Pilot Revalidation

21st August 2009

Effect of initial trim, propulsion and

variable water depth on ship squat

Jonathan Duffy
Australian Maritime College
 Effect of initial trim on ship squat

• Dand and Ferguson (1973) conducted full scale trials and model scale
experiments on full form ships to investigate the effect of initial trim on ship
squat. They found:
– When the ship was fully loaded giving the ship an initial trim by the bow or
stern gave rise to a running trim by the bow.
– A slight initial trim by the bow resulted in an excessive running trim by the
bow, increasing the likelihood of grounding.
– They found that, in ballast draught, once the ship was trimmed by the
stern in the static condition it remained trimmed by the stern whilst
running, making the rudder and propeller susceptible to grounding.

• Barrass (1979) stated that any static trim on a ship will be increased in that
respective direction when underway, that is, if a ship is initially trimmed by the
bow in static condition then the ship will continue to trim by the bow when
underway. Similarly if a ship is statically trimmed by the stern then the ship
will continue to trim by the stern.
 Effect of initial trim on ship squat

0.35m •For a 250m ship the Fn values correspond to:

–Fn=0.041 is 4 knots
0.25m –Fn=0.050 is 4.8 knots

–Fn=0.060 is 5.8 knots

0.15m
–Fn=0.067 is 6.4 knots

•At the lower Froude numbers tested there is only a

small change in midship and forward perpendicular
sinkage due to varying the initial trim angle for a
constant water depth to draught ratio.
h/T=1.1 (static UKC=1.1m) , MarAd L, Schoenhoff (2005) •At the higher Froude numbers tested the midship
sinkage generally decreased slightly as the initial trim
0.45m angle by the stern increased.

•At the higher Froude numbers tested the forward

0.35m perpendicular sinkage generally decreased as the
initial trim angle by the stern increased.
0.25m
•The change in sinkage at the forward perpendicular
was greatest at the highest Froude number for each
water depth to draught ratio investigated.
 Effect of initial trim on ship squat
Stern down
• For a 250m long ship the Fnh values correspond to:

–Fnh=0.21 is 4.9 knots

–Fnh=0.28 is 6.5 knots

–Fnh=0.34 is 7.9 knots

–Fnh=0.43 is 10.0 knots

•The running trim angle is defined as the deviation from a level

static trim.

•A level static trim resulted in a bow down running trim for all
conditions investigated.

Bow down •The running trim changed to stern down as the initial trim by the
h/T=1.3, MarAd L, Schoenhoff (2005) stern was increased.

•The transition between a bow down to stern down running trim

was dependent on the water depth to draught ratio and depth
Froude number (ship speed).

•Generally it was found that as the depth Froude number (ship

speed) was increased level trim occurred at a higher initial trim.

•For an initial trim of 0.25 and 0.5 deg. the running trim by the
stern was greatest for the lowest ship speed.

h/T=1.4, MarAd L, Schoenhoff (2005)

 Effect of initial trim on ship squat

•For a 250m long ship the Fnh values correspond to:

2.25m
–Fnh=0.21 is 4.9 knots

–Fnh=0.28 is 6.5 knots

–Fnh=0.35 is 7.9 knots

1.0m –Fnh=0.44 is 10.0 knots

•As expected, the under keel clearance at both the forward

and aft perpendiculars is reduced when the steady-state
speed is increased.
•The under keel clearance at the forward perpendicular
increased as the initial trim angle increased.
1.25m
•The under keel clearance at the aft perpendicular is
reduced at the higher initial trim by the stern cases.
•The ship model nearly grounded for an initial trim angle of

0.62m
0.5 deg. by the stern at a speed of 10 knots, highlighting
the danger of excessive initial trim by the stern. This may
be accentuated with an operating propeller.

h/T=1.2(static UKC=2.3m), MarAd L, Schoenhoff(2005)

 Effect of initial trim on ship squat
•For a 250m long ship the Fn values correspond to:
1.13m
–Fn=0.041 is 4.0 knots
–Fn=0.050 is 4.8 knots
0.88m –Fn=0.060 is 5.8 knots
–Fn=0.067 is 6.5 knots
–Fn=0.081 is 7.8 knots
0.63m –Fn=0.103 is 9.9 knots

•The results clearly show that an initial trim by the stern can increase the
minimum under keel clearance when underway.

•For a constant Froude number of 0.067 (6.5 knots) at h/T=1.1 using an

initial trim of 0.12 degrees by the stern will result in an increase of
approximately 20% in the minimum under keel clearance compared to the
level initial trim case (bare hull only).

h/T=1.1, Schoenhoff (2005) •This may be increased further with a smaller initial trim angle.

•For a constant Froude number of 0.081 (7.8 knots) at h/T=1.2 using an

initial trim of 0.12 degrees by the stern will result in an increase of
approximately 60% in the minimum under keel clearance compared to the
0.88m
level initial trim case (bare hull only).

0.63m •For a constant Froude number of 0.103 (9.9 knots) at h/T=1.2 the ship
grounded with an initial level trim, but with an initial trim less than 0.5
degrees sufficient under keel clearance was obtained.
0.38m
•When the initial trim by the stern exceeds the optimum value the
minimum under keel clearance is reduced and may cause the stern to
ground, as is the case for a Froude number of 0.103 (9.9 knots) at
h/T=1.2.

h/T=1.2, Schoenhoff (2005)

 Effect of initial trim on ship squat

1.13m •For a 250m long ship the Fn values

correspond to:
0.88m –Fn=0.050 is 4.8 knots
–Fn=0.067 is 6.5 knots
0.63m
–Fn=0.081 is 7.8 knots
–Fn=0.103 is 9.9 knots

•As the ship speed is increased the initial trim

angle to give highest min. UKC increases.

h/T=1.3, MarAd L, Schoenhoff (2005)

•The increase in min. UKC due to an initial
trim is greatest at the higher ship speeds.
1.13m
•For a water depth to draught ratio of 1.4 and
0.88m a Froude number of 0.103 an initial trim by
the stern would result in a 200% increase in
0.63m minimum under keel clearance.

0.38m •However, too large an initial trim by the

stern may place the rudder and propeller at
potential risk of grounding.

h/T=1.4, MarAd L, Schoenhoff (2005)

 Effect of propulsion on ship squat

• The effect of an operating propeller can reduce the pressure in the vicinity of
the propeller and influence the sinkage and trim of the ship.

• It is important to accurately predict the effect of propulsion on vertical plane

motions when simulating the behaviour of a ship operating in restricted water.

• AMC conducted a study into the effect of propulsion on a MarAd L series bulk
carrier model for a range of speeds, at various water depths.

• The model was towed at a constant speed and a range of motor power
settings from zero to an equivalent full-ahead motor power setting was tested
at each ship speed (length Froude number) to simulate a change in thrust
setting.
 Effect of propulsion on ship squat

MarAd L series bulk carrier model with propeller fitted

 Effect of propulsion on ship squat
0.14
•Effective thrust = Xpropelled – Xunpropelled
h/T=1.1
0.12
Non-dimensional heave force (Z")

•For a 250m long ship the Fnh values at

0.1
h/T=1.1 correspond to:
–Fnh=0.327 is 4.6 kn
0.08 –Fn=0.433 is 6.1 kn
–Fn=0.499 is 7.0 kn
0.06

•For a 250m long ship the Fnh values at

0.04 h/T=1.2 correspond to:
–Fnh=0.327 is 4.8 kn
0.02
Fnh = 0.327 –Fn=0.433 is 6.3 kn
Fnh = 0.433
–Fn=0.499 is 7.3 kn
Towards sea floor Fnh = 0.499
0
0 0.5 1 1.5 2 2.5 3 3.5
•Increasing effective thrust increases the
Non-dimensional effective thrust (1000T")
heave force experienced by the ship due to
0.14
the change in pressure over the hull caused
h/T=1.2
0.12
by the operating propeller.
Non-dimensional heave force (Z")

0.1 •The increase in heave force when changing

from zero thrust to an equivalent full ahead
0.08 thrust was in the order of 30% at h/T=1.1 and
20% at h/T=1.2.
0.06

•Figures taken from Duffy (2000)

0.04

Fnh = 0.327
0.02
Fnh = 0.433
Towards sea floor Fnh = 0.499
0
0 0.5 1 1.5 2 2.5 3 3.5
Non-dimensional effective thrust (1000T")
 Effect of propulsion on ship squat
1
• Effective thrust = Xpropelled – Xunpropelled
Non-dimensional pitch moment (1000M")

Bow up
0
Bow down
• Increasing effective thrust reduces the bow
-1
down pitch moment experienced by the ship
due to the change in pressure over the hull
caused by the operating propeller.
-2

• The reduction in pitch moment when changing

-3
from zero thrust to an equivalent full ahead
thrust was in the order of 65% at h/T=1.1 and
-4 Fnh = 0.327 50% at h/T=1.2 (Fnh=0.499).
h/T=1.1 Fnh = 0.433
Fnh = 0.499
-5
0 0.5 1 1.5 2 2.5 3 3.5
• At a Fnh of 0.327 the pitch moment became
Non-dimensional effective thrust (1000T") bow up at the higher thrusts.
1

• Propulsion had a far greater influence on pitch

Non-dimensional pitch moment (1000M")

Bow up
0 moment than heave force.
Bow down

-1 • As the effect of the propulsion is to increase the

draft aft, a ship with an initial trim by the stern
-2 may be caused to ground at the stern in very
shallow water if full ahead thrust is applied.
-3

• Data from Duffy (2000)

-4 Fnh = 0.327
h/T=1.2 Fnh = 0.433
Fnh = 0.499
-5
0 0.5 1 1.5 2 2.5 3 3.5

Non-dimensional effective thrust (1000T")

 Squat over an undulating sea floor
• Edstrand
and Norrbin (1978) investigated squat experienced by a ship when travelling over a ‘platform type shallow’,
whereas Ferguson et al. (1982) investigated squat over a ramp bank leading to a level platform bank.

Platform shallow
(Edstrand & Norrbin (1978))

Bank profile
(Ferguson et al. (1978))

•Both found that the bow dipped towards the bottom as it passed the leading edge of the level platform part of the
respective banks.

•Ferguson et al. found that upon approaching the bank, the transient forces and moments induced an instantaneous
‘overshoot’ of the steady-state sinkage and trim by the bow, which resulted in a maximum instantaneous bow draught
greater than the steady-state draught underway, causing the ship to ground.

•To a lesser extent unsteady squat has been studied at full scale (Queensland Transport 1996; Hatch 1999), but full scale
results are difficult to compare to theoretical predictions due to the complex bottom topography and non-constant ship
speed.

•In comparison to the number of studies conducted to investigate squat in uniform depth, the paucity of data for unsteady
squat is apparent. In addition to this, it was found that the trends from previous studies were not consistent.
 Squat over an undulating sea floor
• Studies have been conducted at the Australian Maritime College (AMC) to investigate the squat of a ship when
travelling over an undulating sea floor.

• In comparison to the cases tested in the recent AMC study, the ramp slope modelled by Ferguson et al. (1982) was
considerably lower and the water depth to draught ratios before the bank were considerably lower.

Schematic of simplified bank geometries for unsteady squat tests at AMC

Squat over an undulating sea floor

Bank model in AMC towing tank

 Squat over an undulating sea floor
•For the cases tested at AMC it was found that the ship experienced a bow down trim leading up to the bank and over
the flat portion of the bank.

•For most cases a stern down trim occurred as the bow left the bank; this was also found by Edstrand and Norrbin (1978)
and Ferguson et al. (1982).
0.2
0.50m
0.1
Bow up
0
Bow down
-0.1

-0.50m -0.2
Trim (%LPP)

Trim over ramp bank, AMC test,

-0.3
MarAd L, Fnh=0.30-0.55 (12 knots),
-1.00m -0.4 h/T=4.3-1.3 (48m-14.7m), Duffy
(2008)
-0.5

-0.6
Measured trim
Ship origin at leading toe of bank
-0.7 Ship origin at leading edge of level bank
Ship origin at trailing edge of level bank
-2.00m -0.8
0 1 2 3 4 5 6 7 8 9 10

Non-dimensional distance (x m)

•Ferguson et al. (1982) found that when the bow had passed the leading edge of the level portion of the bank it
experienced a sudden repulsion, which was counteracted by an increase in stern sinkage; it was found that the trim
changed from bow down to bow up at this point.

•In the present study there was a reduction in bow down trim when the ship origin was approximately half a ship length
past the leading edge of the level portion of the bank.

•This variation in trim behaviour compared to Ferguson et al. (1982) may be attributed to the different bank
dimensions, ship particulars and water depth to draught ratios before and over the bank.
 1 1
Measured - h1/d=4.2-1.2, Fnh=0.25-0.47 Measured - Semi-captive model - h1/d=4.2-1.2, Fnh=0.25-0.47
Static UKC over bank=2.3m Measured - Steady state - h1/d=1.2, Fnh=0.47 Static UKC over bank=2.3m Measured - Steady state - h1/d=1.2, Fnh=0.47
0.9 0.9
Measured - h1/d=4.3-1.3, Fnh=0.25-0.45
Static UKC over bank=3.4m Static UKC over bank=3.4m Measured - Semi-captive model - h1/d=4.3-1.3, Fnh=0.25-0.45
Measured - Steady-state - h1/d=1.3, Fnh=0.45
Measured - Steady state - h1/d=1.3, Fnh=0.45
Midship sinkage (smidships) (%LPP)

0.8 Ship origin at leading toe of bank 0.8 Ship origin at leading toe of bank
Ship origin at leading edge of level bank
Ship origin at leading edge of level bank

Bow sinkage (sbow) (%LPP)

Ship origin at trailing edge of level bank
0.7 0.7 Ship origin at trailing edge of level bank

0.6 1.50m 0.6 1.50m

0.5 0.5

0.4 1.00m 0.4 1.00m

0.3 0.3
0.75m
0.2
0.50m 0.2

0.1 0.1
Towards sea floor Towards sea floor
0 0
0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10
Non-dimensional distance (xm) Non-dimensional distance (xm)

Midship and bow sinkage over ramp bank for different water depths (10.5 knots), MarAd L, Duffy (2008)
•The midship sinkage and bow sinkage increased from a steady-state value as the ship travelled over the ramp;
however, the increase was not notable until after the ship origin (midships) had travelled past the leading toe of the bank.

•There was not an instantaneous overshoot of midship and bow sinkage upon the bow approaching the leading edge of
the level portion of the bank, as measured by Ferguson et al. (1982).

•There was actually a considerable lag in the distance taken for the midship and bow sinkage to reach the equivalent
steady-state sinkage in the same water depth over the bank.

•Heave and pitch oscillations occurred as the ship passed over the flat portion of the bank, as measured by Edstrand and
Norrbin (1978) and Ferguson et al. (1982).

•Generally, the amplitude of the heave and pitch oscillations over the level portion of the bank increased at low water
depth to draught ratios over the bank.

•The oscillations did not decay before the ship reached the end of the bank when the water depth
to draught ratio over the bank was reduced to a value of 1.2.
 Squat over an undulating sea floor
1 1
Measured - h1/d=4.3-1.3, Fnh=0.20-0.36 Measured - h1/d=4.3-1.3, Fnh=0.20-0.36
Static UKC over bank=3.4m Measured - Steady state - h1/d=1.3, Fnh=0.36 8.5 knots Static UKC over bank=3.4m Measured - Steady state - h1/d=1.3, Fnh=0.36
8.5 knots
0.9 0.9
Measured - h1/d=4.3-1.3, Fnh=0.25-0.45
Static UKC over bank=3.4m Measured - Steady state - h1/d=1.3, Fnh=0.45
10.5 knots Static UKC over bank=3.4m Measured - h1/d=4.3-1.3, Fnh=0.25-0.45 10.5 knots
Measured - Steady state - h1/d=1.3, Fnh=0.45
Ship origin at leading toe of bank
Midship sinkage (smidships) (%LPP)

0.8 0.8 Ship origin at leading toe of bank

Ship origin at leading edge of level bank
Ship origin at trailing edge of level bank Ship origin at leading edge of level bank

Bow sinkage (s bow) (%LPP)

0.7 0.7 Ship origin at trailing edge of level bank

0.6 1.50m 0.6 1.50m

0.5 0.5

0.4 1.00m 0.4 1.00m

0.3 0.3
0.75m
0.2
0.50m 0.2

0.1 0.1
Towards sea floor Towards sea floor
0 0
0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10
Non-dimensional distance (xm) Non-dimensional distance (xm)

Midship and bow sinkage over ramp bank for different ship
speeds
•The length of the lag was generally larger at higher ship speeds.

•The lag in the sinkage may be due to viscous effects in the thin layer of
fluid between the hull and the sea floor.

•Generally, the amplitude of the heave and pitch oscillations over the level
portion of the bank was larger at high ship speeds.
 Squat over an undulating sea floor
0.15 0.25
Ramp bank Ramp bank
Step bank Step bank
Ship origin at leading toe of bank (ramp bank only) Ship origin at leading toe of bank (ramp bank only)

Non-dimensional pitch moment (100M')

Ship origin at leading edge of level/step bank Ship origin at leading edge of level/step bank
Non-dimensional heave force (Z')

Ship origin at trailing edge of level/step bank Ship origin at trailing edge of level/step bank
0.12 Bow up
0
Bow down

0.09

-0.25

0.06

-0.5
0.03

Towards sea floor

0 -0.75
0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10
Non-dimensional distance (xm) Non-dimensional distance (xm)

Measurements over ramp bank and step bank, h/T=4.3-1.3, Fnh=0.35-0.63

•The magnitude of the maximum bow down pitch moment was greater for the step case when the ship origin was in the
vicinity of the leading edge of the level/step bank.

•There was a much larger reduction in the bow down moment for the step bank case when the origin was located at a
non-dimensional distance of approximately 4.5.

•For the step bank cases tested in the present study, the midship and bow sinkage did not begin to increase before the
bow reached the step.

•This may be due to the relatively high water depth to draught ratio before the bank in the present study; therefore the
restriction to the flow may not be sufficient for the ‘reflected’ fluid to cause sufficient pressure changes to influence the
sinkage before the ship reaches the bank.

•The differences in sinkage induced by the two banks may be due to the more rapid movement of
fluid associated with transit over the more abrupt depth change for the step bank case.
 References
•Barrass, C.B. 1979, ‘The phenomena of ship squat’, International Shipbuilding Progress, vol. 26, pp.
44-47.

•Dand, I.W. & Ferguson, A.M. 1973, ‘The squat of full ships in shallow water’, Trans. RINA, vol. 115,
pp. 237-255.

• Duffy, J.T. 2008, Modelling of ship-bank interaction and ship squat for ship-handling
simulation, PhD Thesis, Australian Maritime College.

•Duffy, J.T. & Renilson, M.R. 2000, ‘An investigation into the effect of propulsion on ship squat’,
Oceanic Engineering International, vol. 4, no. 1, pp.1-12.

•Edstrand, H. & Norrbin, N. H. 1978, ‘Shallow water phenomena and scale model research – some first
experience from the SSPA Maritime Dynamics Laboratory’, Publications of the Swedish State
Shipbuilding Experimental Tank, Report no. 81, Goteborg, Sweden.

•Ferguson, A., Seren, D. & McGregor, R. 1982, ‘Experimental investigation of a grounding on a

shoaling sandbank’, Trans. RINA, vol. 124, pp. 303-324.

•Hatch, T. 1999, ‘Experience measuring full scale squat of full form ships at Australian ports’,
Proceedings of Coasts and Ports ‘99, Perth, Australia.

•Queensland Transport 1996, ‘Squat measurements in the Port of Cairns’, Transport Technology
Division, Maritime Division, Document no. R-007.

•Schoenhoff, T. 2005, Investigation into the effect of initial trim on steady state ship
squat, Bachelor of Engineering (Nav. Arch.) Thesis, Australian Maritime College.