It was used in order to validate the general equations the team used in purchasing our final design.

=Final Prototype Design=

This Final Prototype Design was selected based on previous testing. It was purchased from U.S. SOLID. The induction unit itself is a US SOLID 15kW high-frequency induction heater purchased directly from the manufacturer, US SOLID. The unit is a purchased whole, so it has several features that ensure ease of use and safety. The unit can operate in both manual and automatic modes turned on or off via button or foot pedal, an adjustable timer for automatic heating, and a digital display shows the supplied amperage. Safety features include an on/off switch, start/stop buttons, a circuit breaker on the back, a low flow warning for the coolant, and a short circuit warning. These features allow the unit to operate up to 15 kW safely and reliably.

=Testing=

Test 1: Validation
Test 1 was used to validate our math models for determining the power required to anneal shells at sub-second speed. The team recorded temurature rise using a thermal camera. The team then recorded this temperature and created the following plot. Results showed a quadratic relationship and this equation was used to find the required power for sub-second annealing was around 1.335kW. This plot allowed the team to experimentally derive an equation to determine the necessary power to achieve the necessary temperature in one second. Once the governing equations were verified, the team proceeded with purchasing an induction unit that could achieve the calculated power for the prototype.

Test 2: pass/fail press testing​
Test 2 was used to determine ball-park amp ratings where annealing occurred in each shell type. To find these ranges shells were annealed for one second at staggered amperage settings and their results were recorded and compared against a control sample set of traditionally annealed shells. It was discovered that when shells were over-annealed they consistently buckled when pressed. When shells were not annealed they would split or butterfly. It was then the goal to find the power setting where around 70% of the shells would buckle. Below are the resulting ballpark figures as well as the press used and various failure modes encountered.

Test 3: Drop Test
This test was used to discover the effects of free-fall annealing. The goal of the drop test was to determine a necessary power setting range for the hardness testing. The procedure for the Drop testing involved dropping a shell though the induction coil and sending it through the press test to see how the case failed. After discovering that the drop method was unable to heat the shells to the necessary temperatures, the team designed a “quasi-stationary” method.

Test 4: “Quasi-stationary” Testing
This modified design was created to allow for continued testing. These modifications would create a controlled time delay in the rate at which shells were dropped, and in how long they were held in the induction coil. This allowed for shells to achieve the needed temperatures necessary for testing.

The procedure for the “quasi-stationary” test involved heating shells at specific power settings, using the results from the pass/fail press test to help set amperage perimeters. We tested perameters until our press results were getting close to matching our control samples.

Test 5: Hardness Testing
The process for the hardness testing was to heat batch sizes of 10 shells at set intervals within the power setting ranges that were determined in the quasi-stationary testing. These shells were then packaged and sent down to Vista Outdoors for hardness testing. There was a preliminary hardness test was done on the high, low and middle power settings for each caliber. This was to determine whether the team had the correct power setting ranges for annealing to occur. The results of the hardness testing showed the team exactly what they expected. The process the team created was able to soften the shells in less than a second.

Due to the variation in the control samples for the hardness testing, the team was unable to determine a power settings that anneals the mouth of the shell. As shown below, the control samples showed little variation in hardness, in fact the annealed cases were on average harder then the pre-annealed samples. These results were not useful in determining anneal quality beyond generalization so another test was needed.



To Test for hardness the cases were cut in half and set in resin. Then micro indents were made along the length of the case. These indentations points are shown below.



Test 6: Instron Testing
Instron (an ITW company) is a manufacturer of test equipment designed to evaluate the mechanical properties of materials and components. Similar to the pass/fail press test, the Instron test involved a load being applied to shell cases. This test used an Instron Compressive strength testing machine at Vista to evaluate the average max load and extension the cases could withstand before failing.

Below are the compiled results for the .40 and.45 case samples. These results show that our testing captured both the lower and upper limits of annealing the brass cases. It is expected that below a certain power rating no annealing will occur and the cases will remain brittle. And above a certain power rating there will be a complete anneal or over anneal which will leave the cases too ductile. This expected S curve is present in our data and can be used to evaluate the quality of shell case annealing.

The Red overlay shows the tested sample max loads and extensions of Vistas current annealing process, with the dashed red lines showing the variability of these samples. The Green lines show the point at which our process achieves the same results as Vista's current operations. These results give a good indication that not only can induction annealing match Vistas current requirements but can get more consistent results.

=Value Streams= The figures below are Value Stream Maps. A Value Stream Map gives a visual representation of a manufacturing process. Value Stream Maps are used to see where in the process can improvements be made to reduce waste. Waste can be anything that does not add value to the product. The inventory step between the inspection/tapering step and the annealing step is waste that can be removed by changing from a batch process to single piece flow. Inventory is the monetary value of a product at a specific point in the manufacturing process. In the current value stream the batch sits in inventory for 16 minutes. For these 16 minutes that the batch is sitting in inventory does not add any value to the process.

The Future State stream is the implementation of the proposed single piece flow annealing system that utilizes induction heating. This single piece flow system can keep up with the production requirement of the tapper-inspection machine. Moving from a Natural Gas oven to the induction heater has a cost savings percentage of 93%, saving $114,856 in every day operations by removing that inventory step as The future State shows below.

Current State Value Stream Map:



Projected State Value Stream Map:



=Future Work= Since the team only produced a proof of concept prototype there is still a considerable amount of work that needs to be done. The key topics to explore for successful integration with the production line are quality assurance, refinement of quasi-stationary feeding system, and final system integration into the production line.

Quality Assurance

A live fire test still needs to be done by annealing a batch of 1,000-2,000 shells of the same caliber. These shells will then be loaded and fired to see if any of the shells fail. There needs to be a near zero defect in the annealing process. A batch size of 1,000-2,000 is large enough to provide relevant statistical data for quality that the induction annealing process can provide.

“Quasi-Stationary” System

The system that controls the “quasi-stationary” system is going to need to be refined. Currently the prototype is using a series of trap doors and motors that are controlled by an Arduino. This system is going to need to be refined before it can be implemented in the manufacturing line.

System Integration

The current prototype will require future work to be implemented into the manufacturing process. The team recommends redesigning the frame that holds the tapper/inspection machine and the hopper that feeds it to accommodate the induction machine.

=Team Members=

=Additional Documentation=