You are on page 1of 10

Rory van Doorn et al.

1




Bow Thruster Currents at Open Quay Constructions
on Piles


Rory van Doorn Dockwise, Breda, Netherlands
Teus Blokland Rotterdam, Engineering Department, Netherlands
Tiedo Vellinga Port of Rotterdam and Delft University of Technology, Netherlands
Henk Verheij Deltares and Delft University of Technology, Netherlands
Henk Jan Verhagen Delft University of Technology, Delft, Netherlands


Summary
The hydraulic loads on the slope at open quay constructions on piles is investigated. The propeller
(bow thruster) jet induces hydraulic loads on the slope, which could result in scour holes and damage
to the quay construction. Performed scale model tests provide details about the hydraulic loads in this
specific situation with an inclining slope and piles. Measurements are compared to calculations with
the equation for an unconfined propeller jet, to include in the design of open quay constructions on
piles.
Introduction
In many ports berths consist of a slope covered with a deck on piles. Usually the slope is protected
with some armourstone. Because of an increase in the use of powerful bow thrusters, the load on the
slope protection increases, which may require heavier armour. However, replacing the armour below a
deck is complicated and consequently very costly.


Bow thrusters at open quay constructions on piles have not been investigated yet, and available
engineering guidelines are based on free propeller jets without piles on a slope (SCHOKKING [2002],
RMISCH [2006]). In this research scale model tests were performed to provide details about the
hydraulic loads in such situations and as result, recommendations are provided for engineering
purposes. This research focuses on the hydraulic loads due to bow thrusters.

Figure 1. Schematic view of open quay construction
Rory van Doorn et al. 2
Hydraulic load
Concluding from the literature review, commonly the Dutch and German engineering guidelines are
used to predict hydraulic loads from propeller jets. PIANC [1997] contains engineering guidelines for
propeller jets and soon a new guideline on this subject is being published, PIANC [2013].
Dutch method
BLAAUW AND VAN DE KAA [1978] and VERHEIJ [1983] derived the following equations, for the
determination of the flow field behind a propeller.


1 3
0 2
0
1.15
w
P
U
D
| |
=
|
\ .
(1)
2
0
, 0 0
U 2.8 exp 15.4 for 2.8
x r
D r
n U x D
x x
| |
| | | |
= > |
| |
|
\ . \ .
\ .

(2)

Equation (2) is valid for situations in which the jet propagation is not restricted by any confinement like
a bottom, a slope or a water surface. If there is a confining surface, the actual velocities near this
surface can deviate from the equation for a unconfined jet.

In the situation of a propeller jet perpendicular to a slope, equation (2) results in the following
x-coordinate of the location of the maximum current velocity on the slope:

( ) o
| |
= + = |
|
\ .
max 2
2 15.4
1 1
cot
U
x K L with K
K
(3)

For this equation x=0 at the outflow opening of the propeller or thruster and x=L at the slope.

In the present study measured flow velocities on a slope (with or without piles) will be compared with
the flow velocities resulting from the equations (1) and (2) for a unconfined jet. The measured
locations of the maximum current will be compared with the result of equation (3).
Protection
Hydraulic loads by propeller jets and the resulting bed protection are commonly approached using an
Izbash type equation, taking into account mean flow velocities combined with a factor to count for
turbulence intensities. Equation (4) is commonly applied in case of propeller jets to determine required
stone diameters according VERHEIJ [1985].

2
,max
50 ,
2
b
Iz cr h
U
D m
g
| >
A
(4)
In (4) maximum mean flow velocities are included and the critical stability coefficient accounts for a
certain turbulence intensity. In the Dutch method |
Iz,cr
= 2.6 to 3.0 for propeller jets (CIRIA, CUR AND
CETMEF [2007]). The required D
50
for the present model tests are calculated using |
Iz,cr
= 3.0.
Influence of turbulence
As the turbulence intensity plays an important role in propeller jets, measuring of the turbulence
intensities was included in the set-up for the scale model tests.
By using the turbulence intensity as an input in the Izbash equation, the adopted = 3.0 is no longer
necessary. This coefficient is used as a safety factor to include the unknown turbulence intensity.
From the scale model tests the turbulence intensity is known as well. With this as an input, follows:

2
2
50
'
2
Iz
h
g D m u p u
|
| |
A = +
|
\ .

(5)
Commonly (SCHIERECK [2004], ROUBOS [2006]) a factor p = 3 is applied. Consequently a 0.13 %
probability of exceeding follows by the Normal distribution. For situations with normal turbulence it is
Rory van Doorn et al. 3
supposed that the relative turbulence intensity r is about 0.1, while | = 0.7. This results in |
Iz,0,p
= 0.41.
This value is used to calculate the required D50 using u & u.

( )
( )
2
2
2
,0,
2
,0,
2
,0,
'
with
1
0.7 1 3 0.1 0.41 0.41
Iz p
Iz p
Iz p
u p u u
p r
| |
| |
|
| |
= +
|
\ .
= +
= + =



Laboratory tests
Starting point for the tests was a 7,000 TEU container vessel, with a 2.75 metre bow thruster diameter.
This vessel and an open quay construction on piles (diameter 0.75 m, centre to centre distance 5 m)
were scaled using a length factor 25. In the Delft Universitys Laboratory of Fluid Mechanics, a basin
of 10 by 2 metres was used to test different scenarios. Ten scenarios were tested using scale model
tests (see Table 1). Each scenario was different by changing one of the following parameters: slope
angle, rough or smooth slope, bow thruster distance to the slope, with or without piles and in case of
piles a displacement parallel to the slope. Most of the tests were carried out with a slope 1:1.5 due to
space limitations.
Furthermore, mean velocities and fluctuations were measured in each scenario using an Acoustic
Doppler Velocimeter (ADV) in order to determine the hydraulic loads on the (non erodible) bed.
Thruster outflow velocity measurements were done, whereas the main part consisted of velocity
measurements near the bed. Using an ADV with a sample frequency of 25 Hz, turbulent velocity
fluctuations were measured, which are part of the hydraulic bed loads.
Findings
Difference between measurements and calculations
Table 1 contains for all scenarios the factor difference between the results from the measurements
and the results according to the Dutch formulas for an unconfined propeller jet and for stone stability
(the engineering guidelines according to the Dutch method). For example, a factor 1.15 means that a
certain parameter is 1.15 times larger in the measurements, compared to the results of the equations.
Table 1: Measurements compared to calculations
Factor difference
measurement/calculation
Scenario Slope Water
depth
L/D
0
Cover Pile
alignment
Mean
velocity
D
50

using u
D
50

using u &
u
S1 1:2.5 0.42 m 6.2 smooth no piles 1.11 1.23 1.17
S2 1:1.5 0.42 m 6.2 smooth no piles 1.07 1.14 2.02
S3 1:1.5 0.63 m 6.2 smooth no piles 1.19 1.43 1.39
S4 1:1.5 0.63 m 6.2 rough no piles 1.26 1.58 1.47
S5 1:1.5 0.63 m 4.0 rough no piles 1.16 1.36 1.07
S6 1:1.5 0.63 m 6.2 smooth axis at piles 1.41 1.98 1.83
S7 1:1.5 0.63 m 6.2 smooth axis mid
between piles
1.36 1.86 0.89
S8 1:1.5 0.63 m 6.2 rough axis mid
between piles
1.63 2.69 1.94
S9 1:1.5 0.63 m 6.2 rough axis quart
between piles
1.72 2.95 1.72
S10 1:1.5 0.63 m 6.2 rough axis at piles 1.58 2.49 2.02
Note: In the model tests D
0
= 0.11 m.
Rory van Doorn et al. 4


F
i
g
u
r
e

2
.

E
x
a
m
p
l
e

o
f

m
e
a
s
u
r
e
d

n
e
a
r

b
e
d

v
e
l
o
c
i
t
i
e
s

&

n
e
a
r

b
e
d

t
u
r
b
u
l
e
n
c
e

i
n
t
e
n
s
i
t
y

-

t
h
e

c
o
n
t
i
n
u
o
u
s

l
i
n
e

i
n

t
h
e

s
i
d
e

p
a
n
e
l
s

i
s

t
h
e

t
h
e
o
r
e
t
i
c
a
l

m
e
a
n

v
e
l
o
c
i
t
y

c
u
r
v
e

a
l
o
n
g

t
h
e

s
o
l
i
d

l
i
n
e
s

i
n

t
h
e

c
e
n
t
r
a
l

p
a
n
e
l

a
c
c
o
r
d
i
n
g

t
o

e
q
u
a
t
i
o
n
s

(
1
)

-

(
3
)

f
o
r

a

f
r
e
e

u
n
c
o
n
f
i
n
e
d

p
r
o
p
e
l
l
e
r

j
e
t
;

-

t
h
e

s
p
i
k
e
s

i
n

t
h
e

s
i
d
e

p
a
n
e
l
s

a
r
e

t
h
e

m
e
a
s
u
r
e
d

m
e
a
n

v
e
l
o
c
i
t
y

(
r
l
e
f
t
)

e
n

m
e
a
s
u
r
e
d

t
u
r
b
u
l
e
n
c
e

i
n
t
e
n
s
i
t
y

(
r
i
g
h
t
)

a
l
o
n
g

l
i
n
e

A


a
n
d

B


i
n

t
h
e

c
e
n
t
r
a
l

p
a
n
e
l
.

t
h
e

v
e
r
t
i
c
a
l

s
o
l
i
d

l
i
n
e

i
n

t
h
e

c
e
n
t
r
a
l

p
a
n
e
l
s

c
o
i
n
c
i
d
e
s

w
i
t
h

t
h
e

a
x
i
s

o
f

t
h
e

p
r
o
p
e
l
l
e
r

j
e
t
,

t
h
e

h
o
r
i
z
o
n
t
a
l

s
o
l
i
d

l
i
n
e

r
e
p
r
e
s
e
n
t
s

t
h
e

l
o
c
a
t
i
o
n

o
f

t
h
e

t
h
e
o
r
e
t
i
c
a
l

m
a
x
i
m
u
m

c
u
r
r
e
n
t

v
e
l
o
c
i
t
y

i
n

x
-
d
i
r
e
c
t
i
o
n

a
c
c
o
r
d
i
n
g

t
o

e
q
u
a
t
i
o
n

(
4
)
.

I
n

s
k
e
t
c
h

a
t

r
i
g
h
t

v
a
l
u
e
s

a
r
e

i
n

m
m
;

v
e
s
s
e
l

i
s

l
e
f
t
,

q
u
a
y

s
t
r
u
c
t
u
r
e

i
s

a
t

t
h
e

r
i
g
h
t

s
i
d
e
.



Rory van Doorn et al. 5
Remarks: it cannot be excluded that maximum bed loads in S1 and S2 are not measured due to un-
usable measurements close to the propeller and a limited number of measurements for these
scenarios. And in scenarios with a rough slope the X-coordinate of axis intersection with slope is 0,096
m different to scenarios with a smooth slope. For all scenarios the height of the propeller jet axis
above the horizontal bottom was the same, namely 0.244 m
Effects due to variations in parameters
The effect of each variation in parameter is provided in Table 2. Here the factor difference between
two scenarios is calculated for the mean velocities and two methods regarding the stone diameter.
The factor difference for D
50
using u is equal to the squared mean velocity factor, because the mean
velocity is used as an input for this D
50
.
Table 2: Effects due to parameter variations
Varied Parameter Value Scenario Mean
velocity
D
50

using u
D
50

using u & u
Slope angle 1:1.5 / 1:2.5 S2 / S1 0.97 0.93 1.73
Water depth 0.63 / 0.42 S3 / S2 1.12 1.25 0.68
Slope angle & water depth S3 / S1 1.08 1.16 1.18
x/D
0
4.0 / 6.2 S5 / S4 0.93 0.86 0.73
Roughness, without piles rough / smooth S4 / S3 1.05 1.11 1.06
Roughness, piles frontal rough / smooth S10 / S6 1.12 1.25 1.11
Roughness, piles mid rough / smooth S8 / S7 1.20 1.44 2.19
Piles, smooth
piles front /
without piles
S6 / S3 1.18 1.39 1.32
piles mid /
without piles
S7 / S3 1.14 1.30 0.64
Piles, rough
piles front /
without piles
S10 / S4 1.25 1.57 1.38
piles mid /
without piles
S8 / S4 1.30 1.69 1.32
piles quart /
without piles
S9 / S4 1.36 1.86 1.17

The most relevant parameters are discussed below.
Variation in slope angle
The factor difference S2/S1 for the mean velocity is 0,97, which is almost equal to 1. This makes it
reasonable that the model tests with further parameter variations using a relatively steep slope of 1:1.5
are also valid for less steep slopes. It however should be remembered that scenarios S1 and S2 have
only a limited number of valid measurements points, which makes the measured factor difference not
very accurate.
Changed water depth
According to equation (3) and (5), S3 requires a stone diameter equal to S2. The water level is raised
from 0.42 to 0.63 metre and measurements show an increase in the maximum load. This can partly be
a spurious effect because of limited measuring points. A physical reason for this could be the
increased branch off by the jet towards the slope. This phenomenon is described in WL [1988].
Consequently the jet hits the inclining slope at a smaller distance from the outflow opening, which
explains the increased slope load.
In addition, because of less measurement results in S2, variation in slope angle and water depth are
compared together. This shows in table 2 a difference factor of 1.08 for mean flow velocities, which is
a relatively small difference, showing the relevance of the performed scale model tests.
Distance to slope
With the outflow opening at a smaller distance to the slope, measurements show results fitting better
to the calculations. This holds for mean flow velocities as well as turbulence intensities.
Rory van Doorn et al. 6
In addition to the change in water depth, the shorter distance of the outflow opening to the slope could
result in less branching off by the jet, because of less distance to encounter branching off.
If branching off by the jet is underestimated using equation (3), this would explain the factor 0.93 for
the mean flow velocity. Due to the fact that the jet is branching off less, the distance between propeller
and the location where the jet hits the slope is larger, where a larger distance results in a larger
decrease in mean flow velocities.
Roughness
For scale model tests with a rough bed, all results show a larger factor difference compared to tests
with smooth slopes. Without piles the factor difference is 1.05 for mean flow velocities, whereas with
piles frontal placed, this increases to 1.12. Largest difference (1.20) is observed for situations with the
jet axis pointed mid between two rows of piles.
These results clearly show that a rough-bed has a significant influence on results from the scale model
tests. No variations in stone diameters (variations in roughness) have been tested.
Piles
In case of a situation with piles compared to a situation without piles, all difference factors are larger
than one, which means that these measurements provide larger hydraulic loads compared to
situations without piles. One value is clearly different; 0.64 for S7 compared to S3. Turbulence
intensities in S7 show less peaks compared to all other scenarios, causing this low value. An
explanation for this deviating result is not found.
Equation (3) for a unconfined jet results in relatively large underestimation factor for situations with
piles, especially for a rough surface.
Pile location
Within the scenarios for situations with piles the vessel and thereby the bow thruster is shifted in the
longitudinal direction, parallel to the slope. Hereby the outflow opening was tested frontal to a row of
piles, as well as mid between two rows of piles. The determining situation for maximum bed load
follows from the measurements. Maximum bed load, including turbulence intensity, occurs when the
bow thruster axis is in line with a row of piles.
Mean velocities for rough bed scenarios, however, are smaller with piles in front, compared to piles at
mid- and quart-position. In contradiction, smooth bed scenarios show larger mean velocities with piles
in front, compared to piles at mid-position. Further research is recommended to explain this
contradiction.
Hydraulic load without piles
VAN DOORN [2012] focuses on the hydraulic load for a situation with piles. From the scale model tests
follows that the current design method however, underestimates the mean flow velocities for an
inclining slope (without piles). Hereby some further details for an inclining slope without piles are
determined, which is the basis for an inclining slope with piles.
The location of maximum flow velocities is different compared to equation (4), which is not unexpected
because equation (4) is derived from equation (3) which describes the flow distribution for a
unconfined propeller jet.
The maximum bed load follows when diffusion of the bow thruster jet is assumed under an angle of
1:2.4 to 1:18.6. This location is for all ten scale model scenarios located at the inclining slope. With the
bow thruster at a larger distance from the inclining slope, also the toe of the bed protection could be
the location of maximum bed loads.
Maximum mean flow velocities are underestimated in all scenarios, using equation (3). In all situations
without piles, mean flow velocities are underestimated with a factor 1.07 to 1.26. Translated into the
hydraulic bed load, with the velocity to the power two, this is a factor 1.14 to 1.58. Analysis of
measurements taking into account the turbulence intensities, results in a hydraulic bed load that is
Rory van Doorn et al. 7
underestimated by a factor 1.07 to 2.02 for situations without piles, which reduces to 1.07 to 1.47 for
rough situations.
Hydraulic load with piles
An important aspect is the mean flow velocity near the slope. In combination with the turbulence
intensities this is determining, in the current design methods, for bed protections. Compared to
situations without piles, piles result in larger mean flow velocities and larger hydraulic slope loads.
Mean flow velocities for situations with piles are underestimated by a factor 1.36 to 1.72, using
equation (3). Translated into the hydraulic bed load, this is a factor 1.86 to 2.95 for the required D
50
.
Including turbulence intensities from the measurements and p = 3, this results in a factor 0.89 to 2.02.
Taking only rough scenarios into account provides a factor 1.72 to 2.02, from Table 1.
As a rough approximation for the mean flow velocities around piles the following can be used:

1 2
P
G D
U U
G
o + | |
=
|
\ .

(9)
Piles in the scale model have a diameter of 0.03 m and a centre-to-centre distance of 0.20 m. For a
situation with piles, the available flow area thereby reduces to 85 % of the original area. If this
approach would be valid, the flow velocity for the situation with piles would be 118 % of the original
flow velocity without piles. This approach seems to be a good approximation, with Table 2 containing
difference factors of 1.14 to 1.36 in case of comparing situations with piles to situations without piles.
Potentially oscillating vectors
The results are discussed with time-averaged vectors. In reality, the flow direction at one single
location is not constant. In the design method for a rock protection, the influence the flow direction with
respect to slope direction of the slope angle and angle of internal friction for stones are included.
According to CIRIA ET AL. [2007], the influence of the slope angle is calculated by:

( )
( ) ( ) ( ) ( ) ( ) ( )
2 2 2 2
tan
cos sin cos tan sin sin
s
h
u s u
m

u o o u o
=
+

(11)
This angle of the velocity component can be calculated from the measurements. Because of a certain
turbulence intensity one would expect an oscillating angle of the flow direction. This can be visualized
by plotting the samples of one measurement time series in a rose.


Figure 3. Oscillating flow direction for scenario 10 at (x,y) = (1250,2195)
Rory van Doorn et al. 8
Figure 3 shows an example plot based on 3340 samples (for one single measurement location),
acquired at a frequency of 25 Hz by the ADV. The angle of the momentary velocity vector varies
between about -10 and +40 degrees. The average angle is 15 degrees, which is the angle for the
average velocity vector. It is not investigated how the magnitude of the velocity vector varies with the
direction of the vector. Probably the maximum current velocity does not occur at the most deviating
directions of -10 and +40 degrees. A certain variation of the velocity vector should be taken into
account when calculating the stability coefficient m
h
in equation (5), for example a variation between 0
and +30 degrees. The average angle of 15 degrees would result in a slope coefficient of 0.688 instead
of 0.670 for flow straight up the slope. However, for an angle of 30 degrees this would result in a slope
coefficient of 0.747. Due to the linear relation with the stone diameter, the required stone diameter for
an angle of 30 degrees is 11% larger than for 0 degrees.
The flow direction can be considered as an average angle with a certain standard deviation, which fits
well to a Normal distribution. Because of the available time for research in Van Doorn [2012], no
extensive analysis of oscillating vectors had been executed. It is recommended to do further research
in this subject, because of significant consequences for bed protection. A remark should be made, that
in the performed scale model tests no measurements were undertaken at locations very close to piles
(< 0.5 times the pile diameter). Very close to the piles the flow direction can deviate more.
Discussion
The Dutch equations for a unconfined propeller jet are compared to the measurements.
Propeller-jet characteristics
First, the outflow velocity is slightly overestimated using the current design method. Calculation of
momentum flux shows a 1.52 m/s outflow velocity in scale model tests, whereas 1.60 m/s was
calculated with equation (1) based on an estimated thrust coefficient. This difference could be due to
measuring errors or an overestimation of the propeller thrust coefficient. It seems that the thrust
coefficient was overestimated, because the applied propeller rotational speed was 1021 rpm while the
propeller was designed to operate at 3000 rpm.
Measured relative turbulence intensities near the bed are up to 0.55 at locations of maximum mean
flow velocity. BLAAUW AND VAN DE KAA [1978] provided relative turbulence intensities up to 0.3 in the jet
axis and up to 0.6 at larger radial distances. It should be mentioned that current engineering
guidelines mainly use 0.3 (SCHIERECK [2004]) as value for the relative turbulence intensity, while
measurements show relative turbulence intensities up to 0.55 at critical locations.
Propeller jet diffusion
According to VERHEIJ [1985], Oebius and Schuster did report in 1975 on diffusion of the jet under an
angle of 18.8 and 20.8 degrees, which is 1:2.94 and 1:2.63 for respectively undisturbed extension of
the jet or a restriction by water level and bed. According to VAN DOORN [2012] angles up to 1:2.4 follow
from the model tests with piles, which introduce extra flow obstruction.
Design methods for open quay constructions on piles need to be updated with this larger diffusion
angle, which effects the location of maximum load. Sideways diffusion is considered less important
and therefore not discussed. However, analysis of a sideways diffusion could be performed using the
available data from VAN DOORN [2012].
Velocity distribution
The measured velocity distribution at the inclining slope deviates from equation (3) for an unconfined
jet. However, besides the value and the location of maximum bed load, the velocity distribution
according to BLAAUW AND VAN DE KAA [1978] is in good agreement with the measurements. Sideways
diffusion by the propeller jet shows a good match with equation (3), also for situations with piles.
Hydraulic bed load
As analysed, the hydraulic load for a situation with piles was found up to 2.02 times larger compared
to a situation without piles. However, this is only valid for this specific combination of scale model
dimensions. Possibly smaller or larger hydraulic loads may occur with different pile diameters or
centre-to-centre distances. Remark: no measurements of hydraulic loads were performed close to the
Rory van Doorn et al. 9
piles, within 0.5 pile-diameter distance from piles. Larger maximum hydraulic loads may occur in this
region.
Concluding, hydraulic loads at inclining slopes with and without piles are underestimated using
equation (3) for an unconfined jet. In addition, hydraulic loads as presented in VAN DOORN [2012]
should be further investigated using a non-fixed bed protection. Recently research is done by HOAN
[2008], which clearly describes potential damage to a bed protection and presents a method to
determine bed transport.
Besides hydraulic loads at rock protection using armour stone, further research is recommended with
other types of bed protection such as concrete mattresses. In order to provide sufficient bed protection,
accurate and correct flow velocities should be applied meaning according to this research higher flow
velocities than up to now. Future research with erodible bed protection is needed to investigate
damage levels and required bed protection.
Conclusions
From the investigations the following can be concluded:
- In all situations the measured maximum time averaged current velocity on the slope is under-
estimated by equation (3) for an unconfined propeller jet. The underestimation factor is about:
1.1 to 1.2 for a smooth slope without piles,
1.25 for a rough slope without piles,
1.4 for a smooth slope with piles,
1.6 to 1.7 for a rough slope with piles.
- It is reasonable that the difference factors measured on a relatively steep slope of 1:1.5 are
also valid for a less steep slope, while the two test with different steepness show comparable
difference factors. However more research on the influence of the slope steepness is
recommended.
- Using the Dutch equations for an unconfined propeller jet, the location of maximum bed load is
calculated higher on an inclining slope compared to the measurements. Depending on the
slope angle, roughness and piles, the location of the maximum load from a bow thruster is at
an angle between 1:18.6 and 1:2.4 from the propeller jet axis; the location of the maximum
bed load furthermore seems to depend strongly on obstacles such as piles on a slope. Piles in
front of the propeller jet result in a larger angle;
- The direction of the time averaged velocity vector on the slope can deviate 15 degrees from
the direction of the propeller jet (which is perpendicular against the slope), while the
momentary velocity shows larger deviations due to turbulent fluctuations. This deviation of the
velocity vector from the jet direction has negative effects for armour stone stability and should
be taken into account when calculating the slope coefficient in the equation of Izbash.
- Measured hydraulic bed loads, including turbulence intensities with p = 3, for a rough bed
without piles are underestimated by a factor 1.07 to 1.47, and with piles by a factor 1.72 to
2.02, using current engineering method. This means that the hydraulic load is a factor up to
1.38 times larger in a situation with piles compared to situations without piles.
References
BLAAUW, H.B. AND VAN DE KAA, E.J. [1978] Erosion of bottom and sloping banks caused by the
screwrace of maneuvering ships, publication 202, WL | Delft Hydraulics, Delft, The Netherlands
CIRIA, CUR AND CETMEF [2007] The rock manual. C683 the use of rock in hydraulic engineering.
CIRIA, London.
HOAN, N.T. [2008] Stone stability under non-uniform flow, Delft University of Technology, Delft, The
Netherlands
PIANC [1997] Guidelines for the design of armoured slopes under open piled quay walls. PIANC,
Brussels, Belgium.
PIANC [2013] Guidelines for berthing structures related to thrusters. PIANC, Brussels, Belgium
RMISCH, K. [2006] Erosion potential of bow thrusters on canal banks (in German). Binnenschifffahrt
ZfB 11.
SCHIERECK, G.J. [2004] Introduction to bed, bank and shore protection, engineering the interface soil
and water. Delft University Press, Delft, The Netherlands
SCHOKKING, L.A. [2002] Bowthruster-induced damage, Delft University, MSc Thesis, Delft, The
Netherlands
Rory van Doorn et al. 10
VAN DOORN [2012] Bow thruster currents at open quay constructions on piles, Delft University of
Technology, MSc Thesis, Delft, The Netherlands
VERHEIJ, H.J. [1983] The stability of bottom and banks subjected to the velocities in the propeller jet
behind ships. Proceedings 8th Int. Harbour Congress, Antwerp, Belgium, pp. 1-11.
VERHEIJ, H.J. [1985] Erosion of bank protections along inland waterways propellers jets and stability
of bed material (in Dutch). WL | Delft Hydraulics, report M1115-VII & Xa, Delft, The Netherlands.
WL [1988] Erosion of bank protections along inland waterways Technical recommendations for bank
protections (in Dutch). WL | Delft Hydraulics, report M1115-VIII, Delft, The Netherlands
List of symbols
D pile diameter [m]
0
D
diameter of the jet just behind the propeller, located at the point of
maximal contraction
0
2 0.71
p
D D D = ~ at thrusters without tunnel
0
0.85
p
D D = propeller-jet combinations in a tunnel
0
1.00
p
D D = at ducted thrusters
[m]
50
D
diameter of stone that exceeds the 50% value of sieve curve [m]
p
D
propeller diameter [m]
G spacing between piles [m]
g

acceleration due to gravity [m/s
2
]
T
K
propeller thrust coefficient [-]
L distance between the thruster outflow opening and the slope, measured
horizontally along the jet axis.
[m]
h
m

slope coefficient to account for stone stability on an inclining slope [-]
n coefficient for the number of propellers:
2 for two propellers
[-]
p
n
rotational speed of the propeller [s
-1
]
P
engine power [W]
p
exceeding factor for Normal distribution ( P = 3 is equal to 99.865 %) [-]
r radial distance to the propeller axis [m]
0
U
flow directly behind a propeller, located at the point of maximal
contraction of the jet
[m/s]
1
U
disturbed mean flow velocity [m/s]
2
U
undisturbed mean flow velocity [m/s]
, x axis
U
flow velocity along jet axis [m/s]
, x r
U
flow velocity distribution [m/s]
x distance in the axial direction [m]
u
u
angle of the velocity vector with respect to the upward slope direction (0
= upwards the slope)
[]
o slope angle []
P
o
pile groups coefficient, 1.0 for alignment used in this research [-]
, Is cr
|
critical stability coefficient (commonly 2.5 3.0) [-]
A relative density of rock [-]
s

angle of repose (for armour stone 40 - 42) []
w

density of water [kg/m
3
]