TC202 WS on Railroad Geotechnics 2015 土木学会 … TC202...TC202 WS on Railroad Geotechnics 2015土木学会第16回舗装工学講演会 Faculty of Engineering, Hokkaido University

土木学会第16回舗装工学講演会TC202 WS on Railroad Geotechnics 2015

1Faculty of Engineering, Hokkaido University

Synergistic effects of moving wheel loads and water content change on

mechanical behavior of substructures at ballasted track

TC202 Workshop on Railroad Geotechnicsat XVI ECSMGE

13 September 2015, Edinburgh, UK

Tatsuya ISHIKAWAHokkaido University, Sapporo, Japan


1. Background of Research

2. Objective of Research

3. Test Apparatuses and Test Samples

4. Results of Multi‐ring Shear Tests on Gravel

5. Results of Multi‐ring Shear Tests on Sand

6. Proposal of Experimental Formula

7. Application of Advanced Soil Testing to Design Standard

8. Conclusions

Table of contentsTC202 WS on Railroad Geotechnics 2015









8. Conclusions


Background of research

4

For saving maintenance costs of transportation facilities at road and railway, it is important to elucidate mechanical behavior of granular roadbed subjected to traffic loads. Sleeper

Wheel

Rail

BallastBallastRoadbed

Running direction

Conventional studies

Following loading tests have been performed in Japan; Model test : Element test :

a

a

rr

Triaxial testfixed-point loading test

Moving load

Bituminous layer

Base courseBase course

Subgrade

Tire

TC202 WS on Railroad Geotechnics 2015

Effect of moving wheel loads

5

Literature Review

Moving wheel loads give a large residual settlement as compared with fixed-point cyclic loads.

RailSleeper

Railroad ballast

fixed-point loading

Moving wheel loading

0.0 0.5 1.0 1.5 2.0 2.5 3.0

0.0

0.5

1.0

1.5

2.0

2.5N=200

N=1

N=100N=10

Rai

l Sea

t For

ce (k

N)

Vertical Displacement (mm)

0.0 0.5 1.0 1.5

0.0

0.5

1.0

1.5

2.0

2.5Poorly-graded

BallastN=200N=100N=10

N=1

Rai

l Sea

t For

ce (k

N)

Vertical Displacement (mm)

Wheel


KEY POINTS Can fixed-point loading tests

simulate actual traffic loads ?

Ishikawa & Sekine 2007

Moving wheel load

6

fixed-point loadingIncrease & decrease of loads

Moving wheel loading+ Principal stress axis rotation

Moving wheel loading testMeasured Ballast Pressure

-800 -400 0 400 800

0.0

0.5

1.0

1.5

2.0

-0.4-0.3-0.2-0.10.00.10.20.30.4

Vertical load

1/5B BallastP=4kN

No.8 Sleeper

Ver

tical

load

, P r

(kN

)

Wheel position (mm)

Shear load

Shea

r loa

d , Q

r (kN

)

1 2 3 4 5 6 7 8 9 10 1511 12 13 14


Normal component remains compression side through loading, while shear component changes the sign according to the position of a loading wheel. Principal stress axes rotate inside railroad ballast during train passage.

Change in water content

7

KEY POINTS Does rainfall affect settlement of

railroad ballast ?


In past studies, model and element tests have been performed mainly under air-dried condition.Deformation modulus of coarse granular materials decreases with the increase in water content. (Coronado et al. 2005, Ekblad & Isacsson 2006)

Literature Review

Fouled ballast New ballast

Full-scale model test

Itou et al. (2014) performed cyclic loading tests of full-scale model ballasted track before and after spraying water on the surface of railroad ballast (10 l/m2).

Results of full-scale model tests

8

Influence of water content

Plastic deformation and elastic deformation of ballasted track during cyclic loading suddenly increase due to water spray. Wetting causes the sample to decrease the shear stiffness, thereby increasing residual settlement.

Residual settlement (Plastic behavior) Settlement amplitude (Elastic behavior)

0 200,000 400,000 600,000 800,0008

6

4

2

0

Res

idua

l set

tlem

ent o

f sle

eper

(mm

)

Number of loading cycles

New ballast Fouled ballast

Water spray

0 200,000 400,000 600,000 800,000

3

2

1

0

Settl

emen

t am

plitu

de o

f sle

eper

(mm

)

Number of loading cycles

New ballast Fouled ballast

Water spray






4. Results of Multi-ring Shear Tests on Gravel

5. Results of Multi-ring Shear Tests on Sand



8. Conclusions


To evaluate the effects of water content on the mechanical behavior of granular materials subjected to moving wheel loads in terms of the cyclic deformation characteristics.

Objectives of Research

Research content Multi-ring shear tests of gravels and sand under various degrees of

saturation.• Fixed-point loading without principal stress axis rotation (FL)• Moving wheel loading with principal stress axis rotation (ML)

10

KEY POINTS Are results obtained from conventional loading tests credible ? How about effects of water content in moving loading tests ?

Research objective










8. Conclusions


Multi-ring shear test apparatus

12Outside

RingInside RingSpecimen

DDM (Axial load)

DDM (Torque load)

Displacement Transducer

Load Cell (Axial load)

Torque Transducer

Water Tank

Remarkable Features• It can evaluate the effects

of rotating principal stress axes under shearing on the deformation-strength characteristics of granular materials.

• It can perform monotonic and cyclic loading tests under various water contents and freeze-thaw history.

Specimen sizeWidth： 60mm

indside D=120mmoutside D=240mm

Height： 60 - 100mm

Test conditionTorsional simple shear with strain-control and stress-control method using direct drive motor (DDM)


Test Samples

Three types of test samples are employed to examine the mechanical behavior of railroad ballast and roadbed at ballasted tracks.To ensure experimental accuracy in terms of specimen size, ballast and gravel have one-fifth and one-fourth similar grading of ballast and base course materials used in Japanese railway and road, respectively. 13

1/5 ballast C-9.5 gravel


Toyoura sand

Sample ρs (g/cm3) ρdmax (g/cm3) ρdmin (g/cm3) D50 Uc Fcinitial (%)

1/5 ballast 2.70 1.650 1.353 8.11 1.52 0.0C-9.5 gravel 2.72 1.730 1.480 2.80 10.8 0.0Toyoura sand 2.65 1.648 1.354 0.18 1.30 0.0

Testing Method

14

Specimens were prepared by tamping a sample with prescribed water content, and a series of multi-ring shear tests was conducted by using fixed-point loading (FL) method and moving-wheel loading (ML) method.

Moving-wheel loading (ML)Fixed-point loading (FL)

Increase & decrease of loads like triaxialcompression test

σ1

σ1 σ1

σ1σ3 σ3

σ3 σ3

σ1

σ1

σ3σ3

0 20 40 60 80 100 120 1400

20406080

100120140160

-30

-20

-10

0

10

20

30

Shear Stress Shea

r St

ress

, a

(kPa

)

A

xial

Stre

ss,a (

kPa)

Time (sec)

Axial Stress

Increase & decrease of loads with principal stress axis rotation

0 20 40 60 80 100 120 1400

20406080

100120140160

-30

-20

-10

0

10

20

30

Shear Stress

Shea

r St

ress

, a

(kPa

)

A

xial

Stre

ss,a (

kPa)

Time (sec)

Axial Stress










8. Conclusions


Results of cyclic loading tests (1)

16

Influence of water content & loading method

Cyclic plastic deformation of wet gravel is more likely to increase with loading cycles than that of oven-dried gravel.The influence of water content on the cyclic plastic deformation for gravel appears more clearly in employing moving wheel loads.

Fixed-point Loading Moving-wheel Loading

0 50 100 150 200 250 300 350 4000.00.51.01.52.02.53.03.54.04.55.05.56.06.5

amax = 114.2 kPa Oven dried Sr = 19 % Sr = 33 % Sr = 48 %

Cum

ulat

ive

Perm

anen

tA

xial

Str

ain,

ap (%

)

Number of Loading Cycles, Nc

C-9.5 gravel

0 50 100 150 200 250 300 350 4000.00.51.01.52.02.53.03.54.04.55.05.56.06.5

amax = 114.2 kPa and amax = 30 kPa Oven dried Sr = 19 % Sr = 33 % Sr = 48 %

Cum

ulat

ive

Perm

anen

t A

xial

Str

ain,

ap (%

)


C-9.5 gravel

Increase & decrease of loads



17

Wetting causes the sample to decrease the shear strength and stiffness, thereby increasing cumulative permanent strain during cyclic loading. When rotational angle of principal stress axis increases with the increase in shear stress amplitude, cumulative permanent axial strain also increases.

Cyclic plastic deformation under different loading conditions and Sr

Influences of principal stress axis rotation on cyclic plastic deformation

Results of cyclic loading tests (2)Influence of water content & loading method


0 10 20 30 40 500

1

2

3

4

5

6

ML test FL test

Axi

al S

train

, a (

%)

Degree of Saturation, Sr (%)

a at Nc=200 cyclesamax=30.0 kPaConstant amax=114.2 kPa

0 5 10 15 20 25 300

1

2

3

4

5

6

amax = 114.2 kPaamax = 0 kPa



K0 = 0.30 Oven dried Sr = 19 % Sr = 33 % Sr = 48 %

Cum

ulat

ive

Perm

anen

tA

xial

Stra

in, ap (%

)

Maximum Rotational Angle ofPrincipal Stress Axis, max (deg.)









8. Conclusions


0 50 100 150 2000.0

0.1

0.2

0.3

0.4

0.5

0.6

Cum

ulat

ive

Axi

al S

train

, a (

%)


amax= 72.58kPa Air-dried w = 7.8 % w = 9.6 % w = 14 %

0 50 100 150 2000.0

0.1

0.2

0.3

0.4

0.5

0.6

Cum

ulat

ive

Axi

al S

train

, a (

%)


amax= 72.58kPa

amax= 18.12kPa Air-dried w = 7.8 % w = 9.6 % w = 14 %

Results of cyclic loading tests (1)

19

Influence of water content & loading method

As seen in the comparison for gravel, cyclic plastic deformation of sand in ML tests is larger than that in FL tests.Cyclic plastic deformation of wet sand is more likely to increase with loading cycles than that of air-dried sand.

Fixed-point Loading Moving-wheel Loading

Toyoura sand Toyoura sand

Increase & decrease of loads



20

Wetting of sand increases cumulative permanent strain during cyclic loading, regardless of loading method. Effects of water content on cyclic plastic deformation under moving-wheel loads differ depending on soil types.

Cyclic plastic deformation under different loading conditions and Sr

Influences of water content on cyclic plastic deformation of gravel and sand

Results of cyclic loading tests (2)Influence of water content & loading method


0 10 20 30 40 500.0

0.1

0.2

0.3

0.4

0.5

0.6

a at Nc=200 cyclesamax=18.12 kPaConstant amax=72.58 kPa

ML test FL test

Axi

al S

train

, a (

%)

Degree of Saturation, Sr (%)0 10 20 30 40 50

0

1

2

3

4

5

Toyoura sand C-9.5

Rat

io o

f axi

al st

rain

in M

L te

stto

axi

al st

rain

in F

L te

st , R s

Degree of Saturation, Sr (%)









8. Conclusions


Effect of principal stress axis rotation

22

MLa c

s FLa c

NR

N

0 40 80 120 160 2001

2

3

4

5

6

71/5A, H=60mm(a)max=80.0kPa

(a)max=12.8 kPa (a)max=25.5 kPa (a)max=38.3 kPa

Rat

io o

f axi

al st

rain

, Rs

Number of loading cycles, Nc

Ratio of axial strain Rs is defined as a ratio of axial strain in ML test εaML to the one in FL test εaFL as;

Rs is almost constant throughout cyclic loading, regardless of loading conditions.Average Rs increases with the decrease in (σa)max or with the increase in (τaθ)max.Effects of principal stress axis rotation can be expressed by the ratio of axial strain Rsave given by

0 10 20 30 401

2

3

4

51/5A Ballast(a)max=80.0kPa (a)max=155.6kPa (a)max=231.1kPa Approximation

Maximum shear stress, (a)max (kPa)

Ave

rage

ratio

of a

xial

stra

in, (R s

) ave

max

max

exp asave

a

R a


Proposal of Experimental Formula

23

Under the same experimental conditions, cumulative axial strain in ML test can be estimated from cumulative axial strain of FL test by using average Rs as follows.

Assumptions

Proposed simple experimental formula is effective in estimating the cumulative strain characteristics of coarse granular materials in the principal stress axis rotation field.

Applicability of proposed formula

max

max

exp aML FL FLa c save a c a c

a

N R N a N

0 50 100 150 2000

1

2

3

4

5

6 H=60mm 100mm(a)max= 80.0kPa, (a)max= 12.8kPa : (a)max= 155.6kPa, (a)max= 25.5kPa : (a)max= 231.1kPa, (a)max= 38.3kPa : Approximation :

Number of loading cycles, Nc

Axi

al st

rain

, ( a

) max

(%)


Note : a is a constant depending on water content.









8. Conclusions


Track condition of ballasted track

25


Train vibration and track irregularity

Design criteria of ballasted trackTrain vibration should be smaller than threshold limit value.

• Traveling safety• Riding comfort

Train vibration increases along with the increase in track irregularity.Train vibration can be estimated by

Precise prediction of track irregularity growth is important for rational design & maintenance works of ballasted track.

a V α: train vibrationV: train speedσ: track irregularity

Change in track irregularity with time

Design procedures of ballasted track

26


Rail settlement and track irregularity

Track irregularity is defined by the difference between rail settlement at rail joint and that at middle part.Standard deviation of track irregularity can be estimated by

Flow of design procedures for ballasted track

3 6m

31

3

m

2Rail joint Middle part

KEY POINTS : How do we calculate track irregularity using settlement? Assumption

Application of advanced soil testing

27


Results of advanced soil testing considering moving-wheel loads and water content change improve the calculation precision of rail settlement.

Calculation of rail settlement in Japanese design standard

Key Point

×Rsave









8. Conclusions


The synergistic effects of principal stress axis rotation and water content strongly influence cyclic plastic deformation of coarse granular materials.

Under the same experimental conditions, cumulative axial strain obtained from moving-wheel loading test can be estimated from that of fixed-point loading test by considering the effects of principal stress axis rotation and water content.

When developing a rational design standard of ballasted track for precisely predicting the mechanical behavior and evaluating the long-term performance, it is important to give a special consideration to the influence of water content, and principal stress axis rotation on the deformation-strength characteristics of granular materials.

Concluding Remarks

29


Thank You for Your Kind Attention



Documents

TC202 WS on Railroad Geotechnics 2015 土木学会 … TC202...TC202 WS on Railroad Geotechnics 2015土木学会第16回舗装工学講演会 Faculty of Engineering, Hokkaido University