Track Substructure Maintenance--From Theory to Practice
by
Ernest T. Selig, P.h.D., P.E.
Emeritus Professor of Civil Engineering at
the University of Massachusetts
and President of Ernest T. Selig, Inc.
105 Middle Street
Hadley, MA 01035
Phone: (413) 586-1449
Fax: (413) 585-9392
[email protected] and
Darrell D. Cantrell
Past President, AREMA
and Senior Consultant
Cantrell Rail Services, Inc.
805 Willow Court
Keller, TX 76248
Phone: (812) 281-3477
[email protected]
Prepared for American Railway Engineering and Maintenance-of-Way Association
Annual Conference
Chicago, Illinois
September 9-12, 2001
Track Substructure Maintenance--From Theory to Practice
by
Ernest T. Selig, P.h.D., P.E. and Darrell D. Cantrell
Abstract
The three main components of track substructure are ballast, subballast and subgrade.
Their functions and how they are achieved will be reviewed. All three components must be
present and functioning satisfactorily for good track performance. Substructure maintenance is
mainly concentrated on the ballast and the need is usually dictated by track geometry
deterioration. Second in priority for substructure maintenance are drainage ditches. The
principles of drainage will be summarized. In practice drainage is often given inadequate
attention. The cost of maintenance or, conversely, the deterioration of track components, is
directly impacted by the condition of the drainage system. Information on these cost factors will
be presented. When normal maintenance operations cannot provide satisfactory track
performance, reconstruction will be needed. This usually involves work on the subballast,
subgrade or drainage system to repair the defects. Subgrade treatment methods are described.
These are more expensive, and more difficult to accomplish tasks, and more disruptive to
revenue traffic. Thus a goal of maintenance planning should be to minimize the need for the
reconstruction by not allowing chronic problems to develop or persist, to the extent possible.
Guidelines for maintenance practice will be presented. The paper also emphasizes the
importance of determining the cause of the maintenance problems in order to apply the proper
solutions. A follow up investigation is recommended to evaluate the results. Finally education
programs are recommended to provide a much needed understanding of the fundamentals of
track substructure behavior.
INTRODUCTION
The substructure consists of ballast, subballast, and subgrade. The need for substructure
maintenance stems from the development of excessive geometry deviations (track
roughness). Track geometry loss results from movements in the ballast, subballast, and
subgrade. However, adjustment of geometry is done by rearranging the ballast. There
are various causes of geometry loss which need to be identified to know how to improve
track performance. The amount of maintenance required is influenced by decisions made
during design and construction or rehabilitation of the track. The railroad may choose to
correct the geometry faults by tamping as needed. This should only be done where
needed, but tamping is often extended to adjacent areas that do not need adjusting.
Tamping is appropriate, but it does damage the ballast and may not be economical in
frequently reoccurring problem spots. Further, tamping does not correct the root cause of
the roughness. Thus the need to find the cause and fix the problem arises (1).
If the tamping improvement does not last long enough, then the ballast condition
should be checked. If the fouling is excessive then undercutting with the addition of new
ballast as needed will be programmed, or alternatively the track will be raised so that
additional ballast can be placed under the ties. If excessive fouling is not the problem or
if the undercutting process does not improve the performance, then an examination is
required to determine the cause of the reoccurring geometry problem.
The drainage of railway track is recognized as a fundamental, yet often neglected
aspect of track design. Track drainage significantly affects track performance and track
maintenance requirements. A complete drainage system must include means of handling
water flowing over the ground surface towards the track, shedding or draining water
falling onto the track, and controlling water from the subgrade (2).
SUBSTRUCTURE COMPONENTS
The ballast is the top structural layer that, among other things, holds the track in place,
reduces the stress transmitted to the subgrade, and facilitates the geometry adjustment
through tamping (3,4). Fouling is the main cause of ballast problems. Once the cause of
the fouling is determined the solutions can easily be identified and applied.
The best choice of ballast for a particular location is the most cost-effective
choice, taking into consideration traffic, environmental conditions, and cost of material
delivered to the site. The best choice, therefore, will not necessarily be the highest
quality material or the material that has the lowest delivered cost. These costs include
purchase and transportation of ballast as well as tamping and undercutting/cleaning.
The subballast is the second structural layer that helps the ballast reduce the stress
to subgrade, maintains separation between the ballast and subgrade particles, and plays an
important role in track drainage (1). Subballast deserves more attention than it gets
because of its importance to track performance. Filter fabric (geotextile) has frequently
been used to fulfill some of these subballast functions, but investigations have shown that
fabric is not generally desirable (1). Asphalt concrete is an alternative or supplement to
sand/gravel subballast materials, but the economics will probably limit asphalt to special
cases.
Subgrade serves as the platform for the structural layers, and so it must be stable.
The subgrade has many modes of failure such as: attrition, squeeze, progressive
compression under repeated loading, consolidation, massive shear, frost heave/thaw
softening and swelling and shrinking from moisture change (1). Investigation is needed to
know which is the cause of the subgrade problem in a specific case in order to choose the
correct cure. This determination may be easy, but often it can be very challenging.
Subgrade problems are clearly more difficult to correct, than ballast or subballast
problems, especially after the track has been constructed. But there are a lot of possible
solutions to subgrade problems.
BALLAST FOULING
Over a period of time in track the ballast gradation typically becomes broader and finer
than the initial condition because the larger ballast particles will break into smaller
particles and additional smaller particles from a variety of sources will infiltrate the voids
between the ballast particles. This process is known as fouling. Fouling deteriorates
ballast performance and eventually requires undercutting/cleaning or track raising with
ballast replacement when tamping improvements do not last long enough
Five categories of fouling material have been identified: 1) particles entering from
the surface such as wind blown sand or coal falling out of cars; 2) products of wood or
concrete tie wear; 3) breakage and abrasion of the ballast particles; 4) particles migrating
upward from the granular layer underlying the ballast; and 5) migration of particles from
the subgrade (1). The main causes of ballast fouling must be identified if proper steps are
to be taken to reduce the rate of fouling. The most frequent cause of ballast fouling is
ballast breakdown, but there are individual situations in which each one of the other
categories predominates.
Geotextiles (filter fabrics) generally have not been found to be useful in solving
ballast fouling problems. A proper subballast layer is the best cure for fouling from the
underlying granular layer and from the subgrade. When subgrade is the source of fouling
material one of two main mechanisms usually is present: 1) abrasion of the subgrade
surface by ballast particles in contact with the subgrade, or 2) hydraulic erosion of water-
filled cracks in the subgrade subjected to repeated train loading.
BALLAST/SUBBALLAST LAYER THICKNESS
The thickness of the ballast layer beneath the ties generally should be the minimum that
is required for the ballast to perform its intended functions. Typically this will be 12 to
18 in.. The minimum subballast layer thickness beneath the ballast should be 6 in. Thus,
the minimum granular layer thickness beneath the ties would be 18 to 24 in. A check
then needs to made to determine whether this is enough thickness to prevent
overstressing the subgrade (5,6). An overstressed subgrade will result in rapid loss of
geometry under train traffic.
BALLAST CLEANING AND RENEWAL
Ballast cleaning and renewal is an expensive and time-consuming process which is also
highly disruptive to train traffic. Thus, the need for this process must be considered
carefully. Basically ballast cleaning and renewal is required when the ballast becomes so
fouled that it cannot fulfill its functions.
Care must be taken during ballast cleaning to prevent removing or damaging the
existing subballast layer. Care must also be taken to ensure that the width and inclination
of the cut surface are such that the water collecting in the cleaned ballast is able to flow
freely across the cut surface and into the track side drainage system. Effort made to
achieve a geometrically smooth cut surface will also improve the stability of the track
geometry under subsequent traffic after the cleaning and renewal process is completed.
The economic benefit of ballast cleaning and renewal will not occur unless the
drainage system is functioning as well (7). Thus, a ballast cleaning and renewal program
should also accommodate installing or repairing the drainage system. It is also desirable
to evaluate the cause of the ballast fouling and consider taking steps to correct the cause
of the fouling in order to reduce the amount of future undercutting.
Shoulder cleaning can lengthen the time before undercutting cleaning is required
(7). The shoulder cleaning must include the full depth of the shoulder all the way out to
the ballast toe. Shoulder cleaning improves drainage by shortening the drainage path of
water out from under the ties and increases the depth of drainage below that which is
typically achieved with undercutting/cleaning. Partial depth undercutting will not result
in effective drainage of water from the uncleaned lower portion of the ballast layer.
Shoulder cleaning also allows migration of fouling materials out from under the
track into the clean voids in the shoulder in order for the fouling material to migrate, the
fouling particles must be much smaller than the voids in the ballast and need a forcing
mechanism such as flowing water or pumping by repeated train loading. When the
fouling material results from breakdown of the ballast in a way that produces a broad
range of particle sizes, migration of the fouling material is not likely to be significant.
SUBGRADE TREATMENT METHODS
The following are a few examples of subgrade remedial treatment methods:
1. Grouting--Some grouts penetrate the voids of the soils (permeation grouting) and
strengthen them or reduce water seepage. Other grouts compact and reinforce the
soils (compaction grouting) to strengthen them or displace the soils to compensate for
settlement. Jet grouting mixes cement with soil to form columns of strengthened soil.
2. Soil mixing--A process in which soil is mixed with augers and paddles to create a
mixture of soil and cement based grout. Soil mixing creates a column of strengthened
soil for compression and shear reinforcement.
3. Lime modification of clay properties--Lime can be used to improve the properties of
clay. Techniques for mixing lime with clay include mechanical mixing (admixture
stabilization) and quicklime piles. Injecting lime into clay in slurry form has been
performed with the expectation of improving the clay properties. This is a common
but not usually effective treatment with often undesirable side effects, e.g. slurry
injection can fracture clay instead of penetrating the voids and also solidifies ballast
(8).
4. Reconstruction--Compaction of existing soils or replacing them with better soils will
give improved subgrade. Chemicals can also be mechanically mixed with the soils in
layers which are then recompacted to form a stronger or less reactive soil. The
chemistry of the soils should be checked or tests performed to verify the effectiveness
of the treatment, because some combinations can be harmful.
5. Reinforcement--Various plastic grids, metal strips or cellular materials placed in the
soils give tensile reinforcement. Alternatively, steel reinforcing can be installed in
the subgrade soils through techniques such as soil nailing, soil doweling and
micropiling.
6. Stress reduction--Increasing the thickness of the ballast and subballast will reduce the
pressure on the weaker subgrade caused by the train loading. Contrary to the
AREMA engineering manual, the allowable pressure is not constant, but varies
widely and must be determined in each case for correct design. The correct strength
considers the magnitude of the repeated loading from the trains and the number of
repetitions. For a given axle load, a high tonnage line would have a much lower
apparent strength than a low tonnage line. Thus the high tonnage line needs a greater
ballast/subballast thickness for the same subgrade properties (5,6).
IMPORTANCE OF PROPER INVESTIGATION
A fundamental limitation of common railroad practice regarding substructure remedial
work is the lack of a proper investigation to diagnose the cause of the problem before
attempting a solution. A proper investigation includes using trained personnel who
understand the fundamentals of what is being observed and recognize appropriate
solutions.
This limitation of inadequate investigation is compounded by not following up to
evaluate the results of the remedial work. Thus, sometimes improvements occur but for
reasons other than the ones expected. Other times improvements do not occur, and
sometimes the problem gets worse. In these cases the reasons will not be known without
a follow up investigation. One consequence of inadequate follow up is that the decision
about whether or not to use the same method elsewhere cannot be reliably made.
The inevitable result of investigation deficiencies is inefficient use of maintenance
resources or a poor return on the investment. An example is that mud in the ballast is
commonly thought to derive from subgrade soil particles. This has led to the
inappropriate use of filter fabric. Investigations have shown that subgrade is rarely the
source of the mud and even in cases where it was, the fabric did not stop the mud.
Although the filter fabric can usually be omitted some other action is still needed to fix
the problem.
NEED FOR GOOD TECHNOLOGY
For reasons of safety, reliability, and profitability railroads need to use the best
technology available. This challenge increases with increasing traffic, increasing axle
loads and higher speed trains, as well as decreasingly available work windows. Mixing
these together on the same track is even a bigger challenge. Another challenge is the
problem of track transitions that occur at turnouts, bridges, and tunnels for example.
A lot of substructure technology is available that is not being applied. One reason
for this situation is the insufficient use of education programs to learn about the
possibilities. Another reason often mentioned is that the investigation and the remedial
work are too expensive. The answer to this is to do a complete cost/benefit evaluation.
Proper use of good technology will improve the economics, and no doubt pay for its
implementation (9).
SOLUTIONS START WITH DRAINAGE
It is usually a good idea to start the investigation by examining the track drainage since
drainage is often a factor causing chronic problems. Achieving proper drainage is not
simply a matter of digging a cross trench and letting water out of the track. If not used
properly, these trenches become longitudinal discontinuities that can cause roughness
increase. They can also allow water to reach further in to the subgrade soil causing more
softening. Furthermore if the trench is filled with ballast, subgrade attrition may occur,
thus creating a new problem.
Drainage is a complex problem (2). Factors to consider include: ballast fouling
condition, subballast gradation, slope of subgrade surface, ditch or pipe depth ditch or
pipe longitudinal slope and expected rainfall characteristics. Remember too that water
drains by gravity and so can only flow to outlets at a lower elevation. Also clay soils will
not drain because of their low permeability.
It would be desirable to have the subballast shed the water draining down from
the ballast to keep the water from the subgrade, but the subballast must also be permeable
enough to drain water which gets in to it. These are contradictory requirements which
can not be met with a single subballast layer.
Drainage of railway tracks is essential to acceptable track performance. Water in
the track substructure originates from three potential sources: 1) precipitation onto the
track, 2) surface flow from areas adjacent to the track, and 3) groundwater flow. A
complete drainage system must include provisions for handling water from all three
sources (2).
Precipitation onto the track will enter the ballast. It will then flow laterally out of
the ballast into ditches or enter the subballast. This water will either drain laterally out of
the subballast or continue downward into the subgrade. Surface ditch drains can collect
water from the ballast and subballast. Subsurface drains are needed to collect water
flowing through the subgrade and may also be needed to help drain the subballast.
Particularly difficult is drainage of water between adjacent tracks or from tracks
surrounded on both sides by other tracks. Not only is the drainage path to the side longer,
but a suitable drainage path is more difficult to maintain. Cross-drains under the outer
tracks or longitudinal drains between tracks may be needed.
Ballast can become saturated from rainfall by a high degree of fouling, which
holds water, or by restricted drainage at the edge of the ballast. Causes of this restricted
drainage include, for example:
• Ballast pocket formation from subgrade settlement
• Fouled ballast shoulder
• Low permeability boundary at edge of ballast
• Ponding of water next to the track from lack of a ditch to carry water away from the
track after exiting the ballast
• Inadequate lateral slope on the subballast surface to direct water to the side of the
track.
The ability of water to drain laterally also requires that the drainage path at the edge of
the ballast and subballast layers not be blocked (2). Two conditions need to be met to
achieve this requirement: 1) the ballast shoulder and the edge of the subballast must be
free draining, and 2) discharged water must be able to flow away from the track that the
toe of the ballast slope is the most critical zone for achieving drainage of water out of the
ballast.
CONCLUSIONS
Clearly there are a lot of remedial techniques for subgrade problems to draw upon. An
investigation is necessary to select the correct solution for each specific case. In most
cases, it is not until the underlying cause of a problem is known can a lasting and
economical solution be found. The subject is complex, but information is available to
help.
Ballast is an important part of track maintenance. It is also a re-occurring part,
and so is designed to be maintained. Furthermore, ballast is the only part of the
substructure which is easily maintained. To keep costs and disruption to traffic low,
ballast maintenance should be minimized.
Ballast properties change over time in track and the ballast becomes increasingly
fouled. The causes of fouling need to be understood in order to reduce the rate of ballast
performance deterioration. Undercutting/cleaning is needed to restore the desired ballast
properties after becoming fouled. However, undercutting/cleaning will not be cost-
effective unless track drainage is functioning properly as well. Shoulder cleaning can
lengthen the time until undercutting/cleaning is needed.
Tamping is needed for geometry corrections. However, tamping should only be
done when and where it is needed to minimize ballast damage and temporary reductions
in track stability.
Ballast and subballast contribute to protection of subgrade from overstressing.
However, subballast, which is always needed, is preferred for completing the total
granular layer thickness beyond the minimum ballast layer thickness required for
fulfilling the other ballast functions.
No factor is more important to track performance than drainage, yet drainage is
often inadequate or neglected.
When unstable subgrade conditions exist, the cost of maintaining track geometry
will be much higher than normal. A number of effective remedial solutions are available.
Even though the cost of these repair methods is often high, low cost solutions may be
ineffective and even detrimental and so may not be cost effective.
Track substructure technology is complex. Its importance both requires and
deserves education programs.
References for More Information
1. Selig, E.T. and Waters, J.M., Track Geotechnology and Substructure
Management, Thomas Telford, Ltd., London, 1994.
2. Heyns, Francois J., “Railway Track Drainage Design Techniques”, Ph.D.
Dissertation, Department of Civil and Environmental Engineering, University of
Massachusetts, May 2000.
3. Selig, E.T., “Ballast’s part: It’s Key Roles and Qualities,” Railway Track and
Structures, March 1998, Vol. 94, No. 3, pp 21-.
4. Selig, E.T., “Ballast Performance: Considering more key factors,” Railway Track
and Structures, July 1998, pp 17-20, 36.
5. Li, D. and Selig, E.T., “Method for Railroad Track Foundation Design. I:
Development,” Journal of Geotechnical and Geoenvironmental Engineering,
ASCE, April 1998, Vol. 124, No. 4, pp 323-329.
6. Li., D., and Selig, E.T., “Method for Railroad Track Foundation Design. II:
Applications,” Journal of Geotechnical and Geoenvironmental Engineering,
ASCE, Vol. 124, No. 4, April 1998 pp 316-322.
7. Selig, E.T., “Characteristics of Ballast Shoulder Cleaning and Ditching,”
Proceedings, 6th International Heavy Haul Railway Conference, Cape Town,
South Africa, April 1997, pp 931-944.
8. Sussmann, T.R., and Selig, E.T., “Lime Stabilization of Railway Track
Subgrade,” Proceedings, 6th International Heavy Haul Railway Conference, Cape
Town, South Africa, April, 1997, pp. 98-112.
9. Selig, E.T., “Substructure Maintenance Management,” Proceedings, 6th
International Heavy Haul Railway Conference, Cape Town, South Africa, April
1997, pp. 796-809.