Paper presented at RINA Marine Renewable Energy Conference 2008

THE ROYAL INSTITUTION OF NAVAL ARCHITECTS THE ADVANTAGES OF COMPOSITE MATERIAL IN MARINE RENEWABLE ENERGY STRUCTURES
M Mohan, Gurit, UK

SUMMARY Composite materials can have many advantages when they are used in marine renewable energy structures. These advantages will be demonstrated in this paper with an example of a 2m prototype of a C-Power Underwater Turbine that was engineered and built at Gurit using composite material. The novel design features of the turbine presented challenges to the structural engineering and will be discussed in this paper. 1. INTRODUCTION The fibres are the part of the composite material that contributes to the strength whilst the matrix holds the fibres together. The fibres generally have a high modulus of elasticity and a high ultimate strength [2]. The fibres can be in continuous form or chopped strand form. With continuous fibres, the orientation of the fibres in a structure can be fixed whilst with chopped strand fibres the fibre orientation is random. Hence fibre orientation tailoring to suit the structure can not be done when using chopped strand fibres. In advanced composite applications, continuous fibres are generally used. Continuous fibres can be made from many different types of materials but the common ones are made out of Glass, Carbon and Aramid. The fibres can come in the form of uni-directionals or woven cloth. The purpose of the matrix is to bind the fibres together, protect the fibres from damage, to transfer the stresses to the fibres and to disperse the fibres [2]. The matrix consists of a resin and examples of common structural resin systems are polyester, vinlyester or epoxy. Polyester and vinlyester resin are low in cost but produce high styrene emissions during production. Although Epoxy resins are more expensive, they generally have superior mechanical properties. All resins will absorb some moisture but the differences between the resins are the way the absorbed moisture reacts with the resin to degrade the laminate. Polyester and vinylester contain the ester group that reacts with the small amount of moisture to start the osmosis cycle which can cause the surface of the laminate to blister. This is less likely to happen with Epoxy resins as they do not contain the ester group. Chopped Strand Mat/Polyester composites are fairly common and have low mechanical properties. Advanced composites such as glass/epoxy or carbon/epoxy are used for high performance applications. Figure 1: A Composite Laminate Cross Section A composite ply consists of a layer of fibres that is impregnated with resin. A composite laminate is formed

With environmental issues such as global warming and the current trend of increasing oil price, it has become imperative to draw our focus onto other forms of energy such as wind power and tidal power. Tidal power can be consistent and predictable if the marine turbine is installed in locations with high tidal ebb and flow or strong ocean currents [1]. There are many marine turbines currently being developed and the prototypes are mostly built with conventional materials. Currently, on a few marine turbines, the only application of composite materials is on the rotor blades. The only commercial scale tidal turbine to be installed has rotor blades made from composite materials. There is a wide scope for using composite materials on other parts of a marine turbine. Gurit has engineered an all composite prototype underwater turbine that has been designed by C Power. The advantages of composites in marine turbine design are discussed in depth in this paper. 2. 2.1 COMPOSITES MATERIAL CONSTITUENT

A composite material is a material that consists of two components: the fibres and the matrix as shown in Figure 1.

Paper presented at RINA Marine Renewable Energy Conference 2008

of several composite plies. These plies can vary in direction (orientation) through the stack of plies as shown in Figure 2.

The wet lay up method is low in cost in terms of materials and tooling required. As the laminate can be cured at room temperature or at low oven temperatures, expensive high temperature tooling is not required. Freezers are also not required as the materials can be stored at room temperature. In comparison to some other composite manufacturing methods, it can be relatively easy to train new operators in the wet layup method. This method is also suitable for the production of large structures. Wet layup is labour intensive and the quality of the laminate is dependent on the skill of the operators. However, high and accurate fibre volume fractions can be obtained by applying known amount of resin. As a general rule of thumb, the weight of resin applied to a square metre of fabric must be no more than the area weight of the fabric. Using a vacuum bag will also help achieve the required fibre volume fraction. Typical resin content in a structure produced using this method is 3540%. Typical amount of void contents in the laminate are between 2-6%. Details on other composite manufacturing methods can be found in the Gurit Guide for Composites [3]. 2.3 ADVANTAGES

Figure 2: A Composite Laminate 2.2 MANUFACTURING

There are many different manufacturing processes available for composite material. These include wet lay up, resin infusion and prepreg. As shown in Figure 3, with the wet layup method, dry fibres are placed in the mould and wet out with the resin using a consolidation roller.

One of the main advantages of composite materials is the ability to choose the material, laminate and manufacturing method to suit the design requirements. This is further discussed in Section 2.4. In general, composite materials have a high strength to weight ratio and high stiffness to weight ratio. This enables lightweight structures to be designed which can help to achieve neutral buoyancy. Also lightweight structures require less expensive lifting equipment for the installation of an underwater turbine.

Figure 3: Wet Layup Method [3] A vacuum bag is used to obtain better consolidation of the fibres. Figure 4 shows the layup of the laminate on a tool with a vacuum bag. Using a vacuum bag enables the production of composite laminates with lower void content and higher fibre volume fraction. The resin flow throughout the fibres is improved which improves the wet-out of the fibres.

By using composite materials and moulds, complex shapes can be made with high geometric tolerances. A composite underwater turbine will also be easy to maintain through out its life cycle as it is resistant to marine boring organisms and resistant to corrosion. Composite materials such as E-glass are also nonconductive which makes it ideal for use in some designs of underwater turbines.

2.4

DESIGN

Figure 4: Vacuum Bag Layup [3]

There are many variables that can affect the cost, quality and weight of a composite structure. By taking these variables into account simultaneously during the design phase, a structure that is optimised for mechanical properties, weight and cost can be produced. These variables include materials constituent, laminate and manufacturing methods- Figure 5.

Paper presented at RINA Marine Renewable Energy Conference 2008

Material Constituent

off production. Therefore the choice of manufacturing method also has a big impact on the cost of moulds and components. When producing large quantities of a product, a more expensive manufacturing method can be justified. For example in wind turbine blade production, 180 blades per year can be made per mould. Hence a mould that can turn out a high number of parts is required which is generally more expensive. Since the choice of manufacturing method has an impact on the cost, weight and quality of a product, the appropriate manufacturing method has to be chosen to meet a customers requirement. For example, the quality of the laminate on an Americas Cup boat has to be much higher than the quality of the laminate on a cruising boat. In this case, a higher technology manufacturing method will be chosen for racing boat which will have its cost implications. 3. STRUCTURAL DESIGN OF THE PROTOTYPE C POWER UNDERWATER TURBINE

DESIGN

Manufacturing

Laminate

Figure 5: Design with Composites 2.4 (a) Material Constituent When using composite material, the type of fibre/resin combination can depend on the structure and its application. For example, for the purpose of this prototype underwater turbine, glass/epoxy composite had the required strength to weight and stiffness to weight ratio. When looking at building the full size underwater turbine, carbon/epoxy will be a consideration as it has a higher stiffness to weight and strength to weight ratio. The type of material chosen also affects the cost and the weight of the product. Hence the appropriate type of material has to be chosen to meet the requirement. 2.4 (b) Laminate One of the main advantages of using composite material is the ability to tailor the laminate to achieve the required mechanical properties by varying the fibre orientation and the position of the ply in the laminate stack. Hence a laminate can be designed to suit the structural requirements. The fibres in the laminate can be aligned in the direction of the main load path so that the fibres are being used efficiently. This way a more optimised structure can be designed. This sort of laminate tailoring can have a significant effect on the cost and weight of the final structure. 2.4 (c) Manufacturing Method With composite materials, the manufacturing method has a large impact on the fibre volume fraction which affects the quality of the laminate. Hence, the type of manufacturing method used affects the strength, stiffness and weight of a composite structure. Some manufacturing methods are most suited to high volume production whilst other methods are suited to one Figure 6: C Power Underwater Turbine The turbines rotor fits inside the turbines stator and a tight tolerance is required between the rotor and the stator. To obtain maximum efficiency, a small gap with high circumferential tolerance had to be maintained between the rotor and the stator during operation. Hence an accurate shape is required to maintain this tight

3.1 GENERAL ASSEMBLY The C Power Underwater Turbine has a novel design and a design code was not available for this underwater turbine at the time of design. The load cases and design criteria were generated by Gurit. The C-Power underwater turbine consists of two main parts the rotor and the stator. These are shown in Figure 6. This turbine was constructed with glass/epoxy and using the wet layup method.

Paper presented at RINA Marine Renewable Energy Conference 2008

tolerance and can be achieved by using composite materials and a mould. For this design, it was important that the rotor was neutrally buoyant and that the entire turbine had to be buoyant. 3.2 STATOR 3.2 a) Water Pressure Load Case In this load case, the turbine is submerged 3m underwater and the water pressure on the stator was used to determine the laminate of the stator structure. In this case, the stator is stiffness driven. Two types of laminates could have been used for the structure of the stator. The laminate could have been a single skin or sandwich construction. A section of the stator was analysed as a panel under this pressure load. It was found that the single skin laminate panel deflected by about 120mm whilst the sandwich panel only deflected by 0.1mm. If a single skin laminate was chosen, webs would have been required to achieve the required amount of stiffness. The sandwich construction was chosen as the stiffness criterion was met relatively easily with little internal structure which simplifies the build. When using a sandwich laminate, a minimum thickness is required for the outer skin laminate so that the structure is robust. The glass skin laminates in the sandwich structure were optimised to obtain sufficient amount of stiffness and strength with this minimum skin thickness. 3.2 (b) Lifting Load Case In this load case the turbine is being lifted by using two lugs. The working load (W1) for this load case is twothirds the weight of the turbine on each lug. This load case ensures that the lifting webs on the stator do not fail when the turbine is being lifted using the lifting points. The Ultimate load was used to design for this load case and the safety factors used are shown in Table 1. All Reserve Factors (RF) for the Ultimate load were more than 1. Factors of Safety 3 2

that buckling of the outlet stator under the weight of the turbine will not be an issue. 3.2 (e) Maintenance Access Load Case The Working Load (W1) for this loadcase is the weight of 2 people (100kg) at 2g bouncing on top of the stator. The safety factors used are shown in Table 2. For the Ultimate load case, it was found that the RF is more than 1. Factors of Safety 1.5 2

W1 to Limit Limit to Ultimate

Table 2 : Factors of Safety for Maintenance Access Load Case 3.3 ROTOR The rotor is modelled as a ring where the behaviour of rotor and the blades can be approximated as a circular ring with the moments, MZ (cause by the blade loads) acting on the ring as shown in Figure 7.

Figure 7: Moments on the rotor The working load (W1) is the Maximum Total Bending Moment on the rotor. A strength analysis was performed on the rotor for the Ultimate load case. The factors of safety used are shown in Table 3. It was found that all the RFs were more than one. It was concluded that the design of the rotor is not strength critical. Factors of Safety 2 1 1.5

W1 to Limit Limit to Ultimate

Table 1: Factors of Safety for the Lifting Load Case 3.2 (d) Buckling Load Case The outlet stator was checked for buckling under the weight of the entire turbine and a factor of safety of 2.5 was used. The buckling RF was obtained by comparing the actual buckling stress to the allowable buckling stress. The RF was sufficiently high enough to conclude

W1 to W2 W2 to Limit Limit to Ultimate

Table 3: Factors of Safety Used for the Rotor Design

Paper presented at RINA Marine Renewable Energy Conference 2008

A stiffness analysis was also performed on the rotor using the bending moments from the blade (MZ). It was found that the design was not stiffness critical as the rotor only deflects by a really small amount. Since the design was not driven either by strength or stiffness, a laminate with a thickness that was easy to manufacture was chosen. 3.4 BLADE The blades were designed for the bending moment (MZ). The load on the blade was about 1.5 kN and the blades were about 0.65m in length. The blades can be made from E-glass, carbon or thermoplastic (rapid prototyping) material. A strength analysis was carried out on a blade made from of each of these materials. It was found that the design of the blade is not strength critical. For the stiffness analysis, the rotor blades were modelled as a cantilever beam. It was found that the deflection of the rapid prototyping blade was too large and the deflection of the carbon blade was fairly small. These deflections are shown in Table 4. The deflection of the Eglass blade was within the acceptable range and since Eglass is much cheaper than carbon, E-glass was chosen as the material for the blade. Type of Material Thermoplastic Glass Carbon Deflection (mm) 152 13 4

were made from glass fibres that were oriented in the 0 and 45 to give the required properties. 4.1 (b) Resin The epoxy resin that was used has been optimised for the manufacture of large structures using the wet lay up method. Large structures are likely to have heavy reinforcements so this type of resin has a low initial mixed viscosity to allow easy wetout of these reinforcements. This resin has been optimised for laminating at temperatures between 18-25C. A standard hardener was used with this resin. 4.2 PROCESSING Due to the tight tolerances that needed to be achieved, a male mould made from a CNC machined tooling block was used to ensure that the correct tolerances were achieved. The required shape and dimension can be achieved with good accuracy by using this type of tooling. For parts of the turbine that were less critical in terms of tolerances and dimensions, a cheaper mould made out of wood can be used. A CNC machined tooling block is shown in Figure 8 whilst tooling fabricated from wood is shown in Figure9.

Table 4 Deflection of Rotor Blade 3.5 WEIGHT In this design, the weights of all the components were critical as the design requirement was that the rotor had to be neutrally buoyant in water whilst the entire turbine had to be buoyant by about 50kg. A detailed weight estimate was made during the design of the turbine. The weights of all the components were checked as the turbine was being built. By doing so changes to the design can be made to ensure that the required weight is achieved. 4. BUILD OF PROTOTYPE

Figure 8: CNC Machined Tooling Partially Gelcoated

4.1 MATERIALS 4.1 (a) Reinforcements Glass fibre was chosen for all the parts of the turbine. The stator was made completely out of fibres that were oriented in the 0, 90 and 45. The rotor and the blades Figure 9: Tooling Fabricated from Wood

Paper presented at RINA Marine Renewable Energy Conference 2008

A layer of gelcoat was applied on to the tooling before laying up the fibres. By using this layer of gelcoat, the finished part will not require filling and fairing which contributes to achieving the required weight and dimensions. The layer of gelcoat also provides an abradable surface for painting. Due to the shape of the male mould used, dry fibres can not be laid onto the mould as it would slide off. In this case, the fibres were wet out with the resin before it was laid onto the mould. The tack of the resin prevented the fibres from sliding of the mould. Good consolidation was obtained by using a consolidation roller and a vacuum bag. The components of the turbine were cured at room temperature (18C) for 20 hours and then a final postcure in an oven at 50C for 16 hours. Post-curing the parts improves the laminates mechanical properties. The parts were then painted and assembled together.

and it was used in this case as the forces were not significantly large in any particular direction. As the size increases, the hoop stress in the rotor of the full sized underwater turbine will be significant. Hence for the full sized stator, the fibre orientation can be tailored so that there are more 0 fibres in the direction of the hoop stress to take this load as in the prototype rotor. Hence tailoring the fibre orientation to suit the type of load on a structure allows for a more efficient and optimised structure to be designed. The laminate is not unnecessarily thick which in turn contributes to the weight and cost saving. The stator and rotor on the prototype underwater turbine were manufactured as a single piece due to their size. With the full sized underwater turbine, it will not be possible to build the stators and rotor is one piece. They will have to be made in sections and joined together. With composite materials, the precision required to obtain a good fit between the sections can be achieved. For the mass production of the full sized underwater turbine, a less labour intensive manufacturing method will be required. For example, the mass manufacture of wind turbine blades uses the prepreg or resin infusion manufacturing method. 6. CONCLUSIONS

5.

DESIGN CONSIDERATIONS OF THE FULL SIZE UNDERWATER TURBINE

For the design of the full size underwater turbine, consideration has to be given to the type of materials used. As the turbine increases in size, the design driver might change. The strength to weight ratio and stiffness to weight ratio could become a critical factor in material choice. With the prototype underwater turbine, a laminate of a minimum thickness was both very strong and very stiff. For the full sized turbine, the larger size means that it probably needs a lot of glass/epoxy laminate to achieve the required strength and stiffness. If the full sized turbine is made from carbon epoxy laminate, a lot less laminate will be required to achieve the same sort of stiffness and strength. This is because carbon/epoxy laminates have a much high strength to weight ratio and stiffness to weight ratio than glass/epoxy. For example, the design of the rotor blade was largely stiffness driven and for the prototype blades it was seen that glass/epoxy laminate provides adequate stiffness. For the full sized blades, due to the increase in length, there is a possibility that the glass/epoxy laminate will not be stiff enough and they might need to be replaced with carbon/epoxy laminate. Cost also plays an important role in material selection as carbon/epoxy laminates are more expensive than glass/epoxy laminate. The decision on which material to choose ultimately relies on a trade off between strength to weight ratio, stiffness to weight ratio and cost. For the rotor of the prototype underwater turbine, the laminate consisted of equal number of plies in 0, 90 +45 and -45. This is known as a quadriaxial laminate

The high strength to weight ratio and high stiffness to weight ratio of composite material makes it an ideal for underwater turbine. Excellent geometrical tolerance can be achieved with composite materials. Structures with complex shapes and geometry can be made relatively easily. Due to the environment in which an underwater turbine operates in, the corrosion resistance of composite materials is an important feature. Composite materials are also resistant to marine boring organisms. With composite materials the designer has the freedom to choose the materials, laminate and manufacturing method to suit the design requirements. This means a more optimised structure can be designed.

7.

ACKNOWLEDGEMENTS

The author would like to express her thanks to C-Power for allowing her to use their underwater turbine as a case study for this paper.

Paper presented at RINA Marine Renewable Energy Conference 2008

8.

REFERENCES

1.

GARDINER, G., Tidal Turbines to Mine Marine Megawatts, Composites Technology, January 2007 HOLLOWAY, L.C., Polymers and Polymer Composites in Construction, Thomas Telford, 1990. http://www.gurit.com/downloads.asp?section=00010 00100220026&sectionTitle=Marine+Downloads&le tter=G

2.

3.

9.

AUTHOR BIOGRAPHY

Menaga Mohan holds the current position of Design Engineer at Gurit UK. She is responsible for structural design with composite material. Her previous experience includes working on the C Power composite underwater turbine and on an all composite 60m yacht Panamax Ketch.