Planar Motion

Mechanism
 Planar Motion Mechanism

PMM is an equipment as well as a measurement

system to experimentally determine all the
hydrodynamic coefficients.
[Link]
PMM thus essentially consists of an equipment where
the model ship is attached to two struts on which there
are dynamometers to measure the forces coming in the
y direction
We can make the struts move together or independently
thus the equation has a phase difference of ε 2
 Planar Motion Mechanism

Thus with 2 dynamometers at the

struts attached on which we are giving
the motion, we are measuring force in
the Y direction, at Y stern point and Y
bow point
Objective is to measure force when all
others are 0 in the taken system

3
 Planar Motion Mechanism

Thus we have tests that help us achieve our objective of finding

the coefficients.
Static tests
q Same as straight line tests
q Considering 𝑦! and 𝑦" as constant values, drift angle 𝛽 is
fixed and we tow the ship
q We can then measure Y/N vs 𝛽 𝑎𝑛𝑑 𝛿 and obtain both linear
and non linear velocity hydrodynamic coefficients 4
 Dynamic Tests

Pure Sway Pure Yaw

Only v, 𝑣̇ , no r,𝑟̇ Only r,𝑟,̇ no v, 𝑣̇

5
 Pure Sway

Considering the same y equation

for the stern and bow points
The model would therefore give
rise to graphs as shown
Thus this test gives the motion of 𝑦 = −𝑎! cos 𝜔𝑡
ship that is being purely swayed, 𝜕𝑦
that is there is no rotation only 𝑣= = 𝑦̇ = 𝜔𝑎! sin 𝜔𝑡
𝜕𝑡
translation. d𝑣
𝑣̇ = = 𝜔" 𝑎! cos 𝜔 𝑡 6
d𝑡
 Pure Sway

Since the graphs of y,v, 𝑣̇

are sine and cosine
curves we can easily get
instances where only v
exists and where only 𝑣̇
exists A : y=0, v= 𝜔𝑎! , 𝑣̇ =0 C: y=0, v= - 𝜔𝑎! , 𝑣̇ =0

B: y= +𝑎! , v=0, 𝑣̇ = - 𝜔" 𝑎! D: y= -𝑎! , v=0, 𝑣̇ = +𝜔" 𝑎!

7
 Pure Sway

Now how would the force look like

and how would it arise?
Some places Y arises only because
of v and some only because of 𝑣̇
Thus the force Y is given by both
sine and cosine curves
Thus one component of Y cosine
arises due to acceleration and sine
component arises due to velocity 𝑌 = 𝑌! cos 𝜔𝑡 + 𝜀 ̅
8
= 𝑌#$ cos 𝜔𝑡 + 𝑌%&' sin 𝜔𝑡
 Pure Sway

𝑌%&' 𝑌" %&' + 𝑌! %&'

Thus the force components 𝑌$ = =
𝜔𝑎( 𝜔𝑎(
arising due to velocity and
𝑥) 𝑌" %&' − 𝑌* +,-
acceleration is given as 𝑁$ =
shown: 𝜔𝑎(
𝑌" ./ + 𝑌! ./
Modulus is for understanding (𝑌$̇ −𝑚) =
but appropriate sign 𝜔 0 𝑎(
3! "# 4 3$ "#
convention is to be followed (𝑁1̇ − 𝑚𝑥2 ) = 𝑥)
5 % 6&
Why (𝑌$̇ −𝑚) and not just 𝑌$̇ ?
9
 Pure Sway

Now we can thus obtain the graph of 𝑌78 vs t arising due to 𝑣.̇
But would it be appropriate if we took only those points where
v=0 for getting the curve?
Would the curve be accurate?

Therefore, we need to find 𝑌78 and 𝑌+,- , using all the data
points and integrate to expression of of Y wrt t from 0 to nt by
multiplying the expression with cos 𝜔𝑡.
10
 Pure Sway
Y = 𝑌#$ cos 𝜔𝑡 + 𝑌%&' sin 𝜔𝑡
Thus integrating,
()
∫! 𝑌 𝑡 cos 𝜔𝑡 𝑑𝑡
()
= ∫! (𝑌#$ cos 𝜔𝑡 + 𝑌%&' sin 𝜔𝑡) cos 𝜔𝑡 𝑑𝑡
()
= ∫! 𝑌#$ cos 𝜔𝑡 " d𝑡
We thus obtain the graph for 𝑌#$
Same for finding 𝑌%&'
11
 Pure Sway

Thus, even though there are only instances where v or 𝑣̇ =0, for the
finding the force we end up using Fourier sort of method analysis and
use all the points of measurement by integrating to avoid a lot of
error.
We did this by presuming the Y vs t obtained to be a sine or cosine
curve and area under the curve multiplied with cosωt or sinωt
remains constant
12
 Pure Yaw

Now for pure yaw

test we need
consider only r and
no v or 𝑣̇ which
means the body v=
*+
=
* ,
𝑦 + 𝑦.
*) *) " -
must go tangential, /
= tan 𝜓 = 𝜓 if 𝜓 is small
0
velocity vector
needs to be
tangential 13
 Pure Yaw

𝑦 = 𝑎! cos 𝜔𝑡
𝜕𝑦
𝑣= = 𝑦̇ = −𝜔𝑎! sin 𝜔𝑡
𝜕𝑡
d𝑣
𝑣̇ = = −𝜔" 𝑎! cos 𝜔 𝑡
d𝑡
𝑌 = 𝑌! cos 𝜔𝑡 − 𝛽
Y = 𝑌#$ cos 𝜔𝑡 + 𝑌%&' sin 𝜔𝑡
𝑌#$ = 𝑌! cos 𝛽
𝑌%&' = 𝑌! sin 𝛽
14
 Pure Yaw

We get,
9: 9 < 0
v= = 𝑦! + 𝑦" 𝜔𝑥)
9; 9; 0 1−
1 𝑢(
=𝜓 cos 𝜖 = 0
= 𝜔𝑥)
1+
𝜓 = v/u 𝑢(
……(1) 5>'
Or, tan 𝜖/2 =
=&
𝜓 = (𝑦" - 𝑦! )/ 2𝑥)
Thus you now try to find out what is the value
…….(2)
of 𝜖 (phase gap) so that the velocity vector v
(1)=(2) is tangential to the path line 15
 Pure Yaw

If we find the 𝜖 and then the 𝜓

will give a sinusoidal function
Thus we have heading angle given
by cosin curve
Rotational velocity sine curve and
𝛹 = −𝜓( cos 𝜔𝑡 − 𝜖E2
rotational acceleration cosine
curves as shown 𝜓̇ =r= 𝜓( 𝜔 sin 𝜔𝑡 − ?⁄0
Then we follow the same 𝜓=̈ 𝑟̇ = 𝜓 ( 𝜔 0
cos 𝜔𝑡 − ?⁄
0
procedure as done for pure sway 16
 Pure Yaw

𝑌" + 𝑌! %&'
%&'
(𝑌@ −𝑚𝑢( ) =
−𝜓( 𝜔
𝑥) 𝑌" %&' − 𝑌* +,-
(𝑁@ −𝑚𝑥2 𝑢( ) =
−𝜓( 𝜔
𝑌" ./ + 𝑌! ./
(𝑌@̇ −𝑚𝑥2 ) =
−𝜓( 𝜔 0
3! "# 4 3$ "#
(𝑁@̇ − 𝐼A ) = 𝑥)
4B& 5%
17
 Pure Yaw

Coming back to the question why (𝑌$̇ −𝑚) and (𝑌@̇ −𝑚𝑥2 )
and not just 𝑌$̇ or 𝑌@̇ ?
It is because when the force is measured it includes the
mass inertia
Thus since the equipment would measure [m+m(added
mass)]x acceleration we subtract in order to obtain the
actual coefficient
18
 Pure Yaw

But if we consider 𝑦! and 𝑦" to be arbitrary

All we need to find are the factors that ensure velocity vector
remains tangential:
q It depends on u
q Frequency of oscillation
q Distance between 𝑦! and 𝑦"

19
 Pure Yaw

𝑦6 (𝑡) = 𝑎 cos 𝜔𝑡
𝑦C (𝑡) = 𝑏 cos(𝜔𝑡+𝜓)
𝑌6 𝑡 = 𝐹6 cos(𝜔𝑡+𝜃6 )
𝑌C 𝑡 = 𝐹C cos(𝜔𝑡+𝜃C )
Solving by adding 𝑌6 + 𝑌C , 𝑌6 − 𝑌C we try to obtain expressions for v,r, 𝜓
we get
:̇ ( D:̇ ) :̇ ( D:̇ )
𝑣= , r= , 𝜓 = (𝑦C - 𝑦F )/2L
0 0E
20
 Pure Yaw

Now taking a=b and equating the sine and cosine terms we get 4 equations
with 8 unknowns
Now taking 𝜓 = 180° we get 4 equations in terms of Y! Y!̇ N! N!̇ as unknowns
Now taking 𝜓 = 0° we get 4 equations in terms of Y# Y#̇ N# N#̇ making it a pure
sway test
So we can not only take 0 or 180 any 2 other angles to generalize the test so
that we can get 8 equations and 8 unknowns
Solve the equations and check!
For submarines similarly we need vertical planar motion mechanism 21
Rudder and
Control Surface
 Rudder and Control Surface

For a surface ship rudder is essentially the control surface

There are primarily 3 rudder types: balanced (supported by both ends); horn
type (hanging and supported by the horn); spade
Modifications are made so as to get extra lift
Now when we consider the section of the rudder, we have movable rudder, we
can either move the entire thing or have a fixed part and moving part
This is done as we have a control surface and a camber. The lift that comes
has 2 effects one is angle of attack attack and other is because of the
camber 23
Rudder and Control Surface

24
 Geometry of a Rudder

Measurements along x is called chord

𝐶( is chord at root section, 𝐶) is chord at tip section
Rudder is parallel to hull to increase lift
It is thicker at root than tip due to the bending moment
acting at the joints
To understand the taper and have a measure of the
tapering we need to look at:
Sweep angle (^) is at the points where ¼ of 𝐶( and 𝐶) is
joined
𝐶* is centre of pressure, its coordinates are 𝐶+ ) , 𝐶* (
25
 Looking closely at Lift and Drag Forces

Rudder angle - angle of attack 𝛼 is lift

angle at the rudder
Lift force is normal to velocity direction
Thus we need to resolve the forces in
proper direction:
𝐹' is total resultant force, 𝐹( is
component of resultant to the x axis of
the rudder

26
 Estimating Rudder forces

Therefore resolving the components according

to different axes we see as in the figure
We need to resolve along all the coordinates
because we look for Y force due to the rudder
and X force the due to the rudder
𝑌G=HHI@ = 𝑌J@ 𝛿𝑟 = ± 𝐿 cos 𝛽G + 𝐷 sin 𝛽G
𝑁G=HHI@ = 𝑁J@ 𝛿𝑟 = 𝑌G=HHI@ . 𝑥G
𝑋G=HHI@ = 𝐿 sin 𝛽G − 𝐷 𝑐𝑜s 𝛽G
𝑥G is distance between 𝐶2 and 𝐶K 27
 Estimating Rudder Forces

While designing the rudder stock we need to know the torque and bending
moment coming so that the rudder stock will have enough strength to withstand
that stress
Hydrodynamic Torque to the stock

𝑄$ = 𝐹% 𝑑 − (𝑐& ( ̅ ) , d = mean distance between leading edge and rudder

stock

Bending moment on the rudder stock = 𝐿) + 𝐷 ) . 𝑐& (̅

28
We see that if we know L&D we can figure all other quantities
 Estimating Rudder Forces
1
Lift coefficient: 𝐶1 = ! "
"
2# 0
3
Drag: 𝐶3 = ! "
"
2# 0
4$
Normal force coefficient: 𝐶$ = ! = 𝐶1 cos 𝛼 + 𝐶3 sin 𝛼
2 0"
" #
9
Moment coefficient about the rudder stock: (𝐶5 ) = 𝐹$ 𝑑 − (𝑐6 ̅ ) / 𝐴: 𝑢" 𝑐̅
8 "
9
Moment coefficient about quarter chord: 𝐶;&% = 𝐹 0.25𝑐̅ − (𝑐6 ̅ ) / 𝐴: 𝑢" 𝑐̅
8 "
'
9
Bending moment coefficient about root section: 𝐿" + 𝐷" . 𝑐6 8̅ / 𝐴: 𝑢" 𝑐̅
"
29
 Estimating Rudder Forces

𝑌J@ 𝛿𝑟 ~ L (𝛼 ≈ 𝛿@ , 𝛽 ≈ 0)
M
~ 𝐶L 𝐴G 𝑢 0
0
𝜕𝐶L
𝐶L = 𝛼
𝜕𝛼
NO*
Thus we get, 𝑌J@ ≅ 𝐹
NP

30
 Estimating Rudder Forces
We can make an estimate of 𝑌,( as a function of 𝐶- and 𝐶. by taking the approximation of 𝛽( = 0,
/0
𝐹 /1! is called slope of the lift curve
Thus we see that with the help of L and D we found:
q L and D are written in terms of 𝐶- and 𝐶.
/0!
q 𝐶- and 𝐶. are written in /1
𝛼 for a given rudder section
q We thus find 𝑌,( 𝑁,(
q Moment torque coming to the rudder or how much force we need to turn it
q Bending moment is important to make the size of the stock
q Although there is correction because of drift angle 𝛽( ,it is usually small 31
 Estimating Rudder Forces

For rudder forces we would look at much simpler semi empirical formula
which would be useful to make a quick estimate of forces etc
Forces along with 𝐶L , 𝐶Q etc depend on
q Geometry influenced by aspect ratio, taper ratio/sweep angle
q 𝑣 0 , 𝜌, 𝛼
q Rudder area 𝐴@
q Cross sectional shape
32
 Estimating Rudder Forces

𝐹G = 577 𝐴@ 𝑣 0 𝑆𝑖𝑛(𝛿G ) Newton

v = 1.3x 𝑣)RST if it is a rudder behind propeller single screw
v= 1.2 x 𝑣)RST for a centreline rudder behind twin screw
𝐹G = 21.1 𝐴@ 𝑣 0 𝛿G Newton, ahead motion
= 19.1 𝐴@ 𝑣 0 𝛿G Newton, astern motion, 𝛿G in degrees
= 18.0 𝐴@ 𝑣 0 𝛿G Newton, for centerline rudder behind twin propellers
v here = 𝑣)RST 33
 Estimating Rudder Forces

/2!
More recent formulas for estimating 7
/1 13!
is called slope of the lift curve

rudder forces: a = effective aspect ratio= 𝑠𝑝𝑎𝑛" / 𝐴( ( 𝐴( is

𝜕𝐶L 𝐶QO 𝛼 0 area)
𝐶L = 𝛼+ ^ = sweep angle
𝜕𝛼 𝑎 57.3
NU* ((.Y)(0[6) 𝛼 = angle of attack in degrees
a = ……(3)
NP PV( (% 𝐶.0 = cross flow drag coefficient dependent
\].^ [(<.`Dab) ^ Dd)]
+,' ^
on both tip-shape and taper ratio
per deg.
= it is the correction term

Now for 𝐶Q : 34
 Estimating Rudder Forces

𝐶L0
𝐶Q = 𝐶Hf +
𝜋𝑎𝑒
𝐶Hf = minimum drag coefficient for a
given foil section
e = “Oswald” efficiency factor
NU*
a = (𝜋/2)a per radian …………(1)
NP PV(
NU* 0[
a = % per radian ……………(2)
NP PV( <D( )
( 35
 Estimating Rudder Forces

𝑐h U/d
𝑐K = 0.25 − 𝑐̅
U̅ 𝑐8
Simple formulas 𝑐K Now this completes all
the required quantities
𝑐K aft of leading edge
to find Lift, drag, force
𝑐K = 0.195 + 0.305 sin(𝛼) and moment
U̅
𝑐K = 0.35 𝑐 ̅ aft of leading edge
U̅
= 0.31 𝑐 ̅ if rudder is in open water 36
 Problem

What would be the bending stress coming, T

and M
Assume:
𝐹@ = 21.1 𝐴@ 𝑣 0 𝛿G
𝑐K = 0.31 chord length
𝛿G =35°
V= 20 knots
37
Hint: Find for each section stress , equivanlent BM, M, T, 𝐹(