CHAPTER 3 EXPERIMENTAL INVESTIGATION
Experimental evaluation of the ability of the semi-active PTMD to reduce floor acceleration versus its passive counterpart was conducted at the Civil and Environmental Engineering Structures and Materials Lab. An existing 8 ft x 30 ft floor was utilized for the experiment. A pendulum-tuned mass damper (PTMD), provided by ESI Inc., was used for this purpose.
Testing was conducted in three separate stages. The floor accelerations were measured with the bare-floor, the passive PTMD added, and then the semi-active PTMD. The results using passive PTMD serves as the baseline for comparison of the performance of the passive and semi-active PTMDs.
The test focused on the reduction in floor acceleration for the first bending mode of the floor. The examination of the acceleration response of the floor in the frequency domain (accelerance) is the primary tool for evaluation of the system performance. The optimal responses of the floor using passive and semi-active PTMDs are compared. The acceleration time histories for walking and heel drop are also investigated.
The robustness of the semi-active PTMD to off-tuning is presented. The floor’s acceleration response due to changes in the floor mass and PTMD spring location are presented. System response to changes in PTMD parameters, both passive and semi-active, is presented. An additional study of the passive and semi-active PTMDs performance when subjected to off-tuning by the addition of a group of people standing, sitting, and standing with knees bent on the test floor is also presented.
Test Specimens and System Identification
This section establishes components used during the experimental testing. Insight into the floor and PTMD geometry and dynamics are presented along with the overall system responses. Experimental equipment will be identified, and the overall arrangement of this equipment for passive and semi-active PTMD testing is given.
The floor is a common slender structural system susceptible to walking vibration. Requirements for floor vibration acceptability are given for floors in the AISC Design Guide 11 (1997).
The floor, shown in Figures 3.1 through 3.3, is composed of a 2 in. deck with a 5 in. normal-weight concrete slab supported on 18LH800 Vulcraft floor joists at 90 in. on center. The joist-girders are 14G4N5.85 Vulcraft long-span joists and span 30 ft. HSS 5x3x5/16 sections were added between each pair of joists seats in “an attempt to attain full composite action” (Warmoth 2002). The joist-girders are placed on W18 hot-rolled sections without any physical connections. The W18’s sit perpendicular to the 30 ft span and span the 8 ft. width between the joist-girders. No pin connection is made at the supports, just bearing. The joist girder bears on the W18 and the wide-flange section bears on the lab floor.
The PTMD used in this study is presented in Figure 3.4 which was provided by ESI Engineering. It offers advantages over the standard sprung mass configuration (Den Hartog 1947). The spring position is adjustable, and also the configuration amplifies the inertial force, spring force, and damper force per the distance squared from the support. The shallow geometry allows it to be placed within floor cavities. A disadvantage to the PTMD is the partial effective mass; this is shown analytically in Chapter 2.
Four stacks of steel plates provide the mass. Two springs, one on each side of the PTMD, provide an adjustable stiffness via the attached slotted plate. Each spring is composed of two springs, one inside the other. They act in series at each spring support. Three viscous dashpots are located at the end of the PTMD, opposite the pin support. The dashpots are manufactured by Airpot Inc., Figure 3.5. The specification is given in the appendix.
A special pin was implemented into this PTMD. It is referred to as a “rigid cross-bar” pin and offers very little friction at the joint. This is significant for two reasons. First, the PTMD can begin to approach the idealization assumed in analytical models. Secondly, this significantly reduced friction found in typical connections and allows the PTMD to respond readily to the floor response and quell vibration more effectively. Any “slop”, or friction, will reduce the effectiveness of PTMD and it is a critical point in a PTMD designed to be effective for small displacements.
The “MR sponge damper” was provided by the Lord Corporation. The damper characteristics are discussed in Chapter 2 from page 49. This damper “provides the necessary on-state damping force when energized and has a reasonably low off-state damping” (Koo 2002). An estimate of the required damping force was made and it was decided that it was within the range of operation of the MR damper. Figure 3.6 shows the allowable range that the sponge damper can provide, and by inspection of the figure, it is sufficient.
The MR damper used is shown in Figure 3.7. The electrical leads supply the necessary current across the damper to energize, the grey, absorbent matrix to obtain a desired level of damping. It is an absorbent matrix to contain the MR fluid by capillary action: open-celled foam, felt, or fabric. The sponge, typically polyurethane foam, keeps the MR fluid in the region of the applied magnetic field. The sponge surrounds a steel bobbin and coil, and it is attached to a shaft that moves axially within a hollow shaft. There are no seals or bearings.
To accommodate the MR damper, an end plate with clevis was designed. This is presented in Figure 3.8. A second clevis was attached to the floor. The damper is connected via pins at the top and bottom. Washers were placed on either side as infill to prevent any lateral motion of the PTMD as the system vibrated.
An impulse force hammer and shaker were used to impart a force on the system. Two accelerometers and a force plate are used to measure the response of the system and the force on the floor respectively. A data acquisition was used for signal processing and monitoring of the system.
A PCB Impulse Force Hammer, Figure 3.9, was used to generate an impulse force. This force excited a desired band of frequency to obtain the frequency response of the system. The force hammer utilizes a piezo-based transducer to capture the imparted force. The sensitivity of the hammer is 0.85 mV/lb.
The primary force generator used on the floor was an APS Electro-Seis, Model 400 Shaker, see Figure 3.10. The dynamic mass of the shaker is provided by four blocks weighing 67.4 lbs suspended by rubber bands. The core of the shaker weighs 170.6 lbs for a total shaker weight of 238 lbs. It has a frequency range from 0 to 200 Hz. From 0.10 to 20 Hz the shaker has a force rating of 100 lbs. The floor was typically excited using a chirp swept from 4 to 15 Hz.
A chirp is a sinusoidal function with linearly increasing frequency as a function of time. Equation 3.2 is the mathematical model used throughout this research. Typically, the chirp is swept from zero to twice the natural frequency of the floor. It is shown in both and time and frequency in Figure 3.11.
An APS, Dual-Mode, Model 144 Amplifier, Figure 3.12, was used to send a voltage signal to the shaker. The desired voltage signal, in the form a desired force excitation, was provided by the data acquisition system. The desired voltage signal is sent to the amplifier that converts this signal to current to drive the shaker.
To measure the response of the floor, two PCB 393C accelerometers were used. A typical accelerometer is shown in Figure 3.13. The accelerometers are piezo-based, have a sensitivity of 1 V/g and can cover a frequency range from 0.025 to 800 Hz.
CHAPTER 1: INTRODUCTION AND LITERATURE REVIEW
1.2 Literature Review.
1.3 Magnetorheological damper
1.4 Purpose of Research
1.5 Scope of Research
CHAPTER 2: ANALYTICAL INVESTIGATION
2.2 System Representation
2.3 Implementation of the MR into the PTMD
2.4 MR Control Policy – Displacement Based Ground-hook
2.5 Analytical Simulation
2.7 Comparison of Off-Tuning Effects for the Passive and Semi-Active PTMDs
2.8 Passive versus Semi-Active Walking Response
CHAPTER 3: EXPERIMENTAL INVESTIGATION.
3.2 Test Specimens and System Identification
3.3 Data Acquisition
3.4 System Configuration
3.5 Signal Processing Methods
3.6 Experimental Control Policy
3.7 Experimental Optimization
3.8 Experimental System Off-Tuning.
3.9 Floor Response to Human Excitation
CHAPTER 4: CONCLUSIONS AND RECOMMENDATIONS
4.1 Summary and Conclusions
LIST OF REFERENCES
GET THE COMPLETE PROJECT
Application of Magneto-Rheological Dampers in Tuned Mass Dampers for Floor Vibration Control