Silicone is stuck on automotive

Soft-touch feel, low-gloss looks, durability, and light transmissivity for clear backlit graphics - whiter than white by day, and brighter than bright by night. These were among the most important design and performance criteria driving the Appearance Technologies (AT) group of Delco Electronics (Kokomo, IN) on a worldwide search of epic proportions for new technologies. It began about four years ago. Delco Electronics was looking for the best materials to use in manufacturing button clusters for audio systems and other trim, appearance, lighting, and color products used in GM North American vehicles.

Also, it was looking for more cost-effective manufacturing methods than costly secondaries like painting and laser-etching to make these quality products. Developmental fine tuning continues today, but the strategic consortium of experts at Delco Electronics is certain it has found the answer - two-shot, rotary-platen overmolding of liquid silicone rubber on engineering thermoplastics in two-shot molds. Finished parts are produced right out of the mold.

General Motors recently announced that Delco Electronics is being integrated into Delphi Automotive Systems. Like Delphi, Delco Electronics must compete with the rest of the world for GM's business, while competing for business with other car makers, and companies in other markets, worldwide. A lot is riding on this project.

Charlie LaRose, Delco Electronics AT's senior project engineer, reminisces: "None of the processes most people were investigating at that time met the true soft-feel expectations our customers want the consumer to find in a luxury vehicle. We started evaluating different materials, and came across silicones. You can add a lot of different fillers, textures, colors, smells, and bat wings to silicones that can open up your design parameters, and solve manufacturing problems."

Delco Electronics investigated overmolding silicone onto plastics like PC, "and found out that silicone did not chemically adhere to plastics. Then we called on George Kipe. We explained to George what we had in mind, and George probably thought, 'What planet do these guys live on?'" George Kipe is president of Kipe Molds Inc. (Placentia, CA), specialists in tooling for liquid silicone injection molding. Actually, they hit it off very well - maybe because LaRose, a self-styled "old grassroots guy off the floor," is a moldmaker himself.

Silicone has a high-quality, soft-touch feel. "It's one of the few materials that closely simulates the feel of human skin," Kipe reminds us. It has an extremely low gloss reflective index and very high light transmissivity, based on tests developed by Delco Electronics' extensive R&D. And, unlike the thermoplastic elastomers, rubbers, olefins, and urethanes among the reigning champs in the soft-touch league, thermosetting silicone shrinks equally in all axes, causing zero knit and flow lines, sink and gate marks, and blush - defects paint can't hide. Also, silicones can be molded at relatively low pressure, 90 psi, so highly detailed products with good mold replication can be manufactured, like little bumps that little buttons can go under.

But silicone doesn't adhere very well to engineering thermoplastics. Silicone's a thermoset. Heat cures it. The high temperatures required to mold engineering plastics, like light-transmissive nylon, nylon 6/6, PC, and Valox, result in hot substrates that can kick off the silicone prematurely when it comes into contact with the substrates' surfaces. And when you heat silicone up to around 230F to initiate cure, you drive volatiles off some engineering plastics, like nylons, that can inhibit the curing of the silicones. Most had written off the idea. "We needed to come up with a harmonious system," LaRose says.

Trials and Tribulations

New silicone materials that "liked" engineering plastics would have to be formulated. LaRose; his Delco Electronics partner Greg Crater, development engineer; and Kipe unsuccessfully queried a few resin suppliers, including those that make both silicone and engineering plastics. Some denied their requests even after exhaustive trials, LaRose recalls. "Then we went to Wacker [Wacker Silicones, Adrian, MI] and addressed the issue. They tested every conceivable silicone formulation with dozens of thermoplastics, running so many DOEs [designs of experiments] that it would hurt your head just to think about them. They came back with a 40 durometer material that shows a great deal of promise."

Meanwhile, Kipe was building experimental tools for molding black silicone, white silicone, and engineering plastics. "They wanted molds to test different design iterations that could be manufactured in different areas of the world," he says. "It was the rear-seat audio control for sport utility vehicles, and it had a lot of shapes - a very intricate design. They wanted to try to mold a window, a white silicone with a legend, and then black silicone over the whole thing except for the legend. I built four silicone molds and three plastics molds. It took the better part of a year to finish."

Initial adhesion trials were run at Kipe's shop, converting his custom-built 80-ton Engel to running plastic and silicone, and back again. Proving out manufacturability was the goal, so Kipe refrained from adding in too much complexity, opting for direct and tab gates, rather than his cold runners, for instance. There were a few shutoffs used to prevent overspill, one mold filled from the back side, and one mold had a hot drop. The trials were exhaustive. If Kipe found out anything, it's that plastics are weird. "I found that certain colors of plastic adhere well to certain colors of silicone. And in gate areas, I found different adhesions than in areas farther in at lower pressures. Also, you can drive silicone into plastics like a glass-filled nylon. I had to use a lower temperature cure."

"We were able to give these molds to Wacker," says Delco Electronics' Crater. "Wacker could shoot the parts in real time and adjust its formulations." For molding systems, Delco Electronics is thinking about three cell designs: a horizontal rotary-platen press with 90¡ liquid silicone injection for low volumes; a horizontal with parallel injection units for bigger rotary molds; and a vertical press with a rotary table for loading premolded trim plate inserts, and with onboard secondary finishing stations.

The Road Ahead

The potential for high-volume application spin-offs from this project is enormous, both in automotive and nonautomotive markets. Another OEM is looking at the system as a means of molding silicone directly onto its connectors. B-pillars and air bag covers are good possibilities, as are valve covers and engine-seal gaskets - anything that needs a gasket. It could change how instrument panels are made, with a skin of soft-touch silicone shot over an entire hard plastic substrate with everything already built in. It could eliminate PVC overmolding and painting and problems like PVC outgassing and UV instability in interiors. Consumer electronics, telecommunications, and medical device markets also could profit from more protective products with no sharp edges and with that soft, high-quality feel.

Crater says the biggest problem explaining the benefits to more immediate customers is in getting them out of the paradigm of making one-to-one cost comparisons. Those involved in conventional paint and laze systems are used to shooting and painting one button at a time. Tooling costs are high for the silicone overmolding process, but when the tooling cost is compared to all the secondaries in paint and laze, and when you see that you're getting all of the buttons in a cluster at once, finished, right out of the mold, the cost savings become obvious.

Silicone costs anywhere from $5.50 to $9.50/lb, but only a thin .5- to 1-mm-thick layer is applied in overmolding. Paint costs $60/gal. Coats are some .003- to .006-inch thick, and they can chip. Then there's all that in-process inventory and those part numbers to keep track of, hazardous materials, and contaminants, and recycling concerns. Then compare the floorspace of a painting area to the floorspace of a molding machine cell.

There are no parts in production yet. That's how new it all is. Parts are being tested in Arizona's hot sun for fading and adhesion after weathering. Once in production, Delco Electronics intends to use cold runner technology. Any candidate who joins the consortium and builds the production tools needs to have the ability to design to Delco Electronics' parameters and simultaneously do the mold design. A product will be selected, a manufacturing cell created, and then the process will be proven out for about a year. The dollars and cents cost reductions that have been achieved to date are closely guarded secrets, but both LaRose and Crater agree that the numbers very strongly support their case. LaRose sums up, saying, "It's the Superplug of interior switches."