I'm working on a project where I need to model a number of airfoils and determine their torsional stiffness. I'm looking at building foils out of various types and combinations of materials and need to determine how the torsional rigidity is affected by the use of different material in different sections of the foil cross section. Similarly, I'm investigating the impact of adding holes (void sections) to the foil interior, again with the goal of understanding the impact on torsional rigidity. I have been unable to figure out how to calculate torsional rigidity of solids in Inventor. If anyone knows how to do that I would appreciate the help.
Alternatively, I could calculate this value if I have access to the shear modulus (G) and torsional constant (J). From what I have read, Inventor determines G based on poisons ratio and youngs modulus. Can Inventor provide the effective youngs modulous for a composite foil? Also, is there any means of calculating the torsional constant or a cross section using Inventor?
Thank you for your help.
I have found that the best method thus far is to perform a static stress analysis with a known torque applied to one end of the member and the other end fixed. I measure the resultant angle and can then deduce torsional stiffness (the product of shear modulus and torsional constant GJ), as I know both applied torque and resultant twist.
Any thoughts are still appreciated.
I am having same problem with you. But I have no clue how you get the twist angle from a static stress analysis. I need to find the moment of area of a shape, but I couldn't found from stress analysis. Do let me know where to locate the twist angle. Thank you.
The angle of twist cannot be directly obtained as far as I can tell. However, if you create two working points in the model at known locations, you can use the results inquiry option to determine the location of each point before and after deformation. This allows you to determine the direction vector between the two points in both cases. The angle can then be deduced using the equation theta= acos(dot(d0,d1)/norm(d0)/norm(d1)) where do&di are the direction vectors before and after deformation.
The issue with using this method is that it assumes the the cross section does not warp. An alternative method is to implement a prescribed rotation through a remote load, which holds the face planar. However, using this method you cannot determine the neutral axis as the implementation of the remote load dictates the axis of rotation.
Hope that helps,
An advisor on here also recomended creating beam elements in the regions you are interested in. Beam elements (unlike bricks and tets) can display nodal rotation values. Once they are included, you can determine angle of twist by accessing Results Contours>Displacement>Rotation
Best of luck
There is another option to display nodal rotations within Stress Analysis by using Shell elements.
Shell elements are introduced R2013. You can make a "dummy" Shell body at the tip of the airfoil, which won't contribute to the overall stiffness. Display rotations as the result type. At least this way, you won't have to calculate angles from the Probe's Cartesian coordinates.
Attached is a quick example.
Hope this helps...
Best regards, -Hugh
Thank you Hugh
Could you please tell me how you're applying your moment in the figure shown?
At the moment I'm using a surface torque, but I'm considering other approaches and I'm intruiged by the forced displayed in your figure.
A Force load can be applied to a face, edge or vertex. In the example, I applied the force loads to two edges, and specified the direction of each by selecting a face (to capture it's outward normal). Alternately, you can specify vector components of the force load.
If there are no edges available in your model, you can create them by splitting the face (a modeling command) using a workplane, sketch, etc. as the Split tool. This "split face method" is also handy if you need to localize a load to a portion of a face.
Log into access your profile, ask and answer questions, share ideas and more. Haven't signed up yet? Register