1 Overview of FRA/Volpe Research on Concrete Ties Sponsored by Federal Railroad Administration, Office of Research and Development U.S. Department of Transportation Research and Innovative Technology Administration John A. Volpe National Transportation Systems Center David Jeong Hailing Yu International Concrete Crosstie & Fastening System Symposium June 6-8, 2012

2 Motivation for Research Rail seat deterioration determined as probable cause of two Amtrak derailments on curved track — Home Valley, WA on April 3, 2005 — Sprague, WA on January 28, 2006 Widespread damage observed on concrete ties on Northeast Corridor and elsewhere Service life of concrete ties appears to be less than original design life (50 years) FRA has awarded several contracts via High-Speed Rail BAA to conduct research on concrete tie performance

3 Research Constituents FRA Track Systems Research Program Volpe National Transportation Systems Center Transportation Technology Center, Inc. FRA High-Speed Rail Broad Agency Announcement (BAA) Program University of Illinois – Urbana-Champaign Kansas State University Silica Fume Association ENSCO NDT Corporation Other Stakeholders Amtrak and North American Railroads Concrete Tie Manufacturers

4 FRA BAA Projects on Concrete Ties University of Illinois at Urbana-Champaign (UIUC) “Improved Concrete Crossties and Fastening Systems for US High Speed Rail and Joint Passenger/Freight Corridors” Kansas State University (KSU) “Quantifying Effect of Prestressing Steel and Concrete Variables in the Transfer Length in Pretensioned Concrete Crossties” Silica Fume Association (SFA) “Development of Optimal High Performance Concrete Mixture to Address Concrete Tie Rail Seat Deterioration” ENSCO “Concrete Tie Machine Vision Inspection” NDT Corporation “Characterizing Damaged Concrete Ties with Nondestructive Pulse Velocity Measurements” Kansas State University (KSU) “Freeze-Thaw Performance of Concrete Railroad Ties”

5 Example of Coordination with BAA Projects Untensioned and Tensioned Pullout Tests Pretensioned Concrete Prism Tests Concrete Railroad Tie Under Load Kansas State UniversityVolpe Center

6 Finite Element Modeling of Concrete Tie Concrete Tie Supported by Ballast and Subgrade Heterogeneity

7 Motivation for Analysis and Modeling Identify potential conditions for failure Provide guidance for testing Interpret test data Extrapolate test results for difficult-to-test conditions Evaluate “what-if” scenarios

8 Flexural Cracking (Center-Binding) Rail Seat Deterioration Fastener Failure Cracking Due To Excessive Tensile Force in Anchorage Zone Common Concrete Tie Failure Modes Others: Environmental degradation (freeze-thaw) Alkali-Silica Reactivity Electrical Isolation Failure

9 Photographs of Failures in Wood and Concrete Ties Plate Cutting in Wood Ties Rail Seat Deterioration in Concrete Ties

10 Examples of Rail Seat Damage Gage Side Field Side Triangular-shaped DamageAbrasion due to Water Intrusion

11 Building Block Approach TestAnalysis Correlation Full-scale Level Component Level Coupon Level GENERIC SPECIMENS NON-GENERIC SPECIMENS CLOSED-FORM ANALYSES AND COMPUTATIONAL MODELING

12 Framework for Analysis Design “A” Load Case Evaluate Compare Effectiveness Of Designs Develop Evaluation Techniques Revise Design “B”

13 Modeling and Simulation Activities Experimental and Testing Activities Confirm Evaluate Y N Develop Evaluation Techniques Development of Evaluation Techniques

14 Establishing Credibility and Confidence Verification — Credibility from understanding the mathematics — Compare computed results to known solutions Validation — Credibility from understanding the physics — Compare computed results to experimental data Uncertainty Analysis — Credibility from understanding the statistical evidence — Quantify uncertainty and variability from all sources

15 Example Applications Wood versus Concrete Ties Untensioned Pullout Tests

16 Untensioned Pullout Test Pullout direction Reinforcement Steel tube Matrix Interface Schematic of KSU TestFinite Element Model (Half-symmetry)

17 Free Body Diagram of Wire Pullout x 2R L P 2r L-x F(x)  (x) P

18 Distributions of Slip, Force and Bond xxx s(0) = s F s(L) = s L F(0) = 0 F(L) = P x 2R L P 2r Slip, s(x) Force, F(x) Bond,  (x)

19 Analysis of Pullout Test Initial Value ProblemTwo-point Boundary Value Problem Prescribed Conditions Calculated Outputs Two first-order differential equations

20 Direct and Inverse Analysis GIVEN: Bond-slip Relation P s DIRECT ANALYSISINVERSE ANALYSIS CALCULATE: Pullout Force vs. Slip Curve CALCULATE: Bond-slip Relation GIVEN: Pullout Force vs. Slip Curve  s Input Output P s  s

21 Inverse Analysis of Untensioned Pullout Tests Bond-Slip Relations Derived from Inverse Calculation Average Pullout Curves with 95% Confidence Band

22 Pullout Test Results and Direct Analysis Results

23 Model Verification and Validation Process Reality of Interest Mathematical Model Computer Model Validation Verification Quantification Implementation Modeling Simulation

24 Recent Volpe Publications H. Yu and D.Y. Jeong, “Railroad Tie Responses to Directly Applied Rail Seat Loading in Ballasted Tracks: A Computational Study,” JRC , August B. Marquis et al., “Effect of Wheel/Rail Loads on Concrete Tie Stresses and Rail Rollover,” RTDF , September H. Yu et al., “Finite Element Modeling of Prestressed Concrete Crossties with Ballast and Subgrade Support,” DETC , August 2011.