Europe's Infusion Pioneer Simplifies Process With Bottom Up Approach
Conyplex BV, an old-line Dutch yacht builder located in historic Medemblik, The Netherlands was the first and remains one of the few European boatbuilders to use vacuum infusion molding to produce large composite hulls and deck structures. Despite high concern for the environment, both in its governmental activities and among its populace — not to mention approaching regulation within the European Union — Europe's composite boat manufacturers historically have viewed closed molding as too risky for something as large as a 5O-ft hull. In 1994, however, after 30 years of hand layup and sprayup, Conyplex partnered with The University of Delft Center of Lightweight Structures (TNO-TUD) to research the viability of resin transfer and resin infusion methods for production of large marine structures. In 1997, the Conyplex production team successfully tested the process on a full hull, for a yacht in the Contest 55CS class. Conyplex produced its last layup/sprayup hull 24 months ago, converting its entire operation to a vacuum infusion method so simple, it required just two days to retrain employees in its use. Today, the boatbuilder routinely infuses hulls, stringers, bulkheads and decks for yachts up to 60 ft in length, using the same process and wonders why the rest of the European marine community has been so slow to follow.
FROM SUPPLIER TO BOATBUILDER
Situated 60 km/37 miles northwest of Amsterdam, the modern Conyplex production complex and full-service boat yard is a few minutes by water from IJsselmeer, Holland's famed inland sea. Owned and operated by the Conyn family since its inception, Conyplex entered the marine market as a supplier of fine woods and marine plywood. The firm first became involved with sailboat production in the late 1950s, supplying wooden parts for a single-masted 20-footer in the Flying Dutchman class. As boatbuilders adopted fiberglass construction, Conyplex produced fiberglass molds in addition to wooden parts for other boatbuilders, but didn't begin building its own boats until the early '60s. The first was the Contest 25, a 25-ft sloop, of which the company eventually produced 400. Forty years and 5,000 Contest yachts later, Conyplex bucks the boatbuilding trend toward mass production, launching 35 to 40 new semi-custom composite yachts in the 40 to 60 ft range each year, with a work force averaging 135.
OPENING UP TO CLOSED MOLDING
When Conyplex and TUD-TNO first looked into closed molding, in 1994, the company's composites department manager Han van Aalst recalls early resistance to the idea among his seasoned sprayup crew: "They thought it would be very complicated." Indeed, as research commenced, it looked very much like their worries would be substantiated. Resin transfer molding (RTM) required construction of a second, rigid mold for each hull, to contain the forces necessary to inject resin under great pressure. That would be a tough sell to senior management, if adapting the new molding process would require significant expense in terms of retooling and equipment or seriously disrupt scheduled production during a technology changeover. Vacuum infusion was attractive, but the long, wide (16.4m/55 ft by 4.5m/14.8 ft) and very high (2.4m/7.9 ft) Contest hulls presented a formidable challenge to infusion in one shot. "When the University saw the big hull, they were a little bit worried about it," quips van Aalst.
Van Aalst, however, was concerned about process complexity. Proponents of resin infusion overcome size challenges by placing dozens of resin inlets in the vacuum bag, fed by resin lines connected to multiple sources of resin strategically placed around the hull. (For instance, see "Inside Manufacturing," CT November/December 2001, p. 33). Prepping for infusion was labor intensive and required technical expertise to determine the best placement for resin inlets and air outlets, to prevent pockets of resin starvation and trapped air. "It was very complicated," says van Aalst.
In hindsight, van Aalst views the move to vacuum infusion as a huge boost to company productivity and product quality. But at the time, Conyplex was not unhappy with the results of its spray-up process. With six sprayup systems in operation, the primary motivation was to do something about the boatyard's impact on the environment and improve working conditions. "We had a lot of solvent emission," van Aalst recalls. "And our worker's exposure was too high." Faced with high turnover and the resulting costs of training new hires, neither Aalst nor the few old-timers among his crew relished the idea of a new system that would require additional expense and expensive training. That would make the cure as painful as the disease. So in all his dealings with TUD-TNO theoreticians, van Aalst preached process simplicity. "That's what I wanted for my people," he says. "And we worked with them to find a way."
With funding contributed by The Netherlands Ministry of Economical Affairs, Conyplex and TUD-TNO enlisted the help of software developer Polyworx (Ammerstol, The Netherlands). The company created digital simulations of hull infusion, using RTM-worx software, a direct descendant of a program (then known as π7) developed at CLC TUD-TNO by Arjen Koorevaar, now Polyworx owner and director, while he was still involved in research at the school. The software was used to predict resin flow in a computer model of a 16.8m/55-ft hull. Flow-front simulations demonstrated that a potential drawback to infusion was the tendency for the flow to be disrupted both by changes in laminate thickness and/or a change in material permeability. Disruptions of the flow front tended to trap air in the laminate. Over the next year-and-a-half, they discovered that in a hull where the laminate and core thickness did not vary, a very uniform flow front could be simulated and, more importantly, easily controlled using the simple resin delivery system described below.
When it came time to infuse the first complete hull (16.8m/55 ft), actual tests had been conducted only on flat sample sandwich panels and small hull sections. There was concern about whether or not the resin would fully impregnate the unseen laminate between the core and gel coat. So Conyplex molded its first hull without gel coat, allowing visual inspection of the hull exterior through the clear resin after demolding. "We were curious to see if the resin was running underneath the balsa core, and we found that it did very well," van Aalst explains, but adds that, for peace of mind, the next nine 55-ft hulls were done in similar fashion, and afterward, were finished with epoxy filler and sprayed with AWL-Grip Erethane paint. Today, molds are gel coated in conventional fashion and the molding process proceeds as follows.
STEP BY SIMPLE STEP
Conyplex builds its own hull molds, using tooling gel coat and polyester tooling resin supplied by Reichhold (Research Triangle Park, N.C., U.S.A.). Molds are prepped for release, as they have been for 40 years, with Carnuba wax, and gel coating is done after regular work hours (to spare the rest of the crew). A two-man team can sprayup an entire mold surface to minimum 0.8 mm/0.031-inch thickness (0.6 mm/0.022 inch, when cured) in three-quarters of an hour. They use Reichhold's NORPOL XTREM gel coat, selected for what van Aalst notes is its high resistance to osmosis and superior color retention in sunlight.
For aesthetic and practical reasons, a single layer of 450 g/m2 chopped strand mat, supplied by Cam Elyaf Sanayii AS (Kocaeli, Turkey), is layed up by hand and wetout with Reichhold isophthalic resin formulated with a special catalyst, which promotes full cure of the initial reinforcement layer in 24 hours. This step provides protection against print-through and acts as an osmosis barrier. In addition, van Aalst notes, "we can walk on the gel for the rest of the process."
On day three, additional dry reinforcements are layed up, beginning with SAERTEX Wagener GmbH & Co. KG (Saerbeck, Germany) triaxial non-crimp stitched fabric, placed with the 0°/90° (longitudinal) layer down first, and the +45° and -45° fibers on the inside (overlapped about 10 cm/4 inches.) "The layup is the same throughout the hull, except for the keel area, which gets two additional layers," says van Aalst.
Conyplex specifies 25.4 mm/1-inch-thick Contourkore from Baltek Corp. (Northvale, N.J., U.S.A.) throughout the hull structure. The end-grain core is delivered in 1m by 0.6m (39.4 inch by 23.6 inch) sheets adhered to open mesh backing on one side. The balsa is cut through to the backing to form approximately 38 mm/1.5-inch squares. Placed mesh-side up, the core conforms to hull contour and does double duty as a resin flow medium, because the slight gaps between the balsa blocks that result from the negative curvature create an intricate resin flow network on the gel-coat side. The core is over layed with another layer of the triaxial fabric.
Next, the resin delivery system is laid out over a layer of peel ply, which enables post-cure removal. Drawn from a single container placed at the stern, the resin passes through an umbilical that is attached to a main resin line that runs lengthwise along the keel, feeding a number of what Conyplex calls "runners," which run at right angles up the sides of the hull to the deck flange. Runners are long steel coils coated to prevent damage from the resin. When vacuum is applied, the coils prevent the channels from collapsing and enable the resin to flow out between the windings into the adjacent reinforcements.
The entire layup is vacuum bagged with polyethylene film (which the Hollanders call "foil") and vacuum lines are attached along the top outside edge of the hull and harnessed to a single vacuum pump. A vacuum is pulled to 100 millibars/m2 (8,000kg/m2 or 11.4 psi). When vacuum pressure stabilizes (indicating no leaks in the bag), Reichhold's low-styrene DCPD (dicyclopentadiene)-modified laminating resin is released and begins to fill the hull from the bottom up.
Infusion takes about 1.5 hours and the infused laminate is hard to the touch in three hours. That leaves an hour-and-a-half window in which the two-man crew can inspect the laminate for dry spots. If one remains, the bag is pierced with a syringe-like fitting on the end of a vacuum hose. "With this device, we can suck the air out in the dry spot," says van Aalst. "It's also very simple." Once the area is thoroughly impregnated with resin, the syringe is removed and the hole is taped over. Vacuum is maintained as a precaution for a full day afterward, and the laminate achieves 75 percent cure within that time.
THE SIMPLE PLAN EXPANDED
While initial efforts focused on the hull, Conyplex and TUD-TNO determined through additional flow-front simulations that the company's system of stringers, bulkheads and its complex, one-piece decks could be infused by the same simple process. In each case a central resin channel runs the length of the structure (flowing from stern to bow) with runners extending out at right angles to the main line.
After 72 hours, technicians rough sand areas where stringers, the mast foundation and two stiffeners that run lengthwise below the water line are fashioned from PVC foam, overlaid with SAERTEX 600g/m2 stitched woven roving and biaxial (±45°) fabric and infused in one piece. Afterward, three bulkheads are made with 450g/m2 chopped strand mat and woven glass roving, also from SAERTEX, and are infused outside the hull, then hand laminated into position inside the hull.
From gel coat to first infusion, hull prep takes two men 15 working days. The stiffeners take an additional workweek, with a third week required for bulkheads. "So in four weeks, two men can make a finished hull," van Aalst maintains. In the next twelve days the same two-man crew infuses the deck in similar fashion.
According to van Aalst, the infusion method saves a total of 160 man-hours, compared to the previous sprayup process. Further, laminate thickness is an average 50 percent thinner, with a fiber-to-resin ratio of 60:40 (previously 40:60), resulting in a savings of 250 kg/551 lb of resin on the hull alone (the average hull consumes about 900 kg/1,984 lb of resin). "Our hulls are much lighter, so there's better handling when sailing, and with less draft they're faster, too," says van Aalst.
Finally, the process not only is simple, but easily adapted for significantly larger structures, as well. "In the future, if we make bigger boats, our filling time of three hours will remain the same," he maintains. In lengthier hulls, only an additional filling line or two might be required to accommodate the additional length. "A boat of 70 or even 90 ft is no problem for us," he says. "In fact, we make more profit on the bigger boats, so the firm is very happy with us."
SAILING TOWARD NEW HORIZONS
In the wake of the successful migration to infusion, Conyplex now actively seeks innovation. "We're open to new ways," says van Aalst. At press time he was testing Lantor BV's (Veenedaal, The Netherlands) new Soric nonwoven surfacing veil, designed to curtail the risk of print-through, and expects to incorporate the veil into future hull layups within three months. Conyplex also has contemplated replacing the current DCPD-modified resin with a vinyl ester. But the big item on his wish list is the next quantum leap to an even lighter, stronger hull that requires less processing time. Given the proliferation of carbon fiber composites he's seen at recent trade shows, in automotive and other previously glass-dominated composites applications, he considers it at least possible that within five years, Contests could be made from carbon. "It costs more," he says. "But if it's what the customer wants, why not?"
Van Aalst laments that fellow Europeans have been so slow to warm up to vacuum infusion. "Conyplex is one of, maybe, three or four, but no one is infusing on this scale." Yet, he's has seen encouraging signs that infusion will soon make a positive impact on his industry, and says that Conyplex is doing everything it can to encourage its use, by inviting boatbuilders to witness the process first hand. "All the companies from Europe are coming out here to see how we work," he says. "We've had more than 200 people here in the last five years. They all say it's not possible." Van Aalst doesn't blame them. He's seen it done everyday for five years and still can say, "It's amazing. It's a miracle every time I look at it."
Durable custom elastomeric bags cut consumables cost and boost rate production.
A double vacuum-bag system and tight process control enable repeatable fiber volumes of 60 to 70 percent and improves consistency of infused laminates.
There are numerous methods for fabricating composite components. Selection of a method for a particular part, therefore, will depend on the materials, the part design and end-use or application. Here's a guide to selection.