Geotechnical Research Geotechnical Research, 2017, 4(3), 178–191

Volume 4 Issue GR3 http://dx.doi.org/10.1680/jgere.17.00011

Paper 17.00011
Benefits of pre-loading temporary props for Received 25/07/2017; accepted 06/09/2017
Published online 12/10/2017
braced excavations
Keywords: design methods & aids/diaphragm & in situ walls/temporary works
Latapie, Winarta and Lloyd
Published with permission by the ICE under the CC-BY 4.0 license.
(http://creativecommons.org/licenses/by/4.0/)

Benefits of pre-loading temporary props for

braced excavations
1 Benoît Latapie MSc, EurIng 2 Terry Winarta MEng (Hons)
Technical Manager, Ground Engineering, WS Atkins & Partners Overseas, Geotechnical Engineer, Ground Engineering, WS Atkins & Partners
Dubai, UAE (corresponding author: [email protected]; Overseas, Dubai, UAE
[email protected]) 3 Steven David Lloyd BSc (Hons)
Development Manager, Groundforce Shorco, Leeds, UK

1 2 3

Using soil profiles typical of excavations in London, UK and Dubai, UAE calculations have been carried out to
compare the performance of temporary steel and proprietary hydraulic propping. Numerical modelling indicates that
pre-loading allows the lower-stiffness hydraulic prop to match the performance (in terms of prop load and wall
displacement) of a higher-stiffness steel prop, without pre-load. The low prop stiffness also helps to mitigate loading
induced by thermal effects. Furthermore, where limiting the horizontal deflection of the wall is critical, the
application of a calibrated pre-load can improve the economy of the temporary works by reducing the prop size.
Pre-loaded props are sometimes associated with increased bending moment and shear force values in the retaining
wall. The finite-element models produced for this paper prove that for the cases considered, the use of pre-loaded
low-stiffness hydraulic props achieved the same performance as stiff steel props, with no impact on the wall’s
reinforcement. Further work is suggested to explore how manipulating and varying the pre-loading values in
multipropped basements might be used to improve further the performance and/or the economy of the retaining
wall and how this might affect the load transfer from temporary to permanent supports in the longer term.

Notation Ppre-load pre-load of a hydraulic prop

A cross-sectional area of a propping element Psteel working load of a steel prop
Ahyd cross-sectional area of hydraulic element of a Rinter strength reduction factor for soil–structure interface
proprietary prop S stiffness of a propping system expressed as stiffness
Asteel cross-sectional area of steel element of a per unit length of wall: (kN/m)/m
proprietary prop S0 stiffness of a single prop: kN/m
c0 cohesion (drained) Sn stiffness of element n within a prop from 1 up to N
cu undrained shear strength (cohesion) t wall’s thickness
d horizontal spacing distance of props wwall wall’s self-weight
E Young’s modulus of materials zLC depth below the top of the London Clay
EA axial stiffness a angle of prop axis away from the perpendicular to
Econcrete Young’s modulus of concrete the wall
Ehyd Young’s modulus of hydraulic element of a at coefficient of linear thermal expansion of steel prop
proprietary prop b empirical factor for percentage degree of restraint
Esteel Young’s modulus of steel element of a against thermal strain
proprietary prop g unit weight of soil/rock materials
(EA)wall wall’s axial stiffness gsat saturated unit weight of soil/rock materials
(EI )wall wall’s bending stiffness DPtemp change in axial prop load due to temperature change
Ko coefficient of at-rest earth pressures Dt change in prop temperature
L0 initial length of a single prop dpre-load positive value of the net horizontal deflection of the
L0,hyd initial length of hydraulic element of a retaining wall at the prop’s location, induced by
proprietary prop applying Ppre-load
L0,steel initial length of steel element of a proprietary prop u Poisson’s ratio
Phyd working load of a hydraulic prop f0 angle of internal friction (drained)

178
Downloaded by [] on [25/07/18]. Published with permission by the ICE under the CC-BY license
 Geotechnical Research Benefits of pre-loading temporary props
Volume 4 Issue GR3 for braced excavations
Latapie, Winarta and Lloyd

Introduction prop. The ram remains an active part of the prop throughout its
Shoring is a necessary part of almost every deep basement service life. Figure 1 shows hydraulic props in service; the
excavation where retention of soil materials and/or sustaining telescopic sleeves containing the hydraulic rams are labelled
groundwater pressures is required. The most common construction ‘hydraulic unit’.
method uses a ‘bottom–up’ technique, where the retaining wall is
constructed initially and the excavation and installation of the Previous literature has often focused on the effect of pre-loading of
support frames/props proceed downwards in stages. In this steel props. Construction Industry Research and Information
situation, the permanent works are constructed from the bottom of Association report C517 by Twine and Roscoe (1999) lists a
the supported excavation, working upwards. An alternative to this number of methods which may be used to introduce pre-load into a
is a top–down construction method, which is not in the remit of conventional steel prop. Such methods include driving wedges to
this paper. The methods used vary greatly in terms of execution, secure the props or the waling (which introduces a small pre-load
but from a design point of view, they tend to follow a very similar to the system) and using a removable jacking device between the
pattern: an embedded retaining wall (e.g. diaphragm wall, secant/ steel prop and waling to push back the wall before installing the
contiguous bored pile wall, steel sheet piles) is supported by one or prop. Practical site experience shows that the use of such remote
more levels of horizontal beam (e.g. concrete capping beam, steel jacking systems requires additional installation time and in turn
waling beam), which is in turn supported by a series of props or additional cost when compared to a modern proprietary hydraulic
anchors (e.g. raking props, flying shores, ground anchors). Shallow prop that has this facility built in. The control and monitoring of the
excavations, which are not in the remit of this paper, are sometimes pre-load value is also much easier and more precise when using the
retained with a retaining wall alone, acting as a cantilever. hydraulic system. Figure 2 shows the high-pressure couplings
housed within the prop casing and the motorised pump that
In this paper, numerical modelling of deep propped excavations is controls the flow of hydraulic fluid during the standard installation
used to investigate the potential benefits of pre-loading (or pre- process. In this particular case, the corner prop shown was installed
stressing) the structural supports on the performance of the to the required position and pre-loaded in less than an hour.
temporary work system. Different excavation depths and ground
conditions are considered in order to ensure that the conclusions Hydraulic actuation provides practical benefits during the
drawn in this paper are not reliant on conveniently chosen installation process and facilitates greater control over any pre-
assumptions. Using past experience in both Dubai (UAE) and loading that needs to be induced into the prop. However, using a
London (UK), representative temporary work scenarios were hydraulic ram to support load results in a prop that is intrinsically
recreated and used in the analyses. The paper demonstrates that less stiff when compared to a structurally equivalent steel prop.
carefully selected values of pre-load may be used as substitutes This is because the stiffness of a hydraulic prop is limited by the
for inherent prop stiffness. A discussion on the benefits of using
props with lower stiffness, particularly with regard to dealing with
the effect of temperature, is also included. Ground anchors and
other support methods are not considered in this paper.
Proprietary hydraulic props are used when analysing the scenarios
where pre-load is introduced in the support system.

A short history of hydraulic propping

Proprietary hydraulic propping systems have been in use in the
UK for well over 30 years. The benefits of using hydraulic rams
to support relatively shallow excavations are widely accepted
there, mostly from an ease of installation and practicality point of
view. The use of this technique to support deep excavations has
led to the development of high-load-capacity props over the last
10 years. The UK shoring industry has reached the point where
proprietary rental equipment, for most temporary work
applications, is a cost-effective alternative to traditional welded
steel framework. In some cases – for example, deep complex
cofferdam layouts where the ground and groundwater loads are unit
aulic
very high – the traditional system remains the preferred (and Hydr
only) option as the current range of proprietary rental equipment ions
extens
cannot accommodate such loads. Steel

The modern proprietary hydraulic shoring props discussed in this Figure 1. Proprietary hydraulic props in service
paper include a hydraulic ram which is an integral part of the

179
Downloaded by [] on [25/07/18]. Published with permission by the ICE under the CC-BY license
 Geotechnical Research Benefits of pre-loading temporary props
Volume 4 Issue GR3 for braced excavations
Latapie, Winarta and Lloyd

Hydraulic strut installed as one unit; Introduction of hydraulic pressure to lock

hydraulic pressure not yet applied prop in position and apply pre-stress

Hydraulic unit extension

Figure 2. Proprietary hydraulic prop in service

stiffness of the column of fluid in compression within the ram expands on this research and demonstrates that pre-load used in
itself. This is discussed in greater detail in the section headed combination with low-stiffness hydraulic props has a beneficial
‘Determination of the axial stiffness of hydraulic and non- effect on the design of temporary retaining walls, in terms of both
hydraulic props’. Figure 3 shows the difference between a economy and performance.
hydraulic prop and a conventional steel prop.
Ground model assumptions
Irrespective of the prop type, all props used to support deep For the purpose of this paper, two different ground conditions have
temporary excavations have two main properties that control their been adopted to explore the benefits of pre-loading temporary
design: structural strength and stiffness. For a traditional steel prop, propping for deep-excavation works. Representative ground
these two properties are directly related, as an increase in steel cross- conditions (stratigraphy and ground parameters) for both London and
section directly translates to a stronger and stiffer prop. This Dubai were chosen by the authors due to their experience working
relationship is not as simple when considering hydraulic props as on deep-excavation projects in these two cities. These ground
strength and stiffness are related to hydraulic ram properties as well conditions have been idealised to ease the comparison between the
as steel section properties. Hydraulic props are able to achieve a different excavation depths considered and are not meant to be
strength comparable to that of a conventional steel prop with a lower directly applicable to any particular site. The assumptions and
overall stiffness. This will be discussed in detail later in the paper. calculations presented in this paper should not be used in place of
site-specific site investigation and project-specific design calculations.
Previous research by Chambers et al. (2016) showed that high
prop pre-load values are not generally necessary to achieve The stratigraphy for typical ground condition in Dubai consists of
satisfactory performance in terms of limiting wall movement. The cemented sand overlying a bedrock of highly weathered to
research suggests that in the case of the Crossrail Paddington moderately weathered calcarenite and calcisiltite. Many Dubai
station box, a smaller (i.e. c. 30% of design load) pre-load would projects were constructed near its coastline where the groundwater
have offered satisfactory results with far greater economy level is relatively high, close to the existing ground level. In this
compared to what was used in the construction. This paper paper, the groundwater level is assumed at 3 m below the existing
ground level for the Dubai ground conditions. The corresponding
ground model is illustrated in Figure 4.
Hydraulic unit Steel extensions
The typical ground condition for the London model consists of
made ground and river terrace deposits overlying the London
Hydraulic ram Clay formation. The groundwater level is considered at 6 m below
(a)
the ground level, which is the interface between the made ground
Steel prop and river terrace deposits. The corresponding ground model is
illustrated in Figure 5.

Welded joints Welded endplate detail The geotechnical parameters assumed for both ground conditions
(b) are summarised in Table 1. These parameters were selected after
reviewing several factual site investigation reports for Dubai and
Figure 3. Comparison between (a) hydraulic prop and (b) typical
steel prop London projects. The authors’ intention is to analyse a set of
typical or representative parameters for each location.

180
Downloaded by [] on [25/07/18]. Published with permission by the ICE under the CC-BY license
 Geotechnical Research Benefits of pre-loading temporary props
Volume 4 Issue GR3 for braced excavations
Latapie, Winarta and Lloyd

5 m DMD 5 m DMD
Groundwater level S1 Groundwater level S1
2 m DMD 2 m DMD 2 m DMD 2 m DMD
Cemented sand
11 m

Cemented sand

11 m
−3 m DMD S2
−4 m DMD
−6 m DMD −6 m DMD
−7 m DMD S3
−9 m DMD
15 m

15 m
Calcarenite Calcarenite
−15 m DMD

−21 m DMD −19 m DMD

−21 m DMD
>50 m

>50 m
Calcisiltite Calcisiltite

(a) (b)

Figure 4. Stratigraphy and analysed model for Dubai: (a) propped cantilever – Dubai (PCD) and (b) multipropped – Dubai (MPD). DMD,
Dubai Municipality Datum

16 m OD 16 m OD
S1 S1
6m

Made ground Made ground

6m

13 m OD 12·5 m OD
Groundwater level
Groundwater level 10 m OD
10 m OD S2
8·5 m OD
5m

River terrace deposits

5m

River terrace deposits 8 m OD

5 m OD 5 m OD S3
4·5 m OD
3 m OD
>30 m

>30 m

1 m OD
London Clay
London Clay

−5 m OD

(a) (b)

Figure 5. Stratigraphy and analysed model for London: (a) propped cantilever – London (PCL) and (b) multipropped – London (MPL). OD,
Ordnance Datum

Table 1. Characteristic geotechnical parameters of soil/rock materials

Lithology f : kN/m3 c0 : kPa cu: kPa e 0: ° E: MPa t Ko
Dubai
Cemented sand 19 2 — 32 50 0·30 0·47
Calcarenite 21 125 — 40 200 0·25 0·36
Calcisiltite 21 300 — 30 750 0·25 0·50
London
Made ground 20 0 — 32 12 0·25 0·61
River terrace deposits 20 0 — 36 100 0·15 0·41
London Clay 20 — 100 + 5·34zLCa — 80 + 4·27zLCa 0·495 1·00
a
These equations highlight the fact that undrained shear strength and stiffness for the London Clay vary with depth, where z is the depth below the top of the London Clay.
They were obtained from a specific past project experience in London, where the values are derived from standard penetration test correlations and triaxial test results

181
Downloaded by [] on [25/07/18]. Published with permission by the ICE under the CC-BY license
 Geotechnical Research Benefits of pre-loading temporary props
Volume 4 Issue GR3 for braced excavations
Latapie, Winarta and Lloyd

Geotechnical modelling construction effects. To model this in Plaxis 2D, an Rinter

To predict the relationship between shoring stiffness and parameter of 0·7 was taken. This corresponds to a 30%
excavation-induced loads and displacements, the authors have reduction in shear strength at the interface to take into account
chosen to use the finite-element software Plaxis 2D to carry out the impact of the construction methodology of the diaphragm
the numerical modelling required to support this paper. wall onto the concrete–ground interface friction. This is in
line with the values recommended by Gaba et al. (2003).
To facilitate the comparison between models and the
interpretation of the results, the authors have tried to keep the The characteristic geotechnical parameters for the soil/rock
analysis as straightforward as possible. A set of assumptions materials are presented in Table 1.
common to all models was defined to ease comparisons of results
and enable drawing meaningful conclusions from their Excavation models
interpretation. The assumptions and general conditions made in For each of the two different ground conditions considered, two
the model are the following. excavation depths were analysed to represent a range of typical
temporary work scenarios encountered in the industry. This gave
■ The retaining wall was chosen as an 800 mm thick diaphragm a total of four models, for all of which an excavation footprint of
wall, with the properties presented in Table 2. 30 m width by 60 m length was assumed. The first geometry
■ The elasto-plastic Mohr–Coulomb constitutive model was corresponds to (relatively shallow) excavation works for the
used for the soil and rock materials. construction of a single basement level, with a final formation
■ All the ground materials are considered as drained apart from level located 8 m below ground level. A propped cantilever wall
London Clay, which was assumed to be undrained. This was used in this case. The second situation represents the deep-
choice was made in the context of the short-term construction excavation works required to construct multiple basement levels,
displacements estimated in the models. with a final formation level located at 15 and 20 m below ground
■ The models were analysed using a two-dimensional plane level for London and Dubai ground conditions, respectively. For
strain symmetry. these deep excavations, three levels of propping were adopted.
■ All analyses were carried out with unfactored geotechnical
parameters to obtain models that behave closely to real A diaphragm wall was adopted for all cases, with the parameters
situations. Where required, partial factors were applied to the presented in Table 2. In accordance with Gaba et al. (2003), the
analyses’ output in order to obtain ultimate limit state values bending stiffness for the diaphragm wall is reduced to 70% of its
and support the structural calculations. theoretical value to take into account the crack width that
■ The shoring systems presented in this paper were analysed to developed during the relatively short construction period. For the
resist full hydrostatic pressure without the benefit of pore temporary props, circular hollow section (CHS) steel members
water pressure redistribution caused by dewatering. were adopted for the analyses, comparing typical steel sections
■ The staged construction sequence was modelled to match against hydraulic proprietary props. The different models were
realistic site conditions – namely, the following. computed iteratively, varying the prop type, spacing and pre-load
levels or, in other words, altering the prop’s stiffness value (per m
■ The groundwater level inside the excavation was set to be run of model) until the target deflection criterion was met. This is
1 m below the formation level for every intermediate discussed in detail in the methodology.
construction stage.
■ A height of 1·0 m for the working space was left below each The position of props supporting the wall and the required wall
prop centre line, before their ‘installation’ in the models. embedment depth to achieve sufficient stability were pre-determined
by way of hand calculations, before running the finite-element
■ Construction surcharges were modelled as uniformly models. The hand calculations were carried out following Rankine’s
distributed, with a magnitude of 20 kPa, acting at ground earth pressure theory to pre-determine the earth pressures acting on
level with a 1 m offset from the edge of excavation. the retaining wall and in accordance with the methods proposed by
■ A reduced friction was assumed at the interface between the Burland et al. (1981) to calculate the overall stability of the walls
retaining wall and the ground behind it to account for against overturning. Both prop levels and toe embedment were
subsequently refined and optimised iteratively in the finite-element
model to ensure a reasonable distribution of bending moments and
Table 2. Summary of diaphragm wall parameters shear forces throughout the depth of the wall while maintaining a
Parameter Symbol Unit Value buildable prop layout. Figures 4 and 5 illustrate the scenarios
Wall thickness t mm 800 analysed for Dubai and London, respectively. In each case, a propped
Young’s modulus Econcrete GPa 32 cantilever and a multipropped geometry are presented.
Axial stiffness (EA)wall kN/m 25·6 × 106
Bending stiffness (EI)wall (kN/m2)/m 955·7 × 103 The construction sequence stages for these models are provided in
Self-weight wwall (kN/m)/m 19·2
Table 3.

182
Downloaded by [] on [25/07/18]. Published with permission by the ICE under the CC-BY license
 Geotechnical Research Benefits of pre-loading temporary props
Volume 4 Issue GR3 for braced excavations
Latapie, Winarta and Lloyd

Table 3. Construction sequence stages

Model/stage PCD PCL MPD MPL
1 Excavate to +3 m DMD Excavate to +14 m OD Excavate to +3 m DMD Excavate to +14 m OD
2 Excavate to 1 m below S1 Excavate to 1 m below S1 Excavate to 1 m below S1 Excavate to 1 m below S1
at +1 m DMD at +12 m OD at +1 m DMD at +11·5 m OD
3 Install S1 at +2 m DMD Install S1 at +13 m OD Install S1 at +2 m DMD Install S1 at +12·5 m OD
4 Excavate to −1 m DMD Excavate to +10 m OD Excavate to −2 m DMD Excavate to +10 m OD
5 Excavate to FEL at −3 m DMD Excavate to FEL at +8 m OD Excavate to 1 m below S2 Excavate to 1 m below S2
at −5 m DMD at +7·5 m OD
6 — — Install S2 at −4 m DMD Install S2 at +8·5 m OD
7 — — Excavate to −8 m DMD Excavate to 1 m below S3
at +3·5 m OD
8 — — Excavate to 1 m below S3 Install S3 at +4·5 m OD
at −10 m DMD
9 — — Install S3 at −9 m DMD Excavate to FEL at +1 m OD
10 — — Excavate to FEL at −15 m DMD —

FEL, final excavation level

Methodology stiffness of the props and the small amount of pre-load. An iterative
In accordance with most international standards, the design of procedure followed where calibrated pre-load values were adjusted
structural members supporting temporary excavations needs to and set to meet the same performance criteria as the conventional
include at least two verifications: serviceability and ultimate limit non-pre-loaded CHS steel props. The horizontal spacings of the
states. Those two limit states are commonly translated to an props in all the analyses were kept identical so that the output data
allowable deflection and structural capacity, respectively. It is from the numerical analyses are directly comparable.
quite common for temporary works, particularly in urban
environments, to be designed to achieve a specified performance A flow chart depicting the analysis is shown in Figure 6.
criterion or, in other words, not to exceed the maximum allowable
wall deflection. This is to minimise the ground-movement- Determination of the structural resistance of
induced damage to existing structures and services adjacent to the tubular hydraulic and non-hydraulic props
site. The criteria used in this paper were set to follow the same Determination of the structural resistance to the axial loading of
philosophy where the maximum wall deflections were limited to the non-hydraulic steel CHS props was carried out by standard
30 and 15 mm for the Dubai and London models, respectively. methodology in accordance with BS EN 1993-1-1:2005 (BSI,
These criteria are based on the authors’ experience working in 2005) and in the case of class 4 (slender) sections by using the
both cities. The limit for London is more stringent, mainly approach outlined by Groundforce Shorco (2015).
because the city is more congested than Dubai and the majority of
its existing infrastructure is very sensitive to ground movements. The resistance of a hydraulic prop was calculated as being the
lesser of (a) its hydraulic components and (b) its resistance to
Using the deflection criterion defined earlier, a staged analysis was buckling as governed by the constituent CHS section capacity.
carried out for all four models, using conventional steel propping and Typically, it is the hydraulic ram capacity that limits overall axial
then pre-loaded hydraulic props to achieve the same performance resistance on short to medium prop lengths and resistance to
criteria (i.e. the same maximum wall deflection). The determination buckling dictates thereafter. This transition is illustrated in the
of the structural resistance of the selected props is discussed in detail typical prop resistance chart shown in Figure 7.
in the next section.
The axial capacities of the conventional steel tubes and the proprietary
First, the model was analysed with conventional non-pre-loaded CHS hydraulic props used in this paper, for a 30 m wide excavation, are
steel props, installed at a horizontal spacing chosen to achieve the summarised in Table 4. Where buckling is the governing factor, steel
specified performance criteria. The size of the CHS member and its and hydraulic props that use an identical CHS section have the same
spacing was selected based on the governing factor of either the structural capacity. The resistance of a hydraulic ram is calculated
structural capacity of the member or the deflection limit set for the based on its cylindrical diameter, length, wall thickness and material,
retaining wall. Subsequently, the same model was analysed with as well as the pressure rating of seals, fixtures and other components.
hydraulic props using their nominal pre-load. This usually leads to a Once the theoretical resistance has been estimated, it is then verified
higher wall deflection than the first analysis, due to the lower by factory testing before a final value is allocated to a specific ram

183
Downloaded by [] on [25/07/18]. Published with permission by the ICE under the CC-BY license
 Geotechnical Research Benefits of pre-loading temporary props
Volume 4 Issue GR3 for braced excavations
Latapie, Winarta and Lloyd

Determination of the axial stiffness of

Run models with conventional
CHS steel props and no pre-load
Amend CHS selection hydraulic and non-hydraulic props
Equation 1 is derived from Hooke’s law and calculates the axial
stiffness of a propping system per unit length of wall, comprising
No multiple struts of initial length L0 acting at an angle a from the
Are the performance criteria met? normal of the plane of the wall and installed at a lateral spacing, d

Yes
EAðcos2 a Þ
Is the structural capacity of the No S¼
1. L0 d
props adequate?

Yes Note: (a) a = 0° for props perpendicular to the supporting wall;

therefore, cos2 a = 1. (b) The effect on overall system stiffness
Run models with hydraulic prop Amend pre-load selection
and calibrated pre-load due to the flexural stiffness of the supported elements (e.g.
(start with nominal pre-load) diaphragm walls, capping or intermediate waling beams) is not
accounted for in Equation 1. It is the authors’ experience that the
flexural stiffness of the wall is generally ignored as it does not
No have a significant effect on the results.
Are the performance criteria met?

Yes Table 4. Typical characteristic geotechnical force on prop type

Is the structural capacity of the No Characteristic
props adequate? geotechnical force,
applied axially,
Prop type Diameter × wall thickness
which will give rise to
Yes 100% structural
Compare the results of all utilisation: kN
analyses and conclude Steel CHS 711 × 12·5 mm 1308
813 × 14·2 mm 2102
1016 × 12·5 mm 2541
Figure 6. Flow chart depicting the numerical modelling analyses
1016 × 14·2 mm 3141
1220 × 20 mm 7174
Hydraulic Groundforce MP150 1748
and the hydraulic prop is produced. The resistance of hydraulic rams Groundforce MP250 2503
Groundforce MP450 4505
is independent of the prop length.

Prop resistance governed by Hydraulic ram

Prop resistance governed by
buckling capacity Buckling
hydraulic ram capacity
Prop overall
Prop axial resistance: kN

Available hydraulic props

Prop length: m

Figure 7. Typical prop resistance

184
Downloaded by [] on [25/07/18]. Published with permission by the ICE under the CC-BY license
 Geotechnical Research Benefits of pre-loading temporary props
Volume 4 Issue GR3 for braced excavations
Latapie, Winarta and Lloyd

The axial stiffness of a single prop is calculated by using the Equation 3 to determine the resultant stiffness of the composite
equation hydraulic prop as shown in the equation

EA 1
S0 ¼ S0 ¼  

2. L0 L0,hyd = Ehyd Ahyd þ L0,steel = Esteel Asteel
4.

Equation 1 is valid for a steel prop modelled as a uniform

element. This equation is used to determine the stiffness of the By substituting this combined stiffness value S0 back into
steel CHS props used in this paper. However, proprietary Equation 1, the stiffness (per metre run) of a series of hydraulic
hydraulic props, as shown in Figure 3, consist of two elements, props installed at spacing d can be calculated from the equation
each of different axial stiffness. To take this factor into account, S0
S¼
the following approach is adopted in order to calculate the overall 5. d
axial stiffness of these prop types.

Equation 3 is also derived from Hooke’s law and states that for The parameters of length, area and Young’s modulus for the
elements connected end to end (in series), the reciprocal of the hydraulic ram elements of the proprietary props used in this paper
resultant stiffness is the sum of the reciprocals of all of the are stated in Table 5. An appropriate specification of hydraulic
connecting elements that make up the series ram is selected to meet the strength requirements stated in
Table 4.

1 1 1 1 It is the length of the column of hydraulic fluid within the ram,

¼ þ þ ⋯þ
3. S0 S1 S2 SN after the strut has been pumped out during the installation
process, that is used to calculate axial stiffness. For example, an
MP250 prop installed at nominal 30 m length requires the
The stiffness of each individual element can be calculated from hydraulic ram to be pumped out by 0·9 m. The calculated prop
Equation 2 by using appropriate parameters and substituted into stiffness values are summarised in Table 7.

Table 5. Characteristics of hydraulic ram

Full bore Young’s Max stroke of Ram Nominal Standard Maximum
Prop Hydraulic ram
area: modulus: hydraulic ram: housing minimum pre- achievable pre- achievable pre-
type type
mm2 N/mm2 mm length: mm load: kN load: kN load: kN
CHyd Groundforce MP150 31 416 1600 1150 3550 325 866 1748
AHyd
BHyd Groundforce MP250 49 087 1600 1066 3100 505 1355 2503
DHyd
EHyd Groundforce MP450 96 212 1600 1050 3100 910 2435 4505

Table 6. Summary of the temporary prop types used in the analysis

Maximum characteristic axial Strut stiffness:
Prop type CHS member Spacing: m
force per metre: kN/m kN/m
A CHS 1016 × 12·5 mm 12 212 689 700
AHyd CHS 1016 × 12·5 mm c/wMP250 hydraulic ram unit 12 209 166 935
B CHS 813 × 14·2 mm 12 175 623 550
BHyd CHS 813 × 14·2 mm c/w MP250 hydraulic ram unit 12 175 162 885
C CHS 711 × 12·5 mm 5 262 1 152 000
CHyd CHS 711 × 12·5 mm c/w MP150 hydraulic ram unit 5 262 250 590
D CHS 1016 × 12·5 mm 5 508 1 655 100
DHyd CHS 1016 × 12·5 mm c/w MP250 hydraulic ram unit 5 501 400 650
E CHS 1016 × 14·2 mm 5 628 1 876 950
EHyd CHS 1016 × 14·2 mm c/w MP450 hydraulic ram unit 5 628 670 605
F CHS 1220 × 20 mm 5 1435 3 166 800
G CHS 1220 × 20 mm 2 3587 7 916 850
H CHS 1220 × 20 mm 3 2391 5 277 900

c/w, combined with

185
Downloaded by [] on [25/07/18]. Published with permission by the ICE under the CC-BY license
 Geotechnical Research Benefits of pre-loading temporary props
Volume 4 Issue GR3 for braced excavations
Latapie, Winarta and Lloyd

Table 7. Summary of results for PCD, PCL and MPD

Working load
Prop Prop Pre-load applied as a percentage of the Strut stiffness: Maximum wall
Analysis output: kN/m
level type working load: % kN/m deflection: mm
[utilisation: %]
PCD S1 A 203 [97] N/A 689 700 14
PCD S1 AHyd 198 [95] 57 166 935 14
PCL S1 B 164 [94] N/A 623 550 10
PCL S1 BHyd 163 [93] 49 162 885 10
MPD S1 C 257 [98] N/A 1 152 000 15
MPD S1 CHyd 242 [93] 58 250 590 15
MPD S2 D 475 [93] N/A 1 655 100 15
MPD S2 DHyd 458 [91] 59 400 650 15
MPD S3 E 626 [99] N/A 1 876 950 15
MPD S3 EHyd 628 [99] 48 670 605 15

N/A, not applicable

Summary of props used in the analysis Results

Based on the methodology flow chart presented in Figure 6, the The analyses have shown that the design of the temporary props is
authors have selected suitable temporary props and the governed either by the structural capacity of the prop (for the PCD,
typical spacing required to satisfy both the deflection limit of the PCL and MPD cases), as presented in Table 7, or the deflection limit
wall and the structural capacity of the prop, which are presented of the retaining wall (MPL case), as presented in Table 8.
in Table 6. The maximum characteristic axial resistance values
stated are based on either the structural capacity of the steel prop The output working loads for the steel and hydraulic prop of the
or the capacity of the hydraulic unit. Steel props are denoted with same model at the same level are very similar, with no more than 6%
a capital letter (i.e. A), and hydraulic props are denoted as AHyd. difference. The analysis results presented in Table 7 show that all
temporary props (steel and hydraulic) are highly utilised (over 90%
Values of nominal pre-load and achievable utilisation ratio) with wall deflections well below the limiting criteria.
pre-load This means that the choice of the props for these scenarios is
Two types of props were proposed for investigation in this paper, governed by the structural (buckling) capacity of the steel section,
namely which does not enable material savings when replacing the
conventional steel props with the proprietary hydraulic props.
■ conventional steel props
■ hydraulic props with calibrated pre-load to achieve the The axial stiffness of the conventional steel props was reduced by
performance of conventional steel prop in terms of wall 64 and 78% for MPD prop levels S3 and S1, respectively, when
deflections. replaced with hydraulic props. This is due to the incorporation of
the hydraulic unit within the prop assembly, which has a lower
The values for nominal minimum and maximum achievable pre- stiffness than the equivalent length of CHS. The benefit of
loads are outlined in Table 5 for each prop type used. Standard reduction in prop stiffness in the context of temperature effects is
achievable values of pre-loads may be reached with standard further discussed in the section headed ‘Benefit of lower prop
hydraulic pumping equipment able to apply 4000 pounds per stiffness for thermal effects’.
square inch (psi) (25·6 MPa). The hydraulic ram can, however, be
pre-loaded up to its capacity (or maximum achievable pre-load) The results in Table 7 confirm that the performance (i.e.
by using special hydraulic equipment tailored to the target pre- maximum wall deflection) of conventional steel props can be
load and able to apply sufficiently high pressures. replicated with the use of hydraulic props with calibrated pre-

Table 8. MPL – comparison between cases for wall deflection and working load
Working load
Prop Prop Pre-load applied as a percentage of the Strut stiffness: Maximum wall
output: kN/m
level type working load: % kN/m deflection: mm
[utilisation: %]
S1 F 190 [13] N/A 3 166 800 15
S2 G 500 [14] N/A 7 916 850 15
S3 H 421 [18] N/A 5 277 900 15
S1 CHyd 199 [76] 60 250 590 15
S2 DHyd 460 [91] 74 400 650 15
S3 DHyd 426 [85] 75 400 650 15

N/A, not applicable

186
Downloaded by [] on [25/07/18]. Published with permission by the ICE under the CC-BY license
 Geotechnical Research Benefits of pre-loading temporary props
Volume 4 Issue GR3 for braced excavations
Latapie, Winarta and Lloyd

loads. This proves that the application of a pre-load enables a pre-loads were introduced to all three levels of the proprietary
lower-stiffness proprietary hydraulic prop to match the hydraulic prop to satisfy the 15 mm limit criterion. Prop sizes
performance (or stiffness) of a conventional high-stiffness steel were reduced to CHS 711 × 12·5 mm combined with a MP150
prop. This principle is illustrated by the equation hydraulic unit for the S1 level and CHS 1016 × 12·5 mm
  combined with a MP250 hydraulic unit for the S2 and S3 levels.
Psteel  L0 Phyd − Ppre-load In comparison with the conventional steel prop, the utilisation
¼ − dpre-load
6. Esteel  Asteel S0 ratio of the structural capacity was increased to between 76 and
91%. This result indicates that the hydraulic proprietary props are
working more efficiently compared to the conventional steel prop.
Psteel and Phyd are the working loads for the conventional steel As shown in Table 9, the use of hydraulic props for the MPL case
and hydraulic props, respectively. These are almost identical for enabled nearly 74% savings in steel quantities.
the cases analysed and presented in this paper. Ppre-load is the pre-
load applied to the proprietary hydraulic prop and dpre-load is the As presented in Table 9, in the MPL case, the axial stiffness of
positive value of the net horizontal deflection of the retaining the conventional steel props was reduced by 92% (prop levels S1
wall, towards the retained side and at the prop’s location, induced and S3) to 95% (prop level S2) when replaced with hydraulic
by applying Ppre-load. After interrogating the analyses outputs, a props, which is due to the incorporation of the hydraulic unit
range of 2·3–4·5 mm was found for dpre-load. Its value is normally within the prop assembly.
higher for the top prop and reduces for deeper prop levels. This is
because this parameter depends on the combined stiffness of the In Table 8, the pre-load applied to DHyd (for S2 and S3) exceeds
retaining wall and the ground behind it in response to the applied the standard achievable pre-load (see Table 5). The pre-load
pre-load. One may calculate Ppre-load by using Equation 6 with values presented may be achieved by using specialist pumping
these approximate values of dpre-load to obtain an initial estimate equipment with hydraulic pressure capacity of 5000 psi
and accelerate the design procedure. (34·5 MPa), which is within the maximum achievable pre-load of
the hydraulic ram.
The authors’ analyses have found that the pre-load required to
match the performance of the conventional steel prop ranges The reduction in prop sizes, combined with increased spacing,
between 48 and 59% of the prop’s working load, which are within indicates a greater efficiency, and economic benefit can be
the standard achievable pre-load defined in Table 5. brought to the design. Based on the analysis, application of pre-
loaded hydraulic proprietary props for deep excavations with
The results of the authors’ analyses of the temporary props in stringent criteria would be beneficial in terms of economy and
multipropped ‘London ground conditions’ showed that their design structural efficiency of the design.
is governed by the 15 mm deflection limit of the retaining wall
rather than their structural capacity. For the purpose of the paper, a Correlation between pre-load force and the
very narrow prop spacing was introduced for conventional steel performance of the retaining wall
prop analysis for S2 and S3 (CHS 1220 × 20 mm with 2 and 3 m Some research has been carried out into the correlation between
spacings, respectively) to satisfy the deflection limit. As prop pre-loading and the design requirements of the supported
summarised in Table 8, the results of this analysis indicate very retaining wall. It is apparent that the advantage of reduced
low utilisation ratios, between 13 and 18%. This means that excavation-induced ground settlements by using prop pre-loads is
choosing conventional steel props in such a situation would lead not universally accepted within the temporary work community.
to inefficient and possibly uneconomical design. This is because prop pre-load has the potential to increase
unnecessarily the forces within the retaining system without
In contrast, in the analysis of hydraulic props under the same offering a significant benefit. If the wall is prevented from
ground conditions, a typical spacing of 5 m was used as less prop moving, then a situation could arise where an at-rest earth
stiffness was required to achieve the deflection criterion. Instead, pressure larger than active earth pressures is mobilised (Bond and

Table 9. Material and stiffness savings brought by the use of hydraulic props
Reduction in steel quantities
Reduction in prop stiffness per
City Excavation type Steel prop Hydraulic prop
metre run:a % Difference: %
weight: kg/m weight: kg/m
Dubai Propped cantilever 76 Steel quantities are equivalent as prop design is governed by
Dubai Multipropped 64–78 structural capacity and the required CHS is identical for the
London Propped cantilever 74 conventional and hydraulic props
London Multipropped 92–95 18 348 4854 74
a
The benefits of reduced prop stiffnesses on their design are discussed in the section headed ‘Benefit of lower prop stiffness for thermal effects’

187
Downloaded by [] on [25/07/18]. Published with permission by the ICE under the CC-BY license
 Geotechnical Research Benefits of pre-loading temporary props
Volume 4 Issue GR3 for braced excavations
Latapie, Winarta and Lloyd

Harris, 2008). This may happen when using excessive pre-loads, Bending moment Shear force
and in extreme cases, if the magnitude of pre-loading introduced 80
to the prop exceeds the at-rest pressures of the surrounding
ground, the wall may be ‘pushed back’, attracting even higher

Pre-load/working load: %
ground loads than at-rest pressures. 70

Using values of pre-load in excess of what is required to maintain

stability and minimise ground movements has the potential to 60
induce unnecessary shear stresses and bending moments into the
retaining wall. Such stress increases would in turn increase the
retaining wall’s reinforcement and reduce the cost-effectiveness of
50
the overall system.

Conversely, any reduction in prop stiffness will encourage

40
greater movement as the dig starts, leading to greater activity 0 1 2 3
and reducing stresses. So in the case of pre-loaded hydraulic Change in bending moment and shear force: %
props, as demonstrated by the analysis results presented in this
paper, the effects of reduced prop stiffness and calibrated pre- Figure 9. Comparison between ratio of applied pre-load to working
load and absolute change of bending moment and shear force
load force can cancel each other out as long as the pre-load is
carefully selected.

This is illustrated in Figures 8 and 9, which show that the props to hydraulic props with pre-load, is negligible (<3%) for all
calculated change in the maximum bending moment or shear the scenarios analysed and presented in this paper. This is also
force induced into the retaining wall, from conventional steel illustrated in Figure 10, which shows that the earth pressures

Conventional steel prop

Pre-loaded hydraulic prop

15 15

10 10
Level: m OD

Level: m OD

5 5

0 0

Prop Prop
−5 −5
−450 −250 −50 150 350 −350 −150 50 250
Bending moment/m run: kN m/m Shear force/m run: kN/m
(a) (b)

Figure 8. Comparison between (a) bending moment and (b) shear force envelope for MPL with conventional steel props and pre-loaded
hydraulic props

188
Downloaded by [] on [25/07/18]. Published with permission by the ICE under the CC-BY license
 Geotechnical Research Benefits of pre-loading temporary props
Volume 4 Issue GR3 for braced excavations
Latapie, Winarta and Lloyd

15
Conventional steel prop
Made ground
Pre-loaded hydraulic prop
6m

Groundwater level
10

River terrace deposits

5m

Level: m OD
5

London Clay 0
>30 m

Prop
−5
0 −100 −200 −300
Mobilised lateral earth pressure: kN/m2

Figure 10. Plots of lateral earth pressures applied to the retaining wall for MPL

applied to the wall in the model using steel props match very well Benefit of lower prop stiffness for thermal
the profile from the pre-loaded hydraulic props. effects
As a prop heats up due to a change in ambient temperature, it will
The results support the authors’ opinion that the use of pre-load in attempt to expand. Being restrained at each end by the waling beams
props, if carefully selected, can be a substitute for stiffness without and the retaining wall, the load in the prop will increase. The
having a detrimental impact on the design of the retaining wall. It is magnitude of this load will depend on the relative stiffness of the
also clear that in order for efficient implementation of pre-load to prop and the retaining wall/waling stiffness. The load variation due to
be possible, the props’ stiffness must also be carefully selected such temperature change is generally expressed by the following equation
that the overall support system is not made unnecessarily stiff.
b
When designing a retaining system, it is therefore important to DPtemp ¼ at  Dt  E  A 
7. 100
ensure that the pre-load values and prop stiffnesses selected are
not too high and compatible, to prevent overstressing the wall
unnecessarily. The results presented in this paper suggest that pre- The term b in Equation 7 is often termed the ‘degree of restraint’
loads of 50–75% of the prop working load will achieve the and represents the overall stiffness of the retaining wall and
required deflection criterion without increasing the structural loads waling system. Recommended b values vary between 40 and 70
or forces on the retaining wall, compared to the conventional steel for stiff walls in stiff ground condition and between 10 and 40 for
prop solution, without pre-load. flexible walls (Twine and Roscoe, 1999).

Table 10. Material and stiffness savings brought by the use of hydraulic props
City Excavation type Prop type Spacing: m E × A: kN/m `t: °C−1 Dt: °C a DPtemp: kN/m
A 689 700 166
Dubai Propped cantilever 12 1·2 × 10−5 40 0·5
AHyd 166 935 40

Table 11. Material and stiffness savings brought by using hydraulic props
City Excavation type Prop type Spacing: m E × A: kN/m Ground load: kN/m DPtemp: kN/m Total working load: kN/m
E 4 2 346 188 214 563 777
Dubai Propped cantilever
AHyd 10 200 322 202 48 250

189
Downloaded by [] on [25/07/18]. Published with permission by the ICE under the CC-BY license
 Geotechnical Research Benefits of pre-loading temporary props
Volume 4 Issue GR3 for braced excavations
Latapie, Winarta and Lloyd

Equation 7 shows that the load increase due to temperature In both cases, the additional ground loads attracted by the higher
is proportional to the prop’s stiffness. This means that the support stiffness (per linear metre of excavation) were less than
stiffer the prop, the more thermally induced load it will attract. 4% and hence negligible.
One of the benefits of using pre-loaded low-stiffness hydraulic
props is that they will attract less thermally induced load than a Conclusions and further studies
stiffer steel prop of a comparable section size. This is an The use of a hydraulic unit as part of a prop enables a reduction
important consideration in the Middle East region, where design in the overall axial stiffness of the prop. In all cases analysed and
temperature gradients vary from 40 to 60°C with thermally presented in this paper, such stiffness reduction could be replaced
induced load being a major factor in the design of temporary by a carefully chosen pre-load. Using soil profiles typical of
propping. excavations in London and Dubai, finite-element analyses have
been carried out to test the performance of steel and proprietary
To demonstrate this, Equation 7 has been applied to the results of hydraulic temporary propping.
the Dubai propped cantilever scenario, with the input parameters
summarised in Table 10. This shows that the conventional steel Numerical modelling indicates that applying a carefully chosen
prop, due to its higher stiffness, would attract over four times pre-load allows low-stiffness hydraulic props to match the
more temperature-induced load than its proprietary hydraulic performance of high-stiffness steel props, in terms of both wall
counterpart. All prop information in Tables 10 and 11 is presented deflection and induced structural shear forces and bending
per linear metre of retained wall in order for the results to be moments. Results for the two kinds of props considered were
directly comparable. remarkably close to each other, which is encouraging in promoting
the use of controlled pre-load with lower-stiffness props to provide
As both struts used in the PCD scenario resulted in high an economic temporary work solution. Based on the results of the
utilisation ratios, they are not able to accommodate the high analyses presented in this paper, pre-loads of 50–75% of the prop
temperature loads typically designed for in Dubai. To determine working load achieved the required performance.
the impact of such temperature loads onto the prop design,
iterative calculations were carried out to find the additional For temporary props that are governed by structural (buckling)
material quantities required to accommodate those loads. This is a capacity, pre-loads have been shown to have small effects on the
tedious exercise because with the temperature loads being steel section. In those cases, the main benefit of using pre-loaded
proportional to the prop’s stiffness, the larger the prop is the more proprietary hydraulic props is to reduce the prop stiffness. In
load that it attracts. However, this trend is a lot gentler for situations where props are not utilised fully in their structural
hydraulic props than steel props, as they are less stiff. Prop capacity, but governed by the deflection criterion of the retaining
sections and spacing were amended to find the most efficient wall, the overall stiffness of the propping system controls the
arrangements which accommodate both the ground and design. Hence, using pre-loaded proprietary hydraulic props
temperature loads. Finite-element analyses were also recalculated allows economy on the prop steel section size and spacing as well
to ensure that the prop forces were correctly estimated as higher as a reduction in the support stiffness.
support stiffnesses tend to attract higher ground loads. Outputs of
this analysis are summarised in Table 11. The effect of thermally induced loads was discussed and estimated
for one of the scenarios analysed. The use of low-stiffness props has
The conventional steel prop section had to be increased to a type been put forward to minimise thermal loads. It was demonstrated that
E with the spacing reduced from 12 to 4 m. The additional steel using proprietary hydraulic props helps reduce the effect of thermal
section required to accommodate temperature effects led to an loads on the design of the overall retaining system. This advantage
increase in temperature loads per linear metre of excavation of was found to have a stronger bearing on the design of support
340%. This means that 72% of the design load comes from systems controlled by deflection. For locations subject to large
temperature effects rather than ground loads. The reduced prop temperature differentials, like Dubai, temperature-induced loads can
spacing provides a congested arrangement for a situation where govern the design of conventional steel props. Conversely, for
ground loads are relatively small (compared to the bottom of a proprietary hydraulic props, thermal loads represent a relatively small
multipropped excavation). proportion of the prop’s total design load.

In the case of the hydraulic prop, temperature loads could be It was proven that applying pre-load by way of a hydraulic prop
accommodated by keeping the prop type with only a small can have a negligible effect on the earth pressures, bending
adjustment of the prop spacing, down to 10 m. This means that moments and shear forces resisted by the retaining wall as long as
the temperature loads per linear metre of excavation increased the pre-load is carefully selected.
only by 20% in response to this additional support stiffness.
Consequently, only 19% of the design load is attributed to Further work is suggested to analyse numerically the empirical
temperature effect, allowing a support arrangement representative stiffness reduction factor b for different wall types to evaluate better
of the magnitude of the ground loads to be maintained. the thermally induced loading in the props. This is important

190
Downloaded by [] on [25/07/18]. Published with permission by the ICE under the CC-BY license
 Geotechnical Research Benefits of pre-loading temporary props
Volume 4 Issue GR3 for braced excavations
Latapie, Winarta and Lloyd

because the factor b has a large impact on the proportion of Chambers P, Augarde C, Reed S and Dobbins A (2016) Temporary
thermally induced loads a propping system ought to be designed propping at Crossrail Paddington station. Geotechnical Research 3(1):
3–16, http://dx.doi.org/10.1680/jgere.15.00009.
with. This would help provide a more scientific derivation of the
Gaba AR, Simpson B, Powrie W and Beadman DR (2003) C580:
stiffness reduction factor b, which is currently arbitrarily selected Embedded Retaining Walls – Guidance for Economic Design.
by the designer from historic observational data. Construction Industry Research and Information Association,
London, UK.
REFERENCES Groundforce Shorco (2015) Technical Note 7: Design Guidance on Class
Bond A and Harris A (2008) Decoding Eurocode 7. Taylor & Francis, 4 Circular Hollow Sections. Groundforce Shorco, Leeds, UK. See
London, UK. https://www.vpgroundforce.com/Groundforce/media/Groundforce/
BSI (2005) BS EN 1993-1-1:2005: Eurocode 3: design of steel structures. Secured/Imported/Technical-Note-7-Guidance-on-class-4-CHS.pdf
General rules and rules for buildings. BSI, London, UK. (accessed 15/06/2017).
Burland JB, Potts DM and Walsh NM (1981) The overall stability of free Twine D and Roscoe H (1999) C517: Temporary Propping of Deep
and propped embedded cantilever retaining walls. Ground Engineering Excavations – Guidance on Design. Construction Industry Research
14(5): 28–38. and Information Association, London, UK.

How can you contribute?

To discuss this paper, please submit up to 500 words to
the editor at [email protected]. Your contribution will be
forwarded to the author(s) for a reply and, if considered
appropriate by the editorial board, it will be published as a
discussion in a future issue of the journal.

191
Downloaded by [] on [25/07/18]. Published with permission by the ICE under the CC-BY license