Click here to see this story as it appears in the April 2017 issue of Modern Casting.

Marshall Miller, supplier development manager at Flowserve Corp. (Rock Spring, Georgia), has a few simple words of advice for metalcasters regarding 3-D printing patterns in plastic: “Just try it.”

Fused deposition modeling (FDM)—also known as fused filament fabrication (FFF), in which a plastic filament is put down layer by layer according to a 3-D model to create a geometric shape, has been around since the 1990s. It has been adopted most widely by the investment casting sector to create patterns, but the technology can be used to make patterns for sand castings, as well.

In the last five years, 3-D printing equipment and material have become more readily available and lower in cost—low enough for a wide range of companies, groups and individuals to purchase and tinker with on their own. Googling “FDM printer” will yield results for printers ranging from $200-$6,000. It coincides with the maker movement, a culture in which technology and do-it-yourself have intersected. And it opens the door for metalcasters to make an entry into the 3-D printed pattern fray.

“FDM machines have become really popular in the last six years, and there’s a lot of open source communities that do a lot of research,” said Matthew Loerwald, foundry engineer, Martin Foundry (Dallas). “You end up with the ability to put together these machines now because people have been working on it and the software for a long time. You can do it low cost, the material is cheap and widely available, and you are not locked into a proprietary system.”

Martin Foundry assembled its own 3-D printer using information available in the open source environment surrounding FFF and FDM. The casting facility, which produces power transmission products like pulleys and gears, thought it might have a use to make quick patterns for replacement parts.

“We knew it would work with a one-off pattern,” Loerwald said. “There is little investment in it besides time. It wasn’t a direct cost. It has worked well.”

Lessons Learned
As a supplier development manager for Flowserve, Miller has been monitoring the progress of 3-D printing and how it might help get parts to his company faster. Flowserve makes pumps, valves and seals for the power, oil and gas, chemical and water industries. Like many companies, Flowserve first used FDM and FFF for development of its parts made via the investment casting process. But in recent years, it has worked with suppliers that are using it to produce patterns for sand molds.

In FDM, a 3-D model is “sliced” in a software program and the output is loaded into the printer, usually via an SD card, which typically uses the same software as a CNC machine. Different polymer material can be used, depending on the application. Polylactic acid (PLA) is perhaps the most common filament and is often used for investment casting and sand casting patterns. Other plastics, such as ABS or nylon, require a heated chamber for printing but are more rigid for sand mold tooling. All are available with “fill” such as graphite, fiberglass or even wood to increase strength and durability.

As companies experiment and test the equipment, better and more effective methods and practices are starting to emerge. Miller recalls an early project to print a replacement tool for a pump that was experiencing core tilt that resulted in 90% scrap.

Flowserve provided a master 3-D model and the tooling was printed via FDM. The projected timeline was four weeks, including 12 days for tool printing. In reality, printing the tool took 24 days because of two failed prints. Ultimately, the castings were produced and verified, with lessons learned about how printed polymers behave in the equipment.

Miller points to two later examples to show how the methods his suppliers are using have improved by using what he calls a “system.”

“Only print what you need,” Miller said. “You are building, you are not cutting away. A lot of foundries run inserts in their pattern equipment. So, you can make a standard size insert for a picture frame or bolster where you drop your print into the frame. You are not building all the other portions of the tool that are standard. It drops the build time a lot.”

Flowserve has worked with at least two suppliers who have incorporated a system Miller describes. In one case, the supplier printed an insert overnight in ABS plastic for $250. For a separate job at another metalcasting facility, castings were provided within a week of receiving the model by printing only what was needed in PLA.

Miller said there are two ways to print tooling for metalcasting. Direct printing means printing or building what will be inserted into the picture frame or bolster. Indirect printing is to mass extrude, or quickly print, a large near net shape of the object that is then machined into shape.

“Both ways work, it just depends on the part and what you want to do,” Miller said.

Do-It-Yourself
Martin Foundry has found success with its scratch-built printer for made-to-order and replacement parts.

“It’s been successful on orders that would otherwise require milling out of a solid sheet of UHMWPE (ultra high molecular weight polyethylene),” Loerwald said. “It’s not quite as fast, but the material cost is next to nothing and we don’t have to buy any tooling or do it by hand.”

Loerwald said that so far, the metalcasting plant will use the 3-D printer on about two jobs per month, but on each of those jobs, the company has saved at least $1,000.

In one particularly successful part, Martin Foundry printed out a piece to modify the aluminum matchplate for a bearing cap.

“The customer wanted to put an oil groove on the bearing cap but we didn’t want to permanently modify the pattern,” Loerwald said. “We printed the corebox on the printer and produced it in two days rather waiting a week for new tooling.

“It works great for when the patternshop is backed up. You can draw up the patterns, press print and move on.”

Stocking the Tool Box
While 3-D printing seems like a natural technology to market to metalcasting, Miller admits many FDM printer manufacturers and equipment users are unfamiliar with the industry and the process. He has invited a few of them to casting facilities and exhibit shows to introduce them to the market, its opportunities and its needs.

“Most of the folks that sell the printers have never been in a foundry and have no idea what happens in a foundry,” he said. “If you are looking for tools like these 3-D printers, I suggest engaging the people who you might consider buying a machine from and ask them to do a test print for your foundry. Once the printer people understand what we need in terms of surface finish, size and tolerances, the relationship works pretty well.”

Miller said he currently sees a break-even point for his parts to be two or three pieces. After three, it usually becomes more cost effective to make a tool. This basically matches up to what Loerwald has seen at Martin Foundry, as well, but with so little overhead, the 3-D printer is a handy machine to have.

“It’s not a magic machine, but it’s another good tool to use,” Loerwald said.

Miller agrees it is one tool of many at the disposal of manufacturers. But for the price point, the barrier to entry for FFF printing is small, and he urges his suppliers to “just get one.”

“There is a cafeteria of choices. There’s print for investment, print for sand, other additive manufacturing methods—none of them are mutually exclusive,” Miller said. “You can put and use these together. That is the fun part. In the end, a company has to be good at 3-D modeling because everything rests on the 3-D model.”