E K I PDF
E K I PDF
E K I PDF
UDC 629.5.035
S. EKINCI
Serkan EKINCI
A Practical Approach for Design of Marine Propellers with Systematic Propeller Series
Original scientic paper Although there have been important developments in marine propellers since their rst use in 1850, the main concept has been conserved. Parallel to the developments in the computer technologies in the past 50 years, methods based on the circulation theory are often used for the design and analysis of propellers. Due to the ability to predict the propulsion performance with just a few design parameters, the use of systematic propeller series based on open water model experiments is still widespread. The design with standard propeller series is usually made with diagrams developed from model experiments. Reading errors during the use of these diagrams are inevitable. In the presented study a practical approach is developed for preventing reading errors and time loss during the design with standard propeller series. A practical approach based on empirical formulas for the design and performance prediction for four-bladed Wageningen B propeller series is presented, and design applications for three propellers with different loading conditions are realized. The results obtained from the presented method are compared with those of open water diagram data and a good agreement between the results is observed. Keywords: propeller design, propeller series, propeller performance, open water diagram
Authors Address (Adresa autora): Yildiz Technical University, Department of Naval Architecture and Marine Engineering, 34349 Besiktas, Istanbul, Turkey E-mail: ekinci@yildiz.edu.tr Received (Primljeno): 2011-01-31 Accepted (Prihvaeno): 2011-03-29 Open for discussion (Otvoreno za raspravu): 2012-06-30
Nomenclature
AE/A0 CTH D J KQ KT n P PD Q Blade area ratio Trust loading coefcient Propeller diameter (m) Advance coefcient Torque coefcient Thrust coefcient Propeller rate of rotation per minute (RPM) Propeller pitch (m) Delivered power (kW) Propeller torque (kNm)
Thrust (kN) Total resistance (kN) Advance speed (m/s) Ship speed (knot) Taylor wake fraction Number of propeller blades Open water efciency Pitch ratio Density of water (kg/m3) Constant number Equation coefcients
62(2011)2, 123-129
123
S. EKINCI
1 Introduction
In ship hydrodynamics, xed pitch propellers, also named screw propellers, have an important place among the propulsion systems to propel a ship. The screw propellers, which showed up in the 19th century, have still secured their position as the best suitable propulsion system since that time. Although there have been signicant developments in both the propeller design and the propulsion systems in this long period of time, any changes in the main concept of screw propellers was observed. It is seen that these propellers will be used for longer periods of time due to their high efciency and suitable use. The aim in the propeller design is to obtain the optimum propeller which applies to minimum power requirements and against maximum efciency conditions at an adequate revolution number. Usually two methods are used in the propeller design. The rst is to use diagrams obtained from open water propeller experiments for systematic propeller series. The second is to use mathematical methods (lifting line, lifting surface, vortex-lattice, BEM (boundary element method)) based on circulation theory. After 1950, due to the developments in the computer technology, great improvements were seen in the second method, above mentioned, for the design and analysis of propellers [1-11]. In the past ten years, three new developments have appeared in the design and analysis of propellers. These are CFD methods (RANS solvers), high speed camera techniques, and PIV techniques [12-17]. In the rst stage of propeller design, usually the open water experiment diagrams of systematic model propeller series are used. These series consist of propellers whose blade number (Z), propeller blade area ratio (AE/A0), pitch ratio (P/D), blade section shape and blade section thickness are varied systematically. The most known and used propeller series is the Wageningen (Troost) series. Besides this series, the Gawn (Froude) series, Japanese AU series, KCA series, Lindgren series (Ma-series), Newton-Rader series, KCD series, Gutsche and Schroeder controllable pitch propeller series, Wageningen nozzle propeller series, JD-CPP propeller series are also mentioned in literature through various studies [18]. When examining the studies about propeller design for the past ten years, it is seen that advantage of the improvements in the computer technology has been utilized. Tanaka and Yoshida in [19] developed a computer program for propeller designers which transforms the dimensionless tables obtained from propeller series experiments into numerical graphics with great accuracy. In a similar study a computer program has been developed by Koronowicz et.al. for the design calculations for a propeller which is analyzed in the real velocity environment [20]. In this study, calculations considered are the scale effects in the velocity eld where the propeller is operating, corrections in the velocity eld due to the rudder, maximization of the propeller efciency, optimization of accurate blade geometry in terms of cavitation and strength, optimization of blade geometry depending on induced pressure forces and the blade numbers. A multipurpose optimization method is made by Benini in [21] for the Wageningen B propeller series using an algorithm for maximizing both; the thrust and torque coefcients under a constraint determined according to cavitation. Unlike the classical lifting line methods, Celik and Guner in [22] suggested an improved lifting line method by modelling the ow deformation behind the propeller with free vortex systems. In his study Olsen in [23] developed a method to
calculate the propeller efciency with the help of energy coefcients including the propeller loss. He compared his ndings with the results obtained from vortex-lattice method. Hsin et.al. in [24] applied in their study a method derived from the adjoint equation of the nite element method to two propeller design problems. Matulja and Dejhalla in [25] realized the selection of the optimum screw propeller geometry with articial neural networks. Roddy et.al. in [26] used the articial neural network method for the prediction of thrust and torque values of the Wageningen B series propeller. Chen and Shih in [27] realized the propeller design by the use of the Wageningen B series propellers by considering the vibration and efciency characteristics using genetic algorithm. A similar study was made by Suen and Kouh in [28]. In this study a practical design approach is presented by using the Wageningen B series propellers for a case where the delivered power (PD), the advance speed (VA) and the revolution number (n) is known. A set of propellers suitable for a wide loading range is developed by the use of polynomials representing the open water diagrams of the Wageningen B propeller series. The set of propellers consists of four-bladed propellers and is developed in such a manner that the whole range of the blade area ratios (AE/A0) and pitch ratios (P/D) of the Wageningen B series are included. The effects of the Reynolds number are not included in the calculations. The propeller design and performance characteristics are presented by means of empirical formulas and diagrams to propeller designers as a practical tool for the use in the preliminary design stage. Besides the propeller design, the diagrams and empirical formulas can also be used for the propeller performance prediction.
(1) (2)
124
62(2011)2, 123-129
S. EKINCI
Thrust coefcient: T KT = 2 4 n D Torque coefcient: KQ = Advance coefcient: v J= A nD Open water propeller efciency: KT J KQ 2 Propeller thrust loading coefcient: (5) Q n2 D5 (4) (3)
The torque requirement of a propeller can be expressed as a V function of the advance coefcient J = A and extracting the nD diameter (D) from this formula, substituting it in the expression Q K Q = 2 5 and nally rearranging the equation the expression n D below can be obtained: KQ = k= PD n 2 5 J = k.J 5 2VA5 (8)
PD n 2 = constant 2VA5
(9)
o =
(6)
CTH =
8 KT J2
(7)
The Wageningen B propeller series is a general purpose series. This series is expressed with open water diagrams obtained from model tests where the KT-KQ-J curves are showed for propellers with constant blade number (Z) and constant blade area ratio (AE/A0) but variable pitch ratios (P/D). Because the open water experiments are made in fresh water, this must be considered in the design calculations. The Wageningen B series propellers are extensively used for the design and analysis of xed pitch propellers.
For the cases where the design variables (PD, n, VA), the k expression (9), are known, the propeller design is carried out as follows. Firstly, the torque requirement curve KQ-J obtained according to Equation (8) is drawn over the Wageningen B open water diagram. The intersection points of this curve which express the torque requirement of the propeller and the KQ curves of different P/D values on the open water diagram describe the possible design solutions. The optimum efciency curve is obtained by drawing vertical lines from the intersection points to the efciency curves. And the maximum point of this curve, which represents the most efcient propeller among the different solutions satisfying the requirements, is read-off. Later the J, P/D, KT, KQ and o values of the optimum propeller are read-off. A computer code based on polynomials of the Wageningen B series is used in the applications of the design method. The main propeller data used for generating the set of propellers for each AE/A0 (0.4, 0.55, 0.70, 0.85, 1.0) are given in Table 1. The set of propellers is generated by considering the PD and n values of the main propeller constant and by changing the advance speed VA (2.5-22.5 knots, with 0.5 step size, in total 39 values) to cover the whole P/D range (0.5-1.4) of the Wageningen B series. For this condition the propeller design is based only on the value of k, so the main propeller data are used to generate different values of k including all probable propeller cases of 4-bladed Wageningen B series. And, P/D, J, KT, KQ and o values due to k are presented in relation to k in Figures 1-5.
Table 1 Main propeller design input data Tablica 1 Glavni ulazni podaci za projekt vijka
Delivered power, PD (kW) Advance speed, VA (m/s) Propeller rate of rotation per minute, RPM Propeller diameter, D (m) Number of blades, Z Blade area ratio, AE/A0
Since the pitch ratio (P/D), advance coefcient (J), thrust coefcient (KT), torque coefcient (KQ) and propeller efciency (o) are given as the function of k in the graphics above, these can be also expressed in a general polynomial form (10). The expressions of the curves tted to the points so as to ignore the
62(2011)2, 123-129
125
S. EKINCI
1.6
AE/A0=0.4
0.5
AE/A0=0.4 AE/A0=0.55
1.4
AE/A0=0.55
AE/A0=0.7
0.4
1.2
AE/A0=0.85
AE/A0=1.0
0.3
0.8
0.2
0.6
0.1
0.4 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
0.5
1.5
2.5
3.5
4.5
k0.2
k0.2
Figure 1 Variation of non dimensional pitch ratio (P/D) in relation to k Slika 1 Promjena bezdimenzijskog omjera uspona i promjera vijka (P/D) u odnosu na k
Figure 4 Variation of torque coefcient (KQ) in relation to k Slika 4 Promjena koecijenta momenta uvijanja (KQ) u odnosu na k
0.8
1.2
AE/A0=0.4
AE/A0=0.4
0.7
AE/A0=0.55 AE/A0=0.7
0.6
Advance coefficient ( J )
0.8
AE/A0=0.85
0.5
AE/A0=1.0
0.6
AE/A0=1.0
0.4
0.4
0.3 0.2
0.2
0.1
0.5
1.5
2.5
3.5
4.5
k0.2
Figure 2 Variation of advance coefcient (J) in relation to k Slika 2 Promjena koecijenta napredovanja (J) u odnosu na k
Figure 5 Variation of open water efciency (0) in relation to k Slika 5 Promjena korisnosti u slobodnoj vonji (0) u odnosu na k
AE/A0=0.55 AE/A0=0.85
Thrust coefficient (K T)
0.19 0.18 0.17 0.16 0.15 0.14 0.13 0.5 1 1.5 2 2.5 3 3.5 4 4.5
k0.2
Figure 3 Variation of thrust coefcient (KT) in relation to k Slika 3 Promjena koecijenta poriva (KT) u odnosu na k
also be executed by reducing the reading errors to a minimum with the use of (10). For the intermediate values of the blade area ratio the interpolation methods can be used. The other propeller parameters (D, T) can be obtained by (J) and (KT) expressions given in (3), (5). In addition to propeller design, these equations and graphics can also be interpreted as useful and timesaving tools for the prediction of the performance characteristics of an existing 4bladed propeller. The values of the coefcients in (10) with the blade area ratio (AE/A0) are given in Table 2. F ( k ) = k 0.2 (ak + bk 4 /5 + ck 3/5 + dk 2 /5 + ek 1/5 + f ) + g (10) PD n 2 ; F(k), function of pitch ratio (P/D), advance 2VA5 coefcient (J) etc. Here; k =
effects of oscillations are determined by using the least square method (LSM). While the design of a propeller with the given PD, n, VA values of k can be carried out with the graphics, it can
126
62(2011)2, 123-129
S. EKINCI
Table 2 Coefcients of the F(k) equation with blade area ratio (AE/A0) Tablica 2 Koecijenti iz jednadbe F(k) s odreenom vrijednosti omjera povrina krila (AE/A0)
AE/A0=0.4 F(k) P/D J KT 10KQ a 0.0469 0.0182 0 0.0236 -0.003 a 0.0437 0.0339 0 0.0205 -0.0009 a 0.0428 0.0323 0 0.0197 -0.0011 a 0.029 0.0263 0 0.0089 -0.002 a 0.0112 0.0182 0 -0.006 -0.0042 b -0.658 -0.2659 0 -0.3287 0.0056 b -0.6128 -0.4778 0 -0.284 0.013 b -0.603 -0.4573 0 -0.2735 0.0154 b -0.4161 -0.3758 0 -0.1301 0.0283 b -0.1747 -0.2659 0 0.0723 0.0574 c 3.6709 1.5607 0.0053 1.818 d -10.402 -4.7324 -0.0535 -5.0795 e 15.846 7.9103 0.1875 7.556 0.0085 e 14.614 11.878 0.1156 6.3058 -0.157 e 14.462 11.541 0.0608 6.0542 -0.1394 e 10.997 9.9928 0.0022 3.5196 -0.3993 e 6.3924 7.9103 -0.1054 -0.3805 -0.9192 f -12.53 -7.1721 -0.25914 -5.7157 -0.4465 f -11.494 -9.7354 -0.1477 -4.6823 -0.3042 f -11.327 -9.5054 -0.0664 -4.4478 -0.3806 f -9.1227 -8.5175 -0.001 -2.866 -0.2119 f -6.0879 -7.1721 0.1517 -0.2892 0.1208 g 4.7135 3.166 0.2633 1.9072 0.9475 g 4.3615 3.7278 0.2038 1.5636 0.9246 g 4.312 3.6688 0.1719 1.4905 0.9324 g 3.8503 3.4603 0.161 1.1633 0.8802 g 3.175 3.166 0.1183 0.5878 0.7864
0
F(k) P/D J KT 10KQ
0
F(k) P/D J KT 10KQ
0
F(k) P/D J KT 10KQ
-0.0331 0.0794 AE/A0=0.55 c d 3.4116 -9.636 2.6798 -7.6654 0.0036 -0.0345 1.5572 -4.3023 -0.0747 0.1959 AE/A0=0.70 c d 3.3695 -9.53597 2.5775 -7.4106 0.002 -0.0192 1.5028 -4.1506 -0.0837 0.2046 AE/A0=0.85 c d 2.3818 -6.9566 2.1437 -6.2669 -0.0005 0.0016 0.7555 -2.2266 -0.1537 0.3935 AE/A0=1.0 c d 1.1005 -3.5816 1.5607 -4.7324 -0.0029 0.0296 -0.3214 0.6202 -0.3057 0.7859
0
F(k) P/D J KT 10KQ
The practical design approach presented in this work allows the design of a four-bladed Wageningen B screw series propeller or the performance prediction of an existing propeller based on just only the k value given as in (9) and the blade area ratio. The present approach is considered as a practical tool for propeller designers for use during the preliminary design stage as diagrams and empirical formulas.
4 Design applications
For verication and to show usability of the presented approach, three propeller designs for medium, low and high thrust
loading cases were realized. Various loading conditions were provided by changing only the revolution number of the propeller to data as given in Table 3. From the design variables of these three propellers, with the use of the open water curves of the Wageningen B screw series and the above presented empirical formulas, the propeller designs were realized. The obtained propeller design and performance results (P/D, KT, KQ, J and 0) are presented in Table 4. The design results of the three propellers with different loading conditions are compared with those of KT and KQ polynomials for the 4-bladed Wageningen B screw series developed in [31], and it is seen that the order of the error values is in an acceptable range.
62(2011)2, 123-129
127
S. EKINCI
Propeller rate of rotation per minute (RPM) Delivered power (kW) Ship speed (knot) Taylor wake fraction Number of propeller blades (Z) Blade area ratio (AE/A0)
Table 4 Comparison of the propellers design results Tablica 4 Usporedba rezultata za projektirane vijke
Propeller 1 k=0.10025 CTH=0.6396 Open Water Diagram 1.0000 6.0000 0.7650 0.1470 0.2630 0.6780 Present Method 1.0032 5.971 0.7687 0.1497 0.2637 0.6762 % Error 0.3183 0.483 0.4854 1.8367 0.2616 0.2602 Open Water Diagram 0.8100 4.1129 0.5580 0.1480 0.2170 0.6050
Propeller 2 k=0.4010 CTH=1.210 Present Method 0.7946 4.21875 0.5440 0.1486 0.2078 0.6024 % Error 1.9049 2.573 2.5099 0.4338 4.2392 0.4281 Open Water Diagram 0.7300 3.28326 0.4660 0.1480 0.1980 0.5550
Propeller 3 k=0.90223 CTH=1.7355 Present Method 0.7265 3.36263 0.4555 0.1490 0.1960 0.5540 % Error 0.4727 2.41 2.2519 0.6703 1.0227 0.1883
P/D D J KT 10KQ
5 Results
Although the propeller designs can be made with computer programs based on circulation theory, the conventional design method using propeller series based on model experiments remains the most applied method. The primary advantage of this method is the practice of the design with a few design variables and the availability of the performance prediction. In this study a practical design approach is presented for the four-bladed Wageningen B series propellers for the cases where the delivered power (PD), the advance coefcient (VA) and the propeller revolution number (n) are known. Further, this method can be applied in a rapid and accurate manner by the use of the given diagrams and empirical formulas as an alternative method to design with Bp-delta diagrams. Additionally, the prediction of the performance characteristics that is not possible with Bp-delta diagrams for an existing propeller can also be available by the present approach. In this study the design applications for three propellers with different thrust loading conditions are taken into account for the presented method. The results obtained from this method are compared with those of the open water diagrams, and a good agreement has been found. The presented method can be expanded to the remaining Wageningen B series (propellers other than the 4-bladed ones) or to other propeller series. A further aim is to present in the future similar diagrams and empirical expressions for practical propeller
design approaches for different propeller conditions for various given design variables.
References
[1] LERBS, H.W.: Moderately Loaded Propellers with a Finite Numbers of Blades and an Arbitrary Distribution of Circulation, SNAME Trans., Vol. 60 (1952). [2] ECKHARDT, M.K., MORGAN, W.B.: A propeller design method, SNAME Trans. Vol. 63 (1955). [3] GLOVER, E.J.: A Design Method for Heavily Loaded Marine Propellers, RINA Trans. Vol. 116 (1974). [4] SZANTYR, J.: An interactive program for design of ship propellers, The Naval Architect (1987). [5] GUNER, M.: A Rational Approach to the Design of Propulsors behind Axisymmetric Bodies, Ph.D. thesis, University of Newcastle upon Tyne, England (1994). [6] GREELEY, D.S., KERWIN, J.E.: Numerical methods for propeller design and analysis in steady ow, SNAME Trans. Vol. 90 (1982). [7] PIEN, P.C.: The Calculation of Marine Propellers Based on Lifting Surface Theory, Journal of Ship Research, No. 2 (1961). [8] KERWIN, J.E., TAYLOR, T.E., BLACK, S.D., McHUGH, G.P.: A Coupled Lifting Surface Analysis Technique for Marine Propulsors in Steady Flow, Proceedings of Propeller/Shafting 1997 Symposium, Virginia, (1997).
128
62(2011)2, 123-129
S. EKINCI
[9] LEE, H., KINNAS, S.A.: A BEM for the Modeling of Unsteady Propeller Sheet Cavitation Inside a Cavitation Tunnel, Journal of Computational Mechanics, Vol. 37, No. 1, p. 41-51 (2005). [10] GHASSEMI, H.: The Effect of Wake Flow and Skew Angle on the Ship Propeller Performance, Scenta Iranca Transacton B-Mechancal Engneerng, Vol. 16 No. 2, p. 149-158 (2009). [11] BAL, S., GUNER, M.: Performance of Podded Propulsors, Ocean Engineering, Vol. 36 No. 8, p. 556-563, (2009). [12] SANCHEZ-CAJA, A., RAUTAHEIMO, P., SIIKONEN, T.: Simulation of Incompressible Viscous Flow Around a Ducted Propeller Using a RANS Equation Solver, TwentyThird Symposium on Naval Hydrodynamics, (2001). [13] KULCZYK, J. S., SKRABURSKI, L., ZAWISLAK, M.: Analysis of Screw Propeller DTMB 4119 using the FLUENT System, Archives of Civil and Mechanical Engineering, Vol. 7, No. 4, p. 129-136 (2007). [14] ARAZGALDI, R., HAJILOUY, A., FARHANIEH, B.: Experimental and Numerical Investigation of Marine Propeller Cavitation, Scientia Iranica Transaction B-Mechanical Engineering, Vol. 16, No. 6, p. 525-533 (2009). [15] CELIK, F., GUNER, M., EKINCI, S.: An Approach to the Design of Ducted Propeller, Scientia Iranica Transaction B-Mechanical Engineering, Vol. 17, No. 5, p. 406-417 (2010). [16] TUKKER, J., KUIPER, G.: High Speed Video Observations and Erosive Cavitation, PRADS Symposium, (2004). [17] KUIPER, G.: New Developments and Propeller Design, 9th International Conference on Hydrodynamics, Shanghai, China, (2010). [18] CARLTON, J.S.: Marine Propellers and Propulsion, Second Edition, Published by Elsevier, (2007). [19] TANAKA, H., YOSHIDA, Y.: An Approach to Transformation of Non-Dimensional Diagrams into Nomograms and its Application to New Propeller Design System, Marine Technology V, Book Series: Marine and Maritime, Vol. 3, p. 91-102, (2003).
[20] KORONOWICZ, T., CHAJA, P., SZANTYR, J.: A Computer System for the Complete Design of Ship Propellers, Archives of Civil and Mechanical Engineering, Vol. 8, No.1, p. 49-59 (2008). [21] BENINI, E.: Multiobjective Design Optimization of BScrew Series Propellers Using Evolutionary Algorithms, Marine Technology and SNAME News, Vol. 40, No. 4, p. 229-238 (2003). [22] CELIK, F., GUNER, M.: An Improved Lifting Line Model for the Design of Marine Propellers Marine Technology and SNAME News, Vol. 43, No. 2, p. 100-113 (2006). [23] OLSEN, A.S.: Energy Coefcients for a Propeller Series, Ocean Engineering, Vol. 31, No. 3-4, p. 401-416 (2004). [24] HSIN, C.Y.: Application of the Adjoint Method to the Propeller Designs, Journal of Hydrodynamics Ser. B, Vol. 22, No. 5, Supp.1, p. 501-506 (2010). [25] MATULJA, D., DEJHALLA, R.: Neural Network Prediction of an Optimum Ship Screw Propeller, Annals of DAAAM and Proceedings, p. 829-830 (2008). [26] RODDY, R.F., HESS, D.E., FALLER, W.E.: Neural Network Predictions of the 4-Quadrant Wageningen BScrew Series, p. 315-335, COMPIT06, (2006). [27] CHEN, J.H., SHIH, Y.S.: Basic Design of Series Propeller with Vibration Consideration by Genetic Algorithm, Journal of Marine Science and Technology, Vol. 12, No. 3, p. 119-129 (2007). [28] SUEN, J.B., KOUH, J.S.: Genetic Algorithms for Optimal Series Propeller Design, Marine Technology III Book Series: Marine and Maritime, Vol. 1, p. 237-246 (2000). [29] TROOST, L.: Open Water Tests Series with Modern Propeller Forms, Trans. NECIES, Vol. 54 (1938). [30] LAMMAREN, W.P.A. van, MANEN, J.D. van, OOSTERVELD, M.W.C.: The Wageningen B-Screw Series, SNAME Trans. (1969). [31] OOSTERVELD, M.W.C., OSSANNEN, P. van: Further Computer Analysed Data of the Wageningen B-Screw Series, ISP, Vol. 22, No. 251 (1975). [32] BURRILL, L.C.: Calculation of Marine Propeller Performance Characteristics, Trans N.E .C.I.E.S. Vol. 60 (1943-44).
62(2011)2, 123-129
129