Long Term Monitoring of

Mass-Spring-Systems (MSS)
Roman Geier1, Helmut Wenzel2
1
Dipl.-Ing.; Project Manager of VCE Holding GmbH., Vienna, Austria – e-mail: [email protected]
2
Dipl.-Ing. Dr.; Managing Director of VCE Holding GmbH., Vienna, Austria – e-mail: [email protected]

ABSTRACT

To reduce noise, and in particular vibrations induced by modern high speed trains, reducing
the stiffness of the conventional superstructure with an elastic layer is very efficiently. The
vibration attenuating capacity of such superstructures, called Mass-Spring-Systems, depend
on the natural frequency and the damping ratio. The source of noise and vibration emissions
of railway lines mainly is the rough contact area of the wheel-rail system. The roughness
causes vibrations which are induced into the superstructure and propagate through the soil to
buildings in the neighbourhood. The paper is limited to low-frequency vibrations and
ground-borne noise reduction. The goal of the developments is to reduce the stiffness of the
conventional superstructure – consisting either of a ballasted track system or of a modern
solid roadway – by using additional elastic members. Finally the superstructure consists of a
mass – and an elastic layer which behaves like a spring.

1. INTRODUCTION

Due to their high transport capacity and their effective use of energy with lowest damage to
the environment railways are one of the most important means of transportation for the
future. In spite of the advantages of railways in comparison with other transport systems as
for example motor cars the acceptance of new railway lines is very low especially by
potential neighbours. One of the most important reasons for that is the fear of irritations from
noise and vibrations induced by modern high speed trains. These problems especially occur
in densely populated areas as in towns, where railway routes are in tunnels with low
overburden and very close to residential buildings. Due to maintenance reasons ballast-less
permanent ways become more and more important, especially for such tunnel lines. With
regard to load carrying capability and to long-time stability of the track these solid roadways
show a lot of advantages. Nevertheless the most important disadvantages of most kinds of
permanent ways are the additional increasing noise and vibration emissions caused by using
such superstructures.
2 GEIER, WENZEL

2. TYPE OF PROBLEM

The source of noise and vibration emissions of railway lines is the rough contact area of the
wheel-rail-system. The roughness causes vibrations which are induced into the
superstructure and propagate through the soil to buildings in the neighbourhood.
The following report is limited to low frequency vibrations and structure- or ground-borne
noise reduction measures. For the further remarks vibrations are defined as perceptible low
frequency oscillations between 1 Hz and 80 Hz, whereas structure- or ground-borne noise
are mechanical vibrations in a audible frequency range between 16 Hz and 20 kHz. Both
kinds of emissions are transmitted through the soil by mechanical wave propagation. The
distinction between noise and vibrations is shown in figure 1.

VIBRATIONS
well perceptible
velocity of vibration [mm/s]

without MSS (KB=0,4)

1 barely perceptible
(KB=0,15)

GROUND-
BORNE
0.1 without
MSS NOISE
with MSS well audible
(50dB(A))
limit of
audible range with
0.01 MSS
barely audible
(20dB(A))

1 2 5 10 20 50 100 200
frequency [Hz]

Figure 1. Distinction between noise and vibrations

In accordance to national codes and laws the allowed intensity of noise and vibrations caused
by railways is limited to certain values. To reach these limits different kinds of superstructure
modifications have been developed within the last 25 years.

3. THEORETIC BACKGROUND

The goal of all these developments is to reduce the stiffness of the conventional
superstructure – consisting either of a ballasted track system or of a modern solid roadway –
by using additional elastic members. Finally the superstructure consists of a mass – including
the rails, the rail carrying system with the sleepers and the ballast bed or an alternative solid
roadway – and an elastic layer which behaves like a spring.
The principle of mass-spring-systems is based on the response amplification factor of a
dynamic system and can be explained in a very simple way by a linear
single-degree-of-freedom system (SDOF-system).
These basic principles of the physics of a dynamic system can be transformed into possible
variation parameters for realisation of a vibration mitigating construction, reduced to a
SDOF system consisting of a vibrating mass m, a spring k and a damper c. These factors
have a strong impact to the insertion loss of the MSS.
GEIER, WENZEL 3

rail pad
baseplate pad

sleeper pad

ballast mat

elastic bearing for

slab track

Figure 2. Elasticity in railway superstructures

The insertion loss ∆LE is defined as the difference of the noise and vibration level induced by
trains on conventional ballast bed track systems and on mass-spring-systems. According to
the effectiveness of mass-spring-systems (MSS) the following division into three groups is
commonly used:
- light-weight MSS with m ≤ 4 t/m, f1 ≥ 15 Hz
- medium-weight MSS with m ≤ 8 t/m, f1 ≥ 10 Hz
- heavy-weight MSS with m > 8 t/m, f1 < 10 Hz
These theoretical thoughts can be summarised by the statement “elasticity in modern railway
superstructures reduces noise and vibrations”.

4. REAL CONSTRUCTION

To put the shown principle into practice different construction philosophies have been
developed during the last 25 years. Depending on where the elastic intermediate layer is
situated a more or less heavy sprung mass can be achieved. The possible locations of the
elastic elements are in the rail fastening system (rail pads or base plate pads), below the
sleepers (sleeper pads), below the ballast bed (sub-ballast mats) or in case of a ballast-less
solid roadway below a floating slab track. The last one mentioned are also known as
classical Mass-Spring-Systems.
The mass consisting of reinforced concrete with a rail carrying system mounted can be either
a chain of short elements connected by hinges or a joint less slab with several hundreds of
meters length. In the first case a single-degree-of-freedom (SDOF) model seems to be an
appropriate simplification, whereas for the second case – the joint-less system – more
detailed investigations on a multi-degree-of-freedom (MDOF) model are necessary.
4 GEIER, WENZEL

4.1 Elastic Elements – Bearings

As mentioned before the dynamic behaviour of Mass-Spring-Systems is dominantly

influenced by the bearing material. For applications different groups of requirements for the
elastic elements have to be defined.
The first group can be summarised under the headline “vibration requirements” – the
dynamic characteristics of the bearing material are needed for the design. The most
important quantity is the stiffness, the frequency and amplitude of excitation, and the inner
damping. Furthermore, changes of the dynamic behaviour of the materials over time should
be considered. Additionally, “mechanical requirements” can be defined. Long-time stability
of the elastic elements has to be ensured for the relevant applied load combinations.
Important for the serviceability of the bearings, are the load carrying capability, the fatigue
behaviour, the deformations due to pressure and the long time settlements.
A third group of requirements must be defined to ensure the “integrity of the elastic
elements” under site conditions. The material has to be stable not only against water but also
against chemicals like diluted alkalis and acids, and against commonly used types of oil and
fat. In summary, one can say that the most important parameter which describes the quality
of a bearing made of natural or artificial rubber is the quotient between the static kstat and the
dynamic stiffness kdyn. The static stiffness is responsible for the deflection of the
mass-spring-system under dead and live loads, whereas the dynamic stiffness is the
key-parameter for the insertion loss, and therefore responsible for the vibration attenuation.

4.2. Mass

In general the sprung mass consists of all elements laying on the elastic element. For
classical Mass-Spring-Systems the sprung mass consists of a reinforced concrete trough or
slab, the rail carrying system (sleepers in a ballast bed or any kind of solid roadway) and the
rails. Beside for any noise and vibration damping reasons most kinds of rail fastening
systems contain an elastic layer (rail pads, base plate pads) which make a defined deflection
of the rail possible. Other kinds of solid roadways include sleeper pads which have the same
function as elastic rail fasteners. Anyway, for the further reflections concerning classical
Mass-Spring-Systems these elasticity will be neglected – the mass is assumed as a rigid
element.
At least for medium- and heavy-weight MSS it is necessary to couple the single elements to
reach a continuous deflection curve. In some cases for light-weight systems the bending
stiffness of the tracks themselves is sufficient to reach a continuous deflection curve. To
reduce the initial investment costs and to limit the maintenance expenditure for
Mass-Spring-Systems, VCE developed a new type based on a continuous design philosophy
which does not need expensive coupling elements any more.
GEIER, WENZEL 5

5. INVESTIGATION PROGRAM

The paper should give an overview about the currently applied investigations performed on
Mass-Spring-Systems in Austria. Basically deformation measurements, temperature
measurements, dynamic response identification and measurement of the ground transmitted
vibrations can be distinguished according figure 3. The long term measurements usually
have a run time of more than 2 years in order to identify the walking line precisely. The
so-called acceptance measurements consists of investigations only once during the whole
measurement program.

Mass-Spring-System

Long-Term Monitoring Acceptance

Deformations Temperature Dynamic Response Deflections

Ground During Test

Air Tunnel Wall MSS MSS
Transmission Operation Measurement

Figure 3. Investigation program for a Mass-Spring-System

5.1 Vertical Deflections of the MSS

This investigation usually is performed before the regular railway traffic starts, therefore a
test train is necessary in order to identify and compare the measured to the calculated results.
In addition tests are performed during regular traffic in order to obtain maximum deflections
under high loads.

5.1.1 Test Measurements

The investigation usually is performed with a well known locomotive passing the MSS with
different velocities. An average value is calculated from these tests; the maximum deflection
for the static as well as the dynamic test is derived. The maximum deflection value usually is
obtained from the static tests (lower bearing stiffness). In addition measurements of the rail
supporting elements is performed in order to obtain a total deflection value of the whole
system.
6 GEIER, WENZEL

Figure 4. Cross section of a Mass-Spring-System

5.1.2 Investigations during regular traffic

In order to identify the behaviour of the system during regular traffic and to identify
maximum displacement values usually a 24-hours measurement is performed. The
deflections are compared to the train weight which is read out from the train schedule if
possible. In order to make a comparison with the finite element calculation possible, several
load cases according to the standards are applied. From the experiences it is well known, that
the maximum deflection value usually is obtained during the passage of the locomotive.
Generally speaking a passenger train leads to lower deflections than heavy goods trains.
Aufnehmer_6

mm
44.369
44.266
44.164
44.061
43.959
43.856
43.754
43.651
43.549
43.446
43.344
43.241
43.138
43.036
42.933
42.831
42.728
42.626
42.523
42.421
42.318
42.215
42.113
42.010
13.83 15.38 16.92 18.46 20.00 21.55 23.09 24.63 26.17 27.72 29.26 30.80 32.35 33.89
s

Figure 5. Sensors installed to the MSS – Results of measurement

GEIER, WENZEL 7

5.2 Long Time Monitoring

To make a long time monitoring of the longitudinal behaviour possible, displacement sensors
are applied to characteristic locations of the MSS. The sensors are installed to critical points,
e.g. longitudinal locking as well as the beginning and the end of the system. All data can be
accessed easily, bye use of a data logger in combination with a GSM-data transmission unit.
In order to identify deviations resulting from creep and shrinkage effects, the monitoring
should least for a minimum time of 2 years. Long time temperature measurements which are
described in chapter 5.5 are an important input for the assessment of the longitudinal
deformations. A comparison between measured and calculated values is highly desired.

3,00

2,00

1,00
Wege [mm]

Weg1

0,00 Weg2

Weg3

-1,00

-2,00

-3,00
27.05.99

27.06.99

27.07.99

27.08.99

27.09.99

27.10.99

27.11.99

27.12.99

27.01.00

27.02.00

27.03.00

27.04.00

27.05.00

27.06.00

27.07.00

27.08.00

27.09.00

27.10.00

27.11.00

27.12.00

27.01.01

27.02.01

27.03.01

27.04.01

27.05.01

27.06.01

Zeitachse

Figure 6. Long time deformation observation

5.3 Dynamic Response of the System

The identification of the dynamic response, which is represented in the modal parameters
natural frequency and mode shapes is necessary in order to identify the vibration attenuation
of the superstructure. Due to a variation of the modal parameters as well as the concrete
mass, the efficiency of the system can be changed. In order to make a comparison to the
static design (desired values) possible, a modified finite element simulation (real stiffness
values for bearings) is used. During calculation the static values will lead to the lower limit
for the natural frequencies, the dynamic stiffness will lead to the upper limit. The measured
values then, have to be situated between these to border lines.
8 GEIER, WENZEL

5.4 Ground Transmitted Vibrations

By measurement of the ground transmitted vibrations outside the tunnel, the efficiency of the
MSS can be assessed easily. A comparison of the measured values is done according the
limits shown in the valid national standard. The measurement usually take place in nearby
buildings, where the foundation area as well as the top level is investigated. During
assessment a differentiation is done for the specific train-type; the ground transmitted
vibrations strongly depend on the train weight, train velocity as well as the quality of the
wheel surface (roughness).

5.5 Temperature Monitoring

In order to make an accurate determination of the concrete temperature possible, several

cross sections of the system are equipped with temperature sensors. Usually the concrete
trough, the tunnel wall as well as the air temperature and humidity is measured. The
observation of the concrete temperature consists of an assessment during the hydration
process as well as acquisition of the long time walking line (> 2 years duration).
Due to several cross sections the temperature behaviour close to the tunnel portal as well as
inside the tunnel can be assessed and compared. The long time temperature measurement in
combination with the longitudinal deformation is an important factor for assessment
according the static design of the system.
Specific equipment is required which should enable long time data acquisition and resist
rough environmental conditions during construction of the MSS and during use. This
measurements are an important factor to interpret and assess the longitudinal deformations of
the system.

24,00
22,00
20,00
Sensor T1 Sensor T3 18,00
16,00
Temperatur °C

14,00
12,00

Sensor T4 10,00
8,00
6,00
4,00
2,00
0,00
-2,00
-4,00
20.10.1998
20.11.1998
20.12.1998
20.01.1999
20.02.1999
20.03.1999
20.04.1999
20.05.1999
20.06.1999
20.07.1999
20.08.1999
20.09.1999
20.10.1999
20.11.1999
20.12.1999
20.01.2000
20.02.2000
20.03.2000
20.04.2000
20.05.2000
20.06.2000
20.07.2000
20.08.2000
20.09.2000
20.10.2000
20.11.2000
20.12.2000
20.01.2001
20.02.2001
20.03.2001
20.04.2001
20.05.2001
20.06.2001

Zeitachse

Figure 7 Long-Time Monitoring of the concrete temperature

GEIER, WENZEL 9

6. SUMMARY

The investigation program described in this paper is sufficient for long-time monitoring as
well as acceptance tests of Mass-Spring-Systems, which are required for the permission of
regular traffic.
The importance of noise and vibration reduction motivated the Austrian railway companies
ÖBB, HL-AG and BEG to establish a research programme. These research and development
works started in 2000 and are going to run over two years. Experts of different sciences –
dynamics, railway engineering, tunnelling, bridge engineering, concrete design – work
together to develop optimised railway superstructures for the future based on the
investigations made until now.
For Mass-Spring-systems further improvements will be achieved with the results of the
research project “Consistent Semiactive System Control - CaSCo” supported by EC (FP5)
started in spring 2000 under the co-ordination of VCE. The research work obtained during
the first two years show very promising results to increase the damping capability and
therefore reduce the ground transmitted vibrations of railway lines.

7. REFERENCES

Clough, R.W., Penzien, J., (1993), Dynamics of Structures, Second International Edition, McGraw-Hill.
Geier, R.; Wenzel, H.: (2001) Noise and vibration attenuating measures for modern railway
superstructures. Proceedings of the 4th European Conference on Noise Control, Euronoise,
Patras, Greece
Geier, R.; Pichler, D.: (2001), Arlberg-Tunnel – Messergebnisse der Erschütterungsausbreitung im
Untergrund, Report to ÖBB, Vienna
Pichler, D., Mechtler, R., and Plank, R., (1997), “Entwicklung eines neuartigen Masse-Feder-Systems
zur Vibrationsverminderung bei Eisenbahntunnels”, Bauingenieur 72, Springer-VDI-Verlag,
pp. 515-521.
Pichler, D., Huber, P., (1997) , “Reduction Measures for Tunnel Lines”, Report for RENVIB II Phase 1
to ERRI, Vienna Consulting Engineers and Rutishauser Ingenieurbüro.
Pichler, D., (1998), “Concrete based floating track slab systems – modelling and reality”, Proceedings
of the EURO-C 1998 Conference on Computational Modelling of Concrete Structures,
Badgastein, Austria, 31 March – (1998), A. A. Balkema, Rotterdam, Brookfield, pp. 665-671.
Pichler, D., Zindler, R., (1999), “Development of artificial elastomers and application to vibration
attenuating measures for modern railway superstructures”, Proceedings of the First European
Conference on Constitutive Models for Rubber, Vienna, Austria (1999), A. A. Balkema,
Rotterdam, Brookfield, pp. 257-266.
Steinhauser, P., (1997), Römerbergtunnel – Ergebnisse der VibroScan Untersuchung auf dem
Masse-Feder-System, Report to HL-AG, Vienna.
Wenzel, H., Pichler, D., and Rutishauser, R., (1997), “Reduktion von Lärm und Vibrationen durch
Masse-Feder-Systeme für Hochleistungseisenbahnen”, Oral presentation at the
D-A-CH-Meeting in Zürich, SIA-Dokumentation D 0145, pp. 123-132.