Hello,
I have another question about this topic: when I do simulations the damping that I see in the simulation output is about half of the damping that I specify in the FAST input.
This possibly relates to Jason’s comment in his post above:
.
To minimize the influence of the other modes I disabled all DOF’s except the tower FA mode. Surely the output damping in that case must correspond to the input damping? Well, I still had the mass of nacelle and rotor not set to zero. Is that fact alone responsible for this factor 2 difference between input and output?
The way I determined the damping in the output was: Turbine running in vacuum (CompAero=false), all DOFs turned off except the first tower mode FA, turbine is parked (so no dynamic damping), initial tower displacement 1 m at the top, then releasing it and inspecting how the amplitude is decreasing.
I did 5 simulations like this, each with a different input damping ratio zeta. I determined the damping from the output by calculating the damping decrement of the decaying tower oscillation.
The simulation results are in the table below (Sorry for the unclear format: it is a table with 1 row header column, 5 simulation result columns and 1 comment column. I could not figure out how to present it more neatly, any spaces and tabs I put disappear).
Simulation 1 Simulation 2 Simulation 3 Simulation 4 Simulation 5 comment
logd 0.01 0.02 0.03 0.04 0.05 Suppose we want these logd’s
zeta 0.0016 0.0032 0.0048 0.0064 0.0080 =logd/(2*pi)
TwrFADmp1 0.159 0.318 0.477 0.637 0.796 =zeta*100 (FAST input value)
x3 9.89E-01 9.79E-01 9.69E-01 9.59E-01 9.49E-01 3rd peak in graph below
x12 9.29E-01 8.64E-01 8.03E-01 7.46E-01 6.94E-01 15th peak in graph below
logd 0.0053 0.0105 0.0156 0.0209 0.0261 =1/12*ln(x3/x12)
zeta 0.0008 0.0017 0.0025 0.0033 0.0041 =logd/(2*pi)
Note the bold numbers, the first bold row is the input logd (which I convert to zeta and % before putting into FAST), the last bold row is the damping determined from the decaying oscillation. There is a factor 2 difference.
Can that be right?