Paper Hiswa v5 THE MARIN SYSTEMATIC SERIES FAST DISPLACEMENT HULLS
Paper Hiswa v5 THE MARIN SYSTEMATIC SERIES FAST DISPLACEMENT HULLS
Paper Hiswa v5 THE MARIN SYSTEMATIC SERIES FAST DISPLACEMENT HULLS
Summary
The paper describes the development of a systematic series of high speed displacement hull
forms. The series is built around a parent model by varying the length to beam ratio, the beam
to daft ratio and the block coefficient. Eventually thirty-three hull forms were built and tested in
calm water up to a maximum speed of Froude number 1.14. Twenty-four of these hull forms
were also tested in head waves. Regression analysis has been done on the experimental results
to allow interpolation for intermediate speeds or intermediate hull forms.
All results of the experiments are now made available in the form of a computer program; the
program also contains the regression equations to allow interpolation. The use of the program is
illustrated by way of a case study.
Introduction
A Joint Industry Project (JIP) on high speed displacement ships was carried out in the period
1979 – 1989 by the Royal Netherlands, US and Australian Navies, Delft University of Technology
and MARIN. Western navies were very interested in high speed vessels in the seventies and they
studied all sorts of advanced concepts ranging from hydrofoil to air cushion supported vessels
and even wing-in-ground effect vehicles. Performance and endurance at high speed in waves
was an issue for most of the studied types, so some people believed that also the monohull
concept could be a viable alternative. This belief was the starting point of the JIP and it was
decided to develop a systematic series around a parent hull form. The main parameters to vary
in the series were length over beam ratio, beam over draft ratio and block coefficient and the
FDS - The MARIN Systematic Series 2
speed range to be considered should reach Fn = 1. Contrary to most existing systematic series,
not only resistance characteristics in calm water was to be measured, but also performance in
waves. Head seas was considered the most severe wave direction and therefore experiments
were restricted to this condition. Most of the models were tested both in calm water and in
head seas, the latter at fixed Froude numbers of Fn = 0.285, 0.430, 0.570, 0.855 and 1.140.
Now, 25 years after closing the project, the results were de-classified by all the partners. It was
considered that the results of the project are still valuable today, not only for the naval market,
but also for the market of the (mega) yachts. Because of this and with the help of a subsidy of
the Dutch Ministry of Economic Affairs (Maritime Innovation Platform) a project was started to
make all results available. The format chosen was to make a PC program for numerical results
and to write a book on the why, how and what of the project. The PC program contains all the
hull forms, the experimental data and the small programs to interpolate in the existing hull
forms (geometry) and to predict the hydrodynamic performance for this newly generated hull
form. The book will explain in detail the background of the project, how all the work was done
and analysed and how the interpolation methods were developed. Some additional topics were
also investigated; some of these are mentioned in this paper, all of them are described in the
book.
This paper gives an overview of the early phase of the project were the parent hull form was
selected, an overview of all the work done and finishes with a case study showing how the
results might be used in a design office.
motions in waves. This hull form was therefore selected as the parent of the systematic series.
The selection procedure is described in more detail by Blok and Beukelman [1984].
Table 1 Relative difference (to model 2) of calm water resistance and various seakeeping
aspects of the six models at a speed of Fn = 0.855
Calm Rel Vert
water Heave Pitch motion accel St Rcalm +
model resistance motion motion St 17 19 Raw Raw
1 +0.1 +7.7 +7.6 +6.2 +3.7 +1.0 -1.0
2 0 0 0 0 0 0 0
3 +5.2 0 -10.6 -9.6 -7.4 -13.6 +4.0
4 -0.4 +2.6 +7.6 +11.9 +3.7 +1.0 0
5 +1.8 +7.7 +10.6 +4.5 +7.4 -3.9 +1.4
6 +4.5 +5.2 +4.5 +7.9 0 -9.6 +3.6
Figure 1 Schematic diagram showing the hull forms tested to select the parent hull form.
FDS - The MARIN Systematic Series 4
0.08 0.08
model 1 model 4
model 2 model 5
0.06 0.06
model 3 model 6
0.04 0.04
0.02 0.02
0.00 0.00
0 1 2 3 4 0 1 2 3 4
Froude displacement number [-] Froude displacement number [-]
1.2 1.2
model 1 model 4
1 model 2 1 model 5
model 3 model 6
0.8 0.8
Pitch RAO [-]
Pitch RAO [-]
0.6 0.6
0.4 0.4
0.2 0.2
0 0
0.5 1 1.5 2 2.5 3 0.5 1 1.5 2 2.5 3
wave frequency [rad/s] wave frequency [rad/s]
of z0. This last aspect involved some manual intervention of the transformation procedure to
arrive at a fair hull form. Figure 4 shows an example of a normal transformation of a section in
the fore body and of a section in the aft body where the contour line had to be changed. Figure
4 also shows that the width at the design waterline, especially in the fore body, changes as a
function of the block coefficient; this effect was accepted as an unavoidable consequence of the
transformation. The main hull form coefficients of 3 models differing in block coefficient are
shown in Table 2; the table shows the magnitude of the changes in the parameters for the
water line.
Figure 4 Example of the geometry transformation method for a section in the fore body (left)
and a section in the aft body (right) where the contour line was lowered.
Using this method a systematic series of 27 models was built. In order to identify the models,
use was made of the fact that just three parameters completely define the shape: L/B, B/T and
cB. The hull forms were represented in the so-called “magic cube”, Figure 5, in which each dot
indicates a hull form.
FDS - The MARIN Systematic Series 6
Figure 5 The “magic cube” showing the hull forms of the systematic series in black dots.
Because the project ran for a period of 10 years, inevitably the ideas about interesting “design
areas” changed. The initial plan involved three “levels” of cB, but later on a fourth level at cB =
0.55 was added to extend the applicability also to larger destroyer type of hull forms. In a very
late stage it was realized that the step from L/B = 4 to L/B = 8 was a very large one, especially
concerning the resistance characteristics. A fourth L/B plane at L/B = 6 was added and a series of
models was designed, built and tested in calm water; not in waves.
Experiments
Experiments in calm water have been carried out on an unappended hull. Turbulence
stimulation at the bow has been applied as customary by way of pins. The models were towed at
the longitudinal position of the CoG and at a height corresponding to the supposed propulsion
drive train. Calm water resistance, trim and sinkage were measured for speeds up to Fn = 1.14
unless excessive trim and spray prohibited such high speeds. Propulsion tests were not carried
out.
Motions and added resistance in regular head waves were measured using a semi-captive set-
up: the model was connected at the CoG to the carriage by way of a gimballed heave post.
Measurements were done at a series of fixed values of the wave length – ship length ratio for a
fixed series of speeds: Fn = 0.285, 0.430, 0.570, 0.855 and 1.140. Again the tests in waves could
not all be done at the highest speeds due to excessive trim and spray; this was especially the
case for the low L/B hull forms.
HISWA Yacht Symposium 7
Interpolation methods
In the final stages of the project the data was further analysed. The idea was to determine the
trends on resistance in calm water and on motions in waves based on the 3 hull form
parameters and, for the results in waves, also based on the fixed values of the Froude number.
This would then allow the prediction of the performance of intermediate hull forms based on
the actually measured values.
For the calm water resistance (and the trim and sinkage) it appeared that regression using the
three hull form parameters as in the independent variables for a series of fixed values of the
Froude-displacement number Fn gave the most accurate results; an example of the accuracy
is given in Figure 6.
For the motions and added resistance in head waves the regression was carried out on the
speeds for which the experiments were done. Also fixed values of a non-dimensional wave
frequency parameter have been used so that the independent variables of the regression are
just the hull form parameters. This approach resulted in reasonably accurate predictions, but the
results, motion RAOs or the quadratic added resistance operator, as a function of the wave
frequency, are not necessarily smooth lines. An example of the result of the interpolation
formulas is shown in Figure 7.
3500 2
Experiment 1.8 Exp - Fn = 0.285
3000
Regression 1.6 Regr - Fn = 0.285
Total Resistance [kN]
Additional studies
Apart from the experiments as described above some additional work has been carried out:
A study into the effect of transom stern wedges for a limited series of hull forms.
A study into the magnitude of the hull roll damping and the effect of bilge keels at these
high speeds.
A check on the choice of the prismatic coefficient by testing two additional hull forms.
Regression analysis for the resistance of an extended database of high speed
displacement hull forms.
FDS - The MARIN Systematic Series 8
0.12 0.11
wa=0 deg
0.1 wa = 5 deg
0.10
Residual resist. coef [-]
0.06
wa=0 deg 0.08
0.04
wa = 5 deg
Figure 8 Residual resistance coefficient for the L/B = 4, B/T = 4, cB = 0.40 model with a wedge,
length 0.02*Lpp, with different wedge angles (wa). Full speed range (right) and zoom-in (left).
Roll damping
As indicated in the introduction, focus of the experiments was on calm water performance and
behaviour in head seas. Nevertheless in the course of the 10-year project attention was paid to
the prediction of the roll motion; more specific the prediction of the proper roll damping for
these high speeds. Roll is a notorious problem; the damping is usually quite low – much lower
than the critical damping. The damping increases significantly for increasing speed; the relatively
flat aft body of these designs create lift forces that dominate the roll damping at high speeds.
Traditional methods to estimate the roll damping, like the one developed by Ikeda (1978) and
Himeno (1981), are based on model experiments with merchant ships. The combination of the
frigate type hull forms and the high speeds of this systematic series makes the traditional
method very inaccurate.
It was therefore decided to develop a new formula to predict the lift damping. A special series of
experiments was designed using an oscillator mechanism to force the model into a roll motion
while it was still free in the heave and pitch modes. Experiments were carried out with a series
of 13 models covering the B/T range in a band from the low cB – low L/B corner to the high cB –
high L/B corner of the “magic cube”. Most of the models were tested with and without bilge
keels; these bilge keels had a length of 25%*Lpp and a height of 1% of Lpp. This work resulted in
a new formula for the roll damping which gave much better predictions of the roll motion at high
HISWA Yacht Symposium 9
speeds as was verified by a series of free running experiments. A detailed description of this
work was published by Blok and Aalbers (1991).
Figure 9 Roll response of the parent hull form in beam seas at different speeds. Results of
calculations using the new method to estimate the roll damping are shown in lines, experimental
results by symbols.
0.1 1.4
Cp = 0.69
1.2
0.08 Cp = 0.63
Heave RAO [-]
Residual resist. coef [-]
1
Cp = 0.56
0.06 0.8
0.04 0.6
Cp = 0.69
0.4
0.02 Cp = 0.63
0.2
Cp = 0.56
0 0
0 1 2 3 4 0 0.5 1 1.5 2 2.5 3
Froude displacement number [-] wave length - ship length ratio [-]
Figure 10 Residual resistance / displacement Figure 11 Heave RAO at Fn = 0.57 for the
for the prismatic coefficient sub-series. prismatic coefficient sub-series.
FDS - The MARIN Systematic Series 10
the NPL High Speed Round Bilge Displacement Hull Series, Bailey (1976), 24 hull forms.
the David Taylor Research Centre Series 64, Yeh (1965), 27 hull forms.
the NRC series, Murdey and Simoes Re (1985), 39 hull forms.
Non-systematic data from the MARIN database, selected for high speed round bilge hull
forms, 36 hull forms.
The MARIN systematic series, 33 hull forms.
In total there were 159 hull forms in this database. The regression of the resistance yields a
specialized tool for fast displacement ships that is much more general than the regression of the
systematic series only.
A case study
In order to illustrate the possibilities using the data and the computer program, a case study is
presented. The base design for the case study is a motor yacht with main dimensions as listed in
Table 3. The hull form belongs to the FDS family. The basis results of the calm water resistance,
powering and motions in head waves are presented in Figure 12.
Table 3 Main dimensions and hull form ratios of the motor yacht
Figure 12 shows that the major resistance component is the residual drag for speeds higher than
10 kts. There is a clear hump for speeds 17 – 21 kts after which the relative importance of the
residual resistance decreases. Figure 13 shows the heave motion in head seas; the peak of the
RAO increases for increasing speed and shifts to lower frequencies. The driving parameter
behind this is the heave excitation which increases for longer waves. Those longer waves are
being encountered at a higher frequency when the speed increases. Since this higher frequency
is closer to heave resonance, the response increases. The wave excitation for pitch has a
maximum for a wave length about equal to the ship length (corresponding to a wave frequency
of 1.4 rad/s); this is the main reason why the response shows a maximum around this frequency.
HISWA Yacht Symposium 11
Figure 12 Calm water resistance of the motor yacht showing the Total, Frictional, Wave
making, Appendage and Wind drag components. The plot on the left gives the absolute values,
the plot on the right gives the relative contributions (plotted in a cumulative manner).
1.6 10
Vs = 10 kts
1.4 Vs = 10 kts
8 Vs = 20 kts
1.2 Vs = 20 kts
Vs = 30 kts
Pitch RAO [deg/m]
Heave RAO [-]
1 Vs = 30 kts
6
0.8
0.6 4
0.4
2
0.2
0 0
0.5 1 1.5 2 2.5 0.5 1 1.5 2
wave frequency [rad/s] wave frequency [rad/s]
Figure 13 Motions in head seas; heave (left) and pitch (right). Results for 10, 20 and 30 kts.
The case study concerns the effect of a variation of the beam and draft while keeping the length
and displacement constant. If the beam is changed independently of the draft, the block
coefficient has to be changed in order to keep the displacement constant. A matrix can be set up
with the B/T ratio on one axis and the cB on the second axis showing either the beam or the
draft, see Table 4 as an example. This table shows that the variation of the beam is quite large
and also the draft varies a lot more (from 1.05 to 1.95 m) than might be possible in real life.
However limits to the ranges can always be added, even after doing the calculations.
Calculations were carried out for a matrix of 100 hull forms, 10 steps on the B/T axis and 10 on
the cB axis. The results considered are the total resistance at the design speed and the significant
heave and pitch motion in the “design” sea state with significant wave height HS = 2.0 m and
average period T1 = 6 s. The results are shown in Figure 14. The upper left hand graph shows that
there is a maximum difference of 24% between the total resistance of the best and the worst
FDS - The MARIN Systematic Series 12
design. The optimum of the design space is the high block coefficient - low B/T corner. This
design has a narrow beam and, since the length is constant, the highest L/B ratio; this design also
has the lowest wetted surface, both aspect help in reducing the resistance. The diagram further
shows that the penalty in resistance when increasing the B/T ratio is not very large; the total
resistance of the cB = 0.55, B/T = 5.5 hull form is only 5% higher than that of the optimum hull
form.
Table 4 Table showing the beam of the motor yacht as a function of the B/T ratio and the cB
B/T
2.5 3.0 3.5 4.0 4.5 5.0 5.5
cB
0.35 4.877 5.343 5.771 6.169 6.543 6.897 7.234
0.40 4.562 4.997 5.398 5.771 6.121 6.452 6.767
0.45 4.301 4.712 5.089 5.441 5.771 6.083 6.380
0.50 4.080 4.470 4.828 5.161 5.474 5.771 6.052
0.55 3.891 4.262 4.603 4.921 5.220 5.502 5.771
Figure 14 Results in calm water and in waves of the case study for the 30 m motor yacht
HISWA Yacht Symposium 13
The motions in waves and the added resistance are also shown in Figure 14. The optimum is
clearly different to the optimum for the calm water resistance. The ship motions show that
having a low value of the B/T ratio is not very good for comfort aspects. Unfortunately the
corner in the design space having the highest resistance in calm water also has the highest pitch
motion! It seems to be a good compromise to aim for a high B/T and a high cB. The price to pay is
the high added resistance in waves, in the design sea state of HS = 2 m it adds 37% to the calm
water resistance. One wonders if this “natural” speed reduction is not a very sensible thing to do
anyway in such a seaway...
This case study shows how the PC program can be used; it can quickly show the consequences of
quite dramatic changes in the hull form in a very early stage of the design process. It also shows
that naval architecture is about making choices, about finding the right balance between
conflicting requirements for a particular design. This balance is dependent on the preference of
the owner of the yacht, the use of the yacht and about the wave climate in which the yacht has
to operate.
Concluding remarks
As explained in the introduction, no real results of the project were published because of the
confidentiality of the project. Permission was granted for some overview publications: Oossanen
(1980), Oossanen and Pieffers (1985) and Robson (1987), and for some publications on specific
details of the project: Koops (1981), Blok and Beukelman (1984), Aalbers and Blok (1991),
Keuning (1990) and a chapter in the PhD thesis of Keuning (1994).
Now all results will be made available, this concerns the definition of the hull forms, the
experimental data and the interpolation routines. Together with the book explaining the details
of why and how the experiments and the analysis were carried out, we hope that the maritime
community will have a useful database of the hydrodynamic behaviour of high speed
displacement hull forms.
Nomenclature
B m Maximum beam on the water line
T m Draft amidships
Lpp m Length between perpendiculars
LcB m Longitudinal position of the centre of buoyancy forward of St 10
LcF m Longitudinal position of the centre of floatation forward of St 10
CoG m Longitudinal position of the centre of gravity forward of St 10
cP - Prismatic coefficient
cB - Block coefficient
Fn - Froude number based on length, Fn v g LPP
1
Fn - Froude number based on displacement Fn v g 3
m3 Displacement volume
FDS - The MARIN Systematic Series 14
References
1. Bailey, D., The NPL high speed round bilge displacement hull series, Maritime Technology
Monograph No. 4., Royal Institution of Naval Architects, 1976.
2. Bailey D., High speed displacement ships: trends in hull form design, Workshop on
developments in hull form design, Oct 1985, Wageningen.
3. Blok J.J. and Aalbers A.B., Roll damping due to lift effects on high speed monohulls, Proc.
1st Fast Conference, June 1991, Trondheim.
4. Blok J.J. and Beukelman W., The high speed displacement ship systematic series hull
forms - seakeeping characteristics, SNAME Annual Meeting, Nov 1984.
5. Himeno, Y., Prediction of Ship Roll Damping - State of the Art, Dept. of Naval Architecture
and Marine Engineering of the University of Michigan, Report 239, 1981.
6. Kehoe J.W., Destroyer Seakeeping: Ours and Theirs, US Naval Inst. Proc, Nov 1973.
7. Keuning J.A., Distribution of added mass and damping along the length of a ship moving
at high forward speed, Int. Shipb. Progress, Vol 37, No. 410, pp. 123-150, 1990.
8. Koops A., Hull form definition and Computer Aided Design, Proc. Comp Appl. In the
Automation of Shipyard Operation and Ship Design (ICCAS), Trieste, Italy, 1985
9. Murdey, D.C. and Butler J., Hull form series for Fast Surface Ships - summary of
seakeeping data, Inst. for Marine Dynamics, Report No. LTR-SH-393, Feb 1985.
10. Murdey, D.C. and Simoes Re, A.J., Hull form series for Fast Surface Ships - summary of
resistance data, Inst. for Marine Dynamics, Report No. LTR-SH-386, February 1985.
11. Murdey, D.C. and Simoes Re, A.J., The NRC Hull Form Series - an Update, Workshop on
developments in hull form design, Oct 1985, Wageningen, Netherlands
12. Oossanen P. van, Resistance prediction of small High-Speed Displacement Vessels: State
of the art, Int. Shipb. Progress, Vol 27, No. 313, Sep 1980.
13. Oossanen P. van en Pieffers J.B.M., NSMB-Systematic series of High Speed Displacement
Ship Hull Forms, Workshop on Development in Hull Form design, Oct 1985, Wageningen.
14. Robson B.L., Systematic series of high speed displacement hull forms for naval
combatants, RINA Australian Division, Sep 1987.
15. Versluis A., Transformation of ship forms, (in Dutch) Report No. 438, Technical University
Delft, Aug 1976.
16. Yeh, H.Y.H., Series 64 resistance experiments on High-Speed Displacement Forms, Marine
Technology, July 1965.