01

DRAFT – WITHOUT PLOTS – 10/8/01 COMPREHENSIVE INVESTIGATION OF NONLINEAR SITE RESPONSE CENTRIFUGE DATA REPORT FOR DKS04 D.K. Stevens, D.W. Wilson, and B.L. Kutter Center for Geotechnical Modeling Data Report UCD/CGMDR-01/03 Date:

October 2001

Date of testing:

January 2000

Project:

Comprehensive Investigation of Nonlinear Site Response: Collaborative Research with UC San Diego, and UC Davis

Contract number:

99HQGR0019

Sponsor:

United States Geological Survey (USGS)

Related reports:

UCD/CGMDR-99/02: Centrifuge data report for DKS02 UCD/CGMDR-99/03: Centrifuge data report for DKS03 UCD/CGMDR-00/04: Centrifuge data report for DKS05

ACKNOWLEDGMENTS The DKS04 model test was funded by the USGS. The contents of this report are not necessarily endorsed by the USGS. The authors would like to acknowledge the suggestions and assistance of Ahmed-Waeil Elgamal, Tao Lai, Kevin Hazleton, Jon Boland, Prianshu Singh, Kiran Manda, Amy Smith, Key Rosebrook, Nason McCullough, Chad Justice, Tom Coker, Tom Kohnke, and Dennis O’Brien. Development of the large centrifuge at UC Davis was supported primarily by the National Science Foundation, NASA, and the University of California. Additional support was obtained from Tyndall Air Force Base, the Naval Civil Engineering Laboratory, and Los Alamos National Laboratories. The large shaker was funded by the California Department of Transportation, the Obayashi Corporation, the National Science Foundation, and the University of California. CONDITIONS AND LIMITATIONS Permission is granted for the use of these data for publications in the open literature, provided that the authors and sponsors are properly acknowledged. It is essential that the authors be consulted prior to publication to discuss errors or limitations in the data not known at the time of the release of this report. In particular, there may be later releases of this report. Questions about this report may be directed by email to [email protected].

i

TABLE OF CONTENTS COVER PAGE ..........................................................................................................................................i PURPOSE AND CONFIGURATION OF THE TEST ...............................................................................1 SOIL PROPERTIES .................................................................................................................................1 PORE FLUID PROPERTIES ....................................................................................................................1 SCALE FACTORS ...................................................................................................................................6 INSTRUMENTATION AND MEASUREMENTS....................................................................................7 KNOWN LIMITATIONS OF RECORDED DATA AND MODEL......................................................... 13 DESCRIPTION OF DISPLACEMENT-COMMAND FILES FOR MODEL SHAKING ......................... 16 ORGANIZATION OF DATA FILES AND PLOTS ................................................................................ 16 REFERENCES ....................................................................................................................................... 24 FIGURES ............................................................................................................................................... 25 DATA PLOTS:

(NOT INCLUDED IN THIS REPORT DRAFT)

................ 29

TIME HISTORIES FOR ALL INSTRUMENTS: NINE SELECTED SHAKING EVENTS .......................................................................................................... PLOT SECTION 1 TIME HISTORIES FOR THE CENTRAL VERTICAL ARRAY AND AVERAGE BASE MOTION: ALL SHAKING EVENTS .................................................... PLOT SECTION 2 ACCELERATION RESPONSE SPECTRA FOR ALL ACCELEROMETERS: ONE SELECTED SHAKING EVENT.............................................................. PLOT SECTION 3 APPENDIX A: DESCRIPTION OF CENTRIFUGE FACILITIES .......................................................A-1

PURPOSE AND CONFIGURATION OF THE TEST This was the third test in a comprehensive experimental/analytical investigation of nonlinear site response. Test results will aid in analyzing the impact of input motion magnitude and frequency content on the nonlinear site response problem. Prepared with Nevada Sand, the DKS04 model was constructed to nearly full-container height in the Flexible Shear Beam 2 (FSB2) container and saturated with a viscous fluid. Instrumentation included longitudinal, transverse, and vertical arrays of accelerometers; pore pressure transducers; vertical and horizontal displacement transducers; and strain gauges. Model configuration is shown in Figure 1. This model configuration is meant to be identical to that used in DKS02, with the notable exception that the current model was saturated. Model testing was performed using the 9-meter radius geotechnical centrifuge at UC Davis. A total of 70 shaking events were applied to the model. Shaking was applied parallel to the long sides of the model container. Ground motions included various frequency sweeps; scaled versions of the Santa Cruz ground motion (UCSC/Lick lab, Ch. 1: 90 degrees; CSMIP website, http://www.consrv.ca.gov/dmg/csmip/) recorded during the 1989 Loma Prieta earthquake; and a scaled version of the Port Island ground motion (83-m depth, NS direction) recorded during the 1995 Kobe earthquake. Actual centrifugal scaling factors (see Scale Factors for method of determination) used for test events are listed in Table 1 and are shown in Figure 2 for shaking events. A chronology of the entire test program is given in Table 1, and the column headings for this table are described under Organization of Data Files and Plots. SOIL PROPERTIES Some of the soil properties of the sand are summarized in Table 2. Mechanical grain-size analyses of the Nevada Sand batch used for this model were previously conducted according to ASTM D422 for the DKS03 model (Stevens et al., 1999b). Figure 3 shows the Nevada Sand grain-size distribution reported in the DKS03 data report. Average relative density is based on measured unit weights and on the results from maximum and minimum density measurements, performed according to ASTM D4253-83 and D4254-83 by Woodward-Clyde (1997). PORE FLUID PROPERTIES The pore fluid used to saturate the DKS04 model was a 1.1% hydroxypropyl methylcellulose (HPMC) solution, where the solution concentration is expressed as the percentage HPMC of the entire solution by mass. Reference was made to a paper by Stewart et al. (1998) in determining the proper proportion of HPMC powder to use to achieve a fluid viscosity of approximately ten times that of pure water. The HPMC powder used in the solution was Methocel, premium grade F, from The Dow Chemical Company. It was mixed with de-ionized water and a quantity of benzoic acid

1

Table 1. Summary of the DKS04 nonlinear site response model test, January 2000 (see Organization of Data Files and Plots for an explanation of column headings, file extensions, channel gain lists, input files, and other info. in this table) Channel gain list (cgl) used

Output file

Rough model peak-to-peak base acceleration (g)

-

-

-

-

-

-

Vertical p-wave array cgl -

vptop1 -

-

-

-

Vertical p-wave array cgl

vptop2

-

Date

Time

Centrifugal scaling factor (see Scale Factors )

Input file name

Thurs., 1/6

-

1

-

Instrument check

-

4:15 PM -

1 1

-

Manual hammer (vertical hit) on surface block; Saturation check Instrument check

4:45 PM

1

-

Manual hammer (vertical hit) on surface block; Saturation check

Mon., 1/10

Tues., 1/11

Event description

West ampl. East ampl. factor factor

-

1

-

Instrument check

-

-

-

-

-

2:08 AM 2:20 AM

1 1

-

Instr. check: Large air hammer (remote switch up) Instr. check: Large air hammer (remote switch down)

-

-

Small shakes cgl Small shakes cgl

inchk1 inchk2

-

2:58 AM 3:07 AM

1 1

-

Instr. check: Mini air hammers (remote switch up) Instrument check

-

-

Small shakes cgl Slow data cgl

inchk3 inchk1.slw

-

10:32 AM

1 1

-

Instrument check Instr. check: Large air hammer (remote switch up)

-

-

Small shakes cgl

inchk4

-

10:48 AM 11:25 AM

1 1

-

Instrument check Instrument check

-

-

Slow data cgl Slow data cgl

inchk2.slw inchk3.slw

-

2:58 PM

1 1

-

Instrument check Instr. check: Large air hammer (remote switch up)

-

-

Small shakes cgl

ham1

-

3:01 PM 3:37 PM 3:42 PM 3:45 PM

1 to 9.3 9.3 9.3 9.3

-

Spin up, 1 to 9.3 g Mini air hammers (remote switch down) Mini air hammers (remote switch up) Mini air hammers (remote switch down)

-

-

Slow data cgl Horizontal p-wave array cgl #1 Upper central array cgl #1 Lower central array cgl #1

dks04_a.slw vp_1 vs_1 vs_2

-

4:33 PM 4:52 PM 5:04 PM 5:17 PM 5:23 PM

9.3 9.2 9.5 9.5 9.5

2sc.txt 2sc.txt 2sc.txt lsc.txt lsc.txt

Santa Cruz 0.4g at20g, 1/2dt and dt, corrected Santa Cruz 0.4g at20g, 1/2dt and dt, corrected Santa Cruz 0.4g at20g, 1/2dt and dt, corrected Santa Cruz 0.4g at20g, 2dt, corrected Santa Cruz 0.4g at20g, 2dt, corrected

0.1 0.2 0.45 0.2 0.1

0.1 0.2 0.45 0.2 0.1

Small shakes cgl Small shakes cgl Small shakes cgl Small shakes cgl Small shakes cgl

dks04_1 dks04_2 dks04_3 dks04_4 dks04_5

0.4 0.9 3 1.5 0.4

5:30 PM 5:39 PM 6:06 PM 6:09 PM 8:39 AM 8:58 AM -

9.9 9.5 9.6 9.6 to 1 1 1 1 1

swp6.txt swp7.txt swp7.txt -

Sweep 80 to 200, 2 g peak Sweep 160 to 400, 2 g peak Sweep 160 to 400, 2 g peak Spin down, 9.6 to 1 g Instr. check: Large air hammer (remote switch up and down) Instrument check Instr. check: Large air hammer (remote switch up and down) Instrument check

0.3 0.8 0.8 -

0.3 0.8 0.8 -

Small shakes cgl Small shakes cgl Small shakes cgl Slow data cgl Small shakes cgl Small shakes cgl -

dks04_6 dks04_7 dks04_8 dks04_b.slw ham2 ham3 -

0.6 0.9 -

2

Comments / Details

DC amp. gains were set to 10 for A24 and A16 A24 gave good signal, but A16 gave none; tech filters were set with a cut-off freq. of 20 kHz for all saturation check events and for all "vs_" and "vp_" mini air hammer tests Switched A16 from DC amp. ch. 33 to ch. 34 Both A24 and A16 gave good signal, but A16's is approx. 12 times smaller than A24's; pwave velocity calculated at approx. 500 m/s between A24 and A16 DC amp. channels zeroed; all accelerometers' cables hooked up to DC amp. and gains set to 10; A24 switched from ch. 32 to 37, and A16 from ch. 34 to 49; cables from linear pots., strain gauges, and ppt's connected to zeroed PVL amp. channels; linear pots. manually zeroed to output near-zero voltage initially, except for D2; strain gauges ch. zeroed; PVL amp. gains set: linear pots., 2; strain gauges, 1000; ppt's, 100 No hit recorded; tech filters were set with a cut-off freq. of 800 Hz for all instrument-check hammer tests and all shaking events; all air hammer tests were run with air supply set at 40 psi Hit recorded; A18 and A20 output no signal A18 and A20 again output no signal from hit; CONVENTION: Triggering the mini air hammers' remote switch in the "up" direction caused the "north" air manifold to be charged with air, which in turn forced the hammers' pistons to the "south", and vice-versa Four accels. switched to different DC amp. ch.: A18 from ch. 33 to 52, A20 from ch. 34 to 62, A7 from ch. 52 to 33, and A28 from ch. 62 to 34 A18, A20, and A28 output good signals; A7 (ch. 33) output only low-amplitude noise Moved linear potentiometers' shafts manually, full range and slowly; all four appeared to work well Similar input and response as for inchk2.slw Re-zeroed linear potentiometers and strain gauges; pore pressure transducers were rezeroed by setting at approx. -2 volts amp. ch. output for all ppt's except for P2 and P7, which were set at -1 volt, thus allowing for vacuum-to-liquefaction pressure range during testing; water is approx. up to the avg. level of the soil surface; ball valve for water was opened to allow water to run into model A7 (ch. 33) still output only noise Spin up; start spinning Day 1; rpm counted manually and matched gauge value well; D1 (PVL amp. ch. 17) output zero-displacement voltage for this event and all others (spinning/shaking) through dks04_k.slw MAH4 hit before MAH3 No MAH4 hit Started small shakes; A31 (ch. 63) showed drifting behavior during this and most other shaking events; A7 (ch. 33) output only noise for this and all remaining test events; DC amp. ch. 33 is probably broken

Rough peak-to-peak base accel. for swp4.txt, swp6.txt, and swp7.txt events were obtained by observing only the main portion of the output time histories, not including any initial and final spikes Sequencer stopped early; no shaking occurred Spun down completely; end spinning Day 1 A7 (ch. 33) and A20 (ch. 62) output no signal A20 switched from ch. 62 to 34; A28 switched from ch. 34 to 62 A20 (ch. 34) still not working; may be broken A20 switched from ch. 34 to 59; A30 switched from ch. 59 to 34

Table 1, cont. Summary of the DKS04 nonlinear site response model test, January 2000 (see Organization of Data Files and Plots for an explanation of column headings, file extensions, channel gain lists, input files, and other info. in this table) Date

Time


Input file name

9:06 AM

1

-

9:22 AM 10:25 AM 10:58 AM 11:08 AM 11:16 AM 11:21 AM 11:37 AM 11:38 AM 11:40 AM 11:45 AM 11:46 AM 11:57 AM 11:59 AM 12:00 PM

1 to 1.4 to 1 1 to 12.9 12.9 13.2 12.8 13.8 13.8 13.8 13.8 13.8 13.8 to 24.7 24.7 24.7 24.7

swp4.txt swp4.txt swp6.txt swp7.txt -

12:16 PM 12:39 PM 12:53 PM 12:59 PM 1:09 PM 1:12 PM 1:43 PM 1:47 PM 1:55 PM 1:59 PM 2:10 PM 2:10 PM 2:11 PM 2:15 PM 2:50 PM

Event description

Instr. check: Large air hammer (remote switch up and down)


Channel gain list (cgl) used

Output file


-

-

Small shakes cgl

ham4

-

-

Spin up, 1 to 1.4 g, and Spin down, 1.4 to 1 g Instrument check Spin up, 1 to 12.9 g Sweep 50 to 125, 2 g peak Sweep 50 to 125, 2 g peak Sweep 80 to 200, 2 g peak Sweep 160 to 400, 2 g peak Mini air hammers (remote switch up) Mini air hammers (remote switch down) Mini air hammers (remote switch up) Spinning at 13.8 g Spin up, 13.8 to 24.7 g Mini air hammers (remote switch down) Mini air hammers (remote switch up) Mini air hammers (remote switch down)

0.5 0.8 0.6 1.2 -

0.5 0.8 0.6 1.2 -

Slow data cgl Slow data cgl Small shakes cgl Small shakes cgl Small shakes cgl Small shakes cgl Horizontal p-wave array cgl #1 Upper central array cgl #1 Lower central array cgl #1 Lower central array cgl #1 Slow data cgl Horizontal p-wave array cgl #1 Upper central array cgl #1 Lower central array cgl #1

dks04_c.slw dks04_d.slw dks04_9 dks04_10 dks04_11 dks04_12 vp_2 vs_3 vs_4 dks04_e.slw dks04_f.slw vp_3 vs_5 vs_6

0.4 0.8 1 1.3 -

24.7 25.4 25.5 25.5 25.2 25.1 25.2 25.1 25.2 25.1 25.1 25.1 25.1 25.1 to 47.2 47.2

2sc.txt 2sc.txt lsc.txt lsc.txt swp7.txt swp7.txt swp4.txt swp4.txt swp6.txt swp6.txt -

Santa Cruz 0.4g at20g, 1/2dt and dt, corrected Santa Cruz 0.4g at20g, 1/2dt and dt, corrected Santa Cruz 0.4g at20g, 2dt, corrected Santa Cruz 0.4g at20g, 2dt, corrected Sweep 160 to 400, 2 g peak Sweep 160 to 400, 2 g peak Sweep 50 to 125, 2 g peak Sweep 50 to 125, 2 g peak Sweep 80 to 200, 2 g peak Sweep 80 to 200, 2 g peak Mini air hammers (remote switch down) Mini air hammers (remote switch up) Mini air hammers (remote switch down) Spin up, 25.1 to 47.2 g Mini air hammers (remote switch up)

0.35 0.2 0.2 0.15 0.7 0.4 0.7 0.9 0.6 0.3 -

0.35 0.2 0.2 0.15 0.7 0.4 0.7 0.9 0.6 0.3 -

Small shakes cgl Small shakes cgl Small shakes cgl Small shakes cgl Small shakes cgl Small shakes cgl Small shakes cgl Small shakes cgl Small shakes cgl Small shakes cgl Horizontal p-wave array cgl #1 Upper central array cgl #1 Lower central array cgl #1 Slow data cgl Horizontal p-wave array cgl #1

dks04_13 dks04_14 dks04_15 dks04_16 dks04_17 dks04_18 dks04_19 dks04_20 dks04_21 dks04_22 vp_4 vs_7 vs_8 dks04_g.slw vp_5

1.2 0.6 1 0.6 1.6 0.8 0.6 1.3 1.3 0.6 -

2:51 PM 2:52 PM

47.2 47.2

-

-

-

Upper central array cgl #1 Lower central array cgl #1

vs_9 vs_10

-

3:05 PM 3:13 PM 3:19 PM 3:23 PM 3:45 PM 3:50 PM 3:56 PM 4:06 PM 4:07 PM 4:08 PM 4:13 PM 5:28 PM

47.2 47.3 47.2 47.2 47.2 47.1 47.1 47.1 47.1 47.1 47.1 to 1 1

2sc.txt 2sc.txt swp7.txt swp7.txt lsc.txt swp4.txt swp6.txt -

Santa Cruz 0.4g at20g, 1/2dt and dt, corrected Santa Cruz 0.4g at20g, 1/2dt and dt, corrected Sweep 160 to 400, 2 g peak Sweep 160 to 400, 2 g peak Santa Cruz 0.4g at20g, 2dt, corrected Sweep 50 to 125, 2 g peak Sweep 80 to 200, 2 g peak Mini air hammers (remote switch down) Mini air hammers (remote switch up) Mini air hammers (remote switch down) Spin down, 47.1 to 1 g Manual hammer (vertical hit) on surface block; Saturation check

0.4 0.2 0.75 0.4 0.25 1 0.7 -

0.4 0.2 0.75 0.4 0.25 1 0.7 -

Small shakes cgl Small shakes cgl Small shakes cgl Small shakes cgl Small shakes cgl Small shakes cgl Small shakes cgl Horizontal p-wave array cgl #1 Upper central array cgl #1 Lower central array cgl #1 Slow data cgl Vertical p-wave array cgl

dks04_23 dks04_24 dks04_25 dks04_26 dks04_27 dks04_28 dks04_29 vp_6 vs_11 vs_12 dks04_h.slw vptop3

1.9 0.9 2.1 0.9 1.7 1.4 1.7 -

5:29 PM

1

-

Manual hammer (vertical hit) on surface block; Saturation check

-

-

Vertical p-wave array cgl

vptop4

-

Mini air hammers (remote switch down) Mini air hammers (remote switch up)

3

Comments / Details

A20 (ch. 59) appears to work now; A30 (ch. 34) also works; these two instruments swapped DT entry numbers for this event ONLY; then cgl was updated and DT entry nos. for A20 and A30 were 10 and 30, resp., hereafter for small and large shakes cgls Start spinning Day 2; air-over-oil pump stopped; remained at 12 rpm to reset sequencer; pump still not working; spun down completely; data file contains a 12-to-0-rpm appended portion with a time gap in-between spin-up and spin-down Air-over-oil pump observed and serviced; okay for now Spin up

No MAH4 hit MAH4 hit before MAH3 Wrong cgl for slow data acquisition Spin up MAH3 hit before MAH4 MAH3 hit before MAH4 Base accelerations on east load frame were slightly larger than on the west; A12 (ch. 42 for now) output only noise for this event through shaking event 22

MAH3 hit before MAH4 MAH3 hit before MAH4 Spin up; on camera, water appeared to totally cover surface now No MAH4 hit; A20 (ch. 59 for now) output only noise for this and all remaining miniair hammer tests (vs_9 through vs_26) No MAH4 hit A20 (ch. 59 for now) output only noise for this and all remaining shaking events (23 through 70); instrument may be broken; A12 (ch. 42) okay for now; A2 (ch. 55) drifting Ball valve for water was closed to stop water from running into model

Finished small shakes No MAH4 hit MAH3 hit before MAH4 Spun down completely; end spinning Day 2 A16 signal very small P-wave velocity calculated at approx. 1500 m/s between A24 and A16; A16 signal very small


Wed., 1/12

Time


Input file name

Event description



Output file


8:49 PM 8:50 PM

1 1 1

-

Instrument check Mini air hammers (remote switch up) Mini air hammers (remote switch down)

-

-

Horizontal p-wave array cgl #1 Horizontal p-wave array cgl #1

vp_7 vp_8

-

10:22 PM

1

-

Mini air hammer: MAH1 ONLY! (remote switch up)

-

-

Horizontal p-wave array cgl #2

vp_9

-

10:39 PM

1

-

Mini air hammer: MAH4 ONLY! (remote switch down)

-

-

Horizontal p-wave array cgl #2

vp_10

-

8:54 AM -

1 1

-

Instr. check: Large air hammer (remote switch up and down) Instrument check

-

-

Small shakes cgl -

ham5 -

-

9:12 AM 9:24 AM

1 1 1

-

Instr. check: Large air hammer (remote switch up and down) Instrument check Instr. check: Large air hammer (remote switch up and down)

-

-

Small shakes cgl Small shakes cgl

ham6 ham7

-

9:53 AM

1 1 to 8.7

-

Instrument check Spin up, 1 to 8.7 g

-

-

Slow data cgl

dks04_i.slw

-

10:51 AM 10:58 AM 11:00 AM 11:00 AM

8.7 8.7 8.7 8.7 8.7

-

Instrument check Instr. check: Large air hammer (remote switch up and down) Mini air hammers (remote switch down) Mini air hammers (remote switch up) Mini air hammers (remote switch down)

-

-

Large shakes cgl Horizontal p-wave array cgl #3 Upper central array cgl #2 Lower central array cgl #2

ham8 vp_11 vs_13 vs_14

-

11:20 AM

8.7

2sc.txt

Santa Cruz 0.4g at20g, 1/2dt and dt, corrected

0.6

0.6

Large shakes cgl

dks04_30

6

11:27 AM 11:35 AM 11:41 AM 12:08 PM 12:23 PM 12:33 PM 1:09 PM 2:15 PM

8.8 9.1 9.3 9.1 9.1 9.2 9.2 to 2.4 2.4 to 6.0 6.0 to 1 1 1

2sc.txt swp7.txt swp7.txt lsc.txt lsc.txt swp6.txt -

Santa Cruz 0.4g at20g, 1/2dt and dt, corrected Sweep 160 to 400, 2 g peak Sweep 160 to 400, 2 g peak Santa Cruz 0.4g at20g, 2dt, corrected Santa Cruz 0.4g at20g, 2dt, corrected Sweep 80 to 200, 2 g peak Spin down, 9.2 to 2.4 g Spin up, 2.4 to 6.0 g Spin down, 6.0 to 1 g Instrument check Instrument check: Null shake

0.75 2 2.5 0.6 0.4 1.5 -

0.75 2 2.5 0.6 0.4 1.5 -

Large shakes cgl Large shakes cgl Large shakes cgl Large shakes cgl Large shakes cgl Large shakes cgl Slow data cgl Large shakes cgl

dks04_31 dks04_32 dks04_33 dks04_34 dks04_35 dks04_36 dks04_j.slw null1

11 5 8.5 9 4 4 -

1 1 1 1 to 9.2

-

-

-

Slow data cgl Slow data cgl Large shakes cgl Slow data cgl

dks04_k.slw null2 ham9 dks04_l.slw

-

9.2 9.2 9.1 9.1 9.1 9.1 9.1 to 18.6

swp6.txt swp4.txt swp4.txt -

1.75 1.7 2.2 -

1.75 1.7 2.2 -

Large shakes cgl Large shakes cgl Large shakes cgl Horizontal p-wave array cgl #3 Upper central array cgl #2 Lower central array cgl #2 Slow data cgl

dks04_37 dks04_38 dks04_39 vp_12 vs_15 vs_16 dks04_m.slw

6 4 8 -

2:28 PM 3:10 PM Thurs., 1/13 8:33 AM 8:59 AM 9:23 AM 9:29 AM 9:36 AM 9:44 AM 9:45 AM 9:46 AM 9:52 AM

Instrument check Instrument check Instrument check: Null shake Instr. check: Large air hammer (remote switch up and down) Spin up, 1 to 9.2 g Sweep 80 to 200, 2 g peak Sweep 50 to 125, 2 g peak Sweep 50 to 125, 2 g peak Mini air hammers (remote switch up) Mini air hammers (remote switch down) Mini air hammers (remote switch up) Spin up, 9.1 to 18.6 g

4

Comments / Details

Inspected air manifold and hoses; some water was in manifolds and in hoses to MAH4; manifold hoses were observed to be slightly bent over and nearly kinked; these were then anchored better to bucket side; also, a little water had spilled over the south end of container during complete spin-down

MAH3 and MAH4 were disconnected from air manifolds for this test; A12 (ch. 42) output only noise; NOTE: DC amp. gains changed to 100 for this test for channels involved MAH1 and MAH3 were disconnected from air manifolds for this test; A12 (ch. 42) output only noise again; NOTE: DC amp. gains changed to 100 for this test for channels involved All DC amp. ch. gains set back to 10 prior to this test; A20 (ch. 59) and A7 (ch. 33) not working; output noise only Switched A20 from ch. 59 to 55, and A2 from ch. 55 to 59 A2 (ch. 59) was okay; A20 (ch. 55) still bad; A12 (ch. 42) drifting; A30 (ch. 34) output relatively low voltage Switched A2 from ch. 59 to 42, and A12 from ch. 42 to 59 A12 (ch. 59) worked well; A2 (ch. 42) also fine; ch. 42 may be a problem channel All DC amp. ch. gains were set to 1 for large shakes; water level observed to be approx. 0.5 cm above soil surface or approx. 2.5 cm below top of container; was probably 1 cm above soil prior to spillage during last spin-down Spin up; start spinning Day 3; rpm counted manually and matched gauge value well Data acquisition screen (onboard computer) froze up; rebooted onboard computer by remote switch; both computers appeared to then work fine Data acquisition was successful All 3 mini air hammers were used for this and all subsequent mini air hammer tests MAH4 hit before MAH3 No MAH4 hit Started large shakes; P0 (PVL amp. ch. 8) and P1 (PVL amp. ch. 11) output smaller dynamic pressures than central-portion counterpart ppt's; hydrostatic pressures okay D1 appeared to not be working; displacements small compared to D0; P0 output smaller dynamic pressures; A28 (ch. 34) drifting P0 output smaller dynamic pressures; A28 (ch. 34) drifting P0 and P1 output smaller dynamic pressures; A28 (ch. 34) drifting P0 output smaller dynamic pressures P0 and P1 output smaller dynamic pressures A28 (ch. 34) drifting Problems with air-over-oil pump after recharge; spin down slightly Spin up; no record; pump still wouldn't come on Spin down completely; no record; end spinning Day 3 Technicians rebuilt air-over-oil pump Obtained current output voltage readings for D0 and D1 D1 appeared to be stuck in mid-stroke, and was probably stuck here for all prior shaking events and spinning; moved D1 up and down to check if still working; it worked fine once moved out of stuck position Re-zeroed D0 and D1 so that more than half of stroke was available to extend Obtained new output voltage readings (zeroes) for D0 and D1 Signals were small for most accelerometers since amp. gains were low Spin up; start spinning Day 4 D1 appeared to function properly now, as compared to D0 output; P0 output smaller dynamic pressures P0 output smaller dynamic pressures P0 output smaller dynamic pressures; A29 (ch. 60) drifting MAH3 hit before MAH4 No MAH4 hit Spin up


Time


Input file name

10:03 AM 10:04 AM 10:05 AM

18.6 18.6 18.6

-

10:21 AM 10:28 AM 10:52 AM 11:00 AM 11:05 AM 11:34 AM

18.6 18.7 18.3 19.0 19.2 20.3

2sc.txt lsc.txt swp4.txt swp7.txt 2sc.txt swp6.txt

11:39 AM 11:49 AM 12:20 PM 12:23 PM 12:26 PM 12:48 PM 12:52 PM 12:57 PM 1:01 PM 1:08 PM 1:09 PM 1:10 PM 1:13 PM 1:34 PM 1:35 PM 1:36 PM 1:44 PM 1:49 PM 1:55 PM 2:17 PM 2:26 PM 2:52 PM 2:58 PM 3:13 PM 3:17 PM 3:21 PM 3:26 PM 3:45 PM 3:48 PM 3:52 PM 3:58 PM 3:59 PM 4:01 PM 4:05 PM 4:14 PM 4:15 PM 4:16 PM

20.2 20.4 17.9 17.9 17.9 18.2 18.1 18.2 18.1 18.1 18.1 18.1 18.1 to 36.8 36.8 36.8 36.8 36.8 37.0 37.5 37.6 37.9 36.2 37.1 37.5 37.4 37.5 37.4 37.4 37.7 37.3 37.3 37.3 37.3 37.3 to 17.2 17.2 17.2 17.2

swp4.txt swp7.txt lsc.txt lsc.txt lsc.txt 2sc.txt 2sc.txt swp4.txt swp6.txt 2sc.txt lsc.txt swp7.txt swp6.txt lsc.txt swp4.txt swp4.txt 2sc.txt 2sc.txt swp6.txt swp7.txt lsc.txt swp4.txt swp6.txt -

4:21 PM

17.2

4:31 PM 4:36 PM

17.6 17.6 to 1

-

1

lsc.txt

Event description

Mini air hammers (remote switch down) Mini air hammers (remote switch up) Mini air hammers (remote switch down)


Output file


Comments / Details

-

-

Horizontal p-wave array cgl #3 Upper central array cgl #2 Upper central array cgl #2

vp_13 vs_17 vs_18

-

Santa Cruz 0.4g at20g, 1/2dt and dt, corrected Santa Cruz 0.4g at20g, 2dt, corrected Sweep 50 to 125, 2 g peak Sweep 160 to 400, 2 g peak Santa Cruz 0.4g at20g, 1/2dt and dt, corrected Sweep 80 to 200, 2 g peak

1.2 1.1 2 3 0.6 2

1.2 1.1 2 3 0.6 2

Large shakes cgl Large shakes cgl Large shakes cgl Large shakes cgl Large shakes cgl Large shakes cgl

dks04_40 dks04_41 dks04_42 dks04_43 dks04_44 dks04_45

23 23 8 11 4 11

Sweep 50 to 125, 2 g peak Sweep 160 to 400, 2 g peak Santa Cruz 0.4g at20g, 2dt, corrected Santa Cruz 0.4g at20g, 2dt, corrected Santa Cruz 0.4g at20g, 2dt, corrected Santa Cruz 0.4g at20g, 1/2dt and dt, corrected Santa Cruz 0.4g at20g, 1/2dt and dt, corrected Sweep 50 to 125, 2 g peak Sweep 80 to 200, 2 g peak Mini air hammers (remote switch up) Mini air hammers (remote switch down) Mini air hammers (remote switch up) Spin up, 18.1 to 36.8 g Mini air hammers (remote switch down) Mini air hammers (remote switch up) Mini air hammers (remote switch down) Santa Cruz 0.4g at20g, 1/2dt and dt, corrected Santa Cruz 0.4g at20g, 2dt, corrected Sweep 160 to 400, 2 g peak Sweep 80 to 200, 2 g peak Santa Cruz 0.4g at20g, 2dt, corrected Sweep 50 to 125, 2 g peak Sweep 50 to 125, 2 g peak Santa Cruz 0.4g at20g, 1/2dt and dt, corrected Santa Cruz 0.4g at20g, 1/2dt and dt, corrected Sweep 80 to 200, 2 g peak Sweep 160 to 400, 2 g peak Santa Cruz 0.4g at20g, 2dt, corrected Sweep 50 to 125, 2 g peak Sweep 80 to 200, 2 g peak Mini air hammers (remote switch up) Mini air hammers (remote switch up) Mini air hammers (remote switch down) Spin down, 37.3 to 17.2 g Mini air hammers (remote switch up) Mini air hammers (remote switch down) Mini air hammers (remote switch up)

2.5 1.8 0.85 0.6 0.35 0.9 0.7 1.6 1.6 1.1 1.2 1.8 2.6 2.2 2.75 2.2 0.85 0.6 1.6 2.4 0.6 1.6 2.1 -

2.5 1.8 0.85 0.6 0.35 0.9 0.7 1.6 1.6 1.1 1.2 1.8 2.6 2.2 2.75 2.2 0.85 0.6 1.6 2.4 0.6 1.6 2.1 -

Large shakes cgl Large shakes cgl Large shakes cgl Large shakes cgl Large shakes cgl Large shakes cgl Large shakes cgl Large shakes cgl Large shakes cgl Horizontal p-wave array cgl #3 Upper central array cgl #2 Lower central array cgl #2 Slow data cgl Horizontal p-wave array cgl #3 Upper central array cgl #2 Lower central array cgl #2 Large shakes cgl Large shakes cgl Large shakes cgl Large shakes cgl Large shakes cgl Large shakes cgl Large shakes cgl Large shakes cgl Large shakes cgl Large shakes cgl Large shakes cgl Large shakes cgl Large shakes cgl Large shakes cgl Horizontal p-wave array cgl #3 Upper central array cgl #2 Lower central array cgl #2 Slow data cgl Horizontal p-wave array cgl #3 Upper central array cgl #2 Lower central array cgl #2

dks04_46 dks04_47 dks04_48 dks04_49 dks04_50 dks04_51 dks04_52 dks04_53 dks04_54 vp_14 vs_19 vs_20 dks04_n.slw vp_15 vs_21 vs_22 dks04_55 dks04_56 dks04_57 dks04_58 dks04_59 dks04_60 dks04_61 dks04_62 dks04_63 dks04_64 dks04_65 dks04_66 dks04_67 dks04_68 vp_16 vs_23 vs_24 dks04_o.slw vp_17 vs_25 vs_26

16 4 15 8 3.5 20 11 6 6 26 19 6 23 46 25 11 14 5 5 11 7 4 12 -

2.2

2.2

Large shakes cgl

dks04_69

45

6 -

6 -

Large shakes cgl Slow data cgl

dks04_70 dks04_p.slw

41 -

MAH3 hit before MAH4 No MAH4 hit Large surface accelerations; A25 momentarily out of range; A25 (ch. 43) and A29 (ch. 60) drifting Very loud shaker movement observed; end of large shakes; A23 (ch. 41), A25 (ch. 43), A29 (ch. 60), and A27 (ch. 61) drifting Spin down completely; end spinning Day 4; end of testing

-

D0 and D1 observed to still contact pads near center of pads, as placed; soft area on sand surface found near middle of west side of container, directly above mini air hammer MAH4 and its hoses, and probably due to air/water leakage from this hammer or its hoses

Santa Cruz 0.4g at20g, 2dt, corrected

kobe0807.txt Kobe 0.3-15 Hz, 100g, 20V/in, corrected Spin down, 17.6 to 1 g

-


Instrument check

-

-

-

5

-

MAH3 hit before MAH4 MAH3 hit before MAH4 Pore pressures not fully dissipated during sampling; P0 output smaller dynamic pressures Sampling duration increased; pore pressures fully dissipated for most ppt's

Shaker movement was heard slightly, and saw small ripples on water surface; A13 (ch. 47) and A25 (ch. 43) drifting P0 output smaller dynamic pressures P0 output smaller dynamic pressures

P0 output smaller dynamic pressures

MAH3 hit before MAH4 No MAH4 hit Spin up No hits recorded No MAH4 hit No MAH4 hit

A13 (ch. 47), A25 (ch. 43), and A29 (ch. 60) drifting Loud shaker movement and widespread water ripples observed; A30 drifting (ch. 34) A29 drifting (ch. 60)

No MAH4 hit No MAH4 hit Spin down

Model

Lab test results

Table 2. Some properties of the soil: lab test results and model characteristics Soil No. 100 Nevada Sand, UCD Batch #3, Delivered March 1999 Supplier Gordon Sand Co., Compton, CA Classification Uniform, fine sand; SP 1 Specific gravity 2.67 Mean grain size, D50 (mm) 0.14 Coefficient of uniformity, CU 1.5 2 3 16.76 Maximum dry unit weight , γd,max (kN/m ) 2 3 13.98 Minimum dry unit weight , γd,min (kN/m ) 3 Average dry unit weight (kN/m ) 16.8 (saturated model) Average relative density (%) 101 (saturated model) Water content during sand pluviation (%) 0.1 3 Saturated unit weight (kN/m ) 20.3 1 2

Arulmoli et al. (1991) for VELACS Nevada Sand Woodward-Clyde (1997) for No. 120 Nevada Sand, UCD Batch #2

(preservative), equal to 2% of the mass of HPMC powder in the solution. The kinematic viscosity of the solution was checked prior to and after model saturation and upon completion of model testing, using a glass Ubbelohde-type capillary viscometer. Fluid samples taken from the model after saturation and after centrifuge testing were passed through a filter to remove any non-dissolved particles prior to viscosity testing. Measured viscosities for pore fluid samples taken from the model, as reported in Table 3, varied between approximately 7 and 9 times that of water. Stewart et al. (1998) reported that HPMC solutions with concentrations less than 5% have a specific gravity within 1% of that for pure water. Therefore, since the HPMC solution used for DKS04 had a concentration that was within this range, it is reasonable to assume a unit weight of 9.8 kN/m3 for the pore fluid. SCALE FACTORS All data presented in this report and included in the accompanying data files (excluding the spin-up/spin-down raw data files) are in model-scale units of g, kilopascals, millimeters, and volts for acceleration, pore pressure, displacement, and strain, respectively. No scale factor has been applied because the test events were conducted at varying g-levels. The centrifugal scaling factor, N, for each event is listed in Table 1 and plotted in Figure 2 for shaking events. These factors used in the model were calculated as 2 N = (rpm ∗ 2 ∗ π / 60) ∗ radius / g . Centrifuge rotations-per-minute (rpm) was determined by back-calculating an average rpm corresponding to measured hydrostatic pore pressures at the beginning of each shaking event. An effective radius, i.e. radial distance to onethird the model depth, of 8.56 m was used to calculate N. Table 4 lists factors that can be used to convert the data to prototype scale.

6

Table 3. Fluid viscosities of the DKS04 HPMC solution at 1 g and related information ViscosityKinematic Temperature HPMC Kinematic Ratio of check HPMC Viscosity of sample in concentration viscosity Event (cSt) viscometer by mass (%) 3 (standard) of solution (°C) pure water at viscosity to the same pure water temperature as viscosity at the viscometer this sample (cSt) temperature Before 12.81 17.0 1.1 1.084 11.8 saturation1 After 9.775 18.3 1.4 1.050 9.3 saturation2 After model 7.704 18.8 1.3 1.037 7.4 testing2 1

Sample was taken from the mixing container prior to transferring to the vacuum reservoir Samples were taken from the base of the model 3 This measurement was made by oven drying the samples; values were reported assuming that approximately all solid mass dissolved in the water was HPMC powder 2

Table 4. Scale factors required to convert the data to prototype units, and other useful scale factors: Model Dimension / Prototype Dimension Time (dynamic) 1/N Displacement, Length 1/N Acceleration, Gravity N Pressure, Stress 1 Velocity 1 Kinematic viscosity N INSTRUMENTATION AND MEASUREMENTS The centrifuge facilities and instrumentation system are briefly described in Appendix A. Instrumentation and facilities used in this test but not described in Appendix A are described below. Model construction involved air pluviation of successive Nevada Sand layers in the model container, followed by moderate leveling of the surface and approximately 8 minutes of surface vibration for every typical 5.7-cm lift. Instruments were placed on these finished lifts according to the planned model layout. The model was weighed at various stages of construction, and the unit weight of the sand was then calculated using this weight and the volume filled. Water content of the sand was also checked frequently during model construction and averaged approximately 0.1% prior to addition of pore fluid. The relative density of the in-place Nevada Sand varied as follows: 99.7% in the depth interval (measured from top of container) of 3.18 to 33.21 cm, 114.3% from 33.21 to 50.38 cm, and 88.6% from 50.38 to 58.20 cm, resulting in an average dry unit weight of 16.9 kN/m3 and an average relative density of 103% prior to model saturation. A thin cover layer of 7

medium sand (γd ~ 14 kN/m3, γsat ~ 19 kN/m3) was added to the model surface to keep sand from drying and blowing out of the model during spinning. After model saturation was complete and after removing the medium sand from the pad areas, pads for surface displacement transducers were placed firmly into the Nevada Sand. The model saturation process involved various steps to ensure removal of air both from the soil voids and from the HPMC solution. Transfer of the solution from the mixing tank to a vacuum reservoir was done with the reservoir under vacuum. During model construction, three small-diameter plastic tubes were placed at each end of the model, along with three small piles of medium sand at the base of these tubes, to assist in the saturation process. Each set of three plastic tubes was connected to a separate metal trough at the model surface. The two troughs were attached to fittings on a vacuum lid that covered the whole container top. The model was brought under vacuum, flushed with CO2 gas, and once again brought under vacuum. The HPMC solution was then slowly pulled from the vacuum reservoir into the model by holding the vacuum pressure in the model higher than that in the reservoir. After full saturation the estimated total or saturated unit weight of the model was 20.3 kN/m3, resulting in an average dry unit weight of 16.8 kN/m3 and an average relative density of 101% during model testing. Measurements of the HPMC solution viscosity were made and are described previously in Pore Fluid Properties. The origin of coordinates for instrument position was located at the inside northwest corner of the FSB2 container, at the top of the container. As shown in Figure 1, “x” was positive in the south direction, “y” in the east direction, and “z” in the down direction. By convention, direction here is determined by considering the centrifuge arm at 1 g to be a compass needle, with the bucket forming the north arrow. Positions of instruments relative to the origin were recorded before and after testing and are presented in Table 5. Distances from the origin to the instruments were measured using rulers and a depth gauge. Specifically, the position of each accelerometer, pore pressure transducer, and displacement transducer was measured at the tip of the instrument, along its centerline. Prior to testing, the pads on which the surface linear potentiometers (D0 and D1) rested were positioned so that the transducers contacted the pads at the approximate pad centers. These instrument/pad-center positions were measured and are listed as the initial positions of D0 and D1 in Table 5. Post-test coordinates for these instruments and their corresponding pads were measured and are given in Table 6 along with the pre-test positions. In addition to giving instruments’ positions, Table 5 lists each instrument’s location number, type, engineering units for reported data, location description, direction, range, serial number, calibration factor, and comments. An instrument location number is an alphanumeric identifier assigned to a specific instrument location and corresponds to an instrument label in Figure 1. In general, a given instrument location number corresponded to a specific instrument type and location in the model. Therefore, if an instrument was replaced in the model by another instrument and in the same general location, the new instrument retained the former instrument’s location number. However, when instruments

8

Table 5. Instrument list used for the DKS04 model test, January 2000 Instrument Instr. Engr. Location No. Type units

Location Description

A0

acc

g's

[Note: Primary array containing instrument is listed in most cases. FSB2 container rings are numbered 1-5 from top to bottom. Fractions of inside length (L) and width (W) below refer to approximate distances from the origin.] Load frame, northeast corner

A1

acc

g's

Load frame, northwest corner

Initial Position: (Origin at NW Final Position: (Origin at NW Instr. Instrument Calibr. Factor1 top inside corner of container), top inside corner of container), Direction Range Serial No. and Units mm mm x (south) y (east)

z (down)

x (south) y (east)

Comments

z (down)

-213

918

520

-213

918

520

H

50g

3158

106.5 mV/g

-213

-131

520

-213

-131

520

H

50g

3161

108.9 mV/g

A2

acc

g's

North outside surface, baseplate, W/2

-195

373

570

-195

373

570

V

50g

3259

109.5 mV/g Output drifted for some events

A3

acc

g's

North outside surface, middle of ring 5, W/2

-149

394

509

-149

394

509

H

50g

3955

106.6 mV/g

A4

acc

g's


-149

393

392

-149

393

392

H

50g

4435

105.8 mV/g

A5

acc

g's


-149

334

275

-149

334

275

H

50g

4534

104.8 mV/g

A6

acc

g's


-149

398

166

-149

398

166

H

50g

3156

A7

acc

g's

North outside surface, top of ring 2, W/2

-126

397

114

-126

397

114

V

50g

5269

109.3 mV/g Output just noise for all test events due to problem with 103.7 mV/g amp. ch. (33)

A8

acc

g's


-4

378

51

-4

378

51

H

50g

3959

104.4 mV/g

A9

acc

g's

In sand, upper horiz. array, middle of ring 2, W/2, L/8

211

393

164

213

395

163

H

50g

3157

109.8 mV/g

A10

acc

g's

In sand, between rings 1 and 2, W/2, L/8

206

424

134

201

429

134

V

50g

3155

Tip points approx. 15 deg. north of vertical during 108.4 mV/g excavation

A11

acc

g's

In sand, north vertical array, middle of ring 4, W/2, L/4

415

448

391

416

450

391

H

50g

3164

106.6 mV/g

A12

acc

g's


418

392

163

422

392

163

H

50g

5275

100.6 mV/g Output just noise for some test events

A13

acc

g's

A14

acc

g's

In sand, central vertical array, middle of ring 2, W/2, 3L/8

659

395

161

665

397

162

H

50g

4437

105.5 mV/g

A15

acc

g's

In sand, transverse array, middle of ring 2, 7W/8, 3L/8

659

682

160

665

683

162

H

50g

4595

104.8 mV/g

A16

acc

g's

In sand, near baseplate, W/2, L/2

820

446

567

822

445

564

V

50g

5267

103.8 mV/g

A17

acc

g's

In sand, central vertical array, middle of ring 5, W/2, L/2

827

395

501

828

397

502

H

50g

3203

108.7 mV/g

A18

acc

g's

In sand, central vertical array, between rings 4 and 5, W/2, L/2

823

394

449

825

396

451

H

50g

3202

103.0 mV/g

A19

acc

g's


834

449

392

834

449

393

H

50g

5276

A20

acc

g's


829

393

333

831

393

333

H

50g

3204

104.6 mV/g Output just noise for more than half of the test events; 107.7 mV/g instr. may be broken

A21

acc

g's


829

393

269

833

394

271

H

50g

3964

105.0 mV/g

A22

acc

g's


826

391

215

831

393

215

H

50g

4596

105.3 mV/g

A23

acc

g's


830

396

108

832

396

108

H

50g

3962

A24

acc

g's


817

458

126

814

460

129

V

50g

3949

106.5 mV/g Output drifted for some events Tip points approx. 5 deg. north of vertical during 106.2 mV/g excavation

A25

acc

g's


826

394

54

821

392

55

H

50g

3951

A26

acc

g's

In sand, between rings 1 and 2, W/2, 7L/8

1449

398

133

1449

395

136

V

50g

4436


421

391

50

419

392

53

H

50g

3166


106.5 mV/g Output drifted for some events Tip points approx. 5 deg. north of vertical during 106.4 mV/g excavation

A27

acc

g's

South outside surface, baseplate, W/2

1849

412

570

1849

412

570

V

50g

4523


A28

acc

g's

South outside surface, top of ring 2, W/2

1787

418

114

1787

418

114

V

50g

5268


A29

acc

g's

South outside surface, crucifix rack, middle of ring 1, W/2

1950

392

54

1950

392

54

H

50g

3948


A30

acc

g's

North manifold

-230

393

712

-230

393

712

V

50g

3159

A31

acc

g's

South manifold

1884

468

712

1884

392

712

V

50g

3160

P0

ppt

kPa In sand, north vertical array, on baseplate, W/2, L/4

418

412

575

418

412

574

100 psi

7722

P1

ppt

kPa In sand, north vertical array, between rings 3 and 4, W/2, L/4

411

414

329

410

416

332

50 psi

7370

109.6 mV/g Output drifted for some events 106.5 mV/g Showed drifting behavior in output most of the time Appeared to not sense dynamic pore pressures well 8238 kPa/V during some shaking events Appeared to not sense dynamic pore pressures well 3900 kPa/V during some shaking events

P2

ppt

kPa In sand, north vertical array, between rings 1 and 2, W/2, L/4

408

394

112

410

394

112

50 psi

P3

ppt

kPa In sand, central vertical array, on baseplate, W/2, L/2

911

463

576

911

463

576

100 psi

8044

8540 kPa/V

P4

ppt

kPa In sand, central vertical array, between rings 4 and 5, W/2, L/2

858

464

444

860

465

446

100 psi

7988

8222 kPa/V

P5

ppt


918

465

331

920

465

332

50 psi

6837

3850 kPa/V

P6

ppt


858

461

213

856

462

214

50 psi

7369

3605 kPa/V

P7

ppt


927

467

109

928

470

109

50 psi

6838

3820 kPa/V

D0

displ

mm Model surface, W/2, L/2

935

392

31

935

392

34

V

1"

LP101

2.62

mm/V

D1

displ

mm Model surface, W/2, 3L/4

1240

397

30

1240

397

34

V

1"

LP102

2.55

Potentiometer shaft of LP102 was stuck in mid-stroke until mm/V manually reset before last day of testing

D2

displ

mm South outside surface, middle of ring 1, W/2

1658

392

54

1658

392

54

H

4"

LP404

10.12 mm/V

displ

H

2"

D3

7367

3498 kPa/V

mm South outside surface, middle of ring 2, W/2

1803

392

167

1803

392

167

LP208

5.06

mm/V

S0

Strn G Volts North outside surface, middle of ring 3, W/2

-149

391

280

-149

391

280

Ring 1

1

V/V

S1

Strn G Volts North outside surface, middle of ring 3, W/8

-149

85

278

-149

85

278

Ring 2

1

V/V

S2

Strn G Volts West outside surface, middle of ring 3, L/6

221

-150

280

221

-150

280

Ring 3

1

V/V

S3


526

-150

280

526

-150

280

Ring 4

1

V/V

S4


829

-150

280

829

-150

280

Ring 5

1

V/V

1. Calibration factor as applied to raw data. The data files included in this report are in model-scale engineering units (g, kPa, mm, volts).

9

Table 6. Pre- and post-test positions of instruments D0 and D1 and their pads Instruments / Pre-test Post-test Pads x (mm) y (mm) z (mm) x (mm) y (mm) z (mm) D0 D0 pad center D1 D1 pad center

935 935 1240 1240

392 392 397 397

31 31 30 30

935 939 1240 1243

392 393 397 395

34 34 34 34

were added to the model, additional location numbers were assigned. The location description is a general description of an instrument’s location in the model, which in some cases includes reference to approximate distances from the origin in the form of fractions of the container’s inside length and width. The direction refers to an instrument’s primary orientation; specifically, “V” and “H” refer to vertical and horizontal orientation, respectively. The range of an instrument is simply its maximum relative output capacity. Calibration factors for instruments were used in converting instruments’ voltage outputs into engineering model-unit data, and were determined either by previous manual recalibration (pore pressure transducers and linear potentiometers) or by using directly the manufacturer’s sensitivities (accelerometers). Comments regarding performance of instruments, specifically, are included in Table 5. During the DKS04 model test the following types of data were recorded: accelerations within the soil, accelerations of the shaker load frames and manifold, accelerations of the FSB2 container rings, pore pressures at various depths, settlement of the soil surface, displacement of the rings, and strain in the third ring. As shown in Figure 1, horizontal accelerometers within the soil were arranged into longitudinal, transverse, and vertical arrays. Each of the two shaker load frames, to which the container baseplate was bolted, also included an accelerometer. On the north outside surface of the container, accelerometers were attached to the five rings. Vertical accelerometers, both those within the soil and those attached to the container and manifold, aided in capturing possible rocking behavior of the model. Pore pressure transducers positioned throughout the central and north portions of the model measured hydrostatic and dynamic pore pressures. Settlement of the soil surface was measured using linear potentiometers, which were attached to racks and extended downward to pads placed at the soil surface. In order to measure longitudinal displacement of the upper two FSB2 rings, additional linear potentiometers were attached to a rack (“crucifix”) on the south end and extended over to anchors on the rings. An accelerometer was attached to the crucifix, in line with the linear potentiometer on the top ring, thus allowing observation of rack vibration at this location. Five fully active, full-bridge strain gauges were previously installed on the outside surface of the third FSB2 ring and were utilized during the test. Data recorded from these strain gauges allowed researchers to observe primarily bending strains in the third ring as internal soil pressures changed.

10

Each instrument had a consistent, positive-output sign convention corresponding to its orientation in the model. In particular, positive displacement represented extension of the linear potentiometers, as depicted in Figure 1. All horizontal accelerometers, except for A29 on the crucifix, were oriented with their tips to the south, and the outputs of these south-pointing accelerometers all have the same sign when the container was accelerated in a given direction. A similar condition exists between the vertical accelerometers, all of which pointed downward. Pore pressure transducers were all placed pointing west, so as to be perpendicular to shaking and to centrifugal acceleration. Positive changes in output from these instruments represent actual increases in pore pressure. Strain gauges on FSB2’s third ring output decreasing voltage as the ring was forced to bend outward. In general, when the long sides of the third ring were forced outward, the short sides were forced inward. Several instrument-check hammer tests were performed throughout the model test using an air-driven, piston-cylinder hammer (known in this report as the large air hammer) attached to the underside of the container. This air hammer consists of a 2.9-cm long, Teflon-ended, steel piston encased by a 25-cm long, aluminum cylinder with an outside diameter of 3.8 cm. It was powered by compressed air passing into the cylinder from either side. By using a remote switch to trigger this hammer, and the data acquisition system to record the response of all instruments, researchers determined which of the accelerometers were performing well. Good performance of an accelerometer was defined as showing a typical waveform and amplitude in the recorded acceleration time history during the hammer tests. In-flight shear- and compression-wave velocity testing was performed on the model using three smaller or "mini-" air hammers. A mini-air hammer, as developed and tested by Arulnathan et al. (2000), consists of a 4.2-cm long, hollow aluminum cylinder with an outside diameter of 0.56 cm, with an internal, 1.9-cm long, Teflon piston. It is capped and fitted with air ports on both ends and is covered with epoxy and fine sand. These air hammers were positioned within the Central Vertical Array and near the north end (see Figure 1) and were oriented along the long axis of the container. They were triggered remotely during the test to introduce high-frequency shear and compression waves to the model, while horizontal accelerations were simultaneously recorded in the targeted arrays. For the case of shear wave tests, outputs from only the upper five (A20-A22, A23, A25) or lower five (A17-A21) accelerometers in the Central Vertical Array were recorded. Compression wave (or p wave) tests using the mini-air hammers involved five accelerometers (A14, A12, A9, A11, A19) horizontally in line with the long axis of one of two hammers (MAH4 and MAH1). Although the three hammers were triggered with the same switch, a slight delay between each mini-air hammer's firing was achieved by simply varying the lengths of tubing running between air manifolds attached to the centrifuge bucket. Air hammer tests were generally conducted before and after a series of shaking events that had similar centrifugal scaling factors. Data from these tests accompany this report. The DKS02 report (Stevens et al., 1999a) provides examples of using air hammer test data to determine the shear-wave velocity profile of a model.

11

Both before initial spin-up and after completion of the small shakes (see Table 1), saturation-check tests were performed using a manual hammer and a metal block, which was placed on the soil directly above vertical accelerometers A24 and A16. The hammer was used to strike the top of the block several times, while at the same time acceleration outputs from instruments A24 and A16 were recorded. Interpretation of the acceleration time histories from these tests indicated that the p-wave velocity of the model was around 500 m/s after vacuum saturation and before initial spin-up, and significantly higher than 1000 m/s after completion of the small shakes. Data from these tests also accompany this report. All data acquisition during the model test included filtering of instrument output using 5th –order analog, low-pass filters prior to analog-to-digital (A/D) conversion. These filters are otherwise known as “tech filters” in this report and were set with a cut-off frequency generally equal to 40% of the sampling frequency. Typical sampling frequencies and corresponding cut-off frequencies, respectively, were as follows: 2000 Hz and 800 Hz for shaking events and instrument-check hammer tests; and 50 kHz and 20 kHz for shearwave and p-wave air hammer tests. Depth measurements of the model surface relative to the origin were made for ten locations before saturation, after saturation, and after testing and are listed in Table 7. The before-saturation surface depth averaged 3.17 cm, the after-saturation surface depth averaged 2.94 cm (slight swell during saturation), and the average after-testing surface depth was 3.30 cm. These manual measurements indicate that approximately 4 mm of average total settlement at the model surface occurred between completion of vacuum saturation and completion of centrifuge testing. Figure 4 shows the surface settlements measured by the linear potentiometers during specific shaking events, as well as cumulative settlements computed by sequentially summing these discrete measurements. From this figure it appears that total settlement at the model surface, according to instrument D0, was also about 4 mm. As recorded in Table 6, the differences between initial and final manual depth measurements for D0 and D1 pad tops indicate that total settlements for these locations were 3 mm and 4 mm, respectively. A discussion regarding the discrepancy between the recorded and the manually measured total settlements of linear potentiometer pads, as well as on problems encountered with instrument D1 during testing, is included in Known Limitations of Recorded Data and Model. Penetration tests were conducted at 1 g before saturation, after saturation, and after centrifuge testing to demonstrate the uniformity of the sand. In a total of six locations on the model surface near the southwest corner, a metal rod having a 6.4-mm diameter and a 60o conical tip was inserted vertically. The force applied to the rod for a given penetration depth was recorded, as given by a force gauge held by the operator at the top of the rod. From these force measurements, penetration resistance was calculated as the total resistance (shaft plus tip), divided by the cross-sectional area of the rod. A plot of penetration resistance vs. depth is given in Figure 5 for the six penetration tests carried out.

12

Table 7. Depth readings (z, centimeters) of the model surface Location 1 2 3 4 5 6 DepthApprox. x check 1376 1376 1101 1101 826 826 (mm) Event

Æ Approx. y (mm) Æ

Before saturation After saturation After testing

262

525

262

525

262

525

7

8

9

10

550

550

275

275

262

525

262

525

3.17

2.96 3.07 3.14 3.21 3.12 3.39 3.22 3.62 2.90 3.05

2.94

2.94 2.97 3.04 3.04 2.82 3.13 2.82 3.05 2.75 2.87

3.30

3.13 3.20 3.26 3.32 3.24 3.57 3.27 3.63 3.13 3.23

Average KNOWN LIMITATIONS OF RECORDED DATA AND MODEL As presented previously, calculated relative density values of the Nevada Sand varied between 88.6% and 114.3%. The low 88.6% relative density of the bottom 7.8 cm of the model was a result of there being six small piles of medium sand in this interval at each end of the model. Since the exact volume occupied and the density of the medium sand piles were not measured, the measured unit weight for this small section of the model represents an average unit weight between the two sand types. However, to calculate relative density for the section, the dry density limits for only the Nevada Sand were used. In addition, the same pluviation and vibration methods were used on the Nevada Sand in this section as were used in the other two sections, where significantly higher unit weights were measured. Based on the unit weight for the majority of the Nevada Sand in the model, the Nevada Sand in the bottom 7.8 cm of the model most likely also had a relative density of about 100%. The high measured unit weight and 114.3% relative density calculated for the next 17.2 cm above this bottom section was a result of slightly longerduration vibration. Despite the differences in measured relative density values for the three model sections, the stated average dry unit weight of 16.8 kN/m3 and the average relative density of 101% for the Nevada Sand, appear to be reasonable values for characterizing the saturated model. Inaccuracies in the manually recorded tachometer readings of gauge rpm were discovered after test completion, by comparing back-calculated rpm (via pore pressure data) for shaking events and the gauge rpm observed during testing. As a result of the rpm problem, back-calculated centrifugal accelerations in the model varied from target values of 10, 20, and 40 g (all previously used for DKS02 and DKS03). Therefore, careful reference to Table 1 or Figure 2 should be used in determining which centrifugal scaling factor to use in converting model-unit data to prototype units. For non-shaking events the centrifugal scaling factors listed in Table 1 correspond to that of a shaking event that occurred either just previous to or just after the given set of non-shaking events.

13

Some water was added in flight during shaking events 1 through 24 (see Table 1). Initially located at approximately the same level as the average soil surface, the water table increased by about 10 mm over the course of these shaking events. This change is reflected in the initial pore pressures in the recorded time histories and was taken into account in the back-calculation of rpm and centrifugal scaling factors. The order in which the three mini-air hammers fired after activation varied throughout the testing sequence. Based on recorded time histories of targeted arrays, MAH1 appeared to hit before MAH3 during p-wave mini-air hammer tests. Recorded time histories from shear-wave mini-air hammer tests showed that MAH4 hit infrequently: sometimes before MAH3, sometimes after it, and sometimes not at all. This sporadic behavior was perhaps due to the fact that the small air hoses leading to MAH4 were observed to be partially filled with water. A leak in the hammer or in the hoses may have caused water to enter these hoses and obstruct the airflow. The observed order between MAH3 and MAH4 during shear-wave mini-air hammer tests is included in the comments/details of Table 1 for each of these events. Two p-wave mini-air hammer tests involved special amplifier gain settings and mini-air hammer arrangements on the air manifolds. As noted in Table 1, only MAH1 was utilized for event vp_9 since MAH3 and MAH4 had been disconnected temporarily. Similarly, event vp_10 only involved MAH4 while the other two mini-air hammers were not active. For both of these hammer tests the amplifier gains for accelerometers involved were changed temporarily to 100, whereas all other test events involved accelerometer amplifier gains of either 10 or 1. Measured p-wave velocity of the model was lower than expected prior to initial spin-up (500 m/s) but was shown to have increased significantly upon completion of the small shakes (1000 m/s, typical of saturated soil). The model, therefore, was apparently fully saturated once initial spinning had occurred and most likely remained as such for the duration of the model test. During observation and excavation of the model after test completion, it was discovered that a small section of the model had become relatively soft during testing. This section was located on the west side of the model (see Figure 1), near the middle of the model surface, and extended down from the surface to about the level of the MAH4 air hoses, and horizontally from about the half-width line of the model over to the far west side of the model. As pointed out previously, the air hoses leading to MAH4 were observed to have water in them during testing. If water entered the hoses via a leak in either the hammer or the hoses during testing, then any escaping air from this leaky system could have been the cause of the loosened soil above it. It is not known how long during testing this area was soft relative to the rest of the model, but the response of instruments in the area (A25, A23, A24) may have been affected by the softened soil. Instrument D1 only output its initial-zero voltage for all test events from initial spin-up through shaking event 36 and instrument-check event dks04_k.slw (see Table 1). It was

14

discovered after this sequence of events that the shaft of the linear potentiometer at D1 was physically stuck in the zero-voltage position, and probably had been since initial spinup. The shaft was manually moved up and down, and the instrument was re-zeroed, after which D1 appeared to correctly output varying voltage according to the type of event being conducted, and in good agreement with the output recorded from D0 also on the model surface (see Figures 1 and 4). Shaking-event surface settlement data displayed in Figure 4 reflects the D1 zero-output problem present through shaking event 36, as well as the apparently good agreement between settlement recordings for D0 and D1 thereafter. There is a slight discrepancy between the recorded total settlements (4 mm for D0; see Figure 4) and the manually measured total settlements (3 mm for D0; 4 mm for D1; see Table 6) of linear potentiometer pads at D0 and D1. The actual cause of the 1-mm difference between the measurements at D0 is unknown but may relate to operator error in manual depth measurements using the depth gauge. Observation of the model after testing showed that the pads below D0 and D1 had moved laterally only a few millimeters, as shown by positions given in Table 6, and the instruments remained in contact with the tops of the pads. Lateral movement of the pads may have caused them to tilt slightly, thus adding to possible error in settlement measurements by the linear potentiometers. Two pore pressure transducers in the north portion of the model, P0 and P1, appeared to not respond well to dynamic pore pressures for some of the large shaking events. This was determined by comparison with the level of dynamic pressures recorded by the corresponding central-portion instruments (P3 and P5, respectively; see Figure 1). However, the hydrostatic pore pressures recorded by P0 and P1 generally did match the levels of those recorded by their central-portion counterparts. As noted in Table 1, the following shaking events resulted in one or both of these transducers recording apparently smaller dynamic pore pressures: 30-35, 37-40, 48, 49, and 53. Output from instrument A31, located on the south manifold (see Figure 1), consistently exhibited low-frequency drifting behavior during the test. Some accelerometers' outputs exhibited drifting behavior during selected shaking events, as indicated in Table 1. These include the following: A2 during event 23; A13 during events 46 and 58; A25 during events 46, 58, 69, and 70; A27 during event 70; A28 during events 31 through 33, and 36; A29 during events 39, 60, 69, and 70; and A30 during event 59. Instrument A7 consistently output only low-amplitude noise during testing. The following additional instruments sometimes output only noise and are listed here together with the events affected (see Table 1): A12 during shaking events 13 through 22, and during miniair hammer tests vp_9 and 10; and A20 during shaking events 23 through 70, and during mini-air hammer tests vs_9 through 26.

15

As noted in Table 5, three of the four vertical accelerometers within the soil were observed during model excavation to have rotated during testing between about 5 and 15 degrees away from vertical. The rotated instruments included A10, A24, and A26. Calibration information for the ring 3 strain gauges on FSB2 is contained in the DKS01 data report. A simple factor of 1 volt/volt is listed in Table 5 for these instruments, in place of any specific calibration factors. DESCRIPTION OF DISPLACEMENT-COMMAND FILES SHAKING

FOR

MODEL

As noted in Table 1, six different input motion files were used as commands to the shaker during in-flight model testing and are described in this section. Input files containing the Santa Cruz ground motion were 2sc.txt and lsc.txt. The 2sc.txt input file included two complete time histories: first, the motion with a time step equal to half the original time step (1/2dt portion); and second, the motion with the original time step (dt portion; original time step and amplitude were scaled for the 20-g case). Similarly, lsc.txt had the same motion, but with a time step equal to twice the original time step. The scaled Kobe ground motion was contained in the input motion file kobe0807.txt and was only used once, for shaking event 70. Frequency-sweep input files included swp4.txt, swp6.txt, and swp7.txt. Each of these included 15 packets of sinusoidal waves in one continuous time history, with each packet being made up of 15 cycles at a particular frequency and amplitude. The first frequency packets for these files were at frequencies of 50, 80, and 160 Hz, respectively. Final frequency packets, also respectively, were at 125, 200, and 400 Hz. Specifically, the following function was used to calculate discrete frequencies of packets subsequent to the first packet: f = pf × ( nf −1) ff . In this function, f is the frequency of a given packet after the first packet, pf is the frequency of the preceding packet, nf is the total number of frequency packets in the file (15), and ff is the frequency factor, i.e. the ratio of the final packet’s frequency to that of the first packet (2.5). Finally, all of the input motions were pre-conditioned based on the expected shaker response. ORGANIZATION OF DATA FILES AND PLOTS In addition to listing the date, time, centrifugal scaling factor, and input file name for each test event, Table 1 gives the event description, amplification factors, channel gain list (cgl), output file, rough model peak-to-peak base acceleration, and comments corresponding to each event. In the case of shaking events, the event description is a shortened version of the full input file description. Other event descriptions include reference to instrument checks, hammer tests (described previously), null shakes, and spinning up or down to a different rpm and corresponding centrifugal scaling factor. An instrument check consisted of zeroing instruments, changing amplifier gains, observing the model at 1 g, and/or otherwise adjusting components of the centrifuge testing system as noted. Null shakes involved acquiring high-frequency data without actually shaking the model and were used to determine the voltages currently output by selected instruments.

16

West and east amplification factors, corresponding to the respective load frames, were applied to each input file commanded to the shaker. These factors determined the general magnitude of shaking for an event. Table 1 lists the amplification factors along with an approximate, model-scale, peak-to-peak horizontal acceleration of the container baseplate (via the shaker load frames) for most of the events. Various cgl’s were used during the model test. These are referenced in Table 1 and are shown in Tables 8 through 18. A cgl was chosen prior to acquiring data. It specified which instruments’ outputs, and in what order, would be recorded for a particular event. Some instrument information is repeated from Table 5 in the cgl’s, along with the order in which instruments were sampled and the overall amplifier gain. The order in which instruments were sampled is also referred to as “DT Entry No.” and corresponds to the column number (beginning with zero) to which data for an instrument was written in the original output file for an event. “Slow” data, acquired at low sampling frequency, was generally recorded during spin-up or spin-down to a different rpm and corresponding centrifugal scaling factor. All slow data sampling involved the slow data cgl; thus, data from only the pore pressure transducers, linear potentiometers, and strain gauges were recorded. Original output files had either the *.slw or *.out extension, corresponding, respectively, to slow and high-frequency data acquisition. High-frequency data sampling occurred during shaking events, air hammer tests, manual hammer tests, and null shakes. Output files listed in Table 1 without an extension were originally *.out files. They were converted from voltage data to engineering model-unit data and were saved as *.txt files. As part of the file-conversion process, all data were zeroed. Acceleration data from a given event and instrument were zeroed with the average of the initial 50 data points from that instrument for that event. Data from pore pressure transducers, linear potentiometers, and strain gauges were zeroed during file conversion, using the average outputs in output file inchk1 (for events prior to initial spin-up) and the average initial outputs in output file dks04_a.slw (for events after initial spin-up). Data from these two files were used in this zeroing process due to the fact that between the two respective events, the amplifier channels were reset to allow for the expected range of pore pressure transducer output. The hydrostatic pressures at 1 g were also added to the respective pore pressure transducers' relative outputs as part of file conversion. Pore pressures represented by data in the model-unit files are in absolute pressure. For events after D1 was unstuck (after dks04_k.slw; see Table 1), data from D0 and D1 were zeroed using the average signals from output files null1 and null2, along with the previous "zeroes" in order to keep the converted data in absolute displacement. Strain data were also corrected for the effects of rpm and of vertical load in the container rings (made possible with results from the DKS01 test). No filtering of data was performed during file conversion. The model-unit files accompanying this report contain data in the same column order as in the *.out files, but include a time vector in model-scale seconds as the first column.

17

Table 8. Slow data channel gain list used for the DKS04 model test, January 2000 Column No. in Original Instrument Output Files Location No. (DT Entry No.)


Overall amp. gain (not incl. DT gain)

Calibr. Factor and Units

[Note: Primary array containing instrument is listed in most cases. FSB2 container rings are numbered 1-5 from top to bottom. Fractions of inside length (L) and width (W) below refer to approximate distances from the origin.] 0

P0

In sand, north vertical array, on baseplate, W/2, L/4

100

8238

kPa/V

1

P1

In sand, north vertical array, between rings 3 and 4, W/2, L/4

100

3900

kPa/V

2

P2


100

3498

kPa/V

3

P3

In sand, central vertical array, on baseplate, W/2, L/2

100

8540

kPa/V

4

P4


100

8222

kPa/V

5

P5


100

3850

kPa/V

6

P6


100

3605

kPa/V

7

P7


100

3820

kPa/V

8

D0

Model surface, W/2, L/2

2

2.62

mm/V

9

D1

Model surface, W/2, 3L/4

2

2.55

mm/V

10

D3

South outside surface, middle of ring 2, W/2

2

5.06

mm/V

11

D2


2

10.12

mm/V

12

S0


1000

1

V/V

13

S1


1000

1

V/V

14

S2

West outside surface, middle of ring 3, L/6

1000

1

V/V

15

S3


1000

1

V/V

16

S4


1000

1

V/V

18

Table 9. Small shakes channel gain list used for the DKS04 model test, January 2000 Column No. in Original Instrument Output Files Location (DT Entry No. No.)



Table 10. Large shakes channel gain list used for the DKS04 model test, January 2000 Column No. in Original Instrument Output Files Location (DT Entry No. No.)


[Note: Primary array containing instrument is listed in most cases. FSB2 container rings are numbered 1-5 from top to bottom. Fractions of inside length (L) and width (W) below refer to approximate distances from the origin.]




[Note: Primary array containing instrument is listed in most cases. FSB2 container rings are numbered 1-5 from top to bottom. Fractions of inside length (L) and width (W) below refer to approximate distances from the origin.]

0

A0

Load frame, northeast corner

10

106.5

mV/g

0

A0

Load frame, northeast corner

1

106.5

mV/g

1

A1


10

108.9

mV/g

1

A1


1

108.9

mV/g

2

A3


10

106.6

mV/g

2

A3


1

106.6

mV/g

3

A4


10

105.8

mV/g

3

A4


1

105.8

mV/g

4

A5


10

104.8

mV/g

4

A5


1

104.8

mV/g

5

A6


10

109.3

mV/g

5

A6


1

109.3

mV/g

6

A8


10

104.4

mV/g

6

A8


1

104.4

mV/g

7

A17


10

108.7

mV/g

7

A17


1

108.7

mV/g

8

A18


10

103.0

mV/g

8

A18


1

103.0

mV/g

9

A19


10

104.6

mV/g

9

A19


1

104.6

mV/g

10

A20


10

107.7

mV/g

10

A20


1

107.7

mV/g

11

A21


10

105.0

mV/g

11

A21


1

105.0

mV/g

12

A22


10

105.3

mV/g

12

A22


1

105.3

mV/g

13

A14


10

105.5

mV/g

13

A14


1

105.5

mV/g

14

A23


10

106.5

mV/g

14

A23


1

106.5

mV/g

15

A25


10

106.5

mV/g

15

A25


1

106.5

mV/g

16

A11


10

106.6

mV/g

16

A11


1

106.6

mV/g

17

A12


10

100.6

mV/g

17

A12


1

100.6

mV/g

18

A13


10

106.5

mV/g

18

A13


1

106.5

mV/g

19

A9


10

109.8

mV/g

19

A9


1

109.8

mV/g mV/g

20

A15


10

104.8

mV/g

20

A15


1

104.8

21

A29


10

105.8

mV/g

21

A29


1

105.8

mV/g

22

A2


10

109.5

mV/g

22

A2


1

109.5

mV/g

23

A16


10

103.8

mV/g

23

A16


1

103.8

mV/g

24

A27


10

104.9

mV/g

24

A27


1

104.9

mV/g

25

A7


10

103.7

mV/g

25

A7


1

103.7

mV/g

26

A10


10

108.4

mV/g

26

A10


1

108.4

mV/g mV/g

27

A24


10

106.2

mV/g

27

A24


1

106.2

28

A26


10

106.4

mV/g

28

A26


1

106.4

mV/g

29

A28


10

106.4

mV/g

29

A28


1

106.4

mV/g mV/g

30

A30

North manifold

10

109.6

mV/g

30

A30

North manifold

1

109.6

31

A31

South manifold

10

106.5

mV/g

31

A31

South manifold

1

106.5

mV/g

32

P0


100

8238

kPa/V

32

P0


100

8238

kPa/V

33

P1


100

3900

kPa/V

33

P1


100

3900

kPa/V

34

P2


100

3498

kPa/V

34

P2


100

3498

kPa/V

35

P3


100

8540

kPa/V

35

P3


100

8540

kPa/V

36

P4


100

8222

kPa/V

36

P4


100

8222

kPa/V

37

P5


100

3850

kPa/V

37

P5


100

3850

kPa/V

38

P6


100

3605

kPa/V

38

P6


100

3605

kPa/V

100

3820

kPa/V

39

P7


100

3820

kPa/V

2

2.62

mm/V

40

D0


2

2.62

mm/V

39

P7


40

D0


41

D1


2

2.55

mm/V

41

D1


2

2.55

mm/V

42

D3


2

5.06

mm/V

42

D3


2

5.06

mm/V

43

D2


2

10.12

mm/V

43

D2


2

10.12

mm/V

44

S0


1000

1

V/V

44

S0


1000

1

V/V

45

S1


1000

1

V/V

45

S1


1000

1

V/V

46

S2


1000

1

V/V

46

S2


1000

1

V/V

47

S3


1000

1

V/V

47

S3


1000

1

V/V

48

S4


1000

1

V/V

48

S4


1000

1

V/V

19

Table 11. Upper central vertical array, channel gain list #1 used for DKS04 air hammer tests, January 2000 Column No. in Original Instrument Output Files Location (DT Entry No. No.)





A25


10

106.5

mV/g

1

A23


10

106.5

mV/g

2

A22


10

105.3

mV/g

3

A21


10

105.0

mV/g

4

A20


10

107.7

mV/g

Table 12. Upper central vertical array, channel gain list #2 used for DKS04 air hammer tests, January 2000 Column No. in Original Instrument Output Files Location (DT Entry No. No.)





A25


1

106.5

mV/g

1

A23


1

106.5

mV/g

2

A22


1

105.3

mV/g

3

A21


1

105.0

mV/g

4

A20


1

107.7

mV/g

Table 13. Lower central vertical array, channel gain list #1 used for DKS04 air hammer tests, January 2000 Column No. in Original Instrument Output Files Location No. (DT Entry No.)





A21


10

105.0

mV/g

1

A20


10

107.7

mV/g

2

A19


10

104.6

mV/g

3

A18


10

103.0

mV/g

4

A17


10

108.7

mV/g

Table 14. Lower central vertical array, channel gain list #2 used for DKS04 air hammer tests, January 2000 Column No. in Original Instrument Output Files Location No. (DT Entry No.)





A21


1

105.0

mV/g

1

A20


1

107.7

mV/g

2

A19


1

104.6

mV/g

3

A18


1

103.0

mV/g

4

A17


1

108.7

mV/g

20

Table 15. Vertical p-wave array, channel gain list used for the DKS04 model test, January 2000 Column No. in Original Instrument Output Files Location (DT Entry No. No.)





A24


10

106.2

mV/g

1

A16


10

103.8

mV/g

Table 16. Horizontal p-wave array, channel gain list #1 used for DKS04 air hammer tests, January 2000 Column No. in Original Instrument Output Files Location (DT Entry No. No.)





A14


10

105.5

mV/g

1

A12


10

100.6

mV/g

2

A9


10

109.8

mV/g

3

A11


10

106.6

mV/g

4

A19


10

104.6

mV/g






A14


100

105.5

mV/g

1 2

A12


100

100.6

mV/g

A9


100

109.8

mV/g

3

A11


100

106.6

mV/g

4

A19


100

104.6

mV/g






A14


1

105.5

mV/g

1

A12


1

100.6

mV/g

2

A9


1

109.8

mV/g

3

A11


1

106.6

mV/g

4

A19


1

104.6

mV/g

21

Sampling frequency and duration of an event were considered in producing the time vector. The model-unit files contain no header lines. The *.slw output files, which also accompany this report, were not converted to modelunit files; hence, their contents remain voltage data. These files include seven header lines of event and cgl information that can be ignored. Converting these data to engineering model units would require consideration of initial instrument voltage output (“zeroes”), overall amplifier gains, and applicable calibration factors. A continuous time reading is at the end of each line of data in these files. Table 19. Order and description of Plot Section 1 events Event ID / Description Output file name dks04_39 Frequency Sweep 50 to 125 Hz, N = 9.1 dks04_53 Frequency Sweep 50 to 125 Hz, N = 18.2 dks04_67 Frequency Sweep 50 to 125 Hz, N = 37.7 dks04_1 Small Santa Cruz, original time step (dt) portion only (as explained in Description of Displacement-Command Files for Model Shaking), N = 9.3 dks04_13 Small Santa Cruz, original time step (dt) portion only, N = 24.7 dks04_23 Small Santa Cruz, original time step (dt) portion only, N = 47.2 dks04_30 Large Santa Cruz, original time step (dt) portion only, N = 8.7 dks04_52 Large Santa Cruz, original time step (dt) portion only, N = 18.1 dks04_55 Large Santa Cruz, original time step (dt) portion only, N = 36.8 Plots of data from various events are presented in the order outlined in the Table of Contents. Data from instruments A7, A20, and A31 are not presented in the report plots since they output only noise or had consistently drifting output for most of the test events. Plot Section 1 contains time histories for all instruments for the nine selected shaking events listed in Table 19. The first three events for which data are presented have similar model-scale peak base accelerations (~ 2 to 3 g). Reference to “small” and “large” Santa Cruz input motions for the six subsequent events refers to the relative magnitude of the input amplification factors and the recorded accelerations for particular events. The three “small” Santa Cruz events for which data are presented have similar prototype-scale peak base accelerations (~ 0.02 to 0.03 g); the same applies between the “large” Santa Cruz events (~ 0.3 to 0.4 g). For each of these nine events, dynamic (shaking) data are reproduced on ten pages of time histories in the order given in Table 20. Plots for shaking events in which significant excess pore pressures developed (39, 30, and 55) include an extra page (8b) that shows both the dynamic and the pressure-dissipation portions of the time histories for pore pressure transducers.

22

Table 20. Instrument arrays and pages for events presented in Plot Sections 1 through 3 Instrument Location Numbers 1 Description Page A16, A0, A1, AB A8, A6, A5, A4, A3, AB A25, A23, A14, A22, A21, A19, A18, A17, AB A19, A4, A13, A12, A11, AB A15, A14, A12, A9, A6, AB A29, A25, A13, A8, AB A26, A24, A10, A28, A27, A16, A2, A30, AB P7, P6, P5, P4, P3, P2, P1, P0, AB P7, P6, P5, P4, P3, P2, P1, P0, AB D2, D3, D0, D1, AB S1, S0, S2, S3, S4, AB

Input Base Motion, Horizontal and Vertical (Central) Accelerations Horizontal Accelerations of FSB2 Rings Horizontal Accelerations, Central Vertical Array

1 2 3

Horizontal Accelerations, North Vertical Array and Ring 4 Elevation Horizontal Accelerations, Ring 2 Elevation and Transverse Array Horizontal Accelerations, Ring 1 Elevation Vertical Accelerations

6 7

Pore Pressures (shaking portion only)

8a

Pore Pressures (including pressure dissipation portion) Surface and Rings' Displacements Ring 3 Strains

4 5

8b 2 9 10

1

Instrument location number AB corresponds to the average horizontal base motion (the average acceleration of instruments A0 and A1) 2 Page 8b is only included in Plot Section 1 for plots of shaking events 39, 30, and 55

At the base of each page in Plot Section 1, the output file name ("file") corresponding to an event is listed together with the page number ("page") and page description ("page_desc"). The top of each page contains the following three labels: “each_tick = … units” refers to the model-unit increment represented by the distance between tick marks on the ordinate axis, where the units are g, kPa, mm, or volts; “inst_from_top” designates a list of instruments for which data is plotted on that page, where left-to-right listing corresponds to top-to-bottom order of instruments in the plot; “pk_to_pk” gives the peakto-peak response, in model units, of each instrument for which data is plotted on that page, in the same order as instruments are listed for “inst_from_top”. Each instrument’s signal was zeroed, for plotting purposes, at the beginning of the time history for an event, and the signal oscillates about an offset “zero” line. Adjacent instrument signals on the plots have been offset so as not to overlap, although a given ordinate-axis scale is constant for all instruments represented on that page. An exception to this rule applies to pages 8a, 8b, 9, and 10, in which the ordinate-axis scale, the "each_tick" value, and the "pk_to_pk" values are not valid for the average base motion. The time axis in the plots is in modelscale seconds and in some cases has been truncated to show only a portion of the time histories for an event.

23

Plot Section 2 is very similar in layout to Plot Section 1 but contains time-history plots in model scale for all shaking events for only the Central Vertical Array and the average base motion. It combines “page 3” from Table 20 and all shaking events listed in Table 1. Therefore, all pages in this section are designated as “page 3” but proceed according to chronological order of events. Acceleration response spectra for time histories from shaking event 52 (dt portion only, as in Plot Section 1, and as explained in Description of Displacement-Command Files for Model Shaking) are presented in Plot Section 3. Spectral acceleration, computed with a damping ratio of 5%, and period are presented in model scale. REFERENCES Arulmoli, K., Muraleetharan, K.K., Hossain, M.M., and Fruth, L.S. (1991). “VELACS Laboratory Testing Program,” Preliminary Data Report to the National Science Foundation, The Earth Technology Corporation, Irvine, CA. Arulnathan, R., Boulanger, R. W., Kutter, B. L., and Sluis, W. K. (2000). "New Tool for Shear Wave Velocity Measurements in Model Tests," Geotechnical Testing Journal, GTJODJ, ASTM, Vol. 23, No. 4, pp. 444-452. Stevens, D.K., Kim, B.-I., Wilson, D.W., and Kutter, B.L. (1999a). "Comprehensive Investigation of Nonlinear Site Response - Centrifuge Data Report for DKS02." Center for Geotechnical Modeling Data Report UCD/CGMDR-99/02, University of California, Davis, CA. Stevens, D.K., Wilson, D.W., and Kutter, B.L. (1999b). "Comprehensive Investigation of Nonlinear Site Response - Centrifuge Data Report for DKS03." Center for Geotechnical Modeling Data Report UCD/CGMDR-99/03, University of California, Davis, CA. Stewart, D.P., Chen, Y.-R., and Kutter, B.L. (1998). "Experience with the Use of Methylcellulose as a Viscous Pore Fluid in Centrifuge Models," Technical Note, Geotechnical Testing Journal, GTJODJ, Vol. 21, No. 4, December 1998, pp. 365369. Woodward-Clyde (1997). “Experimental Results of Maximum and Minimum Dry Densities of Nevada Sand.” Memorandum (Feb. 13, 1997) from Woodward-Clyde to the Center for Geotechnical Modeling, University of California, Davis, CA.

24

1651

A0

A15 Central vertical array

A2 S0

A10 A8

P0-2

A9

A14

MAH3,4

P3,5,7

A28

A31

787

A3-6 A30

A16,24 A19

A11

MAH1

A7

A27

P4,6

A17,18,(19),20-22,(14),23,25

D1

D0

A26

D2

A29

D3

North vertical array

A(11),12,13 S1

y x

A1

S2

S3

S4

Top View 1651 4

Medium Sand cover layer γsat ~ 19 kN/m^3

x

A8 A7 z

A6

A13 P2

A10 A9

Transverse array

A14,15

A12 S0,1

549

A5 A4

A22 A21

P1

A20 A19 A18 A17 A16

MAH1

A3 P0

D1

D0 A24

P7

Water table approx. at the soil surface initially

D2 A26

MAH4

A11

Load frames

A0,1

A25 A23

NEVADA SAND: Dr = 101% γ sat = 20.3 kN/m^3

P5 MAH3

HPMC SOLUTION: Kinematic Viscosity = 7 to 9 X visc. of water γ = 9.8 kN/m^3

P4 P3

A27 Large Air Hammer

A30

N

A28 D3

P6

A2

Manifold

A29

West Side Section View

Crucifix rack

A31 Manifold

Linear potentiometer

Notes: 1. Dimensions shown are in millimeters. 2. An instrument location number is listed next to each instrument, and the instrument description is listed in Table 5. 3. The NORTH end of the model container is designated by the arrow and "N" above. 4. The origin of coordinates is located at the inside northwest corner of FSB2, at the top of the container. 5. Referring to the instrument position measurements in Table 5: x is positive in the south direction, y is positive in the east direction, and z is positive in the "down" direction (toward the base of FSB2). 6. Initial and final position coordinates of instruments are given in Table 5. 7. The positions of accelerometers, pore pressure transducers, and linear potentiometers, as listed in Table 5, were measured at the tips of the instruments.

Horizontal accelerometer Vertical accelerometer Pore pressure transducer Strain gauge Air hammer (large & mini [MAH]) ORIGIN

Figure 1. Initial DKS04 model configuration in the FSB2 container. January 2000 (scale = 1:15) 25

50

45

40

Centrifugal scaling factor *

35

30

25

20

15

10

5

0 0

5

10

15

20

25

30

35

40

45

50

Shaking-event sequence number * Centrifugal scaling factors were calculated using the following parameters: (1) an effective radius, i.e. radial distance to one-third the model depth, of 8.56 m; and (2) the average rpm back-calculated from initial pore pressure transducers' data for each event.

Figure 2. Centrifugal scaling factors for DKS04 shaking events

26

55

60

65

70

100 No. 100, UCD Batch #3, Stevens et al. (1999b) No. 120, UCD Batch #2, Stevens et al. (1999a)

Percent finer

80

60

40

20

0 100

10

1

0.1

0.01

0.001

Grain size, mm

No. 100, UCD No. 120, UCD Batch #3, Stevens Batch #2, Stevens et al. (1999b) et al. (1999a)

Sieve no.

Grain size (mm)

Percent finer

Percent finer

30 40 60 100 140 200

0.6 0.42 0.25 0.15 0.105 0.075

100.0 99.3 94.4 62.4 15.3 2.3

100.0 99.5 95.9 51.7 12.0 2.1

Figure 3. Grain-size distribution of selected Nevada Sand batches at the Center for Geotechnical Modeling, UC Davis, 1999 (Stevens et al., 1999b)

27

Shaking-event sequence number 0

10

20

30

40

50

60

70

-0.5 0.0 0.5 37

Settlement, mm

1.0

58

30

1.5 2.0

D0, discrete

2.5

D1, discrete 3.0

D0, cumulative D1, cumulative

3.5 4.0 4.5 5.0

Figure 4. Surface settlement during DKS04 shaking events

Penetration resistance, MPa 0

2

4

6

8

10

12

14

16

18

0 5

Depth z below origin, cm

10

Location 1 2 3 4 5 6

Description Before saturation Before saturation After saturation After saturation After testing After testing

x (mm) 1171 1411 1171 1411 1291 1531

y (mm) 294 294 98 98 196 196

15 Location 1 Location 2

20

Location 3 Location 4

25

Location 5 Location 6

30 35 40

Figure 5. Results of penetration tests at 1 g for the DKS04 model

28

20

PLOT SECTION 1 TIME HISTORIES FOR ALL INSTRUMENTS: NINE SELECTED SHAKING EVENTS

PLOT SECTION 2 TIME HISTORIES FOR THE CENTRAL VERTICAL ARRAY AND AVERAGE BASE MOTION: ALL SHAKING EVENTS

PLOT SECTION 3 ACCELERATION RESPONSE SPECTRA FOR ALL ACCELEROMETERS: ONE SELECTED SHAKING EVENT

APPENDIX A: CENTER FOR GEOTECHNICAL MODELING

December 1997

CENTER DIRECTORS: James A. Cheney I. M. Idriss Bruce L. Kutter

1983-1989 1989-1996 1996-

FACILITY MANAGEMENT: Bruce L. Kutter D. P. Stewart D.W. Wilson

1983-1990 Managing Director 1990-1996 Associate Director 1995-1997 Facility Manager 1997Facility Manager

DESCRIPTION OF CENTRIFUGE FACILITIES 9 m CENTRIFUGE AND SHAKER In 1978, in response to a broad agency announcement from NSF, UC Davis collaborated with NASA to propose the development of a large geotechnical centrifuge facility. The proposal was successful, and the centrifuge was constructed and installed at the NASA Ames Research Center. The centrifuge was first operated at NASA Ames in early 1984. In 1987, NASA disassembled the centrifuge and hauled it to UC Davis. Using funds from Tyndall AFB, Los Alamos National Laboratories, the National Science Foundation, and the University of California, the centrifuge was installed and operated on a concrete slab in the open air in 1988; in this configuration, the maximum acceleration that could be achieved was 19 g. NSF then funded the construction of an enclosure around the centrifuge which was completed in December 1989; this allowed the centrifuge to achieve accelerations of 50 g. From 1990-1995, a large amount of effort was expended to raise funds, design, and develop an earthquake simulation capability for the large centrifuge. The shaker development was supported by NSF, Caltrans, the Obayashi Corporation, and the University of California. In April 1995, the first proof tests of the shaker were conducted at a centrifugal acceleration of 50 g. In 1996, the focus of the Center finally shifted from facility development to utilization of the shaker and centrifuge facilities. This centrifuge, in terms of radius (9.1 m to bucket floor), maximum payload mass (4500 kg), and available bucket area (4.0 m2) is the largest geotechnical centrifuge in the world. At present, the centrifuge is limited to centrifugal accelerations up to 53 g. The centrifuge capacity in terms of the maximum acceleration multiplied by the maximum payload is 53 g x 4500 kg = 240 g - tonnes. The maximum speed of the centrifuge is limited by available drive torque, thus there is a potential to increase the maximum acceleration by streamlining the enclosure or the arm. There is, however, no shortage of research work that can be conducted at accelerations up to 50 g.

A-1

The data acquisition system on the centrifuge is kept up to date by continual upgrades to the data acquisition systems. Low level signals are conditioned with instrumentation amplifiers at the end of the centrifuge arm. Commercially available programmable gain, programmable cutoff frequency anti-aliasing amplifiers mounted at the center of the centrifuge are used prior to digitization. At present the system has a capability to digitize 64 channels of data at about 5,000 samples per second per channel. The digitization takes place onboard the centrifuge by a commercial data acquisition board (Data Translations DT2839) installed in a 486 IBM compatible computer. The data is stored on a conventional hard disk on the centrifuge computer and then transmitted to another computer in the control room via an ethernet network and remote control software. Some specifications of large servo-hydraulic shaker are summarized here; a more detailed description is given by Kutter et al. (1994). For a rigid shaking mass of 2700 kg (model plus container), the actuators, operating with 3500 psi (24 MPa) oil pressure theoretically have the capacity to produce 14 g shaking accelerations at frequencies up to 200 Hz. In practice, much larger accelerations (up to 30 g) can be obtained because the models do not behave like a rigid mass at high frequencies. The maximum absolute shaking velocity is about 1 m/s and the stroke is 2.5 cm peak to peak. To achieve this, two pairs of single acting actuators provide a net peak actuator force of 400 kN. One actuator pair is mounted on each side of the model container. Each actuator pair (patented by Team Corporation) consists of two - single acting actuators bolted on opposite sides of a two stage servo-valve block. The actuators are controlled by a conventional closed loop feedback control system. A simple correction scheme is used to precondition the command signal to improve the coherence and frequency content of the shaking table motions. In addition to step waves, sine waves and sine sweeps, simulations of ground motions recorded in the Loma Prieta and Kobe earthquakes have been produced in the shaker. For the new shaker, two flexible shear beam containers (FSB1, and FSB2) and a rigid container with windows (RC1) are available. FSB2 and RC1 have dimensions of approximately 1.7 x 0.8 x 0.5 m (Length by Width by Height), with the shaking in the direction of the length. FSB1 has dimensions of about 1.7 x 0.7 x 0.6 m. The FSB container consists of a stack of rectangular rings of aluminum tubing and rubber. The rubber, being flexible in shear allows the container to deform with the soil layer. The stiffness of the rubber is designed to make the natural frequency of the container (as a shear beam) to be less than the natural frequency of a soil layer. SCHAEVITZ CENTRIFUGE AND SHAKER The 1 m radius centrifuge can carry 90 kg models to 100 g or 27 kg models to 175 g. The centrifuge is equipped with a servo-hydraulic actuator which is capable of simulating earthquake-like motions to the base of the model container during flight. The shaker has been used to study the dynamic behavior of retaining structures, liquefiable soil layers, response of soft soil sites, soil-pile-structure interaction, flow slides, embankments and port structures. Over a thousand model earthquakes have been produced by this

A-2

shaker, and it was the primary focus of the centrifuge modeling efforts of the faculty at UCD until completion of the large shaker in 1995. The shaker on the 1 m centrifuge has a model container with dimensions 0.56 x 0.28 x 0.18 m. The shaker, capable of producing 30 g shaking accelerations is designed to operate at centrifuge accelerations up to 100 g. A computer outside the centrifuge is used to control the shaker and to acquire data from the experiment. Sixteen channels of data can be digitized at rates up to 2500 samples per second per channel. Sixteen variable gain amplifiers for strain gage type instruments and eight variable gain charge amplifiers for accelerometers are mounted on the centrifuge to improve the signal to noise ratio. This centrifuge is equipped with four hydraulic rotary joints which may be used for hydraulic oil, water supply, or compressed gas. On board accumulators store energy in compressed gas to provide the short burst of high power required to produce large flow rates of high pressure oil to for the shaker. The hydraulic system has also been used to apply static loads to foundations and other mechanisms on the centrifuge.

FURTHER INFORMATION Anyone interested in using or collaborating in the use of the centrifuge facilities is encouraged to contact: Bruce L. Kutter, Director, Center for Geotechnical Modeling Department of Civil and Environmental Engineering University of California Davis CA 95616 Tel: 916 752 8099 Fax: 916 752 7872 email: [email protected] References Kutter, B.L., Li, X.S., Sluis, W., and Cheney, J.A. (1991) "Performance and Instrumentation of the Large Centrifuge at UC Davis", Centrifuge 91, Ko and McLean (eds.), Balkema, Rotterdam. Kutter, B.L., Idriss, I.M., Kohnke, T., Lakeland, J., Li, X.S., Sluis, W., Zeng, X., Tauscher, R.C., Goto, Y. Kubodera, I., (1994) "Design of a Large Earthquake Simulator at UC Davis, Centrifuge 94, Leung, Lee, and Tan (eds.), Balkema, Rotterdam, pp. 169175.

A-3