The carbon fiber roof on the BMW M3 CSL is made with the auto industry's first automated production process for the material. Weighing in at only 6 kg., the roof is more than 50% lighter than one made of steel.
Fiberforge has developed a high-speed carbon fiber part manufacturing method that could help bring the material into the mainstream. The company's first project is the strut tower brace shown here. Plans call for it to be ready by the end of the year.
It has become fashionable for automakers to disperse bits of carbon fiber reinforced plastic (CFP)—real or faux—around the cockpits of their sportier models. Gear shift knobs and center console panels proudly display the familiar exposed weave in an effort to capture some of the cachet of F1 racers and fighter jets. But outside of these decorative touches, the use of CFP in automotive (below the rarified levels of cost-is-no-object supercars) is practically nil. The chief reason is cost. According to Dr. Rudolf Stauber, vice president, material development, at BMW Corp., a CFP component can cost roughly five times more than the same part made from aluminum. A big part of the difference is in raw materials. Stauber estimates that the combined costs of the carbon fibers themselves and the resin matrix needed to hold them together is in the $30-$35/kg. range. (For reference: aluminum runs about $5-7/kg.) What's more, "Scrap can be around 30% of the material used," says David Dwight, vice president of marketing for Fiberforge Corp. The reason for the high scrap rate is that most companies work with woven carbon fiber fabric that comes in standard roll widths, and any dimensional variation from the standard, which is practically inevitable, leads to waste. But inasmuch as there is no shortage of either carbon or plastic resins, raw material costs could be brought down over time with increased demand.
Even if material costs are reduced, there are still problems. One is that most automotive applications of CFP involve the time-consuming process of hand-positioning every layer of fiber and then placing the resulting resin-impregnated parts in an autoclave for a lengthy bonding process—resulting in a cycle time measured in hours. The aerospace industry has succeeded in automating much of the lay-up process, but given it's two-weeks-per-plane production schedule there is no impetus to speed the process to anything close an automotive mass production level.
BMW and CFP. Still, CFP's combination of light weight, strength and stiffness is so appealing for automotive applications that the search for a viable volume production method is getting increased attention. Among automakers, BMW may be the farthest along in making mass produced CFP components a reality. Its Landshut, Germany, facility serves as both a laboratory and pilot plant and houses what BMW calls "the world's first highly automated production process for CFP body components." To date, Landshut's largest project has been the 1,400 roofs produced for the M3 CSL coupe in 2003. A two-step process was used to make the roofs: five layers of carbon fiber are automatically positioned to form a mat that is then placed in an 1,800-ton press and injected with transparent epoxy in a resin transfer molding (RTM) process. The molded panel is then removed by robot and given a clear coat. BMW claims that by automating the process it can make a roof in less than 20% of the time needed by traditional CFP production methods. But according to Stauber, that still brings total cycle time in at a lengthy 30 to 40 minutes. To reduce that time the company is working with plastics suppliers to develop resins (the focus is on polyester) that can be distributed more quickly during RTM. "To make the infiltration of resin with the carbon fiber is a procedure that up until this time has taken 20 to 25 minutes," explains Stauber. "There is a goal to reduce that down to 12 to 15 minutes. But these resins still have to be developed. It seems to be realistic that within the next five years we will achieve this goal." Stauber says other problems BMW is working on are developing effective repair technologies; defining the components best suited to use the material; and figuring out what to do with them at the end of their useful life, since recyclability is not a strong point for CFP. However, he sees demand for carbon fiber increasing over the next decade and predicts a corresponding drop in raw material prices that could bring the fiber's price down to $10/kg., making the cost factor much less daunting.
Fiberforge. Halfway around the world from Landshut in the automotive mecca of Basalt, Colorado, a start-up company called Fiberforge is also trying to solve the problem of mass-producing CFP structures. It has patented an automated lay-up and thermoforming process that it claims can achieve CFP parts with 80% of the performance of comparable aerospace components at 20% of their cost. That translates to a finished part cost of about $20/lb., though Fiberforge's David Dwight admits that in order to get the automotive industry to take notice, "We need to be in the $10 to $12/lb. range."
The heart of the company's approach is a high-speed automated lay-up process that eschews the use of woven mats in favor of the precise positioning of individual carbon fibers. A proprietary lay-up head design applies 50-mm wide strips of fibers while concurrently impregnating them with resin to form a "tailored blank." Dwight says this method offers several advantages over working with woven fabric. First, scrap is largely eliminated since the head puts the fiber only where it is needed. Second, the precision of the application allows designers to put the exact number of layers at the exact degrees of angle necessary to optimize strength, and to obtain a high fiber to resin ratio (50-60% fiber) which further enhances part strength and stiffness. And since the heads are designed to work with a variety of fibers, material costs can be reduced by producing hybrid parts that combine cheap glass fibers in non-critical areas with carbon fibers.
Once the blank is finished it is moved to a press where it undergoes a traditional thermoforming process. Dwight explains that Fiberforge chose thermoforming in part because it is similar to steel stamping and would therefore be familiar to automotive engineers. But its chief advantage lies in cycle time. "Thermoforming has the greatest development potential for a high-speed process," says Dwight. Currently, at Fiberforge's newly opened pilot plant it has achieved thermoforming cycle times of four minutes, but as the technology is tweaked Dwight says, "We believe we can get in the realm of a one-minute cycle time." At that speed carbon fiber parts in mass production vehicles could move beyond fashion to function.