Professional Documents
Culture Documents
TBM Basics TUdelft
TBM Basics TUdelft
surface
Feasibility study for the Netherlands
A.H. Oldenhave
April 2014
URUP method applied for large
diameter tunnels
Feasibility study for the Netherlands
A.H. Oldenhave
1387286
April 2014
PUBLIC REPORT
Personal information
Student:
Angela Oldenhave
Studienr.: 1387286
Stevinweg 27, Bergschenhoek
Tel.: 06-29442689
angela.oldenhave@gmail.com
Graduation committee:
TU Delft:
Prof. Ir. J.W. Bosch
Sectie Geotechniek, Faculteit Civiele Techniek en Geowetenschappen, TU Delft
j.w.bosch@tudelft.nl
Tel.: +31 (0)15 27 82844
TU Delft:
Ir. S. van der Woude
Sectie Geotechniek, Faculteit Civiele Techniek en Geowetenschappen, TU Delft
s.vanderwoude@tudelft.nl
svanderwoude@vhbinfra.nl
Tel.: +31 (0)15 27 85266
TU Delft:
Dr. Ir. W. Broere
Sectie Geotechniek, Faculteit Civiele Techniek en Geowetenschappen, TU Delft
w.broere@tudelft.nl
Tel.: +31 (0)15 27 81545
TU Delft:
Dr. Ir. C.B.M. Blom
Afdeling Bouw, Faculteit Civiele Techniek en Geowetenschappen, TU Delft
C.B.M.Blom@tudelft.nl
Tel.: +31 (0)15 27 84324
Volker InfraDesign:
Philip Labouchere
Ontwerpmanager
plabouchere@volkerinfradesign.nl
Tel.: +31 (0)348 435111
A.H. Oldenhave 2
Starting a large diameter TBM from surface 28 april 2014
Preface
This report represents my master thesis for the completion of the master Geotechnical Engineering
at the faculty of Civil Engineering and Geosciences, Delft University of Technology.
For my master thesis different projects were available. In retrospect, I am very glad I decided to
choose this subject. This thesis has allowed me to gain a considerable amount of knowledge; not only
in the tunnelling technique, but also in other geotechnical subjects and implementations of concrete.
There are numerous persons I would like to thank. First of all I would like to thank Sallo van der
Woude, who convinced me to engage into this subject and for his dedication and encouragement.
Further I would like to thank Kees Blom and Philip Labouchere who helped me with insightful
knowledge about topics outside my master scope. Also the suggestions, interest and input of Johan
Bosch and Wout Broere during meetings were very helpful during my thesis.
I would also like to thank the colleagues at Volker InfraDesign, who assisted and supported me by
explaining and gaining me information. Finally, special thanks are extended to my friends and family
and particularly my partner Wiebe Eikmans for their encouragement, patience, support, advice and
contribution.
April, 2014
Angela Oldenhave
A.H. Oldenhave 3
Starting a large diameter TBM from surface 28 april 2014
Summary
In the Netherlands, when a bored tunnel has to be constructed, a launch and a reception shaft have
to be excavated to guarantee sufficient cover. This is necessary to ensure face stability, prevent uplift
and risk of blow out. Moreover, especially in the western part of the Netherlands, the first 10 to 20
meters of soil consists of the Holocene layer consisting of soft clays and peat. By starting the
tunnelling process at sufficient depth, the tunnel will be driven through the Pleistocene and and the
soil deformations will remain small.
In 2009, a new method was introduced in Japan: the Ultra Rapid UnderPass (URUP) method. With
this method the tunnel boring machine (TBM) starts tunnelling from ground surface level.
Consequently there is no more need for a launch and reception shaft. This eliminates the need for
expensive shafts which are time-consuming to build, which makes the URUP method interesting to
apply in the Netherlands. However, applying this method will result in small covers at the beginning
and the end of the tunnel lining. Therefore complications are to be expected.
In this report, the feasibility of this method, applied for large diameter tunnels, is assessed by
investigating the main complications associated with this method (tunnelling with marginal cover
method) in the Netherlands. As a basis for the feasibility study a reference project, which is planned
for the near future, is used. The reference project “Rijnlandroute” is situated in the west of the
Netherlands (between Leiden and Voorschoten), where the first 11 meters consist of Holocene soft
soil. For this thesis only the eastern start of the tunnel is considered, that is located in a polder. The
groundwater level is located approximately one meter below surface level and the surrounding area
is largely greenfield.
Firstly, the soil deformations in the Holocene layer, when placing the TBM at surface level, are
determined. It can be concluded that without ground improvement the settlements will become too
big. It is therefore recommended to apply Mixed In Place columns, that have been applied
successfully several times in the past for similar cases. When the TBM reaches the Pleistocene layer,
ground improvement is no longer required and deformations remain small.
Due to the high groundwater table an embankment is necessary in permanent situation to prevent
uplift. Moreover, the tunnel invert has to be filled with a sand ballast as well. During construction
(ballast is not constructed yet) the water level will be drawn down temporarily with a maximum of
two meters.
Due to the absence of cover at start of the tunnelling process, the force distribution in the lining
differs from the conventional method. The axial loads are negligible, whereas bending moments still
exist resulting in opening up of the joints, which might cause leakage. Therefore it is required to
increase the capacity to withstand high bending moments when tunnelling with shallow covers. The
required embankment will increase the normal force. However, this increase is too low to counteract
the occurring bending moments in the lining. The Japanese box, which is a bolt type where the
tension forces are transferred directly to the reinforcement, will be applied (permanently). This bolt
type can withstand higher forces, though this bolt type will increase the construction costs.
It can be concluded that despite these issues, tunnelling with marginal covers is feasible in the
Netherlands. To gain more insight in the relative benefits of this method compared to the
conventional method, both are assessed, considering the costs as well as the construction time. The
A.H. Oldenhave 4
Starting a large diameter TBM from surface 28 april 2014
comparison shows that the total construction costs only differ 1.8 % (the costs of the conventional
method will be 1.8 % higher). However, the construction of the tunnel with the marginal cover
method will result in a shorter construction time. The construction time of the conventional method
is approximately four years whereas the construction time of the tunnel applying the tunnelling with
marginal cover method only amounts approximately three years. Therefore, it will be efficient to
apply this method in the Netherlands. However, despite the feasibility and efficiency of this method
in the Netherlands, it cannot yet be concluded if tunnelling with marginal cover will be
recommended in future. Further research is still required to conclude if the marginal cover method
will be successfully. With this thesis it can only be concluded that the tunnelling with marginal cover
method, applied for large diameter tunnels, is feasible in the Netherlands.
A.H. Oldenhave 5
Starting a large diameter TBM from surface 28 april 2014
Samenvatting
Wanneer in Nederland een geboorde tunnel wordt geconstrueerd, dienen een start- en
ontvangstschacht gerealiseerd te worden alvorens met het boorproces begonnen kan worden. Deze
schachten zijn noodzakelijk om voldoende dekking boven de tunnel lining tijdens het boren te
kunnen waarborgen. Dit is nodig om opdrijving van de tunnel te voorkomen en voldoende
normaalkrachten in de tunnel lining te kunnen garanderen. Bovendien bestaat, vooral in het westen
van Nederland, de eerste 10 à 20 meter uit slappe Holocene grondlagen (zachte klei en veen).
Wanneer het boorproces op voldoende diepte begint, zal de tunnel grotendeels in de sterke
Pleistocene zandlaag geconstrueerd worden, wat resulteert in geringe zakkingen. Wanneer de
geboorde tunnel is voltooid, worden de toeritten naar het geboorde gedeelte aangelegd.
In 2009 is in Japan een nieuwe boormethode gepresenteerd: de Ultra Rapid UnderPass (URUP)
methode. Bij deze methode start de tunnelboormachine (TBM) op maaiveldniveau, boort en
construeert vervolgens de tunnel lining en eindigt weer op maaiveldniveau. Dit resulteert erin dat de
toeritten achteraf niet meer geconstrueerd hoeven te worden, maar al bij het boorproces zijn
inbegrepen en verkort de totale constructietijd van de tunnel. Wegens deze voordelen, wordt het
interessant om deze methode ook in Nederland toe te passen, maar deze methode resulteert in
geringe dekking aan het begin en eind van de tunnel. Daarom zullen bij het toepassen van deze
methode in Nederland complicaties verwacht worden.
In deze thesis wordt een haalbaarheidsstudie van deze methode (boren met marginale dekking)
verricht, toegepast voor grote diameter tunnels in Nederland. Voor deze studie worden de grootste
verwachte complicaties onderzocht, voor een referentieproject genaamd de “Rijnlandroute”. De
Rijnlandroute bestaat deels uit een geboorde tunnel in het westen van Nederland (tussen Leiden en
Voorschoten), waar de eerste 11 meter van de ondergrond bestaat uit slappe Holocene lagen. Voor
deze thesis wordt alleen de oostkant van de start van de tunnel (inclusief open en gesloten toerit) in
beschouwing genomen, gelegen in een polderlandschap. De grondwaterstand ligt gemiddeld één
meter onder maaiveld en er is geen bebouwing in de directe omgeving aanwezig.
Om opdrijven van de tunnel lining door de hoge grondwaterstand te voorkomen, dient een dijk op
maaiveldniveau te worden aangelegd. Bovendien zal het gedeelte onder het wegdek in de tunnel
gevuld worden met een zandballast. Tijdens de constructie van de tunnel (op dit moment kan de
zandballast nog niet worden aangelegd) zal de grondwaterstand tijdelijk maximaal twee meter
verlaagd worden.
De afwezigheid van een dekking boven de tunnel lining zorgt ook voor een andere krachtverdeling in
de lining. De normaalkrachten zijn klein, waardoor de hoge buigende momenten niet kunnen worden
opgenomen. Dit resulteert in het openstaan van de voegen tussen de verschillende segmenten. Er is
A.H. Oldenhave 6
Starting a large diameter TBM from surface 28 april 2014
dus een noodzaak om de momentcapaciteit te vergroten bij boren met marginale dekking. De dijk op
maaiveldniveau zal de normaalkracht verhogen, maar dat zal niet genoeg zijn om de buigende
momenten, met mogelijk lekkage tot gevolg, op te kunnen nemen. Hiervoor zal een ander bouttype
toegepast worden; de Japanse box. Dit bouttype brengt de trekkrachten rechtstreeks naar de
wapening in het beton waardoor het grotere trekkrachten aankan. Een nadeel van dit type zijn de
hogere constructie kosten.
Ondanks bovengenoemde complicaties kan toch geconcludeerd worden dat het mogelijk is om boren
met marginale dekking in Nederland toe te passen. Om ook inzicht te krijgen in de efficiëntie van
deze methode, is een vergelijking gemaakt tussen de conventionele methode en de nieuwe methode
met het oog op de constructiekosten en constructietijd. Uit deze vergelijking blijkt dat de
constructiekosten van de marginale dekking methode 1.8 % minder zal bedragen. De constructietijd
van beide methoden scheelt meer. De constructietijd van de totale tunnel, wanneer de
conventionele methode wordt toegepast, bedraagt ongeveer vier jaar, terwijl de constructietijd voor
het boren met marginale dekking ongeveer drie jaar bedraagt. Dit betekent dat het efficiënt zal zijn
wanneer deze methode kan worden toegepast. Toch kan nog niet geconcludeerd worden dat boren
met marginale dekking in de toekomst in Nederland aangeraden wordt. Hiervoor is meer onderzoek
noodzakelijk alvorens deze conclusie getrokken kan worden. Met deze thesis kan alleen
geconcludeerd worden dat boren met marginale dekking voor grote diameter tunnels in Nederland
mogelijk is.
A.H. Oldenhave 7
Starting a large diameter TBM from surface 28 april 2014
Table of contents
Personal information ..........................................................................................................................2
Preface ...............................................................................................................................................3
Summary ............................................................................................................................................4
Samenvatting......................................................................................................................................6
1. Introduction .............................................................................................................................. 11
1.1. Motivation......................................................................................................................... 11
1.2. Approach methodology of master thesis ........................................................................... 11
2. Literature study......................................................................................................................... 13
2.1. General information .......................................................................................................... 13
3. Reference design “The Rijnlandroute” [4], [25] .............................................................................. 18
3.1. Introduction of the bored tunnel design ............................................................................ 18
3.2. Soil conditions ................................................................................................................... 19
3.3. Lining geometry of bored part ........................................................................................... 19
3.4. Total length of tunnel ........................................................................................................ 19
3.5. Integration of trajectory .................................................................................................... 20
3.6. Tolerance in alignment ...................................................................................................... 20
3.7. Limitations......................................................................................................................... 21
4. Consequences driving with low cover ........................................................................................ 22
4.1. Deformation of Holocene layers due to driving with marginal cover .................................. 22
4.2. Effects on buoyancy of bored tunnel due to driving with marginal cover ........................... 22
4.3. Effects on tunnel lining due to driving with marginal cover ................................................ 22
4.4. Effects on face stability whilst driving with marginal cover................................................. 23
4.5. Choice of TBM ................................................................................................................... 27
4.6. Effects on torque of TBM due to driving with marginal cover ............................................. 28
4.7. Effects on grouting of the tail void for marginal cover ........................................................ 28
4.8. Launching procedure for driving with marginal cover ........................................................ 29
4.9. Cross-connections ............................................................................................................. 30
4.10. Summary ....................................................................................................................... 32
5. Deformations of Holocene layer due to shallow driving ............................................................. 33
5.1. Introduction ...................................................................................................................... 33
5.2. Pressure caused by the TBM .............................................................................................. 33
5.3. Soil stresses ....................................................................................................................... 34
5.4. Moment and point of application of the TBM .................................................................... 36
A.H. Oldenhave 8
Starting a large diameter TBM from surface 28 april 2014
A.H. Oldenhave 9
Starting a large diameter TBM from surface 28 april 2014
A.H. Oldenhave 10
Starting a large diameter TBM from surface 28 april 2014
1. Introduction
In the last century, a variety of bored tunnels has been constructed worldwide. Due to lack of
understanding of tunnelling in soft soil, the first bored tunnel was started in 1995 in the Netherlands
(see Chapter 2). Since then more than ten bored tunnels have been constructed in the Netherlands.
For a bored tunnel a launch and reception shaft have to be excavated to create sufficient cover on
top of the bored tunnel lining. Recently a new method has been developed in Japan called the Ultra
Rapid UnderPass method (URUP method). With this method the TBM starts from ground level
instead of from a launching shaft. Consequently the launching and reception shafts are no longer
necessary potentially reducing costs and construction time. Therefore it is of interest to examine the
feasibility of the URUP method in Dutch ground conditions.
This thesis builds on an earlier study of the URUP method in the Netherlands of Beijer (2010).
However Beijer considered only a small rectangular envelope for a cyclists and pedestrians tunnel
with small lengths (which was concluded as being feasible). In this thesis, the adaptation of the URUP
method in the Netherlands will be extended to large diameter tunnels, called the “tunnelling with
marginal cover method” in this thesis.
1.1.Motivation
By applying the new method (tunnelling with marginal cover method) instead of the conventional
method, it is expected that the construction time and construction costs will decrease. Therefore it
becomes interesting to determine the technical feasibility and efficiency of this method in the
Netherlands. However due to differences between both methods, the expected complications and
possible solutions have to be determined to define the feasibility. In this thesis, the feasibility and
efficiency of tunnelling with marginal covers, applied in the Netherlands for large diameter tunnels
will be determined.
After determination of the minimum requirements for the tunnelling with marginal cover method,
the costs as well as the planning are evaluated (Chapter 8). This gives insight in the efficiency of the
method in comparison with the traditional method. In order to obtain meaningful results for the
A.H. Oldenhave 11
Starting a large diameter TBM from surface 28 april 2014
traditional method, the information about the Rijnlandroute, applied by the Province of South
Holland, is used. Hereafter the conclusion of this thesis as well as the recommendations is given.
A.H. Oldenhave 12
Starting a large diameter TBM from surface 28 april 2014
2. Literature study
As start of this thesis project, the applied literature study is mentioned in this chapter, as well as the
motivation for this thesis. The literature study is divided in a study of the traditional method and a
study of the “tunnelling with marginal cover” method where the tunnelling process starts at surface
level.
2.1.General information
To determine the differences between the conventional tunnelling method and the tunnelling with
marginal cover method, both methods will be described in this paragraph. Hereafter the differences
are described.
2.1.1.1. Technique
The tunnel boring machine (TBM, see figure below) consists of a steel cylinder (mentioned shield
hereafter) with a cutter head on front and a mixing chamber, a conveyor system, a thrust system and
trailing support mechanisms inside. The cutter head, which is the rotating cutter wheel at the front,
removes the soil on the front. The excavated soil, collected in the mixing chamber and acting as
support pressure, is then transported by a conveyor belt and a screw conveyor or pump system,
depending on the type of shield (see further). The lining consists of precast reinforced concrete
segments, which are transported through the already constructed tunnel by a trailer train. These
segments are then placed by an erector and thrust cylinders keep the segments in place.
A.H. Oldenhave 13
Starting a large diameter TBM from surface 28 april 2014
Furthermore, the thrust cylinders or hydraulic jacks are used to advance forward by pushing off
against the concrete segments. The segments will be connected with adjacent segments, by bolts
(permanent at start and end, temporary in between) and dowels, to complete the segmental tunnel
rings of the lining.
The shield of the TBM is slightly larger (about 10 cm) than the tunnel lining and results in gaps
between the tunnel lining and the soil, when the lining leaves the cylinder. To avoid settlements of
the surface, this so called tail void is filled under pressure with grout, which normally takes place in 4
to 6 points around the tunnel lining.
When cohesive soils are present, the soil might stick to the excavation wheel and will slow down the
progress when using the Slurry shield. For these soils the EPB shield is often used. This shield
contains special cutters at the front to prevent sticking of the soil to the bore front. This type of TBM
does not use slurry and therefore a separation plant is not required (reduced construction costs). The
excavated soil itself is capable to maintain the water pressure and horizontal effective ground
pressure at the front, see Figure 2. The excavated soil is mixed in the excavation chamber to become
homogeneous and easy to transport with a screw conveyor. By means of the screw conveyor the
pressure at the bore front is reduced to atmospheric pressure inside the tunnel. A disadvantage of
this shield type is the setting of the support pressure in the mixing chamber. The soil is difficult to
compress, which means that small changes of volume lead to large changes in pressure. To apply
“real-time monitoring” the small changes can be limited, according to Bakker (2008) [9]. This includes
regulating the discharge of the soil and use of additives to set the pressure.
A.H. Oldenhave 14
Starting a large diameter TBM from surface 28 april 2014
In the Netherlands both types of machines have been used (Table 1). The most important difference
between both types of TBM’s is the difference in pressure of the support fluid, required in the Slurry
shield when driving through sand layers, see next paragraph.
When an EPB shield will be used for construction in theory no blowout will occur. However, when air
support pressure might be needed in case of repair works in the excavation chamber (for instance
the replacement of the cutting teeth on the excavation wheels), the minimum cover is required as
well.
To maintain water tightness at the reception shaft, the TBM will stop in the sealing block (low
strength concrete). The space between the sealing block and tail void will be grouted, that will lead
to a watertight connection. After hardening, the last rings can be constructed. A compression lid is
placed in the reception shaft or the building pit is filled with water to maintain the pressure for the
TBM [9].
A.H. Oldenhave 15
Starting a large diameter TBM from surface 28 april 2014
The method was first developed for the purpose of constructing an underpass in a short period of
time to cross an urban area with a high traffic density. The elimination of launching and reception
shafts enables rapid construction of the underpass (installation of shafts, excavation equipment or
transportation of the soil is no longer required), as well as continuously construction of the lining of
the entire underpass during driving of the TBM. Moreover, it alleviates traffic congestion when the
underpass is constructed. In the thesis of Beijer (2010) a description of a pilot project, executed in
Tokyo, is given. The bored tunnels using the URUP method are executed with an EPB shield, due to
the shallow cover on top to delete the risk of blow out of a slurry.
Meanwhile due to further development of this technique, even circular shaped large diameter
tunnels are constructed nowadays, see Figure 4 and Figure 5. The main difference with the
traditional method is the start at surface level resulting in construction of the tunnel with shallow or
even no cover. The TBM launches itself from a launching frame (see also Paragraph 4.8), installed at
surface level as well. In Figure 4 it is shown that still a shallow building pit is necessary. This is due to
the fact that the road alignment inside the tunnel is located higher than the bottom of the tunnel.
Therefore a shallow depth is still required. However, cover on top of the TBM is missing at the start
of the tunnelling process.
A.H. Oldenhave 16
Starting a large diameter TBM from surface 28 april 2014
Figure 4: TBM launching at surface level [3] Figure 5: TBM arrival at surface level [3]
Also other projects and tunnel shapes are constructed with the URUP method. For more information,
a reference is made to the report of Beijer (2010).
To extend above mentioned feasibility project, this thesis will investigate the possibility of
constructing tunnels with large diameters in the Netherlands (indicated as “tunnelling with marginal
cover method”). Again a reference project will be used where the method will be applied, see
Chapter 3.
When starting at surface level, one has to deal with other soil conditions (mostly soft soil), since the
new method starts at the surface instead of the Pleistocene sand layer when applying the
conventional method in the Netherlands. Especially the soil in the western part of the Netherlands
consists of a Holocene layer on top of the Pleistocene.
A.H. Oldenhave 17
Starting a large diameter TBM from surface 28 april 2014
The new road contains a connection between two highways, the A44 (north-west of Leiden) and the
A4 (south east of Leiden). To limit nuisance due to the new connection, the part between Leiden and
Voorschoten will be constructed as a tunnel, see white arrow in the figure below. The bored tunnel
starts at the east side just before the Schiekanaal and ends on the west side behind the railway. The
ramps, given in yellow and orange in the figure, connect the tunnel with the highways and are
located in agricultural area without constructions. In this thesis only the eastern ramp of the project
will be considered.
Figure 6: Overview Rijnlandroute with optimized tunnel location (white arrow) [4]
A.H. Oldenhave 18
Starting a large diameter TBM from surface 28 april 2014
3.2.Soil conditions
Only limited information is available on the different soil types. The subsurface at the location
consists of a Holocene layer (clay with peat layers) from surface level until approximately NAP -11.0
m. On this level the Pleistocene layer starts. Between NAP -2.5 m and NAP -6.5 m sand layers can be
present as well. The surface level is located at NAP -1.0 m to NAP -1.5 m.
When applying the above mentioned geometry and the geometry given in Figure 8, the top surface
of the road will be located approximately 2.5 meters above the bottom of the tunnel lining. This area
will also be used for the required cables and ducts. The concrete quality will be assumed to be
C35/45. Both tunnels (one for each direction) will be designed as a provincial road with maximum
allowable speed of 80 km/h. For further information a reference is made to the feasibility report of
the bored part of the Rijnlandroute of Movares (2012) [4].
Summed, the total length of the tunnel is 2604 m, including two service buildings, constructed after
the tunnelling process in the building pits. In total seven cross connections will be constructed
between the two tunnels. The locations of the cross connections depend on present buildings at
surface level. The maximum distance between two cross connections amounts 250 m, which is in line
with the Dutch regulations [27]. The dimensions of the cross connections can be found in the report
of Movares (2012) as well, but will not be further discussed.
A.H. Oldenhave 19
Starting a large diameter TBM from surface 28 april 2014
3.5.Integration of trajectory
When the tunnelling with marginal cover method will be applied for this case, sufficient area has to
be present to start the tunnelling process at surface level. In the figure below the start location of the
tunnel is given. When applying the traditional method, the bored part will start just before the
waterway. When applying the tunnelling with marginal cover method, the tunnelling process will
start 300 meters earlier.
3.6.Tolerance in alignment
When tunnels are constructed, deformations will appear in soil (subsidence) as well as in tunnel
lining. The maximum deformations depend on the tolerance in the alignment. In the figure below the
gauge of the tunnel is given; the height is 4700 mm and the width is 7850 mm. According to
Rijkswaterstaat [28] dimensions are dependent on the category of a road. The Rijnlandroute is
designed as a primary road with a design speed of 80 km/h. This leads to category IV and a lane
width of 3.10 meters. However, when the tunnel is located in agricultural area, other (wider) vehicles
can be expected as well. It will be expected that no tolerance is acceptable in horizontal direction.
The minimum required height for a tunnel amounts 4.50 meters. This consists of a maximum height
of a truck of 4.00 m, a vertical movement of 0.20 m and a safety margin of 0.30 meter. However, in
Figure 8 4.70 meters is present. This means a tolerance in vertical direction of 20 cm is allowed.
To calculate the maximum tolerance in diagonal direction, the influence of the minimum required
height and the present height are compared:
Minimum diameter is √ m.
A.H. Oldenhave 20
Starting a large diameter TBM from surface 28 april 2014
This results in an allowable tolerance of 9.15 – 9.05 = 0.10 m = 100 mm. This is in accordance with
the report of the Rijnlandroute.
However, the total deformations are the result of settlements of the tunnel in the soil (when the
tunnel will be drilled at an angle of 4.5 % but the tunnel lining will settle and results in adjustment of
the height of the lane) and deformations of the tunnel lining. For the tolerance in deformations of
the tunnel lining typically 1 % of the radius will be assumed, amounting 50 mm. It will be stated that
the maximum settlements in the soil amounts 50 mm as well as the maximum deformations in
tunnel lining.
3.7.Limitations
On the next page the scope of this thesis is drawn (see blue part). For this feasibility project only the
length of the eastern ramp (start side) of the Rijnlandroute is considered. It is assumed that when
constructing the tunnel with the tunnelling with marginal cover method on this side is feasible, it will
not give complications at the west side of the tunnel (finishing side). The bored part of the traditional
method is pretended to remain the same. Moreover, as limitation the data (dimensions, location
etc.) are assumed to remain the same as the original Rijnlandroute. Therefore no other tunnels
(shape, diameter) or locations are considered.
It has to be said that this thesis only serves as a first investigation of the adaptation of the tunnelling
with marginal cover method in the Netherlands. Further research is strongly recommended.
A.H. Oldenhave 21
Starting a large diameter TBM from surface 28 april 2014
A.H. Oldenhave 21
Starting a large diameter TBM from surface 28 april 2014
For this thesis it is assumed that the first three major effects will determine the feasibility of the
method in the Netherlands. When these three major effects cannot be solved, feasibility in the
Netherlands is assumed to be unattainable. The three main consequences will be discussed
respectively in more detail in subsequent chapters.
The measures result in a modification of the soil conditions (stresses, strength). The modified soil
conditions will be taken into account for further calculations. Therefore the deformations of the soil
will be determined at the start.
To prevent uplift of the tunnel lining, measures can be applied as well. This can be done by applying
extra weight on top, lowering the groundwater table, applying thicker lining segments or applying
extra ballast inside the tunnel. For this thesis however it will be assumed that all assumed dimensions
for the traditional method will also be applied for the new method (tunnelling with marginal covers).
Therefore, the tunnel lining segments as well as the diameter of the tunnel remain the same.
In Chapter 6 the upward forces and downward forces will be calculated for the first part of the tunnel
where shallow or no covers are present. Safety factors will be taken into account as well. With these
calculations it can be concluded if the buoyancy forces can be counteracted to retain the feasibility of
the tunnelling with marginal cover method.
A.H. Oldenhave 22
Starting a large diameter TBM from surface 28 april 2014
will result in unequal force distributions (low normal forces and high bending moments). The normal
force on top is missing which might result in high deformations in the lining. Moreover, due to high
bending moments this can result in gaps between the joints (the pressure zone decreases). These
two effects have to be countered.
In Chapter 7 the effects on the tunnel lining will be calculated at different depths. Due to the number
of calculations only two critical cross sections will be used as well as the cross section where the
traditional conditions apply (+ 1D cover). The measures that are discussed in Chapters 5 and 6 will be
taken into account. The deformations for each depth as well as the amount of reinforcement and
bolts will be determined with the help of a framework analysis to calculate the normal forces, the
shear forces and the bending moments in the tunnel ring.
The jacking forces can be arranged separately for each jack. This results in lower jacking forces in the
top of the ring and higher forces at the bottom (the soil pressures increase by depth). For each depth
the minimum and maximum support pressures can be calculated. As starting point the starting depth
of 2.5 meter below surface will be calculated. Furthermore the three cross sections used for
calculations in Chapter 7 will be calculated as well.
A.H. Oldenhave 23
Starting a large diameter TBM from surface 28 april 2014
To prevent collapse of the soil above the top or deformations at surface level, the jacking pressure
has to be smaller than the passive support pressure. The passive support pressure on top is [10]:
( ) ( )
( ) ( )
Above calculated support pressure give the limits in which the jacking pressures have to remain.
When the jacking pressure becomes too high, the soil will not be excavated which might lead to
deformations. When the jacking pressure becomes lower than the active support pressure, the soil
will react towards the TBM.
When applying the four different cross sections (see Table 2 and Chapter 7; start, half below surface,
small cover and conventional start), the minimum and maximum pressures can be calculated for
each depth (in Table 18 the estimated values for each depth are given. For the starting point the
tunnel is only located in the first layer, for these values see Chapter 5). The mean, minimum and
maximum groundwater levels are all calculated as well as the minimum and maximum pressures on
top and at the bottom of the tunnel lining. The required ground improvement is taken into account.
The results are given in the table below.
Groundwater levels [m Pmin, top Pmax, top Pmin, bottom Pmax, bottom
below surface] [kN/m2] [kN/m2] [kN/m2] [kN/m2]
bottom tunnel 0.6 0 0 30.5 52.1
2.5 m below 1 0 0 28.8 54.9
surface level 1.4 0 0 27.2 57.7
bottom tunnel 0.6 0 0 57.3 228.2
2.5 m + 3.82 m 1 0 0 55.7 231.0
embankment 1.4 0 0 54.0 233.8
bottom tunnel 0.6 0 0 106.4 350.0
6 m below 1 0 0 104.7 352.8
surface level 1.4 0 0 103.1 355.6
0.47 m cover 0.6 16.7 93.9 162.1 391.3
1 16.7 93.9 160.4 394.0
1.4 16.7 93.9 158.8 396.8
10.47 m cover 0.6 126.4 234.3 257.9 563.6
1 124.8 237.1 256.2 566.4
1.4 123.1 239.9 254.6 569.2
It is clear that especially at the start the range between the minimum and maximum pressures is the
smallest. Here the jacking pressures have to be controlled very precisely in such a way that the
pressure outside the boundaries will not be reached. When the depth of the TBM increases, the
range for possible jacking pressures becomes wider due to the larger range of active and passive
A.H. Oldenhave 24
Starting a large diameter TBM from surface 28 april 2014
support pressure and fewer problems will be expected. To increase the range between the minimum
and maximum pressure, the required embankment to prevent uplift (see Chapter 4) is extended
towards the start location. This results in a bigger range due to higher soil pressures on the bore
front, see Table 2, and therefore an extension of the required embankment will be recommended to
prevent failure of the support pressure.
For this calculation it is assumed that the groundwater tables are not lowered. However, after
calculation it appeared that a groundwater level decrease of 2 meters is necessary. This is not taken
into account in this calculation, because this effect will only increase the range between the active
and passive support pressure. Moreover, the critical point is located at the start where the
groundwater level is not lowered yet. However, the required embankment (see Chapter 6) and
ground improvement (see Chapter 5) are included.
( )
( ( ) )
r’ is the horizontal effective soil pressure at the centre of the bore front (or at the centre of
the part below surface level);
fw is the reduction factor due to friction between soil and steel, about tan() = tan(22) = 0.4
according to Tan et. al. (2008) [16];
f is the stress reduction factor due to the conical shape of the TBM (0.8);
is the internal friction angle for the soil;
a is the adhesion between the soil and the steel frame of the TBM, which amounts 0.853 kPa
when the temperature of the soil is 5C [17];
fbent is the reduction factor for smearing medium (assumed to be 1 for EPB);
lTBM is the length of the TBM;
DTBM is the diameter of the TBM;
R is radius of the TBM;
H is depth of the bottom of the TBM below surface level.
The horizontal effective soil pressure at the centre of the TBM depends on the depth. In the table
below the friction force for the four different depths (see Table 2) is calculated. By dividing this force
by the area of the tunnel lining, the minimum required jacking pressure to overcome the friction is
calculated, see last column in Table 3. When the TBM is not totally covered, the part above surface
level will not contribute to the friction force. This is taken into account as well (last part of the
formula).
A.H. Oldenhave 25
Starting a large diameter TBM from surface 28 april 2014
Depth bottom TBM [m] Friction part [-] Fshield, friction [kN] Pshield, friction [kN/m2]
2.5 (no embankment) 0.324 257 20
2.5 (with embankment = 6.32) 0.564 1122 88
6 (with embankment = 9.82) 0.833 2069 163
11 (with embankment = 12.77) 1 2616 206
21 1 2995 235
The decisive cross section is the starting point. To increase the range of the possible jacking pressure
(during the tunnelling process, the jacking pressure can fluctuate) and to guarantee water tightness,
the embankment has to be extended towards the starting point of the tunnelling process. In the
table below the pressures are transferred into forces (when the tunnel is located only partially in the
ground, it is easier to calculate the forces instead of pressures). The minimum jacking force to
guarantee water tightness is 955 kN. The minimum force to overcome the friction between the steel
frame and the soil and embankment is 1122 kN. When the maximum support pressure at the bottom
amounts 228 kN/m2 (see Table 2) and decreases to 0 where the TBM crosses the embankment, this
results in a force of approximately 6220 kN. It can be concluded that the minimum force to overcome
the friction and guarantee water tightness of the segments is smaller than the maximum support
A.H. Oldenhave 26
Starting a large diameter TBM from surface 28 april 2014
force, see table below. It is also clear that without an embankment the friction forces and jacking
forces cannot be counteracted by the maximum support pressure.
Depth bottom TBM [m] Fmin, jack [kN] Fshield, friction [kN] Fmin, support [kN] Fmax, support [kN]
2.5 (no embankment) 955 257 240 410
2.5 (with embankment = 6.32) 955 1122 1570 6220
6 (with embankment = 9.82) 955 2069 2990 9860
11 (with embankment = 12.77) 955 2616 4700 11330
21 955 2995 7480 16340
In the table above also the minimum support forces are given. The jacking forces have to overcome
this minimum support force to guarantee excavation of the soil plus the friction of the shield. When
the maximum jacking forces are in the order of some MN (compare with minimum jacking force of 55
kN when applying two jacks per segment), it will not be expected that the friction and the minimum
support pressure cannot be overcome by all the jacking forces.
When applying the embankment at start level, the jacking forces can provide enough force to
counteract the friction and the soil pressure on front. It will be recommended to apply a jacking
pressure equal to the mean support pressure on front Pmean, support (see formula below) + Pfriction if this
jacking pressure is higher than the minimum jacking pressure of 75 kN/m2 to guarantee water
tightness.
( ( )) ( ( ))
kN/m2
Remark
When the minimum jacking pressure of 75 kN/m2 will be applied on top of the TBM, the moment will
increase. That influences the total settlements underneath the TBM (see Paragraph 5.4.1 and
further). However, after calculation, the maximum settlements below the TBM will only increase by 1
mm. Due to the high weight of the TBM the extra jack force will not influence the moment and
settlements significantly.
4.5.Choice of TBM
Originally the bored part of the Rijnlandroute would start approximately 16 meter below surface. The
top of the tunnel is then still located in the Holocene. However, most of the tunnel lining is already
located in the Pleistocene sand, see Figure 9 (grey part). Therefore for the Rijnlandroute the slurry
shield will be used. However, when applying the tunnelling with marginal cover method, the bored
part will already start entirely in the Holocene. Moreover, to delete the risk of a blowout an EPB
shield will be used.
The application of the EPB shield in the Holocene will not result in problems. However, when the
TBM reaches the permeable Pleistocene, requirements to prevent inflow of groundwater are
necessary. For the Botlekspoortunnel an EPB shield was used (Table 1). The launching and arrival
A.H. Oldenhave 27
Starting a large diameter TBM from surface 28 april 2014
shaft were still located in the Holocene, but during the tunnelling process the TBM was driving full
face through medium to coarse grained Pleistocene sands. For this project foam conditioning was
successfully applied to support the Pleistocene sands with higher water pressures. This foam is
biological and therefore separation of the soil is not necessary. The foam ensures good
transportation through the excavation chamber and more important creates impermeability of the
earth paste in the excavation chamber. When the Pleistocene at the Rijnlandroute becomes too
permeable for transportation of the soil and the tunnelling process, foam will be added as well.
When the tunnel lining is located partly above surface level, this results in complications to inject the
grout in the tail voids. The grout will flow out on surface level. By applying fast curing mortar instead
of normal grout, the outflow will decrease. Also when more locations for injecting the mortar will be
applied (normally 4 to 6, see Chapter 2) and will be arranged separately, only the bottom will be
injected at start. During increase of depth of the tunnel more and more injections will be activated,
see Figure 10.
A.H. Oldenhave 28
Starting a large diameter TBM from surface 28 april 2014
Figure 10: Activated injection locations when tunnel lining is partly above surface
Figure 11: Launching of the TBM at steel frame (yellow) in Figure 12: Steel frame (yellow) connected to the sheet pile
launching shaft [40] wall on front [40]
When the TBM will launch on the steel frame, this will result in a force distribution given in Figure 14.
According to this distribution, the steel frame will have tension forces at front resulting in failure of
A.H. Oldenhave 29
Starting a large diameter TBM from surface 28 april 2014
the concrete floor when reinforcement will not be constructed. However, when the steel frame will
be poured into the concrete and connected with reinforcement present in the concrete floor (see
Figure 15), the structure will only fail when the yielding stress of steel is reached. For this thesis it will
be assumed that the forces acting on the steel frame are below the boundaries of failure (yielding of
steel or failure of the concrete floor by the pressure forces). However further research is
recommended. To counteract the pressure forces on the concrete floor, foundation piles will
transport these forces towards the strong Pleistocene layer, see figure below.
4.9.Cross-connections
According to the Dutch regulations the maximum distance in longitudinal length between two cross
connections in a tunnel amounts 250 meters [27]. The “first” cross connection will be constructed at
A.H. Oldenhave 30
Starting a large diameter TBM from surface 28 april 2014
surface level where the alignment of the road is at the same height, see Figure 16. At this location
the bottom of the tunnel is constructed at 2.5 meters below surface, i.e. the launching shaft. At this
location the tunnel lining above surface level will be removed after completion of the bore process.
Therefore, for this cross connection no real constructions are needed. Thus, at this location the two
tunnels can be constructed with enough distance to minimize any influence on the other tunnel
(unknown for the new procedure).
The real tunnel part starts when the tunnel lining above surface level will not be removed anymore.
According to Table 52 the tunnel lining removal ends when the bottom of the tunnel is located 4
meters below surface level; 1.5 meters deeper. This results in a start of the tunnel at 33 meters
distance from the launching shaft. Therefore, the next cross connection will be constructed 283
meters forward from launching shaft. The bottom of the tunnel is located at 15 meters below
surface; on top of the tunnel lining a cover of approximately 5 meters is present. It will be expected
that the cross connection at this location can be constructed as well as the other cross connections
designed by Movares (see Figure 17). At this location the tunnels are less far removed from each
other to minimize the construction costs of the connections.
For the construction of the second cross section with shallow cover, the freezing method will be
used. The shallow depth is advantageous for this method; the freezing lances can be injected
underneath the tunnel lining more easily, resulting in better freezing of the whole soil around the
tunnel. When the lining is located deeper into the soil, the freezing method becomes less reliable.
Then it is more unlikely that the whole soil underneath the lining is thoroughly frozen. This can result
in groundwater inflow during construction of the connection. For the deepest cross connections it is
therefore recommended to inject the soil with jet grouting, which is also advised by Movares for the
Rijnlandroute.
A.H. Oldenhave 31
Starting a large diameter TBM from surface 28 april 2014
4.10. Summary
In this chapter nine consequences are described when applying the new method (tunnelling with
marginal covers) instead of the traditional method:
The first three mentioned consequences are expected to determine the feasibility of this method in
the Netherlands; when these three major effects cannot be solved, the tunnelling with marginal
cover method cannot be applied in the Netherlands. For this thesis these three major consequences
will be further evaluated in Chapter 5, 6 and 7. The other six aspects are mentioned in this Chapter
and can be solved relatively easily. Therefore, it can be concluded that when the first three major
consequences can be solved, the remaining six aspects will not result in problems for the feasibility
of the tunnelling with marginal cover method in the Netherlands.
A.H. Oldenhave 32
Starting a large diameter TBM from surface 28 april 2014
5.1.Introduction
When starting the tunnelling process at surface level, i.e. on top of the Holocene layer, large
settlements will be expected due to the extreme weight of the TBM. As start the soil characteristics
will be determined. Unfortunately soil investigation was not available. With help of DINO loket [37],
which is a Dutch software program, the soil characteristics and its profile are estimated (see
Appendix B). This program contains all the information of soil investigations of the past. This
estimation will be used to calculate the settlements and to determine possible solutions.
1
Was not applied at the Hubertustunnel (slurry shield was used) so the weight of the screw conveyor at the
Botlekspoortunnel is applied.
A.H. Oldenhave 33
Starting a large diameter TBM from surface 28 april 2014
It is assumed that the part A+B+C causes the decisive deformations. The cutting wheel is connected
to this part via a bearing structure and therefore also contributes to the deformations. So the total
weight is 955 ton which works over a length of 10.235 meters. Furthermore, it is assumed that the
same width which contributes to the break-up calculation (see Appendix E), contributes here to the
soil deformations. So width = ( ) ( ) m.
Now all the data are known to determine the pressure caused by the heavy part of the TBM. The
pressure for this part becomes:
5.3.Soil stresses
Different methods can be used to estimate the stresses of the soil below a foundation. In Appendix A
Newmark’s chart is explained as well as the Boston Rule for uniform footing (B x L) [6]. It can be
concluded that both methods result in the same order of soil stresses. For this thesis the Boston Rule
will be applied to determine the required soil stresses.
A.H. Oldenhave 34
Starting a large diameter TBM from surface 28 april 2014
When the depth increases, the area of the stress distribution also increases and therefore the value
of the additional stresses 'z will decrease. Therefore 'z is depending on the depth. With the
following formula 'z can be calculated:
So when depth z increases, the additional stresses decrease. The total stresses in the new situation
become 'z = 'z;0 + 'z.
This results in Table 6. The value for the effective specific weight, ’, is estimated to be 14 kN/m3 for
the dry Holocene layer and 14-10=4 kN/m3 for the saturated Holocene layer.
A.H. Oldenhave 35
Starting a large diameter TBM from surface 28 april 2014
The increase of the stresses in the soil is smaller when it increases with depth. Therefore the
additional stresses (due to additional weight of the TBM) have the biggest consequences in the first
meters.
When, per part of the TBM, the weights are known, the total force and the point of application can
be determined, see Figure 20 and Figure 21. Here, only the known forces will be used to calculate the
point of application. The real force working on the soil will be assumed to be 9,550 kN. The width of
the TBM will be disregarded.
∑ ( ) ( ) (
) tonmm
To determine the exact location where this resultant force works, the moment has to be divided by
the total force: mm.
The total length of the TBM is 11,245 mm = 11.245 m. This means the geometric centre of the TBM
lies at 5.6225 m (see figure). The length between the point of application and the centre is 5.6225 -
3.8193 = 1.8032 m.
A.H. Oldenhave 36
Starting a large diameter TBM from surface 28 april 2014
Remark
The jacking forces influence the moment. However, it will be expected that these forces are in the
same order and for this reason disregarded. Nevertheless, when the first segments of the tunnel ring
have to be placed, the jacks temporary have to be released and no jacking forces can be carried out.
The ring consists of seven segments + one keystone. Due to this, it will be assumed that the
remaining jacks will carry the released jacking force. Therefore this influence can be disregarded.
( )
kN/m2, see Figure 23.
A.H. Oldenhave 37
Starting a large diameter TBM from surface 28 april 2014
5.5.Soil investigation
The deformation of the soil depends on the stiffness of the different soil layers (see Terzaghi’s
formula, next paragraph) and therefore a soil investigation is required.
To determine the structure of the soil at the desired location, six different cone penetration test
results were used from DINO loket [37], which collects all kind of information about the underground
of the Netherlands. Unfortunately no information at the exact location is present yet. So by
interpolating the information (CPT results) of the surroundings a reliable estimation of the soil can be
made.
Table 7: Friction ratio and cone resistance for different kind of soil types [6]
The results of each CPT can be found in Appendix B: CPT results. The six different locations show
diversity in the layer structure of the soil. Especially the CPT’s measured along a road have stronger
layers at the top. The two CPT’s measured in the polder (S30H00674 and S30H00677) contain
especially clay and peat with a stronger silty layer between approximately 4 and 7 meters below
surface. This is in accordance with the geological information of DINO loket, see Appendix C:
Geological results. Therefore, the two CPT’s of the polder are used to estimate the mean soil
structure.
In Figure 24 and Table 8 the estimated soil structure and its description can be found. It will be
assumed that this soil structure is plausible for the location.
A.H. Oldenhave 38
Starting a large diameter TBM from surface 28 april 2014
( ) ( )
With this formula the deformations at each depth can be calculated using the stress in that layer and
the primary compression coefficient Cp and the second compression coefficient Cs.
( )
Where:
( )
The values for the compression coefficients are given in NEN 6740 (Table 1) [30] and below in Table
9. It is assumed that after one month (i.e. 30 days) the compression is equal to at least 90 % of the
total settlements of the cohesive layer. Another assumption is the infinity of the stiffness of the
Pleistocene layer. It will be assumed that the settlements will only be located in the Holocene layer.
It appears that when a layer of 10 meters sand will be added to the calculation, this layer will only
settle 2 mm. It can be concluded that assuming the Pleistocene sand layer as an infinite stiff layer is a
correct assumption. The calculations will be executed for the Serviceability Limit State (SLS). For the
Ultimate Limit State (ULS) the compression constants have to be divided by the partial factor m =
1.3.
The TBM will start at 2.5 meters depth. This means only 0.5 meters of peat remains below the TBM
and part of the TBM is present in the groundwater (the differential effective stresses ’z will
decrease, therefore total settlements will decrease). For this calculation the lowest groundwater is
taken into account (1.4 meters below surface level). This means the smallest upward pressures and
A.H. Oldenhave 39
Starting a large diameter TBM from surface 28 april 2014
therefore the maximum settlements. It can be concluded that even without the peat layer the
settlements remain too big, see Table 9. The tunnel cannot be constructed without any ground
improvement.
The calculations are executed for the centre of each layer. For the Boston rule the round TBM is
supposed to be a plate with dimensions 8.066 m x 10.235 m. When the bottom of the TBM is below
groundwater level, the buoyancy forces are very small due to the small volume below groundwater
level. For this calculation the height from bottom TBM to width 8.066 m is removed from the depth
below groundwater level. The figure below clarifies this phenomenon.
According to the figure, only when the bottom of the TBM is located more than 1.88 meters below
groundwater level, the buoyancy forces will counteract the weight of the tunnel. For the start (2.5 –
1.4 = 1.1 meters below groundwater level) no buoyancy forces are included.
For example the differential effective stress of layer 3 (see Table 9) is calculated as:
( ) ( )( ) ( )( )
kPa.
Table 9: Total settlements Holocene layer with TBM at 2.5 meter depth
A.H. Oldenhave 40
Starting a large diameter TBM from surface 28 april 2014
improvements are a soil replacement, an embankment, jet grouting or mixed in place. These
phenomena will be summed below.
Table 10: Total settlements Holocene layer with embankment (5 meter) and soil replacement
The soil can be stabilized by columns (column stabilization), see figure, or by applying MIP for the
whole ground volume (block stabilization). Both methods can be combined as well, but require
different machines (the block stabilization does not consist of columns). The columns can be applied
until a depth of 25 meters. The block stabilization (mixing a ground volume) can only be applied for 5
meters depth. The columns can be constructed using the ‘dry’ and the ‘wet’ method. With the dry
method binding suspension will be injected with air pressure; for the wet method the suspension will
be mixed with water (it becomes a slurry) before it enters the soil.
A.H. Oldenhave 41
Starting a large diameter TBM from surface 28 april 2014
When applying stabilized columns the strength will contribute to the surrounding non-stabilized soil
when the strength of the column is not higher than circa 20 times the strength of the surrounding
soil (to prevent that the columns will operate as a pile foundation). When a load is applied on top of
the mixed soil, the columns will contribute more in the load distribution, causing an arc effect, see
Figure 27 (the arrows show the contribution of the non-stabilized soil to the MIP columns).
Figure 27: Force distribution in the soil (brown is MIP columns, blue is non-stabilized soil)
The advantages of applying MIP as ground improvement are the low construction costs (production
rate, no soil removal) and the improvement of the geotechnical properties of the Holocene layer.
Disadvantages are the limitations in depth and hardening time of the mixed soil. Moreover, it is
difficult to monitor the increased strength of the soil. This is depending on the homogeneity of the
mixing process and the properties of the soil layers. However, different reference projects show good
results.
5.6.1.3.1. Botlekspoortunnel[33]
The first reference project is the Botlekspoortunnel. This tunnel is mostly located in the Pleistocene
layer. The ends however, connecting the bored part to the shafts, are located in the Holocene. At the
east side the strength of the clay and peat layers were too small and ground improvement was
necessary to decrease the settlements to an acceptable level and create sufficient support for the
concrete tunnel lining. Initially jet grouting was scheduled. However, due to financial benefit, better
soil conditions to bore through the improved part and less removal of soil, the MIP method became
an interesting option.
A.H. Oldenhave 42
Starting a large diameter TBM from surface 28 april 2014
The two possible methods for the MIP (wet and dry method) were prior to the construction on site
tested in a pilot project next to the location of the tunnel. For the wet method 100 % cement was
injected in the soil. The dry method used 80 % cement and 20 % anhydrite, together blown into the
soil. After the pilot project and different tests it was concluded that the wet method would be
recommended. This method ensures more homogeneity in the improved soil, as well as in horizontal
direction as vertical direction, and the quality of the ground improvement can be checked more
easily.
The stabilized area was 11 x 30 x 125 meters and the columns were executed from NAP – 8.0 meters
to NAP – 19.0 meters (23.5 meters below surface). In total 4,500 columns are installed with a
diameter and hart to hart distance of 800 mm. For every kind of soil the amount of cement is
measured. After installation of the columns measurements were performed in order to determine
the compressive strength of the improved layers. The measurements are given in the table below.
After this project it was concluded that the required strengths are achieved. The homogeneity of the
soil was enough and the settlements remained below the limits. Yet, additional investigation is
recommended to determine the influence of the surrounding area and the variation in the results.
The construction was carried out in 10 months, by two machines fulltime. A third machine was not
used to decrease the amount of heave. The total settlements remained below 50 mm.
A.H. Oldenhave 43
Starting a large diameter TBM from surface 28 april 2014
were too weak to construct a tunnel lining. The stability of the soil had to be increased by MIP
columns.
The columns had to be constructed singular, in rows with overlap, in a triangle pattern and as block
stabilization. The columns were 14 to 18 meters long, with diameter 600 and 800 mm. The block
stabilization was executed until the Pleistocene layer. For this, it was expected that the settlements
were minimal.
After installation of the columns some measurements were executed. It could be concluded that due
to the MIP method the strength of the soil was increased.
For the Rijnlandroute the wet method will be recommended, especially due to the more
homogeneity in vertical direction. The differences in strength of the layers become smaller and
therefore less differential settlements will be expected. The last clay layer above the Pleistocene
layer determines the greatest settlements. Therefore it will be recommended to apply the MIP
columns until the Pleistocene layer. The Pleistocene layer will be assumed as an infinitely stiff layer.
When assuming the same compressive strength of the Botlekspoortunnel (Table 11), the maximum
present pressure (281.0 kPa) remains lower than the strength of the soil. This means the soil deforms
elastically and it follows . Again in Table 12 below, the soil characteristics are estimated
with table 1 of NEN 6740. The TBM will start at 2.5 meter depth, resulting in a remaining peat layer
of 0.5 meter. To estimate the bedding stiffness Eoed of the soil, the following formula is used:
( )
( )( )
A.H. Oldenhave 44
Starting a large diameter TBM from surface 28 april 2014
As earlier mentioned the difference between the strength of the soil and the strength of the columns
may not differ more than 20 times to keep the contribution of the strength of the soil. The strength
of the soil or the stress is linear with Young’s Modulus according to Hooke’s law and therefore the
difference between Eoed; soil and Eoed; column is smaller than 20 times as well. According to CUR report
2001-10 it is assumed that column = soil. When assuming a difference of 15 times, the mean Eoed of the
columns can be assumed: kPa.
The strength increase of the soil depends on the amount of columns (hart to hart distance) and the
diameter of the columns. This is reflected in an enhancement factor a. When applying a triangular
pattern, this factor can be calculated according to:
Where:
In final situation the settlements of the soil is equal to the settlements of the columns (the system is
in balance), due to the higher stresses on top of the columns and smaller stresses on top of the soil.
The total settlement of the surface can be calculated with:
∑
( )
Where q is the load applied on top of the soil (281 kPa due to weight of the TBM) and w m is the
settlement of the soil as well as the columns. When now assuming a hart to hart distance and a
diameter of the columns, the settlements can be calculated, see further. These settlements have to
remain below 50 mm (see Paragraph 3.6).
According to the CUR report the settlements for the soil and columns can also be calculated
separately, depending on the load for each kind. By assuming that w m = w1 = w2, where w1 is the
settlement of the column and w2 is the settlement of the soil, the loads q1 and q2 can also be
calculated.
kPa.
A.H. Oldenhave 45
Starting a large diameter TBM from surface 28 april 2014
When assuming a maximum bearing capacity ult;col of 440 kPa, the maximum allowable load for one
column amounts . So q1 < q1;max.
With above mentioned conditions finally the optimal dimensions of the columns can be calculated,
see bold row in Table 13. It appears that the maximum allowable load q1;max is decisive for the
calculations. When applying 800 mm columns with a hart to hart distance of 900 mm the total
settlements only amount 26 mm (< 50 mm). However, q1;max is 284 kPa and q1 is 274 kPa. The
additional soil stress q2 is 7 kPa. In the table below other possibilities are given as well.
In Appendix D the settlements for different depths of the TBM are calculated. When depth increases
indeed the additional stress becomes smaller, however even when applying the TBM half a meter
above the Pleistocene in the clay layer, the settlements remain too big (80 mm > 50 mm), see Table
47. Therefore it can be concluded that ground improvement (Mixed In Place) is necessary when the
bottom of the TBM is still located in the Holocene layer. To conclude whether the bearing capacity of
the Pleistocene is still high enough, an extra calculation is added, see Table 48. When the bottom of
the TBM rests at the Pleistocene the TBM will only settle 1 mm. The assumption that the Pleistocene
layer is infinitely stiff is therefore correctly assumed.
When the TBM starts at 2.5 meters depth, reaches the Pleistocene at 11 meters depth and the TBM
bores at an angle of 4.5 %, the total length the ground improvement has to be applied amounts
( )
m.
According to the Boston rule, the distribution of the stresses in the soil develops like the figure below
(2:1 in depth). At the starting point the depth of the Holocene layer is 8.5 meters. However, when
assuming a rectangular plate this plate will be lying 1.88 meters higher. Therefore the thickness of
the Holocene layer amounts 8.5 + 1.88 = 10.38 m. The total width of the influence area becomes
meter. The total area of MIP for the east side is 200 x 18.5 m2
for each starting point (two areas at the east side, see also Chapter 8). At the west side the tunnels
will end in the same building pit, 40 meters width. There the total width for applying MIP is 5.19 + 40
+ 5.19 50.5 meters.
A.H. Oldenhave 46
Starting a large diameter TBM from surface 28 april 2014
On the next page the required ground improvement is schematized in a figure (Figure 32),
representing the total research area of this thesis, i.e. the eastern ramp of the Rijnlandroute. It can
be concluded that when applying MIP columns, the deformations remain below the maximum of 50
mm. Therefore, the jacking forces do not have to participate to control the alignment of the lining
with high forces. Still small differences will be applied due to the higher support pressures and
friction at the bottom of the lining.
A.H. Oldenhave 47
Starting a large diameter TBM from surface 28 april 2014
Figure 32: Length of required ground improvement for application of the “tunnelling with marginal cover” method
A.H. Oldenhave 48
Starting a large diameter TBM from surface 28 april 2014
6.1.Introduction
The main reason to start the bore process on a specific depth in a building pit is to prevent uplift
forces. When starting the bore process on surface level, the high groundwater table (0.6 m below
surface is decisive) will cause uplift forces. Therefore, when starting at surface sufficient
countermeasures have to be taken into account to prevent any upwardly directed forces at any
location of the tunnel.
The road alignment in the tunnel will be located 2.5 meters above the bottom of the tunnel lining.
Therefore the actual bore process will start at 2.5 meters depth instead of surface level (see Figure
33).
For the calculations the ground improvement is taken into account. In CUR 2006-2 the amount of
added cement is given for each soil type (see fifth column in Table 14). This added by the initial
weight factors leads to the following results, required for the uplift calculation.
A.H. Oldenhave 49
Starting a large diameter TBM from surface 28 april 2014
The soil structure given in the table above is only applied for the part where the tunnel lining is still
located above the Pleistocene layer. When the top of the tunnel is located 0.47 meter below surface,
the lining will cross the Pleistocene and the ground improvement is not necessary any more (4th
column of Table 14 will then be used for calculation). This results in less weight of the untreated soil
on top of the lining and therefore higher buoyancy forces, see tables in Appendix E.
The possible countermeasures (embankment, lowering of groundwater table, ballast) are different
for the temporary and the permanent situation, see next paragraphs.
6.2.Safety factors
In the Netherlands directives for the design of constructions (in Dutch: “Richtlijnen Ontwerp
Kunstwerken) are drawn. In this report a description of guidelines can be found for each kind of
construction including safety factors. According to these directives the following safety factors have
to be used for the uplift calculations [35]:
The safety factor of the groundwater level is 1.0. This safety factor applies for the “worst case
scenario” or the highest groundwater table for buoyancy calculations. According to the report of the
Rijnlandroute the highest occurring groundwater table lies at NAP -2.10 m. The surface level varies
between NAP -1.0 m and NAP -1.5 m. Therefore for the uplift calculation a groundwater level of 0.6
m below surface level will be assumed.
Moreover, according to the Dutch “Richtlijnen Ontwerp Kunstwerken, ROK” [35] a temporary
removal of a soil layer with a thickness of 0.5 m above the tunnel should be taken into account. This
is only taken into account for the permanent situation.
A.H. Oldenhave 50
Starting a large diameter TBM from surface 28 april 2014
Two different starting points are possible; to minimize the change in groundwater level and to limit
the height of the temporary embankment. Both scenarios are calculated, including the above
mentioned safety factors. However, in final situation an embankment is always necessary and
therefore this will be used as starting point. As limitation the height of the permanent embankment
is taken into account. When further measures are required, the water level will be decreased. This
will lead to the following calculation; see Appendix E and Table 15.
With the calculations it can be concluded that as well as an embankment as water level decrease is
necessary to avoid uplift. In the table below a summary for the temporary situation is given. In Figure
35 the summary of the minimum required measures is more clearly presented in a drawing. For the
construction of the tunnel with the tunnelling with marginal cover method, a higher embankment
will be recommended. In the figure only the minimum required heights of the embankment are
drawn which clarifies the winding line.
A.H. Oldenhave 51
Starting a large diameter TBM from surface 28 april 2014
A.H. Oldenhave 52
Starting a large diameter TBM from surface 28 april 2014
Figure 36: Cross section tunnel with present volume for ballast [4]
Sand ( = 16 kN/m3);
Slag ( = 19 kN/m3);
Concrete ( = 23 kN/m3).
For each kind of ballast a calculation is made to determine whether or not an embankment is
required and whether the tunnel lining above surface level can be removed (see Appendix E). For this
calculation the safety factor of 0.9 will be used for all the downward loads (tunnel lining, ballast and
weight of the ground). In Appendix E.2 the measures for each zone can be found. Due to the missing
ground improvement, a ballast is required in the whole assumed area (even when 1D cover is
present). In Table 16 a summary of the three tables is given. Again, in Figure 37, Figure 38 and Figure
39 the situation for each kind of ballast is presented in a drawing as well.
A.H. Oldenhave 53
Starting a large diameter TBM from surface 28 april 2014
Slag ballast
Zone 1 10.0 to 7.5 Sand ballast
Zone 2 7.5 to 6.5 Slag ballast
Zone 3 6.5 to -5.0 Slag ballast + embankment (+ 250 m long)
Zone 4 -5.0 to -6.0 Slag ballast
Zone 5 -6.0 to -8.0 Sand ballast
Sand ballast
Zone 1 10.0 to 7.5 Sand ballast
Zone 2 7.5 to -6.0 Sand ballast + embankment (+ 290 m long)
Zone 3 -6.0 to -8.0 Sand ballast
A.H. Oldenhave 54
Starting a large diameter TBM from surface 28 april 2014
A.H. Oldenhave 55
Starting a large diameter TBM from surface 28 april 2014
A.H. Oldenhave 56
Starting a large diameter TBM from surface 28 april 2014
A.H. Oldenhave 57
Starting a large diameter TBM from surface 28 april 2014
7.1. Introduction
When applying a round shaped tunnel lining with small covers on top of the lining, the normal force
in the elements will decrease. When the normal force with respect to the bending moments in the
lining will become too small, this can result in high tension stresses. These stresses can be absorbed
by reinforcement. Still, an upper limit for the applied reinforcement in percentages of the total area
of the concrete element does exist (to prevent failure of the pressure zone of the concrete before
yielding of the reinforcement, this will result in spontaneous failure). Therefore the exact amount of
required reinforcement has to be calculated to check if this requirement is fulfilled.
Between the segments in ring direction the forces have to be translated by bolts. To determine the
required amount of bolts as well as the required reinforcement two critical cross sections will be
chosen for calculation as well as the starting depth of the traditional method for a good comparison
between the different methods. Hereafter it can be concluded if a construction with small covers is
feasible with respect to the tunnel lining.
Figure 40: Continuum model Figure 41: Beam spring model Figure 42: Crown reduction
A.H. Oldenhave 58
Starting a large diameter TBM from surface 28 april 2014
A more common method is the analytical method where the ring will be modelled as a beam spring
model (also used at framework programs, see Figure 41). The lining has certain stiffness (i.e. EI and
EA) and the springs will represent the stiffness of the soil (instead of a soil continuum). The horizontal
and vertical soil loads acting on the lining can be converted into radial and tangential loads, see
Figure 43. In theory it is expected that both loads are acting on the lining. However, in practice it
appears that when assuming both radial and tangential loads, the lining is sometimes over
dimensioned. When tangential load is omitted, calculations still satisfy according to different results
in practice. Apparently the lining does not give tangential support in some cases. According to Blom
(1995) it can be assumed that for deep locations (H > 3*D) the tangential support is present (‘bond’)
while for shallow locations of the tunnel (H < 2*D) the tangential support is omitted (‘slip’ or ‘no
bond’). When for this case only the shallow locations are within the scope, it will be assumed that
tangential support is omitted and therefore only the radial load works on the tunnel lining. H is the
length between the centre of the tunnel and the surface.
For this case it is assumed that the analytical beam spring model will give reliable estimations of the
forces and bending moments in the tunnel lining. A comparison with a numerical model will be
recommended when applying this method.
Figure 43: Transformation of the horizontal and vertical loads to radial and tangential loads [11]
Another assumption is the omitting of the soil support on top of the tunnel lining for shallow
locations, see Figure 42. At these depths the cover is minor resulting in too low resistance along the
sliding lines to counteract deformations and loads with the soil stiffness. By omitting the springs on
the top part over an angle of 90 degrees, the soil stiffness will be even estimated to be zero (no
resistance). According to Blom (1995) the soil directly above the tunnel lining does not influence the
stiffness when H < 2D (cover is 1.5D).
A.H. Oldenhave 59
Starting a large diameter TBM from surface 28 april 2014
be derived. As earlier mentioned it will be assumed that no tangential support will occur. Therefore
only the radial stresses are taken into account.
The radial soil stress can be calculated with (see also Figure 43):
( ) ( )
[kN/m3];
[kN/m3].
Duddeck (1980) also mentions that soil support may only be assumed when the tunnel lining
deforms externally to the surrounding soil for depths H < 2D. According to Ahrens (1982) and
Erdmann (1983) the soil support on top of the tunnel can be even assumed to be zero in the worst
case, regardless of depth. As earlier mentioned, for small covers (H < 2D) the so called crown
reduction (over an angle of 90 degrees) of the soil stiffness will be assumed.
To estimate the bedding stiffness Eoed of the soil, the following formula is used:
( )
( )( )
For this case a ground improvement is applied (MIP columns). Therefore the Young’s modulus of the
soil is different for the columns and for the soil. To estimate the mean Es for the total soil (columns
included) the area of the columns and the soil is calculated for one triangle (the columns will be
applied in triangular pattern). The area of the columns becomes:
m2;
√ √ m2;
This means Ecolumn (which is 15 times Es) has to be multiplied with this factor and the remaining 28 %
is determined by Es:
The required parameters for the calculations of this chapter are summarized in the table below:
A.H. Oldenhave 60
Starting a large diameter TBM from surface 28 april 2014
Table 17: parameters of the different soil layers (according to NEN 6740)
It is clear that many parameters vary with depth (bedding stiffness Eoed, natural earth pressure
coefficient K0 and the saturated soil weight sat). Moreover, for each depth the height of the
embankment on top is different as well (see previous chapter and Appendix E.2). As a simplification
for the calculations, the parameter in the centre of the tunnel lining is assumed to be valid for all the
points of the tunnel lining. An estimation of Eoed can now be calculated for each depth as well as the
required bedding stiffness kr representing the soil support.
Table 18: Estimated parameters for the three calculated cross sections (see further)
Mean dry Mean sat Mean Mean K0 Mean Es Mean Mean Eoed
[kN/m3] [kN/m3] [°] [-] [MPa] [-] [MPa]
bottom tunnel 15.4 18.31 22.59 0.616 20.367 0.254 24.634
6m under
surface level
0.47 m cover 15.4 15.23 20.22 0.654 54.598 0.362 92.67
10.47 m cover 15.4 19.70 25.57 0.568 23.792 0.305 32.493
7.3.4. Self-weight
The last load acting on the tunnel lining is the weight of the lining itself. Assuming a self-weight of 24
kN/m3 for concrete, the total weight can be calculated by multiplying this value with the volume of
the ring. In theory the self-weight does not influence the results of the forces acting in the tunnel
A.H. Oldenhave 61
Starting a large diameter TBM from surface 28 april 2014
lining for deep locations. However, this scope especially focuses on shallow depths, so this load shall
be taken into account as well.
For the ULS the following safety factors are given by Walraven (2006) [12]:
For the calculations only the permanent loads will be taken into account. Therefore the unfavourable
loads (tensile forces and bending moments) will be multiplied by = 1.35 and the favourable loads
(pressure forces) by = 0.9.
7.4.LDesign [39]
Nowadays a variety of framework programs is available where simple beam spring models can be
modelled and calculated. Unfortunately creating a ring model is more complex in these programs.
For this, Dr. Ir. C. Blom decided to make an own framework program especially for tunnel linings,
called LDesign. This program calculates with the input the different loads acting on each point and
determines the normal forces, shear forces, bending moments and deformations for each point. In
Appendix F an explanation of the working of the program is given.
7.5.Output
With the generated input in LDesign (data of the Rijnlandroute are used) the model can be calculated
and the different forces in the tunnel lining can be presented by N, M and V lines (normal forces,
bending moments and shear forces). These forces will act when the segments are installed. During
installation, transportation or storage of the segments, other forces and bending moments will also
work on the segments. However, these forces are not taken into account.
To limit the amount of output three different sections have been analysed; two critical sections
which can be compared with the traditional method starting at approximately 1D cover. Furthermore
the minimum and maximum groundwater levels are taken into account as well.
A.H. Oldenhave 62
Starting a large diameter TBM from surface 28 april 2014
Figure 45: Situation for lowest groundwater level Figure 46: Situation for highest groundwater level
After calculation it appears that the lowest groundwater table (1.4 m below surface level) is decisive
for calculations. The N, M and V-lines are given in the figures below.
Figure 47: Normal forces [39] Figure 48: Shear forces [39] Figure 49: Bending moments [39]
It is expected that the most onerous condition is located in the top of the lining (due to the extreme
values located in the top, see moment line) or at the bottom. To eliminate doubts the forces are
compared for each point in the lining, see tables below. Safety factors are included.
The forces are depending on the length of the longitudinal joint; when the longitudinal joint is more
opened i.e. lt is small, the deformations will increase but the forces will decrease. On the other hand,
when the tunnel is acting more stiff and the longitudinal joint remains closed (l t = 170 mm), the
forces will increase.
As previously mentioned the maximum allowable deformation of the tunnel lining caused by the
internal forces is stated as 50 mm. Moreover, according to Beijer (2010), when the length of the
longitudinal joint lt is 100 mm, water tightness is guaranteed.
Underneath the forces for as well as a joint length of 170 mm as for 100 mm is given. This will only
influence the magnitude. Therefore above figures are given for only one case.
A.H. Oldenhave 63
Starting a large diameter TBM from surface 28 april 2014
Table 20: Forces acting in the lining when lt = 170 mm Table 21: Forces acting in the lining when lt = 100 mm
It is clear that despite higher deformations for lt = 100 mm (which is the most onerous situation,
smallest forces), the deformations remain below the limit of 50 mm and therefore these values will
be used for calculation.
7.5.1.1. Results
In Table 21 the decisive normal forces and bending moments are given. It appears that the bottom is
decisive (M=105 kNm). These values determine the minimum required reinforcement in each ring.
A.H. Oldenhave 64
Starting a large diameter TBM from surface 28 april 2014
The shear force determines the amount of reinforcement in longitudinal direction. Both calculations
are given in Appendix G.
Between the joints reinforcement is absent and bolts have to prevent failure caused by the forces
acting in the tunnel lining. Again, calculations are explained in Appendix H.
According to the calculations the decisive points are located at node 12 and 18 (sides of the bottom).
Here the normal force amounts 154 kN (pressure force) and the bending moment 105 kNm.
Reference is made to Appendix G for more explanation about the calculations. With these forces the
minimum amount of reinforcement amounts 486 mm2 for one side. However, according to the
regulations a minimum reinforcement is always required to prevent a sudden collapse of the
concrete (see Table 23 for further explanation). The minimum reinforcement is depending on the
concrete class and the height of the element and amounts 726 mm2. For prevention of construction
failures reinforcement will be applied on both sides of the element. The total reinforcement A s,total
will be As times two (1452 mm2). This is equal to two rows (see figure) of 10 x 10. The length
between two bars is 92 mm which is lower than the maximum bar distance of 120 mm (see section
7.5.5). Also the diameter remains below the maximum of 13 mm (again see section 7.5.5) and
therefore the crack width will remain below the maximum width to protect the reinforcement.
The maximum shear force in the tunnel lining is 82 kN. When assuming a concrete class C35/45, the
maximum shear force without reinforcement is N (see Appendix G), so
no reinforcement is required. However, in order to guarantee dimensional stability of the
reinforcement cage, brackets will be applied with a maximum distance of 300 mm. To prevent
A.H. Oldenhave 65
Starting a large diameter TBM from surface 28 april 2014
construction failures 12 will be applied as well. By constructing 16 bars in one segment, the
distance between each bar is 282 mm. Again on both sides the longitudinal reinforcement will be
applied to connect the longitudinal reinforcement to the ring reinforcement, see figure below.
Figure 52: Cross section Japanese box method, dimensions in mm Figure 53: Cross section Japanese box method with
[14] reinforcement bars (connected to applied reinforcement),
dimensions in mm [14]
A.H. Oldenhave 66
Starting a large diameter TBM from surface 28 april 2014
For this connection two bolts and six reinforcement bars per box will be applied, in total two boxes
per joint (see Figure 53). The bolts are constructed in the middle of the steel plate to prevent
construction failures. The most generous location for the decisive point (the top) will be at the
outside of the lining, see moment line; however construction at this location is impossible. By
applying the bolts in the centre of the lining, these bolts can counteract the present bending moment
in the joint. The rotation point is located at the centre of the pressure zone of the concrete.
However, after calculating the reinforcement and its pressure zone, it was stated that the pressure
zone of the concrete is small with respect to the total thickness of the lining. For this calculation it
will be assumed that the length between the bolts and the rotation point is ½ h - 10 mm (190 mm).
The pressure forces will be transferred through the surrounding concrete.
In Appendix H the total calculation is given. For this cross section M30 bolts of class 8.8 will satisfy,
together with a steel plate with a thickness tp of 20 mm. The six reinforcement bars behind the steel
plate are of 20 (see figure).
A.H. Oldenhave 67
Starting a large diameter TBM from surface 28 april 2014
Figure 55: Applied bolts Figure 56: Location and Figure 57: Oblique bolt [14]
in Sluiskiltunnel number of bolts in
Sluiskiltunnel
According to the TNO report of Siemes (2001) [8] the maximum friction force between two profiles
during construction is 50 kN/m. Per element this is equal to 236 kN. When the bolts can guarantee
this force, the water tightness will remain. The oblique bolts will be constructed under an angle of 30
degrees, see Figure 57; therefore the required force is √ kN. According to Figure 56
above, four bolts between the rings will be placed over one element. Therefore the total force of one
bolt has to be 66 kN. An M20 of class 6.4 will satisfy.
When also taking into account the maximum absorbing bonding stress of the concrete (prevention of
reinforcement pulled out of the concrete), the minimum length can be determined (safety factors
included, for calculation see Appendix H.2). The length of the bolts will be 450 mm.
Summary
Underneath in Table 22 the minimum reinforcement and bolts are given for one segment.
As ring 10 x 10 x 2
As longitudinal 16 x 12 x 2
Bolts in longitudinal joints 2 x M30, class 8.8
tp 20 mm
reinforcement 20 mm
Bolts in ring joint 4 x oblique bolt M20, class 6.4, length 450 mm
The amount of required measures (reinforcement and bolts) can be found in the table above. In this
table the minimum and maximum reinforcement is used as well. In order to prevent a sudden
collapse (concrete behaves brittle) a minimum reinforcement is always necessary. In the table below
the values for the minimum reinforcement percentage are given for each concrete quality and
height. According to this table the minimum amount of reinforcement per side is 0.22% =
mm2 = As,min (see Appendix G).
A.H. Oldenhave 68
Starting a large diameter TBM from surface 28 april 2014
A maximum reinforcement is also necessary to prevent failing of the pressure zone of the concrete
before the yield stress is reached. The maximum reinforcement is independent from height and given
in the last column of Table 23. This leads to a maximum As =
mm2 for each side of the column.
At last when deformations in the tunnel lining become too high, this can cause problems with the
alignment. According to the outcome (Table 20 and Table 21) these deformations remain below the
maximum of 50 mm.
Figure 58: Situation for highest groundwater level Figure 59: Situation for lowest groundwater level
After calculation it appears that without ground improvement the values become too high.
The mean Eoed without MIP columns amounts 8.4 MPa but with ground improvement it increases to
92.7 MPa. This results in stiffer soil springs. Moreover, the soil weights also increase due to ground
improvement. The calculation without ground improvement results in a displacement of 259 mm
which is absolutely not acceptable any more. It can be concluded that ground improvement is not
only necessary to limit the settlements but also to reduce the deformations in the tunnel lining and
A.H. Oldenhave 69
Starting a large diameter TBM from surface 28 april 2014
to guarantee feasibility of construction of the tunnel lining. Thus, for this cross section ground
improvement will be applied as well.
For this situation the highest groundwater table (0.6 m below surface) is decisive. Dead weight is
included and with LDesign the following data is calculated. See figures below and Table 24 and Table
25.
Figure 60: Normal forces [39] Figure 61: Shear forces [39] Figure 62: Bending moments [39]
A.H. Oldenhave 70
Starting a large diameter TBM from surface 28 april 2014
Table 24: Forces acting in the lining when lt = 170 mm Table 25: Forces acting in the lining when lt = 100 mm
According to the outcome, the deformations, when assuming a longitudinal joint opening of 100 mm,
are too high (> 50 mm). Therefore, the values when assuming an l t of 170 mm will be used for
calculation. It is clear that this cross section is decisive for determination of the feasibility to
construct a tunnel lining at shallow depth.
In Table 24 the maximum bending moment amounts 420 kNm and the associated normal force
amounts 346 kN. To determine whether this normal force reduces the bending moment, the stresses
due to these forces are calculated:
N/mm2
A.H. Oldenhave 71
Starting a large diameter TBM from surface 28 april 2014
kNm.
According to above mentioned calculation the bending moment will not decrease despite the
occurring normal force. Therefore the bending moment will be used for calculation only.
7.5.2.1. Results
The amount of reinforcement and bolts are determined with the measures.
A.H. Oldenhave 72
Starting a large diameter TBM from surface 28 april 2014
Summary
Comparing this cross section with previous critical cross section shows that this cross section will be
the decisive section. The forces are higher resulting in more reinforcement and bolts, given in Table
26 for one segment.
As ring 13 x 16 x 2
As longitudinal 16 x 12 x 2
Bolts in longitudinal joints 2 x M48, class 10.9
tp 40 mm
reinforcement 40 mm
Bolts in ring joint 4 x oblique bolt M20, class 6.4, length 450 mm
Figure 64: Situation for highest groundwater level Figure 65: Situation for lowest groundwater level
The highest groundwater table is decisive again. Dead weight is included and the data given in the
figures below and Table 27 and Table 28 are calculated with LDesign.
A.H. Oldenhave 73
Starting a large diameter TBM from surface 28 april 2014
Figure 66: Normal forces [39] Figure 67: Shear forces [39] Figure 68: Bending moments [39]
When reinforcement is required, the top of the lining will be decisive (due to crown reduction).
However, at this depth the traditional method starts. Therefore it is expected that less or even no
reinforcement is required to guarantee stability in final situation (due to adequate normal forces to
balance the bending moments).
A.H. Oldenhave 74
Starting a large diameter TBM from surface 28 april 2014
Table 27: Forces acting in the lining when lt = 170 mm Table 28: Forces acting in the lining when lt = 100 mm
7.5.3.1. Results
A.H. Oldenhave 75
Starting a large diameter TBM from surface 28 april 2014
Summary
In the table below again the minimum reinforcement as well as the dimensions of the bolts per
segment are given.
As ring 10 x 10 x 2
As longitudinal 16 x 12 x 2
Bolts in longitudinal joints 2 x M30, class 8.8, length 500 mm
Bolts in ring joint 4 x Oblique bolt M20, class 6.4, length 450 mm
7.5.4. Conclusion
It can be concluded that construction at sufficient depth will decrease the amount of required
measures. Regular bolts will satisfy which will decrease the costs as already expected.
On the next page the total measures, required to prevent collapse of the tunnel lining, are
schematized in one figure, see Figure 70. It has to be remarked that due to the fact that only three
cross sections are used for calculation, the length of the zones are a rough estimation. Further
research is recommended.
A.H. Oldenhave 76
Starting a large diameter TBM from surface 28 april 2014
A.H. Oldenhave 77
Starting a large diameter TBM from surface 28 april 2014
7.5.5. Cracking
As already mentioned, the design of the reinforcement is depending on different criteria. The most
decisive criterion is the cracking width.
When reinforced concrete segments are subjected to bending, the concrete will crack before the
reinforcement will adopt the forces. If these cracks will become too wide, this can influence the
water tightness and the protection zone of the reinforcement (this can result in rust of the
reinforcement). A good crack distribution (more cracks, smaller crack width per crack) will reduce
these influences. The amount of cracks is depending on the bonding strength between the steel and
concrete. To increase this bonding the reinforcement bars can be roughened. Moreover, more bars
and a higher amount of reinforcement will also increase the bonding. This taken into account by the
detailing of the reinforcement can influence the amount of cracks and therefore the water tightness
of the tunnel lining.
In the reader of Walraven (2006) tables are given for the maximum allowable bar diameter and
maximum bar distance. If these requirements are met, it can be assumed that the crack width is
permissible. The tables are depending on the environmental classes. For a tunnel lying (partly) below
groundwater level, the environmental class is XC2: exposure caused by carbonation; wet, rarely dry.
A mean steel stress of 300 N/mm2 will be a reliable value for the serviceability limit state. This gives
according to Table 30 and Table 31 a bar diameter km of 13 mm and a bar distance s of 120 mm. In
theory only one of the requirements is enough to assume that the crack width will remain below the
maximum width.
A.H. Oldenhave 78
Starting a large diameter TBM from surface 28 april 2014
8.1.Costs
By adding the differences in costs between the traditional method and the tunnelling with marginal
cover method, insight in the efficiency of this new method is given. For a rough estimate of the
differences in total costs, the Sluiskiltunnel project is used as reference [42]. In this thesis only the
difference in percentages will be mentioned, where the costs of the traditional method will be used
as reference year.
Difference in % - 3.3 %
8.2.Planning
With previous paragraph it can be concluded that the total costs between both methods do not
differ much. However, the building time of a method also influences the efficiency (and the costs).
With the new method (tunnelling with marginal covers) the tunnelling process already starts at
surface level and deep ramps are not necessary any more. This decreases the building time.
For both methods a planning is made [43]. In Appendix I the total building time for each method is
given, including all required proceedings. In Table 33 both methods are summarized and in Figure 71
the critical paths are shown. The critical path implies that these proceedings determine the minimum
construction time due to requirement that the first proceeding is finished before the start of the next
A.H. Oldenhave 79
Starting a large diameter TBM from surface 28 april 2014
proceeding. The preparations include the procurements of the TBM. In the Appendix the planning is
more precisely explained.
Table 33: Comparison building time traditional and “tunnelling with marginal cover” method (1)
Traditional method
Start Work days Calendar days End date
Total building time 1-1-2015 1065 1491 31-1-2019
Preparations 1-1-2015 306 428 4-3-2016
Bored part 4-3-2016 576 624 18-11-2017
Finishing ramps 18-11-2017 314 439 31-1-2019
Figure 71: Comparison building time traditional and tunnelling with marginal cover method (2)
It can be concluded that the “tunnelling with marginal cover” method will take less time to build. The
total building time when constructing the tunnel with the traditional method will amount 4 years and
1 month. When applying the tunnelling with marginal cover method, the total building time will
amount 3 years and 1 month. This represents a gain of 1 year when starting the tunnelling
proceeding at surface level. Especially the finishing of the tunnel results in the smaller construction
time of the tunnelling with marginal cover method, see figure above.
8.3. Conclusion
With above mentioned costs and planning, it can be concluded that the tunnelling with a marginal
cover method will be more efficient due to less building time. The costs however will not differ very
much. Still, when construction time will be less for the tunnelling with marginal cover method, this
A.H. Oldenhave 80
Starting a large diameter TBM from surface 28 april 2014
will also decrease the total construction costs. Moreover, the indirect costs are not taken into
account. These costs include the costs, depending on the building time (wage of staff, rental costs of
equipment, etc.). Probably when applying the new method (tunnelling with marginal covers), this will
result in smaller indirect costs (due to smaller time of construction). Moreover, the site is smaller and
less staff is required during the tunnelling process than during construction of the ramps. For this
thesis only the construction costs are taken into account. When the indirect costs will be calculated
as well, the tunnelling with marginal costs method will probably differ more than 3.3 % with the
traditional method.
A.H. Oldenhave 81
Starting a large diameter TBM from surface 28 april 2014
9.1.Conclusion
In Chapter 4 nine expected complications when applying the tunnelling with marginal cover method
in the Netherlands were mentioned:
The first three consequences are of major interest when the feasibility of the new method (tunnelling
with marginal covers) will be determined. Therefore these three were further explored for the scope
of this thesis (see Figure 9). It can be concluded that the remaining six effects will not result in
complications for the construction of the tunnel with the tunnelling with marginal cover method.
Starting with the soil deformations in the Holocene layer, it can be concluded that ground
improvement is necessary when starting the tunnelling process on surface level. This is due to the
fact that especially in the western part of the Netherlands the first 10 – 20 meters consists of
Holocene deposits (soft soils, clay, peat). When applying a TBM on top of the Holocene layer, this will
result in high deformations that are not allowed for the control of the starting process. For this case
especially the last layer in the Holocene was very soft. Only when the bottom of the TBM reaches the
Pleistocene layer, which is a strong sand layer, the deformations will remain below limits without
ground improvements. Due to experience in applying MIP (Mixed In Place) columns for decreasing
settlements in the soil, this ground improvement is recommended.
A second and maybe the main complication in the Netherlands due to the high groundwater tables
are the buoyancy forces acting on the tunnel lining. When a cover on the tunnel lining is not yet
present, the upward forces cannot be counteracted by the soil and tunnel lining. Therefore measures
have to be applied. A sand ballast in the bottom of the tunnel lining with a maximum of 2.5 meters
and an embankment on top is recommended. Other kinds of ballast can be applied as well. However,
due to increase of costs it will not be advised. During construction of the tunnel the groundwater
level will be lowered temporarily, because of the missing ballast inside the tunnel during construction
(ballast will be applied when TBM is at sufficiently distance).
Constructing a tunnel without cover will also result in complications of construction of the tunnel
lining. The soil forces acting on the lining are absent on top. This results in high bending moments in
the lining and a missing normal force. To prevent open joints between the segments (and infill of
groundwater) bolts of sufficient dimensions have to be applied. For the shallow locations of the
tunnel normal bolts will not satisfy. For these locations the longitudinal joints will be connected using
A.H. Oldenhave 82
Starting a large diameter TBM from surface 28 april 2014
the Japanese box connection. This connection ensures force transmission directly towards the
reinforcement resulting in the applicability of transmitting high bending moments. However, other
bolt types are not taken into account and may be possible as well.
It can be concluded that despite above mentioned main complications the tunnelling with marginal
cover method is feasible in the Netherlands. However, despite the omitted building pits, the
construction costs will not differ for the two methods. Still, the construction time of the tunnelling
with marginal cover method will be almost one year shorter than the traditional method. Therefore it
can be concluded that application of the tunnelling with marginal cover method in the Netherlands
will be feasible and efficient.
9.2.Recommendations
As previously mentioned, this thesis is only intended to determine the adaptation of the tunnelling
with marginal cover method for large diameters in the Netherlands by applying this method for a
specific reference project. The optimal solution has not yet been reached. The calculations are made
in the same order as Chapter 4. This means first the settlements are calculated, determining the
input (soil structure) for further calculation. Subsequently, the measures to prevent uplift of the
tunnel lining are determined, resulting in a required embankment and temporarily lowering of the
groundwater table. The soil structure and embankment are then used for the calculations of the
tunnel lining. These three subjects are the main part of this thesis. After determination of the
required measures for these three subjects other calculations were made, see Chapter 4. It appeared
that some outcomes resulted in new input for the first three calculations (expanded embankment
necessary for the jacking forces for example). However, this thesis is only a first rough research to
determine if the tunnelling with marginal cover method is plausible for a Dutch situation. To
determine the optimization when applying this method in the Netherlands, further research is strictly
recommended. However, the scope of this thesis is to determine the feasibility of the tunnelling with
marginal cover method in the Netherlands. This report cannot be used as a design guide for the
applied method.
For this thesis some limitations are considered. Primarily, when for this thesis only the Rijnlandroute
(even only the eastern side) is considered, it cannot be concluded that the new method (tunnelling
with marginal covers) is feasible for every location in the Netherlands. When this method will be
applied in the Netherlands, again a feasibility research is recommended for the specific location in
the Netherlands. Moreover, due to inadequate geotechnical data of the soil, a rough assumption of
the soil structure is made. When applying the tunnelling with marginal cover method at this location,
site investigation (CPT’s, groundwater measurements) is recommended.
In the scope of this project only three complications are explored in detail. When applying the
tunnelling with marginal cover method, further research for the other complications will be
recommended as well. Moreover, the tunnel lining analysis is only executed in 2D for a single ring. It
will be recommended to apply the calculation with continuum models and 3D models as well.
Additionally, for the tunnel lining analysis only three cross sections were considered. More
calculations for extra cross sections will be suggested. This can also result in smaller application areas
of the measures required for this complication. Moreover, other bolt types are not considered in this
thesis. It will be recommended to analyse if another bolt type instead of the Japanese box will also be
probable. This can reduce the construction costs as well.
A.H. Oldenhave 83
Starting a large diameter TBM from surface 28 april 2014
To conclude, if this method will be recommended in future in the Netherlands, the risks have to be
determined as well. It is concluded that the total construction time will be decreased by one year
when applying the tunnelling with marginal cover method instead of the traditional method,
however when adapting a new method this will also entail additional risks. Therefore it cannot be
concluded yet if this method will be recommended instead of the traditional method.
A.H. Oldenhave 84
Starting a large diameter TBM from surface 28 april 2014
10. Sources
Readers
[9] Bakker, K.J., et al., (2008). Bored and Immersed Tunnels. Delft: TU Delft
[10]Broere, W. & van der Woude, S. (2010). Trenchless Technology. Delft: TU Delft
[11]Blom, C.B.M. (2009). Concrete linings for shield driven tunnels. 2nd edition. Delft: TU Delft
[12]Walraven, J.C. (2006). College gewapend beton. Delft: TU Delft
[13]Abspoel, R., Bijlaard, F.S.K. & Samsor, R. (2008). Constructieleer 2B. Deel Staalconstructies.
Delft: TU Delft
A.H. Oldenhave 85
Starting a large diameter TBM from surface 28 april 2014
Internet sites
[23]Herrenknecht AG. (2011). TBM/Tunneling Video.
http://www.youtube.com/watch?v=qx_EjMlLgqY. Consulted 15-05-2013
[24]CSR News (2013). URUP (Ultra Rapid Under Pass) Method wins the Prime Minister’s Prize in
the Fifth Monodzukuri Nippon Grand Award Enable tunnel boring machine to be launched
and arrive at surface level. http://www.obayashi.co.jp/english/news/csr/2013101000.html
Obayashi Corporation. Consulted 21-02-2014
[25]Provincie Zuid Holland (2011). Rijnlandroute. http://www.zuid-holland.nl/rijnlandroute
Consulted 09-02-2014
[26]University of waterloo. Vertical Stress in a Soil Mass.
http://www.civil.uwaterloo.ca/maknight/courses/CIVE353/Lectures/stress-load.pdf
Consulted 08-06-2013
Regulations
[27]VRC. Veiligheidsrichtlijnen deel C (2009). Rijkswaterstaat
[28]Specifieke Aspecten TunnelOntwerp (2005). Rijkswaterstaat
[29]CUR 162 (1993). Construeren met grond. Grondconstructies op en in weinig draagkrachtige
en sterk samendrukbare ondergrond. Gouda
[30]NEN 6740 (2006). Geotechniek – TGB 1990 – Basiseisen en belastingen.
[31]CUR-report 2001-10 (2001). Diepe grondstabilisatie in Nederland. Handleiding voor
toepassing, ontwerp en uitvoering.
[32]CUR 2006-2 (2006). Innovatieve aardebaan. Snel gebouwd, blijvend vlak.
[33]CUR 199 (2001). Handreiking toepassing No-Recess technieken.
[34]CUR-report 95-7 (1995). Preadvies omtrent het gebruik van kalk-cementkolommen in
Nederland.
[35]ROK 1.0 (2011). Richtlijn Ontwerp Kunstwerken. Rijkswaterstaat
[36]NEN-EN 1992-1-1+C2 (1992). Eurocode 2: Ontwerp en berekening van betonconstructies –
Deel 1-1: Algemene regels voor gebouwen
Software
[37]TNO (2014). DINO loket. Data en Informatie van de Nederlandse Ondergrond.
www.dinoloket.nl Consulted 13-08-2013
[38]Geodan (2013). Actueel Hoogtebestand Nederland. http://ahn.geodan.nl/ahn/ Consulted 06-
11-2013
[39]Blom, C.B.M. (2002). LDesign (v5). Delft: TU Delft
Received information
[40]Pictures of Botlekspoortunnel of S. van der Woude
[41]Pictures of Japanese box method of A.J. Beijer
[42]Cost estimation of project Sluiskiltunnel of J. Schenk
[43]Planning estimation of project Sluiskiltunnel of S. Wierikx
A.H. Oldenhave 86
Starting a large diameter TBM from surface 28 april 2014
Appendices
In this chapter the appendices can be found, which contain more detailed calculations of the
complications mentioned in the report. This chapter contains:
A.H. Oldenhave 87
Starting a large diameter TBM from surface 28 april 2014
The method of Boussinesq normally applies to a uniformly loaded circular area. Depending on the
radius of the area the increase of stresses for each depth z can be determined with
An increase in radius means an increase in stress. With this in mind Newmark developed a drawing
with different radii, each ring subdivided in 20 boxes where every box equals 0.005 p, see Figure
72–Figure 75. When the stresses on a certain depth underneath an object have to be calculated,
Newmark’s chart can be used for a first estimation.
In the figures on the next page the dimensions of the TBM (L x B = 10.235 x 8.066 m) are drawn in
Newmark’s chart to a scale where distance AB on the chart equals the desired depth. In Figure 72
and Figure 74 the stress right under the centre is determined for different depths, in Figure 73 and
Figure 75 the stresses underneath a corner at a specific depth. By multiplying the amount of marked
squares by the area of the object n (in this case the TBM) with the stress on top and 0.005 which is
the stress of each box I (Influence area), the total stress caused by the extra load can be determined.
In formula: .
A.H. Oldenhave 88
Starting a large diameter TBM from surface 28 april 2014
Figure 72: Stresses at z = 5 m below centre (130 squares) Figure 73: Stresses at z = 5 m below corner (45 squares)
Figure 74: Stresses at z = 10 m below centre (60 squares) Figure 75: Stresses at z = 10 m below corner (30 squares)
A.H. Oldenhave 89
Starting a large diameter TBM from surface 28 april 2014
Boston Rule
In the report this method is used to determine the soil stresses. The outcome can be found in the
table below.
Comparison
When the stresses, found with Newmark’s chart and the Boston rule, are drawn and compared in a
graph (see Figure 76), the Boston rule indeed gives a good estimation of the mean stress for each
depth. Therefore, to calculate the total deformation in the Holocene layer, these stresses will be
used. Only the increase of stresses (4th column inTable 34) will cause deformations.
60
40
20 z=5 (Newmark)
0 z=5 (Boston)
0 2 4 6 8 10 12
Length [m]
20
z=10 (Newmark)
0 z=10 (Boston)
0 2 4 6 8 10 12
Length [m]
A.H. Oldenhave 90
Starting a large diameter TBM from surface 28 april 2014
The figures show that CPT S30H02476 and CPT S30H02477 are closest to the start of the tunnelling
process. However, these penetration tests show stronger layers in the first 10 meters than the two
CPT’s measured in the polder (S30H00677 and S30H00674). Moreover, the closest penetration tests
are measured along a road, as well as S30H00659 and S30H00668. A ground improvement can be
expected. Therefore, these results will not contribute to the expected soil layers at the start of the
tunnelling process.
A.H. Oldenhave 91
Starting a large diameter TBM from surface 28 april 2014
A.H. Oldenhave 92
Starting a large diameter TBM from surface 28 april 2014
A.H. Oldenhave 93
Starting a large diameter TBM from surface 28 april 2014
A.H. Oldenhave 94
Starting a large diameter TBM from surface 28 april 2014
A.H. Oldenhave 95
Starting a large diameter TBM from surface 28 april 2014
A.H. Oldenhave 96
Starting a large diameter TBM from surface 28 april 2014
A.H. Oldenhave 97
Starting a large diameter TBM from surface 28 april 2014
A.H. Oldenhave 98
Starting a large diameter TBM from surface 28 april 2014
Underneath the total settlements of the Holocene layer is given when applying the TBM at different
depths. The TBM will start at 2.5 meters below surface. When the TBM rests on the Pleistocene layer,
the settlements are almost zero. Therefore it is proven that assuming the Pleistocene layer as
infinitely stiff is a good assumption.
Table 41: Settlements when applying TBM at depth 2.5 meters. Soil replacement and embankment of 4.19 meters
Soil type Layer Thickness Mean ' 'z;0 'z 'z Cp Cs [-] h
[m] depth [m] [kN/m3] [kPa] [kPa] [kPa] [-] [m]
Sand (gone) 1 5.19 2.595 18 47 0 47 600 0
Sand (gone) 2 1.5 5.94 10 101 0 101 600 0
Sand 3 0.5 6.94 10 111 266 377 600 0
Clay, silty 4 1 7.69 7 117 228 345 15 160 0.08
Silt 5 3 9.69 9 134 158 292 45 1300 0.05
Clay 6 4 13.19 7 161 95 257 7 30 0.36
Sum h 0.49
Table 42: Settlements when applying TBM at depth 4 meters. Soil replacement and embankment of 4.19 meters
Soil type Layer Thickness Mean ' 'z;0 'z 'z Cp Cs [-] h
[m] depth [m] [kN/m3] [kPa] [kPa] [kPa] [-] [m]
Sand (gone) 1 5.19 2.595 18 47 0 47 600 0
Sand (gone) 2 1.5 5.94 10 101 0 101 600 0
Sand 3 0.5 6.94 10 111 0 111 600 0
Clay, silty 4 1 7.69 7 117 0 117 15 160 0
Silt 5 3 9.69 9 134 201 335 45 1300 0.06
Clay 6 4 13.19 7 161 114 275 7 30 0.41
Sum h 0.47
Table 43: Settlements when applying TBM at depth 7 meters. Soil replacement and embankment of 3.44 meters
Soil type Layer Thickness Mean ' 'z;0 'z 'z Cp Cs [-] h
[m] depth [m] [kN/m3] [kPa] [kPa] [kPa] [-] [m]
Sand (gone) 1 4.44 2.22 18 40 0 40 600 0
Sand (gone) 2 1.5 5.19 10 87 0 87 600 0
Sand 3 0.5 6.19 10 97 0 97 600 0
A.H. Oldenhave 99
Starting a large diameter TBM from surface 28 april 2014
Sum h 0.57
Table 44: Settlements when applying TBM at depth 7.53 meters. Soil replacement and embankment of 3.24 meters
Soil type Layer Thickness Mean ' 'z;0 'z 'z Cp Cs [-] h
[m] depth [m] [kN/m3] [kPa] [kPa] [kPa] [-] [m]
Sand (gone) 1 4.24 2.12 18 38 0 38 600 0
Sand (gone) 2 1.5 4.99 10 84 0 84 600 0
Sand 3 0.5 5.99 10 94 0 94 600 0
Clay, silty 4 1 6.74 7 100 0 100 15 160 0
Silt 5 3 8.74 9 117 0 117 45 1300 0
Clay 6 0.53 10.505 7 132 0 132 7 30 0
Clay 7 3.47 12.505 7 146 168 314 7 30 0.51
Sum h 0.51
Table 45: Settlements when applying TBM at depth 8.53 meters. Soil replacement and embankment of 2.84 meters
Soil type Layer Thickness Mean ' 'z;0 'z 'z Cp Cs [-] h
[m] depth [m] [kN/m3] [kPa] [kPa] [kPa] [-] [m]
Sand (gone) 1 3.84 1.92 18 35 0 35 600 0
Sand (gone) 2 1.5 4.59 10 77 0 77 600 0
Sand 3 0.5 5.59 10 87 0 87 600 0
Clay, silty 4 1 6.34 7 93 0 93 15 160 0
Silt 5 3 8.34 9 110 0 110 45 1300 0
Clay 6 1.53 10.605 7 128 0 128 7 30 0
Clay 7 2.47 12.605 7 142 177 319 7 30 0.38
Sum h 0.38
Table 46: Settlements when applying TBM at depth 9.53 meters. Soil replacement and embankment of 2.43 meters
Soil type Layer Thickness Mean ' 'z;0 'z 'z Cp Cs [-] h
[m] depth [m] [kN/m3] [kPa] [kPa] [kPa] [-] [m]
Sand (gone) 1 3.43 1.715 18 31 0 31 600 0
Sand (gone) 2 1.5 4.18 10 69 0 69 600 0
Sand 3 0.5 5.18 10 79 0 79 600 0
Clay, silty 4 1 5.93 7 85 0 85 15 160 0
Silt 5 3 7.93 9 102 0 102 45 1300 0
Sum h 0.24
Table 47: Settlements when applying TBM at depth 10.53 meters. Soil replacement and embankment of 1.99 meters
Soil type Layer Thickness Mean ' 'z;0 'z 'z Cp Cs [-] h
[m] depth [m] [kN/m3] [kPa] [kPa] [kPa] [-] [m]
Sand (gone) 1 2.99 1.495 18 27 0 27 600 0
Sand (gone) 2 1.5 3.74 10 61 0 61 600 0
Sand 3 0.5 4.74 10 71 0 71 600 0
Clay, silty 4 1 5.49 7 77 0 77 15 160 0
Silt 5 3 7.49 9 94 0 94 45 1300 0
Clay 6 3.53 10.755 7 120 0 120 7 30 0
Clay 7 0.47 12.755 7 134 198 332 7 30 0.08
Sum h 0.08
Table 48: Settlements when TBM rests on Pleistocene layer (depth = 11 meters). No ground improvement
Sum h 0.001
Appendix E: Buoyancy
When calculating the amount of upward forces and the amount of downward forces acting on the
tunnel, it can be calculated if buoyancy will act (upward forces > downward forces). This is the so
called uplift calculation.
Another effect which may occur is the break-up failure, see Figure 89. This failure occurs when the
horizontal ground pressures become too high with respect to the vertical ground pressures. When
the shear force of the wedge on top of the tunnel is too small, the tunnel does not have counter
pressure on top and the tunnel lining will ovalize as presented in the figure. In this case however the
tunnel is located very shallow in the soil, resulting in small ground pressures. Moreover in most cases
a cover on top is even absent. It will be assumed that the break-up failure will not be decisive for the
shallow applied locations of the tunnel lining. Therefore only the uplift calculation is executed.
In Figure 91 the situation for the Rijnlandroute is schematized where the tunnel is located just below
the highest occurring groundwater level (decisive for these calculations). In this figure also the ballast
is drawn. The maximum height amounts 2.5 meters. By applying different kind of ballasts (with
different unit weights) the remaining upward pressure is different for these situations. Therefore for
each situation and each depth separately the forces are calculated, see next paragraph.
By summing the uplift force A (negative), the weight of the tunnel lining G1, the ground force G2 and
the ballast, the remaining upwards force has to be compensated by an embankment on surface level.
Safety factors however have to be taken into account in these calculations (see Chapter 5).
Underneath the formulas are given (for this cross section as example) which are used to calculate the
total uplift force (see Table 50 where top of tunnel is 1 meter below surface). For G2 the soil next to
the top of the tunnel also has to be taken into account. By calculating the area of that part of the
tunnel (given in blue in the formula), the wet soil area can be calculated.
kN/m
( ) ( ) kN/m
kN/m
( ) ( ) ( )
( ) (( ) ) ( ) kN/m
( ( )) ( )
( ( )) ( )
kN/m
( ) ( )
Where ( ) ( ) degrees
kN/m
( ) ( )
m.
Table 50: Permanent situation with safety factors and concrete ballast
Concrete ballast
Permane Top of Depth tunnel Upward Weight Ballast Height Weight Weight Upward - Height Permanently
nt zone tunnel below s tunnel [kN/m] ballast dry wet Downward embankme (after
below groundwater pressure lining [m] ground ground with safety nt with tunnelling) with
surface [m] level [m] [kN/m] [kN]m] [kN/m] [kN/m] [kN/m] safety [m] safety
Zone 1 10 10.53 870.9 305.5 127.8 1.6 75.8 560.2 -91.6 -0.04 Sand ballast
9 10.53 870.9 305.5 156.8 2.1 75.8 531.0 -91.3 -0.04
8 10.53 870.9 305.5 196.2 2.4 75.8 495.2 -94.6 -0.06
Zone 2 7 10.53 870.9 305.5 235.3 2.1 75.8 455.6 -94.2 -0.06 Concrete ballast
6 10.53 870.9 305.5 274.3 2.3 75.8 409.8 -88.0 -0.02
Zone 3 5.6 10.53 870.9 305.5 314.6 2.5 75.8 362.8 -82.0 0.01 Concrete ballast
No removal of tunnel lining above surface
5 10.53 870.9 305.5 314.6 2.5 75.8 314.3 -38.4 0.27 + embankment
4 10.53 870.9 305.5 (length is + 225
314.6 2.5 75.8 227.5 39.7 0.74
m)
3 10.53 870.9 305.5 314.6 2.5 75.8 144.3 114.6 1.18
2 10.53 870.9 305.5 314.6 2.5 75.8 85.8 167.3 1.49
1 10.53 870.9 305.5 314.6 2.5 75.8 45.5 203.6 1.71
0.47 10.40 868.8 305.5 314.6 2.5 94.2 63.9 168.5 1.50
0 9.93 851.1 305.5 314.6 2.5 66.9 44.0 193.2 1.65
-1 8.93 787.4 305.5 314.6 2.5 33.5 18.8 182.2 1.98
-2 7.93 703.6 305.5 314.6 2.5 17.0 6.8 124.1 2.33
-3 6.93 607.8 305.5 314.6 2.5 7.1 1.6 41.9 2.69
-3.53 6.40 554.0 305.5 314.6 2.5 3.7 0.1 -7.6 2.89
-4.53 5.40 449.6 305.5 314.6 2.5 0.4 0.0 -108.8 3.26
Zone 4 -5.53 4.40 344.8 147.9 254.6 2.2 0.0 0.0 -17.5 From here Concrete ballast
Zone 5 -6.53 3.40 243.2 129.2 151.5 1.8 0.0 0.0 -9.4 no tunnel, Sand ballast
-7.53 2.40 149.4 109.5 64.5 1.0 0.0 0.0 -7.2 so no 0.5 m
required
-8.03 1.90 107.0 87.7 38.1 0.7 0.0 0.0 -6.3
Table 51: Permanent situation with safety factors and slag ballast
Slag ballast
Perma Top of Depth tunnel Upwards Weight Ballast Height Weight Weight Upward - Height Permanently
nent tunnel below pressure tunnel [kN/m] ballast dry wet Downward embankment (after
zone below groundwater [kN/m] lining [m] ground ground with safety with safety tunnelling)
surface [m] level [m] [kN]m] [kN/m] [kN/m] [kN/m] [m] with safety
Zone 1 10 10.53 870.9 305.5 127.8 1.6 75.8 560.2 -91.6 -0.04 Sand ballast
9 10.53 870.9 305.5 156.8 2.1 75.8 531.0 -91.3 -0.04
8 10.53 870.9 305.5 196.2 2.4 75.8 495.2 -94.6 -0.06
Zone 2 7 10.53 870.9 305.5 233.0 2.4 75.8 455.6 -92.1 -0.05 Slag ballast
Zone 3 6 10.53 870.9 305.5 249.1 2.5 75.8 409.8 -65.4 0.11 Slag ballast +
5.6 10.53 870.9 305.5 249.1 2.5 75.8 362.8 -23.1 0.36 embankment
(length is +
No removal of tunnel lining above surface
Table 52: Permanent situation with safety factors and sand ballast
Sand ballast
Perman Top of Depth tunnel Upward Weight Ballast Height Weight Weight Upward - Height Permanently
ent tunnel below s tunnel [kN/m ballast dry wet Downward embankment (after
zone below groundwater pressure lining ] [m] ground ground with safety with safety tunnelling)
surface [m] level [m] [kN/m] [kN]m] [kN/m] [kN/m] [kN/m] [m] with safety
Zone 1 10 10.53 870.9 305.5 127.8 1.6 75.8 560.2 -91.6 -0.04 Sand ballast
9 10.53 870.9 305.5 156.8 2.1 75.8 531.0 -91.3 -0.04
8 10.53 870.9 305.5 196.2 2.4 75.8 495.2 -94.6 -0.06
Zone 2 7 10.53 870.9 305.5 209.7 2.5 75.8 455.6 -71.1 0.08 Sand ballast +
6 10.53 870.9 305.5 209.7 2.5 75.8 409.8 -29.9 0.32 embankment
5.6 10.53 870.9 305.5 209.7 2.5 75.8 362.8 12.4 0.57 (length is +
290 m)
5 10.53 870.9 305.5 209.7 2.5 75.8 314.3 56.0 0.83
No removal of tunnel lining above surface
F.1. Input
When starting the program a lot of different input parameters are asked (Figure 92). Starting with
the dimensions of the lining and the amount of required points (amount of segments times amount
of segment fields) the coordinates of each point can be calculated and the ring with its points can be
drawn. In Figure 93 the black dots imply the points in the segment while the green dots indicate the
longitudinal joints. As earlier mentioned, in these longitudinal joints a contact stiffness of
is present. The contact length of the joint lt is assumed to be 170 mm which is an experience number.
Figure 93: Tunnel lining with 28 springs (purple lines) and forces
Figure 92: Input caused by the soil (red lines)
Under the input box of the lining geometry the double ring can be checked. This means that instead
of assuming a single ring analysis, the second ring and its interaction with the first ring can also be
taken into account. Again, due to previous experiences with this calculation method it can be
concluded that this interaction does not have major influences on the results. Therefore in this thesis
only the single ring analysis is performed.
As a simplification different approaches exist to determine the forces acting in the lining. One of
these approaches implies that due to the upward pressure of the groundwater (floating) the soil
stresses on top increase and the stresses at the bottom decrease. This leads in the steady state to
equal pressures on top and at the bottom. This approach therefore assumes:
.
Starting a large diameter TBM from surface 28 april 2014
However, in this case the tunnel lining is only partly located in the groundwater. Therefore this
approach does not apply.
Stabilizing supports as springs is nothing more than stabilizing the tunnel ring to prevent rolling of
the ring. However, when this box is not ticked, one support still remains (see blue triangle on top in
Figure 93).
As previously mentioned the key segment is of minor influence and can be omitted in the calculation.
After giving the mean Eoed at the centre, the soil characteristics can be filled in. The heights of the
different parameters are with respect to the centre of the lining.
There are different methods to calculate the required data. Duddeck is a commonly used method but
only feasible for deep positions. Therefore only increasing gravity load will be used. For the shallow
positions (H < 2D) the crown reduction will be used as well.
Given in the previous chapter the dead weight lining is of minor influence for deep positions.
However for this case the influences cannot be disregarded any more.
The last part is outside this scope. Furthermore, the values are all given in N and mm.
Eventually LDesign uses the input to generate a table which is used for the final calculations.
Underneath the table is given. Coloured rows are explained below:
1
49 28 21
Yellow (1):
0 5065 1 0 0 0 0 0 0 -9995.18 0 1 1 7.9E+10
1127.06853 4938.00988 0 0 0 0 0 0 -2779.83 -12179.2 0 1 1 1
2197.62114 4563.40732 0 0 0 0 0 0 -8305.64 -17246.8 0 1 1 1
3157.97585 3959.97646 0 0 0 0 0 0 -17840 -22370.6 0 1 1 1
3959.97646 3157.97585 0 0 0 0 0 0 -32282.5 -25744.4 0 1 1 7.9E+10
4563.40732 2197.62114 0 0 0 0 0 0 -50062.2 -24108.7 0 1 1 1
-4938.0099 1127.06853 0 0 0 0 0 0 69836.74 -15939.8 0 1 1 1
tunnel lining and the soil points; the second column gives 5543.96704
4421.16618
3076.66959
1577.89594
-4421.16618
-5543.96704
-6388.77024
-6913.21383
1
1
1
1
1
1
1
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
-5.43E-12 -7091 1 1 0 0 0 0 0 0 0 1 1 1
1.312E+10 1.7493E+14 1 2
Orange (3):
1.312E+10 1.7493E+14 2 3
1.312E+10 1.7493E+14 3 4
1.312E+10 1.7493E+14 4 5
1.312E+10 1.7493E+14 5 6
1.312E+10 1.7493E+14 6 7
1.312E+10 1.7493E+14 7 8
Here the radial forces of the soil are calculated for each 1.312E+10
1.312E+10
1.312E+10
1.312E+10
1.7493E+14
1.7493E+14
1.7493E+14
1.7493E+14
8
9
10
11
9
10
11
12
1.312E+10 1.7493E+14 18 19
6
9 33 1 1000 9378462
10 34 1 1000 9378462
11 35 1 1000 9378462
16 40 1 1000 9378462
Blue (5):
17 41 1 1000 9378462
18 42 1 1000 9378462
19 43 1 1000 9378462
20 44 1 1000 9378462
21 45 1 1000 9378462
22 46 1 1000 9378462
The first column gives the EA which is the force per unit 23
24
25
47
48
49
1
1
1
1000
1000
1000
9378462
9378462
9378462
Officially LDesign was designed for traditional bored tunnels where the tunnel was positioned at least
1D under surface level. However, with this new method (tunnelling with marginal covers) the tunnel
lining already starts above surface level. Therefore additional adjustments in the general input were
necessary. In LDesign this is possible by manipulating the table correctly for the applied situation.
When more than the upper quarter of the tunnel lining was located above surface level, more springs
had to be removed. With this method the upper half of the springs can be removed, see Figure 94.
However, after calculation of the buoyancy it was concluded that at shallow depths an embankment
is necessary to avoid uplift forces. For the calculations with LDesign it is therefore not necessary any
more to reduce the amount of springs in the model (the embankment covers these parts of the
tunnel lining).
Moreover, the heights of the soil (h s(soil)) and groundwater table (h water) are with respect to the
centre of the tunnel. When the centre of the tunnel is located above these heights the input has to
be manipulated by hand as well (negative values are not possible in LDesign). For H=0.735 (see
further) and a groundwater level of 1.4 m below surface, the forces of the soil acting on the different
points are calculated by hand and are replacing the forces given in the table. This will result in more
accurate results.
When pressure forces and bending moments are present, two equations have to be valid:
In Figure 95 a cross section of a column and its forces are schematised. The equation for the sum of
the normal forces is:
(0.75 is the so called fullness factor and valid when c = cu = 3.5 ‰)
In above formulas two unknowns are present; As and x (length of pressure zone, see figure).
Therefore summation of the bending moments (at the top, right point in figure) is necessary as well:
( )
The length between Nc and top (a) is depending on the strain distribution. In most cases the
maximum strain in the concrete is reached before the steel yields. Therefore c = cu is plausible for
the calculation of the amount of reinforcement. In this case the value of a is equal to 0.39 x. Now
with two equations and two unknowns the final reinforcement area A s can be calculated. Values are
given in Paragraph 7.5. The total amount of reinforcement is .
Figure 95: Cross section column with bending and pressure forces. On the right side the strain distribution and the forces
acting on the column are drawn. The normal force acting in the right reinforcement bars are zero (reinforcement cannot
react as pressure force).
Remark
The maximum strain of concrete amounts 3.5 ‰ and steel yields at 27.5 ‰. In most cases the
maximum strain of concrete is reached (cracks will be present) before the steel will yield. This is
desirable, i.e. the maximum strain at the tension side amounts 27.5 ‰. With the maximum pressure
strain of 3.5 ‰, the minimum length of the pressure zone can be calculated.
mm.
This implies that when the value of x < 45.2 mm, the steel will yield and the tunnel lining will fail.
However, it appears that the strain distribution is different. The strain at the pressure zone is less
than 3.5 ‰ (but bigger than c,pl = 1.75 ‰, so still cracks will occur) and the length of the pressure
zone is bigger, see Figure 96. Now the concrete will still yield before reinforcement will fail.
According to Walraven (2006) the limit of concrete (i.e. maximum shear stress of concrete) to absorb
the shear forces is 0.4 fb, see Table 53.
To determine the maximum shear force where no reinforcement (brackets) is required, the value d =
0.4 fb has to be multiplied by the area of a segment, i.e. b x d.
When the occurring shear force is higher than the maximum shear force of the concrete, the
remaining shear force has to be taken up by the reinforcement. By dividing the remaining shear force
Vs by the yielding strength of steel (435 N/mm2), the amount of reinforcement can be calculated:
( )
To determine the dimension and strength of the bolts, insight of the present forces in the lining is
required. Two critical sections are chosen (see Paragraph 7.5, the lowest groundwater levels are
decisive). The forces are depending on the length of the longitudinal joint.
Limitations
As previously mentioned the maximum allowable deformation of the tunnel lining caused by the
internal forces is stated as 50 mm. Moreover, according to Beijer (2010), when the length of the
longitudinal joint lt is 100 mm, water tightness is guaranteed. Adding these two limitations to the
outcome of the two critical cross sections, the deformations remain below 50 mm when applying an
opened joint (lt = 100 mm) which is the most onerous situation (smallest forces). This leads to the
following maxima:
H = 0.735 m H = 5.735 m
lt [mm] 170 100 lt [mm] 170 100
Mmax [kNm] 114 105 Mmax [kNm] 420 379
ymax [mm] 13 17 ymax [mm] 45 54
According to the table above the cross section located just below surface level will be normative. This
is due to the fact that ground improvement is not constructed at this cross section any more.
However, in this Appendix the first cross section will be used for calculations (H=0.735).
Due to self-weight of the tunnel lining, the normal forces in the joints act like pressure forces. This is
onerous for the joints, however negligible with respect to the bending moment (for example:
M/W=6.24, N/A=0.55). Therefore for these cross sections normal bolts will not be applied.
Bolt type
For this connection the Japanese box method will be applied (see Chapter 6), which can counteract
higher bending moments. For a first calculation it will be assumed that the length between the bolts
and the rotation point is ½ h (200 mm). The pressure forces will be transferred through the
surrounding concrete.
Calculations [13]
Two different failures of the bolt connection are possible; Failure of the bolt or failure of the steel
plate (steel will punch). To determine which failure will be decisive, both mechanisms are calculated
for a M30, 8.8 bolt, screwed into a steel plate with a thickness tp of 20 mm:
According to EC3 the maximum tensile force to prevent punching of the bolt through the plate is:
The maximum tensile force in the bolts is decisive and amounts 326 kN.
When Mmax = 105 kNm and arm = 0.2 m, the bolts have to guarantee a force of kN. Divided
by two both bolts have a tensile force of 263 kN. It can be concluded that M30 bolts of class 8.8 and a
steel plate with a thickness of 20 mm will satisfy the requirements. In Figure 53 and underneath a
detailed design of the connection is given.
A more precise calculation takes into account the different lengths of the arms of the two bolts.
When applying a distance between both bolts of 70 mm (see figure), the arms will be 0.155 and
0.225 m. The rotation point is located 10 mm under the top due to the centre of the pressure zone of
the concrete. The maximum bending moment these two bolts can take into account amounts:
Reinforcement
The force of the two bolts has to be translated by the reinforcement. In total 6 reinforcement bars
will be applied and the maximum force of both bolts amounts kN. Every bar has to
transfer 109 kN. This is equal to mm2 reinforcement, which equals a reinforcement bar
with diameter 18. Reinforcement bars with the common applied diameter of 20 will be applied.
Where fctd is the design value of the tensile strength of concrete according to:
The coefficients 1 and 2 are depending on the quality of the bonding conditions. 1 amounts 1.0 for
‘good’ conditions and 0.7 for all the other cases. When h > 250 mm, one side will always result in
‘bad’ conditions, so here 0.7 is used. For 2 a value of 1.0 may be used when the diameter is smaller
than 30 mm.
So, when assuming the first cross section where the bolt forces between the rings amount 66 kN per
bolt, the maximum bonding strength amounts:
The minimum required area present in the concrete amounts m2 (the maximum force
in the bolt amounts 66 kN). This results in a minimum length of the bolt of m.
In this Appendix the different proceedings will be shortly explained, for both methods.
When the preparations of the east ramp are finished in theory the tunnelling of the tunnel can start.
However, for the cross passages it will be assumed that jet grouting is required to make construction
of the passages possible. The jet grouting will already start when preparations of the east ramp are
still in progress. This results in starting of the bored part when the preparations are finished. The jet
grouting for the last cross passages are still in operation, however due to adequate difference in
length, this will not influence the tunnelling process. The jet grouting is finished 4 months after the
start of the first bored tunnel. 6 months later the tunnelling process is finished as well. These two
proceedings will therefore not influence each other. It will be assumed that the excavation rate per
day amounts 8.5 meter, which is a reliable estimation for soft soils according to Bouygues
Construction (2009) [21]. In this rate interventions for disc cutters changes are included. Moreover,
for the North-South line in Amsterdam (corresponding soil conditions) a mean rate of 8.5 meters and
12 meters were achieved. Therefore an excavation rate of 8.5 meters per day (24 hours per day, 7
days a week) is a reliable and safe assumption.
After finishing the first tunnel, the TBM has to be disassembled and transported back to the
launching shaft. This will approximately take 8 weeks to disassembly the TBM and 8 weeks to
transport and assemble the TBM again. The second tunnel will now be drilled. When the TBM has
crossed the location of the first cross passages, these passages can already be constructed before the
tunnelling is finished.
When the tunnels are finished, the TBM will be disassembled and transported again and the
construction of the ramps can start (construction of the concrete floors, walls and decks (only closed
section)). It will be assumed that to limit the costs only one side of the ramps will be constructed at
the same time. When the east ramp is finished, the west ramp will be finished afterwards. Now only
backfill of the building pits remain. When the closed parts of the ramps are refilled with soil until
surface level the whole tunnel connection is finished.
The traditional method will take 4 years and 7 months. In Table 55 and Figure 97 the planning of this
method is given.
Dec-16
Dec-17
Dec-18
Sep-16
Sep-17
Sep-18
Apr-15
Apr-16
Apr-17
Apr-18
Apr-19
Oct-15
Jan-15
Jan-16
Jul-15
Jul-16
Jul-17
Jul-18
Jul-19
Ramps
Ramp east
Insertion sheet piles
Excavation building pit & construction struts
Soil transportation
Construcion underwater concrete floor
Placing reinforcement
Pouring concrete
Hardening time concrete
Pumping of water
Construction plug
Finishing ramp
Construction service building
Construction launching frame & installing TBM
Construction concrete first floor
Construction concrete walls
Construction concrete deck (closed section)
Backfill building pit
Ramp west
Insertion sheet piles
Excavation building pit & construction struts
Soil transportation
Construction underwater concrete floor
Placing reinforcement
Pouring concrete
Hardening time concrete
Pumping of water
Construction plug
Finishing ramp
Construction concrete first floor
Construction concrete walls
Construction concrete deck (closed section)
Backfill building pit
Bored part
Tunnel boring 1st part
TBM return & reinstalling
Tunnel boring 2nd part
Disassembly TBM
Dec-16
Dec-17
Dec-18
Sep-16
Sep-17
Sep-18
Apr-15
Apr-16
Apr-17
Apr-18
Apr-19
Oct-15
Jan-15
Jan-16
Jul-15
Jul-16
Jul-17
Jul-18
Cross passages
Cross passage 1
Installation jet
Drilling and jet grouting
Construction
Dismantling & transportation jet
Cross passage 2
Installation jet
Drilling and jet grouting
Construction
Dismantling & transportation jet
Cross passage 3
Installation jet
Drilling and jet grouting
Construction
Dismantling & transportation jet
Cross passage 4
Installation jet
Drilling and jet grouting
Construction
Dismantling & transportation jet
Cross passage 5
Installation jet
Drilling and jet grouting
Construction
Dismantling & transportation jet
Cross passage 6
Installation jet
Drilling and jet grouting
Construction
Dismantling & transportation jet
Cross passage 7
Installation jet
Drilling and jet grouting
Construction
Dismantling & transportation jet
To limit the settlements and to make construction at shallow depths possible, it was concluded that
ground improvement is necessary (Chapter 5), as well as an embankment at surface level. These
proceedings require most of the preparation time. Afterwards small sheet piles will be installed
(depth building pit is 2.5 meters, so the sheet piles will be 3 x 2.5 = 7.5 m) and the soil within will be
excavated. The groundwater has to be lowered temporarily as well.
For the construction time of the MIP columns, CUR rapport 95-7 is used in which the amount of
construction of MIP columns is estimated as 25 mm per rotation and 150 rotations per minute.
Assuming a work day of 8 hours, it can be calculated how much MIP columns can be constructed per
day. As mentioned in Paragraph 7.3.2 the amount of MIP columns per m2 is 71.6 %. Assuming a
ground improvement area of 200 x 18.5 m2 x 2 at the eastern side (two different building pits) and an
area of 200 x 50.5 m2 at the west side, this leads to a total ground improved area of 17500 m 2. When
71.6 % of that area will be applied with MIP columns of more than 10 meters, this results in a total of
130060 MIP columns that have to be constructed. Using above mentioned values the total
construction time for the MIP columns can be calculated:
days.
After these proceedings and installing the TBM (which is ordered and delivered on site in
approximately 10 months) and launching frame, the bored process can start. For the traditional
method a mean excavation rate of 8.5 meters per day was assumed. However, when the bored part
becomes longer, the mean rate will also increase (assumption). Therefore for this method a mean
rate of 10 meters per day is assumed.
For the traditional method 7 cross passages are necessary. For this method however (due to longer
tunnel length) 9 cross passages have to be constructed. This results in a higher construction time for
the cross passages. Only when the last cross passage is finished, the back fill of the west “ramp” can
start. Hereafter the tunnel lining above surface partly will be removed (see Appendix E).
The total construction time of the Rijnlandroute using the tunnelling with marginal cover method,
amounts almost 4 years. In Table 56 and Figure 98 the planning for this method is given, including all
proceedings.
Remark
This is only a first general assumption of the total construction time for the tunnelling with marginal
cover method. It is for example assumed that at the location of all cross passages jet grouting is
required. However, when the first and last cross passage is still located in the MIP columns jet
grouting may not be necessary any more. Moreover, freezing for shallow depths can also be
interesting. The area for the MIP columns can also differ. For a more precise assumption of the total
construction time of this method it will therefore be recommended to expand the amount of
calculations at certain depths. This will influence the amount of precautions and therefore the
planning (as well as the costs).
Table 56: Detailed planning construction time tunnelling with marginal cover method (1)
Dec-16
Dec-17
Sep-16
Sep-17
Apr-15
Apr-16
Apr-17
Apr-18
Oct-15
Jan-15
Jan-16
Jul-15
Jul-16
Jul-17
Jul-18
Ramps
Construction MIP columns
Placing embankment
Insertion sheet piles
Excavation building pit
Soil transportation
Drainage of groundwater
Finishing ramp
Construction service building
Construction launching frame & installing TBM
Backfill building pits east
Backfill building pit west
Bored part
Tunnel boring 1st part
TBM return + reinstalling
Tunnel boring 2nd part
Disassembly TBM
Cross passages
Cross passage 1
Installation jet
Drilling and jet grouting
Construction
Dismantling & transportation jet
Cross passage 2
Installation jet
Drilling and jet grouting
Construction
Dismantling & transportation jet
Cross passage 3
Installation jet
Drilling and jet grouting
Construction
Dismantling & transportation jet
Cross passage 4
Installation jet
Drilling and jet grouting
Construction
Dismantling & transportation jet
Cross passage 5
Installation jet
Drilling and jet grouting
Construction
Dismantling & transportation jet
Cross passage 6
Installation jet
Drilling and jet grouting
Construction
Dismantling & transportation jet
Dec-16
Dec-17
Sep-16
Sep-17
Apr-15
Apr-17
Apr-16
Apr-18
Oct-15
Jan-15
Jan-16
Jul-15
Jul-16
Jul-17
Jul-18
Cross passage 7
Installation jet
Drilling and jet grouting
Construction
Dismantling & transportation jet
Cross passage 8
Installation jet
Drilling and jet grouting
Construction
Dismantling & transportation jet
Cross passage 9
Installation jet
Drilling and jet grouting
Construction
Dismantling & transportation jet
Tunnel lining removal
Tunnel lining removal east
Tunnel lining removal west
Figure 98: Detailed planning construction time tunnelling with marginal cover method (2)
List of figures
Figure 1: Tunnel boring machine [23] ................................................................................................ 14
Figure 2: Required support pressure in EPB shield [9] ....................................................................... 15
Figure 3: Principle of the URUP method [24] ..................................................................................... 16
Figure 4: TBM launching at surface level [3] ...................................................................................... 17
Figure 5: TBM arrival at surface level [3] ........................................................................................... 17
Figure 6: Overview Rijnlandroute with optimized tunnel location (white arrow) [4] .......................... 18
Figure 7: Integration of trajectory ..................................................................................................... 20
Figure 8: Cross section tunnel Rijnlandroute with dimensions [4] ...................................................... 21
Figure 9: Scope of the thesis (eastern ramp) ..................................................................................... 21
Figure 10: Activated injection locations when tunnel lining is partly above surface ........................... 29
Figure 11: Launching of the TBM at steel frame (yellow) in launching shaft [40] ................................ 29
Figure 12: Steel frame (yellow) connected to the sheet pile wall on front [40] .................................. 29
Figure 13: Launching procedure “tunnelling with marginal cover” method ....................................... 30
Figure 14: Force distribution in launching frame ............................................................................... 30
Figure 15: Steel frame connected (welded) to reinforcement ............................................................ 30
Figure 16: “First” cross connection at surface level ........................................................................... 31
Figure 17: Cross connection as designed by Movares [4] ................................................................... 31
Figure 18: Weights and lengths TBM (EPB shield) .............................................................................. 34
Figure 19: Boston Rule (approximation method) [26] ........................................................................ 35
Figure 20: Schematic overview forces acting on soil .......................................................................... 36
Figure 21: Resultant force causing a moment .................................................................................... 36
Figure 22: Stress distribution due to eccentricity ............................................................................... 37
Figure 23: Prevention of tilting .......................................................................................................... 37
Figure 24: Estimated soil structure .................................................................................................... 39
Figure 25: Height bottom tunnel - plate assumption ......................................................................... 40
Figure 26: Possible configurations of stabilized MIP columns ............................................................ 42
Figure 27: Force distribution in the soil (brown is MIP columns, blue is non-stabilized soil) ............... 42
Figure 28: Soil structure Botlek ......................................................................................................... 43
Figure 29: Triangular pattern ............................................................................................................ 44
Figure 30: Optimal MIP column pattern ............................................................................................ 46
Figure 31: stress distribution in Holocene layer ................................................................................. 47
Figure 32: Length of required ground improvement for application of the “tunnelling with marginal
cover” method .................................................................................................................................. 48
Figure 33: Starting depth tunnelling process ..................................................................................... 49
Figure 34: Estimated soil structure .................................................................................................... 50
Figure 35: Temporary situation ......................................................................................................... 52
Figure 36: Cross section tunnel with present volume for ballast [4] ................................................... 53
Figure 37: Permanent situation with concrete ballast........................................................................ 55
Figure 38: Permanent situation with slag ballast ............................................................................... 56
Figure 39: Permanent situation with sand ballast .............................................................................. 57
Figure 40: Continuum model ............................................................................................................. 58
Figure 41: Beam spring model ........................................................................................................... 58
Figure 42: Crown reduction ............................................................................................................... 58
Figure 43: Transformation of the horizontal and vertical loads to radial and tangential loads [11] ..... 59
List of tables
Table 1: Large diameter bored tunnels in the Netherlands ................................................................ 13
Table 2: Minimum and maximum jacking pressures for different depths ........................................... 24
Table 3: Friction force for the different depths .................................................................................. 26
Table 4: Total forces acting in/on TBM for different depths............................................................... 27
Table 5: Weights different parts TBM [5]........................................................................................... 33
Table 6: (increase of) stresses of the soil ........................................................................................... 35
Table 7: Friction ratio and cone resistance for different kind of soil types [6] .................................... 38
Table 8: Soil structure of location ...................................................................................................... 39
Table 9: Total settlements Holocene layer with TBM at 2.5 meter depth ........................................... 40
Table 10: Total settlements Holocene layer with embankment (5 meter) and soil replacement ........ 41
Table 11: Results of measurements at Botlekspoortunnel [32] .......................................................... 43
Table 12: Soil characteristics ............................................................................................................. 44
Table 13: Optimal MIP column pattern and other possibilities .......................................................... 46
Table 14: Soil structure of location [30] ............................................................................................. 50
Table 15: Summary of the temporary situation ................................................................................. 51
Table 16: Summary of the different zones ......................................................................................... 53
Table 17: parameters of the different soil layers (according to NEN 6740)......................................... 61
Table 18: Estimated parameters for the three calculated cross sections (see further) ....................... 61
Table 19: Safety factors ..................................................................................................................... 62
Table 20: Forces acting in the lining when lt = 170 mm ...................................................................... 64
Table 21: Forces acting in the lining when lt = 100 mm ...................................................................... 64
Table 22: Requirements for one segment.......................................................................................... 68
Table 23: Minimum and maximum reinforcement [12] ..................................................................... 69
Table 24: Forces acting in the lining when lt = 170 mm ...................................................................... 71
Table 25: Forces acting in the lining when lt = 100 mm ...................................................................... 71
Table 26: Requirements for one segment.......................................................................................... 73
Table 27: Forces acting in the lining when lt = 170 mm ...................................................................... 75
Table 28: Forces acting in the lining when lt = 100 mm ...................................................................... 75
Table 29: Requirements per segment................................................................................................ 76
Table 30: Maximum allowable mean bar diameter km [mm] ........................................................... 78
Table 31: Maximum allowable bar distance s [mm] ........................................................................... 78
Table 32: Summary costs and differences between both methods (rounded to hundreds of
thousands) ........................................................................................................................................ 79
Table 33: Comparison building time traditional and “tunnelling with marginal cover” method (1)..... 80
Table 34: (increase of) stresses of the soil ......................................................................................... 90
Table 35: Soil structure of location 1 ................................................................................................. 92
Table 36: Soil structure of location 2 ................................................................................................. 93
Table 37: Soil structure of location 3 ................................................................................................. 94
Table 38: Soil structure of location 4 ................................................................................................. 95
Table 39: Soil structure of location 5 ................................................................................................. 96
Table 40: Soil structure of location 6 ................................................................................................. 97
Table 41: Settlements when applying TBM at depth 2.5 meters. Soil replacement and embankment of
4.19 meters ...................................................................................................................................... 99
Table 42: Settlements when applying TBM at depth 4 meters. Soil replacement and embankment of
4.19 meters ...................................................................................................................................... 99
Table 43: Settlements when applying TBM at depth 7 meters. Soil replacement and embankment of
3.44 meters ...................................................................................................................................... 99
Table 44: Settlements when applying TBM at depth 7.53 meters. Soil replacement and embankment
of 3.24 meters ................................................................................................................................ 100
Table 45: Settlements when applying TBM at depth 8.53 meters. Soil replacement and embankment
of 2.84 meters ................................................................................................................................ 100
Table 46: Settlements when applying TBM at depth 9.53 meters. Soil replacement and embankment
of 2.43 meters ................................................................................................................................ 100
Table 47: Settlements when applying TBM at depth 10.53 meters. Soil replacement and embankment
of 1.99 meters ................................................................................................................................ 101
Table 48: Settlements when TBM rests on Pleistocene layer (depth = 11 meters). No ground
improvement .................................................................................................................................. 101
Table 49: Temporary situation with safety factors ........................................................................... 104
Table 50: Permanent situation with safety factors and concrete ballast .......................................... 105
Table 51: Permanent situation with safety factors and slag ballast .................................................. 106
Table 52: Permanent situation with safety factors and sand ballast ................................................ 107
Table 53: Limit of shear stresses in concrete ................................................................................... 113
Table 54: Maxima critical cross sections .......................................................................................... 114
Table 55: Detailed planning construction time traditional method (1) ............................................. 118
Table 56: Detailed planning construction time tunnelling with marginal cover method (1) .............. 124