Millikan Shake Test Results (test date : 28Aug2002)

Case has a website with a lot of this data in movie format, which is very useful to look at the phase of the two NS channels..

Mode Shapes:

each mode analysis for each loading condition is represented by 3 plots -

the modes we investigated were chosen using results from a frequency sweep
of the building from 1Hz - 10Hz performed in June 2002.
a frequency response curve consisting of force-normalised maximum amplitudes
for each of the three ROOF channels
a "snapshot" (ie synchronous time) of the building taken at the time of the
maximum displacement (for all 3 roof channels)
a plot of the channel displacements throughout the building for the
maximum and minimum roof excursion for each channel (not synchronous time)
--- The maximum and minimum displacements for the same time period
recorded at the TriNet station MIK, a tri-axial EpiSensor located on the
East side of the 9th floor, are also on this plot.

NOTES:

the data in these plots are bandpassed displacements
(bandpass is 1 octave above, and 1 octave below the shaking frequency)
2 channels in N-S direction, E side, at floors 2 and 8, are dead.
data from these channels is linearly interpolated from channels directly above and below

Primary Modes

Mode	Shake/Weight Combination	Resonance Peak, Hz	Normalised Displacement, (x10,000,000cm/N)	% of roof displacement due to basement rocking, translation
East-West 1	EW with (i) full weights, (ii) 2x2 side weights	1.11 1.14	~200(EW)	3%
North-South 1	NS with (i) full weights, (ii) 2x2 side weights	1.64(?) 1.67	~80(NS)	30%
Torsion 1	EW with (i) 2x2 side weights, NS with (i) full weights, (ii) 2x2 side weights	2.38 2.35(?) 2.38(?)	~25(NS) ~5(NS) ~5(NS)	2% ***
East-West 2	EW with (i) no weights	4.93	~2(EW)	1%
North-South 2	NS with (i) no weights	7.22(?)	~0.8(NS)	-21%
Torsion 2	EW with (i) no weights NS with (ii) no weights	6.57(?) 6.70(?)	~0.4(EW) ~0.15(NS) ~0.5(NS)	23% ***
East-West 3	EW with (i) no weights	7.83	~0.6(EW)	0%

*** : % of rotation recorded at roof due to basement rotation

Some Other Interesting Frequencies:

High Frequencies Data from EW shake
(i) no weights
And Data from NS shake
(ii) no weights
Mid Frequencies Data from NS shake
(i) no weights ~6.55Hz
(ii) no weights ~6.30Hz
(iii) no weights ~5.30Hz

Building Response for each Frequency

Plots are presented for each of the 6 shaking experiments:

NOTES:

the data in these plots are bandpassed accelerations
(bandpass is 1 octave above, and 1 octave below the shaking frequency)
2 channels in N-S direction, E side, at floors 2 and 8, are dead.
data from these channels is linearly interpolated from channels directly above and below

the plots have the following legend:

dashed blue line : raw output in cm/s^2
solid green line : component of motion due to translation and rocking of base
solid black line : motion of building due to bending and shear
the red line is the zero acceleration line.
thicker lines indicate maximum excursions over the 20 sec resonant period
thinner lines are from a sampling of 10 'snapshots' taken at any particular time

Experiment 1 - EW shake, full weights includes: Torsional 1, East West 1

Experiment 2 - NS shake, full weights includes: Torsional 1, North South 1

Experiment 3 - NS shake, 2x2 side weights includes: Torsional 1, North South 1

Experiment 4 - EW shake, 2x2 side weights includes: Torsional 1, East West 1

Experiment 5 - EW shake, no weights includes: High Frequencies (8.2 - 8.6Hz), East West 3, Torsional 2,
East West 2

Experiment 6 - NS shake, no weights includes: High Frequencies (8.2 - 8.6Hz), North South 2, Torsional 2,
Miscellaneous frequncies near 6.55, 6.30 and 5.30Hz.

Full Frequency Sweeps

Frequency Sweep In July a full frequency sweep of the building, from 10Hz to 1Hz, was performed with varying weights (we removed weights as we shook at higher frequencies) for both the EW and NS shaking directions. In this test we did not record the full 36 channel strong motion system. We only used the TriNet station MIK located on the West side of the 9th floor (triaxial EpiSensor), and a temporary Ranger located on the roof about 5m N and W of the SE corner of the building (just to the E of the shaker).
This test was primarily to identify the resonance peaks of interest for us to then shake the building at when we had access to the full instrumentation.
Plots are presented for each of the E-W, N-S and Vertical directions during both the E-W and N-S shakes:

John Clinton

Last modified: Tue Nov 26 13:30:01 PST 2002