-
TENDER DOCUMENT SPECIFICATIONS FOR PIPE UMBRELLA INSTALLATION
METHODS
Volkmann, Gnther M. ALWAG Tunnelausbau Gesellschaft m.b.H.,
Pasching/Linz, Austria
[email protected]
Schubert, Wulf Graz University of Technology, Institute for Rock
Mechanics and Tunneling, Graz, Austria
[email protected] Abstract: Pipe Umbrella support systems are
effectively used as pre-support systems in tunneling to increase
stability in the working area and decrease deformations induced by
the construction. Its main application area is in weak ground
tunneling. Either a Pre-Drilling or a Cased-Drilling method can
install this system. Due to their characteristics, both
installation methods influence stability demands and deformations
in a different way. The negative influences can be minimized by
choosing the adequate installation method or pipe type. This
article discusses differences inherent in installation methods of
apparently equal Pipe Umbrella systems. 1. INTRODUCTION Many
subsurface infrastructures are constructed in urban areas. In these
regions, ground conditions are often weak. For this reason,
conventional tunneling needs additional measures to ensure stable
conditions in the working area during construction. Often used
measures are face bolts, spiles, pipe umbrellas, pipe roofs and
jet-grouted columns. The application of these systems increases the
stability and decreases the construction-induced deformations. A
Pipe Umbrella support consists of pipes, which are installed from
the actual face to the front. Typically, the pipes are 12.0 m or
15.0 m long and arranged at the outer shape of the tunnel. The
outer diameter of the pipes ranges from 70 mm to 200 mm. This
support system is primarily working ahead of the primary lining so
its influence on stability conditions and deformations is difficult
to observe. The application of a horizontal inclinometer chain
allows measuring deformations of installed pipes so that the
results of the state-of-the-art geodetic surveys can be
complemented with the in-situ data measured ahead of the face. The
data can be used to control and optimize the ongoing construction
process and to detect any new phenomena. Two construction-relevant
issues that were detected with this measurement technique are
discussed in the following. These are the influences of the
installation method on stability conditions and deformations as
well as the used pipe type. 2. INCLINOMETER MEASUREMENT SYSTEM
Geodetic observations at the tunnel level only survey deformations
in the already supported section. Because Pipe Umbrella systems
influence the ground support interaction mainly ahead of this
section, horizontal inclinometers were utilized to monitor the
settlements of the Pipe Umbrella roof pipes. The used
instrumentation consists of 10 horizontal inclinometer links. Each
of these links is 2 m long and
Volkmann G.M. & Schubert W. 2008. Tender Document
Specifications for Pipe Umbrella installation methods. In Proc. of
the 34th ITA-AITES World Tunneling Congress, Agra, India, 22-24
September 2008, pp. 285-293
-
Figure 1: The measured time - settlement line shows the
influence of each construction step on the settlements in the crown
(1).
connected to the next one. Like this, a continuous horizontal
inclinometer chain monitors the changes in inclination in an up to
20 m long section ahead of the face. The instrumentation is
connected to a data acquisition system, which stores the measured
data in pre-defined time intervals. The data acquisition system can
either be situated in the tunnel or outside in an office.
Evaluations presented in this publication are based on data stored
in minute intervals so that the development of deformations can be
observed in detail. This instrumentation records the full path of
settlements (including pre-settlements) and it is possible to
analyze, control and optimize the tunnel construction process
depending on project requirements with adequate evaluations (1).
The quality of the evaluated data is shown exemplarily in figure 1.
It shows the development of settlements near the face due to one
excavation round. This round consists of 5 excavation steps and the
installation of different support components for the primary
lining. The increase of settlements can be seen during excavation
phases as well as the characteristic time-dependent stabilization
process afterwards. The installation of the steel arch and the
primary shotcrete layer did not interrupt this stabilization
process after excavating phase 3 but the outstanding point in this
diagram is the significant settlement increase during the
installation of radial bolts and micropiles. The installation of
these two support systems caused 3 - 4 mm of additional settlement.
Compared to this value the 16 mm for the entire excavation round
looks comparatively small. Weak ground is therefore sensitive on
the dynamic loading of drilling and the stress transfer induced by
drilling boreholes. 3. INSTALLATION METHODS Two methods are
available for installing Pipe Umbrellas. These are called the
Pre-Drilling method and the Cased-Drilling method (2). Their
characteristics are explained in the following paragraphs. 3.1
Pre-Drilling Method The Pre-Drilling Method is characterized by a
two-step installation procedure for each pipe. At first, a hole is
drilled and then in a second step the pipe is pushed into the
pre-drilled hole. Sub-contractors with specialized drilling
machines and extra crews are usually required to install pipes with
a Pre-Drilling method. These specialized machines have one long
boom and the hole is drilled in one piece after rigging the boom
(figure 2). The cooling of the drill bit and the flushing can be
performed with water or air. The flushing medium is moved to the
front inside the drill rod and removes the cuttings in the annulus
between the drill rod and the borehole walls. After finishing the
drilling process, the drill bit is repeatedly moved forward and
backwards to clean and reshape the drill hole, if necessary. After
replacing the drill rod, a one-piece pipe is placed into the
borehole. Usually, a few boreholes are prepared before the pipes
are placed into the holes. That way less time is consumed for
replacing the drill rod.
Volkmann G.M. & Schubert W. 2008. Tender Document
Specifications for Pipe Umbrella installation methods. In Proc. of
the 34th ITA-AITES World Tunneling Congress, Agra, India, 22-24
September 2008, pp. 285-293
-
Figure 2: Installation of Pipe Umbrella support by different
machines. The Pre-drilling method uses a special machine (left
side) while conventional drill jumbos install the pipes by using
the Cased-Drilling method (right side)
3.2 Cased-Drilling Method The Cased-Drilling Method is
characterized by using only one-step for the installation. By using
this method, the pipe follows directly behind the drill bit during
drilling and stays in place after completing the drilling process.
The crews can install this system by using conventional drill
jumbos. Drill jumbos in Sequential Excavation Tunneling (SEM) have
usually 2 booms that can be used parallel during installation
(figure 2). This fact decreases the time necessary for the
installation. John & Mattle (3) mention that one Pipe Umbrella
installation (600 m of pipes) can be finished in a 24-hours shift.
This time includes the grouting process. A Pipe Umbrella support is
longer than a boom so the drilling process is interrupted typically
every 3 meters. The drilling process starts again after connecting
an extension pipe and drill rod. Mostly water is applied as
flushing and cooling media, respectively. The water moves through
the drill rod to the front and removes drill cuttings from the
borehole inside the casing pipe. When the drilling process for the
last pipe piece is finished, the installation is completed by
removing the drill-rod pieces. 3.3 Grouting Process The grouting
process for the Pipe Umbrella support starts after installing the
pipes. The grout is used to fill the pipe, the annular gap between
the pipe and the surrounding ground, and open pores (soil) or
joints (rock) surrounding the bore hole. In general, the grout
consists of a water-cement suspension with a water-cement ratio
ranging from 0.45 to 0.80. This suspension is usually pumped into
the pipes at a low pressure level (< 10 bar). For
quality-control reasons the grout volume as well as the grout
pressure is generally monitored during the entire grouting process.
The grout causes different effects: The grout inside the pipes
increases the flexural strength and ensures that the shape of the
cross section stays un-deformed when the pipes are loaded during
construction. The grout in a remaining annular gap between pipe and
ground increases the load transmission between ground and support
during construction. Whenever grout can penetrate the surrounding
ground, its strength and stiffness properties are positively
influenced as soon as the grout hardens.
Volkmann G.M. & Schubert W. 2008. Tender Document
Specifications for Pipe Umbrella installation methods. In Proc. of
the 34th ITA-AITES World Tunneling Congress, Agra, India, 22-24
September 2008, pp. 285-293
-
Figure 3: The inclinometer measurements resulted in differing
settlement amounts during installing Pipe Umbrellas. The measured
settlement values for the Cased-Drilling method (left side) are in
average clearly smaller than the settlement values for a
Pre-Drilling method (right side).
3.4 On-site Measurements Boreholes for Pipe Umbrellas start at
the face and are orientated to the front. When this construction
step causes deformations, the commonly performed 3-dimensional
geodetic survey in tunnels does not monitor most of these
deformations, which occur mainly around the boreholes ahead of the
face. An additional measurement system, the above-explained
inclinometer measurement system, allows recording deformations
ahead of the face during pipe installation as well. The
deformations are measured about 1.0 to 1.5 m above the boreholes
with this instrumentation. Measurements were performed during pipe
installations of both described installation methods. The
Cased-Drilling method was monitored at the Birgl Tunnel (Austria)
while the Pre-Drilling method was observed at the Trojane Tunnel
(Slovenia). A comparison of the measured results is feasible
because the tunnel size as well as the ground quality is similar.
Results of these campaigns are shown in Figure 3. The displayed
settlement values are settlements that were recorded for installing
and grouting of one Pipe Umbrella support. Each line represents the
result after one support installation. In the diagrams, the tunnel
construction proceeds from the left to the right side. The
measurements are shown relative to the face position so a positive
chainage means ahead of the face while minus values are situated in
the already excavated and supported tunnel section. The left
diagram shows settlements monitored at the Birgl Tunnel, where the
Cased-Drilling method was used for installing the pipes. The
maximum value, measured during installation, is less than 10 mm and
its position is near but ahead of the face. The characteristics of
these lines are similar to those measured for an excavation step
and in fact, the evaluation of the associated time settlement lines
shows a small, still ongoing, and abating time-dependent settlement
increase, which was mainly caused by the last excavation step. The
right diagram presents a completely different picture. These are
settlements recorded at the Trojane Tunnel during pipe installation
with the Pre-Drilling method. The monitoring clearly shows that the
settlements occur primarily ahead of the face. The amounts vary
with distance to the face and do not show any logical distribution.
These additional deformations are definitely caused by the drilling
process because the increase in settlements happened only in
periods of drilling. The associated time settlement lines can prove
this. Four out of six monitored installations increased the
settlements relatively small (< 15 mm) but the settlements
measured for two installations increased the settlements ahead of
the face up to nearly 40 mm in the monitored section. The
monitoring of the following excavation rounds explained the cause
for this increase. Wherever the settlements during drilling
increased, the ground conditions got softer. 3.5 Numerical
Simulation A preceding study, which used simple analytical
formulations, indicated that the stress redistribution caused by
single boreholes (150 to 200 mm in diameter) nearly causes an
overloading of the ground. The analytical formulation could not
consider an interaction of adjacent holes so additional numerical
simulations were performed with UDEC. The 2-dimensional calculation
was performed using a 40 m wide and 40 m high model. The symmetric
conditions were used so only half of the tunnel was calculated. The
cover above the crown pipe was
Volkmann G.M. & Schubert W. 2008. Tender Document
Specifications for Pipe Umbrella installation methods. In Proc. of
the 34th ITA-AITES World Tunneling Congress, Agra, India, 22-24
September 2008, pp. 285-293
-
Figure 4. Resulting deformations induced by different
installation methods calculated by UDEC. The un-deformed mesh on
the left side deforms less with Cased-Drilling methods (centre)
than with Pre-Drilling methods (right). Each plot is 30 cm x 30 cm
and the displayed deformations are scaled to absolute values
(4).
set to 15 m and all together, a number of 15 boreholes (half
side) was considered. The maximum length of the generated mesh was
defined to be smaller than 5 mm near the boreholes and in the areas
that developed plastic zones during calculating stability. The
results of two exemplary calculations are shown in figure 4. The
left picture in this figure presents a 300 mm x 300 mm sized detail
of the model. It shows an un-deformed borehole, a pipe, and the
annular gap between pipe and ground. Starting with the un-deformed
mesh, two calculations were carried out until stability. The first
one illustrates a Cased-Drilling method with a pipe in the borehole
(centre) and the second one the Pre-Drilling method without the
pipe (right). The stability calculation for the Cased-Drilling case
resulted in deformations, which closed the annulus between pipe and
ground nearly over the entire perimeter. However, as soon as the
ground touches the pipe, the pipe supports the ground so no further
deformations occur. In the second case the borehole walls can
deform unhampered and the borehole closes due to the induced stress
transfer. In fact, the calculated settlements had the same
magnitude as the measured settlements at the Trojane Tunnel for the
same construction process. The remaining hole is definitely smaller
than a pipe so the hole must be reshaped before the pipe can be
installed. Again, this process induces stress redistribution and
deformations respectively. This amount of deformation is expected
to be smaller than the initial one but should not be neglected.
Summarizing, this simple 2-dimensional simulation proves that the
deformations induced by removing ground from boreholes can cause
considerable settlement amounts in weak ground. 3.6 Consequences /
Influence on Stability Conditions The stability of the unsupported
span is an important issue for using a Pipe Umbrella method as an
additional pre-support system. For this reason, a proper
installation of all pipes is necessary because the Pipe Umbrella
protects the workers during a later construction stage from falling
ground (like an umbrella that protects people from getting wet in
the rain). The characteristics of a Cased-Drilling method cause a
shear force between the pipe and the ground during installation so
the machine that is chosen for installation must have enough power
to overcome the created retaining force. State-of-the-art drill
rigs do usually have enough power so this should only be a problem
in special cases. The closure of the borehole when using a
Pre-Drilling method can create a serious problem when it is not
detected or detected too late. Once the borehole is smaller than
the pipe, the strongest machine cannot push a pipe into the
borehole because the pipe fails first by buckling (Figure 5).
Additionally, time-consuming measures must be applied in the
section where the pipe could not be installed to ensure the safety
of the workers. Two new pipes need to be installed adjacent to the
failed one or additional spiles can be installed later during
construction. The ground that is additionally supported by a Pipe
Umbrella is normally sensitive on water. The flushing removes the
drill cuttings between drill rods and ground when using a
Pre-Drilling method so the ground properties can be influenced
negatively by water flushing over the entire length of the
borehole. This negative influence may cause instabilities at the
face region during subsequent
Volkmann G.M. & Schubert W. 2008. Tender Document
Specifications for Pipe Umbrella installation methods. In Proc. of
the 34th ITA-AITES World Tunneling Congress, Agra, India, 22-24
September 2008, pp. 285-293
-
Figure 5: One Pipe Umbrella pipe that could not be installed
properly with a Pre-Drilling method.
construction steps. For this reason, an air flushing should be
preferred as soon as the ground is sensitive on water influence.
When using a Cased-Drilling method, the negative influence of water
on the ground properties is limited to a short section in the area
of the drill bit, because, as mentioned above, the flushing is
performed inside the casing pipe. 3.7 Influence on Subsidence Pipe
Umbrella systems are not only installed for stability reasons but
also to decrease the settlements induced by the construction. The
decrease in settlements cannot be numbered clearly but when the
installation of a support system increases the settlements up to 40
mm the advantage of decreasing settlements abates or disappears.
On-site measurements as well as numerical simulations show that
this effect does not appear when a pipe supports the borehole
immediately behind the drill bit. Hence, a Cased-Drilling system
should be preferred where subsidence is of relevance. Particularly,
when a stability and deformation check raises suspicion for
conditions similar to the above described. The way of transporting
drill cuttings may also increase subsidence when using a
Pre-Drilling method because of the erosion effects in a borehole.
This erosion may increase the borehole in diameter. Consequently,
the deformations due to the boreholes increase as well. 4. PIPE
UMBRELLA PIPES Pipes used for a Pipe Umbrella support system do not
only have a certain outer diameter and wall thickness but also the
pipes do have some additional features, which may influence the
strength and stiffness parameters of the installed support. For
this reason, bending tests were performed with grouted and
un-grouted steel pipe samples. Figure 6 shows an outline of the
dimensions of the testing apparatus. The relative displacement of
three measurement points measures the samples deflection. This
value was used as a feedback command controlling the test
procedure. Back calculations from the on-site data showed that the
pipes are generally loaded in the elastic range. However, higher
loads leading to a failure of the pipes cannot be excluded. Thus,
the tests focus was set to the elastic range. Nevertheless, the
samples were loaded until failure or a significant drop in
resistance occurred. 4.1 The Influence of Grout A first test series
had the goal to investigate the influence of grout on the pipes
flexural strength. The water-cement ratio was decided to be 0.45
for all tests. 5 different sample types were tested: empty pipes,
pipes 100% filled with grout (7 days and 28 days hardened), pipes
filled 75% with grout, and pipes filled 75% with grout and the
remaining volume filled with sand. The last sample type
replicates
Volkmann G.M. & Schubert W. 2008. Tender Document
Specifications for Pipe Umbrella installation methods. In Proc. of
the 34th ITA-AITES World Tunneling Congress, Agra, India, 22-24
September 2008, pp. 285-293
-
Figure 6. Bending test arrangement
Figure 7. Bending test results of a grouted regular pipe and
samples with injection holes
the case when drill cuttings are not totally flushed out of the
pipe during installation. Therefore, the bottom of the horizontal
pipe is filled with drill cuttings while the remaining area is
filled with grout. The last but one sample type represents a
non-perfect grouting. Results of this test series showed that the
flexural strength of all sample types did not vary much. The only
mentionable difference was observed at empty pipes. The shape of
the cross section slowly changed at higher loading levels. This
ovalization could only be seen at greater outer diameters (114.3 mm
& 139.7 mm) and resulted in a 15% decrease of flexural
strength. For this reason, the Pipe Umbrella pipes should be fully
grouted before the excavation starts. This ensures that the cross
section of pipes stays un-deformed and buckling effects cannot
decrease the theoretical strength. 4.2 The Influence of Injection
Holes The pipes for the Pipe Umbrella support are usually provided
with injection holes for grouting. This enables the grout to fill
the annular gap between pipe and ground. Grout can also infiltrate
open joints or pores in the ground around the borehole. The
injection holes reduce the moment of inertia so this
Volkmann G.M. & Schubert W. 2008. Tender Document
Specifications for Pipe Umbrella installation methods. In Proc. of
the 34th ITA-AITES World Tunneling Congress, Agra, India, 22-24
September 2008, pp. 285-293
-
Figure 8. Bending test results of a grouted regular pipe and
different connection types
feature may reduce the strength and stiffness properties of
grouted pipes. For this reason, fully grouted pipes (7 days
hardened) with injection holes in the middle of the sample were
tested in a second test series. The test results of this series are
shown in figure 7. The solid line presents the test result of a
regular, grouted pipe (114.3 mm x 6.3 mm) while the dashed and
chain dotted line show the test results of samples with injection
holes. As can be seen in the pictures of figure 8 the injection
holes were tested at neutral axis as well as outer fiber position,
but no significant difference could be monitored. Therefore, the
injection holes do not influence the flexural strength at
construction relevant loadings. 4.3 The Influence of Connections
Typically, Pre-Drilling methods install a one-piece pipe so this
case is only important for Cased-Drilling methods. At this method,
3 m long pipe pieces are connected to one another by thread
connections during installation. The moment of inertia is reduced
in the threaded section as well, so the third test series bended
grouted samples with different connection types in the middle. This
test series clearly identified the usually used thread connection
as the weakest link in a Pipe Umbrella support. The picture on the
right side in figure 9 shows the thread connection after failure.
The flexural strength as well as the load of failure is lower than
those of regular pipes so the strength and stiffness used for the
design must be limited to the values of the thread connection for
safety reasons. This fact reduces the performance of Pipe Umbrella
pipes, which leads to the development of a new pipe connection
type, the so-called Threaded Nipple Coupling. As can be seen in
figure 8 the sample with this connection type resolves the problem
by showing a comparable flexural strength as regular pipes.
Additionally, the load of failure was always higher for the
Threaded Nipple Couplings than for regular pipes during testing.
Therefore, the Threaded Nipple Couplings compensate the weakness of
pipe connections and increase the effectiveness of piecewise
installed Pipe Umbrella pipes to the level of regular pipes.
Volkmann G.M. & Schubert W. 2008. Tender Document
Specifications for Pipe Umbrella installation methods. In Proc. of
the 34th ITA-AITES World Tunneling Congress, Agra, India, 22-24
September 2008, pp. 285-293
-
5. CONCLUSION Pipe Umbrella systems are effectively used to
increase the stability in the working area and to reduce the
construction induced settlements in weak ground conditions. A
longitudinal transfer of loads achieves these effects. Each pipe
transfers the loads from the supported areas to the less critical
areas, which are used as abutments. At weak ground conditions, even
boreholes for bolts or pipes may be unstable and drilling those
increases the overall induced deformations related to construction.
Due to this fact, the definition of the installation method may get
an important point. Particularly in urban areas, subsidence often
rules the design. A Cased-Drilling method should therefore be
preferred at weak ground conditions because it is less susceptible
to create additional deformations during installation.
Cased-Drilling methods are installed by using conventional drill
rigs so each Pipe Umbrella pipe is installed piecewise. When the
pipe is loaded during construction, the usually used thread
connections are a weak link in the system. This leads to an
un-effective utilization of the regular pipe sections. This can be
countered by changing to Threaded Nipple Couplings, which have the
same flexural strength as regular pipes. The proper choice of
installation method and pipe type must agree with the project
requirements. A Cased-Drilling method with Threaded Nipple
Couplings is recommended if the project requirements include
settlement limitations in strain-sensitive ground. This choice
combines the advantages of different methods during installation
and construction. References (1) Volkmann, G.M., & W. Schubert.
2005. The Use of Horizontal Inclinometers for the Optimization
of the Rock Mass Support Interaction. In Proc. of the 31st
ITA-AITES World Tunneling Congress, Underground Space Use: Analysis
of the Past and Lessons for the Future, 7-12 May 2005, Istanbul,
Turkey; page 967-972; eds. Y. Erdem & T. Solak; A.A. Balkema
Publishers, London, ISBN 04 1537 452 9
(2) Volkmann, G. 2004. A Contribution to the Effect and Behavior
of Pipe Roof Supports. In Proc. of EUROCK 2004 & 53rd
Geomechanics Colloquium. Ed. Schubert, 2004 VGE, ISBN
3-7739-5995-8
(3) John, M. & B. Mattle 2002. Design of Tube Umbrellas.
Tunel, 11. Ronk, . 3/2002. Magazine of the Czech Tunnelling
Committee and Slovak Tunnelling Association.
(4) Volkmann, G.M., & W. Schubert. 2006. Optimization of
Excavation and Support in Pipe Roof Supported Tunnel Sections. In
Proceedings of the 32nd ITA-AITES World Tunneling Congress, Safety
in the Underground Space, eds. In-Mo Lee, Chungsik Yoo &
Kwang-Ho You, April 22-27 2006, Seoul, Republic of Korea, ISSN
0886-7798, Elsevier Ltd.
Volkmann G.M. & Schubert W. 2008. Tender Document
Specifications for Pipe Umbrella installation methods. In Proc. of
the 34th ITA-AITES World Tunneling Congress, Agra, India, 22-24
September 2008, pp. 285-293
-
BIOGRAPHICAL DATA
In 2001, Mr. Volkmann graduated in Civil Engineering at Graz
University of Technology. As Researcher and Lecturer at the
Institute for Rock Mechanics and Tunneling, he started his PhD on
Pipe Umbrella Support Systems. Since 2005, he also works for ALWAG
Company, a subsidiary of DYWIDAG-Systems International. At ALWAG,
he is also responsible for research and development for the AT -
Umbrella Systems. The still ongoing research resulted in more than
10 international publications until now. CERTIFICATE The author
certifies that the paper titled Tender Document Specifications for
Pipe Umbrella Installation Methods and submitted for consideration
for WTC 2008 on July 11, 2008, to be held in from 22-24 September
2008 at Agra in original and has not been published or presented at
any other forum.
....................... Signature
Place: Graz, Austria Dated: July 11, 2008
Volkmann G.M. & Schubert W. 2008. Tender Document
Specifications for Pipe Umbrella installation methods. In Proc. of
the 34th ITA-AITES World Tunneling Congress, Agra, India, 22-24
September 2008, pp. 285-293