I’ve got outputs from Numad + Precomp and trying to create my own BeamDyn input file to use in FAST.
Majority of needed data are obtained from Numad’s output but there are some missing pieces and confusion.
questions:
1_how can I obtain edge and flap shear stiffnesses?
Page 15* in BeamDyn manual decsribes what Damping Coefficient is but I can not fully understand it.
2_Should I put a standard value for Damping Coefficient or do I need to calculate my own Damping Coefficient by using the formula in page 15?
PreComp does not model the shear terms, so, it cannot be used to calculate the edge and flapwise shear stiffness. NuMAD should be able to calculate the shear stiffness, but I’m not familiar enough with the software to know how to output it.
Basically, BeamDyn uses stiffness–proportional structural damping. A setting of 0.01 for all coefficients is reasonable in the absence of more information.
I have noticed that you recommend a setting of mu = 0.01 as reasonable here, whilst for the NREL5MW reference blades mu = 0.001 is preset. I lack a good feeling for the physical meaning as well as the values of these damping coefficients, as they differ from typical modal damping or Rayleigh damping values used in aeroelastic analyses. Is there any method to have a rough estimate of the damping values for BeamDyn, e.g. based on modal damping?
You could select the stiffness-proportional damping in BeamDyn based on a damping ratio (zeta) and natural frequency (f, in Hz) of a given mode using the standard relationship mu = zeta/(pi*f), with the understanding that stiffness-proportional damping produce damping ratios that scale linearly with frequency. For example, a damping ratio of 2% and a frequency of 0.6 Hz results in mu of a approximately 0.01 s.
I don’t recall why mu is set to 0.001 s for the example BeamDyn model of the NREL 5-MW reference wind turbine, but I agree that this is inconsistent with the specification for this turbine. Perhaps the structural damping was reduced to compensate for the numerical damping intrinsically included in BeamDyn’s generalized-alpha time integrator (depending on the setting of rhoinf).
thanks for your quick help. I’ve had never worked with stiffness-proportional damping and was confused with mu being a non-dimensional damping ratio, as the BeamDyn_Blade input file header of the NREL5MW suggests that mu didn’t have any unit.
Currently, I am struggling with stand-alone BeamDyn input files( Dvr_5MW_static.inp, Static_BeamDyn_input_5MW.inp, 5MW_Blade_IEC.inp). I wonder if you could help me to change the length of the blade in the input file and also I wanted to know the difference between nodes and number of key points on member(kp_total). because for instance I want to consider 6 nodes on the member and then applying force on each node.
It looks like you’ve posted the same question in OpenFAST issues (github.com/OpenFAST/openfast/issues/140), which has now been answered by Jelmer Polman, with a reference to the following forum topics:
I’d like to follow up on the calculation of the damping coefficient used in BeamDyn (mu).
I’ve got some measurement data of a rotor blade and would like to use the measured damping ratios in BeamDyn to improve my simulations.
Based on the measured natural frequencies and damping ratios I’m able to calculate the damping coefficient for a given mode as described above.
I’ve got data for the following eigenmodes:
Rigid Body Heave, Roll, and Pitch
1., 2. and 3. bending flapwise
1., 2. and 3. bending edgewise
and 2. torsion.
My problem now is, that I’m not certain how to transfer those calculated mu into BeamDyn; i.e. the data of which mode to use for which mu (1,1 – 6,6).
Also the calculated mu are decreasing drastically with increasing mode order (no linearly scaling) and the mu of the rigid body modes are about 100-times greater than the other (0.01 vs. 0.0001).
This behavior also produces the following warning when running BeamDyn with the mu based on the measurement:
FAST_InitializeAll:BD_Init:BD_ValidateInputData:Damping values in blade file are not of similar
order of magnitude. BeamDyn may not converge!
Is there any additional documentation on this besides the ‘damping coefficient’ section in the BeamDyn User Guide?
I’m not sure what you mean by damping of rigid-body modes. Damping is not really defined for rigid body modes and BeamDyn itself does not support rigid-body modes.
BeamDyn only supports stiffness-proportional damping, so, you’ll be able to match the damping of one mode (likely the first mode), but not all modes. With stiffness-portional damping, the damping will increase with increasing frequency. If I have known damping values for the flapwise, edgewise, and torsion modes, would normally tune mu55 to the correct damping of the first flapwise mode, mu44 to the correct damping of the first edgewise mode, and mu66 to the correct damping of the first torsion mode. The values of mu11, m22, and m33 are likely less critical, but you could set them to values similar to mu44, mu55, and mu66.
