Structural Solutions to Save Weight in Future Blades
Recent work at Risø DTU on wind turbine blades has shown that more failure mechanisms need to be taken into account than just the classical ones such as buckling, material failure, etc. An example of one of the failure modes, which is not part of a certification process, is the non-linear out-of-plane deformation of the load-carrying cap laminate, which introduces interlaminar shear stresses in the load-carrying laminate. This could be the reason for some of the failures in today’s wind turbine blades. This failure mechanism is taken into account in the design of a new, innovative 40m load-carrying box girder presented in this article.
By Find M. Jensen, Risø National Laboratory for Sustainable Energy, Denmark
The concept blade represents a significant structural improvement with regards to the design criteria established in references 1 and 2. The first design criterion only allows a certain (e.g. 4–6mm for a 34m blade) out-of-plane deformation of the load-carrying cap laminate in order not to have too high interlaminar stresses and/or too high transverse tension stresses in the unidirectional layers.
The design only takes static criteria into account and does not consider any kind of fatigue-related issues. Furthermore, the tip deflection is increased considerably and, without compensating for that (by pre-bending, coning or tilting the blade), tower clearance will be a problem. A study of how much tip deflection can be allowed is not part of this study, only the potential for weight saving in the load-carrying laminates using static criteria is taken into account. As a proof of concept, a 40m load-carrying box girder has been designed and manufactured. The box girder includes three structural solutions patented at Risø DTU (see references 4, 5 and 6). Two of the reinforcements reduce the inward out-of-plane deformation of the load-carrying laminate and the third reduces the transverse shear distortion by inserting a diagonal between two opposite box corners. By use of these inventions the thickness of the load-carrying laminates was decreased by 40% and the box was prevented from distorting in the transverse direction.
Failure Mechanism and Innovative Design
The results from a full-scale test performed on a 34m blade from SSP Technology A/S suggest that an important non-linear failure mechanism has been addressed (see reference 1). In this phenomenon, the caps reveal inward out-of-plane deformations which decrease the load-carrying capacity of the structure (see Figure 1).
The out-of-plane deformation could also cause tension failure in the unidirectional (UD) layers which dominate the load-carrying laminate. In Figure 2 the transverse strains were measured back-to-back in a 34m wind turbine blade tested in the flapwise loading direction.
The strain measured in the transverse direction is approximately twice as high as in the longitudinal direction (not shown here). This is mainly due to the fact that the flexural lateral stiffness is very low. This is because there are no fibres in the transverse direction here and the epoxy resin needs to transfer the out-of-plane loads. These are the non-linear Brazier loads caused by the longitudinal curvature. Explanation of this issue and its relevance for wind turbine blades can be found in references 1 and 2.
In Figure 3 an example from the literature (see reference 7) shows how the critical strains were found for a unidirectional glass fibre laminate. The figures show the maximum allowable strains, with (a) and (b) showing tension perpendicular and parallel to the fibre direction respectively. An outstanding difference, by a factor of four, can be noticed between the maximum strain level in the longitudinal and transverse directions. When comparing the measured transverse strains compensated for by the membrane contribution (Figure 2), with the maximum allowable strain from reference 7, it can be observed that the blade has exceeded the maximum strain level by approximately 1,000?S (=6,000–5,000?S).
Based on this observation, a structural solution that could prevent the inward cap deformations was invented (see Figure 4). The corners are coupled, which limits their relative displacement in the transverse direction of the profile. Again, the reinforcement is only to carry tension. Fixing the corners prevents curved panels from flattening, decreases transverse strains and increases the flexural stiffness of the cap. Consequently, the interlaminar shear failure is prevented and the buckling load is increased. The proof for this solution is presented in reference 2. The primary conclusion drawn from this study is that there is a significant improvement in the structural response of the blade when the structural solutions described here are used.
Manufacturing
The cap reinforcement and a diagonal stiffener were introduced in the concept blade, with the box girder being manufactured by SSP Technology A/S. Figures 5 and 6 show the assembled box girder and its manufacture.
The reinforcements were added to the blade using pre-manufactured parts, so that they were ready to install in the box girder as soon as the curing was completed and before the box girder was assembled. Some of the parts were manufactured using the pultrusion process, which is very efficient and delivers high quality results. Furthermore, the decrease in the material thickness by 40% obtained by the new design is beneficial for the manufacturing process because it is less time consuming and prevents defects in the finished laminate. Moreover, the curing process is easier to control since thin laminates reach lower temperatures during the exothermic reactions.
Summary and Commercialisation Conditions for Licensing the Inventions
The new box girder design utilises three patented reinforcements in order to prevent interlaminar failure in the caps and transverse shear distortion of the blade. These reinforcements allowed material savings in the cap of up to 40% when only the two new failure criteria together with the two classical ones (buckling and in-plane material failure in longitudinal direction) are considered. The three inventions have been developed at Risø DTU over the last 2–4 years. Risø DTU is offering licences to the wind turbine industry for the right to apply these inventions to their own blade designs. Currently, the inventions are being evaluated, and a number of wind turbine manufacturers have already asked for specific licensing offers, which should be followed by licensing agreements during 2011 and 2012.
Acknowledgements
The research was financially supported by the Danish Energy Agency through the EUDP-Programme. The support is gratefully acknowledged. Both the research with the 34m blade and the 40m box girder was conducted in cooperation with SSP Technology A/S and their work is very much appreciated. Also, my thanks to colleagues at Risø DTU who have contributed to this work.{/access}
Recent work at Risø DTU on wind turbine blades has shown that more failure mechanisms need to be taken into account than just the classical ones such as buckling, material failure, etc. An example of one of the failure modes, which is not part of a certification process, is the non-linear out-of-plane deformation of the load-carrying cap laminate, which introduces interlaminar shear stresses in the load-carrying laminate. This could be the reason for some of the failures in today’s wind turbine blades. This failure mechanism is taken into account in the design of a new, innovative 40m load-carrying box girder presented in this article.By Find M. Jensen, Risø National Laboratory for Sustainable Energy, Denmark
{access view=!registered}Only logged in users can view the full text of the article.{/access}{access view=registered}An extensive wind turbine blade testing programme at Risø DTU has led to numerous conclusions regarding the structural design of wind turbine blades and has drawn attention to failure mechanisms and loading configurations that are often not required in design and certification tests (see reference 1).
The aim of this work is to provide a blade design that can address the failure mechanisms observed recently (see references 1 and 2). The recently observed failure modes are the non-linear out-of-plane deformations of the load-carrying cap laminate. The two classical failure criteria, buckling and in-plane material failure (static only) in the longitudinal direction, are also considered in this work. Designing a wind turbine blade is a trade-off between improving the performance and reducing the weight. Thus, it is necessary to address the failure modes that are occurring by means of solutions that will not increase the weight.The concept blade represents a significant structural improvement with regards to the design criteria established in references 1 and 2. The first design criterion only allows a certain (e.g. 4–6mm for a 34m blade) out-of-plane deformation of the load-carrying cap laminate in order not to have too high interlaminar stresses and/or too high transverse tension stresses in the unidirectional layers.
The design only takes static criteria into account and does not consider any kind of fatigue-related issues. Furthermore, the tip deflection is increased considerably and, without compensating for that (by pre-bending, coning or tilting the blade), tower clearance will be a problem. A study of how much tip deflection can be allowed is not part of this study, only the potential for weight saving in the load-carrying laminates using static criteria is taken into account. As a proof of concept, a 40m load-carrying box girder has been designed and manufactured. The box girder includes three structural solutions patented at Risø DTU (see references 4, 5 and 6). Two of the reinforcements reduce the inward out-of-plane deformation of the load-carrying laminate and the third reduces the transverse shear distortion by inserting a diagonal between two opposite box corners. By use of these inventions the thickness of the load-carrying laminates was decreased by 40% and the box was prevented from distorting in the transverse direction.
Failure Mechanism and Innovative Design
The results from a full-scale test performed on a 34m blade from SSP Technology A/S suggest that an important non-linear failure mechanism has been addressed (see reference 1). In this phenomenon, the caps reveal inward out-of-plane deformations which decrease the load-carrying capacity of the structure (see Figure 1).
The out-of-plane deformation could also cause tension failure in the unidirectional (UD) layers which dominate the load-carrying laminate. In Figure 2 the transverse strains were measured back-to-back in a 34m wind turbine blade tested in the flapwise loading direction.
The strain measured in the transverse direction is approximately twice as high as in the longitudinal direction (not shown here). This is mainly due to the fact that the flexural lateral stiffness is very low. This is because there are no fibres in the transverse direction here and the epoxy resin needs to transfer the out-of-plane loads. These are the non-linear Brazier loads caused by the longitudinal curvature. Explanation of this issue and its relevance for wind turbine blades can be found in references 1 and 2.
In Figure 3 an example from the literature (see reference 7) shows how the critical strains were found for a unidirectional glass fibre laminate. The figures show the maximum allowable strains, with (a) and (b) showing tension perpendicular and parallel to the fibre direction respectively. An outstanding difference, by a factor of four, can be noticed between the maximum strain level in the longitudinal and transverse directions. When comparing the measured transverse strains compensated for by the membrane contribution (Figure 2), with the maximum allowable strain from reference 7, it can be observed that the blade has exceeded the maximum strain level by approximately 1,000?S (=6,000–5,000?S).
Based on this observation, a structural solution that could prevent the inward cap deformations was invented (see Figure 4). The corners are coupled, which limits their relative displacement in the transverse direction of the profile. Again, the reinforcement is only to carry tension. Fixing the corners prevents curved panels from flattening, decreases transverse strains and increases the flexural stiffness of the cap. Consequently, the interlaminar shear failure is prevented and the buckling load is increased. The proof for this solution is presented in reference 2. The primary conclusion drawn from this study is that there is a significant improvement in the structural response of the blade when the structural solutions described here are used.
Manufacturing
The cap reinforcement and a diagonal stiffener were introduced in the concept blade, with the box girder being manufactured by SSP Technology A/S. Figures 5 and 6 show the assembled box girder and its manufacture.
The reinforcements were added to the blade using pre-manufactured parts, so that they were ready to install in the box girder as soon as the curing was completed and before the box girder was assembled. Some of the parts were manufactured using the pultrusion process, which is very efficient and delivers high quality results. Furthermore, the decrease in the material thickness by 40% obtained by the new design is beneficial for the manufacturing process because it is less time consuming and prevents defects in the finished laminate. Moreover, the curing process is easier to control since thin laminates reach lower temperatures during the exothermic reactions.
Summary and Commercialisation Conditions for Licensing the Inventions
The new box girder design utilises three patented reinforcements in order to prevent interlaminar failure in the caps and transverse shear distortion of the blade. These reinforcements allowed material savings in the cap of up to 40% when only the two new failure criteria together with the two classical ones (buckling and in-plane material failure in longitudinal direction) are considered. The three inventions have been developed at Risø DTU over the last 2–4 years. Risø DTU is offering licences to the wind turbine industry for the right to apply these inventions to their own blade designs. Currently, the inventions are being evaluated, and a number of wind turbine manufacturers have already asked for specific licensing offers, which should be followed by licensing agreements during 2011 and 2012.
Acknowledgements
The research was financially supported by the Danish Energy Agency through the EUDP-Programme. The support is gratefully acknowledged. Both the research with the 34m blade and the 40m box girder was conducted in cooperation with SSP Technology A/S and their work is very much appreciated. Also, my thanks to colleagues at Risø DTU who have contributed to this work.{/access}
