Section 3 Model experiments

3.1.1 Model experiments should generally follow the procedures recommended by the ITTC (International Towing Tank Conference) for conduct of sea keeping tests to accurately predict the motions, accelerations and structural loads on the vessel.

3.2.1 Model scale should be at least 1:50 and its size such that it avoids test tank wall interference. Weight distribution along both the longitudinal and transverse directions of the model must be reproduced as correctly as possible to measure the global loads like vertical and transverse bending moments, torsion loads and shear forces with reasonable accuracy. Model construction should consider scaling of bending stiffness as and where applicable to enable accurate measurement of hydro-elasticity dependent loads or suitable corrections may be incorporated in the analysis of measurements.

3.2.2 The measured data during model tests in waves should be obtained for sufficiently large number of wave encounters, especially in the case of irregular seas to enable reasonably reliable and accurate results.

3.2.3 The data sampling frequency should be sufficiently high, especially for the measurement of loads induced by impact or structural vibration, to enable the dynamic property of the loads be recorded reasonably accurately.

3.3.1 Loading conditions to be chosen for model tests are as specified in Vol 4, Pt 1, Ch 3, 2.2 Input requirements 2.2.1.

3.4.1 The sea area considered to produce the worst operating environment for a vessel in unrestricted service is the North Atlantic. The most onerous wave conditions in the North Atlantic with a return period of 20 years are generally to be used. The wave data and spectrum are as specified by IACS Recommendation 34. The most onerous wave conditions, which are often presented as wave contour, are shown in Figure 3.3.1 North Atlantic Wave Contours in 20 years return period.

Figure 3.3.1 North Atlantic Wave Contours in 20 years return period

3.4.2 For a vessel in restricted service, the testing wave conditions are to be determined from the wave data of the specified service area. A prior approval of the wave data by LR is required.

3.4.3 The minimum model test matrix which is required to be carried out is given in Table 3.3.1 Minimum model test matrix to consist of the following.

3.4.4 Selection of the speed should consider operation limitation of the vessel in the selected testing sea state with agreement of LR. In beam, stern quartering and stern seas, the testing sea states can be reduced to the most onerous wave conditions with return period of 1 year in the North Atlantic, or in the specified service area if the vessel is in restricted service.

3.4.5 The theoretical analysis can be used to select critical conditions to reduce the scope of model tests.

3.5.1 The basis on which the parameters are chosen for investigation is to be submitted for approval at the earliest opportunity.

3.5.2 In addition to those quantities which are normally measured in a model experiment, the following data are to be obtained where practicable:

1. Motions;

2. Accelerations at:

   Bow – 10 per cent of LPP aft of FP

   CG – (depends on the condition tested)

   Stern – 10 per cent of LPP forward of AP;

3. Global loads: as given in Vol 4, Pt 1, Ch 3, 2.3 Load and motion parameters 2.3.1 to (g);

4. Local loads at bow and wet deck.

3.5.3 Measurement of hull motions should preferably be non-intrusive to avoid the effect of instrumentation on the body motions. Measurement of the encountered waves are more desirable than that of the stationary waves, and non-intrusive method of measurement should be preferred to avoid water run-up on the forward side and ventilation of the back side of the wave probes.

3.5.4 Range, natural frequency, frequency response and linearity of the accelerometers are to be included when reporting acceleration data apart from their sampling rate.

3.5.5 Segmented models are to be used to measure the global loads like, bending and torsion moments, shear forces in hull and cross deck structure or hydro elastic effects like springing and whipping.

1. Segmented model is to be built in a number of stiff segments connected with force transducers. The side-hulls and the connection between the side-hulls (bridging structure) should also be segmented and joined with force transducers, in the same manner as for the main hull girder. Stiffness in connections must be enough either (i) to ensure no effect of flexibility or (ii) to give correct eigenfrequencies depending on whether a stiff or hydroelastic model will be applied.

2. A model purely for motion response measurement requires only global mass property to be represented, including mass, centre of gravity and radii of inertia.

3. A model for load measurement is constructed by modelling both the global mass property and mass distribution as accurately as practicable. Location and number of cuts in the model are to be determined such that they are able to measure the most critical load components and the loads at the critical structural locations.

4. A hydroelastic model requires modelling the structural properties of the full scale vessel (bending stiffness, eigenfrequencies and eigenmodes) also.

3.5.6 Measurement of local loads should preferably be based on both the point pressures and also forces on a sensibly chosen area on the vessel. Measurements on panels with scaled ‘dimensions and flexural properties’ or with suitable corrections are recommended compared to the point pressure measurements:

1. For measurement of local loads on bow and wet deck, hydroelasticity of the local plating is to be correctly modelled or suitable corrections to be incorpo

2. Location of transducers, diameter of the sensor face, the range, frequency response and linearity of the transducer are to be reported.

3. Measurements should indicate sampling rates, rise time for the experiment data collection system. Generally sampling rates in the order of ‘kHz’ would be required.

4. Motions and velocities at bow and under wet deck relative to incoming wave crest are to be measured also. They can then be used as basis to analyse or verify the impact pressures based on either dropping test results or theoretical analysis approved by LR.

3.5.7 Test duration for each irregular sea state should be not less than 3 hours in full scale time to determine the critical motions, accelerations and global loads, and six hours for the critical local loads. It is recommended to repeat the most critical test cases for each response to improve reliability of the measured data.

3.6.1 Appropriate steps are to be taken to ensure the unwanted noises due to instruments or model vibrations be removed from the measured raw data, and the high frequency components, that do not induce local or global structural responses, due to wave impacts be filtered out.

3.6.2 The processed data are to be analysed by appropriate statistic model as agreed by LR, to establish the best statistic fits to the measured responses in critical sea states versus the probability of occurrence.

3.7.1 The following details are to be submitted:

1. A summary of the model details including its size, weight distribution and construction.

2. A summary of the testing arrangements and procedures.

3. A summary of the tank facilities and test equipment.

4. Details of the different instrumentation used during testing and their calibration including calibration procedures.

5. Details of the wave/sea state generation, measurements of waves, responses and loads, definitions and notations.

6. Details of data acquisition, reduction and analysis procedures.

7. Tabulated and plotted output.