Learning Forum

[dbruneau]

Hi, I am trying to run an undamped Jeffcott rotor at speeds (1700-2000 rad/sec) above my critical speed (which is around 1070 rad/sec), but the support loads are unreasonably high (6000kN) in my transient analysis. Deformations are also unreasonably high. Below or slightly under critical speed they seem reasonable. I have tried modifying the majority of available parameters (ramping up the rotational velocity vs starting at high speed, decreasing the timestep, modifying internal damping, using line vs. solid models). The main reason for running the analysis in transient is to eventually estimate the bearing loads, including damping forces. Which analysis settings are important to include for supercritical rotors?

[Dave Looman]

In an undamped system, the steady state response at resonance is infinite, so the only thing limiting reaction forces is how quickly you pass through the critical speed. It would probably be best not to use large deflection as it will artificially increase the centrifugal force associated with lateral displacements. The rotational velocity is assumed to be centered on the undeformed axis when in actuality the rotor will spin about its center of mass. As a result lateral displacements produce huge centrifugal forces which aren't physical. Example 7.6 in the Rotordynamics Guide is a transient start-up analysis.n

[sbo]

nI am surprised to see that much reaction force in your analysis. Because there is no force in your analysis and only rotational velocity is defined. If you want see unbalance response in transient analysis, you should look transient analysis example in ansys manual. I cant remember example name. n