rss icon  Keep up with Madison Scholar. Click the orange icon for the RSS feed.

A Mold For Teaching
Layer By Layer, Rapid Prototype Machine Will Build A Cell

By Eric Gorton, JMU Media Relations

The RP machine uses computer drawings like this to turn powder into solid 3-D objects. Click on the image to see a larger version.

The way Dr. Dawn Tamarkin envisions it, a model cell made up of plastic parts that can be handled and manipulated would be an effective aid for teaching high-school and college students some basic cell biology.

The National Science Foundation agrees with Tamarkin, a biology instructor at Springfield (Mass.) Technical Community College, but there was a hitch when she tried to secure a grant to make her idea a reality: The grant would not cover the $250,000 cost of mass producing the parts. The NSF "wanted us to be able to test the products in a classroom, but they did not want to pay for production of the product," Tamarkin said.

That decision left Tamarkin in a bind. Her college has machinery capable of producing a few models that she could demonstrate at schools, but it could not produce enough for the schools to have their own. "And there are definitely schools that would benefit immediately from these, like schools for the blind and schools for people with learning disabilities," she said. "So we wanted to have some available … so I could get it to them while we figure out how to really mass produce these down the road."

Tamarkin got the break she needed when an NSF program director referred her to other successful grant candidates for help. After a few more referrals, Tamarkin found the answer to her dilemma at JMU: An integrated science and technology lab that can produce the parts she needs without exceeding the $400,000 grant.

"What (JMU does) is make the molds differently, they make the molds in a more affordable manner by prototyping the mold," Tamarkin said.

The cost difference even surprised Dr. Ronald Kander, head of ISAT and a founder of the Center for High Performance Manufacturing, a collaborative established in 2001 by JMU and Virginia Tech. "We were amazed to find out that cheaper was actually under $100,000," he said.

The lower cost results from reduced labor and reduced time to create the molds. To produce Tamarkin's cells, metal molds have to be made first. Once the molds are made, the cell parts will be created by injecting polyethylene, a soft plastic used in products such as milk jugs and stadium cups, into the metal molds. Using traditional manufacturing methods to create the molds, it would take a tool and die machinist hours of manual labor to create each one, especially those for the most intricate pieces, Kander said. With rapid prototyping, a laser builds the molds layer by layer based on a 3-D computer image. Once the molds are created by the rapid prototype machine, they are put in a furnace for two days and infiltrated with bronze to harden.

Lab technician Dwight Dart pours powder into the RP machine. The machine uses a laser to build an object out of the powder based on 3-D computer images.

"During those two processes, there's no person required to be doing anything with it, so the labor is really getting it started (in the RP machine) and then pulling it out and putting it in the furnace and then pulling it out of the furnace," Kander said. "So you're not talking about three or four days of continuous labor."

Another benefit, Kander said, is that the molds can be made in three or four days rather than months that it would take making them the traditional way. And since the cost is so much less, "you can afford an iteration or two in there if necessary, which in a traditional manufacturing of the mold, a hardened steel mold, if there's an error in that mold, you've sunk $50,000 into it."

The rapid prototype process does have some challenges and a traditional manufacturer may still be needed to make molds for the smallest parts due to their fine detail. The main difficulty with the RP process, said RP lab technician Dwight Dart, is that the molds are very brittle when they come out of the machine.

"You have to be meticulous with the extraction part," Dart said, or the parts could crumble. "That's almost a day where you just take your time because the part is really brittle when it's first done and it has to go over into the oven and be infiltrated with bronze to harden."

Objects produced in the RP machine start out as batches of nylon, elastomer or metal powder (steel powder will be used for the cell molds). The RP machine uses a laser to sinter the powder into the shape of 3-D objects drawn with computer-aided design (CAD) software. "The laser knows the shape of that cross section, so it melts the powder together to form the shape," Kander said. "So everywhere else in that layer that you haven't put the laser is loose powder. So when you're done, you have a cake of loose powder and somewhere in the middle of that cake is your solid part. And depending on how fine the features are on that solid part sort of dictates how careful you need to be. If it's got a lot of lattice and shapes and little holes and stuff, then you've got to be very careful."

Added Dart, "What will happen is just a little tiny piece will chip off and that part is shot."

Dart said he's most concerned about the outer shell, the largest part of Tamarkin's cell model, because its wall is very thin at the top. Once the molds are hardened, however, they are strong enough to be used for at least a limited time, which in many cases is good enough.

Despite the potential challenges, Dart said he is eager to get started and he anticipates successful completion.

"Now it's true that these parts won't be as robust as say a hardened-steel mold manufactured in the traditional manner, but that's why, if you're going to make a million of something, you would still eventually have to make a hard mold for it, but for what's called limited-run manufacturing, these molds make a lot more financial sense," Kander said.

Tamarkin said each cell will consist of about 40 pieces and she's hoping the molds will be strong enough to make several hundred kits. "We'll see. It gets us started and it's exciting too because we will be finishing the whole grant by making sure the people in my community are interested."

The project is exciting for Kander because it's the first large, long-term grant-funded project the lab has secured. "It's our first real hit," he said. "It's the first time we're getting to really do exactly what we built the lab for."

Tamarkin is now working on the CAD files for the cell parts. If all goes according to schedule, the cell kits could be ready in spring 2008.

Published June 2007 by JMU Media Relations