design evaluations for the time dependent …docs.trb.org/prp/14-0322.pdf · bezgin, o.n. 1 1 trb 2...

Bezgin, O.N. 1

TRB 1

14-0322 2 3 4

DESIGN EVALUATIONS FOR THE TIME DEPENDENT CONTRACTIONS OF 5

PRESTRESSED CONCRETE HIGH SPEED RAILWAY SLEEPERS 6

N. Özgür Bezgin (corresponding author – [email protected]) 7

İstanbul University 8 Avcilar Campus, Civil Engineering Department 9

34320, Avcilar, İstanbul, Turkey 10 Phone: 00.90.533.663.9755 11

Fax: 00.90.212.473.7176 12 [email protected] 13

14 Number of words = 4.249 15

Number of tables = 2 16 Number of figures = 11 17

Total equivalent number of words = 7.499 18 19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

TRB 2014 Annual Meeting Original paper submittal - not revised by author

Bezgin, O.N. 2

ABSTRACT 34 Prestressed high performance concrete railroad sleepers are frequently used in ballasted high speed 35 railways. There are strictly enforced dimensional production tolerance values for the concrete sleepers on 36 the scale of millimeters. The creep and shrinkage of the prestressed high performance concrete sleepers 37 within regions of low humidity have the potential to reduce their gauge lengths below their allowed 38 tolerance values. The probability of rail abrasion, friction and derailment of a train under a given speed 39 increases with the level of reduction of the gauge length that requires an evaluation of the time related 40 dimensional changes of a sleeper before the finalization of its design. Ankara - Konya High Speed 41 Railway located in Turkey is a ballasted railway that became operational in 2011. An analytical 42 engineering evaluation that was conducted for the prestressed high strength B70 class sleepers showed 43 that the cumulative dimensional changes of the sleepers due to the effects of prestressing, shrinkage and 44 creep after demolding had the potential to reduce their shoulder length and the associated gauge length 45 below the allowed tolerance within two months of their production. The sleeper molds were dimensioned 46 to include an estimate for the time dependent contractions. The contractions of a number of sleepers 47 produced by the adjusted molds were measured for two months, confirming the requirement to modify the 48 production lengths of the sleepers to be used in high speed railways proposed along routes within arid 49 climates. 50

INTRODUCTION 51 Sleepers align and secure the rails at a fixed distance apart and transfer the loads that are imposed onto the 52 rails to the underlying supportive materials. They are exposed to cyclic and dynamic train loads under the 53 effects of environmental conditions. The sleepers of a high speed railway line must be designed and 54 produced to levels of precision higher than typical structural concrete elements. Attainment of this 55 precision provides for the safe and reliable operation of a high speed line. Train speed is heavily 56 dependent on the railway track qualities and the loads are generated by the interactions of the wheels with 57 the rails. The interaction between the flanges of a train wheel with the railhead is among these 58 interactions. The gauge length tolerances for normal speed railways are typically specified as +10 mm 59 (+0.39 in) and -3mm (-0.12 in) (1). The gauge length tolerances for high speed railways built according to 60 the international standards are typically specified as +2mm (+0.079 in) and -1mm (-0.039 in) (1). The 61 desired contact condition between the wheel and the rail is between the top surface of the railhead and the 62 bottom conical face of the rail as shown in figure 1 (2). However, due to gauge length variations, the 63 interactions can take place between the flange and the inner side of the railhead as well, thereby 64 increasing the number of possible points of contact between the wheel and the rail, provoking lateral and 65 frictional forces, disturbing the ride comfort and ride stability and inducing wheel and rail abrasions (3). 66

67

FIGURE 1 Train wheel on a railhead. 68


Bezgin, O.N. 3

Due to the conic design of the wheels, a laterally displaced axle with respect to the motion of a train is 69 realigned along the rails. When the track gauge length is larger than the design value, sidesway motion of 70 the train travelling at a certain speed, which is also known as the “Klingel Movement” (1, 4), is provoked. 71 As the train motion acquires a motion lateral to the direction of travel, the train begins to oscillate 72 between the rails. These oscillations may seize and reach a point of stability under constant speed and 73 linear train motion. However, as the train accelerates into higher speeds or moves along curvatures, these 74 oscillations increase accordingly and the oscillating behavior develops into a recurring impact behavior 75 when the flanges of the wheels begin to hit and rebound from the inner sides of the railheads known as the 76 “Hunting Movement” that induces lateral impacts onto the railheads (1, 4). The laterally induced motion 77 generates a rotational behavior along the z – axis as well that is perpendicular to the x-y plane. Such 78 rotations around the z-axis develop an angle “α” referred to as the “angle of attack” (5) along which the 79 wheels tend to cut into the rail rather than smoothly roll along the rail. The laterally induced motions are 80 represented through a sketch presented in figure 2. If the gauge lengths are larger than the intended 81 values, these oscillations increase, which induce discomfort to the passengers before transforming into 82 impact behavior. The coincidence of the developed frequency content of these increasing oscillations to 83 the natural frequencies of the rolling stock reduces the stability of the train motion. If the gauge lengths 84 are smaller than their design value, these oscillations quickly develop into producing forces due to lateral 85 impacts that is increased by the developed angle α, developing friction forces at the instant of impacts. 86 87 88

89 90

FIGURE 2 Lateral motion and rotation of an axle along the perpendicular axis. 91 92

The outcome of such lateral train behavior is either the high speed trains have to be operated at lower 93 speeds to prevent this behavior (6) or the trains are operated with an increased risk of track and wheel 94 abrasion or derailment. A gauge length smaller than the design value can increase the rate of 95 development of friction between the flange and the railhead thereby increasing the probability of wheel 96 and rail abrasion and derailment of a train travelling at a certain speed (7, 8, 9). 97 98 DIMENSIONAL ASPECTS OF PRESTRESSED CONCRETE SLEEPER DESIGN 99 Shrinkage and creep are two characteristics of concrete that result in changes in the dimensional aspects 100 of a concrete structure (10). In prestressed concrete design, determination of prestressing losses is the 101 primary concern since under extremely dry conditions these losses could be as high as 40%. Prefabricated 102 high speed railway sleepers are prestressed structural elements that need to be evaluated within a scale 103 that is much more precise than other structural elements. Therefore, along with the prestressing losses, the 104 changes in the sleeper dimension must become a concern as well. The locations for the attachments of 105 rails onto the sleepers are predetermined within millimeters and the value of the distance between the 106 attached rails of a sleeper is set within a positive and a negative tolerance value, the precision of which is 107

x

y

z

α


Bezgin, O.N. 4

inversely proportional to the design speed of the train. International high speed railway structures with a 108 shoulder length of 1813 mm (71.4 in) and a gauge length of 1435 mm (56.5 in) are accepted within 109 tolerances of +2mm (+0.079 in) and -1mm (-0.039 in). The allowed positive tolerance is to limit the 110 excessive lateral vibration which could also lead to derailment and the allowed negative tolerance is to 111 limit the rail and wheel flange abrasion and the development of frictional forces that could also lead to 112 derailment. The sleepers are designed and produced with established locations for rail attachments. In 113 terms of production and construction efficiency, the sleepers are rapidly cast, cured and demolded within 114 1 to 3 days. 115

Even though a prefabricated prestressed high strength concrete sleeper attains the design 116 strength and durability requirements within a month, its dimensional values have not yet 117 stabilized due to the ongoing creep and shrinkage the majority of which lasts up to a year. The 118 ambient humidity values along a proposed high speed line have a significant effect on their creep 119 and shrinkage (10). The climatic conditions along a proposed route is unique and two routes that 120 are identical in terms of geographical aspects, geological aspects, axle loads, train speed and 121 operational frequencies is likely to require a unique mechanical and dimensional design for the 122 sleepers due to the effects of climatic variances. The international high speed railway experience 123 up to date have been mainly located in the Northern Hemisphere such as the Shinkansen of Japan, 124 TGV of France, ICE of Germany, Acela Express of United States and AVE of Spain. 125

The climatic conditions; especially in Northern Europe and Japan where a major portion 126 of the internationally applied design experience of high speed railways is located, are mainly 127 humid, which may not apply for projects located within arid climates. It is determined 128 analytically and experimentally that the time dependent contractions of prestressed concrete 129 sleepers in countries where the climate is humid are within the allowed tolerance values. 130 However, it was determined for the Ankara-Konya High Speed Railway that under the existing 131 and projected local climatic conditions, the allowed tolerance for the sleepers had the potential to 132 be surpassed within 2 months of production and was estimated to reach a level that was twice the 133 allowed value within 1 year. 134 135 THE ANKARA – KONYA HIGH SPEED RAILWAY 136 Ankara - Konya High Speed Railway Line; operated by the Turkish State Railroads (TCDD), connects the 137 capital with a population of 4.8 million and the sixth largest city of the country with its 2.1 million 138 inhabitants. It is the second high speed railway in the country after the Ankara – Eskişehir High Speed 139 Railway which was completed in 2009 and the first railway to be constructed in the region outside the 140 European Union. The construction of the double line, 212 km (131 mile) long high speed railway, which 141 was designed for 22,5 Ton (49.6 kip) axle load commuter trains with a maximum train speed of 250 142 km/hour (155 mile/hour), began in late 2008 and the line became operational in late 2011 (11). It was a 143 national high speed railway that was designed and constructed by a national company. An approximate 144 count of 700,000 type B70 class sleepers was produced for the project. The prestressed pretensioned 145 sleepers shown in figure 3 were produced in the city of Konya within confined and well insulated quarters 146 by the circular production method, which is also known as the “caroussel method”, where the consecutive 147 prestressed concrete production steps were applied to the mold sets that moved from one station to 148 another along a closed loop until the sleeper was removed and the molds were returned back to the initial 149 station where they were cleaned and the prestressing wires with the required attachments were inserted. 150


Bezgin, O.N. 5

151

FIGURE 3 B70 type, high performance concrete sleepers. 152 153

EVALUATION OF THE LOCAL CLIMATE AND HUMIDITY CONDITIONS 154

The Central Anatolia region of Turkey, where the line is located is an arid region within a climatic shift 155 that is projected to end with desertification. The expected level of losses due to shrinkage and creep is 156 15% to 20% higher than the expected losses for the high speed railway lines located in Northern Europe 157 and the sleepers were expected to be exposed to arid conditions estimated to produce dimensional 158 contractions that had the potential to surpass the allowable tolerances. At the early sleeper design stages, 159 it was determined that under the humidity conditions equal and below RH = 50%, the sleepers could 160 contract more than 2 mm within a year (11). This was an issue that was not considered for the high speed 161 railway projects located within humid conditions since the amount of contractions due to elastic 162 shortening, creep and shrinkage are typically absolved; sometimes unknowingly, within the 1 millimeter 163 negative tolerance. However for this project, the prospect of derailment with high speeds acting on a 164 gauge length with excessive contractions under the arid conditions was evaluated as a risk and remedial 165 actions were reflected on the production dimensions of the sleeper shoulder length. 166

During the design of the sleepers, a meteorological investigation was conducted and the drought 167 characteristics of the country were determined from the relevant sources (12). The drought characteristics 168 of the country during 1971-2000 and for 2007 are shown in figure 4 and figure 5. The cities of Ankara 169 and Konya are indicated with bold spots on the maps. The current level of humidity along the proposed 170 route was determined to fluctuate annually between RH: 50% and RH: 85% with high levels of humidity 171 between November and March, followed by an abrupt fall in humidity to arid levels between June and 172 November. The expected level of decline in the near future was RH: 50% and perhaps below. 173


Bezgin, O.N. 6

174

FIGURE 4 Annual mean drought levels between 1971 and 2000. 175

176

FIGURE 5 Annual mean drought levels for 2007. 177

ANALYTICAL EVALUATION OF THE TIME DEPENDENT CHANGES OF THE SLEEPER 178

The B70 sleepers were designed with C65 grade concrete with design strength of 65 MPa (9.400 psi) and 179 a mean strength of 73 MPa (10.580 psi). The concrete was cemented by CEM 52,5R type, high early 180 strength white cement, with a cement compressive strength of 52,5 MPa (7.614 Psi). The concrete sleeper 181 was pretensioned under a load of 350 kN (78.7 kips) applied through count-8, 7 mm (0.276 in) diameter 182 high strength wires via plates anchored 30 mm (1.38 in) deep within the ends of the sleepers. Each sleeper 183 was 260 cm (8.5 ft) long and weighed 280 kg (617 lb). The sleeper shoulder length was specified at 1813 184 mm (71.4 inch) with a tolerance value of +2mm and -1mm (+0.787 inch and -0.394 inch) (11). 185

Following the design of the concrete mixture for the sleeper and structural design of the 186 prestressed section, the time related dimensional changes due to creep and shrinkage of the sleeper were 187 analytically evaluated according to the Eurocode – 2 concrete standards (13, 14). 188 189


Bezgin, O.N. 7

Creep 190

Eurocode – 2 includes a detailed qualitative and quantitative evaluation of creep (13).The “time 191 dependent creep coefficient” is represented as a function of the “base creep coefficient” and the “time 192 dependent variation of creep”. 193

194 Time dependent creep coefficient: ( ) ( ) ( ) (1) 195

Base creep coefficient: ( ) ( ) (1.1) 196

Time dependent variation of creep: ( ) (( )

( ))

(1.2) 197

“Base creep coefficient” is represented as a function of “effect of relative humidity on creep”, “effect of 198

concrete strength on creep” and “effect of concrete age during the transfer of prestressing on creep”. 199

Effect of relative humidity on creep: [

√ ] (1.1.1) 200

[

]

and [

]

(1.1.1.1) 201

(1.1.1.2) 202

Effect of concrete strength on creep: ( )

√ (1.1.2) 203

Effect of concrete strength during prestressing transfer on creep: ( )

( )

(1.1.3) 204

Where, ho = Representative dimension of the concrete strength, Ac = Concrete cross section area, u = 205 Concrete cross section perimeter, β(fcm) = Effect of concrete strength on creep constant, fcm= Mean 206 concrete strength at the end of 28 days, β(to) = Effect of concrete age at the instant of loading on the 207 creep constant. 208

The time dependent variation of creep is a function of, humidity, representative cross section 209 dimension, concrete strength and concrete age. 210

[ ( ) ] (1.2.1) 211

[

]

(1.2.1.1) 212

(

)

(1.2.2) 213

∑ (

[ ( )] )

(1.2.2.1) 214

{

[ ]

[ ] [ ]

} (1.2.2.2) 215

Where, βH = Constant based on relative humidity, representative element dimension and concrete 216 strength, t0= Age of concrete based on the ambient temperature during prestressing (days), t= Age of 217 concrete (days), tT = Effective concrete age, T(∆ti)=Temperature in terms of centigrade within ∆ti, 218 ∆ti =Number of days when the temperature is T, ho= Representative dimension of the concrete element, 219


Bezgin, O.N. 8

fcm=Average concrete cylinder strength, fck=Characteristic concrete cylinder strength, R = Rapidly 220 hardening high early strength cement, N = Normally hardening cement, S = Slowly hardening cement. 221

The total creep shortening is determined by the multiplication of the elastic shortening by the 222 time dependent creep coefficient. 223

Total shortening due to creep: ( ) (

) (2) 224

225 An analytical evaluation of the expected contraction within the shoulder length of the sleeper was 226

conducted. The sleeper is a non-prismatic section. To this end, the sleeper shoulder length is represented 227 within three regions as shown in figure 6. The representative sections within these regions are: section A, 228 B and C, which are the sleeper rail region, the sleeper transition region and the sleeper central region 229 respectively. The cross sections are presented in figure 7. 230

231 232

FIGURE 6 Regions along the shoulder. 233

234 235

FIGURE 7 Sleeper cross sections. 236

The cross section areas of the sections A, B and C are AA = 49.233 mm2 (7.631 in

2), AB = 37.400 mm

2 237

(5.797 in2) and AC = 33.425 mm

2 (5.181 in

2), their perimeters are UA = 884 mm (34.8 in), UB = 780 mm 238

(30.7 in), UC = 737 mm (29 in) and h0A = 111 mm (4.4 in), h0B = 96 mm (3.8in), h0C = 91 mm (3.6 in). 239 The creep characteristics for RH=75%, which was the mean indoor humidity level reported for the 240 experimental measurement presented in the next section, can be evaluated through the numerical analysis 241 presented in Appendix A. 242 243 244 245 246 247


Bezgin, O.N. 9

Shrinkage 248 The shrinkage characteristics are presented in Eurocode-2 (13). Total “concrete shrinkage” is related to 249 “drying shrinkage” and “autogenous shrinkage”. 250

Drying Shrinkage: 251

Drying shrinkage: ( ) (3) 252

( ) ( )

( ) √ (3.1) 253

[( ) (

)] (3.2) 254

[ (

) ] (3.2.1) 255

fck = 65 MPa (Characteristic cylinder strength), fcm= Average concrete strength = 73 MPa, fcm0=10 MPa, ts 256 =Time of cure termination (ts =0,5 Days), t = Time (Days), RH =Relative humidity (%) and RHo =100% 257 258

{

} , {

} (3.2.2) 259

Autogenous Shrinkage: 260

Autogeneous shrinkage: ( ) ( ) ( ) (4) 261

( ) ( ) (4.1) 262

( ) ( ) (4.2) 263

264

Estimate of Total Elastic and Time Dependent Contractions 265

The evaluation of the development of creep and shrinkage can be determined through the use of 266 spreadsheets. Tables 2 and 3 that are presented in Appendix A, show the development of creep and 267 shrinkage along the shoulder of the sleeper under a relative humidity of 75%. Such tables are constructed 268 for varying levels of relative humidity and the resultant variations in time summarized in a single figure. 269 Figure 8 shows that under relative humidity conditions of 80% and less, the expected time dependent 270 contractions within the shoulder, exceed the 1 mm negative tolerance within 1 month of the application of 271 prestressing. 272


Bezgin, O.N. 10

273

FIGURE 8 Shoulder contraction estimates in time with relative humidity. 274

EXPERIMENTAL EVALUATION OF ELASTIC AND TIME DEPENDENT CONTRACTIONS 275

Prior to the production of the sleepers for the Ankara - Konya high speed railway, the sleeper molds were 276 produced according to the numerically estimated time dependent contractions for the designed sleeper. 277 The numerical estimate for the annual time dependent dimensional change for RH = 50% was 2 mm 278 which was 100% higher than the allowed negative tolerance of -1mm. Based on the estimates and the 279 allowed tolerances of +2 mm and -1 mm, the molds were dimensioned for a shoulder length of 1814,5 280 mm that was 1,5 mm longer than the required design value of 1813 mm. In order to test the numerical 281 contraction estimates, 24 sleepers from the daily production of May 27, 2009 was selected and measured 282 indoors, with a calibrated digital gauge under a reported indoor humidity mean level of RH=75%.Figure 9 283 shows the shoulder length measurements of an array of sleepers with the calibrated digital gauge. 284

The first measurements were conducted 1 day after demolding at 08.00 AM of the 285 following day when the external humidity was at its daytime high, followed by a monthly 286 measurement and a final measurement at the end of 60 days at the early day hours before the 287 routine daily shift progressed. The measurement values are presented in figure 10. 288 289

0,47

1,11

1,53

2,03

2,35

0,84 0,96

1,12 1,23

0,00

0,20

0,40

0,60

0,80

1,00

1,20

1,40

1,60

1,80

2,00

2,20

2,40

2,60

0 30 60 90 120 150 180 210 240 270 300 330 360 390

Sho

rte

nin

g (m

m)

Time (Days)

RH 10% RH 20% RH 30%

RH 40% RH 50% RH 60%

RH 70% RH 80% RH 90%


Bezgin, O.N. 11

290 291

FIGURE 9 Measurement of sleeper shoulder length. 292

293 294

FIGURE 10 Time dependent contractions for the collected 24 samples. 295 296 The measurements were statistically evaluated based on the Gauss Distribution. The mean values were 297 determined, followed by the determination of the variance and the values for the 95% and the 99% 298 confidence intervals. After the comparison of the mean, the 95% CI value and the 99% CI value with the 299 theoretical estimates, it was determined that the numerical estimates was approximately in 100% 300 agreement with the 99% CI values. The numerical evaluations of the time dependent contractions over-301 estimated the measured 2 month mean contraction values by 30% and the 2 month 95% CI value by 14%. 302 The statistical values are presented in figure 11. 303

The 2 mm shoulder analytical contraction estimate for RH=50% was revised to 1.5 mm (0.06 in) 304 and the shoulder production length was set at 1814.5 mm (71.44 mm). The random measurements from 305 the railroad gauge length taken a year after panel placement yielded values 1435 mm ± 0.3 mm. (56.5 in ± 306 0,012 in). 307

308

0,00

0,20

0,40

0,60

0,80

1,00

1,20

1,40

0 2 4 6 8 10 12 14 16 18 20 22 24

Sho

rten

ing

(mm

)

Sample number

1 day

1 month

2 months


Bezgin, O.N. 12

309

FIGURE 11 Comparison of predicted and measured time dependent contractions 310 . 311

CONCLUSIONS 312

Although a dimensional shift on the scale of millimeters could be considered to be irrelevant for an 313 ordinary civil engineering concrete structure, for a high speed railway; the components of which should 314 be considered more as a mechanical piece rather than a structural element, it is an important design 315 variation that could jeopardize the high speed operations of the railway or cause structural damage and 316 failure. The issue of time dependent gauge length shortening was raised for the Ankara-Konya High 317 Speed Railway, which was not previously addressed for other high speed railway lines. The presented 318 numerical analysis and the measurements provide evidence for the initiative to maintain the dimensional 319 quality of the high speed railway sleepers in arid climates. 320

The measurement of the dimensional changes along the sleepers were not pursued 321 beyond two months based on an understanding that estimates were indicative and the allocations 322 for the expected time dependent dimensional changes were sufficient. 323

In conclusion, the following are proposed for the design of a prestressed concrete sleeper 324 destined for use in a high speed railway proposed within an arid climate: 325

1. The existing and the projected humidity levels for the proposed locations of the high 326 speed railways must be established. 327

2. The shrinkage and creep characteristics of the sleeper design concrete must be 328 experimentally determined along with its strength and stiffness parameters. 329

3. Under RH values equal and below 75%, the -1 mm shoulder tolerance is determined to 330 be lost in two months to time dependent contractions 331

4. The mold shoulder length must be dimensioned to include the estimated time 332 dependent contractions for projects proposed in countries with arid and desert conditions. 333

5. Further studies for wheel and rail interaction is required. 334 335 336 337 338

0,47

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,9

1

1,1

1,2

1,3

1,4

0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1 1,1 1,2 1,3 1,4

Me

as

ure

d c

on

rra

cti

on

s (

mm

)

Predicted contractions (mm)

ElasticMean95% CI99% CILine of equality

1 day

1 month

2 months


Bezgin, O.N. 13

APPENDIX A 339 Curing time for the sleepers: 7 hours (0.3 days), waiting time until transfer of prestress to the sleepers: 5 340 hours (0.2 days) 341

(

[ ] )

(

[ ] )

1.35 342

(

)

(

)

(High early strength cement: CEM 52,5 R) 343

( )

(High early strength cement: CEM 52,5 R) 344

( )

√

√

[

] [

] and [

] [

] 345

[

√ ] [

√ ] , and 346

347

( ) ( )

, 348

349

350 TABLE 1 Elastic Shortening Values Along the Sleeper 351

352

Elastic shortening within the shoulder

Region on sleeper

A B C

E at transfer (MPa) 33.000 33.000 33.000

F (kg) 35.714 35.714 35.714

Cross section (cm2) 334 374 492

Cross section perimeter (mm) 400 600 814

Δ(mm) 0,127 0,17 0,175

Total Δ (mm) 0,47

353 354 For RH = 50%: 355 Δs-A= 1.623 * 0.13 =0.21 mm, 356 Δs-B = 1.616 * 0.18 =0.29 mm, 357 Δs-C = 1.596 * 0.18 =0.28 mm 358 Total expected creep for RH=50%, ΔSt = 0.76 mm. 359 360

361

362

363


Bezgin, O.N. 14

TABLE 2 Creep and Shrinkage Between the Shoulders for RH = 75% 364

365

ACKNOWLEDGEMENTS 366

I would like to respectfully acknowledge Dr. Ersin Arioglu of Yapı Merkezi Inc. and Prof. Dr. Mustafa 367

Karasahin of İstanbul University. 368

REFERENCES 369 1. Esweld, C. Modern Railway Track, Second Edition, Delft University of Technology, Zaltbommel, 370

2001 371 372

2. Samuels, M.J., (2008), The Freight Railroad Renaissance, Transportation Infrastructure 373 374

3. Presle, G., Hanreich, W., and Paul Mittermayr. Austrian Track Testing and Recording Car EM 250 375 Source for Wheel-Rail Interaction AnalysisIn Transportation Research Record: Journal of the 376 Transportation Research Board, No.1713, Transportation Research Board of the National 377 Academies, Washington, D.C., 2000, pp.22 – 28 378 379

4. Güngör, E. Railways, First Edition, Birsen Yayinevi, Istanbul, 1999 380 381

5. Wu, H., Shu, X., Wilson, N.. TCRB Report No.71, Track-Related Research Volume 5: Flange Climb 382 Derailment Criteria and Wheel/Rail Profile Management and Maintenance Guidelines for Transit 383 Operations, Transportation Technology Center, Inc. (TTCI), 2005 384

385 6. López-Pita, A., and Robusté F. Effect of Very High-Speed Traffic on the Deterioration of Track 386

Geometry. In Transportation Research Record: Journal of the Transportation Research Board, No. 387 1825, Transportation Research Board of the National Academies, Washington, D.C., 2003, pp.22-27. 388

7. Liu, X., Saat, M., and Barkan, C. Analysis of Causes of Major Train Derailment and Their Effect on 389 Accident Rates. In Transportation Research Record: Journal of the Transportation Research Board, 390

Interval Days A B C A B C A B C

to (end of cure) 0 - - - - - -

1 Day 1 17,3% 17,2% 16,8% 0,00009 0,00008 0,00006 0,04 0,05 0,05 0,13

1 Month 30 46,9% 46,6% 45,6% 0,00024 0,00021 0,00016 0,10 0,13 0,13 0,35

2 Months 60 56,5% 56,1% 55,0% 0,00029 0,00026 0,00019 0,12 0,15 0,15 0,42

1 Year 365 82,0% 81,7% 80,7% 0,00042 0,00037 0,00028 0,17 0,22 0,23 0,62

5 Years 1825 95,0% 94,8% 94,5% 0,00049 0,00043 0,00032 0,20 0,26 0,26 0,72

40 Years 14600 100,0% 100,0% 100,0% 0,00051 0,00046 0,00034 0,21 0,27 0,28 0,76

Interval Day A B C A B C A B C

to (end of cure) 0 - - - - - -

1 week 1 40% 40% 40% 0,00012 0,00012 0,00012 0,05 0,07 0,10 0,22

1 Month 30 62% 60% 54% 0,00019 0,00018 0,00016 0,07 0,11 0,13 0,31

2 Months 60 73% 71% 66% 0,00022 0,00021 0,00020 0,09 0,13 0,16 0,38

1 Year 365 93% 92% 90% 0,00028 0,00028 0,00027 0,11 0,17 0,22 0,50

5 Years 1825 98% 98% 98% 0,00029 0,00029 0,00029 0,12 0,18 0,24 0,53

40 Years 14600 100% 100% 100% 0,00030 0,00030 0,00030 0,12 0,18 0,24 0,54

Percentage of regional total Strain Shortening (mm)

Time Region on sleeper

TOTAL

CREEP

Time Region on sleeper

TOTAL

Percentage of total Strain Shortening (mm)SHRINKAGE


Bezgin, O.N. 15

No. 2289,Transportation Research Board of the National Academies, Washington, D.C., 2012, pp. 391 154–163. 392 393

8. Liu, X., Barkan, C., and Saat, M. Analysis of Derailments by Accident Cause Evaluating Railroad 394 Track Upgrades to Reduce Transportation Risk. In Transportation Research Record: Journal of the 395 Transportation Research Board, No. 2261,Transportation Research Board of the National 396 Academies, Washington, D.C., 2011, pp. 178–185. 397 398

9. Barkan, C., Dick, C., and Anderson, R. Railroad Derailment Factors Affecting 399 Hazardous Materials Transportation Risk. In Transportation Research Record: Journal of the 400 Transportation Research Board, No. 1825, Transportation Research Board of the National 401 Academies, Washington, D.C., 2003, pp.64 - 73 402 403

10. Idriss, R., and Solano, A. Effects of Steam Curing Temperature on Early Prestress Losses in High-404 Performance Concrete Beams. In Transportation Research Record: Journal of the Transportation 405 Research Board, No.1813, Transportation Research Board of the National Academies, Washington, 406 D.C., 2002, pp.218 – 228 407

408 11. Bezgin O.N., Ankara – Konya High Speed Railway Sleeper Design, 2008, İstanbul 409

410 12. General Directorate of Turkish Meteorological Studies, MGM, Drought Maps, 1971 – 2000, 2007 411

412 13. Eurocode – 2 – Design of Concrete Structures, November 2002 413

414

14. Bezgin O.N., Yilmaz V., Rayton Sleeper Contractions, 2009, İstanbul 415


design evaluations for the time dependent …docs.trb.org/prp/14-0322.pdf · bezgin, o.n. 1 1 trb 2...

Documents