All-carbon bicycle rims cost-competitive with aluminum

Iterative design process improves performance and manufacturability of one-piece carbon rims.

In the bicycle world, carbon fiber isn't just high-tech ... it's hot.

Virtually every bicycle manufacturer has incorporated carbon fiber into a portion of a frame or introduced an all-carbon frame (e.g., see "Aluminum Frame Builder Incorporates Carbon Fiber Tubes,"HPC January 2006, p. 44). Successful frame applications have encouraged composites manufacturers such as MacLean Quality Composites (MQC, Jordan, Utah) to use carbon in a number of other bicycle components as well, including handlebars, water-bottle cages, front-wheel forks, seat posts and even wheel rims. While carbon rims have been available for almost a decade, their high cost has limited their use primarily to racing applications, where lightweighting is a critical design driver and price is rarely an issue. Under the brand name Reynolds Cycling, however, MQC has added to its complete line of carbon parts for road and mountain bikes, nine wheel models that feature one-piece carbon rims manufactured in a net-shape bladder molding process. Recently, MQC refined its process to the point that its newest wheel set is price-competitive with high-end aluminum wheels.

Why carbon?

Traditionally, rims, spokes and hubs have been made from aluminum because it is one of the lightest extrudable metals, yet durable enough for daily riding. An aluminum rim also is relatively easy to manufacture: The desired cross-section is extruded, then cut to length, rolled and joined at the ends to create the rim. However, because it is an isotropic material, aluminum has the same value – 68 GPa (10×106 psi) – for two key rim properties, tensile and compressive modulus. By contrast, a carbon composite's fiber architecture can be designed anisotropically, exhibiting much higher tensile modulus – 140 GPa to 240 GPa (20.3×106 psi to 34.8×106 psi) for standard modulus carbon/epoxy composites – but with a compressive modulus as low as 11 GPa to 15 GPa (1.6×106 psi to 2.2×106 psi). Carbon is able to achieve these values at lower rim weight.

These factors make it possible to significantly improve bike performance with carbon rims, says Reynolds Cycling's director of sales and marketing Jonathan Geran. Carbon's higher tensile modulus increases rim stiffness, which reduces power loss as the rider's energy is transmitted to the rear wheel. Greater stiffness in an aluminum wheel requires the use of more material, adding weight and producing a rougher ride. In addition, the less stiff aluminum wheel requires a greater number of spokes to maintain true wheel shape, avoiding side-to-side "wobble"as the wheel turns and "hop,"which is caused by an out-of-round rim. Both problems are addressed by adjusting spoke tension, and a greater number of spokes permits finer adjustments when truing a less stiff rim. However, the cost, again, is greater weight. The stiffer carbon rim requires fewer aluminum spokes to keep the wheel true. This reduces the rotational weight of the wheel, which means the rider expends less energy during acceleration. The cumulative difference in acceleration – and deceler-ation, for that matter – is often compared to the difference between driving a sports car and an economy car.

According to Geran, reducing rotational weight is the best way to improve riding performance because the rider can easily perceive the difference in ride. Carbon's very low compressive modulus, in the meantime, does a better job than aluminum of damping vibration on the road or rough terrain, a key factor in rider comfort. According to Geran, ride comfort is one of the most saleable characteristics in the bicycle industry today.

MQC achieves these stiffness and damping benefits in a one-piece rim, using unidirectional prepregs from Newport Adhesives and Composites (Irvine, Calif.), J.D. Lincoln Inc. (Costa Mesa, Calif.) and Hexcel (West Valley City, Utah). The prepregs, number of plies and their orientations and, therefore, the modulus for each wheel is dependent on the type of rim and its intended use. The MQC design team selected the architecture for each rim, while keeping in mind that carbon fiber acts much better in tension than compression. Thus, the layup is a mixture of plies that are angled to stiffen the wheel as needed depending on the rim type and shape, yet improve damping properties. In addition, the layup is optimized by adding reinforcement in areas where it is needed – again specific to rim type. For example, extra material is used to reinforce the areas where the spokes interface with the rim to prevent pull-through due to the highly concentrated point load. Fatigue, flex/bending and impact testing as well as field trials have been an integral to the design cycle for each rim type.

A producible part

Since the processing of composites is still a young technology compared to that of metal, it is constantly evolving, says MQC's custom products manager David Erickson. When a composite part replaces a metal part, a host of design implications must be considered. Thus, the greatest challenge MQC's design team faces is manufacturability, says operations manager Jimi Paulsen. Since the rim is a hand-layed part, repeatability was a key concern. Rick McMillan, owner of Industri-Tech LLC (Salt Lake City, Utah) and supplier to MQC of Newport material, notes that using three different prepreg systems in a single rim actually was fairly straightforward. The three prepregs differ little in fiber content and, while their different resin systems are necessary to achieve wheel design goals, they all can be processed successfully at the same temperature and cure cycle duration. Modulus can be matched, while the use of prepreg provides consistent fiber/resin ratios, which enables layup optimization.

To maintain part-to-part consistency in the bladder molding process, the prepreg is nested and cut using an automated cutting table from Gerber Technology (Tolland, Conn.), which ensures repeatable ply shape and fiber angle. The team also developed a kitting sequence that gives the proper ply to the operator when it is needed, reducing layup time and eliminating placement error.

The design team spent significant time refining the bladder molding portion of the process as well. Since the part is one piece with a complex shape, the tooling is quite complex – 11 pieces, including the bladder. Because the rims are sold "naked"(unpainted), it is imperative that the rim comes out of the mold in a saleable form. To achieve this end, the team worked with a coating company to develop a proprietary permanent coating for the tool surfaces that facilitates part extraction while improving the rim surface finish. Reynolds says it is currently the only carbon wheel manufacturer that can produce a cosmetically acceptable naked rim. The design team also developed a proprietary method for enclosing and holding the layup in the tool during the cure process. Once the mold is closed, controlled inflation pressure and rate ensures that the bladder is properly pressurized and creates a part with very few voids, explains Paulsen. This process has recently been enhanced by the integration of a new bladder type, developed with the independent assistance of both Airtech International (Huntington Beach, Calif.) and Richmond Aircraft Products (Norwalk, Calif.). Paulsen explains that the new bladder results in more uniform compression throughout the rim, which improves part repeatability.

Clincher machining

Early carbon rims were of the tubular type, so named for the less common tube-within-a-tire construction, the choice of most serious racing enthusiasts. The company was one of the first to offer the more complex clincher style in 2003. Clinchers are the much more common and less expensive technology found on consumer-level bike models, featuring a separate tire and tube. Clinchers are easier to remove and replace and, when flat, usually require replacement of only the tube. Clincher rims also are relatively easy to manufacture in extruded metal, but the clincher rim's "hook beads,"which hold the tire in place (see illustration on p. 44), are difficult to mold in a composite.

MQC addressed this challenge by using a second bladder to create the cavity where the tube and tire are secured. After demolding, the rim bead is cut in a secondary process. For initial prototypes, the beads were routed by hand, but when the wheel went into production, MQC automated the process, using a point on the rim's brake track (see illustration) to guide the router. However, this method soon proved inadequate. As the mold tools wear and approach the end of their service life, Paulsen explains, demolded rims show a very slight deviation in flatness and thickness of the brake track, which yielded no noticeable change in riding performance, but introduced problematic variation into the routing process. These variations caused the router to leave a nonuniform lip, based on the one-point guide. To solve this problem, MQC designed a three-point guidance system and a continuous feedback loop that now enables its automated CNC routing. The result is a uniform hook bead regardless of mold/brake track condition. The clincher wheel that this process produces is able to handle the tire pressure of an aluminum rim, but has all the benefits of a carbon rim. Despite the additional processing steps, these refinements have reduced processing cost sufficiently to permit Reynolds to offer its latest clincher style, available in 2007, at a cost comparable to competing all-aluminum clinchers.