The Woodruff School

SUMMARY

Broadband, variable, and random excitations are often suppressed using active vibration absorbers (AVAs). While AVAs can be effective, they also are expensive and subject to instability when the disturbance is ill defined. A state-switched absorber (SSA) can be used for these same vibration classes while reducing the expense and instability because an SSA is only allowed to be active at discrete instances. SSAs are spring-mass-damper devices in which at least one element is controllably variable. The work presented in this dissertation evaluates the properties of magnetorheological elastomers (MREs) to assess their use in SSAs as variable springs. MREs are elastomers doped with magnetically permeable material, generally iron. They are modeled as lossy springs, and have stiffness and loss factor components. Natural frequency and stiffness behavior, and their relationships to static displacement, iron content, and forcing frequency and amplitude were determined. Loss factors were found to be independent of MRE content, configuration, and static displacement. This was confirmation that MREs are in fact controllable springs. Natural frequencies changed in the presence of magnetic fields by as much as 360%. The corresponding change in static displacement could not account for this frequency change. Transient data was found by determining the length of time it took for an MRE to achieve quasi-steady state oscillation behavior when subjected to a harmonic excitation. This time was referred to as the characteristic response time. The characteristic response time correlated to the ratio of the forcing frequency to the zero-field natural frequency. When a magnetic field was turned on, the characteristic response time on average was found to be consistently longer than when the magnetic field was turned off, regardless of iron content or configuration. The difference between these two characteristic response times is caused by the particles' mechanics. To form a chain, a magnetic field must both be set up, and particles must move to join together. When a chain is broken, the magnetic field must merely be removed. However, this difference gives opportunities for future research to be conducted on controlling MREs' transient responses.

SUBJECT:	Ph.D. Dissertation Defense

BY:	Anne-Marie Lerner

TIME:	Thursday, August 14, 2008, 1:30 p.m.

PLACE:	MRDC Building, 4211

TITLE:	Stability and controllability investigation of MREs in longitudinal vibration absorbers

COMMITTEE:	Dr. Kenneth A Cunefare, Chair (ME) Dr. Nader Sadegh (ME) Dr. Christopher Lynch (ME) Dr. Reginold DesRoches (CE) Dr. Massimo Ruzzene (AE)

SUMMARY Broadband, variable, and random excitations are often suppressed using active vibration absorbers (AVAs). While AVAs can be effective, they also are expensive and subject to instability when the disturbance is ill defined. A state-switched absorber (SSA) can be used for these same vibration classes while reducing the expense and instability because an SSA is only allowed to be active at discrete instances. SSAs are spring-mass-damper devices in which at least one element is controllably variable. The work presented in this dissertation evaluates the properties of magnetorheological elastomers (MREs) to assess their use in SSAs as variable springs. MREs are elastomers doped with magnetically permeable material, generally iron. They are modeled as lossy springs, and have stiffness and loss factor components. Natural frequency and stiffness behavior, and their relationships to static displacement, iron content, and forcing frequency and amplitude were determined. Loss factors were found to be independent of MRE content, configuration, and static displacement. This was confirmation that MREs are in fact controllable springs. Natural frequencies changed in the presence of magnetic fields by as much as 360%. The corresponding change in static displacement could not account for this frequency change. Transient data was found by determining the length of time it took for an MRE to achieve quasi-steady state oscillation behavior when subjected to a harmonic excitation. This time was referred to as the characteristic response time. The characteristic response time correlated to the ratio of the forcing frequency to the zero-field natural frequency. When a magnetic field was turned on, the characteristic response time on average was found to be consistently longer than when the magnetic field was turned off, regardless of iron content or configuration. The difference between these two characteristic response times is caused by the particles' mechanics. To form a chain, a magnetic field must both be set up, and particles must move to join together. When a chain is broken, the magnetic field must merely be removed. However, this difference gives opportunities for future research to be conducted on controlling MREs' transient responses.