Wheel wear of a two axle freight vehicle - KTH/Menu/general... · Wheel wear of a two axle freight...

Wheel wear of a two axle freight vehicle

Carlos Casanueva, Sebastian Stichel, Per-Anders Jönsson Division of Rail Vehicles, KTH 17th Nordic Seminar on Railway Technology October 3, 2012

Layout

• Project description • Wear modelling • Experimental results • Validation and discussion • Conclusions and further work

2012-10-02 KTH - Spårfordon - Division of Rail Vehicles 2

Objectives

• Design of a new wheel profile for freight wagons - Reduce maintenance costs caused by wear and fatigue - Reduce the effect of low frequency instability

• Analyse the influence of freight vehicle modelling on uniform wear and RCF - Increased axle loads - Energy dissipation elements

2012-10-02 KTH - Spårfordon - Division of Rail Vehicles 3

Wear Modelling

• Archard’s wear law

2012-10-02 KTH - Spårfordon - Division of Rail Vehicles 4

HsNkVw =

Wear Modelling

• Archard’s wear law

• Define Operational Case • Perform dynamic simulations • Calculate wheel wear • Update wheel profile

2012-10-02 KTH - Spårfordon - Division of Rail Vehicles 5

HsNkVw =

Stop when desired mileage is reached

Reference Case: Trätåg Timber Wagons

• Laaps wagons, Unitruck running gear

2012-10-02 KTH - Spårfordon - Division of Rail Vehicles 6

Reference Case: Trätåg Timber Wagons

2012-10-02 KTH - Spårfordon - Division of Rail Vehicles 7

LS

LOM

VO

GMS

BLG

HLF

LSE

HFJ

GÄ SUR

Experimental results: flange height hf

2012-10-02 KTH - Spårfordon - Division of Rail Vehicles 8

0 2 4 6

x 104

28

29

30

1r R2=0.6577

0 2 4 6

x 104

28

29

30

1l R2=0.4484h f (m

m)

0 2 4 6

x 104

28

29

30

2r R2=0.83296

mileage (km)0 2 4 6

x 104

28

29

30

2l R2=0.71513

mileage (km)

h f (mm

)

1l

2l 2r

1r

Experimental results: flange thickness tf

2012-10-02 KTH - Spårfordon - Division of Rail Vehicles 9

1l

2l 2r

1r

0 2 4 6

x 104

26

28

30

32

1r R2=0.60143

0 2 4 6

x 104

26

28

30

32

1l R2=0.46515t f (m

m)

0 2 4 6

x 104

26

28

30

32

2r R2=0.49669

mileage (km)0 2 4 6

x 104

26

28

30

32

2l R2=0.529

t f (mm

)

mileage (km)

Experimental results: flange inclin. qr

2012-10-02 KTH - Spårfordon - Division of Rail Vehicles 10

1l

2l 2r

1r

0 2 4 6

x 104

6

8

10

121r R2=0.68258

0 2 4 6

x 104

6

8

10

121l R2=0.57731

q r (mm

)

0 2 4 6

x 104

6

8

10

122r R2=0.75477

mileage (km)0 2 4 6

x 104

6

8

10

122l R2=0.59601

q r (mm

)

mileage (km)

Experimental validation

2012-10-02 KTH - Spårfordon - Division of Rail Vehicles 11

Experimental validation

2012-10-02 KTH - Spårfordon - Division of Rail Vehicles 12

Experimental validation

2012-10-02 KTH - Spårfordon - Division of Rail Vehicles 13

Experimental validation

2012-10-02 KTH - Spårfordon - Division of Rail Vehicles 14

Experimental validation

2012-10-02 KTH - Spårfordon - Division of Rail Vehicles

Contact geometry at the switch

2012-10-02 KTH - Spårfordon - Division of Rail Vehicles 16

17 18-19-20

21-22 23

Switch simulation

2012-10-02 KTH - Spårfordon - Division of Rail Vehicles 17

-0.06 -0.05 -0.04 -0.03 -0.02 -0.01 0 0.01 0.02 0.03 0.04

0

100

200

300

400

500

600

700

posw (m)

Dis

sipa

ted

ener

gy(J

/m)

cpt11l

cpf11l

cpt11r

cpf11r

17

A

B C

D

New profile Worn profile

Switch simulation – wear

• Average: 3.14km per switch

2012-10-02 KTH - Spårfordon - Division of Rail Vehicles 18

-0.06 -0.04 -0.02 0 0.02 0.04 0.060

0.5

1

1.5

2x 10

-6

dz(m

)

y(m)

wear/(km/turnuot)wear/turnout S1002-newwear/turnout S1002-worn

Full simulation

• Laden • 3.14km simulation • Switch simulation • Repeat until laden run is ended

• Unladen • 3.14km simulation • Switch simulation • Repeat until unladen run is ended

• Laden • ...

2012-10-02 KTH - Spårfordon - Division of Rail Vehicles 19

Full simulation

• No way, simulations are: - c.a. 100km per wear step - c.a. 500 wear steps - c.a. 15 days computation.

• Switch simulation - c.a. 45 min. ~c.a. 22.5 days. - Non-smooth wheel profile will increase simulation time

2012-10-02 KTH - Spårfordon - Division of Rail Vehicles 20

Conclussion

• Wheel wear on freight vehicles cannot be studied considering only track geometry - Heavy wear at the tread end and flange top is caused by

switch geometry - Block brakes will cause additional wear at the tread

• Wear pattern at the tread end could be detected with Archard’s wear model but not with Tγ model

• Promising preliminary results

2012-10-02 KTH - Spårfordon - Division of Rail Vehicles 21

Further work

• Switch simulation - Straight run - Unladen run - Worn switch geometry

• New vehicle – Y25 bogies • Block brake influence

2012-10-02 KTH - Spårfordon - Division of Rail Vehicles 22

Wheel wear of a two axle freight vehicle

Carlos Casanueva, Sebastian Stichel, Per-Anders Jönsson Division of Rail Vehicles, KTH 17th Nordic Seminar on Railway Technology October 3, 2012 Thank you for your attention carlosc@kth.se