Interface - The Journal of Wheel/Rail Interaction

FRICTION MODIFICATION

Top of Rail Friction Modification in Tough Terrain (continued)

Due to bi-directional traffic requirements, the 4.5-degree test curve was elevated to 3.5 inches, resulting in a balance speed of 33 mph. Analysis of actual train speeds from the L/V measurement site revealed that 100% of the eastbound coal loads were operating below balance speed. This meant that 95% of the trains were effectively encountering curves with more than 2.5 inches of excess superelevation. Scatter plots of lateral forces versus speed indicated that the lateral forces increased substantially at very low speeds (8 - 10 mph), resulting in an apparent increase in the frequency of large (over 10 kips) lateral forces. This is consistent with the general understanding that operating well below balance speed can aggravate truck warp and generate very high forces.

Figure 11 shows the fraction of lateral forces exceeding 10 kips for loaded coal trains in speed ranges of 7-10, 10-13, 13-16 and 16-19 mph for baseline and TOR-treated conditions. As shown, the application of TOR friction modifier does reduce lateral forces, but a clear trend toward higher forces at lower speeds is evident (high rail force data revealed the same trends). While the trend in lateral forces shown in Figure 11 appears to be quite evident, the distribution of speeds for loaded coal traffic is quite narrow. Therefore, the analysis was extended to include a wider range of traffic (from under balance speed to over balance speed).

Intermodal/autorack traffic operating at speeds between 12 and 40-plus mph was identified as the most promising category for supplemental analysis. After verifying a reasonable distribution of vertical loads over the entire speed range (to avoid skewing the analysis), and noting that normalization of forces (by total axle load) was required to account for varying vertical loads in these categories, 90th percentile normalized forces were plotted against excess superelevation (see Figure 12).

As Figure 12 shows, minimum values of low and high rail normalized forces occur with 0-1 inches of cant deficiency (i.e. target running speeds on most North American freight railways). As excess superelevation is increased (i.e. speed decreased), forces begin to climb. At a value of 2 - 3 inches of excess superelevation (the range at which loaded coal traffic was operating in the test area), the 90th percentile forces increased by as much as 66% versus balanced conditions.

Future Work
The results of this test program demonstrate that the use of KELTRACK® Trackside Freight wayside TOR friction control can effectively reduce lateral forces in heavy grade areas, with the application rate and/or spacing adjusted to compensate for tread braking (on descending grades) and locomotive sanding (on ascending grades).

The effects of train speed and superelevation on lateral force levels (specifically the impacts of under balance speed conditions) emphasize that wheel/rail interaction is a system issue. Therefore, a systems approach must be considered to maximize the benefits of any single technology or procedure designed to provide comprehensive solutions.

Norfolk Southern’s ongoing evaluation of TOR friction control will further quantify the ability of this technology to reduce rail wear and gauge widening in curves. TOR friction modifier application equipment will continue to be improved. Equipment spacing and application rate models will be refined as wayside TOR friction control is expanded into new areas.

Kevin Conn is Research Engineer, Norfolk Southern Corp. Kevin Oldknow is Vice President of Applications and Operations, Kelsan Technologies Corp.

References
(1) Conn, K., “Wayside Top-of-Rail Friction Control Performance” AREMA Annual Conference, September 2005.
(2) Eadie, D.T., Oldknow, K.D., Maglalang, L., Makowsky, T., Reiff, R., Sroba, P. and Powell, W., “Implementation of Wayside Top of Rail Friction Control on North American Heavy Haul Freight Railways,” Proceedings of the World Congress on Railway Research, June 2006.

Acknowledgements
The authors would like to express their thanks to the following people, whose assistance and support was critical to the successful completion of the test program described in this article: Bob Blank, Craig Webb and Dwayne Gibson (Norfolk Southern); Don Eadie (Kelsan Technologies); Ward Powell, Kevin Adkins, Steve Singleton and Mike Jirka (Portec Rail); and Rich Reiff (TTCI).

< PREVIOUS PAGE | PAGE 4 OF 4