You are here

Validation of marine structures – Testing approaches

News International-French

25 Feb 2011

This paper presents some of the tests and controls used to validate high-performance marine structures. The materials used to manufacture these structures are mainly carbon/epoxy composites with monolithic or sandwich construction. Non-destructive tests based on dynamic structural analysis are discussed more particularly here.

(Published on August-September 2010 - JEC Magazine #59)


In recent years, composite materials have been used widely in pleasure and racing boat manufacture, due to their high specific properties. For racing boats, which are at the leading edge of design innovation, the challenge facing designers and manufacturers is to find the best combination of performance and safety.


In this context, it is essential to know structural material properties and to validate the mechanical behaviour of structures. Here, this is illustrated with examples of projects conducted on structural elements of racing boats.


Mechanical testing

The performance of racing multihulls can be increased significantly by introducing hydrofoil parts on the floats. The main advantage of a hydrofoil is that it contributes along with profile lift to reduce the contact surface area between the floats and the water. These structures are subjected to very high mechanical loadings and dynamic stresses.


French manufacturer FMC recently produced the hydrofoils for the maxi trimaran Sodebo. To reduce weight, the project designers retained a hollow monolithic structure (Figure 1). Following significant development work, the manufacturing process was validated using quality controls and mechanical testing on samples and hollow structures. During the transfer of hydrodynamic lift forces to floats, the hydrofoil is subjected to high reaction forces. Therefore, the mechanical behaviour of the hydrofoil was studied in detail, using compressive tests on a cross section. A picture of a cross section after failure and an example of a force/displacement curve recorded during a compressive test are presented in Figure 2.



In addition, videos were recorded during the tests to provide information on the failure modes. A structural effect was observed due to the specific geometry of the section: the final failure was initiated in flexure on the thickest section of the sole and propagated inside the cells. It is important to note that the failure is not governed by the compressive failure of the transverse partitions.


The force/displacement curve shows that a significant loss of rigidity occurs at 480 kN. The final failure was observed at 770 kN.


Non-destructive tests based on modal analysis

Non-destructive methods based on modal analysis offer many advantages, both for measuring dynamic properties and for estimating the global rigidity of structures. Three examples of structural tests are presented on keel, mast and cable rigging elements.


With the optimization of the IMOCA Open 60 racing monohull, a resonance phenomenon appeared on the keel. The problem results from a coupling effect between the vibration behaviours in flexion and torsion, and occurs at high speed when the torsional rigidity of the keel is much lower than the flexural rigidity and the ballast weight. Measurements during sailing showed that the vibration modes concerned are the first modes in flexure and torsion.


The dynamic behaviour of ballast keels was investigated using a vibration analysis protocol. This test is now recommended by the IMOCA class rules. The objectives are as follows:

  • Validating the first natural frequencies in flexion and torsion in relation to the theoretical values of the design office. Non-coupled values of these frequencies are expected in order to ensure a good dynamic behaviour of the keel.
  • Implementing periodic controls. Structural modifications of the keel or a loss of rigidity due to damage can change the natural frequency values. For example, a higher ballast weight compared to the initial keel configuration leads to lower natural frequencies.



As shown in Figure 3, these vibration measurements were performed using accelerometers and video analysis. Similarly, NCD developed a measuring protocol for masts. The aim is to validate the dynamic signatures of new masts with respect to calculated values from the design office and to perform periodic controls. The measurements are especially useful to validate a construction for standardized production. The test protocol includes two sets of measurements in longitudinal and transverse configurations:

  • simply supported: the mast is set on cradles,
  • free/free: the mast is hanging from elastic cables.


NCD is an independent consulting firm that was created in Brest, France in 2007. NCD specializes in composite materials engineering. Its services include composite design, quality control and surveys, research on the damage tolerance of composite materials, and non-destructive testing of structures using modal analysis techniques. Established in Brest, France since 2002, FMC specializes in the manufacture of composite parts and structures. The company offers a number of manufacturing processes, from Class 1 autoclave prepregs to infusion moulding and including RTM, LRI, RFI, and compression moulding. Its main customers are in the defence, oceanography, nuclear energy and racing yacht sectors.


The events are recorded using accelerometers and the signals are analysed by FFT (Fast Fourrier Transform) and analytical models. Based on the signals and natural frequencies measured on the mast, two parameters are estimated:

  • global flexural rigidities between the limit conditions of the test,
  • dynamic coefficients of damping as indicators of response time.



This mast protocol has been validated on several types of racing boats, such as Olympic 470 dinghies, Bénéteau Figaro 2 class, and IMOCA Open 60 class (Figure 4). Modal analysis methods can also be used to estimate tensions in rigging cables. The aim is to ensure that the tensions are well balanced and to optimize the rig adjustments. This method is based on the use of analytical models. The tension is calculated from the natural frequencies of the cable and physical parameters such as length, mass per unit length and Young’s modulus. The first model, presented in equation 1, is simple and suitable for long cables.


where n is the number of the vibration mode considered, L the length of the cable, m the mass per unit length and T the tension applied. Equation 1 is not applicable for short cables and the effects of cable rigidity and limit conditions need to be taken into account in this case. The beam models for carbon composite cables were improved using tensile tests. The natural frequencies of the cables are measured as a function of applied force. Figure 5 shows that the beam model provides a good estimation of experimental data. Following this validation step, the method was applied to rigging cables (main shroud and deck spreader) for an IMOCA racing boat.




Several testing approaches for racing boats are presented in this article, including the use of non-destructive methods based on modal analysis to control keel, masts and rig cables. These methods, and more classical ones like physico-chemical analysis and mechanical tests, provide essential information during the validation process to ensure manufacturing quality and mechanical behaviour of highperformance composite structures.



Peter Davies and Dominique Choqueuse : Ifremer Brest, Pascal Casari : IUT de St Nazaire, Team Veolia (Kairos) Team Foncia (Mer Agitée)