The SMC plant of the '90s.

The SMC Plant of the '90s

What will characterize an advanced SMC plant in the 1990s? The last 10 years have brought about improved surface quality, shorter cycle times, advanced presses, and increased automation. The competition of the next decade between manufacturers and different manufacturing processes and materials--especially in the automotive body-panel arena--promises to continue these trends. That's the picture that emerges from discussions with half a dozen of the top molders of SMC automotive parts, and with officials at General Motors, which has been and remains a driving force behind SMC technology.

Companies who wish to lead in the next decade will have to go beyond the role of the traditional SMC processor. Charles Wilkes, v.p. of technology at GenCorp Automotive, Akron, Ohio, feels that there will be two types of reinforced-plastics suppliers in the '90s: "There will be a small number of very large suppliers who will have the lion's share of the business," he says. "These companies will be competent in design, materials, and processing, and they will have the ability to produce a modular part or all of the panels on a vehicle." He expects small molders without a good, strong materials background to be confined to making high-production, low-cost parts requiring very little engineering or materials optimization. "Anyone can buy presses, and the OEM's usually design the parts, so it is the knowledge of mateirlas and how to make them perform best for a given application that will separate the men from the boys," Wilkes says.

Irvin Poston, manager of the Composites Group of the Advanced Engineering Staff at General Motors' Technical Center in Warren, Mich., says that "if the last 10 years were any indication of the next 10 years, there are still considerable process improvements to be made in terms of press controls. Improved parallelism, force velocity control of the presses, and better heat-transfer characteristics of the mold have all been adopted by our suppliers," Poston says.

An important issue of the next decade for both automakers and SMC processors will be reducing lead times. Steve Rinehart, president and CEO of Eagle-Picher Plastics Div., Grabill, Ind., says that the price and quality concerns of the past two decades will be overshadowed by competition in tooling up for a project as quickly as possible. "We must cut lead time," he says, "not necessarily with more capital, but by scheduling and using our resources more efficiently. We can probably cut 25-40% of our lead time that way, although it will be four to five years before we see a dramatic difference."

The calls for statistical process control, just-in-time, and computer-integrated manufacturing of the last decade have led to real improvements in many plant operations. From recent visits to The Budd Co.'s new Kendallville, Ind., plant and GenCorp's new Shelbyville, Ind., plant, it appears that any advanced SMC plant of the '90s will have mechanisms to track parts back to a particular manufacturing shift, molding shot, roll of SMC, and raw-material shipment, by using information maintained on a computer system. Material formulation and molding information will be referenced on a small slip of paper molded into each part, while information about secondary operations will be identified using etched-in serial numbers or other forms of labeling.

"We can generate a 'birth certificate' for every part and keep track of data such as viscosity, molding cycle time and temperature/pressure profiles. These are all recorded, and when the data are analyzed over time, it will lead to much better Class A panels," says GenCorp's Wilkes. The challenges will be in fulfilling the promise of these systems and keeping them up to date as more downstream and secondary operations are automated in the '90s.

SURFACE QUALITY: ARE WE

THERE YET?

Some processors feel that SMC surface quality is as good as that of any competing material currently available, and that there is no more need to improve it than for any other material. According to Phillip Kusky, director of sales, marketing and advanced engineering for Rockwell International's Plastics Products Business, Troy, Mich., "The material chemistry and press technology will provide a surface that is equal to or better than steel, depending on part shape, as an everyday occurrence." He thinks that obtaining surface quality will become easier due to a combination of material chemistry, press parallelism and process controls, and priming/paint systems. "If you make a good part and you can't paint it well, you still have a bad surface," Kusky notes.

Premix/EMS in Lancaster, Ohio, will be concentrating on "Orange peel and paint surface cosmetics, not SMC cosmetics," to improve surface quality, according to Frank Bradish, v.p. and corporate technical director. "Right now, we have materials that meet or exceed the surface smoothness of sheet metal, and I do not believe that it is a prudent expenditure of dollars to continue to pursue the direction."

Although surface quality will undoubledly be easier to obtain in this decade than in the last one, this was made possible by the tremendous effort and commitment of resources for materials and pressing research in the late '80s. "We are just scratching the surface as far as the capability of new presses and material formulations is concerned, and I cannot say that further improved surface quality will come without agony," says Mostafa Ismail, manager of advanced engineering for Complax Corp., Cobourg, Ontario.

To really attack the surface problem, molders agree that there should be a standard way to measure it. GenCorp's Wilkes points out that "part of the problem with surface quality is that there is so much about it that is qualitative. There are guys in automotive paint plants that can see things that the average consumer can't see, and they can reject parts for almost unknown reasons. However, there are a large number of ways that you can get some quantitative measure of waviness."

Like other molders, GenCorp is performing studies with surface analyzers such as the Diffracto D-Sight from Diffracto Ltd., Windsor, Ontario, and the LORIA device from Ashland Chemical Co., Columbus, Ohio. "We are trying to put all of the tools on some uniform basis so we all know what we're talking about," says Wilkes. "A dozen variables can affect surface characteristics in very complex and unclear relationships. We see these methods being used to improve surface quality in a quantitative fashion within the next couple of years."

However, there is still much indecision on what measurement instrument, if any, might become the standard for surface analysis. "There will still be a lot of discussion and disagreement for the next two or three years, because of the already installed base of instruments," Wilkes says. "I can't say today that we have the one instrument that will do the job. It's just not here today, although I hope that in the next two or three years we will reach a decision on one instrument. In order for polymer composites to really make inroads into steel, we're going to need consistent surface quality. To get it, we have to agree on how to measure it so we can attack the variables that affect it."

As advances in materials and processes become harder for the naked eye to see, many molders see die quality as a frontier for surface improvement. "If the material and the press are perfect, you will get an exact replica of the tool surface on the part. I think the majority of the surface improvements in the next decade are going to come from tooling improvements," says Eagle-Picher's Rinehart. The hardness, dimensional stability, surface quality, and heat-transfer properties of tool steel figure to be more closely examined in the next decade.

LIFE WITHOUT IMC?

Opinions differs on the future of using in-mold coating (IMC) to ensure surface quality. GM's Poston says that although it would be desirable to be able to do without it, IMC currently "insures against porosity for a smooth surface. IMC is what brought about a quantum leap in SMC surface appearance in the last 10 years." Efforts to eliminate IMC would require reductions in inherent SMC porosity, and Poston notes that "there is also the argument that a little porosity helps SMC flow better."

Molders agree on the benefits that would result if IMC was not needed in the molding process, but not everyone is optimistic about the prospects for IMC disappearing. "IMC has proven to be a good solution for the past 10 years," says Complax's Ismail, "and I don't see it diminishing, because it provides a guarantee of a good surface with no porosity. As tool surfaces and materials improve, you might reduce IMC a little bit. Ideally, the objective is to produce a good surface without IMC, but I don't see it in the next 10 years."

GenCorp's Wilkes feels that everyone would like IMC to disappear, "from the standpoint that it adds a significant amount of time to the molding cycle. If you go out into a lot of plants today where production started two to five years ago, you will see an IMC process step that adds 30 sec to the cycle time. We now have IMC that has a 15-sec injection and curing time, but 15 sec added to a minute SMC cycle is still a large fraction. We would rather eliminate IMC and use something like vacuum molding. However, there are a lot of problems with molding a very large part without IMC and getting the whole surface perfect. Although we have work going on trying to eliminate IMC, we also have work going on trying to improve it."

Likewise, Al Trueman, director of sales and marketing for the Plastics Div. of The Budd Co., Madison Heights, Mich., maintains that "if our customers want IMC, then we will be experts in IMC. However, it requires the application of surface treatment during the most expensive part of the process, and you cannot achieve cycle times today of under 60 sec when you in-mold coat. IMC will overcome porosity, but you can still have blisters underneath the coating that make the part unacceptable. If you use vacuum molding, you may or may not need to in-mold coat."

Other companies are also working toward making the basic SMC material free of porosity using vacuum mixing and molding of the SMC. Rockwell's Kusky feels that IMC is "a temporary fix" and that his company is "a couple of years away from eliminating it in 1992 or 1993." Premix's Bradish also feels that "IMC will start to disappear in the next 10 years."

Eagle-Picher's Rinehart feels that IMC is going to vanish "rather quickly in new applications." He expects porosity to be reduced and the conductivity of IMC to no longer be required when powder coating becomes economically viable for paint priming. "Powder coating has no voltile organic compounds (VOC's), and it is fairly close to being economically viable. In three to four years, major automotive programs will be specified for powder coating," he predicts. Eagle-Picher is currently shipping a product that is powder coated by PPG Industries, Inc., Pittsburgh, off-site and eventually painted electrostatically. "At the moment, I would have to do both IMC and powder coating on-site, but I would rather build a new paint shop to powder coat 400-600 parts/hr than keep IMC." Because of the elimination of paint sludge and VOC's, Rinehart feels that powder coating "could eliminate 90% of the environmental issues that our company will face in the next 10 years."

SMC AND THE ENVIRONMENT

Environmental issues promise to be as important to SMC processors as they will be to all manufacturers in the next decade. "I don't think we are less safe than anybody else, but with more and more legislation, it is hard to keep up," says Rinehart of Eagle-Picher.

Lowell Bain, general manager of GenCorp's newest SMC plant in shelbyville, Ind., points to the plant's advanced ventilation system as something that will become more common in the future. A current objective at the Shelbyville plant is to eliminate respiratory irritants that may result from dust generated during secondary (trimming, drilling, routing) operations. As both tools and symbols, Bain has shovels placed prominently throughout the plant to support his motto: "Everybody cleans up."

Increased styrene fume-handling precautions are probably also likely. An example would be the fully enclosed sheet impregnation line built two years ago by Finn and Fram, Inc., Pacoima, Calif., for GM's B-O-C plant in Lansing, Mich. (see PT, May '88, pp. 19, 21). The fume-removal system draws heavier-than-air styrene vapors from the sides of the machine and down through its base.

It also seems likely that an advanced SMC plant of the '90s will have some provision for scrap material recycling. Pressure from thermoplastic competition, as well as municipalities' growing concern over vanishing landfill capacity, would appear to make this a near-certainty.

A trial experiment completed last month and currently being studied by the SPI's SMC Automotive Alliance, Troy, Mich., looked at the products generated by subjecting SMC to thermal decomposition in the absence of oxygen (pyrolysis). This produces "pyro-gas," which contains methane, among other substances. According to Donald Norris, general manager of molding materials for Eagle-Picher's Plastics Div., the gas may be used to sustain the process, and some may be left over for other use as fuel. During cleaning and scrubbing, condensation occurs, which may yield a usable oil. The ash consists of ground limestone, glass fiber, and carbon. Rinehart says there is potential in using the ash for SMC formulations, polymer concrete, or roof tiling. He thinks it will be "a year to two years before we will actually see a site doing this in the Midwest."

Another possibility is for SMC molders to simply incinerate scrap as boiler fuel. There has also been speculation that cured SMC can be ground up and used as filler in fresh compound. However, some sources raise the objection that this might not yield the tight particle-size control normally required of SMC fillers.

CYCLE TIMES: HOW SHORT?

During the '80s, increasing productivity meant taking minutes out of the molding cycle, whereas the objective of the '90s will be to trim seconds out of the cycle. Budd's Trueman predicts that we will see a 40-sec cycle time without IMC in the next decade, and GenCorp's Wilkes expects a 60-sec cycle time with IMC on large Class A parts by 1993.

Molders agree that cycle times will continue to decrease, although the incentives for reduction are changing. While the molding cycle was previously a processing bottleneck, recent investments in advanced presses have insured that the molding cycle will tie up more capital than time in the '90s. "The more capital you spend on a facility, the less you can afford scrap--not that you can ever afford it," Budd's Trueman said. "A fully automated facility with automatic charge preparation, loading-unloading, control of the press cycle, recording of data out of the press, deflashing, and transferring, will tie up a higher percentage of capital.

It also appears that faster presses will be no more important than the controls that are on them and the automation that surrounds them. Eagle-Picher's Rinehart feels that current robotic loading/unloadng equipment is inadequate to address the industry's needs. "Over the next five to six years, we will see some much more creative thought applied to automation, which will help take some of the cycle time away. Very little additional reduction will come from fast-acting presses," he says. Trueman expects charge preheating to continue at Budd, along with the possible elimination of manual edge-trimming of charges from automatic slitters in the '90s. "Once we have achieved that, there would be no reason why we could not go from a prepared charge right into the mold," he says.

"Cutting charges and loading/unloading will become a bottleneck after faster cycle times are achieved," according to GenCorp's Wilkes. "Working around a hot press is a tough job that people really don't enjoy, like standing inside a paint line. It makes sense to automate it," he says.

Although shorter cycle times will boost productivity, the secondary operations must keep pace, and sufficient demand volumes must be present. Decreasing cycle times further may create new bottlenecks in SMC processing downstream. "We can mold in 45-50 sec with fast-acting presses, but it may not make sense to go faster now because our secondary operations for finishing and bonding would have to be done in the same time frame to make it worthwhile. Right now, we are tuning everything to the 60-sec cycle time," Kusky says.

Some new technologies that can speed up secondary operations have already emerged. For example, adhesive bonding times have been cut from 6-1/2 min to 45-50 sec with the "Budd-Tron" radio-frequency curing system developed by Budd, and with the hot-air-impingement system developed by Heat Transfer Technologies, Van Nuys, Calif., in conjunction with RP/C Machinery Corp., Howell, Mich. (see PT, Nov. '87, p. 11).

A new high-speed trimmer/router designed specifically for SMC parts, and an automated flexible automation cell that would handle trimming, routing and inserting without the expense of dedicated fixtures, are both described in the news story on p. 25 of this issue.

To achieve the kind of cycle times they are looking for, Rockwell needs "total automation," according to Kusky. "All the SMC molders are pretty much at the same point with the current automation technologies of robotic bonding, loading/unloading, transfer lines, and automatic paint systems. In the next generation, we want to put material in one end and have a part come out of the other, with practically no human hands touching the part." Kusky says that this is already being done in Europe, although on a much smaller scale. By the end of the decade, he predicts that Rockwell will be close to doing it on high-volume jobs. "In five or six years, we will see some major changes in press conrols, computer checking of part-to-part variances, and centralized control above all the presses at a location isolated from the molding area," he says.

Another company working on "total automation" is Goodyear Tire & Rubber Co. (see PT, Nov. '88, p. 19). At its new Composites Technical Center in Jackson, Ohio, the company is developing a full production-scale pilot line that takes material from SMC roll to bonded assemblies ready for painting in 1 min or less, fully automatically.

Some molders stress that 1-min cycle times are definitely not the standard of the industry today, expecially for large, complex parts. "Some fairly simple parts are made in 1-min cycles." Wilkes says, "and development laboratories can do it, but even with a new production like the General Motors APV minivan, parts are not being made in any of the four suppliers' plants on a 60-sec cycle." He predicts that well into the '90s, we will still see plenty of parts being made in 2-3 min cycles.

Complax's Ismail says that the 1-min cycle remains elusive because new plants have the presses or the material-handling capability necessary to achieve it. "In the next 10 years, we will see a larger percentage of molders getting 1-min cycle times via faster-curing systems, better presses, and better material handling. By the end of the decade, I hope that 70-80% of SMC operations will have achieved the 1-min cycle," he says.

"We will see automation systems that will place charges and take the part out quicker than the present manual load/unloading which is still prevalent. Even existing automated systems are too slow. secondary finishing will not disappear, but the objective will become to produce a mold design that minimizes the need for finishing." Ismail says.

Premix's Bradish feels that "the 1-min time frame is a reasonable objective, but we've got to focus more on process consistency than cycle-time reduction. I don't think that going from 1 min to 45 sec in the next 10 years is going to be a critical factor," he says.

Processors expect presses to become even faster, but also to contain more in-mold instrumentation and microprocessor closed-loop control. Advanced parallelism figures to be a standard feature from now on, and everyone expects faster curing materials to appear, either through better use of existing catalysts or development of new ones. Companies will attempts to balance the need for a faster cure with the need to completely fill the mold. "We have had the advanced chemistry ability for years, but the press technology did not allow us to use it. Lower cycle times really represent a convergence of press and material technologies." according to GenCorp's Wilkes.

UPSTREAM PROCESS CONTROL

Of the 150 or so variables to control in SMC manufacturing, a good number of them occur in the sheet impregnation stage. It seems clear that more on-line sensing, and even closed-loop control, will be applied to the SMC machine than has been standard to date. This direction is already apparent in the sophisticated SMC machine that was recently installed at GM's Lansing B-O-C plant. It includes a load cell at the doctor blade to maintain consistent past level, a gamma-backscatter gauge (adopted from thermoplastic extrusion) on a traversing mechanism to monitor sheet weight per unit area, and continuous weighing of each box as the SMC is festooned into it.

One of the pioneers in using gamma-backscatter gauges is Premix Inc.'s Molding Compound Div. in N. Kingsville, Ohio, which makes all the compound for SMC molding both at that plant and at Premix/EMS plants in Lancaster, Ohio, and Portland, Ind. The nuclear gauge on its automotive SMC machine takes 240 readings/min, and the company has developed software to integrate the gauge into closed-loop control of the glass chopper and paste pump to achieve a consistent glass ratio. The gauge can send a signal to shut down the line if the ratio wanders out of range. This has helped provide a 60% improvement in sheet consistency and 70% reduction in start-up scrap, according to v.p. David Clavadetscher.

Premix now plans to put as many as five nuclear gauges for closed-loop control of another SMC line. Besides providing continuous readings of total mat weight and glass content, the gauges will also regulate the upper and lower doctor-blade settings.

WHAT'S NEXT: COMPUTER PROCESS

MODELING

SMC processors agree that the use of computer modeling, simulation, and control will continue to increase in the '90s. Some of the advances will come from the transfer of technology from high-performance aerospace applications to high-volume automotive applications. For instance, Rockwell's automotive group is working with the firm's West Coast space science center on the analysis of adhesives for bonding inner and outer SMC parts. "There has not been enought work yet on the relation between adhesive and SMC parts, and since aerospace can not afford to make many mistakes, their analysis methods will help the automotive end," Kusky says.

Making a good part depends on flow, and molders agree that there should not be any black magic to it, that material flow should be predicted by computer, as is increasingly common in injection molding today. Rockwell has spent the last year and a half entering information into a process-development computer program from Ultramax Corp., Cincinnati. "We are using the software for SMC processing, to accept inputs from daily activities, such as the use of the LORIA surface analyzer," Kusky says. "The program will allow us to perform 'what-if' analyses without actually running parts." Rockwell hopes the use of the system will yield the best parameters for press control and optimal material loading patterns.

Rockwell is not alone in computerizing the SMC process. GM's Warren tech center has worked with Ultramax and has also developed its own heat-transfer modeling software for predicting temperature patterns across the mold surface (PT, May '88, p. 25). GM has made this software available to its custom vendors through International Technegroup Inc., Milford, Ohio.

And GenCorp is working with its own mold-analysis program, according to Dr. George N. Hartt, v.p. for technology in reinforced plastics in Marion, Ind. "It was developed by Charles Tucker at the University of Illinois, and we have customized it to our needs. We have tailored it for IMC, SMC and RTM so far, so we can do all three mold-flow analyses."

MATERIALS TRENDS

It appears that polyesters and vinyl esters will remain the materials of choice for SMC body panels, although molders expect use of phenolics and epoxies to grow in acceptance for under-the-hood and structural applications, respectively.

Most molders do not see many prospects for non-glass reinforcements making inroads in SMC, for reasons of both cost and performance. SMC continues to compete directly with steel for automotive applications, making fibers other than glass prohibitively expensive. Because the temperature in a paint oven can reach 400 F, a lot of organic fibers are unsuitable for automotive SMC as long as current painting methods continue. The fibers that can withstand these conditions, such as high-performance carbon fibers and aramid, are suitable for some structural applications, but are too costly.

However, GenCorp's Wilkes does not rule out the possibility of small amounts of carbon, aramid, or extended-chain polyethylene (Allied-Signal's Spectra) fibers being used in high-stress portions of a part. "Right now, such use is miniscule, but it will boom in the mid-'90s," he says.

"Continuous-glass XMC could also grow faster than regular SMC," Wilkes adds. And Complax's Ismail notes, "XMC could work in some structural, non-appearance parts like door inners or door frames. These parts are still very heavy, and although the piece price might be more expensive with XMC, the weight could be reduced."