See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/319947494

Marine Propeller Design Method based on Lifting Line Theory and Lifting
Surface Correction Factors

Article in Procedia Engineering · December 2017

DOI: 10.1016/j.proeng.2017.08.132

CITATIONS READS

8 178

3 authors, including:

Ataur Rahman Mashud Karim

Bangladesh University of Engineering and Technology Bangladesh University of Engineering and Technology
2 PUBLICATIONS 9 CITATIONS 74 PUBLICATIONS 584 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Ataur Rahman on 18 April 2024.

The user has requested enhancement of the downloaded file.

 Available online at www.sciencedirect.com

ScienceDirect
Procedia Engineering 194 (2017) 174 – 181

10th International Conference on Marine Technology, MARTEC 2016

Marine Propeller Design Method based on Lifting Line Theory and

Lifting Surface Correction Factors
Ataur Rahmana,∗, Md Refayet Ullahb , Md. Mashud Karimb
a Senior Surveyor, DNVGL Bangladesh. HBFC Building (4th floor), 1/D Agrabad C/A, Chittagong 4100. Bangladesh
b Professor, Department of Naval Architecture and Marine Engineering, Bangladesh University of Engineering and Technology, Dhaka-1000,
Bangladesh

Abstract
In the present paper, a marine propeller design method has been presented based on lifting line theory and lifting surface
correction factors. The developed method has been applied for the propeller whose blade outline is symmetrical about the mid-
chord and the mean line chosen is parabolic which is also symmetrical about the mid-chord. So no pitch correction is necessary.
Only camber correction is required. The blade outline has been chosen symmetric to avoid cavitation. The pitch and camber has
been determined by the requirement that the designed load distribution to be obtained. The method has been used for wake adapted
propeller with an adaptation of the vortex lattice method. Assumption about the loading is that the circulation goes to zero at hub
and tip in both lifting line and lifting surface calculation. Also in lifting surface calculation, the circulation goes to zero at L.E and
T.E. For the calculation of induced velocity at any point in the flow field, the Biot-Savart law has been used. The pitch has been
calculated in lifting line program. Trial and error method has been followed for the program.

©c 2017
2017TheTheAuthors.
Authors. Published
Published by Elsevier
by Elsevier Ltd.is an open access article under the CC BY-NC-ND license
Ltd. This
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the organizing committee of the 10th International Conference on Marine Technology.
Peer-review under responsibility of the organizing committee of the 10th International Conference on Marine Technology.
Keywords: marine propeller; lifting line; lifting surface; wake adapted propelle; vortex lattice method

1. Introduction
Transportation across the oceans has engaged the attention of humankind since the dawn of history. Ships started
thousands of years ago as simple logs or bundles of reeds and have developed into the huge complicated vessels of
today. Wooden sailing ships are known to have appeared by about 1500 BC. The first mechanical propulsion device,
widely used in ships, was the paddle wheel which was gradually superseded by screw propellers for the propulsion of
oceangoing ships from the latter half of the 19th Century. To develop propeller theory for design of screw propeller
it is important to understand the terms used and to have some knowledge of certain principles of hydrodynamics [3].
The circulation theory or vortex theory provides a more satisfactory explanation of the hydrodynamics of propeller
action than the momentum and blade element theories. The lift provided by each propeller blade is explained in terms
of the circulation around it. The circulation theory has been used for designing propellers for over seventy years. If
the propeller blade is represented by a vortex line or lifting line, the effect of the finite width of the blade is neglected
[1,5]. The variation of the induced velocity along the chord of a blade section causes a curvature of the flow over the
blade resulting in changes to the effective camber of the blade sections and the ideal angle of attack.Dueto the low

∗ Corresponding author. Tel.:+880 1755531369

E-mail address: [email protected]

1877-7058 © 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the organizing committee of the 10th International Conference on Marine Technology.
doi:10.1016/j.proeng.2017.08.132
 Ataur Rahman et al. / Procedia Engineering 194 (2017) 174 – 181 175

aspect ratio of most marine propeller blades, the use of lifting line theory results in unacceptably large errors unless
supplemented by a lifting surface correction of some kind. The modern development of the circulation theory is based
on considering the propeller blades as lifting surfaces rather than lifting lines. The simplest mathematical model of a
lifting surface theory was used by Ludwig and Ginjel in 1944. They presented their well-known camber correction
factors, with which the result of a lifting line calculation can be corrected. Instead of a calculation of the camber
distribution over the chord, the maximum camber is calculated only. In 1961, Pien [6] reported a method to solve the
lifting surface equations for the steady case using the continuous vortex distribution. The second generation of lifting-
surface methods was developed around the late 1970s when sufficient computer power became widely available.
Kerwin in 1978 used a vortex-lattice method to solve the lifting surface equations, a method which was earlier used
by Falkner [2]. The lifting surface is represented by a discrete lattice of vortices and the induced velocities are
calculated over an interval from about the quarter chord to the three-quarter chord to determine an average camber
correction factor. The lifting surface theory is used today to determine the camber and pitch angle at the different radii
of the propeller.

Nomenclature

C Blade section chord length, m

ur Radial induced velocity, m/s
CL Lift coefficient
ut Tangential induced velocity, m/s
CT Propeller thrust loading coefficient
u∗ Displacement velocity
f Maximum camber of mean line
x Non-dimensional radius [=r/R]
f¯ Non-dimensional camber
z Number of Propeller blades
G Non-dimensional Circulation
Greek symbols
Kc Camber correction factor
α Angle of attack of section relative to βi
l Chord length of expanded section
ᾱ Non-dimensional pitch correction
R Propeller diameter, m
βi Hydrodynamic pitch angel, deg
r Radius of control point, m
Γ Circulation, m2 /s
u Induced velocity, m/s
γ Vortex sheet strength
ua Axial induced velocity, m/s
ρ Density of fluid kg.s2 /m4
un Normal induced velocity, m/s

2. Basic assumptions

The fluid is assumed to be frictionless and incompressible. The inflow velocity is assumed to be axial and a
function of radius only. For the solution of lifting surface problem, the free vortex system is assumed to lie on a
helical surface whose pitch is determined from lifting line theory with the same radial load distribution. The blade
surface is assumed to be approximately on the helical reference surface. The problem is linearized to the extent that
the boundary condition is applied on the helical surface rather than on the blade itself and the induced velocities are
assumed to be small relative to the resultant inflow. It is assumed that the Kutta condition holds, i.e. that the bound
circulation is zero at the trailing edge. It is also assumed that the bound circulation is zero at the blade tip and at hub.

3. Theoretical Model
3.1. Velocity induced by vortex line

The velocity induced by a vortex line of arbitrary shape may be expressed in terms of an integral taken along the
vortex line by means of Biot-Savart’s law.
  
Γ dlX S
u = (1)
4π S3
176 Ataur Rahman et al. / Procedia Engineering 194 (2017) 174 – 181

Where, Γ = Vortex strength (m2 /sec)

S = Vector distance from vortex element to control point (m)
 = Vector element of distance along the vortex (m)
dl
u = Vector induced velocity (m/sec)

3.2. Lifting line formulation

Using the velocity diagram of Fig.1, the velocity normal to a point on the helical surface with the pitch angle βi at
a radius r can be expressed as [8]

un = −ua . cos βi + ut . sin βi (2)

Fig. 1: Velocity diagram Optimum lifting line propeller

The strength of the bound vortex representing each blade is given by

Γ(ρ)  I
G(ρ) = = ai sin iρ (3)
2πRu∗ i=1
1 + xh − 2x
Where ρ = cos−1 [ ] (4)
1 − xh
From the above equations, we need to find out the solution, the radial distribution of circulation to produce a free
vortex sheet of true helical shape in homogeneous flow, i.e. the optimum propeller. Using the vortex lattice method
and boundary condition, the non-dimensional normal induced velocity at a point r p by a set of semi-infinite helical
vortices originating from each blade with radius rom can be derived as

4πr p 2x p (un )mp

(un )mp = (un )mp . = (5)
Γm u∗Ḡm
The requirement that the relative flow at the lifting line be of constant pitch can be seen from Fig 1 to be
ut ua un
u∗ = = = = Constant (6)
sin βi cos βi cos βi cos βi
2

For the wake adapted propeller, the normal induced velocity can be derived as


I 
M+1
 
ai (ūn )mp ζmp sin i.ρm − ζm−1,p sin i.ρm−1 = 2x p .(cos βi ) p (7)
i=1 m=1
u∗m  (tan βi )m − (tan β)m  rm
here ζmp = = (8)
u∗p (tan βi ) p − (tan β) p r p
and u∗ = ωr(tan βi − tan β) (9)
 Ataur Rahman et al. / Procedia Engineering 194 (2017) 174 – 181 177

3.3. Lifting surface formulation

The vortex sheet strength γ can be converted to a non-dimensional quantity s by dividing by the displacement
velocity U∗ as defined in the preceding section. It is assumed that S can be represented by a series of the form
4  σ  
I I J
S (ρ, σ) = l
Cio sin iρ.cot + Ci j sin iρ sin jσ (10)
D i=1
2 i=1 j=1
 π
1 ds
also G(ρ) = S (ρ, σ) dσ (11)
πD 0 dσ
The value of camber (f/l) and Angle of attack (α) at each radius are determined by the boundary condition that the
flow be tangent to the mean line at each control point.
1   M N
f
α p − hq . p =− ∗ umnpq − (βi − β) p (12)
l V p m=1 n=1
where, hq is the slope of the mean line with unit camber ratio.
From Kutta-Joukowski’s Law,
dL = V ∗ Γdr (13)
The lift coefficient,
dL 2Γ 2πG u∗
CL = ∗ 2
= ∗ = l . ∗) (14)
(1/2(V ) l.dr lV d
V
Using the boundary conditions and from Kutta-Joukowski’s Law,
 4π(b1 ) p x p   4π(b1 ) p x p hq  
M 
N 
J 
I

l
ᾱ − l
f¯ + ζmp ūmnpq μn j Ci j sin iρm
Dp Dp m=1 n=1 j=2 i=1
(15)

M 
N 
I
= −2x p (cos βi ) p − ζmp ūmnpq μn1 Ci1 sin iρm
m=1 n=1 i=1

Here ᾱ = CαL and f¯ = ( Cf /l)

L
, also the chord load factors μn j are constants and the values are given by Falkner [2].
The final set of equations for symmetrical blades is practically containing the same equation except that the terms
containing the pitch correction ᾱ are no longer present.

M 
2N 
J 
I
4π(b1 ) p x p hq f¯ − ζmp ūmnpq μn j Ci j sin iρm
m=1 n=1 j=2 i=1
(16)

M 
2N 
I
= 2x p (cos βi ) p + ζmp ūmnpq μn1 Ci1 sin iρm
m=1 n=1 i=1

The set of equation can be written as matrix notation

AKL .XL = BK (17)

f f CL
As, ( )uncorrected = ( )2D = (for parabolic camber line)
c c 4π
f f
( )corrected = KC ( )uncorrected
c c
f f
( )3D = Kc ( )2D
c c
Therefore camber correction factor
( cf )3D ( cf ) ( cf )
KC = = = 4π f as f =
¯ ¯ (18)
( cf )2D C L 4π CL
178 Ataur Rahman et al. / Procedia Engineering 194 (2017) 174 – 181

4. Design Methodology
The design procedure followed for a wake adapted propeller is provided below [7].
Step 1: Collect the Known Data: Thrust to be produced, diameter, no. of blades, shaft revolutions, ship speed,
wake variations and blade area ratio.
Step 2: Calculate: Advance ratio, thrust loading coefficient and ideal thrust loading coefficient
Step 3: Calculate: Ideal efficiency using KRAMER DIAGRAM.
Step 4: Calculate: Hydrodynamic pitch angles without induced velocities.
Step 5: Calculate: Development of compute program for the calculation of optimum hydrodynamic pitch angles
considering induced velocities through iterations using Lifting Line Theory. The induced velocities will be calculated
using Biot-Savart Law. The computer program will also include calculation of thrust, torque, thrust coefficient, power
coefficient and delivered power.
Step 6: Calculate: B.A.R considering cavitation
Step 7: Calculate: Section chord, section lift coefficient, uncorrected section pitch and section camber
Step 8: Calculate: Development of computer program for the calculation of section pitch and section camber
considering correction factors using Lifting Surface Theory.

5. Application of theoretical formulation

Based on the mathematical model and design methodology, the design method of a marine propeller has been
developed and relevant calculations has been performed considering the basis data taken from [4] of a US Navy
combat DTMB 5415 ship.
Propeller Basis Data (Input)
Thrust, T = 237321 lb = 107649 Kg
Ship Speed, Vs = 21.1 knots = 10.8538 m/s
Propeller Diameter, D = 21.1 ft = 6.4329 m
Density of water, = 104.49 kg sec2 / m4
No. of blades, z = 4
Radius (non-dimensional) Wake distribution Chord distribution
x ω(x) c(x)/D
0.2 0.424 0.220
0.3 0.344 0.249
0.4 0.288 0.271
0.5 0.249 0.284
0.6 0.221 0.289
0.7 0.200 0.284
0.8 0.185 0.260
0.9 0.174 0.209
1.0 0.161 0.000

Output

Table 1: Produced data and corresponding calculations using the Lifting line program

Hydrodynamic uncorrected Chord Uncorrected

Radius Lift coefficient
Pitch Angle, Pitch Diameter c.CL/D Distribution Section camber,
x CL
βi (deg) ratio, (P/D) c/D f/c
0.2 57.6914 0.9936 0.0000 0.2200 0.0000 0.0000
0.3 47.4401 1.0264 0.0952 0.2490 0.3823 0.0304
0.4 39.8164 1.0476 0.1085 0.2710 0.4004 0.0319
0.5 34.0524 1.0616 0.1066 0.2840 0.3754 0.0299
0.6 29.6131 1.0714 0.0979 0.2890 0.3388 0.0270
0.7 26.1282 1.0787 0.0854 0.2840 0.3007 0.0239
0.8 23.3193 1.0834 0.0697 0.2600 0.2681 0.0213
0.9 21.0349 1.0873 0.0493 0.2090 0.2359 0.0188
1.0 19.1579 1.0914 0.0000 0.0000 0.0000 0.0000
 Ataur Rahman et al. / Procedia Engineering 194 (2017) 174 – 181 179

Table 2: Produced data and corresponding calculations using the Lifting surface program (correction factor)

uncorrected corrected Uncorrected Camber corrected

Radius
Pitch Diameter Pitch Diameter Section camber, correction Section camber,
x
ratio, (P/D) ratio, (P/D) f/c factor, Kc f/c
0.2 0.9936 0.9936 0.0000 1.130 0.0000
0.3 1.0264 1.0264 0.0304 1.180 0.0359
0.4 1.0476 1.0476 0.0319 1.230 0.0392
0.5 1.0616 1.0616 0.0299 1.270 0.0380
0.6 1.0714 1.0714 0.0270 1.315 0.0355
0.7 1.0787 1.0787 0.0239 1.345 0.0321
0.8 1.0834 1.0834 0.0213 1.360 0.0290
0.9 1.0873 1.0873 0.0188 1.380 0.0259
1.0 1.0914 1.0914 0.0000 1.380 0.0000

6. Result and discussion

In our design calculations number of blades, propeller diameter, ship speed, thrust to be delivered, radial wake
variation and section chord length variation are taken as known data as same as stated by Kamal & Sallah [3]. But our
methodology of design is different. The validity of the method can be checked from comparison of the two data.

a b

c d

Fig. 2: (a) Comparison of radial variation of Hydrodynamic pitch angle (βi ); (b) Comparison of radial variation of Non-dimensional Circulation
(G);(c) Comparison of radial variation of Pitch-Diameter ratio (P/D); (d) Comparison of radial variation of Camber-Chord ratio ( f /c)

The radial variation of hydrodynamic pitch angle (βi ), radial variation of Circulation (G), radial variation of Pitch-
Diameter ratio (P/D), radial variation of Camber-Chord ratio ( f /c) are computed by the present method are compared
with that predicted by Kamal & Sallah [4]. It is observed that the calculated hydrodynamic pitch angle (βi ) value and
the calculated Circulation (G) value at hub differs slightly with the predicted value but it fully agree with that value at
the tip. The difference in section hydrodynamic pitch angle (βi ) and the calculated Circulation (G) is due to different
180 Ataur Rahman et al. / Procedia Engineering 194 (2017) 174 – 181

boundary condition of loading. On the other hand the calculated Pitch-Diameter ratio (P/D) value and the calculated
Camber-Chord ratio ( f /c)) value at hub fully agree with the predicted value but it differs slightly with that value at the
tip. The difference in section Pitch-Diameter ratio (P/D) and the calculated Camber-Chord ratio ( f /c) are due to the
choice of different mean line. In the present design, parabolic mean line is used where in reference [4] NACA mean
lines is used.
Initially the ideal efficiency (ηi ) is predicted by using Kramer’s Diagram (taken as ηi = 0.77) and several iterations
are done to find out final CT . The same CT value is obtained from any suitable arbitrary prediction of efficiency (ηi ),
this time it is taken as ηi = 0.70. The comparison of the two methods is presented in Figure 3. It is observe that for
different initial value of ideal efficiency the CT i values differs initially but after some iterations both CT i values & CT
values converge to a same definite value. So the iterations can be started from any initial value of ideal efficiency. The

Fig. 3: Convergence of CT i & CT values at different iteration from different initial iteration values

Pitch-Diameter ratio distribution and the Camber-Chord ratio distribution are calculated for the designed thrust to be
produced and it is compared with the same distribution calculated by varying ±10% of the designed thrust. The results
are presented in Figure 4a and 4b. It is observed that the increase of thrust value results the increase of Pitch-Diameter
ratio and the decrease of the thrust value results the decrease of the Pitch-Diameter ratio, similarly the decrease of the
thrust value represents the decrease of the Camber-Chord ratio. These are very usual.

a b

Fig. 4: (a) Effect of thrust requirement on radial Pitch-Diameter ratio distribution (P/D); (b) Effect of thrust requirement on radial Camber-Chord
ratio distribution ( f /c)
 Ataur Rahman et al. / Procedia Engineering 194 (2017) 174 – 181 181

7. Conclusion

In this paper, the development of a design method of a wake adapted marine propeller using lifting line theory with
lifting surface correction factors has been presented. The major outcomes have been summarized as below:

• It appears that the present method of calculation of propeller blade outline with camber correction factors using
lifting surface theory gives good result.
• The output from this design method reflects very near to the practical values because of wake distribution
adaptation.
• From the comparison of values of radial variation of Hydrodynamic pitch, Circulation, Pitch-Diameter ratio,
Camber-Chord ratio with non-dimensional radius indicates the very close agreement with the published data.
• The Pitch and Camber distribution is proportionate to the thrust validate the design method is in good order.

References
[1] Celik, F. and Guner, M., An Improved Lifting Line Method for the Design of Marine Propellers, Marine Technology, 43 (2) (2006).
[2] Falkner, V. M., The solution of lifting-plane problems by vortex-lattice theory, Aeronautical Research Council, R&M no. 2591, 1947.
[3] Ghose, J.P. and Gokarn, R.P., Basic Ship Propulsion, Allied Publishers Pvt. Ltd., India, 2004.
[4] Kamal, I. Z. M. &Sallah, M. S. B. M., Wake adapted propeller design based on lifting line and lifting surface theory, MIMET, 2011.
[5] Kerwin, J. E., The Solution of Propeller Lifting Surface Problems by Vortex Lattice Methods, MIT, Ph. D. Thesis, 1961.
[6] Pien, P. C., The calculation of Marine Propellers based on lifting surface theory, Journal of Ship Research, 5 (2) (1961).
[7] Rahman, A., Development of a marine propeller design method based on lifting line theory and lifting surface correction factors., M. Sc. Engg.
Thesis, BUET, Bangladesh, 2015.
[8] Ullah, M. R., The application of three-dimensional wing theory to the analysis of marine propeller performance, Ph.D. Thesis, University of
Newcastle Upon Tyne, UK, 1984.

View publication stats