This type of crossing surface is made of separate concrete panels, usually about six to seven inches thick and eight to 10 feet long, reinforced with an angle iron frame around the perimeter. The panels sometimes are laid on top of the ties, which are also made of concrete. The panels fill the space created by the tracks, up to the level of the road pavement.

Misalignment with the mating surfaces allows the heavy concrete panels to rock and rub against the ties as traffic passes.

Observing that cracking was a widespread problem among railroad crossings, Performance Polymers Inc., a Cambridge, Ontario, rubber design and manufacturing company, wanted to come up with a better solution.

Cracking Up

Heavy trucks, traveling at high speeds, are a major cause of damage to the panels that make up concrete railroad crossings. According to Frank Williams, manager of product development at PPI, "Transport trucks hit these crossings at 50 to 60 miles per hour."

The resulting impact damage is related to gaps and irregularities in the fit of the panels to the ties. "If you sat three ties together and put a straightedge to them, you could end up with an eighth-of-an-inch gap," Williams said.

Although concrete ties are manufactured to tight tolerances where the rail sits, requirements are less critical outside the rail seating area. The concrete panels, too, may have slight irregularities in their surfaces, causing them to sit unevenly on the ties. A rigid panel suspended across four ties may touch only the surfaces of two, leaving the panel suspended in between.

"If you get loaded transports going across, it's going to deflect the concrete," Williams said. "That leads to cracking."

The existing method of reducing the damage has been to place a solid rubber sheet, about 0.125 inch thick, between the panels and the ties. After evaluating the problem, PPI set out to develop a better way of cushioning the concrete panels.

The project had two basic design goals, Williams said. One was to try stabilizing the crossing by aligning the panels and distributing the load over all the ties. The other was to reduce the chance of breakage by reducing the impact transmitted through the panels to the ties.

A rendering of a truck crossing a railway that is cushioned by the original ribbed pad designed by Performance Polymers Inc.

PPI came up with an initial part concept that was based on classic rubber design guidelines. The approach, worked out by Domi Bruyn, PPI's vice president of engineering, replaced the solid rubber sheet with a design whose underside had bands of rubber, interspersed with open spaces to cushion the impact. Using traditional calculations and a spreadsheet pointed the company in the right direction, but PPI needed to expedite the development of a fairly advanced prototype to try out in the field.

PPI approached Windsor Industrial Development Laboratory Inc., a design analysis and testing company in Windsor, Ontario, to refine the part. As a starting point, the lab, using a sketch supplied by PPI, analyzed the dampening effects of the ribbed pad and those of a flat sheet of rubber. Over the next two months, using finite element analysis, the lab evaluated a number of different shapes and analyzed what was happening to the part under load.

The first step in developing the part was to characterize potential materials for making the pad. Windsor Lab tested three kinds of rubber: EPDM (ethylene propylene dimonomer), SBR (styrene butadiene rubber), and natural rubber, in hardness ranging from 50 to 75 Shore A. Both SBR, an early synthetic rubber, and natural rubber are used extensively in dampening applications.

As a hyperelastic material, which deforms significantly but regains its shape, rubber requires a battery of tests to identify its properties, according to Ben Chouchaoui, president of Windsor Lab. Rubber samples can undergo four modes of deformation--uniaxial, equibiaxial, planar, and volumetric--under tensile or compressive conditions. The materials also had to be studied under dynamic loads and for dampening characteristics.

EPDM rubber exhibits high energy absorption and has been proven in applications such as marine fenders, Williams said. EPDM was also cost-effective, and PPI had long experience in using the material. These considerations and the test results led the company to choose EPDM rubber, with a hardness of 70 Shore A, for the pad.

The next step was to build CAD renderings and FEA models of the pad. The approach allowed the pads to be optimized on the computer through virtual testing, according to Chouchaoui. Rubber requires FEA software capable of nonlinear analysis. Windsor Lab used an FEA package named MARC, supplied by Marc Analysis Research Corp. of Palo Alto, Calif.

Although solid rubber will flex, it is nearly an incompressible material when contained. This is one reason why a flat rubber pad had limited cushioning effects. The purpose of the band design on the underside of the pad was to give the rubber room to change shape under load.

Trying Different Ideas

Using finite element analysis, Windsor Lab tested different ideas to satisfy design goals worked out with PPI. The pad should fill the gaps between the tie and panel. It should provide uniform support and withstand heavy trucks (16,000 pounds per axle) crossing the rails at speeds from 20 to 80 mph. Finally, the pad should have a controlled vertical deformation, so automobile passengers do not feel an "ejection" from their seats when passing over the rails.

To distribute the load more evenly, PPI decided to replace some of the solid rubber bands in PPI's original design with hollow cells that would collapse under the weight of the panels. "The idea was to distribute the initial load of the panel itself more evenly and stabilize it a little more," Williams explained. The cells, which are deeper than the solid bands of the original pad, also help to distribute the load over all the ties, rather than at a few contact points.

Windsor Lab analyzed four cell shapes: circular, square, triangular, and a combination of square and triangular. Deformation patterns of the triangular cell were most effective in absorbing the mismatch between the ties, said Chouchaoui. Each pad was designed with three hollow cells, one at the center and one at each side.

The solid rubber bands remain in the final design. Bands are about 0.125 inch deep and one inch wide, and take the main loads of heavy transport trucks, said Williams. In addition, the band design provides a more controlled deflection of two to three times that of solid rubber. In fact, later tests conducted by Windsor Lab showed that solid rubber pads actually cause the concrete panel to bounce slightly after a vehicle passes. Although the deflection in the new design is very slight--only about 0.03 inch--it's enough to provide significantly improved dampening, he said.

Together, Windsor Lab and PPI were able to arrive at the right combination of material and shape. The material absorbs and dissipates the dynamic loads, while the combination of hollow cells and solid bands provides proper deflections and cushioning.

PPI's initial pad lay on top of each tie to support the concrete panel, and wrapped around the chamfer of the tie to stay in place without adhesives.

Computer simulations showed that the cross-section of the pad widened under load and that the edges lifted up. Windsor Lab proposed a modification to the wraparound design. The space left between the chamfer and the panel would be filled with a segment of triangular section. The section also was narrowed so that under load it could deflect to the width of the tie.

Two months of finite element analysis were followed by parts testing at Windsor Lab. The virtual testing enabled the company to produce a prototype in a matter of months, rather than years, noted Williams.

The pad is currently undergoing field testing at a California site, and the company is arranging tests at another site in the Pacific Northwest.

home | features | news update | marketplace | departments | about ME | back issues | ASME | site search

© 2000 by The American Society of Mechanical Engineers