An update on revenue service tests of bainitic steel rail crossing diamonds: continued maintenance reduction and predicted longer service life are the results in year two of diamond testing - Ttci R&D - Transportation Technology Center

Railway Track and Structures, Jan, 2003 by David D. Davis

In the second year of revenue-service trials, bainitic steel rail crossing diamonds continue to show promising results. Three crossing diamonds are still in service at locations maintained by Canadian National/Illinois Central and Union Pacific. Confirming last year's findings, reduced running surface deformation and maintenance have been reported. (RT&S, October 2001) The findings are based on measurements of the current test bainitic and previous crossing diamonds, as well as interviews with local maintenance personnel.

Besides the reduction in maintenance required, the bainitic rail diamonds are expected to provide a longer service life than conventional diamonds at these locations. In the first two years of service, maintenance has been reduced by more than 50 percent. Based on current wear rates, the expected wear lives of the diamonds will be two to three times the service lives of the previous pearlitic rail diamonds. Railroads are currently spending more than $250 million annually on frog replacements and repair. Thus, the potential savings from improved frog materials are significant.

Operating environment

The test sites (Figure 1) all carry a significant amount of heavy axle load traffic, which is defined as cars with loaded gross rail weights of 263,000 pounds or more. Each site is located on level open ground. Gilman and Zeigler, Ill., have good drainage conditions, but the Hoxie, Ark., site has road crossings nearby that limit the effectiveness of track drainage. Table 1 describes the test sites, which are located on mainlines in the Midwest.

All three diamonds tested have high crossing angles and were selected for their severe service environments. Crossings with angles above 60 degrees have frogs that cannot support the wheel across the flangeway gaps. These crossings generate some of the highest vertical loads on the railroad. Dynamic loads that are three-to-four times static wheel load are typically measured for conventional high-angle frogs at the Transportation Technology Center's Facility for Accelerated Service Testing, Pueblo, Colo. (1)

Performance results

Figure 2 shows each of the test site's average running surface height loss versus tonnage. The measurement -- indications of combined metal flow and wear -- was made one inch from the flangeway of each common corner. The height loss rates are declining with tonnage as the running surfaces work-harden. For comparison, the measured height loss values from the two previous Gilman diamonds are shown. These were conventional head-hardened pearlitic thick-web rails.

A corresponding increase in surface hardness has been measured with running surface height loss. Figure 3 shows average running surface hardness with tonnage for the test sites. The significant increase in average hardness at Gilman during the last measurement corresponds with an increase in surface cracks and spalls.

In addition to the comparison of conventional rail and bainitic rail, additional test variables have included:

* Running rail section: Standard AREMA (RE) vs thick-web (TW)

* Traffic distribution: Main vs. crossing line

* Traffic rate: 40 to 110 mgt/yr

It is assumed that the thick-web rail section is more durable for crossing applications as it strengthens the head-web connection in the rail. Man crossing-diamond rail failures result from head-web area defects. Thick-web rail also requires thinner filler bars. These bars connect the frog rails across flangeways and provide bending strength. To date, there has been no significant difference in deformation between RE and TW sections. One TW section running rail was replaced due to a bolthole crack. No RE section running rails have been replaced to date.

The effect of traffic distribution appears to be small over the three test sites. Running surface height loss rates are similar for Gilman (three percent crossing traffic), Zeigler (about 33 percent crossing traffic) and Hoxie (about 45 percent crossing traffic). However, the height loss rates for all three diamonds are higher than the rates seen for two bainitic diamonds tested at FAST. The FAST tests are conducted under higher wheel loads and similar speeds. The FAST diamonds had no crossing traffic.

Previous analysis has shown that a small amount of cross traffic contributes to a large increase in wear rate for the common corners (i.e., the one corner of each frog that sees traffic from both lines). (2) Further analysis with the latest data also shows that the common corners have a significantly higher height loss rate than the other running surfaces. Figure 4 shows these rates for each crossing.

The effect of traffic rate has not been significant in this test. Running-surface height-loss rates (per mgt) at all three sites are similar. Load application rate can affect height loss rates if there is insufficient track time allowed for inspection and timely maintenance.

Maintenance records

The bainitic rail crossing diamonds have required significantly less initial maintenance than previous three-rail and AMS casting diamonds at these locations. Having less running surface deformation appears to benefit the rest of the diamond with lower dynamic loads, less surfacing and fewer loose fasteners. This is not to say that the diamonds are not deteriorating under traffic. Figure 5 shows a deformed common corner on one of the test diamonds. Hollow-worn wheels are deforming the running rails and easer rails.