Lecture 2

Download as pdf or txt
Download as pdf or txt
You are on page 1of 42

M.

GRDAMAND MOHAMMED
[email protected]

Lecture 2
Lecture Contents
 TRACKBED DETERIORATION / MEASUREMENT
 BALLAST PROPERTIES
 PERMANENT DEFORMATION WITHIN BALLAST
 BALLAST FOULING

2
TRACKBED DETERIORATION / MEASUREMENT

1. Longitudinal Track Profile

• Cost-effective maintenance can only be carried out if there


is sufficient good information about the condition of the
track.
• This is a complicated and developing area. However, the
principal tool is the New Measurement Train (NMT).
• The NMT uses a system of lasers to obtain a continuous
profile of the track under load.
• This is expressed as the standard deviation of the track level
relative to a moving mean level.

3
TRACKBED DETERIORATION / MEASUREMENT

• The greater the standard deviation, the greater the


unevenness of the track.
• The NMT also carries video cameras to allow visual track
inspection.

4
TRACKBED DETERIORATION / MEASUREMENT

2. Visual Inspection
• Regular visual inspection (including video inspection from
vehicle mounted cameras) is vital so that all safety related
problems can be corrected rapidly.
• This particularly includes rail breaks.

• Note: a rail break is not an inevitable disaster; the two rail


ends are still rigidly attached to adjacent sleepers and can
continue to carry train load; however, the stresses are
greatly increased and full failure (a second break) will soon
occur if the rail is not replaced.

5
TRACKBED DETERIORATION / MEASUREMENT

2. Visual Inspection
• Visual inspection will also identify areas of ballast which
have become badly contaminated.
• A dirty slurry will be visible at the surface.
• There will also be splash marks where the pressure of a
passing train has caused the slurry to spray up (termed
pumping).
• The result is very obviously visible.

6
TRACKBED DETERIORATION / MEASUREMENT

3. Ballast / Sub-ballast / Subgrade Investigation


• In order to take the right maintenance decisions it is often
necessary to physically inspect the materials in the
trackbed, not only from the surface.
• The traditional way of carrying this out is to dig a trial pit.
Samples can be taken back to a laboratory and analysed.
• Insitu tests can be carried out (cone penetrometer or shear
vane for subgrade strength).

7
TRACKBED DETERIORATION / MEASUREMENT

A quicker, cheaper method is termed ‘ballast sampling’


• A tube (the ballast sampler) is driven through the ballast
and removed. The sample taken is definitely disturbed but
still gives a good visual idea of the condition of the material,
i.e. the degree of contamination.

8
TRACKBED DETERIORATION / MEASUREMENT

4. Radar Survey
Radar is becoming
increasingly used for
thickness and condition
assessment on railways.
It has advantages over
ballast sampling in that
a continuous profile can
be derived. It is also a
fairly rapid process,
speed being dependent
on the degree of
accuracy required.

9
TRACKBED DETERIORATION / MEASUREMENT

4. Radar Survey
Issues:
 It is not possible to discriminate between layers of
similar material;
 Radar normally has to be calibrated against ballast
sample data;
 Radar can also pick up moisture changes – at least
approximately.

10
TRACKBED DETERIORATION / MEASUREMENT

5. Structural Evaluation
It would be useful to
establish the strength or
stiffness of the trackbed
directly, measuring it in
some way. Nowadays,
this is possible using the
‘Falling Weight
Deflectometer’ (FWD).

11
TRACKBED DETERIORATION / MEASUREMENT

5. Structural Evaluation
Falling weight applied to
a sleeper

Sensors pick up
movement of loaded
sleeper and others

 A pulse load applied via a falling weight (20-30msecs)


 Magnitude 5-20 Tonnes (50-200kN)
 Deflection measured by ‘geophones’
 Typically a few hundred microns

12
TRACKBED DETERIORATION / MEASUREMENT

5. Structural Evaluation
By assessing the relative deflections of different sleepers or
points on the ballast surface, it is possible to deduce
(approximately) the stiffness of both the ballast and the
subgrade. A knowledge of subgrade stiffness is particularly
important for maintenance design.

13
TRACKBED DETERIORATION / MEASUREMENT

14
BALLAST PROPERTIES

1. Particle Strength
• Even rock is not infinitely strong! Rocks used for railway
ballast have a crushing strength of something over 100MPa.

• (a train axle load – say 200kN distributed over more than one
sleeper – generates an average stress in the ballast of around
150kPa; BUT the ballast is not solid rock – the local stress at
the contact between the sleeper and individual particles will
be much, much higher, even over 100MPa)
15
BALLAST PROPERTIES

1. Particle Strength
• If an individual particle is overstressed, it will fracture and
the ballast particles will re-orientate themselves.
• Fracturing will continue until re-orientation means that
there are enough particle contacts, for each not to be at too
high stress.

16
BALLAST PROPERTIES

2. Granular Material Behaviour


The following is the sort of stress / strain relationship
which any granular material or soil will give under
repeated loading.

17
BALLAST PROPERTIES

2. Granular Material Behaviour

Stress
1st 10th 100th B 1000th

Strain
A

18
BALLAST PROPERTIES

19
BALLAST PROPERTIES

• Two important aspects of behaviour can be identified:


A. Represents the stiffness of the material. The real
behaviour is obviously non-linear and complex but the
slope of the line shown (stress / strain) could be used as
an approximate ‘Young’s Modulus’. A better description is
‘Resilient Modulus’. A typical value for railway ballast
would be 100MPa (compare steel at 210GPa, concrete at
30GPa).

B. Represents the tendency of the material to strain


permanently. This property is related to the shear
strength of the material in a shear failure test. In a railway,
it relates to the rate at which settlement of the sleepers
can be expected.
20
BALLAST PROPERTIES

• What is going on?


• The key form of deformation is frictional slip between
particles. Friction  energy loss – hence the loop in the
stress/strain behaviour (termed a hysteresis loop).
• Note: Don’t think that friction only means permanent
deformation – it takes place during normal repeated
stress/strain behaviour too, first one way then back again.
• But, a small part of the slip is not recovered each time. It is
this small part which forms the permanent deformation
(i.e. settlement of the trackbed).

21
BALLAST PROPERTIES

3. Properties of an Ideal Ballast


a) High stiffness  reduces stress to weaker underlying materials
(not so important)  reduces track deflection under load
 reduces bending stresses in the rails

b) High Strength  reduces build-up of settlement under traffic loading


(very important)  gives horizontal stability to the track

c) Crushing Resistance  no degradation of highly stressed stones under sleeper


(very important) (or during tamping – see maintenance notes)

22
BALLAST PROPERTIES

3. Properties of an Ideal Ballast


d) Not affected by:  dirt from above
 soil contamination from below
 rainfall
 frost
 chemical weathering
e) Absorbs noise and surplus energy from vibration
f) Allows easy maintenance (i.e. it is sufficiently workable)

23
PERMANENT DEFORMATION WITHIN BALLAST
1. Strength and Permanent Deformation
Ballast is a frictional soil. Therefore it has an angle of friction.
Normal Force
Shear Movement of
Stress stone
Shear
Force

B
Note: this effect does not require
A the stones to break (although that
would make it easier!)
 - typically about 50

Normal Stress
So, stress at or above the line shown (Point C) will definitely cause failure. This means
that there is enough shear stress to force one stone to ‘ride up’ over another.
24
PERMANENT DEFORMATION WITHIN BALLAST

1. Strength and Permanent Deformation


• But what happens when the stress state (ratio of shear to
normal stress) is still on the safe side of the failure line but it
gets close (Point B)?
• In reality, there are an infinite (nearly) variety of different
particle contact details. Some can easily slip.

25
PERMANENT DEFORMATION WITHIN BALLAST

1. Strength and Permanent Deformation


• In other cases it is much harder.
• So, as the stress is applied, a few slips will take place very
easily indeed,
• And this will transfer a little more load onto those which do
not slip.
• But by definition, if the stress state is less than the stress
ratio at failure tan Ø, then there will still be enough
awkward shaped particle contacts to prevent full failure
developing.

26
PERMANENT DEFORMATION WITHIN BALLAST

1. Strength and Permanent Deformation

100%
% of slipped
particle
contacts
(on a
particular
Ultimate shear
shear plane)
strength = tan ×
normal stress

increasing shear
A B C

27
PERMANENT DEFORMATION WITHIN BALLAST
1. Strength and Permanent Deformation
OK. So the train load causes a certain percentage of particle
contacts to slip (i.e. fail locally). Then it goes away. What
happens? There are 2 possibilities.
a) if the slip is not too great, then the unloading may allow
the particle to slip right back to where it started,
b) if the slip is too great then, unless there is a load the other
way (unlikely), the particle never finds its way back;
irreversible deformation (though only local) has occurred.

28
PERMANENT DEFORMATION WITHIN BALLAST
1. Strength and Permanent Deformation

• Even if only 0.1% of slips remain after the load is


removed, there will be deformation in a few places.
• The particles then re-arrange themselves slightly.
• Next time the load comes, this different arrangement
gives a different set of slips.
• Contact points where nothing slipped before will now
feel a slightly different set of forces because of the
slight difference in particle arrangement.

The result: a whole new set of possible slips

29
PERMANENT DEFORMATION WITHIN BALLAST
1. Strength and Permanent Deformation
The effect of all this:
Shear
strain C
B

Number of load applications

A: If the shear is not too great, almost all slips ‘recover’ – negligible deformation.
B: More shear and some will never recover – some deformation every load cycle.
C: Approach failure stress and many will not recover – rapid accumulation of
deformation.
30
PERMANENT DEFORMATION WITHIN BALLAST
Other Effects

• Crushing of stones makes permanent deformation


easier. If an intact stone slips over an intact stone, then
it may be able to find its way back. If one of the stones
breaks there is probably no way back!

• Stiffness of stone has only small effect. However,


there is some elastic deformation at particle contacts.
The bigger this deformation, the easier slip becomes.

31
BALLAST FOULING

1. Causes A. Dirt from


above (trains,
wind-blown
dust)
B. Fines from
ballast breakage
under traffic
C. Fines from
ballast breakage
during tamping
D. Soil particles
from subgrade
32
BALLAST FOULING
A. Dirt is deposited on the track
from the atmosphere. It may
be wind-blown from some
distance or it may fall directly
from passing trains. This may
amount to as much as a
millimetre or two a year. It is
then washed down to the bottom of the ballast layer, until it
encounters less permeable material.
B. There are very high contact stresses between the sleepers and
individual ballast particles and between the particles
themselves. This will lead to splitting of individual stones and to
a gradual breakdown of the material. The rate depends on
stone strength as well as track stiffness and number and
weight of axle loads.

33
BALLAST FOULING

C. Tamping damages the ballast severely, causing large numbers


of fractures in critical areas.
D. If the subgrade soil is not properly separated from the ballast
(e.g. by a sand blanket or a geotextile), then mixing of the two
materials will take place. Ballast stones will tend to sink into the
soils as soil (in the form of slurry) is pumped by the action of
passing trains up into the ballast layer.

34
BALLAST FOULING
• The result is that certain parts of the ballast layer change
from being a single sized material to being much more
broadly graded.
• The exact nature of the resulting material depends on the
nature of the fines present.
• If clay has come up from the subgrade, then the properties
are likely to be particularly bad.

• Fouling tends to build up at the bottom of the ballast layer,


but can also occur immediately under the sleeper where
ballast break-down is most likely to occur and to the sides,
where tamping damage takes place.

35
BALLAST FOULING
2. So What is the Problem? - WATER

Water tends not to drain well


from the area around the
sleeper

Railway ballast starts off as single


sized (about 40-50mm). Even when
it is closely packed, this means that
there is about 25% air voids in it. It
is permeable, so rain-water has no
effect at all, as long as it can escape
to the sides or down into a
permeable soil.
36
BALLAST FOULING

Effect of Fouling
Dry: ballast is a
pure frictional
material
Wet: water near
contact points; may
give suction
Very Wet: positive
pore pressures =
reduced strength &
stiffness
37
BALLAST FOULING

When it is fouled it is much less permeable. Rain-water cannot


escape freely and it remains within the ballast for long periods.

Suction: improved
properties

Positive Pore
Pressure: much worse
properties

38
BALLAST FOULING

3. What is the Answer?


Hard Rock (flint, granite) 
good crushing resistance
Angular Stones (crushed rock) high shear strength

Frictional Stones (limestone)
Open Grading

high stiffness
non-moisture-susceptible

but high crushing forces
Dense Grading high shear strength
but moisture-susceptible
Large Stones stiffer, stronger material
but high crushing forces 
39
BALLAST v CONCRETE
So why use ballast anyway?
Two fundamentally different ways of doing railways:
1. Granular (i.e. ballast)
It will deform  settlement  need for maintenance
BUT it can be restored again

2. Rigid (i.e. concrete)


It MUST not deform because it is so difficult to restore

  
Ballast is cheaper, easier and requires no awkward decisions.
  
Concrete guarantees a better, safer, faster, less disrupted
 
train service BUT it costs more.
40
BALLAST FOULING
In practice,
• the ‘easy maintenance’ requirement dictates the use of
granular material.
• The fouling problem requires the use of an open grading
and the crushing resistance is provided by the use of very
hard stones.
• But there is an on-going debate as to whether the best
solution has been reached.
• This is particularly important when ballast is re-used after
cleaning.

41
…next lecture
 SUB-BALLAST AND SUBGRADE
 INHERENT TRACK QUALITY
 ‘BEAM ON ELASTIC FOUNDATION’ ANALYSIS

42

You might also like