Passive Balancer Development for Rotor Systems in Icing Conditions
Open Access
- Author:
- Blessington, Ryan
- Graduate Program:
- Aerospace Engineering
- Degree:
- Master of Science
- Document Type:
- Master Thesis
- Date of Defense:
- August 05, 2020
- Committee Members:
- Jose Palacios, Thesis Advisor/Co-Advisor
Amy Ruth Pritchett, Program Head/Chair
Edward Smith, Committee Member
Amy Ruth Pritchett, Committee Member - Keywords:
- Passive
Balancing
Passive Balancer
Structural
Dynamics
Structural Dynamics
Icing
Adverse environment
Rotor
rotorcraft
helicopter
imbalance
mass balance - Abstract:
- For a rotary system, the minimization of detrimental fixed-frame forces and rotor vibration is critically important to ensuring the safety of the system and its operator. For modern VTOL aircraft, rotors are precisely fabricated to certify that their mass is evenly distributed about their axis of rotation. However, significant mass eccentricity can arise in a rotor system, operating in adverse icing conditions, that experiences an asymmetric ice shed event, as ice accretes on the blades and is subsequently removed by centrifugal force. This mass eccentricity creates large dynamic loading and vibration, leading to shortened fatigue life, component failure, or catastrophic failure. As VTOL flight continues to evolve, there is a need for a low-power, low-complexity solution to rotor mass eccentricity that arises during supercritical rotor operation. A solution for mass eccentric vibration is presented in the form of a passive balancing device. A passive balancing device consist of several masses that are free to move circumferentially along a track concentric to the axis of rotation. Beyond the shaft first flexural natural frequency speed, referred to as supercritical operation, the phase of the system response is oriented 180 degrees relative to the asymmetric loading caused by the imbalance, which causes the masses to assume positions opposed to the direction of the imbalance and reduces vibration. A comprehensive model of this nonlinear phenomenon was developed in previous studies and it was experimentally validated for a partitioned ball-type balancer on small scale applications. In the study outlined in this thesis, a partitioned ball-type passive balancing device was investigated as a possible solution to address the fixed-frame loading and rotor vibration created by asymmetric ice shed events. A mathematical model of a flexible, passively balanced rotor system was employed for the prediction of system fixed frame loading in response to a given mass imbalance condition. Three rotor configurations were considered for icing experiments in the Adverse Environment Rotor Test Stand (AERTS), consisting of a representative rotor mounted to a cantilevered shaft inside an icing chamber. Icing experiments consisted of spinning up the rotor to its supercritical operating region and introducing realistic icing conditions in the continuous icing envelope to observe the system vibration response by measuring the fixed frame loading following each ice shed event. An analytical model for the ice mass of each shed event was developed from commercially available techniques to facilitate the modeling of each observed mass imbalance condition. An unbalanced rotor configuration was tested to provide the baseline for the vibration response without a passive balancing device. An initial device was designed from design trends observed in literature that dictate maximizing balancing authority, minimizing distance between the balancer and imbalance planes, and implementing multiple track partitions. The initial device was fabricated and tested to determine its vibration reduction performance and identify key design parameters limiting its effectiveness. The final design was formulated from an iterative design process, which varied key design parameters to improve predicted performance, provided by the model, until convergence. The final design was fabricated and tested to determine its vibration reduction performance. Experimental results from all configurations were compared to model predictions to validate the mathematical model of the system. The performance of the final passive balancer was compared to the initial to determine the effectiveness of implementing an iterative design process based on parametric analysis. Experiments conducted on the unbalanced rotor configuration revealed a closely linear correlation between mass imbalance magnitude and fixed frame loading, with some mass imbalance conditions approaching the design loading limit of the system. The model prediction of the unbalanced rotor configuration loading was 18.7% different on average, underpredicting the load in most cases. Experimental results for the initial passive balancer configuration exhibited a close correlation to unbalanced rotor results in trend and magnitude. The model prediction of the initial passive balancer configuration loading was 22.5% different on average, overpredicting values for the higher imbalance cases. It was calculated that the implementation of the initial passive balancer increased fixed frame loading by 8.63% on average, illustrating that a design formulated without numerical analysis to precisely tune parameters can be detrimental to the system. Examining loading time history and balancing mass positions, it was determined that excess resistance force between the ball and track outer wall were fixing the masses in place prior to reaching supercritical speeds where passive balancing behavior occurs. In addition, the balancer was mounted too close to the fixed boundary to allow for significant displacement in response to asymmetric loading, which is necessary for passive balancing. Key design parameters were identified as the passive balancer track radius, axial position along the shaft, frictional dynamics, and balancing authority. Parametric analysis was conducted on each parameter successively in an iterative design loop until performance predicted by the model converged. The final design was fabricated to conform to the results of the parametric analysis for the most ideal model prediction. Experimental results for the final passive balancer illustrated the effectiveness of the device with a significant reduction in fixed-frame loading in all observed mass imbalance conditions, as compared to prior configurations. The model prediction of the final passive balancer configuration loading was 64.8% different on average; however, the practical significance of this value is diminished when considering the magnitude of the observed loading. The model prediction was within 20 lbf. on average, which is well within the loading tolerance limits for most rotor systems, illustrating that the model remains practically accurate. It was calculated that the implementation of the final passive balancing device reduced fixed frame loading by 78.8% on average. The significant performance improvement between the initial and final passive balancing devices illustrates that a passive balancing device, when carefully designed using numerical analysis, can effectively reduce mass eccentric rotor vibration in response to asymmetric ice shed events.