Low-pressure alternatives for molding large automotive parts.
Low-pressure injection molding has lately attracted wider interest as a potential alternative approach to making large automotive (and other) parts. Attractive benefits like reduced tonnage requirements, less costly molds and presses, and lower stresses in the finished part have led a number of automotive suppliers in the U.S., Japan. and Europe to adopt various low-pressure molding techniques for large interior and exterior parts such as bumper fascias and door panels.
Yet, despite the obvious potential benefits of low-pressure molding, an aura of mystery still surrounds this technology. To some extent, continuing puzzlement on the part of many processors has hampered the widespread adoption of low-pressure molding, for which technology has been available for years from a number of sources. This article attempts to clarify the options available.
One of the most widely publicized, but perhaps least understood low-pressure injection molding technologies is the Thermoplastic Solid Molding ("TSM") process of Hettinga Technologies in Des Moines, Iowa. Dramatic claims are made for what reportedly is a radical departure from conventional high-speed, high-pressure injection molding. That, and the company's desire to protect its proprietary know-how, have prompted considerable speculation and some skepticism among molders and others in industry. Hettinga says it has more than 300 machines installed worldwide, most of them in Asia and Europe.
Company founder Siebolt Hettinga claims that his low-pressure process permits clamp pressures to be reduced as much as two-thirds below those of conventional high-pressure molding. He says his system typically requires clamp tonnage of about 300 psi for unfilled materials and about 600 psi for filled resins.
The Hettinga process is said to be based on carefully controlled, low-speed injection, which reportedly can fill a cavity as effectively as conventional high-speed injection--but without the accompanying stresses in the part. The keys to the process, claims Hettinga, lie in a number of details of equipment and process design that come into play "between the hopper and the finished part." Here is the extent to which Hettinga explains these details of his process.
Hettinga characterizes his system as "coupled" molding. This differs from conventional injection molding, which separates or "decouples" the injection part of the cycle from the packing and holding part of the cycle. In contrast, Hettinga fills the cavity slowly, and claims not to pack out the part--ever. The centerpiece of the system is its controller, which programs screw backpressure, injection pressure, and injection speed.
During plastication, the Hettinga process is said to be more akin to extrusion than conventional injection molding. In extrusion, all the resin is subjected to the same plasticating conditions as it moves continuously toward the die, resulting in a high degree of melt uniformity. This may not necessarily be true of the first and last portions of a shot plasticated on a reciprocating screw. But Hettinga's process allows programming screw rpm and backpressure to control the viscosity throughout the entire shot. One particular point Hettinga noted is that the screw speed or back-pressure may be deliberately varied so as to raise the temperature of the last portion of the shot. Thus, the last melt to enter the mold, which has to travel the farthest, has a better chance of staying fluid longer, even though it's traveling slower and at a lower pressure than in conventional injection molding.
Then the fully plasticated and uniform melt is injected gently into the mold cavity. The injection speed is adjusted continually and automatically to maintain proper forward pressure and speed. Forward movement of the screw during injection is controlled by a linear transducer, which divides the shot into 4096 increments. This reportedly provides excellent resolution for shot-to-shot consistency, despite the lack of a cushion.
FILLING WITHOUT PACKING
According to Hettinga, the pressure does not pack back to the gate during his process--which he adds is the main principle behind low-pressure molding. In conventional molding, he explains, pressure begins at the end of fill and backs up to the gate. The Hettinga process, by contrast, quits filling before the pressure has a chance to back up to the source. Hettinga concedes that freezing off at the gate does constitute "some type of holding," but it is minimal.
Slow filling of the mold is said to avoid the stresses encountered during conventional injection molding. During highspeed molding, explains Hettinga, the melt front starts to break up as soon as it enters the cavity. As more pressure is applied, hydraulic backpressure starts to build up and eventually chokes the flow. A typical remedy is to pack out the part, which often produces molded-in stresses.
Hettinga reports that his process does just the opposite. Resin enters the cavity slowly and in a controlled manner from the gate to the end of fill. The pressure curve reportedly never peaks as it does in conventional high-pressure molding. The gentle flow assures that the melt front remains intact, avoiding molded-in stress. The key difference here is that packing is never used. Although injection time is somewhat longer, Hettinga claims that overall cycle time is the same or even shorter because the packing phase is eliminated.
Hettinga says his low-pressure process does not use a hotter mold to achieve better flow through the cavity. He adds that successful low-pressure molding demands lower melt temperatures--typically 10-22[degrees] F lower than in high-pressure molding.
Gating is also an important component in Hettinga's process. Large gate diameters reduce the effects of frictional heat and shear, increasing material flow into the cavity. Sequential hot-runner valve gating is often recommended for large parts. Hettinga claims that his process is somewhat unique in that the sequencing of each gate's opening and closing is controlled volumetrically by ram position rather than by time. Injection speed and pressure through each valve gate is controlled individually, reportedly providing more control over the flow of resin through the part and eliminating knit lines. Although large gate diameters and sequential gating are helpful to the process, neither is essential to it, Hettinga says.
Hettinga explains that as the resin enters the mold, it sets up an initial skin, which insulates the newer melt entering the cavity. This is one reason why the melt does not freeze off before the cavity is filled, despite slow injection. "Cooling happens much slower than people envision," he remarks, adding that he has measured the temperature of the material at the cavity wall and at its center. Hettinga adds that the minimum wall thickness appropriate for his process is 40 mils. "Thinner than that, there is no advantage over high-pressure molding, because the skin sets up so fast that there is a need for high pressure."
Hettinga claims that the lower stresses encountered during his process allow parts to be demolded at a higher temperature, and therefore earlier. In standard highpressure molding, he says, natural shrinkage is restrained by the mold walls, which induce shrinkage stresses in the part. The early demolding possible with his process reportedly allows the part to shrink unsconstrained outside the mold.
Shrinkage is also said to be kept to a minimum by relatively uniform pressure and temperature throughout the part. Pressures may vary only 20%, compared with differentials of 100% to 300% in conventional molding, says Hettinga. Because there is no packing or holding, gate pressures "are typically lower than at the end of fill in low-pressure molding." This reportedly was demonstrated in molding a 26-in.-long automotive scuff plate with a wall thickness of 1 in. According to Hettinga, the pressure rise measured at the far end of the part is proof that there is no packing back to the gate during his process (see graphs).
Hettinga claims that low-pressure molding cannot be accomplished by merely dialing down the clamp pressure in a standard injection machine. "If you just dial down the clamp without having the control, you'll have difficulty." The glue that holds the process together, he says, is its high-resolution control. Both injection velocity and pressure reportedly are programmed to fill the mold without packing. "There is a very fine line between freezing off and supplying adequate pressure. If I retract the pressure even momentarily, or choke it off with the hydraulic effect, I'm done for."
Hettinga claims that his system achieves a higher degree of control, using more hydraulic oil than conventional injection machines. Oil viscosity is kept constant with a separate pump that recirculates the oil. Injection pressure is read at the pump rather than in the mold, reportedly allowing the computer to react quicker to changing conditions, and injection speed and clamp pressure are adjusted together. A "zoom" feature on the controller reportedly provides exceptional control of the melt front's movement during injection. If a difficult area of the mold is encountered, control resolution can be broken down into 10 more steps.
Hettinga says that his process has many potential automotive interior and exterior applications, including dashboards, bumper systems, interior panels, body cladding, floor coverings, and seating. One area of interest is to embed nonwoven textiles in a molded part as reinforcement. Hettinga's softflow, low-pressure process has also been used to incorporate fabric surfacing materials and other components into molded parts, eliminating secondary operations. Automotive headliners are a potential use for this capability. However, one molder using the Hettinga process, Rick Lickwala of Blue Water Plastics, Marysville, Mich., points out that plenty of trial and error may be required before the process can be used successfully.
One company that has recently analyzed low-pressure molding is Dow Plastics, a business group of the Dow Chemical Co. in Midland, Mich. "Our initial evaluations at Hettinga's site used a new 288-sq-in. developmental plaque mold with ribbed inserts to obtain process data for comparison with conventional molding and to begin to understand how and why the process works," explains Wayne Ruhl of Dow's Injection Molding Core Technology Team. Although the mold was designed to run on a 750-ton press, it was run on a 285-ton press at Hettinga, with new cavities not previously run on a conventional machine. He points out that while Hettinga's process "cannot change physics," it is probably optimizing certain factors to fill the part successfully. Some of these factors, he believes, are generous gating, sequential gating, optimum fill time, and possibly materials with a higher melt-flow rate and lower glass-transition temperature to allow the polymer to remain molten longer. (Hettinga, for his part, disputes the potential role of lower glass-transition temperature, maintaining that the insulating effect of plastic along the side wall of the cavity keeps the melt flowing.)
Based on the preliminary results of Dow's testing, Ruhl believes the process warrants additional study to further establish its validity. For instance, because it is difficult to predict exactly how the melt behaves in the barrel, sprue, runners, and cavities, Ruhl questions the effectiveness of backpressure profiling in controlling melt viscosity throughout the shot.
Hettinga suggests that parts can be demolded earlier because of less molded-in stress. Ruhl agrees that molded-in stress may affect the point at which a part can be demolded, although his lab experience has yet to verify that it's enough to make a significant difference in the cycle time.
Ruhl also questions the assertion that conventional molding often results in a broken flow front. "It may be that with high speeds and highly filled materials, you get a broken flow front. But for the most part, the melt is unlikely to reach a velocity of turbulent flow, where the flow front would break up."
Dow tested polycarbonate and ABS on a Hettinga machine to understand the material requirements for this process. Using the test-plaque mold, the process data obtained suggests that a lower-viscosity resin is probably helpful to the process. Dow plans additional evaluations to test the same tooling on a conventional machine and to analyze the process with tooling designed especially for the Hettinga process.
Other questions center on the pressure distribution in the cavity specifically, Hettinga's claim that pressure at the end of the cavity is higher than at the gate. One possible explanation may be that some sort of compression molding or coining effect is taking place inside the mold, suggests John Bozzelli, a former Dow molding specialist who is now a senior plastics specialist at RJG Technologies, Inc., Traverse City, Mich. Bozzelli was not involved with the Dow tests described above, and prefaces his comments with the fact that he has never visited the Hettinga plant or had the opportunity to instrument a Hettinga machine to analyze the process.
Says Bozzelli, "Plastics, unless you are doing compression molding, will always have higher pressure at the gate than at the end of fill. If you don't have that, you are violating the second law of thermodynamics." Hettinga denies that his process is related to injection-compression or coining, adding that coining could not achieve the same tolerances as his process.
Bozzelli sees one possible key to Hettinga's process in good injection control. With such control, "The flow front unrolls like a rug and the new material has to come in from the center, in a fountain flow. It's insulated because it's traveling between the plastic skins." This, he says, works to Hettinga's advantage, but is also possible in conventional injection molding with good process control. Bozzelli surmises that the Hettinga process is what he calls "decoupled molding," using the term in a different sense from Hettinga. Bozzelli means that injection speed is the primary control variable, and the machine is left free to summon the necessary pressure to achieve the velocity setpoint. "Hettinga is providing just enough pressure just when he needs it--which is what I call decoupled molding. If he needs a little more pressure, it's available, just like a car engine on cruise control."
Bozzelli claims that achieving minimum pressure drop between the gate and end of fill--the big advantage claimed for low-pressure molding--can also be achieved in conventional molding with the right controls. Bozzelli claims he can program a conventional injection machine to control fill speed to achieve that end. The control injects rapidly, taking advantage of the lower viscosity of the material when it's pushed fast. Just before it hits the end of the cavity, it slows down, slowly builds up pressure, and packs. "By that, we have tremendous ability to hold the cavity pressure from the gate to the end of the fill to very minimal numbers."
Hettinga states that he disagrees strongly with Bozzelli's opinions on the Hettinga process.
ANOTHER OEM'S VIEW
Franz Strohmaier, v.p. of engineering at Engel Canada, Inc., Guelph, Ont., which also sells low-pressure injection molding systems, listed what he considers some key factors in low-pressure molding, some of which appear to support Hettinga's points. One of these is that successful low-pressure is very much a function of components in the whole system.
Another of these is gating. "Gates and the number of gates are the first elements in getting the mold pressures and required clamp forces down," he asserts. In the case of multiple gating for large parts, Engel offers a "cascade" valve-gate hot-runner control system, which uses special software to open and shut the gates sequentially, reportedly eliminating flow lines on multiple-gated parts such as bumpers.
Still another is elevated melt temperatures compared with conventional injection molding. (This is contrary to Hettinga's claim.) Another is producing homogeneous melt for easier flow which in his view is a function of good screw design. In addition, good shut-off behavior at the screw tip contributes to consistent shot size. Not least important is a good controller--i.e., one with enough resolution to utilize the capabilities of the machine.
Other options for low-pressure molding are also commercially available from numerous machine suppliers and are finding their way into automotive applications. One such process is injection-compression, during which the melt is injected into a slightly opened mold, then subsequently distributed and compressed through a compression stroke. Injection-compression has been used for years on a variety of thermoset parts and also for molding CDs--in the latter case to reduce the stresses that produce birefringence. Injection-compression can be used with virtually any injection machine that is suitably programmed.
In molding automotive wheel covers, injection-compression surpasses standard injection molding, according to comparison trials by GE Plastics, Pittsfield, Mass. Injection-compression molded wheel covers of Noryl GTX, a PPO/nylon alloy, were said to have lower residual stresses and more consistent impact strength. An expected additional benefit, according to GE would be increased paint adhesion. GE found that with engineering thermoplastics, injection-compression clamptonnage requirements are 20-50% less than for conventional injection, depending on wall thickness.
According to GE global technical leader Jack Avery, injection-compression may be best suited to stiffer materials. "Easy-flow materials may move faster than you want them to in certain areas, so you may not get a uniform and consistent filling pattern."
One version of the injection-compression process is the so-called "SP-mold" technique developed by Sumitomo Chemical Co. of Japan and licensed to JSW Plastics Machinery, Santa Fe Springs, Calif. In Japan, several car companies, including Nissan, Toyota, Mitsubishi, Mazda, and Isuzu, are using the SP injection-compression process for door-panel applications, according to Phillip Townsend Associates of Houston, which recently conducted a study on automotive interiors.
During the SP process, melt is fed into a clearance between the upper and lower halves of the mold in a vertical clamp. The mold is closed so that the resin is spread into the cavity; then the cavity is pressurized. According to Kiyoshi Sawai, business manager of plastics for Sumitomo Chemical in N.Y.C., the SP process differs from conventional injection-compression molding in a number of key ways: Sawai claims that the mold is opened to a lesser extent, keeping the melt hotter; and melt fed in through the bottom of the mold, both resulting in better flow into large parts. Also, the SP process is done on a vertical press, said to be an advantage when molding in skins such as vinyl.
Typical applications have included interior door panels and, more recently, instrument panels on the Mitsubishi Mirage. Sumitomo says its SP process is applicable to most thermoplastics including PP and ABS. For automotive applications, it has developed a special easy-flow (45 MFI) PP for its SP process.
In the U.S., Ford Motor Co. has adopted the SP process for molding interior door trim of the 1994 Thunderbird. The process also has been used to produce door panels for the Mercury Villager. These parts are suitable for highmelt-flow polyolefins. One advantage of the SP process, according to Ashir Thakor, materials specialist at Ford's Plastics and Trim Products Div. in Dearborn, Mich., is that the resin has a chance to flow a bit before the tool is fully closed. This helps when molding in vinyl or fabric coverings because it dissipates some of the heat and reduces some of the pressure, he says. Thakor adds that the SP process utilizes resin with melt index significantly higher than with conventional injection molding. Thakor says another advantage of the process is that it offers greater freedom to vary wall thickness throughout a part without introducing excessive stresses.
One injection-compression process, which is now being tested in Europe, has been developed by Krauss-Maffei. The Vertical Injection Molding (VI-M) machine is a modular system consisting of horizontal injection unit and vertical clamp above a rotary turntable. During this process, a "preform" of material is deposited into the open mold cavity. Next, the turntable is indexed into the clamp station, which compresses the material into the cavity. The turntable includes a third station in which fabric or other substrate can be placed in a frame over the cavity prior to compression.
One advantage of the process, says business manager Carey Brayer, is that it uses a precise charge of material that has consistent flow characteristics. The preform reportedly can be molded to 80% of its ultimate shape without touching the decorative material. The system uses material pressures of under 1450 psi and is suitable for easy-flow material that will be less likely to disturb a delicate substrate. Cycle times are said to be around 50 sec. The VI-M process reportedly avoids stresses encountered during one-step in-mold lamination.
Krauss-Maffei claims that its process avoids stresses caused by high temperatures and high pressures, as well as disruptive flows beneath the decor material, such as are typically encountered with single-step in-mold laminating. In Carey's opinion, these stresses result from the gates, runners, and geometry of the mold during filling in conventional injection molding.
Other variants of injection-compression have been developed by Engel as alternative approaches to its so-called Textilmelt process, which keeps cavity pressures below 2000 psi as plastic is injected over a layer of fabric or other decorative material. One version for large flat parts injects behind the fabric layer. Where multiple gates are used, Engel's "cascade" hot-runner valve-gating control is said to provide a more even distribution of the melt over the plastic. Another version is compression molding, in which a stretcher frame holding the fabric is placed over a preform of resin in the mold, then compressed.
A third approach is a new version of compression molding known as Melt Application Compression Molding. It uses a nozzle system to extrude the melt onto the lower mold half in a vertical clamp. Known as "open-face deposition," the nozzle is servo-controlled in the x-y-z directions to provide even distribution of plastic over the contours of the mold surface. After depositing is complete, the upper half of the mold comes down and compresses the material.
Another low-pressure option is gas-assist or gas-injection molding. That's because--in some versions of the process--the mold is short-shot with melt, and then high-pressure gas is introduced to pack out the part. (The mold can also be fully injected with melt, some of which is allowed to fill an overflow well after the gas is introduced.)
According to Jon Ericson, v.p. and general manager of Gain Technologies, Mt. Clemens, Mich., gas-assist can typically reduce clamp-tonnage requirements by half to two-thirds. In automotive applications, that can mean that the steel required between the cavity wall and the outside of the mold typically can be reduced from 5-6 in. to about 3 in. In addition, tools often can be simplified, eliminating cores and slides.
Cycle times are generally improved, he adds, because of shorter cooling times. Also, parts tend to exhibit less warpage, eliminating the need to use cooling fixtures. However, there are some design limitations to gas injection. The part must have thick sections or ribs to form channels cored out by the gas.
Ericson says automotive engineers have now started to specify gas assist, though not yet in a lot of parts. Besides Gain, the best-known source of gas injection technology is Cinpres Ltd. of England, with an office in Ann Arbor, Mich. Hettinga also has its own variant of gas-assist, which it calls the "Helga" system (PT, Aug '91, p. 36; Sept. '91, p. 47). Most major injection machinery suppliers have some experience with gas injection, and several offer proprietary expertise.
|Printer friendly Cite/link Email Feedback|
|Author:||de Gaspari, John|
|Date:||Sep 1, 1993|
|Previous Article:||News update: flame retardants '93.|
|Next Article:||How they make PEs.|