Tech Trends for Digital Design
A customer’s quality and lead-time issues were solved through the use of new technology for stress analysis and mold-making.
Shannon Wetzel, Managine Editor
(Click here to see the story as it appears in the September/October issue of Metal Casting Design & Purchasing.)
All 10 castings in the initial trial run showed the crack. It was at the corner of a rib, sometimes in several points. Filling and solidification modeling had indicated no issues. The manufacturere of industry air compressors had a case to solve, and time was running short.
The part was a 442-lb. gray iron gear case cover housing used in air compressors located in free-standing boxes at worksites to generate industrial air and power. It was a prototype casting to go into production in late 2016, and further development of the air compressor assembly hinged on its successful and timely manufacture.
Dalton Foundry, which is a division of Neenah Enterprises Inc., had built production intent metal prototype tooling for the part after casting process modeling showed no filling or solidification problems with the design. Unfortunately, the crack appeared in the first prototype poured.
“We went through several gating design changes and process changes to see if there was a way to eliminate the crack,” said Rob Burita, tooling engineer at Dalton Foundry. “Every time we ran one of the changes, we still had the crack. We started realizing it was something inherently built into the design of the part.”
Due to the nature of the crack and its location at the corner of the rib, Burita suspected it could be caused by stress during solidification. Dalton Foundry attempted to alter the design of the casting on its own by manually filing the core to alter its shape, but the crack persisted. The metalcaster now had the unpalatable task of telling the customer their design was the culprit.
“We knew there would be a pretty significant design change,” Burita said.
Recently, Burita had been exposed to the MAGMAstress extension module for use in casting process modeling, and he knew it would help identify what was causing the stress and how to resolve it. Although it ran MAGMA solidification software, Dalton Foundry did not own that particular stress module.
“After we exhausted all the process, gating and design changes and had a pretty good suspicion it was stress-related, we finally went to the customer and said, based on our experience with the stress simulation, we think we need to run the part in the stress program,” Burita said.
The customer agreed and Dalton obtained the software to use through the design iterations of the part. According to Burita, the first simulation in the stress program using the original design predicted stress in the locations the crack had been appearing perfectly, confirming the origin of the crack was built into the customer’s design and not the result of poor casting practice.
“The big thing was the significance of the change that had to be made,” Burita said. “We wanted to make sure we were confident the design was going to work because we had metal tooling. Before we cut it, we wanted to make sure we were right.”
The engineers discovered the stress in the casting was occurring in the original design because the base was so large and voluminous it took much longer to solidify than the rest of the casting. The bottom was pulling down on the ribs that connected to the top rim, while the top was resisting because it was already frozen solid. This caused a pull where the rib attached to the sidewall, leading to the stress cracking.
“There is nothing you could do in the process to prevent, change or alter that,” Burita said. “The simulation helped prove that to the customer. Seeing the actual picture really sold it.”
Based on the simulation results, Dalton engineers, along with the customer’s engineers, began making modifications to the design, eventually working through 10 design changes before settling on one that indicated zero stress issues in the simulation.
The final, successful gear case housing design featured thinner ribs that were more curved in shape with a little more draft angle, which provided a more solidification-friendly transition from the diameter of the casting’s top, around the windows of the rim, and down to the massive base on the bottom.
Making Up Lost Time
Because the prototypes were going into production, Dalton had originally built metal tooling. The plan was to cut the metal tooling again, but because of the time it took to redesign the part, the customers were in urgent need of the cast parts as soon as possible. The gear case housing was one of five prototype parts Dalton Foundry was producing for the customer’s final assembly. Four of the parts were ready. The customer wanted to move on to the next stage of development for its assembly to stay on schedule for production in late 2016.
Metal tooling modifications in the corebox would have taken several weeks. Like the use of stress simulation, Dalton Foundry turned to new technology to solve the problem. The metalcaster opted to make the mold and cores without tooling using 3-D sand printing.
“Our customer was up against the wall needing parts. We were aware of 3-D printing and that a printed core could be turned around in less than a week,” Burita said. “But this was our first use.”
In the customer’s situation, Dalton Foundry saw how it could take advantage of the ability to print directly from the model without upfront tooling cost, particularly with a prototype that had a history of crack defects. If more changes were necessary, they could be done quickly and a new core could be printed within days.
“Speed is a factor,” said Harold DeVaux, national account manager, Neenah Enterprises Inc. “You can turn these 3-D-printed cores around in a short amount of time. Accuracy is very close to production-likeness. Using the customer’s model, you get a casting that is going to be very close to the production model, plus it allows you to do whatever types of changes are necessary inexpensively.”
The design changes made with the aid of the stress simulation proved on target. The first pour of the gray iron part using the 3-D-printed core package was successful. Eventually, the entire prototype order of 20 pieces was made without defect using the 3-D-printed cores, and the lead-time was reduced to a few days rather than several weeks if produced with metal tooling.
“The success of the part required two leaps of technological faith: stress analysis and using printed cores,” Burita said. “Both worked out great.”
After the first success using the printed cores, Dalton Foundry decided to try its next prototype for a different customer using the same method. Now, just about every prototype the metalcasting facility makes is via 3-D sand printing.
“We had been looking at 3-D printing for awhile and this issue provided us the perfect opportunity to try it out in our own process,” said Chuck Fennell, program manager, Dalton Foundry. “It was the catalyst for us to do all our prototypes now with the new technology.”