© 
 
 U S Department 
 of Transportation 
 
 Federal Railroad 
 Administration 
 
 Investigation of Rail 
 Fastener Performance 
 Requirements 
 
 Office of Research and 
 Development 
 Washington, DC 20590 
 
 DOT/FRA/ORD-82/10 
 
 March 1982 
 Interim Report 
 
 This document is available to the 
 U.S. public through the National 
 Technical Information Service, 
 Springfield, Virginia 22161. 
 
NOTICE 
 
 The United States Government does not endorse 
 products or manufacturers. Trade or manufacturers' 
 names appear herein solely because they are con- 
 sidered essential to the object of this report. 
 
 NOTICE 
 
 This document is disseminated under the sponsorship 
 of the Department of Transportation in the interest 
 of information exchange. The United States Govern- 
 ment assumes no liability for its contents or use 
 thereof. 
 
Technical Report Documentation Pc- 
 
 1. Report No. 
 
 FRA/0R&D-82/10 
 
 2. Government Accession No. 
 
 3. Recipient's Cotolog No. 
 
 4. Title and Subtitle 
 
 Investigation of Rail Fastener Performance 
 Requirements 
 
 5. Report Date 
 
 March 1982 
 
 6. Performing Organization Code 
 
 Battel! e 
 
 7. Author's) 
 
 8. Performing Organization Report No. 
 
 Francis E. Dean 
 
 9. Performing Organization Name and Address 
 
 Battelle Columbus Laboratories 
 505 King Avenue 
 Columbus, Ohio 43201 
 
 10. Work Unit No. (TRAIS) 
 
 11. Controct or Gront No. 
 
 DOT-FR-9162 
 
 12. Sponsoring Agency Nome and Address 
 
 Federal Railroad Administration 
 Office of Research and Development 
 400 Seventh Street SW 
 Washington, DC 20590 
 
 13. Type of Report ond Period Covered 
 
 Interim Report 
 
 October 1980 - June 1981 
 
 14. Sponsoring Agency Code 
 
 FRA 
 
 15. Supplementary Notes 
 
 16. Abstroct 
 
 An investigation was conducted to develop qualification requirements 
 which rel iably duplicate the service performance of rail fasteners. 
 
 The study included a review of available data from qualification tests, 
 measurements of rail/tie deflections and fastener clip strains at the 
 Facility for Accelerated Service Testing (FAST), and laboratory tests 
 at Battelle which simulated the FAST environment. Several aspects of 
 service performance at FAST were successfully duplicated in the 
 laboratory tests. Recommendations are made for the design of improved 
 fastener qualification tests. 
 
 17. Key Words 
 
 Rail Fasteners, Fastener Performance 
 
 Railroad Track 
 
 18. Distribution Statement 
 
 Document is available to the U.S. 
 public through the National Technical 
 Information Service, Springfield, VA 
 22161. 
 
 19. Security Clossif. (of this report) 
 
 Unclassified 
 
 20. Security Classif. (of this page) 
 
 Unclassified 
 
 21. No. of Pages 
 
 70 
 
 22. Price 
 
 Form DOT F 1700.7 (8-72) 
 
 Reproduction of completed page authorized 
 
 i 
 
DC 
 O 
 Q- 
 UJ 
 CC 
 
 CO 
 
 U 
 •r— 
 
 U 
 +-> 
 
 cu 
 
 CD 
 -t-> 
 
 fO 
 
 .a 
 o 
 
 E 
 E 
 
 E 
 
 3. 
 
 
 S- 
 
 
 (L) 
 
 
 cu 
 
 
 C_) 
 
 
 -t-j 
 
 
 
 
 01 
 
 
 00 
 
 
 E 
 
 c 
 
 CD 
 
 S- 
 
 •i — 
 
 o 
 
 <L> 
 
 <u 
 
 r— 
 
 +J 
 
 S- 
 
 4-> 
 
 I— 
 
 3 
 
 en 
 
 O) 
 
 :E 
 
 cu 
 
 2: 
 
 CU 
 
 o 
 
 4-> 
 
 <D 
 
 C 
 
 o 
 
 a 
 a» 
 co 
 
 i 
 +j 
 
 4-> 
 
 O 
 
 s- 
 
 CU 
 
 +-> 
 
 CU 
 
 c c 
 
 O •!- 
 
 5 s- 
 
 cu +-> 
 
 2: co 
 
 Q 
 UJ 
 CO 
 
 CO 
 
 h- 
 1 — 1 
 
 co 
 
 (— < 
 
 _I 
 z: 
 
 UJ 
 
 u. 
 o 
 
 co 
 
 .a 
 
 
 
 
 CM 
 
 
 
 
 CO 
 
 00 
 
 
 CM 
 
 
 ■^ 
 
 
 00 
 
 1 
 
 
 
 «^- 
 
 *3- 
 
 
 CO 
 
 • 
 
 •s* 
 
 Ll_ 
 
 • 
 
 ID 
 
 • 
 
 
 
 
 CM 
 
 «3" 
 
 
 00 
 
 00 
 
 
 ^1- 
 
 in 
 
 r— 
 
 
 
 CO 
 
 lo 
 
 CO 
 
 CO 
 
 r-«. 
 
 en 
 
 LO 
 
 O r— 
 
 UJ 
 
 o 
 
 00 
 •1 — 
 
 CXI 
 
 4- 
 
 Q. I 
 
 c 
 3. 
 
 cu 
 
 -M 
 
 CL 
 
 
 
 
 
 4-> 
 
 
 
 
 
 •^ 
 
 
 
 
 
 cu 
 
 
 
 
 
 .c 
 
 
 
 
 
 c 
 
 
 
 
 * — ■* 
 
 cu 
 
 
 
 
 cu 
 
 S- 
 
 
 
 
 u 
 
 -c 
 
 
 
 
 j- 
 
 fO 
 
 
 
 
 
 
 u_ 
 
 
 
 
 4- 
 
 
 
 
 
 
 00 
 
 cu 
 
 • . 
 
 
 
 -a 
 
 cu 
 
 CO 
 
 -t-> 
 
 .c 
 
 c: 
 
 J- 
 
 +J 
 
 O 
 
 
 
 3 
 
 cx> 
 
 r— 
 
 O 
 
 sz 
 
 
 
 cu 
 
 C 
 
 u. 
 
 >— 1 
 
 a. 
 
 
 
 o 
 00 
 
 CO 
 
 
 O 
 
 -a 
 
 JZ 
 
 
 
 
 
 c 
 
 
 
 
 
 O 
 
 3 
 
 C 
 
 
 00 
 
 1 — 
 
 
 
 1 — 1 
 
 
 4-> 
 
 
 Q_ 
 
 "^ 
 
 c 
 
 •r— 
 
 11 
 
 1 
 
 ■0 
 
 •r— 
 
 C 
 
 
 -*-> 
 
 c 
 
 to 
 
 ID 
 
 a. 
 
 
 
 Z3 
 
 i- 
 
 
 •r- 
 
 
 
 O 
 
 +J 
 
 "O 
 
 ^ 
 
 u. 
 
 Q_ 
 
 CO 
 
 C1J 
 
 
 
 
 
 c 
 
 
 
 
 
 •r- 
 
 
 
 
 
 -Q 
 
 
 
 
 
 E 
 
 
 
 
 
 O 
 
 
 
 
 
 c_> 
 
 
 
 
 
 n 
 
PREFACE 
 
 This report completes an investigation to identify improvements in 
 performance requirements for rail fasteners used in U.S. mainline service. 
 Battel le Columbus Laboratories conducted the study under Contract DOT-FR- 
 9162 entitled "Tie and Fastener Verification Studies." The overall program 
 was sponsored by the Improved Track Structures Research Division of the 
 Federal Railroad Administration (FRA). In a related phase of the same 
 program, Battel le investigated the effects of tie pad stiffness on the 
 attenuation of dynamic loads in concrete ties installed on the Northeast 
 Corridor track. 
 
 Mr. Howard G. Moody of the FRA served as the Contracting Officer's 
 Technical Representative and contributed significantly to this effort by 
 obtaining timely delivery of test materials and by editing the suggestions 
 of the reviewers in preparation of the final draft. Several representatives 
 of railroads and suppliers provided meticulous reviews of the draft report. 
 The efforts of all of these people are greatly appreciated. 
 
 m 
 
Digitized by the Internet Archive 
 
 in 2012 with funding from 
 
 University of Illinois Urbana-Champaign 
 
 http://archive.org/details/investigationofr8210dean 
 
TABLE OF CONTENTS 
 
 Page 
 
 PREFACE iii 
 
 INTRODUCTION 1 
 
 Review of Fastener Performance Problems 2 
 
 Wood Tie Fasteners 2 
 
 Concrete Tie Fasteners 5 
 
 FASTENER PERFORMANCE SPECIFICATIONS 7 
 
 REVIEW OF EXISTING FASTENER TEST DATA 14 
 
 Rail Clip Force-Deflection Properties 14 
 
 Tie Pad Stiffness 14 
 
 Fastener Longitudinal Restraint 16 
 
 Fastener Lateral/Rollover Restraint 17 
 
 Lateral and Longitudinal Restraint of Wood 
 
 Tie Fasteners 18 
 
 FASTENER PERFORMANCE EXPERIMENTS 21 
 
 Track Measurements at FAST 21 
 
 Clip Force-Deflection-Strain Characteristics 23 
 
 Comparison With Track Measurements 23 
 
 Method of Determining Clip Yield Load 25 
 
 Tie Pad Compression Tests 26 
 
 Longitudinal Restraint Tests 33 
 
 Lateral/Rollover Restraint Tests 36 
 
 Fastener Fatigue Tests 41 
 
 Concrete Tie Fastener Fatigue Test 42 
 
 Wood Tie Fastener Fatigue Test 47 
 
 DISCUSSION AND RECOMMENDATIONS 50 
 
 APPENDIX A - DATA FROM EXISTING FASTENER PERFORMANCE TESTS 54 
 
 APPENDIX B - STRAIN-VOLTAGE RELATIONSHIP OF INSTRUMENTED CLIPS 61 
 
 APPENDIX C - LABORATORY TESTS OF INDIVIDUAL CLIPS 66 
 
 REFERENCES 69 
 
LIST OF ILLUSTRATIONS 
 Figure Page 
 
 1 Examples of Nonconventional Wood Tie Fasteners 4 
 
 2 Examples of Concrete Tie Fasteners 6 
 
 3 Clip Force-Deflection Properties 14 
 
 4 Two Methods for Measuring Tie Pad Stiffness 15 
 
 5 Measurement of Longitudinal Restraint 17 
 
 6 Measurement of Fastener Lateral and Rollover Restraint .... 17 
 
 7 Comparison of Longitudinal Resistance With and Without 
 
 Rail Vibration 20 
 
 8 Field Measurements at FAST: Fastener Types and Displacement 
 Components 22 
 
 9 Vertical Clip Deflection Vs. Clip Strain From Track 
 Measurements and Two Laboratory Tests 24 
 
 10 Typical Clip Force-Deflection Characteristics 27 
 
 11 Determination of Clip Yield Point for Two Clips 28 
 
 12 Loading Arrangement for Tie Pad Compression Tests 29 
 
 13 Vertical Load-Deflection Characteristics for the 
 
 Polyethylene Pad 30 
 
 14 Effect of Cycle Rate on Load-Deflection Characteristic 
 
 of Grooved "Duraflex" Pads 31 
 
 15 Effect of Cycle Rate on Load-Deflection Characteristic of 
 Grooved Synthetic Rubber Pad 32 
 
 16 Loading Fixtures Used for Longitudinal Restraint Tests .... 34 
 
 17 Influence of Pads and Insulators on Longitudinal 
 
 Restraint of Concrete Tie Fasteners 35 
 
 18 Longitudinal Restraint of Wood Tie Fastener System 37 
 
 19 Loading and Measurement Schematic for Lateral/Rollover 
 Restraint Tests 38 
 
 20 Applied Load Vs. Displacement for Lateral Restraint Tests 
 
 at 20 and 30 Degrees, Hard Polyethylene Pad 39 
 
 21 Applied Load Vs. Displacement for Lateral Restraint Tests 
 
 at 20 and 30 Degrees, Grooved Synthetic Rubber Pad 40 
 
 22 Loading Schematic for Concrete Tie Fastener Fatigue Test ... 43 
 
 23 Typical Load-Deflection and Strain-Deflection Plots From 
 
 Fatigue Test of Concrete Tie Fastener 45 
 
 24 Locations of Clip Cracks Developed During Concrete Tie 
 
 Fatigue Test 46 
 
 vi 
 
LIST OF ILLUSTRATIONS (Continued) 
 
 Figure Page 
 
 25 Loading Schematic for Wood Tie Fastener Fatigue Test 48 
 
 B-l Schematic of Type A Clip Instrumentation 63 
 
 B-2 Measurement of Clip Response Voltage Vs. Deflection in 
 
 Laboratory Fixture 65 
 
 C-l Fixture for Measurement of Clip Force-Deflection-Strain 
 
 Properties 68 
 
 Table 
 
 LIST OF TABLES 
 
 1 Summary of AREA and Amtrak Fastener Performance 
 Specifications 8 
 
 2 Amtrak and AREA Fastener Qualification Test 
 
 Sequence 13 
 
 3 Results of Gauge Widening and Longitudinal Restraints Tests 
 
 on Four-Tie Panels of Wood Ties 19 
 
 A-l Fastener Clip Force-Deflection Properties 55 
 
 A-2 Fastener System Uplift Properties 56 
 
 A-3 Tie Pad Compression Load-Deflection Properties 57 
 
 A-4 Fastener Longitudinal Restraint Properties 58 
 
 A-5 Fastener Lateral Restraint Properties 59 
 
 A-6 Fastener Rollover Restraint Properties 60 
 
 vn 
 
INTRODUCTION 
 
 This investigation was conducted to identify potential improvements 
 in performance requirements for rail fasteners. Improvements are needed to 
 reduce the greater tie and fastener maintenance demands which have emerged 
 with increased levels of lateral and vertical track loading. 
 
 Maintenance problems on highly loaded wood tie track (tie plate 
 cutting and spike killing of the ties, rapid deterioration of surface and 
 alignment) have led to two developments: (1) the trial use of nonconventional 
 fasteners on wood ties, and (2) the introduction and expanding use of con- 
 crete ties. While some of the new systems have brought improvements in the 
 critical area of gauge widening, many have experienced additional problems. 
 These include the failure of components (concrete tie fastener clips, tie 
 pads and insulators, wood tie plates and holddown spikes), the lack of ade- 
 quate longitudinal restraint, and the cracking of concrete ties. 
 
 Performance specifications for concrete tie fasteners have been 
 developed by several railway associations and rail transit authorities. In 
 most cases, the specifications require a series of qualification tests in 
 which the fastener system is subjected to static and dynamic loads. The 
 retention of fastener strength and resistance to permanent deformation are 
 determined by static measurements before and after fatigue tests. While such 
 tests have served to differentiate among candidate systems, they have often 
 not provided reliable indications of performance in track. The value of the 
 tests is limited by a lack of information about the fastener service 
 environment. 
 
 To develop fastener performance requirements which better represent 
 the service environment, a research program was carried out in the following 
 phases: 
 
 a. A review of fastener performance problems, existing 
 performance requirements and available data from 
 laboratory tests was conducted. This review is pre- 
 sented in Reference [1]. 
 
 b. To define a representative fastener loading environment, 
 a field test program was carried out at the Facility 
 for Accelerated Service Testing (FAST). Measurements of 
 rail/tie deflection and fastener clip strains were made 
 on 5-degree curves of wood and concrete tie track. 
 Results of the field measurements and supporting labora- 
 tory tests are reported in Reference [2]. 
 
 c. A laboratory study was conducted to: 
 
 (1) define more realistic fastener fatigue tests. 
 This was accomplished by subjecting two of the 
 
 1 
 
fastener systems to simulated service environ- 
 ments (based on the measured fastener deflections and 
 strains) and comparing the results with observed 
 performance at FAST. 
 
 (2) improve the determination of basic fastener per- 
 formance characteristics. These included the 
 fastener yield load and the stiffness of a con- 
 crete tie pad. 
 
 This report summarizes the preliminary review, presents the 
 essential results of the field measurements at FAST, and describes the sub- 
 sequent laboratory tests. On the basis of this investigation, recommenda- 
 tions are made for changes to current qualification tests of concrete tie 
 fasteners. If adopted, the changes will provide for: 
 
 a. more realistic tests of fastener resistance to fatigue 
 loads 
 
 b. the definition of pad stiffness over the range of pad 
 loads expected in service 
 
 c. elmination or simplification of other tests used in the 
 current specifications. 
 
 Review of Fastener Performance Problems 
 
 Wood Tie Fasteners 
 
 The conventional wood tie fastener in U.S. service consists of: 
 
 a. a tie plate to transfer loads from the rail to the tie 
 
 b. cut spikes to constrain the plate and rail against gauge 
 widening and to constrain the rail against rollover 
 
 c. rail anchors—spring steel clamps which, when attached to 
 the rail base, constrain longitudinal rail-to-tie 
 movement. 
 
 Vertical uplift motion of the rail is allowed through the development of free 
 play between the rail base and the rail line spikes. The amount of free play 
 is adjusted naturally as the rail deflects upward in front of the passing 
 wheels and the line spikes yield slightly from their original anchorage in 
 the tie. Tie pumping is held to a minimum by this development of free play. 
 
This basic fastener has been in use for many years and remains 
 predominant in U.S. track. However, the introduction of 100-ton cars with 
 roller bearings and long unit trains has caused rapid gauge widening, tie 
 plate cutting and spike killing of ties with the conventional fastener. 
 Wear begins in the form of spike pullout and enlargement of spike holes, 
 causing lateral yielding of the tie plate and rotation of the rail. Often 
 the track must be regauged by plugging or filling the spike holes and redriv- 
 ing the spikes. The process accelerates with repeated maintenance and 
 eventually necessitates tie replacement. Finally, frequent transposing and 
 relaying of rail on curves require removal of the line spikes and contribute 
 to spike kill ing. 
 
 Various modifications of the conventional fastener have been 
 introduced to alleviate these problems. These include: 
 
 a. additional spikes 
 
 b. additional spikes with larger tie plates 
 
 c. special screw or locking type spikes as holddown fasteners. 
 
 Some railroads have installed test sections of wood tie fasteners, 
 which represent a major departure from the conventional plate-and-spike 
 fastener system. Examples are shown in Figure 1. 
 
 The greatest departure from conventional design consists of the use 
 of elastically deforming clips to constrain the rail. The clips may be 
 detachable from the tie plate, may be anchored through the plate by screw 
 spikes or other holddown devices, or may be integral with or permanently 
 fixed to a pair of spikes. One major objective in the use of such clips is 
 to eliminate rail anchors. Wood tie fastener test data shown later indicate 
 that some fastener designs actually exceed the longitudinal restraint of rail 
 anchors, but to date this has not been verified in the field. 
 
 Rigid clip designs have also been introduced. Some of these pro- 
 vide a gap between the clip and the rail base to permit rail uplift and 
 thereby reduce pumping. Since the gap eliminates longitudinal restraint by 
 the clip, rail anchors are required. However, detachable clips of either the 
 rigid or elastic type have a major advantage in that the rail can be trans- 
 posed or replaced without respiking. 
 
 Tests of nonconventional wood tie fasteners in revenue service 
 have provided early evidence of improved performance [3]. However, tests at 
 FAST under severe and accelerated conditions have produced the following 
 problems [4]: 
 
 a. elastic clips broken or loose 
 
 b. tie plates broken at clip attachments 
 
 | Refer to References 
 
00 
 LxJ 
 I— 
 
 <c 
 
 _J 
 
 Q_ 
 
 I— I 
 O 
 
 LU 
 Q. 
 O0 
 
 
 o 
 
 
 Q- 
 
 O 
 
 
 
 I— 
 oo 
 
 Q 
 
 o 
 
 O 
 
 o 
 
 o 
 o 
 
 o 
 
 X 
 
 ID 
 
c. screw spikes broken off in the tie 
 
 d. spikes pul led out. 
 
 Concrete Tie Fasteners 
 
 Fastening systems for concrete ties generally consist of: 
 
 a. a pair of detachable clips, either elastic or rigid 
 
 b. clip anchorages or shoulders either embedded in the concrete 
 or inserted into a threaded sleeve 
 
 c. a pair of insulators in the form of separate pieces inserted 
 between the clip and the rail or of material bonded to the fastener shoulders 
 
 d. a pad of elastomeric, rubber or composite material to provide 
 vertical resilience and prevent tie abrasion. 
 
 Examples of concrete tie fasteners are shown in Figure 2. 
 
 Problems identified at FAST and in revenue service test segments of 
 concrete ties have included: 
 
 a. dislocation, fall-out and fracture of elastic clips (pri- 
 marily at the inside-gauge position of the FAST Section 17 
 five-degree curve) 
 
 b. cracking, dislocation and deterioration of pads and 
 insulators 
 
 c. excessive tie skewing 
 
 d. shoulder loosening (Northeast Corridor). 
 
 Problems in skewing of concrete ties are aggravated by the fact 
 that rail anchors are not commonly part of the fastener system except where 
 rigid clips are used with a clip-to-rail gap. However, rail anchors were 
 introduced at FAST to prevent skewing and bunching at the bottom of a 2-per- 
 cent grade in the 5-degree curve of Section 17. Since concrete tie fastener 
 systems are already \iery expensive, it is expected that rail anchors would 
 only be used in the most severe loading environments. 
 
 There is one major performance problem which is peculiar to con- 
 crete tie track. Concrete tie construction creates a track with much higher 
 vertical stiffness than does wood tie construction on a similar roadbed. 
 This must be compensated by the resilience of the rail pad. The Japanese 
 National Railway (JNR) has determined that for its service requirements, 
 this vertical stiffness must be maintained below rather restrictive levels 
 to prevent excessive ballast settlement, ballast particle degradation, 
 growth of rail corrugations and transmission of noise into the passing 
 vehicle [5,6]. 
 
 5 
 
a 
 
 LU 
 Q 
 
 _i 
 r> 
 o 
 
 3Z 
 OO 
 
 Q 
 UJ 
 
 I— 
 <C 
 _J 
 =D 
 OO 
 
 en 
 o 
 
 oo 
 
 LU 
 
 oo 
 
 Q 
 
 <: 
 
 en 
 o 
 
 
 
 LU 
 
 oo 
 
 LU 
 
 OO 
 
 •— i 
 
 < 
 
 i— 
 
 -J 
 
 
 LU 
 
 LU 
 
 
 h- 
 
 
 LU 
 
 
 a: 
 
 
 o 
 
 
 ■z. 
 
 
 o 
 
 
 o 
 
 
 Ll_ 
 
 
 o 
 
 
 oo 
 
 
 LU 
 
 
 _J 
 
 on 
 
 O 
 
 C^ 
 
 1— 
 
 5 
 
 c£ 
 
 X 
 
 _l 
 
 LU 
 
 r> 
 
 
 oo 
 
 
 z 
 
 • 
 
 •-^ 
 
 CM 
 
 LU 
 
 LU 
 
 »— 
 
 cm 
 
 2 
 
 
 < 
 
 i— i 
 
 Q. 
 
 U_ 
 
 LU 
 
 
 OO 
 
 
 a 
 
 
 z 
 
 
 <c 
 
 
 O- 
 
 
 o 
 
 oo 
 
 q. 
 oo 
 
Low vertical stiffness on concrete tie track can only be attained 
 with a pad specifically designed for this purpose. However, many soft pads 
 (stiffness about 500,000 - 1,000,000 pounds/inch) are less durable than 
 harder alternatives (stiffness about 3-5 million pounds/inch). This conflict 
 between requirements for low stiffness vs. durability constitutes one of the 
 principal challenges in the production of cost-effective concrete tie 
 fasteners. 
 
 The continued occurrence of service performance problems for the 
 improved, or nonconventional , designs of wood and concrete tie fasteners 
 indicate the need for the development of laboratory tests which can predict 
 performance to be expected in the field. 
 
 FASTENER PERFORMANCE SPECIFICATIONS 
 
 Chapter 10 of the AREA manual [7] provides specifications for con- 
 crete ties and fasteners. Other agencies and railroads, most notably Amtrak, 
 have incorporated the basic AREA fastener tests into an expanded set of 
 requirements which feature the sequenced repetition of several tests [8]. In 
 Table 1 the performance tests of both the AREA and Amtrak specifications are 
 summarized and compared. Table 2 defines the testing sequence of the Amtrak 
 specification. 
 
 Both specifications require static and dynamic tests on the 
 fastener systems and components to determine the following characteristics: 
 
 a. Strength of the fastener anchorage 
 
 b. Resistance to permanent deformation after cycling through 
 compression and uplift loads 
 
 c. Resistance to fatigue loading 
 
 d. Longitudinal strength against rail creep after the application 
 of cycl ic loads 
 
 e. Stiffness against gauge widening and rail rollover under 
 vertical and lateral loads 
 
 f. Vertical resilience 
 
 g. Electrical impedance. 
 
 Other than the requirements of the FRA Track Safety Standards in the 
 maintenance of track geometry limits, there are no performance specifications 
 for wood tie fasteners. Chapter 5 of the AREA manual specifies requirements 
 for the strength and durability of plates and spikes. However, it is the 
 interaction of these components with the tie which causes most wood tie 
 problems. Since many nonconventional wood tie fastener components are also 
 used on concrete ties, their design is affected by specifications for concrete 
 tie fasteners. 
 
 7 
 
o- 
 
 LU 
 
 OO 
 O 
 
 
 c_> 
 
 LU 
 
 oo 
 
 <_> 
 
 ■z. 
 
 <c 
 s: 
 
 Qi 
 O 
 U_ 
 OC 
 LU 
 CL 
 
 Q£ 
 
 LU 
 
 LU 
 
 I— 
 
 oo 
 
 a: 
 
 Q 
 
 LU 
 
 >- 
 on 
 
 oo 
 
 S- 
 
 to 
 
 c 
 
 O OJ 
 
 . -I- +J 
 +->+-> cu 
 
 t- TO S- 
 
 cu E u 
 
 W i. c 
 
 coo 
 
 •r- 4- U 
 
 <D 
 
 4- "O 4- 
 
 O O 
 
 ■*-> 
 
 (U C Ol 
 
 Did) C 
 
 03 C -r- 
 
 Q. ro -^ 
 
 O- E U 
 
 •i— S- 03 
 
 r- <U S- 
 
 to Q. u 
 
 O O O 
 
 s 
 co 
 
 o 
 
 4- 
 Ol 
 
 -a 
 
 O) 
 
 c 
 
 E 
 s- 
 cu 
 
 Q. 
 
 ai 
 
 cu 
 
 s- 
 o 
 c 
 o 
 o 
 
 S- +-> 4- 
 
 O 5- O 
 
 CD 
 
 o c C 
 
 03 4- 
 
 +J o 
 
 o 
 
 s- c 
 o 
 
 o 
 
 03 
 
 o 
 
 cJ 
 
 CO 
 
 S- 
 
 ai 
 
 c 
 
 c 
 
 c: 
 cu 
 to 
 o 
 o 
 
 CL 
 
 +J 
 
 c 
 
 O) 
 
 c: 
 o 
 a. 
 
 E 
 
 o 
 u 
 
 >> 
 
 c 
 to 
 
 4- 
 O 
 
 CU 
 
 s- 
 +-) 
 
 CJ 
 03 
 
 s- 
 
 4- 
 
 CU 
 I/) 
 tO 
 
 a> 
 'a 
 
 5- 
 
 (0 
 
 S- 
 
 LU 
 
 a: 
 
 o 
 o 
 cc 
 
 4- +J 
 
 O </> 
 
 to 
 -a cu 
 
 03 i — 
 
 o 
 
 i— +j 
 to 
 
 to ~a 
 
 o i— 
 
 •i- o 
 +-> .c: 
 
 i- l/> 
 
 CU "O cu 
 
 > c +-> 
 
 to Z5 
 
 >> c 
 
 f (/I •!- 
 
 O. O- E 
 CL-f- 
 
 LU 
 
 on 
 
 CO 
 
 < 
 
 -Q 4J 
 
 I— :3 
 
 I C 
 
 4- E 
 
 O CO 
 LO 
 
 CM 4-> 
 
 I/) 
 
 4- to 
 
 O CU 
 
 a 
 
 cr 03 
 s_ 
 
 o -o 
 
 +-> I— 
 
 o 
 
 >■>-£- 
 
 "0.-0 
 o_ c 
 <C 03 
 
 LU 
 
 a: 
 
 CO 
 
 
 1 -0 
 to to 
 s- o 
 
 tO 1— 
 
 a, 
 
 cu cu 
 
 to to 
 
 «3 
 
 -a cu 
 
 03 1— 
 Q. CU 
 
 S- CL 
 
 cu -a 
 c 03 
 
 •1- o 
 
 E r- 
 
 s- 
 
 cu c 
 
 +-> o 
 
 cu •<- 
 
 Q 4-> 
 
 Cu 
 LO 
 
 4- 
 o 
 
 "O • 
 03 to 
 O Q- 
 
 X 
 
 to o 
 
 >> o 
 
 c— +-> 
 
 0. 
 a. o- 
 
 cu 
 
 ^ 
 
 g 
 
 to L 
 
 CM 
 
 CO 
 
 LU 
 
 CC 
 LU 
 
 00 
 
 "%i 
 
 1 
 
 o 
 
 LO 
 CM 
 
 -o 
 
 03 
 O 
 
 00 
 
 ^ 4-> 
 
 •— • 3 
 
 C3 ° 
 
 ^ r— 
 
 "-< 3 
 
 2^ Cu 
 
 00 
 
 I/) 
 CU 
 
 CU 
 
 zs 
 a- 
 i~ 
 o 
 
 l-H 
 
 -(-) 
 
 1 
 
 <o 
 
 Q. 
 
 s_ 
 
 ZD 
 
 03 
 
 
 O. 
 
 CT 
 
 cu 
 
 Z 
 
 00 
 
 1 — 1 
 
 
 Z 
 
 -a 
 
 LU 
 
 03 
 
 I— 
 
 CL 
 
 OO 
 
 
 ■=c 
 
 II 
 
oo 
 o 
 
 
 LU 
 
 oo 
 
 C_3 
 
 §£ 
 
 o 
 ll- 
 cs: 
 uj 
 
 Q. 
 
 00 
 
 a; 
 
 UJ 
 
 a: 
 
 >- 
 
 o 
 
 c_) 
 
 CO 
 
 
 E 
 j= 
 
 
 . 
 
 4-> 
 
 
 i 
 
 
 
 
 O 
 
 
 o 
 
 C 
 
 
 ^o tJ 
 
 
 
 l 
 
 
 1 
 
 CD 
 
 
 o 
 
 O ^ 
 
 
 
 ^ 
 
 
 V 
 
 o 
 
 
 o 
 
 a 
 
 
 
 o 
 
 
 o 
 
 Q. 
 
 
 Q. 
 
 § § 
 
 
 
 C\J 
 
 
 <>J 
 
 E 
 
 
 s 
 
 
 oo 
 
 
 
 
 O 
 
 
 o 
 
 +? 
 
 
 1— 
 
 +-> 
 
 
 ■V 
 
 o 
 
 
 o 
 
 a 
 
 
 
 CO 
 <T3 
 
 ai 
 
 
 3 
 
 ft) 
 
 c 
 
 
 ^1 
 
 a 
 
 ^O "!*■ 
 
 
 LU 
 
 i — 
 
 
 N 
 
 fO 
 
 
 ci 
 
 
 
 Qi 
 
 
 
 
 
 
 
 t- 1 . ^Or— 
 
 
 i — i 
 
 ID 
 
 tO 
 
 
 ^ 
 
 O 
 
 
 O 
 
 °^ °a 
 
 
 o- 
 
 
 
 
 
 
 
 en o o ci 
 
 
 LU 
 
 CD 
 
 
 cJ 
 
 CD 
 
 
 CO 
 
 a a cn-s; 
 
 
 CC 
 
 
 
 o 
 s: 
 
 S- 
 Z3 
 
 
 | 
 
 •5 o ci -K> 
 
 "c a a 
 
 
 
 <o 
 
 
 « 
 
 +-> 
 
 
 V 
 
 ^s a ■>< 
 
 
 
 +-> 
 
 
 V 
 
 Q. 
 
 
 a. 
 
 
 
 CO 
 
 
 ■^ 
 
 3 
 
 
 
 
 
 CO 
 
 
 s 
 
 
 
 
 ^ CO 
 
 
 
 CD 
 
 
 o 
 
 
 o 
 
 OS O J^ 
 
 
 
 CC 
 
 
 c* 
 
 z: 
 
 
 z 
 
 ^ c{ S en 
 
 
 
 . 
 
 
 
 
 . 
 
 .^r 
 
 
 
 
 CD -O 
 
 
 tsl 
 
 3: 
 
 to • 
 
 Ll_ 
 
 H- a 
 
 
 U_ 
 
 
 s- s- c: c ^ -^ 
 
 § 
 
 
 O C\J 
 
 o 
 
 ^ ^ 
 
 
 Ci 
 
 
 CD CD •> »p- n3 a 3 
 
 <2> 
 
 r— 
 
 «^ 
 
 ^ S 
 
 
 ^J- 
 
 
 C 4-> r— E -4-> O 
 
 s; 
 
 vO CJ 
 
 +-> 
 
 1 — 
 
 O Cr 
 
 
 
 
 CD <0 tO CO cj -a 
 
 Cj 
 
 O 
 
 C CO 
 
 
 N") O 
 
 , 
 
 
 
 -t-> 3 > LO -i- 'V 
 CO O i — CO V oo 
 
 V 
 
 
 O CD «+- 
 
 £= 
 
 O 
 
 -Q 
 
 a 
 
 
 3 
 
 v^o 
 
 •^-1— O 
 
 ia 
 
 >« ^ 
 
 a. 
 
 ci 
 
 
 rn C E CD 
 M- •>- CD S- S- -CS -^ 
 
 \ s- o a o 
 
 A3 4 t/) 
 
 j= 
 4_> 
 
 ° cr. 
 
 ^ 
 
 -a 
 
 
 
 S- O CD 
 
 
 $) -o 
 
 - i 
 
 1 a 
 
 
 CD CO M- i — <rf ^sd 
 
 •r- S_ S- -r- 
 
 ^ 
 
 <5j P O 
 
 tO i— 
 
 CL CO O 
 
 CO 
 CO 
 
 ^ 1 
 
 t 
 
 
 +-> 3 CD • <a -0 < 
 
 w 
 
 a. 5) *e 
 
 CD +J >> 
 
 CD 
 
 ^ cj 
 
 o ^5 
 
 — J <Z> 1 O 
 
 
 • •ty 
 
 sx ■<} 
 
 CO i — u 
 
 i — 
 
 T-^ 
 
 <V 
 
 
 CD -S= M- 31 1 O "W U. 
 +-> n3 O V « 
 
 CD CO O 4-> « 3 •< 
 i — S- CO 1 =\J C 
 
 Q.S- 3 „ r- d ~C c 
 
 ■ 3 
 
 td S ft) 
 
 Z5 ^ 
 TD CO C^^ 
 
 CD 
 
 a^ 
 
 
 cJ 
 
 LU 
 
 $ -3 
 
 ) *e 
 
 ra CD O CO 
 
 S- 
 4-> 
 
 .3 o 
 c3 ~a 
 
 
 C£. 
 
 eo o r •>- <> ^ *e 
 
 O «Cll 
 
 i — CD i — ^ 
 
 fO 
 
 c3 v 
 
 =* <3 . tQ 
 
 .c> w « 
 
 rD 
 
 Ot--C^ fO S N CI tj 
 
 -s; os 
 
 •r- CO •!— 
 
 S- 
 
 .3 ^> 
 
 "<*■ «-^ t 3 
 
 Q 
 
 >, •— o a ts <3 isi 
 
 -x! 
 
 n3 =5 E O 
 
 O) 
 
 I •$ 
 
 
 c3 
 
 LU 
 
 
 5- OO 
 
 Q. 
 
 
 
 a 
 
 <_) 
 
 CD i — i — CD CJ O 
 
 =N> 
 
 s: . 
 
 C OO 
 
 E 
 
 N-) •> 
 
 
 s 
 
 O 
 
 1/1X1 C S~ «3'VQ cO ~Q 
 
 s- E •>- >> 3 Poo ;> s 
 
 S «JS 
 
 CD tO - 
 
 CD 
 
 ^W - 
 
 
 cJ 
 
 a; 
 
 ^ r^ 
 
 C O >> . 
 
 +-> 
 
 ^ S- 
 
 <— U~l LTl 
 
 o_ 
 
 CDCD-Cr— CO 0~CS'- 
 
 O Cj 
 
 ■r- * — i — ^** 
 
 
 =N> S ^- 
 
 2b.cs.t2 
 
 . Ci C^l 
 
 
 E co +j a. re S 
 
 S o 
 
 
 ^ ao. 
 
 ■a 
 
 a;^ 
 
 r- ^*- 
 
 
 Eco-1-Q.CD £S!+J cj •<? 
 
 CD • a. co 
 
 tO 
 
 a ^ 
 
 
 ^ 
 
 
 •— i to 3 io IE t— >•>?<: 
 
 i OC « 
 
 2 -Q — 
 
 QQ. < • 
 
 Q. 
 
 < ^j- 
 
 
 fi 
 
 
 V 
 
 
 
 
 
 v: 
 
 
 
 
 «=t < 
 
 
 
 i! 
 
 
 < 
 
 
 
 
 £ s 
 
 
 
 LU 
 
 
 tx. 
 
 
 
 
 
 
 CC 
 
 
 h~ 
 
 
 
 
 <c s 
 
 
 
 < 
 
 
 S 
 
 
 
 
 < 
 
 
 
 
 
 < 
 
 
 
 
 ( 
 
 "V_ 
 
 ^J~ 
 
 
 1— co 
 
 
 
 
 
 
 
 
 
 * 
 
 oo a. 
 
 
 
 
 
 
 
 
 
 • 
 
 LU •!- 
 
 I- ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 s — "N 
 
 
 
 
 
 
 
 
 CO 
 
 LU 
 
 1— 
 
 H 
 
 
 _ j 
 
 
 
 hC 
 
 \1& 
 
 ,. 
 
 D- 
 Q *J3 O 
 
 <C ' oo 
 
 
 
 -^ 
 
 
 
 r?^ 
 
 )*Sl 
 
 
 o 
 
 
 
 
 o 
 
 
 
 
 
 V ^0 
 
 
 >v \ 
 
 
 
 h- 
 oo 
 
 LU 
 
 Q 
 LU 
 
 
 
 n 
 
 
 
 *— sO 
 
 
 LU 1 V 
 
 I— JV^ \- 
 
 1 1 1 ^ ^ 
 
 
 
 
 
 
 
 
 v 
 
 : 
 
 
 
 
 
 
 
 
 ^s ( >~ ' 
 
 
 1— 
 
 s: 
 
 
 
 
 > 
 O 
 
 
 
 
 3 
 
 
 D_ \ 
 
 
 
 
 
 
 
 I— t 
 
 
 
 
 1 
 
 
 
 
 
 LU ° 
 
 
 
 
 
 
 
 < 
 
 H 
 
 
 
 Uh 
 
 
 
 
 H 
 
 
 a: o 
 
 CNJ 
 CD 
 
 z: 
 
 
 
 
 
 
 
 •—i 
 
 
 — 
 
 
 
 
 
 
 
 i — i 
 
 
 
 < 
 
 < 
 
 
 
 h- 
 
 C_) "* 
 
 LU ac «s .o 
 
 _l "*— — 
 
 LU 
 
 3 « 
 
 
 
 ^ <: 
 
 S s 
 
 2 < 
 
 
 
 
 
 
 
 oo 
 
 
 
 tT 
 
 
 
 
 
o 
 
 ■=c 
 
 CJ 
 
 c_> 
 
 OO 
 
 C_) 
 
 o 
 
 u_ 
 
 CC 
 LU 
 D_ 
 
 cc 
 
 oo 
 
 cc 
 
 <: 
 
 <C 
 UJ 
 CC 
 
 <c 
 
 >- 
 
 CC 
 
 <c 
 
 o 
 
 CJ 
 
 CQ 
 
 <c 
 
 
 
 
 
 
 
 
 
 T3 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 C CO 
 
 
 
 ■; 
 
 
 
 
 
 
 
 
 
 
 
 <a +-> 
 
 
 
 i — 
 
 
 
 
 
 - 
 
 
 
 
 
 
 c 
 
 
 
 Ci LTl 
 
 
 
 
 
 ^3- 
 
 
 
 
 
 
 ■a cd 
 <a E 
 
 
 
 
 T3 • 
 
 S SI 
 
 
 
 
 "" 
 
 o 
 
 
 
 CD 
 
 
 CD CD 
 .C CJ 
 
 
 
 S ^ 
 
 ■sJ 
 
 
 CO 
 
 
 <T3 CO 
 
 SZ CD CD 
 
 +-> cn+J 
 
 
 
 4-> 
 CO 
 
 Z3 
 
 
 fO 
 
 •■- a. 
 
 fO CO 
 
 
 
 ^•^J 
 •^ 2 ^ 
 
 '■3 2 
 
 
 t— 
 
 LU 
 
 
 CD 
 
 ro _3 
 
 a. c: 
 
 Q.'i- 
 
 "" E -S 
 
 l/> LO ^ 
 
 
 
 E 
 ■u) 
 
 
 S- -r- 
 
 -o • 
 
 
 
 
 Su 
 
 CJ 
 
 CJ ~CS 
 
 
 LU 
 
 cc 
 
 
 fO CO r 
 
 CD CD 
 
 
 
 C 
 CD 
 
 E • 
 
 
 CD ■— «* 
 
 cd <a "^ 
 
 2 S-r— 
 
 
 
 
 O CJ 
 
 T3 cJ 
 
 
 l-H 
 
 
 S- +-> 
 
 1— cj 
 
 
 
 CD = 
 
 
 -u) CD 
 
 
 
 *< o <3 
 
 
 
 _0 
 
 LU 
 
 cc 
 
 
 CD 13 r 
 
 C 
 CD T- 
 cr> E 
 
 CO ,i-> 
 
 o -o 
 
 
 
 > 00 
 
 O s-s. 
 E r- 
 
 
 cd +-> -a 
 
 r— CD 
 CD O 
 
 
 
 
 
 
 
 ro 
 
 t— S- ^ 
 
 
 
 CD C 
 
 
 U CD X 
 
 
 
 S O 
 
 -c 
 
 
 
 
 Q.O0 +-> CD <y 
 
 
 
 <s> ra 
 
 
 C CO CD 
 
 
 
 V -y 
 
 C> CJ 
 
 
 
 
 Q. -r- -a Cj 
 
 ■r- C -O C jJ 
 
 1 '!— "O 'I— 
 
 
 
 
 
 CD ra 
 CD O 
 
 
 
 
 «3 « 
 
 
 
 
 CO -C 
 
 fo ra cJ 
 O CD § 
 
 
 
 i — CO 
 •r- CO 
 
 
 ct- ,— C 
 4- •.- c 
 
 
 
 
 ^^ 
 
 
 
 
 O -r- 
 
 
 
 fO CD 
 
 
 •i— fO ra 
 
 
 
 ci s^ ^j 
 
 cj s; 
 
 
 
 
 ^ S ^ J- oo 
 
 
 
 CC .— 
 
 
 Q L U 
 
 
 
 fX O o 
 
 Sl-y 
 
 
 
 
 T3 
 
 1 
 
 V 
 
 
 CO 
 CO = 
 
 
 .c 
 
 
 
 1 
 
 
 s: 
 
 
 CO 
 
 ra 
 
 CD 5 
 
 s 
 
 
 CD 00 
 C r— 
 
 
 O r— 
 
 
 
 3 3 
 
 
 3 
 
 
 "O 
 
 CD 
 
 
 
 3 
 
 
 S- ra 
 
 
 
 s . 
 
 s; 
 
 cj 
 
 
 ra 
 
 CO 
 
 5- cj • 
 
 NO 
 
 
 CO 
 
 
 -C s- 
 
 
 
 W 
 
 cJ 
 
 
 o 
 
 ra 
 
 CD cl tj) 
 
 o 
 
 
 »-o 
 
 
 ■M CD • 
 
 
 
 S*ti 
 
 
 cj 
 
 
 1 — 
 
 -Q 
 
 ra •• «J C 
 
 
 
 CO CD 
 
 
 -u> +J 
 
 
 
 •-J « 
 
 •g 
 
 s<3 
 
 
 
 
 3 
 
 
 Q. CD 
 
 
 CO ra C 
 
 
 
 o 
 
 SL 
 
 cJ • 
 
 
 ■o 
 
 CD 
 
 ^ 
 
 CO • Sc. <3 
 
 
 •r- CJ 
 
 ^: x 
 
 
 CLr— •!- 
 
 •i- o 
 
 
 
 S cj 
 
 ■*j 
 
 •^-o- 
 
 ■*•£ 
 
 
 +-> 
 
 ra ' 
 
 ra CO cJ -J^ cJ 
 
 •o o 
 
 
 CD 
 
 
 Ji£ CD CL 
 
 
 
 s 
 
 ^ 
 
 ~C! cJ 
 
 
 ra 
 
 i- "> 
 
 CD W 
 
 •w o 
 
 
 ( — 
 
 
 +-> 
 
 
 
 t3 O • 
 
 ^f 
 
 <rf S 
 
 
 CD 
 CD- 
 
 CD 
 
 O t- cJ <3 *? 
 
 ■y s 
 
 c5 d 
 
 
 
 
 LO WD 
 • C ra 
 
 
 
 o -<? ci 
 
 
 Q. cJ 
 ^1 
 
 
 CD 
 
 n3 3 
 
 r- T3 
 
 
 O CD 
 
 
 O -r- O 
 
 
 
 CJ -Q O 
 
 o c> 
 
 
 C. 
 
 C -P ItJ ^ -Q ^C3 
 
 S>. cr 
 
 
 +-> E 
 
 
 CM E ■— 
 
 • 
 
 . 
 
 O CJ 
 
 V — 
 
 LU 
 CC 
 
 O) 
 
 • •!— U CD a) cJ O 
 
 Ci) Ci) 
 
 
 CD 
 
 a. > 
 rs o 
 
 
 M- r^ +-> 
 
 O CD ra 
 
 ^ 
 
 o 
 
 -u) 
 CD 
 
 ii3 
 
 
 ow 
 +5 ts 
 
 LU 
 
 Ct- 
 
 lo M- <d ""ci Si. <rf 
 
 O i — CD CD cj Si. cJ 
 
 Ci) 3 
 
 
 E 
 T3 
 
 
 TD O CD 
 
 i 
 
 E 
 CD 
 
 2 5 V 
 
 *cs •<> 
 
 5 8 
 
 CJ O 
 
 CJ 
 
 ra 
 
 O T3 (O O !rf V 
 
 
 fO CD 
 
 j^j 
 
 ra ■*-> O 
 
 j^ 
 
 S_ 
 
 3 cJ 
 
 « +v 
 
 ^ cj 
 
 O 
 
 
 "3" s- 
 
 en o -q 
 
 ^U 
 
 
 O CO 
 
 \ 
 
 O C 
 
 O 
 
 •r- 
 
 O ^ til 
 
 o s 
 
 
 q: 
 
 E 
 
 C\J o 
 
 CD *e -o S 
 t- O 2. WW 
 
 
 r— ra 
 
 3 
 
 i — CO ra 
 
 3 
 
 3 
 
 tj ao 
 
 Cj CJ 
 
 CJ~S 
 
 Q. 
 
 S- 
 
 C|— 
 
 S 
 
 
 JD 
 
 Cr- 
 
 S- +-> 
 
 cy 
 
 o~ 
 
 •y 
 
 £ 
 
 ^^ 
 
 
 o • 
 
 >> 
 
 ^ 3 
 
 Ci) o 
 
 
 >, 
 
 O 
 
 >> CD CO 
 
 CJ 
 
 CD 
 
 CJ *"". -Q 
 Si. 
 
 ^ cJ 
 
 
 s_ to 
 
 Q-r- 
 
 LO -t-> Ci) ~c$ ft) 
 
 ra s <3 s: 
 
 § ° 
 
 
 CLT- 
 
 *e 
 
 Q., — CO 
 
 «e 
 
 s- 
 
 US 
 
 s ^ 
 
 
 CD CD 
 
 a. o 
 
 CD M- « O W 
 
 -s; s; 
 
 
 Q. n3 
 
 o 
 
 Q_ o CD 
 
 o 
 
 o 
 
 Sl v -= o 
 
 Q. s; 
 
 ^^ 
 
 
 Q. +-> 
 
 <c -c : 
 
 \— w 
 
 
 <C S- 
 
 2 
 
 < LS. 
 
 2 
 
 z 
 
 < O "^ 
 
 <W 
 
 
 
 
 V 
 
 
 
 # , 
 
 V 
 
 # 4 
 
 ic 
 
 
 ii 
 
 
 
 
 < 
 
 
 •< 
 
 
 
 ■=C 
 
 < 
 
 < 
 
 < 
 
 cC 
 
 •< 
 
 
 
 
 LU 
 
 
 P 2 
 
 
 
 LU 
 
 (M 
 
 LU 
 
 CC 
 
 LU 
 
 CX 
 
 
 
 
 CC 
 
 
 h- 
 
 
 
 o; 
 
 r~ 
 
 QC 
 
 \~ 
 
 DX 
 
 K 
 
 
 
 
 < 
 
 
 < 
 
 
 
 =c 
 
 < 
 
 < 
 
 ■< 
 
 < 
 
 < 
 
 
 
 
 h- 
 
 
 
 
 
 Q. 
 
 a 
 
 b. 
 
 
 
 
 
 
 
 Z 
 
 
 
 
 
 
 vjp 
 
 W. 
 
 
 
 
 
 
 
 t— i 
 
 _j 
 
 
 
 
 ^ 
 
 n 
 
 V^ 
 
 
 
 
 
 
 
 el 
 
 < 
 
 
 
 
 
 > 
 
 ^- J 
 
 
 
 
 
 
 
 q; 
 
 ZZl 
 
 
 
 
 5 I 
 
 w^S^ 
 
 ^_J1^V^ 
 
 -\ 
 
 I— 
 
 
 
 
 
 i— 
 
 1 — 1 
 
 
 
 i— 
 
 — L, 
 
 <~Cz&^ 
 
 M ^'li / 
 
 ? 
 
 00 
 
 
 
 
 
 oo 
 
 Q 
 
 
 
 oo 
 
 
 JX^X 
 
 vS g; fk ( 
 
 
 LU 
 
 
 
 
 
 LU 
 
 ID 
 
 
 
 
 LU 
 
 
 /\J 
 
 •Z Vx> 
 
 
 r— 
 
 
 
 
 
 Ct 
 
 (— 
 
 
 
 
 1— 
 
 -1 
 
 1 
 
 i^ > *^ 
 
 
 
 
 
 
 
 
 _1 
 
 i — i 
 CD Q 
 
 
 I— 
 
 
 h- 
 
 
 
 in 
 
 
 2: 
 o 
 
 
 
 
 ** 
 
 
 <c 
 
 z: <c 
 
 
 ^ oo 
 
 
 ■2L 
 
 +-> 
 
 c" 
 
 o 
 
 
 1— i 
 
 
 
 
 ^ 
 
 
 z 
 
 o o 
 
 
 LU Ci 
 
 
 t — i 
 
 c 
 
 
 CNJ 
 
 
 r— 
 
 
 
 
 
 
 1 — 1 
 
 _l _l 
 
 
 ^1 LU 
 
 
 <x. 
 
 CD 
 
 
 
 
 C_) 
 
 
 
 IY 
 
 
 
 Q 
 
 
 
 LU C_> 
 
 
 CC 
 
 E 
 
 
 C 
 
 
 LU 
 
 ^^^ 
 
 
 1 
 
 •s. 
 
 
 1— 
 
 
 
 C_) ZD 
 
 
 00 
 
 CD 
 
 > 
 
 
 o 
 
 
 Lu 
 
 _^p 
 
 — 
 
 / 
 
 
 (— 
 
 • — i 
 
 
 
 _i oo 
 
 
 LU 
 
 O 
 
 
 +J 
 
 
 LU 
 
 I / 
 
 F^ ■" — ^^mt 
 
 \ 
 
 
 oo 
 
 CD 
 
 
 
 a. zz 
 
 
 CC 
 
 s: 
 
 
 ra 
 
 
 Q 
 
 \—l 
 
 
 \ 
 
 
 LU 
 
 ^ 
 
 
 
 U~) <C 
 
 
 
 
 
 +-> 
 
 
 
 
 
 \ 
 
 
 1— 
 
 O 
 
 I 
 
 C3 
 
 
 
 p— i q: 
 
 
 Q 
 
 o 
 
 _J 
 
 rO 
 
 i- 
 0) 
 
 
 o 
 
 CC 
 
 
 Q 
 <C 
 O 
 
 _l 
 
 
 
 *OT 
 
 ^. 
 
 
 Z 
 
 
 
 ^-/J* 
 
 
 
 4-> 
 
 
 •^~ 
 
 
 
 
 
 
 
 i — i 
 
 
 
 ^T" 
 
 
 _1 
 
 fO 
 
 
 n3 
 
 
 Q 
 
 
 
 
 
 ■Z. 
 
 
 
 s^^ K 
 
 
 <c 
 
 _l 
 
 
 oct 
 
 
 cC 
 
 
 
 
 
 LU 
 
 
 
 
 
 CC 
 
 
 
 
 
 Cl. 
 
 
 
 
 
 1— I— 
 
 
 
 
 
 LU 
 
 
 
 
 
 
 
 
 
 
 oo oo 
 
 
 
 
 
 1— 
 
 , — ^ 
 
 
 *— ^. 
 
 
 LU 
 
 
 
 
 
 CC LU 
 
 LU h- 
 
 
 
 
 
 cC 
 
 fO 
 
 
 JD 
 
 
 
 
 
 
 
 
 
 
 — 1 
 
 CO 
 
 
 
 
 
 
 
 
 10 
 
o 
 
 § 
 
 o 
 
 
 (_) 
 
 en 
 
 < 
 
 o 
 u_ 
 
 LxJ 
 
 ex. 
 
 
 
 UJ 
 
 >- 
 on 
 
 o 
 c_> 
 
 CO 
 
 UJ 
 
 IS 
 
 
 *e 
 
 
 v= 
 
 
 *e 
 
 . 
 
 o 
 
 3 
 
 C71 
 
 a) 
 
 s; 
 
 si 
 
 3 
 
 o 
 
 3 
 
 
 •v 
 
 O 
 
 Si 
 
 
 
 in 
 
 o 
 
 S 
 
 -2 
 
 « 
 
 ft) S -Q 
 
 OH 
 
 a 
 
 UJ 
 
 o 
 
 o 
 
 CX, 
 
 ai 
 
 E 
 CD 
 
 s_ 
 
 O" 
 QJ 
 
 on 
 
 < 
 
 +-> 
 cu 
 
 E 
 (U 
 
 i- 
 
 CT 
 O) 
 
 i- 
 
 < 
 
 c: 
 
 0) 
 
 E 
 QJ 
 
 i- 
 
 cr 
 
 s- 
 
 on 
 
 
 
 00 
 
 11 
 

 
 
 
 o\o 
 
 
 
 
 
 
 
 
 s 
 
 u-> tsi 
 
 
 
 
 
 
 
 
 O o\Q 
 
 V *-~ -^ *•« -Q 
 
 
 
 
 
 
 
 
 =vj s; 
 
 £ s; <rf o s; 
 
 
 
 
 
 
 
 
 <3 o -u . so 
 
 
 
 
 
 
 
 
 O T3 -Ed 
 
 ti ci) O 
 HV cJ-V 
 
 «v S? _, cj -o ^ -o 
 -s; <^ W < s: « o 
 
 
 
 
 
 
 oo 
 
 
 £ x -w 
 
 o ' sx o o sx • 
 
 
 
 
 
 
 1— 
 
 
 •>> Cl) 
 
 «rf 'fiA^ -Cos 
 
 
 
 
 
 
 LU 
 
 
 O V « 
 
 O ^i — V tj (J cJ 
 
 
 
 
 
 
 s: 
 
 
 •^ O <C 
 
 <N> -Q O O O 
 
 
 
 
 
 
 LU 
 
 
 s; 
 
 
 
 
 
 
 
 en 
 
 
 cj "tJ 
 
 
 
 
 
 
 1— 1 
 
 
 i'V^I 
 
 
 
 
 
 
 
 ID 
 
 
 -a <Q o 
 
 
 
 
 
 OO 
 
 o- 
 
 
 s .§ 5 
 
 2=n» ^ x*c -vdu~i 
 
 S cj cj <v +? ^ 
 
 
 
 
 
 •z. 
 
 LU 
 
 
 
 
 
 
 O 
 
 a; 
 
 
 o o *e 
 
 oo o^-<> 3ss: 
 c cJ . , SX -^ cJ +v 
 
 
 
 
 
 »— t 
 
 
 
 •>) sx 
 
 
 
 
 
 1— 
 
 
 
 <? O S 
 
 
 
 
 
 <c 
 
 
 
 •<> Si. o 
 co -v o 
 
 
 
 
 
 c_) 
 
 
 
 
 
 
 
 1 — 1 
 
 
 
 
 
 
 
 i— t 
 
 
 
 s 
 
 i ~c ^ 2 o »i Ts *e 
 
 
 
 
 
 o 
 
 
 
 O NO^O 
 
 ■"=*• cJ tt)^: S . a) o 
 
 
 
 
 
 LU 
 
 
 
 C_) O O 
 
 ^-*e p."Od *~ *e s 
 
 
 
 
 
 Q- 
 
 
 
 
 
 
 
 
 
 00 
 
 
 
 
 
 
 
 
 
 LU 
 
 
 
 
 
 
 O 
 
 
 
 -s; 
 
 
 
 
 
 
 ■z. 
 
 
 
 3 ~§ 
 
 
 
 
 
 
 <c 
 
 
 
 ti 
 
 
 
 
 
 s: 
 
 
 
 3 - ^ 
 
 s; 
 
 
 
 
 
 q: 
 
 
 
 cj 
 
 « 
 
 
 
 
 
 o 
 
 U- 
 
 
 
 3 3"S 
 
 • -5 -T3 
 
 
 
 
 
 o: 
 
 
 
 cj O c; 
 
 o 2 si S • 
 
 
 
 
 
 LU 
 
 
 
 S a) o 
 
 
 
 
 
 Q_ 
 
 
 
 cj oo <^> 
 
 ■•^ S ti 2 s; 
 c^ o <? a 
 
 
 
 
 
 OL 
 
 
 
 ti cJ 
 
 
 
 
 
 LU 
 
 
 • 
 
 c^ • O 
 
 
 
 
 
 ■z. 
 
 
 +-> 
 
 sx~cs V 
 
 
 
 
 
 LU 
 
 
 sz 
 
 -o cJ 
 
 S3 <)^ Si. Ci) 
 
 
 
 
 
 1— 
 
 oo 
 
 LU 
 
 a) 
 
 E 
 
 o =^5 
 
 O cj O 
 
 a) C '^ «N> 
 
 ~C3 'vj cj . -£J S= 
 
 -Ci O s; cj 
 
 
 
 
 
 «C 
 
 a: 
 
 CD 
 
 
 
 
 
 u_ 
 
 CD 
 
 s- 
 
 ~CS +V O »o "CJ 
 
 
 
 
 
 ^ 
 
 Q 
 LU 
 
 Z3 
 
 oJ s: ci) 
 
 ^ ^4- cj -W t- cO 
 c^ ^ d d) ^ § 
 
 
 
 
 
 a: 
 
 o 
 
 
 N a 3 
 
 
 
 
 
 h- 
 
 ai 
 
 S- 
 
 "§3 3 
 
 X3 S2_ « so D_ ^ 
 
 
 
 
 
 s: 
 
 Q_ 
 
 
 
 
 
 
 <c 
 
 
 o 
 
 ci) ci cj 
 c* ^> S 
 
 S Si. cj cj Si. cJ 
 
 
 
 
 
 Q 
 
 
 
 
 
 
 
 
 • 
 
 Z 
 
 
 
 
 
 
 
 
 Csj 
 
 <C 
 
 
 < 
 
 < 
 
 
 
 
 
 Ci) 
 
 <C 
 
 
 LU 
 
 e* 
 
 
 
 
 
 «n) 
 
 LU 
 
 
 q; 
 
 P 
 
 
 
 
 
 -Q 
 
 a: 
 
 
 < 
 
 s 
 
 
 
 
 
 ,«d 
 
 < 
 
 
 
 < 
 
 
 
 >. 
 
 u 
 o 
 
 r- 
 
 Ll_ 
 
 
 
 
 
 s: 
 
 O 
 
 
 
 
 
 
 [q 
 
 s >, 
 
 •v 
 
 >- 
 
 
 
 
 
 
 E 
 
 
 "Ci 
 
 cc 
 
 
 
 
 
 <+- 
 
 <D 
 
 CD -Q 
 
 Cl) 
 
 <=C 
 
 
 
 
 
 
 o 
 
 n3 CO 
 
 •■- E 
 
 s; 
 
 s: 
 
 
 
 
 
 
 
 CO 
 
 +J CD 
 
 •v 
 
 s: 
 
 
 
 
 
 •• 
 
 I/) 
 
 O "3 
 
 CO 
 
 S3 
 
 ZD 
 
 
 
 
 
 
 4-> 
 
 +J 
 
 (T3 CO 
 
 Cj 
 
 00 
 
 
 
 
 
 - 
 
 to 
 
 
 CD 
 O 
 
 "d 
 
 
 
 
 
 
 (ro 
 
 s 
 
 CO 
 
 sz 
 
 CD C 
 
 •i- CD 
 
 <D 
 
 •^ 
 
 • 
 
 
 
 
 
 s 
 
 o 
 
 I— +-) 
 
 "O C 
 
 
 +J 
 
 
 
 
 
 
 u 
 
 Q. 10 
 
 CD CD 
 
 cJ 
 
 c 
 
 
 
 
 
 
 
 Q. (O 
 
 •r- +-> 
 
 O 
 
 o 
 
 
 l— 
 
 
 ^ // * 
 
 
 CD 
 
 CO 4- 
 
 i — CO 
 
 s; 
 
 O 
 
 
 00 
 
 
 » * 
 
 
 U 
 
 "^^ 
 
 CL fC 
 
 cJ 
 
 
 
 LU 
 
 
 I / 
 
 
 c 
 
 CO CD 
 
 Q.4- 
 
 d 
 
 
 
 1— 
 
 
 
 s 
 
 <1J 
 
 • i— 
 
 ra 
 
 Cr- 
 
 , 
 
 
 
 
 \ J / 
 
 rs 
 
 o ■*-> 
 
 CD 
 
 cj 
 
 
 
 >- 
 
 
 v^ ^s * 
 
 
 cr 
 
 +J 
 
 U3 i— 
 
 -<? 
 
 LU 
 
 
 _J 
 00 
 
 ^^ 
 
 ^^^^^^^^™~ b 
 
 
 CD 
 
 to 
 
 CD 
 
 r— -M 
 
 1 CD 
 
 
 _1 
 
 ^^ 
 
 
 
 V 
 
 CQ 
 
 (— 
 
 s: 
 
 
 
 
 
 
 CD 
 
 •r— 
 
 3 
 
 < 
 
 oo 
 
 LU 
 
 
 
 * 
 
 +J 
 
 CO i — 
 
 CO CO 
 
 1— 
 
 LU 
 
 OO 
 
 
 / i / 
 
 
 CO 
 
 +-> Q. 
 
 -»-> 
 
 v 
 
 
 I— 
 
 OO 
 
 
 1 
 
 V 
 
 0) 
 
 co E 
 
 CO "O 
 
 
 
 
 e£ 
 
 
 
 
 +-> 
 
 CD O 
 
 CD C 
 
 v 
 
 
 
 
 
 s* 00 ' A\ 
 
 
 
 I— u 
 
 1— fO 
 
 •< 
 
 
 
 O 
 
 
 y^ ^y^\\m * 
 
 
 < 
 
 
 
 tv 
 
 
 
 ^ 
 
 
 
 
 LU 
 
 
 
 i— 
 
 
 
 t— i 
 
 
 
 ^ 
 
 CH 
 
 • 
 
 • 
 
 s 
 
 
 
 ^ 
 
 
 
 V 
 
 < 
 
 fC 
 
 -Q 
 
 -< 
 
 
 
 LU 
 
 
 
 S 
 
 
 
 
 
 
 
 h- 
 
 
 
 "« 
 
 
 
 
 
 
 
 00 
 
 
 
 V 
 
 • • 
 
 
 
 
 
 
 <C 
 
 
 
 S 
 
 oo 
 
 
 
 
 
 
 Ll_ 
 
 
 
 . 
 
 LU 
 
 
 
 
 
 
 
 
 
 I— 
 
 
 
 
 
 
 . 
 
 
 
 o 
 
 
 
 
 
 
 , — 
 
 
 
 z 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 
 12 
 
NOTE: 
 
 TABLE 2. AMTRAK AND AREA FASTENER QUALIFICATION TEST SEQUENCES 
 
 (a) Amtrak Sequence 
 
 Toe load measurements are to be made before and after all tests in 
 the sequence below. Tests will be performed on both rail seats of 
 a tie/fastener assembly. Fastening Insert Test will be performed on 
 one rail seat prior to beginning the sequence. 
 
 Fastening assembly test 
 
 Rail clip load/deflection test 
 
 Tie pad load/deflection test 
 
 1. 
 
 2. 
 
 3. 
 
 4. 
 
 5. 
 
 6. 
 
 7. 
 
 8. 
 
 9. 
 10. 
 11. 
 12. 
 13. 
 14. 
 15. 
 
 Electrical impedance test 
 
 Fastening uplift test 
 
 Fastening longitudinal restraint test 
 
 Fastening repeated loads test 
 
 Fastening uplift test 
 
 Fastening longitudinal restraint test 
 
 Fastening push-pull test 
 
 Fastening uplift test 
 
 Fastening longitudinal restraint test 
 
 Electrical impedance test 
 
 Rail clip load/deflection test 
 
 Tie pad load/deflection test 
 
 (b) AREA Sequence 
 (1) Tests on a Completely Equipped Tie (2) Tests on a Tie Block 
 
 a. Fastening Insert Test 
 
 b. Fastening Uplift Test 
 
 a. Fastening Repeated Load Test 
 
 c. Electrical Resistance and 
 Impedance Test 
 
 b. Fastening Longitudinal 
 Restraint Test 
 
 c. Fastening Lateral 
 Restraint Test 
 
 13 
 
REVIEW OF EXISTING FASTENER TEST DATA 
 
 Available data from the preceding qualification tests and other 
 studies were reviewed to determine the areas where performance improvements 
 appear most needed. Data summary tables are compiled in Appendix A. The 
 major problems indicated by the data are discussed in the following sections 
 
 Rail Clip Force-Deflection Properties 
 
 SPRING RATE AT 
 1WSTAUEP PEFLECT10M 
 
 [Data In Table, A- 1 ) 
 
 VERTICAL CLIP QUIICTIOH (INCH 
 
 FIGURE 3. CLIP FORCE-DEFLECTION PROPERTIES 
 
 A large and uniform clip force (toe load) is required at installed 
 deflection to prevent longitudinal rail/tie creep. Therefore, the nominal 
 clip toe load should be sufficiently below the yield point of the clip to 
 assure that dynamic displacements and construction tolerances will not cause 
 yielding. For two clip designs, clip force-deflection data are available 
 from tests conducted before and after repeated loads tests (Table A-l). 
 Permanent deformation is indicated in both cases by a loss of toe load and a 
 reduction of the clip deflection produced by installation. The results show 
 a definite need to identify the force at which the clip yields and for a 
 criterion which limits the nominal clip toe load to a percentage of yield 
 load. 
 
 Tie Pad Stiffness 
 
 Figure 4 illustrates two current methods of measuring tie pad 
 stiffness. There is an uplift test requirement in both the AREA and Amtrak 
 series; the Amtrak tests develop a compressive curve up to 44 kips plus the 
 
 14 
 
precompression load but apply no spring rate requirement to it. The compres- 
 sive spring rate is being used as an index of compressive stiffness for an 
 investigation of the effect of pad stiffness on the attenuation of impact 
 strains in concrete ties [9]. Experience to date indicates that neither of 
 the two methods provides a good predictor of the other. Pad stiffness for 
 either method can vary substantially with load and rate of application. 
 
 Spiing Rate. 
 
 ~ noo • 
 
 (Data In Tables A- 2 and A- 3) 
 
 CompJieAA4,v& 
 Spring Roto. 
 
 PAP 
 
 10AP 
 
 (LB) 
 
 VERTICAL UPLIFT PEFLECTION (INCH 
 
 PAP COMPRESSION 
 SPRING RATE 
 
 PAP PEFLECTION (INCH 
 
 FIGURE 4. TWO METHODS FOR MEASURING TIE PAD STIFFNESS 
 
 Several conflicting issues are involved in the selection of an 
 optimum pad stiffness for a given track and traffic. All of the issues 
 involve the compressive load-deflection properties of the pad, either for 
 normal wheel passage or for abnormal wheel load conditions. The issues are: 
 
 a. Protection of concrete ties against cracking . Recent discover- 
 ies of rail seat cracks in concrete ties on the Northeast Corridor (NEC) and 
 in other revenue service test segments have made protection against crackinq 
 an issue. Rail seat cracking occurs where conditions of track support and 
 traffic combine to produce high levels of impact strain in the ties. The 
 
 15 
 
cracks can lead to tie failure. In a concurrent Battel le study, it has been 
 determined that a flexible pad can significantly attenuate impact-produced 
 tie strain [9]. Such pads must have a dynamically measured* compressive 
 spring rate between 500,000 and 1,000,000 lb/in compared with a value of 
 5 million lb/in for the EVA tie pad currently used on the NEC. 
 
 In general, the compressive spring rate of Figure 4 has provided a 
 good indicator of impact strain attenuation. However, where the pad is 
 shaped so that its load-deflection curve turns sharply upward in the region 
 above 10,000 pounds, the attenuation of large impact loads will be much less 
 than expected based on data from the compression test. Therefore, where 
 impact loads are a principal issue, spring rates measured at a high 
 pad load range (40,000 - 50,000 pounds) may be required. 
 
 b. Maintenance of ballast support conditions . The JNR [5] has 
 established criteria for tie pad stiffness based on the rate at which sur- 
 facing maintenance is required. Its recommended stiffness for the NEC 
 
 track is approximately one-fifth that of the present NEC pad. The JNR measures 
 stiffness between compressive loads of 1 and 10 metric tons (2200 - 22,000 
 pounds) . 
 
 c. Limiting Rail/Tie Deflection . Laboratory tests of pad stiff- 
 ness vs. rail/tie deflection show a sharp interdependence where tests are 
 conducted on a single fastener system. In this case the bending and torsion 
 of the rail cannot resist the deflection and distribute load to adjacent 
 fasteners. However, the results of measurements on the 5-degree curve 
 
 of concrete tie track at FAST [2] indicate that pad stiffness may have \/ery 
 little effect on the magnitudes of rail/tie deflections. Although data from 
 other locations are required to verify this finding, it is possible that 
 this issue is much less important than commonly believed. 
 
 d. Pad Durabil ity . In general, the hard pad materials (polyethylene, 
 polyurethane, EVA, hard Neoprene) are more resistant to permanent compression, 
 abrasion, and tearing than are most softer materials (soft Neoprene and 
 rubber). The grooving and shaping of pads, which is often required to produce 
 spring rates below 1,000,000 lb/in, may also contribute to pad deterioration, 
 especially where rail rollover causes loading of the pad by the edge of the 
 rail. A grooved pad has less area to resist this concentrated load. 
 
 Fastener Longitudinal Restraint 
 
 Specifications of longitudinal restraint tests for concrete tie 
 fasteners require the fastener system to sustain 2400 pounds without slip. 
 Data in Table A-4 show that some fastener systems cannot consistently meet 
 this requirement, particularly after being subjected to the repeated loads 
 and push-pull tests of the Amtrak series. In addition, there are indications 
 
 * Slope of the load-deflection curve between 4,000 and 20,000 pounds for 
 loading applied at 9-10 cycles per second. 
 
 16 
 
from experience on the South African Railway (SAR) that even higher levels of 
 restraint are required to prevent tie skewing and rail creep under severe 
 conditions in service [10]. The SAR cites a significant improvement in 
 fastener performance when the average toe load was increased from 1800-1900 
 pounds to 2600 pounds. The longitudinal restraint produced by a pair of 
 clips usually approximates the toe load of a single clip. Figure 5 is a 
 schematic of the method for measuring longitudinal restraint. 
 
 {Data -in Table. A- 4 
 
 LONGITUDINAL 
 LOAD 
 
 DISPLACEMENT 
 TRANSDUCERS 
 
 FIGURE 5. MEASUREMENT OF LONGITUDINAL RESTRAINT 
 
 Fastener Lateral/Rollover Restraint 
 
 LATERAL 
 RESTRAINT TEST 
 
 VERTICAL APPLIED LOAD 
 (VARIABLE) 
 
 20.5 Kips 
 
 ROLLOVER RESTRAINT 
 TEST 
 
 [data -in Tables A- 5 and k-6\ 
 
 FIGURE 6. MEASUREMENT OF FASTENER LATERAL AND ROLLOVER RESTRAINT 
 
 The final test of the AREA series (No. 6 of Table 1) applies loads 
 (Figure 6) to a rail segment which is mounted on a tie block oriented at an L/V 
 angle of 30 degrees. Lateral displacement of the rail base is limited to 1/8 
 inch at a load of 41 kips, while the rollover displacement (difference between 
 rail head and rail base lateral displacements) is limited to 1/4 inch at 
 20.5 kips. The data of Tables A-5 and A-6 show that: (1) the lateral dis- 
 placement of the rail base was easily constrained by all systems tested, and 
 (2) the rollover displacement requirements were also met by all systems 
 tested, but rollover results were highly dependent on pad stiffness and on 
 the geometry of the rail cross-section. However, the field measurements at 
 FAST (discussed later) show that there is not necessarily a direct dependence 
 between pad stiffness and rail/tie displacement in track. 
 
 17 
 
This test would constitute a determining factor in the qualifica- 
 tion of a fastener system only if a clip were to break under the high roll- 
 over displacements imposed during the qualification for lateral rail base 
 displacement. This test requires loads up to 41 kips, and Table A-5 shows 
 that it caused rollover displacements up to 0.4 inches. Most clips will 
 yield under much lower displacements [2]. 
 
 Lateral and Longitudinal 
 Restraint of Wood Tie Fasteners 
 
 Table 3 presents results of lateral and longitudinal restraint 
 tests conducted on a 4-tie panel of wood ties. The fastener systems range 
 from the standard 4 cut spikes per plate to several advanced configurations 
 which incorporate elastic or rigid clips and screw or lock spikes. One con- 
 figuration includes rail anchors on two of the four ties in the panel. 
 Longitudinal restraint loads, developed before the rail slipped by one inch, 
 range from the expected near-zero levels for cut spikes without anchors to 
 6310 pounds per rail seat for a combination of rigid clips and screw spikes. 
 The configuration with rail anchors produced only 3735 pounds per rail seat. 
 It is significant that 6 of the 8 nonconventional systems without rail anchors 
 produced restraint loads higher than this level, ranging from 7 to 68 percent. 
 However, these results do not reflect the demonstrated loss of restraint 
 loads by clip-type fasteners when installed in track. 
 
 To measure lateral restraint, the test panel was subjected to rail- 
 to-rail lateral load until the gauge was widened by 1 inch. An exception was 
 made in the test of the K-fastener, where the test was suspended at a total 
 load of 31,500 pounds with a lateral deflection of 0.58 inches. The deflection 
 was caused by the tie bending rather than by fastener failure. The maximum 
 load was more than three times the load which failed Configuration 1, the 
 standard arrangement of four cut spikes per plate. In contrast to the per- 
 formance of the K-fastener, five of the nonconventional configurations 
 produced inceases in resistance over the standard fastener ranging from 41 to 
 64 percent. These values can be compared with a 27 percent addition produced 
 by the addition of a single cut spike. 
 
 Figure 7 shows the results of tests conducted in Europe [12] where 
 it was demonstrated that the application of a single static load may be a 
 poor measure of longitudinal restraint under vibration simulating train action. 
 Tests involved a common elastic clip fastener with a hard masonite pad placed 
 over a steel tie plate on a wood tie. A vibrator was attached to the rail to 
 simulate rail vibration in parallel with the statically applied vertical and 
 lateral wheel loads. Results with and without vibration are shown in Fig- 
 ure 7. The vibration had the effect of reducing the mean longitudinal load 
 at the initiation of slip from about 1600 kg to about 800 kg (3500 to 
 1750 pounds). If this effect should be consistently produced in the labora- 
 tory, the addition of vibration to the longitudinal restraint test should 
 be seriously considered. 
 
 18 
 
Si 
 
 -i S 
 
 — LU O O 
 2 W Q. 
 
 01 «t CO 
 
 h- 
 oo 
 
 00 
 
 <: 
 q; 
 
 h- 
 oo 
 
 LU 
 
 cc: 
 
 00 
 
 J— 
 
 00 
 
 o ■= 
 
 IB 3 
 
 c_3 
 
 
 
 
 
 41 
 
 a. 
 
 
 -o 
 
 
 
 
 
 
 co co 
 
 <_> 
 
 
 01 
 
 0) r— 
 
 UJ 
 
 
 Ol o 
 
 S 
 
 
 CO 
 
 
 
 b 
 
 - n 
 
 
 ' $ 
 
 i- -* 
 
 
 «* u 
 
 Ol -i- 
 
 
 Q. C 
 
 5 00 
 
 
 •r- < 
 
 
 
 4-* 
 
 
 cj c 
 
 CO jt 
 
 
 
 ITJ 0> 
 
 
 c 
 
 u. J- 
 
 
 O co 
 
 u 
 
 
 ■r- Ol 
 
 :*: co 
 
 
 l/l -^ 
 CO •>- 
 
 11 
 
 
 Ol Q. 
 i. CO 
 
 1- 
 
 CD CD 
 
 
 O 3 
 
 
 O <_> 
 
 — * 
 
 
 cm 1*1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Ifl 
 
 Ol 
 
 
 
 
 
 CO 
 TJ 0) 
 
 
 
 
 
 
 
 
 
 
 
 
 
 J£ 
 
 
 
 
 
 
 
 
 CL 
 
 
 a. 
 
 
 
 
 CL. -r- 
 
 
 
 
 
 
 
 
 
 
 CL 
 
 
 D- 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 
 : CO 
 
 
 
 
 CJ 01 
 
 
 CJ 
 
 
 CO 
 
 
 
 
 co 
 
 
 
 
 
 
 
 
 
 
 s 
 
 
 >* s 
 
 
 
 
 
 
 e 
 
 
 B 
 
 
 01 
 
 
 >— Ol 
 
 
 
 
 o cc 
 
 
 O co 
 
 
 O 1/1 
 
 
 E. 
 
 
 , C 
 
 
 
 
 
 
 •r- Ol 
 
 
 ■,- O) 
 
 
 u 
 
 
 1 o 
 
 
 
 
 
 
 
 
 wi -* 
 
 
 l/l CO 
 
 
 CO CO 
 
 
 
 
 l/l c 
 
 
 l/l -»- 
 
 
 i/i •»- 
 
 
 a. 
 
 
 a. 
 
 
 
 
 o> 1- 
 
 
 O) a 
 
 
 O) Q. 
 
 
 
 
 
 
 
 
 
 
 i- CO 
 
 
 C CO 
 
 
 
 
 
 
 
 
 e f 
 
 
 a. 
 
 
 CL 
 
 E 4-> 
 
 
 CJ £ 
 
 
 CJ E 
 
 
 ■o S- 
 
 
 O ITJ 
 
 
 O 3 
 
 
 O 3 
 
 
 LU 0) 
 
 
 UJ Of 
 
 
 
 
 
 
 CJ CJ 
 
 
 <_> CJ 
 
 
 O CD 
 
 
 O CD 
 
 
 a. =c 
 
 
 CM CM 
 
 
 CM CM 
 
 
 <— CM 
 
 
 CM M- 
 
 CM *3" 
 
 
 cm co 
 
 
 l/l (/I 
 Ol o> 
 
 O 
 
 c: -Q 
 
 10 <u 
 
 oi i- 
 i— +j 
 
 Q 
 
 Z OO 
 
 <t LU 
 
 I— I 
 
 C3H- 
 
 •— i Q 
 2:0 
 LU O 
 
 OS 
 
 O 
 LU 
 CD O0 
 
 ■CC LU 
 U_0_ 
 
 o 
 
 LU 
 00 1—1 
 
 I— I— 
 
 __1 I 
 
 OO 
 
 LUZ 
 
 CO 
 
 CO 
 
 00 
 
 I— 
 00 
 
 OL 
 I— 
 O0 
 LU 
 
 en 
 
 1— 1 
 
 IT) 
 I— 
 1— 1 
 CD 
 
 o 
 
 
 
 CJ 
 
 
 
 
 
 
 
 
 
 
 
 00 
 
 Q. Z 
 
 
 
 >-■ =1 
 
 
 
 _l O 
 
 
 td 
 
 co a. 
 
 
 •VJ 
 
 (— 1 
 
 
 r^ 
 
 1— 
 
 
 C> 
 
 UJ <£ 
 
 
 
 CJ UJ 
 
 
 
 CC CO 
 
 
 
 O 
 
 
 
 Cl 1 
 
 
 
 
 
 
 -J < 
 
 
 
 ■a: a: 
 
 
 
 z 
 
 
 
 
 •— ce 
 
 
 
 O UJ 
 
 ro 
 
 
 =3 O. 
 
 CD 
 
 
 a; 
 
 •1— 
 
 
 
 \— c 
 
 
 C\J 
 
 M- 
 
 1 
 
 O C 
 
 
 
 
 a> c 
 
 a 
 
 1 — 
 
 rr 
 
 1 
 
 ro 
 
 +-> 
 
 +J 
 
 CO 
 
 CO 
 
 ro 
 
 •1 — 
 
 1 — 
 
 CO 
 
 1 — 
 
 ai 
 
 ro 
 
 CC CO 
 
 s 
 
 * 
 
 
 
 
 
 
 
 L. 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 i- 
 
 
 
 
 
 O 
 
 
 CO 
 
 
 
 
 
 
 U 
 
 
 1/1 
 
 
 
 
 
 
 
 1 01 J= 
 
 01 -C 
 
 
 1/1 01 
 
 
 
 
 
 
 O 1/1 
 
 
 01 
 
 
 (^ 
 
 
 t. 
 
 
 
 - <-> 
 
 c 
 
 
 _ — : 
 
 
 
 
 M 
 
 
 CO JC Ol 
 
 
 C -^ 
 
 
 T3 Ol 
 
 
 O CO 
 
 
 
 ' -i- C 
 
 •i- c 
 
 
 Ol ■•- 
 
 
 CL 
 
 
 Cl 
 
 
 CL O f— 
 
 
 4) -r- 
 
 
 ID H 
 
 
 JC Ol C 
 
 
 
 _l < 
 
 —1 <t 
 
 
 c a. 
 01 CO 
 
 
 #— 1/1 
 
 
 2^ 
 
 
 •>- C O 
 ^ < 31 
 
 
 C D 
 01 CO 
 
 
 . CL 
 
 
 c 
 c 1- x: 
 
 * IT) 
 
 
 
 
 
 
 ' 1 1 
 
 
 
 
 
 
 
 
 1 1 
 
 
 +-> 
 
 
 cj 01 
 
 
 CJ 
 
 
 CJ 
 
 
 «J 
 
 
 — 01 
 „ c 
 
 
 <C — 1 <— 
 
 C Ol 
 
 
 
 
 
 
 
 
 
 co x 
 
 
 _* 
 
 
 
 
 1 
 
 
 co S 
 
 
 
 1 1 c * 
 
 •r- J)*; 
 
 
 2 
 
 
 
 
 1 1/) CO 
 
 t/1 t/1 
 
 
 03 0) 
 
 
 Sa 
 
 
 c 
 
 
 c 
 
 
 « Ol 
 
 
 
 ' ' <: c 
 
 S_ •!- 
 
 
 Ol 
 
 
 
 c 
 
 1 CU CD 
 
 01 01 
 
 
 U. c 
 
 
 
 O co 
 
 
 O CO 
 
 
 U. c 
 
 
 
 \A t/> IC 
 
 CL CL 
 
 
 l/l c 
 
 
 
 (/I 
 
 -* -^ 
 
 ^L -^ 
 
 
 
 
 
 *- CI 
 
 
 ■t- o> 
 
 
 — 01 
 
 
 Cc (J 
 
 
 
 01 Ol r— D_ 
 
 CO CO 
 
 
 CL CJ 
 
 
 
 •1- 01 
 
 
 
 
 i^ CO 
 
 
 (/I 
 
 
 L/l Ji 
 
 
 
 
 m -i- LT> 
 
 
 CO ,S! 
 
 Cc 1 " 
 
 
 -i«: Ji -r- 
 
 1_ 
 
 
 1- CO 
 
 
 
 (/> ^£ 
 
 1 CC CL 
 
 CL CL 
 
 
 
 
 to C 
 
 
 
 
 l/> -r- 
 
 
 
 
 
 •r- •!- 10 +J 
 
 r— C O 
 
 
 
 
 
 l/l •»- 
 
 CO CO 
 
 CO CO 
 
 
 c c 
 
 
 ai ■•- 
 
 
 Ol O. 
 
 
 01 CL 
 
 
 C CJ c 
 
 
 •■- c 
 
 
 CL Q- or U1 
 
 O ••- -C 
 
 
 CJ c 
 
 to 
 
 
 CD CL 
 
 
 
 
 ITS (O 
 
 
 c 0. 
 
 
 t. CO 
 
 
 1- CO 
 
 
 m <a 
 
 
 »— (O 
 
 
 co co a 
 
 C Q. U 
 
 
 to 
 
 01 
 
 
 I- CO 
 
 1 4-> +-> 
 
 4-> 4-» 
 
 
 EE 
 
 
 CL C 
 
 
 i*j 
 
 
 Cc 
 
 
 ElJE 
 
 
 CJ E 
 
 
 C h- 
 
 T3 l_ C 
 
 
 ■ E 
 
 a 
 
 
 Q. Cc 
 
 3 3 
 
 3 3 
 
 
 
 E f 
 
 
 
 E 4-> 
 
 
 
 
 
 *J 4-> n- 
 
 c -^ <: 
 
 
 cu c 
 
 
 
 E -^ 4-1 
 
 I CJ CJ 
 
 CJ O 
 
 
 Ol 01 
 
 
 O « 
 
 
 O 3 
 
 
 O 3 
 
 
 Ol ■ Ol 
 
 
 UJ 01 
 
 
 3 3 10 c 
 
 fO *o 
 
 
 • 01 
 
 a 
 
 
 O 1— 3 
 
 
 
 
 CD CD 
 CM «3- 
 
 
 CJ I 
 CM CM 
 
 
 CJCJ 
 CM CM 
 
 
 CJ CJ 
 CM ro 
 
 
 CD O CD 
 
 f— f— «T 
 
 
 Q CD 
 
 cm *r 
 
 
 CJ CJ Cc •.- 
 CM CM CM 
 
 a- dz 1 
 
 CM CO 
 
 
 Q CD 
 
 CM <J" 
 
 CO 
 
 
 CJ CJ CJ 
 . — CM 
 
 CM CO 
 
 CM CM 
 
 
 II 1 
 
 
 
 1 « 
 
 19 
 

 
 1 
 
 
 
 l— 
 
 \ 
 
 \l 
 
 
 
 t— i 
 
 3 
 
 \ 
 
 \\ 
 
 
 
 Q 
 
 \ 
 
 \\ 
 
 
 
 o 
 
 
 
 
 
 _l 
 
 1 
 
 l\\ 
 
 
 
 -J 
 
 
 
 
 
 < 
 
 
 *v 
 
 
 
 1—4 
 
 Q 
 
 ZD 
 
 
 Aj\ 
 
 
 ■"— ■ 
 
 t~ Z 
 
 
 M* 
 
 
 
 ^ o 
 
 
 
 
 <-D _ 
 
 in 
 
 . 1 
 
 
 
 
 00 
 
 n* 
 
 
 
 — ' Ci 
 
 uu 
 
 '. \ I 
 
 
 
 
 H 
 
 m\ 
 
 
 ___ 
 
 
 ro 
 
 
 D. 
 
 
 
 O 
 
 _J 
 
 
 < d; 
 
 z 
 
 1 1 
 
 
 >- 
 
 LU 
 
 2 \\ a 
 
 o 
 
 
 -j 
 < 
 o 
 i— i 
 
 i— 
 
 i— 
 
 y 
 
 
 1- 
 
 LU 
 CO 
 
 O 
 
 h 
 
 
 . , 
 
 1 
 
 1 «l 
 
 1 t 
 
 
 o 
 o 
 
 IT) 
 
 o 
 o 
 o 
 
 CM 
 
 o 
 o 
 
 U"> 
 
 o 
 o 
 o 
 
 o 
 o 
 
 LO 
 
 (6>|) 3DcJ0J d33U3 
 
 
 
 Q 
 
 
 O 
 
 ro t 
 
 O 
 
 
 3 
 
 LU 
 
 
 s: 
 
 z 
 
 LU 
 
 o 
 
 C_> 
 
 
 <t 
 
 C£ 
 
 -J 
 
 ►— LU 
 
 Q. 
 
 ID Z 
 
 CO 
 
 O LU 
 
 CM — > 
 
 n: i— 
 
 a 
 
 I— u-i 
 
 
 HH< 
 
 D_ 
 
 3U_ 
 
 LU 
 
 
 LU 
 
 Q D_ 
 
 C£. 
 
 z: ■—. 
 
 O 
 
 <t _i 
 
 
 <_> 
 
 
 31 
 
 i— 
 
 1— O 
 
 
 H^ 1-^ 
 
 
 3 1— 
 
 
 CO 
 
 
 LU <£ 
 
 
 O _l 
 
 
 Z LU 
 
 
 < 
 
 
 1— u_ 
 
 O 
 
 OO O 
 
 1— ( 
 
 
 CO CO 
 
 
 LU 1— 
 
 
 Q£ CO 
 
 
 LU 
 
 
 _J h- 
 
 
 < 
 
 
 z z 
 
 
 t-.cC 
 
 
 Q LU 
 
 
 ■=> Q. 
 
 
 1— O 
 
 
 i— i OL 
 
 
 CD ID 
 
 IT) 
 
 Z LU 
 
 
 O^— 
 
 
 _1 
 
 
 z 
 
 
 u_ o 
 
 
 O i-" 
 
 
 1— 
 
 
 ZcC 
 
 
 OQ£ 
 
 ^d- 
 
 CO QQ 
 
 i— i •— i 
 
 
 Cd > 
 
 
 < 
 
 
 D 1 
 
 *^^ 
 
 2: •— 
 
 P 
 
 o«=t 
 
 fc 
 
 <_> Q£ 
 
 —I 
 
 D. 
 
 CO 
 
 C\J i— i 
 
 o 
 
 Q_ 
 LU 
 LU 
 
 q: 
 o 
 
 
 (6l) 30d0d d33cJD 
 20 
 
FASTENER PERFORMANCE EXPERIMENTS 
 
 In an effort to develop fastener qualification requirements which 
 would provide correlation between observed laboratory and service performance, 
 Battel le conducted an experimental program in three phases: 
 
 a. Rail/tie deflections and fastener clip strains were 
 measured at FAST to define representative, severe 
 fastener loading environments. This effort is 
 described in Reference 1 . 
 
 b. Static measurements of fastener load-deflection 
 characteristics were conducted in the laboratory 
 to: 
 
 (1) determine the load combinations required 
 to reproduce the maximum strains and 
 deflections measured at FAST 
 
 (2) compare the track-measured strains and 
 deflections with those produced during 
 conduct of current qualification tests 
 
 (3) examine the current methods of determining 
 basic clip and pad characteristics. 
 
 c. Fatigue tests were conducted on one concrete tie 
 fastener system and one wood tie fastener system, 
 both of which are among the types used at FAST. 
 This was done to compare the results of the tests 
 with performance observed in service. 
 
 The following sections discuss the results of these experiments as they 
 relate to the development of fastener performance requirements. 
 
 Track Measurements at FAST 
 
 To define severe fastener loading environments which could be 
 simulated in the laboratory, rail/tie deflections and fastener clip strains 
 were measured on 5-degree curves of concrete and wood tie track at FAST. 
 Two concrete tie fasteners and two wood tie fasteners were examined, 
 Figures 8(a) and 8(b). Measurements were made at three sites in each 
 fastener subsection. Each fastener used an elastic clip to constrain the 
 ra i 1 . 
 
 Deflection measurements, Figure 8(c), consisted of three vertical 
 
 rail/tie deflections at the rail base, one lateral at the rail base, one 
 lateral at the rail head, and one longitudinal. In one subsection of each 
 
 type of track where a common clip (Type A) was installed, fastener clip 
 
 strains were also measured. The instrumentation of clips to measure strain 
 is described in Appendix B. 
 
 21 
 
Type A Clip 
 
 Elastic Clip/Spike 
 
 Type A Clip 
 
 (a) Concrete Tie Fasteners 
 
 (b) Wood Tie Fasteners 
 
 / ! 
 
 lf>_< 
 
 I' j 
 
 Attaakrmnt to T-le. 
 
 / i 
 / / 
 
 B- 
 
 
 (c) Rail/Tie Deflection Measurements 
 
 FIGURE 8. FIELD MEASUREMENTS AT FAST: FASTENER TYPES 
 AND DISPLACEMENT COMPONENTS 
 
 22 
 
Train loads were produced by a consist of two locomotives and 
 twenty loaded 100-ton hopper cars. Train runs were made in both clockwise 
 and counterclockwise directions for measurements in the concrete tie section, 
 where a 2-percent grade contributed to greater deflections for counterclock- 
 wise (upgrade) travel. Only clockwise runs were made over the wood tie 
 section where the grade and its effect were much lower. 
 
 The major results of the track measurements were: 
 
 a. Peak lateral deflections between the rail head and tie approached 
 0.100 inches on both types of track. 
 
 b. Vertical rail/tie deflections of rail clips in the concrete 
 tie section approached 0.040 inches in both gauge side uplift and field side 
 compression. 
 
 c. Measurements in the concrete tie section were made in a sub- 
 section containing a very rigid polyethylene pad (7.5 million lb/in spring 
 rate) and in a subsection containing a relatively flexible pad (1.3 million 
 lb/in spring rate). Peak deflections were about 30 percent higher for the 
 hard pad than for the soft pad. 
 
 d. Vertical rail/tie deflections in the wood tie section reached 
 0.100 inches in field side compression. However, where the fastener clip 
 was attached to the tie plate (Type A clip), most of the vertical deflection 
 took place through tie plate bending rather than through clip deflection. 
 
 The peak vertical rail /tie deflections at the fastener clips were 
 calculated from the three vertical deflection measurements illustrated in 
 Figure 8(c). The data points in Figure 9 summarize the peak vertical deflec- 
 tions and the simultaneously occurring peak clip strains for the Type A 
 clip. The next section compares the measured data with the strain-deflection 
 relationships measured in the laboratory and also shown in Figure 9. 
 
 Clip Force-Deflection-Strain Characteristics 
 
 Comparison With Track Measurements 
 
 Track measurements of clip strain vs. clip deflection were compared 
 with similar data produced by two laboratory methods: 
 
 Method 1 - by vertical loading of an individual clip, as 
 described in Appendix C 
 
 Method 2 - by loading through a rail segment to simulate 
 the lateral and vertical components experienced 
 in track. The lateral restraint fixture used for 
 this purpose is shown later in Figure 19. 
 
 23 
 
(1 Volt = 166 microinches/inch) 
 
 -5 
 -6 
 -7 
 
 Ctip StucU.n-Ve.file.cXAjjn ToAt 
 
 — [hleXhod 1) — 
 
 TRACK MEASUREMENTS 
 O Gauge Clip 
 □ Field Clip 
 
 ■60 -50 -40 -30 -20 -10 10 20 30 40 50 60 70 
 
 CLIP DEFLECTION (MILS) 
 
 FIGURE 9. VERTICAL CLIP DEFLECTION VS. CLIP STRAIN FROM TRACK MEASUREMENTS 
 AND TWO LABORATORY TESTS 
 
 24 
 
Figure 9 shows schematics of the two loading arrangements and the three 
 strain-deflection relationships. The results show that: 
 
 a. The peak clip strain level was found in the track at 
 vertical deflections between 0.030 inches and 0.037 
 inches of vertical rail/tie deflection 
 
 b. The reproduction of this peak clip strain required over 
 0.060 inches of vertical clip deflection by Method 1 but 
 only 0.038 inches by Method 2. The clip strain-deflection 
 relationship obtained from Method 2 correlates with the 
 field results much better than does that of Method 1. 
 
 It can be concluded that the clip receives substantial 
 strain from lateral loading as well as from vertical 
 loading when installed in track. Thus, the vertical 
 rail/tie deflection of the clip will not provide a good 
 indicator of the level of clip strain. 
 
 The data in Figure 9 can be used to explain an apparent anomaly 
 between the field measurements of rail/tie deflection and the results of 
 fatigue tests conducted by the clip manufacturer [13]. It is known that 
 substantial numbers of these clips have fractured in service on the 5-degree 
 curve of the concrete tie section at FAST. Therefore, the clip strains 
 experienced in track must exceed the fatigue limit of the clip. 
 
 The manufacturer subjected the clip to fatigue tests with a loading 
 arrangement equivalent to that of Method 1. Tests were conducted by imposing 
 cyclic vertical deflections (measured relative to the nominal installed clip 
 position) at levels of 0.020 to 0.060 inches in 0.010-inch increments. With 
 a 2000-pound toe load and cyclic deflection of 0.050 inches, the clips did 
 not fail in tests up to 15 million cycles. With the same toe load and cyclic 
 deflection of 0.060 inches, the clips failed in less than one million cycles. 
 An increase in toe load to 2400 pounds caused failures within one million 
 cycles at 0.040 inches of cyclic deflection. 
 
 Clip toe loads measured by the manufacturer at FAST did not exceed 
 1580 pounds [14]. The vertical rail /tie deflections measured under this 
 program did not exceed 0.037 inches. However, the fatigue limit of many of 
 the clips in track was exceeded. The apparent anomaly between service per- 
 formance and the manufacturer's fatigue tests can be explained by observing 
 the differences in strain levels produced by the previously described lab- 
 oratory Methods 1 and 2. It is evident that the clip strain levels imposed 
 by combined vertical and lateral loads, either in track or simulated in the 
 laboratory (Method 2) do exceed the fatigue limit of some of the Type A 
 clips. This limit is reached at 3.0-3.5 volts clip strain on the scale of Fig- 
 ure 9, or 0.050-0.060 inches deflection by Method 1. 
 
 Method of Determining Clip Yield Load 
 
 A simple and repeatable method for determining the vertical yield 
 load of an individual clip was suggested by a manufacturer* and duplicated 
 
 *Portec, Inc. 
 
 25 
 
at Battel le. The method requires a fixture for the vertical loading of an 
 individual clip, such as the arrangement described in Appendix C, and a 
 vernier caliper or dial gauge capable of displacement measurements to 0.001 
 inches. The following procedure is used: 
 
 a. Place the clip on a flat surface and measure the 
 height of the clip toe or other characteristic 
 dimension. 
 
 b. Select a value of vertical force which is known 
 
 to be less than the yield load. Apply the vertical 
 load to this point and release it. Typical load- 
 deflection curves are shown in Figure 10. 
 
 c. Repeat step a. 
 
 d. Repeat step b with the load increased by 100-200 pounds. 
 
 This process is continued until the characteristic dimension begins 
 to change and several post-yield points are collected. Straight-line curve 
 fits of pre-yield and post-yield data will intersect at the yield load. 
 Typical data for 2 types of clips are illustrated in Figure 11. 
 
 The clip yield load should be compared with the nominal toe load 
 of the clip. A sufficient margin between toe load and yield load should be 
 maintained to assure that yield will not occur under the worst combinations 
 of displacements produced by train loads and construction/assembly tolerances. 
 
 Tie Pad Compression Tests 
 
 The fixture shown in Figure 12 has been used to perform both static 
 and dynamic compression load tests on a wide variety of pads which differ in 
 material, thickness and shape factor (grooving or molding to reduce stiffness). 
 Examples of pad stiffness are shown in Figure 13 through 15. Some general 
 trends from the tests were: 
 
 a. Compressive stiffness is highly dependent on the rate of loading 
 for some pads but almost independent of loading rate for others. Extremes are 
 shown in Figure 14 (dependent) and Figure 15 (independent). In general, the 
 hard pad materials (polyethylene, polyurethane, EVA) have stiffnesses which 
 are relatively independent of loading rate. Neoprene is relatively independent 
 except where severe shaping causes a sharp change in stiffness as the pad is 
 compressed. A loading rate of 10 cycles per second is recommended. 
 
 b. Load-deflection curves should never be recorded until at least 
 several complete load cycles have been applied, even when the load applica- 
 tion is quasistatic. Load cycles should vary from low load to the maximum 
 desired, rather than from zero load to maximum load. Substantial differences 
 will occur in any definition of the "zero" load-deflection point. 
 
 c. Shaping of pads to achieve flexibility (lower stiffness) can 
 
 be detrimental to the objective of tie impact load attenuation if the shaping 
 
 26 
 
3600 
 
 3200 
 
 2800 
 
 2400 
 
 Q 
 
 2 2000 
 
 O 
 
 o 
 
 
 1600 
 
 1200 
 
 800 
 
 400 
 
 FIGURE 10. 
 
 0.2 0.3 0.4 
 VERTICAL CLIP DEFLECTION (INCH) 
 
 TYPICAL CLIP FORCE-DEFLECTION CHARACTERISTICS 
 
 27 
 
 0.5 
 
 0.6 
 
1.585 
 
 c_> 
 
 cd 1.580 
 i— i 
 
 LU 
 
 LU 
 O 
 
 1.575 
 
 14 
 
 CLIP TOE HEIGHT 
 
 CLIP 
 LOAD 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 >^^~ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ■^"^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 •\ 
 
 r^ 
 
 
 
 
 
 
 
 
 — c 
 
 ■^ ^ 
 
 J ^ 
 
 -> ^ 
 
 
 
 J 
 
 
 
 
 
 
 
 . ._ 
 
 
 
 . 
 
 
 
 l 
 
 16 18 20 22 24 26 28 
 
 MAXIMUM CLIP LOAD FOR EACH TEST (HUNDREDS OF POUNDS) 
 
 5 0.090 
 
 
 uj en 
 
 CD LU 
 
 0.080 
 
 0.070 
 
 23 0.060 
 
 14 
 
 z^y 
 
 CLIP CENTER HEIGHT 
 
 16 18 20 22 24 26 
 MAXIMUM CLIP LOAD FOR EACH TEST (HUNDREDS OF POUNDS) 
 
 FIGURE 11. DETERMINATION OF CLIP YIELD POINT FOR TWO CLIPS 
 
 28 
 
FIGURE 12. LOADING ARRANGEMENT FOR TIE PAD COMPRESSION TESTS 
 
 29 
 
35 
 
 30 
 
 25 
 
 2.0 
 
 
 Q 
 O 
 
 
 
 o 
 
 15 
 
 10 
 
 1 c 
 
 YCLE/MIN O.i 
 
 i CYCLES/SEC 
 
 1 CYCLE/SEC 
 
 
 
 
 
 
 
 
 
 
 
 
 
 — t 
 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 1 
 
 CO 1 
 
 
 
 1 
 
 
 I 
 
 
 °-l 
 
 i 
 
 
 
 J 
 
 
 1 
 
 r 
 
 i 
 
 
 
 0.010" 
 
 
 
 
 
 
 
 
 COMPRESSIVE DEFLECTION 
 
 FIGURE 13. VERTICAL LOAD-DEFLECTION CHARACTERISTICS 
 FOR THE POLYETHYLENE PAD 
 
 30 
 
CO 
 
 Q 
 
 
 o 
 o 
 
 o 
 
 o 
 
 
 CO 
 
 
 i— i 
 
 
 C£ 
 
 
 LU 
 
 <"-^. 
 
 J— 
 
 co 
 
 o 
 
 LU 
 
 DC 
 
 2 
 
 (_) 
 
 «=r 
 
 z 
 
 re 
 
 •— 1 
 
 o 
 
 
 z 
 
 z 
 
 o 
 
 o 
 
 h- * 
 
 t— 1 
 
 1— 
 
 »— 
 
 o 
 
 o 
 
 LU 
 
 LU 
 
 _l 
 
 -J 
 
 u. 
 
 LU- 
 
 LU 
 
 LU 
 
 o 
 
 Q 
 
 1 
 
 
 Q 
 
 O 
 
 <x 
 
 < 
 
 o 
 
 D. 
 
 _J 
 
 _l 
 
 7" 
 
 <C 
 
 o 
 
 o 
 
 
 CJ 
 
 >- 
 o 
 
 Lu. 
 
 o 
 
 \— 
 
 
 (sdi>i) avtn avd ivoiib3A 
 
 31 
 
40 
 
 30 
 
 c 
 
 < 
 O 
 
 w 20 
 
 to 
 io 
 
 LJ 
 
 Q- 
 
 s: 
 o 
 o 
 
 10 
 
 COMPRESSIVE DEFLECTIW 
 
 FIGURE 15. EFFECT OF CYCLE RATE ON LOAD-DEFLECTION 
 
 CHARACTERISTICS OF GROOVED SYNTHETIC RUBBER PAD 
 
 32 
 
does not allow for the gradual transition of stiffness with increase in 
 compression. The pad shown in Figure 14 is made of moderate durometer mate- 
 rial and is grooved to achieve a radical shape factor. However, the pad 
 grooves "bottom" at less than 15,000 pounds. The effectiveness of this pad in 
 attenuating large impact loads is quite low in comparison to other pads with 
 approximately the same average stiffness [9]. This experimental pad shape has 
 been abandoned by the manufacturer. 
 
 Longitudinal Restraint Tests 
 
 Longitudinal restraint tests were conducted to determine whether a 
 dependence could be found between longitudinal restraint and either (1) tie 
 pad stiffness, or (2) the presence or lack of external insulators. It was 
 quickly discovered that the dependence of longitudinal restraint on any single 
 test factor was very difficult to isolate. Problems encountered in the 
 development of an acceptable test procedure are discussed as follows. 
 
 Early trials with the fixture of Figure 16(a) revealed that the 
 fastener clips could not be depended upon to provide consistent toe loads. 
 Successive tests with the same pad resulted in losses of longitudinal slip 
 load by up to 15 percent. The installation and removal of clips caused clip 
 deformation and wear of the fastener shoulders and insulators. Where 
 insulators were not used, the clip toe and rail base became polished. There- 
 fore, it was necessary to devise a method of applying controlled vertical 
 loads to the clip toe areas of the rail base. This was done with the fixture 
 shown in Figure 16(b). Controlled vertical loads were applied by a mechan- 
 ical test machine to a fixture with two bearing surfaces which simulated toe 
 loads. The vertical load was applied through rollers to prevent reaction 
 of longitudinal load by the vertical load fixture. Also, the vertical fix- 
 ture was constrained against longitudinal displacement by reaction of the 
 fixture against the fastener shoulders. Longitudinal load was applied and 
 measured by placing a hand-pumped hydraulic cylinder in line with the load 
 cell. 
 
 After vertical load control was established, it was found that test 
 repetitions with the same pad and insulator would not consistently provide 
 the same results. A sequence of tests yielded a general downward drift of 
 longitudinal restraint under identical test inputs. To maintain comparable 
 values of slip load, it was necessary to change the test specimens (pads and 
 insulators) after each measurement. This process was continued until three 
 values of slip load were obtained for each combination of pad, insulator 
 and vertical applied load. The mean of slip loads obtained with identical 
 test inputs was used to form comparisons. 
 
 Figure 17 presents the results of tests conducted on two pads which 
 represent extremes in pad stiffness and coefficient of friction among those 
 tested. The two pads also produced extremes in longitudinal slip load as a 
 function of vertical applied load. Tests were run with and without insulators 
 of the metal -plastic shim type. The tests with insulators yielded higher 
 loads for both pads, but the difference with and without insulators was much 
 greater for the rigid pad than for the flexible pad. It is possible that the 
 difference in longitudinal stiffness of the two pads causes the insulators to 
 interact differently. 
 
 33 
 
(a) Toe Load Provided by Clips 
 
 (b) Toe Load Provided by Test Machine 
 
 FIGURE 16. LOADING FIXTURES USED FOR LONGITUDINAL RESTRAINT TESTS 
 
 34 
 

 
 
 
 
 
 
 
 
 
 
 ( 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 GROOVED DURAFLEX PAD 
 
 WITH INSULATORS — x 
 
 i i X 
 
 /o / 
 
 / 
 
 
 
 
 
 
 W I THOU 
 
 T INSULATORS 
 
 •»— ^ ™ 
 
 y / 
 
 
 
 
 
 
 
 
 
 
 
 ^ / 
 
 
 
 
 
 
 
 
 
 
 
 
 / l 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 >y 
 
 / 
 
 
 
 
 
 
 
 
 
 &/ 
 
 r/ 
 / - 
 
 >\ 
 
 
 
 
 
 
 
 
 ■if A 
 
 ft 
 
 1 y 
 
 
 i!\ ,<■ 
 
 \ POLYETHYLENE f 
 
 >AD 
 
 
 
 
 < 
 
 &y 
 
 /* 
 
 > 
 
 *A 
 
 o x y* 
 
 vT. 
 
 WITH INSULATORS 
 ITHOUT INSULATORS 
 
 
 
 (£ 
 
 // 
 
 // 
 
 r 
 
 / 
 
 Y 
 
 / 
 
 
 
 
 
 
 
 
 Fa 
 
 y/ 
 
 t S 
 
 y 
 
 
 
 
 
 
 
 
 
 
 // 
 
 V 
 
 S 
 / 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 s 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 fs 
 
 s 
 
 
 
 
 
 
 
 
 
 
 
 
 1000 
 
 2000 3000 4000 
 VERTICAL LOAD (LB) 
 
 5000 
 
 6000 
 
 FIGURE 17. INFLUENCE OF PADS AND INSULATORS ON LONGITUDINAL 
 RESTRAINT OF CONCRETE TIE FASTENERS 
 35 
 
While the two pads vary widely in stiffness, they also differ in 
 the shape and texture of the bearing surfaces. The polyethylene pad has 
 solid and very smooth bearing surfaces. The Duraflex pad is made of a soft 
 polymer with a comparatively rough surface texture and is grooved to lower 
 stiffness. Typically, the Duraflex pad would permit about twice the longi- 
 tudinal rail -tie deflection before the onset of slip (0.010 inches vs. 0.005 
 inches for 4000-pound vertical load). This longitudinal flexibility could 
 be a significant factor in the creep resistance of installed ties. The FAST 
 measurements described earlier showed that longitudinal rail/tie deflections 
 under train loads did not exceed 0.010 inches. 
 
 Similar longitudinal restraint results are shown in Figure 18 for 
 the wood tie fastener which uses a Type A clip. The restraint at a given 
 vertical load falls between those of the concrete tie fasteners shown in the 
 previous figure. However, it should be noted that the onset of slip is 
 almost instantaneous for this case where the rail contacts a steel tie plate. 
 Most measured deflections before slip fell below 0.0002 inches. The system has 
 no flexibility to permit longitudinal rail/tie deflection without slip. 
 
 Lateral /Roll over Restraint Tests 
 
 Lateral/rollover tests were conducted with a range of L/V angles 
 from 20 to 30 degrees and with pads of varying stiffness. The primary pur- 
 pose of this effort was to determine the combinations of load and L/V angle 
 which could most closely simulate the maximum rail/tie deflections and clip 
 strains found in the FAST measurements. This also provided an opportunity 
 to evaluate the current qualification tests for lateral/rollover restraint and 
 repeated loads. 
 
 The fixture used to vary the L/V angle, apply vertical loads, and 
 measure rail/tie deflections is illustrated in Figure 19. Curves for rail 
 head lateral displacement and gauge clip uplift are shown in Figure 20 for 
 the rigid polyethylene pad and in Figure 21 for the flexible synthetic rubber 
 pad. These pads were installed in the FAST concrete tie subsections where 
 field measurements were made. 
 
 The data display strong influences of both L/V angle and pad stiff- 
 ness on rail head lateral displacement and gauge clip uplift. Since rail 
 base lateral displacements were relatively small, the rail head lateral 
 displacement provides a good indicator of rail rollover. The load range was 
 limited to avoid destroying the clips. However, the data indicate that the 
 flexible pad may not have limited the rail to the rollover restriction of 
 0.25 inches rollover displacement with the vertical load of 20.5 kips applied 
 at a 30-degree angle. It should be noted that this rubber pad is not among 
 
 36 
 
4000 
 
 5000 
 
 1000 2000 3000 4000 
 TOTAL VERTICAL LOAD (LB) 
 
 FIGURE 18. LONGITUDINAL RESTRAINT OF WOOD TIE FASTENER SYSTEM 
 
 37 
 
 6000 
 
r rcr 
 
 / 
 
 zzzzz 
 
 r 
 
 o. 
 
 o 
 
 FIGURE 19 LOADING AND MEASUREMENT SCHEMATIC FOR 
 LATERAL/ROLLOVER RESTRAINT TESTS 
 
 38 
 

 
 
 
 
 
 
 I 
 
 
 
 
 
 
 
 
 V 
 
 
 Q 
 
 
 
 
 
 
 
 
 Q. 
 
 
 
 
 
 
 (vr "" 
 
 
 ,_. 
 
 
 
 
 
 
 
 
 r^. 
 
 
 
 
 
 
 
 
 o 
 
 
 
 
 
 
 
 y _#^C^^^ 
 
 
 ■ — 
 
 
 
 
 
 LU 
 
 : 
 
 ^•^^^"v V^ 
 
 
 i — 
 
 
 
 
 
 _l 
 
 
 
 Q. 
 
 > 
 
 
 
 
 ■ZL 
 
 
 
 '^>-<^^^ x\_v" _ 
 
 z A 
 
 
 1 
 
 
 
 
 O 
 i— i 
 
 < T 
 
 D 
 
 
 
 LU 
 
 
 
 
 I— 
 
 >» \ ..^ 
 
 ^^ ' _ 
 
 
 2T 
 
 
 
 
 O 
 
 
 
 LU 
 
 
 
 
 LU 
 
 _i ^^ 
 
 
 
 _J 
 
 
 
 
 _l 
 
 
 
 
 >- 
 
 
 
 
 U_ 
 
 
 __.-, 
 
 
 n: 
 
 
 
 
 LU 
 
 
 
 
 l— 
 
 
 
 
 Q 
 
 
 
 
 LU 
 
 
 
 
 
 
 
 
 
 >- 
 
 
 
 
 _) 
 
 
 
 _ 
 
 
 _l 
 
 
 
 
 < 
 
 
 
 
 
 O 
 
 
 
 
 C_ 
 
 
 
 
 
 Q_ 
 
 
 
 
 LU 
 I— 
 
 < 
 
 Q 
 
 
 
 - 
 
 
 
 
 
 
 
 <c 
 
 
 
 
 
 
 
 
 z: 
 
 
 LU 
 
 
 
 
 
 
 
 
 o 
 
 
 DC 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 I 
 
 
 
 
 
 
 
 
 «t 
 
 
 t— I 
 
 
 
 
 
 
 
 
 C_) 
 
 
 < 
 
 
 
 
 
 
 
 
 I — I 
 
 
 c_ 
 
 
 
 __, 
 
 
 
 
 
 _l 
 
 
 
 
 
 
 
 
 
 
 Q_ 
 
 
 
 
 
 
 
 
 
 
 Q- 
 
 
 
 
 
 
 
 
 
 
 <c 
 
 
 
 
 
 - 
 
 
 
 
 
 Q 
 
 
 
 
 
 
 
 
 
 
 < 
 
 
 
 
 
 
 
 
 
 
 O 
 
 
 
 
 
 
 
 
 
 
 _J 
 
 
 
 
 
 
 
 
 
 
 o 
 
 
 
 
 
 
 
 
 
 
 LU 
 
 
 
 
 
 
 
 
 
 
 Q 
 
 
 
 
 
 
 
 
 
 
 I 
 
 o 
 
 
 
 
 
 
 
 
 
 ^ 
 
 en 
 
 
 
 
 
 
 
 
 
 o 
 
 
 
 
 
 
 
 
 
 1— 1 
 
 
 
 
 
 
 
 
 z 
 
 1— 
 
 
 
 
 
 
 
 
 o 
 
 o 
 
 
 
 
 
 
 
 
 t— 1 
 
 LU 
 
 
 
 
 
 
 
 
 2t 
 
 1— 
 
 _l 
 
 LU\ 
 
 
 
 
 
 
 
 o 
 
 o 
 
 u_ 
 
 CJ3 \ 
 
 
 
 
 _ 
 
 
 
 
 LU 
 
 LU 
 
 r_> _X 
 
 
 
 
 
 
 
 1— 
 
 _1 
 
 Q 
 
 <=C \ 
 
 
 
 
 
 
 
 <c 
 
 U_ 
 
 
 cs 2: 
 
 
 
 
 
 
 
 C_) 
 
 LU 
 
 1 
 
 
 
 
 
 
 «_ 
 
 
 
 1— 1 
 
 Q 
 
 < 
 
 _i i— i 
 
 
 
 
 
 
 
 _l 
 
 
 CtL 
 
 <=C h- 
 
 
 
 
 
 
 
 D_ 
 
 LU 
 
 LU 
 
 c_> 
 
 
 
 
 
 
 
 Q_ 
 
 CJ3 
 
 1— 
 
 i— I LU 
 
 
 
 
 __, 
 
 
 
 < 
 
 <: 
 
 
 I— —I 
 
 cm u_ 
 
 
 
 
 
 
 
 Q 
 
 _-f 
 
 o 
 
 1—1 
 
 LU LU 
 
 
 
 
 
 
 
 o 
 
 "_i" 
 
 <: 
 
 
 
 
 
 
 
 
 _l 
 
 o 
 
 1 — 1 
 
 LU 
 
 IE 
 
 
 
 
 
 
 
 
 o 
 
 h- 
 
 _l 
 
 
 
 
 
 
 
 
 LU 
 
 ai 
 
 i — i 
 
 
 
 
 
 
 
 
 1 
 
 LU 
 
 > 
 
 c_ 
 
 
 
 
 
 
 
 
 o 
 
 
 
 
 
 
 
 
 
 
 CM 
 
 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^^ ___ 
 
 
 
 ' 
 
 
 J _____ 
 
 - — -. 
 
 — — — 
 
 _____ 
 
 
 -1 — 7^>T-^ 
 
 1 
 
 
 1 
 
 i— 1 — 
 
 — I 
 
 I I 
 
 I I 
 
 | 
 
 
 LO 
 Cvl 
 
 O 
 
 cm 
 
 LD 
 
 OO 
 
 C_ 
 CD 
 
 O 
 CO 
 
 < 
 
 o 
 
 C\J 
 
 00 
 
 I— 
 
 c_ 
 I— 
 00 
 
 LU 
 C_ 
 
 c_ 
 
 o 
 
 c_> _r 
 
 <C lu 
 
 _i 5: 
 
 Q_ LU 
 
 o w 
 
 .— .-. < 
 
 • Q _l 
 
 Q. Q 
 
 _J OO «C 
 
 <| I— 1 Q. 
 
 c— q 
 
 OO LU 
 
 > _l 
 >- 
 Q DC 
 < I— 
 O LU 
 
 o o 
 
 LU Q. 
 
 C_> 
 
 lo 
 o 
 
 _i Q 
 a. c_ 
 
 Q- <C 
 
 o 
 
 CM 
 
 
 o 
 
 o 
 00 
 
 o 
 
 CM 
 
 (sdi>i) avoi a3nddv 
 39 
 

 
 1 
 
 
 — 
 
 
 
 ^ 
 
 
 - 
 
 1 
 
 
 rcC 
 
 
 
 =**= 
 
 
 Vw- 
 
 
 
 Q 
 
 
 \>^ 
 
 
 
 < 
 
 
 
 y^\t~ 
 
 
 
 LU 
 UJ 
 
 LU 
 
 
 br% 
 
 
 — 
 
 -8! 
 
 1 "': 
 
 O 
 
 ^^ 
 
 ^ 
 
 i\J V 
 
 
 
 LU 
 
 
 v^ 
 
 
 
 21 
 
 1 
 
 s^ 
 
 1* 
 
 \ »-■* 
 
 
 
 
 
 
 \ V", 
 
 — 
 
 
 
 
 
 \ <J> 
 
 
 
 
 
 
 
 \ v* 
 
 
 
 
 
 
 r^ 
 
 \ -i 
 
 
 
 
 
 
 O 
 
 t— t 
 
 
 - 
 
 
 
 
 
 h- 
 
 \ ^ 
 
 
 
 
 2: 
 
 
 
 
 \ ^ 
 
 _ 
 
 
 
 i— < 
 
 
 »— * 
 
 \ c£ 
 
 
 
 
 1— 
 
 
 1 
 
 Q 
 
 \ ^ 
 
 
 
 
 CJ 
 
 
 \ v "- 
 
 
 
 
 "LU " 
 
 
 
 \ «3- 
 
 
 
 
 _l 
 
 
 < 
 
 \ -J 
 
 
 
 
 u_ 
 
 LU 
 
 
 a 
 
 \ C3 
 
 
 
 
 O 
 
 
 ■sC 
 O 
 
 \ <*- 
 
 \ vjJ 
 
 
 
 
 _J 
 
 
 1 
 
 \ ^ 
 
 
 
 
 <C 
 
 
 
 
 
 
 
 d2 
 
 
 • 
 
 \ -i 
 
 ~ 
 
 
 
 LU 
 
 
 CD 
 
 \ ri 
 
 
 
 
 | — 
 
 
 LU 
 
 \ <*- 
 
 
 
 
 <c 
 
 
 O 
 
 \ ^ 
 
 
 
 
 
 
 
 1 
 
 O 
 
 co 
 
 S \ 
 
 
 
 
 < 
 
 
 
 *"* \ 
 
 
 
 
 LU 
 
 
 V-, 
 
 
 
 
 :r 
 
 
 ^ \ 
 
 
 "■»» 
 
 
 _i 
 
 
 . -^ \ 
 
 
 
 ^^ 
 
 >— 1 
 
 < 
 
 
 \ £ \ 
 \ ^ \ 
 
 - 
 
 2^ 
 
 
 
 ">. 
 
 
 \ ^ 
 
 V 
 
 O 
 
 
 
 
 \ 
 
 
 \ 
 
 i— 1 
 1— 
 
 
 z 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 *««* 
 
 \ ** 
 
 \ 
 
 1— 1 
 
 
 1 — 1 
 
 
 ^^ 
 
 \ y. 
 
 \ 
 
 _1 
 
 
 1— 
 
 
 *»». 
 
 \ c^> 
 
 \ 
 
 Q_ 
 
 
 
 
 
 
 ^ \ uJ 
 
 \ ^ 
 
 D_ 
 
 
 LU 
 
 
 
 \ \ y 
 
 \ 
 
 < 
 
 
 _J 
 U_ 
 
 
 
 \ \ 
 
 \ 
 
 O 
 
 
 LU 
 
 
 
 v \ 
 
 \ 
 
 < 
 O 
 
 
 Q 
 
 
 
 \\ 
 
 V 
 
 1 
 
 
 LU 
 CD 
 
 
 
 
 \ _ 
 
 CD 
 
 
 ZD 
 
 
 
 V 
 
 s \ 
 
 LU 
 
 O 
 
 1 
 
 
 CD 
 
 _l 
 
 
 
 
 \N \ 
 
 O 
 Csl 
 
 
 
 
 
 
 
 \ \ \ 
 
 
 1— H 
 
 f- 
 
 
 
 
 \ -\ \ " 
 
 
 cn 
 
 UJ 
 
 
 
 
 \ \\ 
 
 1 
 
 > 
 
 r 
 
 ~~i i 
 
 " 1 ~ r -1- - 
 
 •— h-^l — 1 
 
 1 1 rv^ 
 
 LT; 
 CM 
 
 o 
 
 (XI 
 
 LT) 
 
 oo 
 
 DC 
 
 CD 
 
 o 
 
 CO 
 
 
 o 
 
 CXI 
 
 «=c 
 
 I/O 
 
 I— 
 
 l>0 
 
 l— 
 00 
 
 LU 
 
 DC 
 
 dc 
 
 
 LU 
 
 cc 
 
 
 CJ 
 
 
 
 
 < 
 
 u_ 
 
 
 _l 
 
 
 
 O. 
 
 1— _ 
 
 
 OO 
 
 ^s 
 
 
 t— 4 
 
 1 1 1 <c 
 
 
 a 
 
 LU 
 
 
 _i 
 
 u^ 
 
 
 <£ 
 
 <r lu 
 
 
 
 a: 
 
 _1 CD 
 
 ' 
 
 LU 
 
 Q_ CQ 
 
 • 
 
 1— 
 
 *" 2 
 
 
 -=c 
 
 •-. OC 
 
 
 _l 
 
 LIED LOAD VS. D 
 OVED SYNTHETIC 
 
 LO 
 
 
 CL O 
 
 O 
 
 
 Q. OH 
 d CD 
 
 C\J 
 
 LU 
 
 DC 
 
 CD 
 
 O 
 
 O 
 CO 
 
 O 
 C\J 
 
 (Sdix) avoi cmiddv 
 
 40 
 
the most flexible pads in use on mainline railroads in Europe and Japan. For 
 adequate attenuation of impact strain under high speed traffic, it may be 
 necessary to use a much softer pad [9]. The results show a need for the 
 evaluation of the gauge widening and rollover allowed by very flexible pads 
 on curved track under 100-ton traffic. This would provide the information 
 required for selection of rollover restraint criteria which assured track 
 safety without representing an unnecessary restriction on the use of flexible 
 pads. 
 
 The most important information in Figures 20 and. 21 concerns the 
 specifications for repeated loads tests (Test 4 of Table 1). The AREA 
 repeated load test is conducted at an L/V angle of 20 degrees, while the 
 effective L/V angle for the Amtrak test is 18 degrees. From Figure 20 it 
 can be seen that a test conducted at the 20-degree L/V angle to a maximum 
 load of 30 kips produced about 0.015 inches of rail head lateral deflection. 
 The results in Figure 21 for the flexible pad produced very little vertical 
 rail/tie deflection, although substantial rail head lateral deflection 
 occurred (80 mils at 30 kips). Most of this deflection resulted from 
 lateral translation of the rail segment. 
 
 The vertical displacement is partially compensated in the AREA 
 tests by the uplift load of 0.6 x pad separation load. However, the strain- 
 deflection results presented in the next section show that the uplift load 
 primarily affects field clip uplift deflection while the compressive load 
 primarily affects gauge clip uplift deflection. Uplift loads were required 
 for the fatigue tests to produce a balance of the peak-to-peak vertical 
 rail-to-tie deflections on the field and gauqe sides. 
 
 The lateral restraint tests indicate that the repeated loads cur- 
 rently conducted at a 20-degree loading angle do not provide rail /tie dis- 
 placements representative of a severe loading environment for rigid pads. 
 The tests also indicate that rail /tie displacements are highly sensitive to 
 L/V angle and pad stiffness. Given these sensitivities, it is reasonable to 
 assume that the tests can be affected by the details of test fixtures which 
 apply the same nominal loads. For example, a variation of the height of the 
 loading rod attachment to the rail segment or the presence of pivot friction 
 would change the load-deflection relationship. These problems led to the 
 approach, described in the following section, by which the fatigue environ- 
 ment is directly monitored through rail/tie deflections. 
 
 Fastener Fatigue Tests 
 
 One of the major objectives of this program was to define fastener 
 qualification tests which could simulate the service environments of fast- 
 eners. To define the service environments in a way which could be repro- 
 duced in the laboratory, rail /tie deflection measurements were made on four 
 fastener systems installed at FAST. 
 
 Two of the fastener systems, one on concrete ties and one on wood 
 ties, had required substantial replacements of components. Each system used 
 
 41 
 
the Type A clip. The concrete tie system had experienced clip fallouts and 
 fractures. In addition, where flexible pads of grooved synthetic rubber or 
 corded rubber were installed on the concrete tie track, some abrasive wear 
 and delamination of the pads occurred. On the wood tie system, a few of the 
 tie plates and many of the screw spikes used with the Type A system had 
 fractured. The cause of all component fractures was fatigue loading. 
 
 The following subsections describe fatigue tests conducted on one 
 concrete tie fastener system and one wood tie system, both of which use the 
 Type A clip. Loads and L/V angles were adjusted so that the loading arrange- 
 ment (using an available fixture) reproduced the range of two important 
 rail/tie deflections measured at FAST. In each case, the tests resulted in 
 fatigue failures identical to some of those which have occurred at FAST. 
 
 Concrete Tie Fastener Fatigue Test 
 
 Figure 22 shows a schematic of the loading and deflection measure- 
 ment arrangement used for the concrete tie fastener test. Details of the 
 loading fixture are provided in Appendix A. A trial and error search was 
 conducted to determine the combination of L/V angle and load range which 
 would most closely duplicate the rail head lateral deflection and vertical 
 clip deflections measured at FAST. 
 
 No attempt was made to conform to the loading geometry of the AREA 
 test which is conducted in an approximately similar manner. It should be 
 pointed out that in no case are the loads applied to a single fastener using 
 a short rail segment equivalent to the wheel/rail loads experienced in 
 track. Through bending and torsional resistance, the rail in track acts to 
 distribute loads to several adjacent ties and fasteners. 
 
 The objective of the loading arrangement was to simulate, as 
 closely as possible, the following combination of rail /tie deflections for 
 both rigid and flexible pads: 
 
 a. Gauge and field clip vertical deflection: 0.040 inches 
 
 peak-to-peak 
 
 b. Rail head lateral deflection: 0.100 inches 
 
 peak-to-peak 
 
 After initial trials with L/V angles between 20 and 27 degrees, a final 
 selection of 24 degrees was made. To simulate the deflection goals within 
 approximately 10 percent, the following ranges of loads were required for 
 hard and soft pads: 
 
 a. Rigid polyethylene pad: 20 kips compression, 2400 pounds uplift 
 
 b. Flexible synthetic rubber* 
 
 or grooved Duraflex pad: 13 - 16 kips compression, 1600 - 
 
 2000 lb uplift. 
 
 *This is the flexible pad used in Subsections 17-J2 and -Kl of the FAST 3- 
 degree curve. 
 
 42 
 
o 
 o 
 
 on 
 o 
 
 I— 
 
 
 o 
 
 Q 
 
 o 
 
 C\J 
 C\J 
 
 CJ3 
 
 43 
 
During the early stages of fatigue testing, frequent minor adjust- 
 ments of load levels were required to maintain the desired deflection levels. 
 Two changes to the loading fixture were made during this period: 
 
 a. The clevis holes were honed and a pin of hardened 4340 
 steel was substituted for the original mild steel pin. 
 
 b. Grease fittings were added to the clevis. 
 
 These measures stabilized the loading arrangement so that the deflections 
 produced by given load levels would remain stable for 8 - 10 hours. Greasing 
 was performed at a maximum interval of 8 hours. A load rate of 2 cycles 
 per second was maintained throughout the initial trials and tests. This 
 rate could have been doubled or halved without significantly affecting the 
 short-term load-deflection relationship. However, a slight heat buildup 
 (about 5 degrees on the rail segment) developed at the 2 Hz rate. This 
 prevented an increase in loading rate, and economics prevented a decrease in 
 the rate. . 
 
 Three pads were used during the test. The flexible 
 synthetic rubber pad was used through 20,000 cycles when a routine inspection 
 was made. The pad was badly abraded where it contacted the field side edge 
 of the rail base. The hard polyethylene pad was substituted and maintained 
 until 160,000 cycles were completed. No damage occurred. Finally, a flexible 
 grooved Duraflex pad was inserted and retained for the duration of the test 
 (653,000 cycles). The loads were changed as indicated previously to maintain 
 constant rail/tie deflections. An inspection after test completion revealed 
 that this pad had also compressed and abraded where it contacted the field 
 side rail base. An additional circular worn area about 1 inch in diameter 
 was caused by a rough spot of the tie surface. 
 
 Figure 23 shows typical load-deflection and strain deflection 
 relationships recorded with the Duraflex pad. These were collected by 
 temporary substitution of instrumented clips during shut-downs for inspections 
 The desired peak-to-peak clip vertical deflections were very nearly maintained 
 The rail head lateral deflection was recorded along with the load levels on 
 a strip chart. The lateral peak-to-peak deflection varied between 105 and 
 115 mils. 
 
 At 653,000 cycles it was discovered that the clip had cracked in 
 the location 1 shown on Figure 24. Subsequent microscopic examination re- 
 vealed that a second crack had formed at location 2. Clips in track had 
 fractured at both of these locations. The test was terminated at this point. 
 
 -fi 
 The FAST train, which produced 33 x 10" MGT per axle, would develop 
 
 21.5 MGT by the passage of 653,000 axles. The first group of clip failures 
 in track began about 40 MGT after installation. After the track rebuild at 
 425 MGT, clips from two different batches began to fail immediately. The 
 test clip came from one of the latter batches. The test represented a severe 
 loading environment since the deflection levels applied in the lab occurred 
 for only a small percentage of axles in track. Considering possible varia- 
 tions in clip properties and loading conditions, the degree of representation 
 of track performance can be judged acceptable. 
 
 44 
 
T-14 
 
 Gauge Clip 
 
 Field Clip 
 
 -40 
 
 30 -20 -10 
 (1 Volt = 166 we) 
 
 10 
 
 r 
 
 £ 3 
 
 C 
 
 1— 
 
 V 
 
 
 
 
 
 
 
 
 
 
 
 
 
 o 2 
 
 > 
 
 c 
 
 
 
 
 
 
 
 
 
 
 Q. 
 
 
 
 
 
 
 
 
 
 
 Gauge C 
 
 o 
 
 
 
 
 
 
 
 
 
 
 y\ — 
 
 -1 
 
 
 
 
 
 -40 -30 -20 -10 
 
 10 
 
 s— Up-tc^ CompA.. 
 
 Gauge Clip Vertical 
 Deflection (mils) 
 
 -20 -10 10 20 
 
 - UpLL^t CompxzA6<Lcn— * 
 
 Field Clip Vertical 
 Deflection (mils) 
 
 FIGURE 23. TYPICAL LOAD-DEFLECTION AND STRAIN-DEFLECTION PLOTS 
 FROM FATIGUE TEST OF CONCRETE TIE FASTENER 
 
 45 
 
Location 1 
 
 Location 
 
 FIGURE 24. LOCATIONS OF CLIP CRACKS DEVELOPED 
 DURING CONCRETE TIE FATIGUE TEST 
 
 46 
 
Wood Tie Fastener Fatigue Test 
 
 Figure 25 shows the loading schematic for the wood tie fatigue test. 
 The desired vertical deflections were relatively large compared to those for 
 concrete tie track. The rail/tie deflection goals were: 
 
 a. Rail-to-tie vertical deflection at 
 
 gauge and field clips: 0.100 inches peak-to-peak 
 
 b. Rail head lateral deflection: 0.100 inches peak-to-peak. 
 
 It was also necessary to induce most of the vertical deflection through bend- 
 ing of the tie plate rather than through flexing of the clips. A clip strain 
 limit of about 1.5 volts had been seen in the track. A goal of 2 volts peak- 
 to-peak (about 333 microinches/inch at the measurement location) was adopted 
 to remain well below the strain limit of the clips (about 3.6 volts uplift 
 strain above the installed strain). 
 
 Loading trials with the new tie block revealed that those deflec- 
 tion levels could not be reached within reasonable load limits without 
 severely flexing the clips. This necessitated a partial adzing of the tie 
 at the rail seat to reproduce the conditions observed in track, where most 
 of the vertical deflection took place through flexing of the tie plate over 
 the irregular surfaces of the service ties. The adzing was sufficient to 
 produce the following conditions: 
 
 a. Field clip vertical deflection: 88 mils peak-to-peak 
 
 b. Gauge clip vertical deflection: 50 mils peak-to-peak 
 
 c. Rail head lateral deflection: 60 mils peak-to-peak. 
 
 This condition was adopted since it reproduced the essential phenomena of 
 tie-plate bending at the field side. Greater deflections would have required 
 significantly higher loads. The loads under these conditions were: 
 
 Compression: 16 kips 
 
 Uplift: 800 pounds. 
 
 As the test proceeded, the tie surface compressed and the deflections grad- 
 ually increased to the following maximum values: 
 
 a. Field clip vertical deflection: 97 mils peak-to-peak 
 
 b. Gauge clip vertical deflection: 60 mils peak-to-peak 
 
 c. Rail head lateral deflection: 70 mils peak-to-peak. 
 The screw spikes required tightening about eyery 200,000 cycles. 
 
 47 
 
CD 
 
 i— 
 
 Li_ 
 
 00 
 
 Q 
 O 
 O 
 
 O 
 
 O 
 
 t— I 
 
 o 
 
 C3 
 
 Q 
 
 o 
 
 C\J 
 UJ 
 
 en 
 
 48 
 
The test was terminated at 1,090,000 cycles when it was discovered 
 that the tie plate had fractured across its width at the location shown in 
 Figure 25. This was the same type of failure which had occurred in service 
 at FAST [4]. However, the test simulated only 36 MGT of traffic, while the 
 plate failures began in track at 155 MGT. The partial adzing of the tie 
 rail seat probably created more bending strain than occurred at FAST under 
 the same deflection levels. The test represented loading events which 
 occurred in track only a small percentage of the time, but was successful 
 in reproducing two principal performance problems (plate fracture and spike 
 loosening) . 
 
 The plate used in this study has since been redesigned and 
 strengthened where the fractures occurred. None of the redesigned plates 
 has yet been tested at FAST. 
 
 49 
 
DISCUSSION AND RECOMMENDATIONS 
 
 The results of this study have led to recommendations for modifications, 
 deletions, additions or retention in current form of the AREA/Amtrak fastener 
 system qualification tests listed in Table 1. The recommendations are presented 
 and discussed as follows. 
 
 1 . Fastening Insert Test 
 
 Retain in current form. 
 
 2. Fastening Uplift Test 
 
 The fastening uplift test provides a good indicator of the ability 
 of the clips to absorb uplift load and to provide a continuous level of rail 
 clamping force while flexing under train action. To better define fastener 
 uplift capacity, the following changes are recommended: 
 
 a. Clip dimensional checks of the type illustrated in Figure 11 
 should be made before and after the test. Retention of the 
 original clip dimension should be required within 0.005 inches. 
 
 b. The maximum uplift point should be specified in terms of an uplift 
 displacement past pad separation, in a manner similar to the 
 procedure of the current Amtrak test. The value of 0.050 inches 
 uplift displacement is recommended. The tests under severe 
 conditions at FAST revealed no requirement exceeding this value. 
 In addition, the test should not proceed past the maximum load 
 
 of 6000 pounds. This limit will prevent the bias of the test 
 against rigid fasteners. 
 
 c. For each test run, the pad separation load should be determined 
 as the mean of at least three successive trials. This mean 
 load should not be allowed to vary more than 20 percent for 
 measurements before and after the repeated loads tests, as is 
 currently in the Amtrak series. 
 
 d. The current Amtrak test requires the measurement of an uplift 
 spring rate which must fall within a specified range. Pad spring 
 rate is of interest because it indicates the ability of the pad 
 to absorb impact loads. However, there is no correlation between 
 the relatively low uplift loads used to measure uplift spring 
 rate and the ability of pad to absorb impact. It is recommended 
 that this requirement for a uplift spring rate be deleted and 
 that a requirement for a compressive spring rate, measured during 
 the pad compression test, be substituted. 
 
 50 
 
3. Fastening Repeated Loads Test 
 
 The results of this study show that the current repeated loads tests 
 do not represent the worst fastener loading conditions identified on the FAST 
 track. The AREA test is conducted at an L/V angle of 20 degrees and the Amtrak 
 test involves an effective L/V angle of 18.4 degrees. The lateral restraint 
 tests conducted as part of this study show that such angles produce rail head 
 lateral deflections and vertical clip deflections much smaller than the maximum 
 values found at FAST. It is recommended that the L/V angle of the repeated 
 loads test be set to reproduce the maximum rail-to-tie deflections expected 
 in service. The values used for the concrete tie fastener repeated loads test 
 described in this report were 0.040 inches peak-to-peak vertical clip deflection 
 and 0.100 inches peak-to-peak rail head lateral deflection. These deflections 
 could be reproduced in either of the following ways: 
 
 a. Beginning with the nominal loads application of the Amtrak 
 test, lower the vertical load until the maximum deflections 
 reach the desired levels. 
 
 b. Beginning with the nominal loads application of the AREA 
 test, raise the plane of incline of the fastener test 
 fixture until the desired displacement levels are attained. 
 
 4. Fastening Longitudinal Restraint Test 
 
 The current longitudinal restraint test is adequate to establish 
 the relative longitudinal resistance of fastener systems to thermal loading 
 of unoccupied track. Some improvements in test procedures are needed. A 
 requirement for two dial indicators [7] should be removed and replaced with 
 a requirement for a displacement device accurate to 0.0005 inches and mounted 
 on the centerline of the load actuator, at the opposite end of the test rail 
 segment. To obtain this sensitivity, the limit of slip displacement should 
 be defined at 0.10 inches, rather than at the current limit of 0.25 inches. 
 The maximum longitudinal restraint load is normally obtained with a few 
 mils of longitudinal displacement. 
 
 The longitudinal load should be applied to the point of slippage 
 
 or to maximum load attainment, whichever requires the higher load. The 
 
 current test stops at 2400 pounds. The extension will provide a factor for 
 the relative evaluation of fastener systems. 
 
 Further study is needed of the effect of vibratory loading on the 
 longitudinal restraint of fasteners. If vibratory loading changes the per- 
 formance ranking of a representative selection of fasteners, then an inde- 
 pendent test of longitudinal restraint under vibratory loading should be 
 considered. 
 
 5. Lateral Load Restraint Test 
 
 This AREA test applies loads to the test rail segment at an L/V angle 
 of 30 degrees. An initial load cycle to 20 kips removes the slack which may 
 
 51 
 
exist between the rail base and the field side fastener shoulder. The first 
 part of the test examines rail base lateral restraint by applying the vertical 
 load through a wooden block to a maximum of 41 kips. This test is adequate 
 to assess the structural integrity of the insulator and fastener shoulder. 
 
 The second part of the test examines the restraint of the fastener 
 system to rollover displacement (difference between rail head and rail base 
 lateral displacement). Rollers are used to eliminate any lateral restraint 
 between the load actuator and the rail head. Rollover displacement is restricted 
 to 1/4 inch under a load of 20.5 kips. 
 
 The results shown earlier in Figure 21 indicate that fasteners using 
 flexible pads (static stiffness below about 1.4 million pounds per inch) may be 
 unable to restrict the rollover displacement to the required 1/4 inch. The 
 measurements at FAST indicated that rail head lateral displacements in track 
 are a yery weak function of pad stiffness. On the other hand, the initial 
 displacements in the laboratory setup are a s/ery strong function of pad stiffness, 
 which controls rollover restraint until the upper corner of the rail base 
 lifts off the pad. At this point the stiffness and strength of the clip begin 
 to contribute significantly to rollover restraint. Thus, a flexible pad may 
 require a stronger clip to meet the requirements of this test. Because the 
 laboratory test shows a strong dependence of pad stiffness on rollover dis- 
 placement while the track measurements do not, it is recommended that this 
 test be deleted. 
 
 6. Tie Pad Load/Deflection Test 
 
 Recent discoveries of rail seat bending cracks in concrete ties in- 
 stalled on the Northeast Corridor and in several revenue service test segments 
 have led to an investigation of the effect of pad stiffness on the levels of 
 tie bending strain produced by impacts from wheel irregularities [9J. The 
 study has provided several important conclusions: 
 
 a. Where ties are installed with rigid pads, the maximum tie 
 bending strains produced by wheel irregularities on high-speed 
 traffic can far exceed the levels required to cause rail seat 
 bending cracks. 
 
 b. The occurence of crack-producing impact strain levels can be 
 reduced or eliminated by major reductions in tie pad stiffness. 
 For example, laboratory tests showed that a reduction in 
 dynamically measured compressive pad stiffness from 5 million 
 pounds per inch to 500,000 pounds per inch can attenuate 
 maximum impact strains by 40 percent. 
 
 c. The compressive stiffness of some pad materials is significantly 
 affected by the rate of load application. 
 
 On the other hand, it should be pointed that there are some loading 
 environments where pad stiffness is not a major concern. Some ties on the FAST 
 track have endured almost 600 MGT of heavy-haul traffic without cracking. The 
 FAST track and train are regularly maintained, and the maximum speed of the train 
 is 45 mph. Other test segments have been in revenue service for several years 
 without significant cracking. Finally, there are some loading environments where 
 cracked ties can remain in service for many years without impairment of function. 
 
 52 
 
Cracked ties from the Kansas Test Track have seen almost 600 MGT of service at 
 FAST without further deterioration. 
 
 Therefore, it is recommended that the selection of an appropriate 
 tie pad stiffness for a given railroad application be based on a knowledge 
 of the track loading environment. The selection should be based upon the 
 frequency with which crack-producing wheel/rail loads or tie strains are 
 expected to occur. Data for such determination will be available from current 
 supporting investigations. Data from the Northeast Corridor track [15] contain 
 the highest frequency of crack-producing tie strains yet found in U. S. revenue 
 service. Bending strains above the level which can cause cracking at rail 
 seats were found to occur at a typical site about 0.1 percent of the time. 
 This represents an average frequency of occurrence of one event with the 
 potential to produce a crack about every 1 to 2 days. 
 
 As a general guideline, it can be stated that the dynamically 
 measured compressive stiffness of new tie pads should not exceed 1 million 
 pounds per inch where any of the following conditions are expected to exist: 
 
 a. the speeds of normally maintained freight and passenger 
 traffic will exceed 60 mph. "Normal" maintenance will permit 
 advanced wheel irregularities to develop. 
 
 b. track defects such as severe engine burns exist. 
 
 c. special track work (grade crossings, turnouts) may produce 
 unusual dynamic loading. 
 
 It is recommended that a guideline of this type be included in fastener qualification 
 specifications, and that the dynamic compressive tie pad stiffness be included 
 as part of tie pad load/deflection test. 
 
 53 
 
APPENDIX A 
 
 DATA FROM EXISTING FASTENER PERFORMANCE TESTS 
 
 54 
 
TABLE A-l. FASTENER CLIP FORCE-DEFLECTION PROPERTIES 
 
 SPRING RATE AT 
 INSTALLED DEFLECTION 
 
 VERTICAL CLIP DEFLECTION (INCH 
 
 FASTENER CLIP TYPE 
 
 VaXa 
 Sowice. 
 
 1 
 
 INSTALLED 
 
 DEFLECTION 
 
 (INCH) 
 
 CLIP FORCE AT 
 INSTALLED DEFL. 
 (LB) 
 
 SPRING RATE AT 
 INSTALLED DEFL. 
 (LB/IN) 
 
 PANDROL 601 A 
 
 A. DESIGN CURVE 
 
 B. QUALIFICATION TESTS 
 
 INITIAL TESTS* - MEAN 
 RANGE 
 
 FINAL TESTS* - MEAN 
 RANGE 
 
 A-l 
 
 .590 
 
 .598 
 .562- .637 
 
 .590 
 .570- .637 
 
 2200 
 
 2290 
 1750- 2780 
 
 1940 
 1530-2350 
 
 5640 
 
 4840 
 3820-6120 
 
 5160 
 3485 - 6980 
 
 HIXSOH (QUALIFICATION TEST) 
 
 A-2 
 
 .400 
 
 2800 
 
 7270 
 
 VSD (QUALIFICATION TEST) 
 INITIAL TEST 
 
 FINAL TEST 
 
 A-3 
 
 .449 
 .412- .478 
 
 .306 
 .286- .326 
 
 2520 
 2250 - 2860 
 
 1425 
 1270- 1580 
 
 6800 
 6580- 7210 
 
 5330 
 5240 - 5420 
 
 SPRINGLOCK (MANUF. DATA) 
 CS3 CLIP 
 CS5 CLIP 
 
 A-4 
 
 
 .41 
 
 .41 
 
 1700 
 2500 
 
 <&.■:■>& ■V , '«^'- 
 
 4100 (Avg.) 
 6100 (Avg.) 
 
 DE-SPRINGCLIP (MANUF. DATA) 
 RANGE 
 
 A-4 
 
 .394 
 
 1650- 2650 
 
 4200- 6720 
 
 RN - CLIP (MANUF. DATA) 
 
 A-4 
 
 .157 
 
 1800 
 
 :lHi|£4 
 
 El = 2460 
 
 (< .161") 
 
 E2 = 7480 
 
 (> .161") 
 
 SIDEWINDER (TEST DATA) 
 
 A-5 
 
 No data 
 
 2042 - 2884 
 
 No data 
 
 Between initial and final clip force-deflection tests, tie pad load-deflection, fastening uplift, 
 longitudinal restraint, repeated loads, and push-pull tests were conducted. 
 
 55 
 
TABLE A-2. FASTENER SYSTEM UPLIFT PROPERTIES 
 
 VERTICAL UPLIFT VET-LECTIOH (INCH) 
 
 FASTENER TYPE 
 
 Data 
 SouAce 
 t 
 
 PAD SEPARATION 
 
 FORCE I DEFLECTION 
 (LB) I (INCH) 
 
 MAXIMUM 
 
 FORCE 
 (LB) 
 
 DEFLECTION 
 (INCH) 
 
 UPLIFT 
 
 SPRING RATE 
 
 (LB/IN) 
 
 PANDROL 601A/ EVA PAD (QUALIFICATION TESTS) 
 INITIAL TESTS* - MEAN 
 
 FINAL TESTS* - MEAN 
 
 A-l 
 
 4830 
 4065 
 
 .0172 
 .0175 
 
 6080 
 5370 
 
 , , ,' ■ ... iHIVJU 
 
 2,310,000 
 1,940,000 
 
 HIXSON/ GROOVED RUBBER PAD 
 
 A-2 
 
 4000- 
 5400 
 
 .018- 
 .042 
 
 8500 
 
 .17 
 
 235,000 
 
 V SD/ 6 mm NEOPRENE PAD (QUALIFICATION TESTS) 
 INITIAL TESTS* - MEAN 
 FINAL TESTS* - MEAN 
 
 A-3 
 
 6860 .027 
 
 4290 
 
 .024 
 
 8690 
 6745 
 
 1,509,000 
 678,000 
 
 SIDEWINDER/ POLYURETHANE PAD 
 
 A- 5 
 
 5545 No data 
 
 11,090 I No data 
 
 No data 
 
 Between initial and final uplift tests, longitudinal restraint, repeated loads and 
 push-pull tests were conducted. 
 
 56 
 
TABLE A-3. TIE PAD COMPRESSION LOAD-DEFLECTION PROPERTIES 
 
 PAD 
 
 LOAP 
 
 (LB) 
 
 PAP COMPRESSION 
 SPRING RATE 
 
 ///)//// f t/iiif/// 
 
 PAD DEFLECTION (INCH 
 
 FASTENER TYPE 
 
 VaXa. 
 SouA.ce. 
 
 INSTALLED 
 LOAD 
 (LB) 
 
 INSTALLED 
 
 DEFLECTION 
 
 (INCH) 
 
 PAD 
 SPRING RATE 
 (LB/IN) 
 
 DUPONT VINYL EVA PAD (WITH PANDROL CLIP) 
 
 A-l 
 
 
 
 
 INITIAL TEST - MEAN 
 
 
 4915 
 
 .0033 
 
 3,700,000 
 
 FINAL TEST - MEAN 
 
 
 4210 
 
 .0060 
 
 3,850,000 
 
 PANDROL TESTS (TO 4500 LB ONLY) 
 
 A-4 
 
 
 
 
 POLYETHYLENE PAD 
 
 
 4000 
 
 .0020 
 
 4,100,000 
 
 FLAT NEOPRENE PAD 
 
 
 4000 
 
 .0051 
 
 2,800,000 
 
 GROOVED NEOPRENE PAD 
 
 
 4000 
 
 .0087 
 
 1,800,000 
 
 PANDROL SPECIFICATION (AT 20 METRIC TONS, 
 CURVES INTERSECT . 
 
 BOUNDARY 
 4 and .8 mm) 
 
 4000 
 (ASSUMED) 
 
 .0034- 
 
 .0114 
 
 2,300,000 - 
 1,230,000 
 
 VSD/6 mm NEOPRENE PAD 
 
 A-3 
 
 
 
 
 INITIAL TESTS - MEAN 
 
 
 5180 
 
 .0214 
 
 865,000 
 
 FINAL TEST* (ONE TEST) 
 
 
 4000 
 
 .0220 
 
 770,000 
 
 MANUFACTURER'S DATA 
 
 
 4000 
 (ASSUMED) 
 
 .0044 
 
 825,000 
 
 SHINKANSEN DATA ENVELOPE 
 
 A-6 
 
 4000 
 (ASSUMED) 
 
 .0035 - 
 .0112 
 
 740,000 - 
 840,000 
 
 Between initial and final pad compression tests, fastening uplift, longitudinal restraint, 
 repeated load, push-pull and rail clip load-deflection tests were conducted. 
 
 57 
 
TABLE A-4. FASTENER LONGITUDINAL RESTRAINT PROPERTIES 
 
 LONGITUDINAL 
 LOAD 
 
 DISPLACEMENT 
 TRANSDUCERS 
 
 FASTENER TYPE 
 
 Data 
 SouAce. 
 
 PANDROL 601 A/ EVA PAD 
 
 FASTENER SYSTEM 1 - INITIAL* 
 
 - FINAL* 
 FASTENER SYSTEM 2 - INITIAL 
 
 - FINAL 
 
 PANDROL 601A WITH EXTERNAL INSULATOR 
 (Mean of 2 Tests) 
 
 A-l 
 
 A-7 
 
 POLYETHYLENE PAD 
 FLAT NEOPRENE PAD 
 GROOVED NEOPRENE PAD 
 
 PANDROL 601A WITH INTERNAL INSULATOR 
 (Mean of 2 Tests) 
 
 POLYETHYLENE PAD ln i^ M }j d '■ 
 
 FLAT NEOPRENE PAD SHouLdeA 
 
 GROOVED NEOPRENE PAD II 
 
 PANDROL 401 WITH EXTERNAL ?n,sulatqr 
 (Mean of 2 Tests) 
 
 POLYETHYLENE PAD 
 FLAT NEOPRENE PAD 
 GROOVED NEOPRENE PAD 
 
 HIXSON/ GROOVED RUBBER PAD 
 
 A-2 
 
 VSD/ 6 mm NEOPRENE PAD 
 TWO INITIAL TESTS 
 TWO FINAL TESTS 
 
 A-3 
 
 SIDEWINDER/ POLYURETHANE PAD 
 
 MAX LOAD 
 
 A-5 
 
 DE SPRINGCLIP/MASONITE PAD 
 (ON STEEL TIE PLATE & WOOD TIE) 
 
 MEAN OF 3 TESTS 
 
 A-8 
 
 LONGITUDINAL 
 LOAD 
 (LB) 
 
 2400 
 1690 
 1620 
 1710 
 
 3950 
 4000 
 4000 
 
 3300 
 4950 
 4650 
 
 3650 
 3700 
 3650 
 
 2400 
 3100 
 
 2400 
 1900, 2000 
 
 3100 
 
 3480 
 3580 
 
 SLIPPAGE 
 (INCH) 
 
 .016 
 
 CONTINUOUS 
 CONTINUOUS 
 CONTINUOUS 
 
 SLIP 
 SLIP 
 SLIP 
 
 SLIP 
 SLIP 
 SLIP 
 
 SLIP 
 SLIP 
 SLIP 
 
 .020 - .024 
 CONTINUOUS 
 
 .038 - .0385 
 CONTINUOUS 
 
 START SLIP 
 
 START SLIP 
 CONTINUOUS 
 
 TOE LOAD BEFORE TEST 
 (AVG. OF 2 CLIPS) 
 
 (LB) 
 
 2370 
 2150 
 2350 
 2085 
 
 2800 
 
 2100, 2940 
 1490 
 
 
 2200 (Nominal) 
 
 Between initial and final longitudinal restraint tests, repeated loads, uplift, 
 and push-pull tests were conducted. 
 
 58 
 
TABLE A-5. FASTENER LATERAL RESTRAINT PROPERTIES 
 
 LATERAL 
 RESTRAINT TEST 
 
 VERTICAL APPLIED LOAD 
 (VARIABLE) 
 
 20.5 Kips 
 
 ROLLOVER RESTRAINT 
 TEST 
 
 FASTENER TYPE 
 
 Data. 
 Sou/ice 
 t_ 
 
 VERTICAL 
 
 APPLIED 
 
 LOAD 
 
 (KIPS) 
 
 LATERAL DEFLECTION AT 
 
 GAGE HEIGHT 
 (INCH) 
 
 BASE 
 (INCH) 
 
 DIFFERENCE 
 (INCH) 
 
 SIDEWINDER/ POLYURETHANE PAD 
 
 Tniulattd ShowidzA- 
 
 A-5 
 
 41 
 20.5 
 
 .128 
 
 .028 
 .014 
 
 .114 
 
 PANDROL 601 A 
 
 (1) 
 
 140 RE RAIL 1 
 
 T.niuJLaXe.d ■ 
 ShouldeA 
 
 (Mean of 2 Tests) 
 
 EXTERNAL INSULATOR 
 POLYETHYLENE PAD 
 FLAT NEOPRENE PAD 
 GROOVED NEOPRENE PAD 
 
 INTERNAL INSULATOR 
 POLYETHYLENE PAD 
 FLAT NEOPRENE PAD 
 GROOVED NEOPRENE PAD 
 
 PANDROL 401 (3) , BS 113A RAIL (4) 
 (Mean of 2 Tests) 
 
 EXTERNAL INSULATOR 
 POLYETHYLENE PAD 
 FLAT NEOPRENE PAD 
 GROOVED NEOPRENE PAD 
 
 (1) 7/8" Diameter Clip 
 
 (2) 7 5/16" Height x 6" Rail Base Width 
 
 (3) 13/16" Diameter Clip 
 
 (4) 6.25" Height x 5.5" Rail Base Width 
 
 A-7 
 
 40 
 40 
 40 
 
 40 
 40 
 40 
 
 40 
 40 
 40 
 
 .103 
 .376 
 .494 
 
 .061 
 .159 
 
 .343 
 
 .075 
 .139 
 .150 
 
 .026 
 .051 
 .090 
 
 .014 
 .018 
 
 .002 
 
 .020 
 .043 
 .044 
 
 .077 
 .325 
 .405 
 
 .048 
 .142 
 .341 
 
 .055 
 .096 
 .106 
 
 CS-5 LEAF SPRING FASTENER (1971 British Tests, 
 
 probably BS 113A rail) 
 
 POLYETHELENE PAD 
 FLAT NEOPRENE PAD 
 
 GROOVED NEOPRENE PAD 
 
 A-7 
 
 22.4^ 
 38.1^) 
 51.5< 3 > 
 14 <«> 
 
 (1) Test terminated due to spalling of concrete shoulder. 
 
 (2) Crushing and spalling of concrete shoulder. 
 
 (3) Test terminated due to equipment limitations 
 and excessive gage widening. 
 
 (4) Test terminated due to concrete shoulder failure 
 
 .159 
 .197 
 .29 
 
 59 
 
TABLE A-6. FASTENER ROLLOVER RESTRAINT PROPERTIES 
 
 MAX LOAD = 20.5 KIPS 
 
 ROLLOVER RESTRAINT 
 TEST 
 
 FASTENER TYPE 
 
 VERTICAL 
 
 APPLIED 
 
 LOAD 
 
 (KIPS) 
 
 LATERAL DEFLECTION AT 
 
 GAGE HEIGHT 
 (INCH) 
 
 BASE 
 (INCH) 
 
 DIFFERENCE 
 
 (INCH) 
 
 SIDEWINDER/ PCLYUkCTHANE PAD 
 
 ItUmtaAcd ShouldeA 
 
 PANDROL 601A (1) , 140 RE RAIL^ 
 (Mean of 2 Tests) 
 
 EXTERNAL INSULATOR 
 POLYETHYLENE PAD 
 FLAT NEOPRENE PAD 
 GROOVED NEOPRENE PAD 
 
 INTERNAL INSULATOR 
 POLYETHYLENE PAD 
 FLAT NEOPRENE PAD 
 GROOVED NEOPRENE PAD 
 
 PANDROL 401 (3) , BS 11 3A RAIL^ 
 (Mean of 2 Tests) 
 
 EXTERNAL INSULATOR 
 POLYETHYLENE PAD 
 FLAT NEOPRENE PAD 
 GROOVED NEOPRENE PAD 
 
 A-7 
 
 hUtutatzd 
 ShoatdcA 
 
 20.5 
 
 20 
 
 20 
 20 
 
 20 
 20 
 20 
 
 20 
 20 
 20 
 
 ,128 
 
 .056 
 .097 
 .148 
 
 ,032 
 .037 
 .081 
 
 .024 
 .054 
 .070 
 
 ,014 
 
 .017 
 ,022 
 .048 
 
 .009 
 .011 
 .010 
 
 .007 
 ,026 
 .029 
 
 .114 
 
 .039 
 .075 
 .100 
 
 .023 
 .026 
 .071 
 
 .017 
 .028 
 ,041 
 
 (1) 7/8" Diameter Clip 
 
 (2) 7 5/1 b" Height x 6" Rail Base 'Width 
 
 (3) 13/16" Diameter Clip 
 
 (4) 6.25" Height x 5.5" Rail Base Width 
 
 60 
 
 
APPENDIX B 
 
 STRAIN-VOLTAGE RELATIONSHIP OF INSTRUMENTED CLIPS 
 
 61 
 
APPENDIX B 
 STRAIN-VOLTAGE RELATIONSHIP OF INSTRUMENTED CLIPS 
 
 Figure B-l shows the schematic of the 4-arm bridge containing two 
 active gauges and two completion resistors. Changes in resistance of the active 
 gauges produce response e n according to the equation 
 
 e o 1 T AR 1 AR 2 1 ( R = 120 "' nominal f °r / R 
 — = 4 |_-p- p-J • all arms of bridge) \V-\) 
 
 The linear strain at each gauge is related to the fractional change in resist- 
 ance of the gauge by 
 
 AR, aR ? 
 
 -~ = GF e y -£■ =GFe 2 . (B-2) 
 
 A ratio between e, and t, was established in laboratory tests conducted in 
 preparation for the field measurements described in Reference [14]. For bi- 
 axial gauges placed at the position indicated in Figure B-l, it was consistently 
 found that 
 
 e 2 = -0.44 E] • (B-3) 
 
 Thus an effective Poisson ratio of 0.44 was found. Substitution of (B-2) and 
 (B-3) into (B-l) yields 
 
 *f= f (1.44) e } . (B-4) 
 
 Shunt resistance R placed across two opposite arms of the bridge 
 
 ill produce the following ratio of response to excitation voltage 
 
 o 1 R 
 
 wi 
 
 e 
 
 Amplification of e yields 
 
 e 2 R + R 
 c 
 
 T= K T = lRTr- < B " 5 ) 
 
 62 
 
Red 
 
 [Colon* k<l{vi to 
 
 CUmplifaeA dOYlYiddULOYUi . ) 
 
 120 fi 
 
 Black 
 
 App/ioXsLmato. Location o& KulxajolI Gauge. 
 
 FIGURE B-l. SCHEMATIC OF TYPE A CLIP INSTRUMENTATION 
 
 63 
 
Because the installation strain and dynamic strain were of such different 
 magnitude, two separate shunt resistances were used. To determine K in each 
 case, the amplification was adjusted so that 
 
 _£ = i = 1. B-5 
 
 e 5 
 
 Then the two amplification factors were determined as follows: 
 
 a. Installati on St rain: R = 24,900 
 c 
 
 K. , - 2(120 ; 24,900) . 417 , „_ 5 
 
 inst. 120 
 
 b. Dynamic Strain : R = 100,000 
 
 „ . 2(120 + 100,000 ) lc7n R c 
 
 K dyn " T20 = 167 °- B " 5 
 
 Finally, the linear strain at Sauge 1 can be expressed in terms of output 
 voltage as 
 
 - 4 F 
 e l ' K GF (1.44) e o B-6 
 
 a. Installation Strain : 
 
 e l ■ (417)(2)(1.44)(5) E o = °- 000666 E o (1n/in) B " 6 
 
 = 666 E (yin/in). 
 
 b. Dynamic Strain 
 
 £ 1 = (1670)(2)(1.44)(5) E o = °- 000166 E o (in/in) B " 6 
 
 = 166 E (yin/in) . 
 
 To determine the relationship between strain gauge bridge output and 
 vertical clip deflection, the clips were subjected to the laboratory tests il- 
 lustrated in Figure B-2. An arbitrary strain "zero" was established with a 
 vertical clip load of 1700 pounds. The load was then cycled between 1400 and 
 2000 pounds. The slope of the curve of strain (volts) vs. deflection (inches) 
 was reasonably linear and very consistent, as shown in the figure. The mean 
 slope of the two clips illustrated is 0.0173 inches per volt, when the clips 
 are given shunt calibrations equivalent to those applied for dynamic measurements 
 
 in the field. 
 
 64 
 

 .10 
 
 
 .08 
 
 _ 
 
 .06 
 
 CO 
 
 
 LxJ 
 
 
 
 .04 
 
 z 
 
 
 ►— < 
 
 
 — 
 
 .02 
 
 2: 
 
 
 
 
 
 •— < 
 
 
 
 h- 
 
 
 O 
 
 
 UJ 
 
 
 1 
 
 -.02 
 
 LU 
 
 
 O 
 
 
 a. 
 
 -.04 
 
 1— 1 
 
 
 —j 
 
 
 -.06 
 
 
 -.08 
 
 
 -.10 
 
 -3 -2 
 
 1 -1 
 
 CLIP OUTPUT (VOLTS) 
 
 00 
 
 LU 
 
 o 
 
 2: 
 
 2T 
 
 o 
 
 t— I 
 
 o 
 
 LU 
 _J 
 LU 
 LU 
 Q 
 
 C_ 
 1 — t 
 —I 
 C_3 
 
 1 2 
 CLIP OUTPUT (VOLTS) 
 
 FIGURE B-2. MEASUREMENT OF CLIP RESPONSE VOLTAGE VS. DEFLECTION 
 IN LABORATORY FIXTURE 
 
 65 
 
APPENDIX C 
 
 LABORATORY TESTS OF INDIVIDUAL CLIPS 
 
 66 
 
APPENDIX C 
 LABORATORY TESTS OF INDIVIDUAL CLIPS 
 
 The fixture illustrated in Figure C-l was constructed to permit the 
 measurement of vertical load-strain and load-displacement relationships of 
 individual clips. Loading is applied through a vertical push rod which is 
 supported by a ball bushing. The ball bushing was required to overcome 
 friction created by a lateral component of the clip toe load. This lateral 
 component is produced because the contact with the clip toe is made through 
 a small segment of rail base which has slope of 1:4. 
 
 The fixture can be used with any combination of load cell and actuator 
 Deflection is measured with a DCDT as illustrated. Strain of the Type A clip 
 was measured as described in Appendix B. The fixture is adaptable to a variety 
 of clips by interchange of clip inserts. The stem of the insert passes through 
 a slotted hole in the lower horizontal plate and is constrained by a ring 
 containing set screws. 
 
 67 
 
VERTICAL LOAD — 
 
 / / / / / // 
 
 //////// 
 
 FIGURE C-l FIXTURE FOR MEASUREMENT OF CLIP FORCE-DEFLECTION- 
 STRAIN PROPERTIES 
 
 68 
 
REFERENCES 
 
 [1] Dean, F.E., "Research Plan for the Development of Improved Rail 
 
 Fastener Performance Requirements," report by Battelle's Columbus 
 Laboratories to the Improved Track Structures Research Division, 
 Federal Railroad Administration, Contract DOT-FR-9162, April 1980. 
 
 [2] Dean, F.E. , "Measurements of Rail/Tie Deflections and Fastener Clip 
 Strains at the Facility for Accelerated Service Testing," report by 
 Battelle's Columbus Laboratories for the Improved Track Structures 
 Research Division, Federal Railroad Administration, Contract DOT-FR- 
 9162, March 1981. 
 
 [3] "What Fastening System for Use on Curves?," Railway Track and 
 Structures , June 1978. 
 
 [4] Moody, H.G., "Comparison of Wood Tie Fastener Performance at FAST," 
 paper presented at the January 1980 meeting of the Transportation 
 Research Board. 
 
 [5] Takahara, Kiyosuke, "Track Degradation Prediction for the NEC," 
 July 31, 1979. Author is Deputy Director, Information Systems 
 Department, Japanese National Railways. 
 
 [6] Yasuo, Y., and Kakegawa, H., "Effects of Rubber-Pad Under Concrete 
 Tie on Pulverization of Ballast, Vibration and Noise," JNR Quarterly 
 Reports, Vol. 18, No. 1, pp 10-14, 1977. 
 
 [7] "Manual for Railway Engineering," American Railway Engineering 
 Association, Chicago, Illinois (Revised 1981). 
 
 [8] Ball, C.G., "Amtrak Qualification Tests of Prorail VSD Fastener 
 System," Portland Cement Association, Research and Development, 
 Construction Technology Laboratories, 5420 Old Orchard Road, Skokie, 
 Illinois 60077, August 1978. 
 
 [9] Dean, F.E. and Harrison, H.D., "Effect of Tie Pad Stiffness on the 
 Attenuation of Impact Strains in Concrete Ties," report by Battelle 
 Columbus Laboratories to the Improved Track Structures Research 
 Division, Federal Railroad Administration, Contract DOT-FR-9162, 
 FRA/ORD-82/19, May 1982. 
 
 [10] Lombard, P.C. and Wildenberger, L.A., "Concrete Sleepers and Fastenings 
 on the South African Railways," Civil Engineering Department, Track 
 Development, Report No. 24.597/08/09. 
 
 [11] Letter with Attachments from W.S. Autry, Chief Engineer, Santa Fe 
 Railway, to H.G. Moody, FRA, February 6, 1980. 
 
 [12] "Report on Static and Pulsating Creep and Torsion Tests Carried Out 
 on the DE-Springclip (Clamping Force 1000 kp)," Institut fur Bau von 
 Landverkehrswegen der TU Munchen, Report No. 524a, February 7, 1972. 
 
 69 
 
[13] "Pandrol Rail Clips Type PR601A - Stability of Properties with 
 
 Dynamic Loading," Report No. 380, Pandrol, Inc., Holborn, London, 
 EC1N2NE. 
 
 [14] Hadden, J. A., Prause, R.H., and Harrison, H.D., "Tests for the 
 
 Structural Evaluation of Pandrol Rail Fasteners in Section 17 at FAST," 
 report by Battelle's Columbus Laboratories to U.S. Department of 
 Transportation, Transportation Systems Center, Contract No. DOT-TSC- 
 1595, March 1979. 
 
 [15] Tuten, J.M., "Technical Memo on Analysis of Dynamic Loads and Strain 
 Data from the northeast Corridor Track," by Battelle's Columbus 
 Laboratories to the Federal Railroad Administration, Improved 
 Track Structures Research Division, Contract D0T-FR-9162, May 1981. 
 
 [A-l] Hanna,A. N., and Ball, C.G. "Amtrak Qualification Testing of Concrete 
 Ties and Fastenings - Testing of Fastenings," report to Santa Fe 
 Pomeroy/San Vel Concrete Corporation Joint Venture, by Portland Cement 
 Association, Research and Development, Construction Technology 
 Laboratories, July 1978. 
 
 [A-2] Lai, J.S., "A Summary Report for Evaluation of Hixson Concrete Cross- 
 Tie Fastening System*" report to Transit Products Company, Inc., from 
 School of Engineering, Georgia Institute of Technology, Atlanta, 
 Georgia, October 11, 1978. 
 
 [A-3] Ball, C.G., "Amtrak Qualification Tests of Prorail VSD Fastener 
 System," Portland Cement Association, Research and Development, 
 Construction Technology Laboratories, 5420 Old Orchard Road, Skokie, 
 Illinois 60077, August 1978. 
 
 [A-4] "Elasticity of Fasteners," Attachment A of a report by Bechtel , Inc. 
 on fastener performance (undated). 
 
 [A-5] Letter to W.R. Hamilton, Portec, Inc., from John W. Weber, Portland 
 Cement Association, March 22, 1976. 
 
 [A-6] Kiyosuke, Takahara, "Track Degradation Prediction for the NEC," 
 July 31, 1979. Author is Deputy Director, Information Systems 
 Department, Japanese National Railways. 
 
 [A-7] Way, G.H. , "Developmental Tests of Concrete Tie and Tie Fastening 
 
 Systems," report by C&0/B&0 Railway Co. to Department of Transportation, 
 Office of Research, Development and Testing, Contract No. D0T-FR- 
 20015, May 4, 1973. 
 
 [A-8] "Report on Static and Pulsating Creep and Torsion Tests Carried Out 
 on the DE-Springclip (Clamping Force 1000 kp)," Institute fur Bau von 
 Landverkehrswegen der TU Muchen, Report No. 524a, February 7, 1972. 
 
 70