Deepsoil User Manual v6 PDF
Deepsoil User Manual v6 PDF
Deepsoil User Manual v6 PDF
April 8, 2015
USER MANUAL
Youssef M. A. Hashash
TABLE OF CONTENTS
1 Program Background and Installation ................................................................................ 6
1.1 About the Program .............................................................................................................. 6
1.2 Historical Development ...................................................................................................... 7
1.3 Program installation .......................................................................................................... 10
2 Program Organization ....................................................................................................... 11
2.1 Profiles Tab ....................................................................................................................... 13
2.2 Motions Tab ...................................................................................................................... 13
2.2.1 Baseline Correction ................................................................................................. 14
2.2.2 Response Spectra Calculation Methods................................................................... 15
2.2.3 Fourier amplitude spectrum calculation and averaging ........................................... 17
2.2.4 Arias Intensity.......................................................................................................... 18
2.2.5 Housner Intensity ..................................................................................................... 19
2.2.6 Convert Input Motion (to be updated) ..................................................................... 19
2.2.7 Adding New Input Motions ..................................................................................... 20
2.3 Analysis Tab ..................................................................................................................... 21
3 Analysis Flow ................................................................................................................... 22
3.1 Analysis Definition: Step 1 of 6........................................................................................ 22
3.1.1 Equivalent Linear Analysis ..................................................................................... 24
3.1.2 Deconvolution via Frequency Domain Analysis ..................................................... 24
3.1.3 Non-Linear Analysis ................................................................................................ 24
3.2 Defining Soil Profile & Model Properties: Step 2a of 6 ................................................... 26
3.2.1 Creating/Modifying Soil Profiles ............................................................................ 28
3.2.2 Maximum Frequency (for Time Domain Analysis only) (Step 2b) ........................ 28
3.2.3 Implied Strength Profile (Step 2b)........................................................................... 29
3.3 Define Rock Properties: Step 2c of 6 ................................................................................ 31
3.4 Output and Motion Selection: Step 3 of 6 ........................................................................ 32
3.5 Viscous Damping .............................................................................................................. 34
3.5.1 Viscous Damping Formulation in Nonlinear Analysis (Time Domain) (Step 4) .... 34
3.5.2 Viscous Damping in Equivalent Linear Analysis (Frequency Domain) (Step 5) ... 39
3.6 Analysis Control Parameters: Step 5 of 6 ......................................................................... 39
3.6.1 Frequency domain analysis ..................................................................................... 40
3.6.2 Time domain analysis .............................................................................................. 41
3.7 Output: Step 6 of 6 ............................................................................................................ 42
3.7.1 Output data file ........................................................................................................ 43
3.7.2 Summary Profiles .................................................................................................... 44
3.7.3 Displacement profile and animation ........................................................................ 45
3.7.4 Convergence results (Equivalent Linear Analyses Only) ........................................ 46
3.7.5 Input Summary ........................................................................................................ 47
4 Soil Models ....................................................................................................................... 48
4.1 Backbone Curves .............................................................................................................. 48
4.1.1 Hyperbolic / Pressure-Dependent Hyperbolic ......................................................... 48
4.1.2 GQH Model ............................................................................................................. 49
4.2 Hysteretic (unload-reload behavior) behavior .................................................................. 49
4.2.1 Masing Rules ........................................................................................................... 49
8.8 Example 8 Non-linear Analysis / Multi-Layer, Elastic Rock, Pore Water Pressure
Generation and Dissipation ............................................................................................. 113
8.9 Example 9 Equivalent Linear Frequency Domain Analysis / Multi-Layer, Elastic Rock,
Bay Mud Profile .............................................................................................................. 116
8.10 Example 10 Non-linear Analysis / Multi-Layer, Rigid Rock, Treasure Island Profile .. 116
8.11 Example 11 Non-linear Analysis / Multi-Layer, Elastic Rock, MRDF.......................... 116
LIST OF FIGURES
Figure 1. DEEPSOIL Main Window and Key Tabs ..................................................... 11
Figure 2. DEEPSOIL Options Window. ....................................................................... 12
Figure 3. Motion Viewer (Plots) ................................................................................... 13
Figure 4. Motion Viewer (Tables) ................................................................................ 14
Figure 5. Baseline Correction. ...................................................................................... 15
Figure 6. Input Motion Conversion. ............................................................................. 20
Figure 7. Converted Motion.......................................................................................... 20
Figure 8. Step 1/6: Choose type of analysis. ................................................................ 22
Figure 9. Step 2a/6: Input Soil Properties. ................................................................... 26
Figure 10. Profile Summary.......................................................................................... 30
Figure 11. Step 2b/6: Input Rock Properties................................................................. 31
Figure 12. Step 3/6: Input Motion and Output Layer(s) (Time History Plots Tab) ...... 33
Figure 13. Step 3/6: Input Motion and Output Layer(s) (Spectral Plots Tab) .............. 34
Figure 14. Step 4/6: Small-Strain Damping Formulation. ............................................ 35
Figure 15. Step 5/6: Analysis Options for Frequency Domain or Time Domain
Analysis................................................................................................................. 40
Figure 16. Step6/6: Analysis Results - Plot Output for Layer. ..................................... 43
Figure 17. Summary Profiles ........................................................................................ 44
Figure 18. Column Displacement Animation ............................................................... 45
Figure 19. Convergence Check. .................................................................................... 46
Figure 20. Input Summary ............................................................................................ 47
DEEPSOIL is a one-dimensional site response analysis program that can perform: a) 1-D
nonlinear time domain analyses with and without pore water pressure generation, and b) 1-D
equivalent linear frequency domain analyses including convolution and deconvolution.
DEEPSOIL was developed under the direction of Prof. Youssef M.A. Hashash in collaboration
with several graduate and undergraduate students including Duhee Park, Chi-Chin Tsai, Camilo
Phillips, David R. Groholski, Daniel Turner, Michael Musgrove, Byungmin Kim and Joseph
Harmon at the University of Illinois at Urbana-Champaign.
When referencing the DEEPSOIL program in a publication (such as journal or conference papers,
or professional engineering reports) please use the following reference format:
Hashash, Y.M.A., Musgrove, M.I., Harmon, J.A., Groholski, D.R., Phillips, C.A., and Park, D.
(2015) “DEEPSOIL 6.0, User Manual” 116 p.
The program is provided as-is and the user assumes full responsibility for all results. The
use of the DEEPSOIL program requires knowledge in the theory and procedures for seismic site
response analysis and geotechnical earthquake engineering. It is suggested that the user reviews
relevant literature and seek appropriate expertise in developing input of the analysis and
interpretation of the results.
Initial development of DEEPSOIL was based on research supported in part through Earthquake
Engineering Research Centers Program of the National Science Foundation under Award
Number EEC-9701785; the Mid-America Earthquake Center. Any opinions, findings, and
conclusions or recommendations expressed in this material are those of the authors and do not
necessarily reflect the views of the National Science Foundation. The authors gratefully
acknowledge this support.
By using this program the user(s) agree to indemnify and defend Youssef Hashash and the
University of Illinois against all claims arising from use of the software and analysis results by
the user(s) including all third party claims related to such use.
DEEPSOIL implements the Armadillo C++ linear algebra library (Sanderson, 2010). Armadillo
is open-source software released under the Mozilla Public License 2.0. A copy of this license is
available at https://www.mozilla.org/MPL/2.0/. You may obtain a copy of the Armadillo source
code at http://arma.sourceforge.net/download.html.
DEEPSOIL has been under development at UIUC since 1998. The driving motivation of the
development of DEEPSOIL was and continues to be making site response analysis readily
accessible to students, researchers and engineers worldwide and to support research activities at
UIUC.
As with any development, DEEPSOIL has benefited from many prior contributions by other
researchers as well as current and former students at UIUC. For the interested reader, a detailed
description of many of the theoretical developments and the background literature can be found
in the following publications:
Hashash, Youssef M. A., and Duhee Park (2001) "Non-linear one-dimensional seismic ground motion propagation
in the Mississippi embayment," Engineering Geology, Vol. 62, No. 1-3, pp 185-206.
Hashash, Y. M. A., and D. Park (2002) "Viscous damping formulation and high frequency motion propagation in
nonlinear site response analysis," Soil Dynamics and Earthquake Engineering, Vol. 22, No. 7, pp. 611-624.
Hashash, Y. M.A., Chi-Chin Tsai, C. Phillips, and D. Park (2008) "Soil column depth dependent seismic site
coefficients and hazard maps for the Upper Mississippi Embayment," Bull. Seism. Soc. Am., Vol. in press.
Hashash, Y.M.A., Phillips, C. and Groholski, D. (2010). "Recent advances in non-linear site response analysis",
Fifth International Conference on Recent Advances in Geotechnical Earthquake Engineering and Soil Dynamics,
Paper no. OSP 4.
Park, D. (2003) "Estimation of non-linear seismic site effects for deep deposits of the Mississippi Embayment," Ph.D.
Thesis. Department of Civil and Environmental Engineering. Urbana: University of Illinois, p 311 p.
Park, D., and Y. M. A. Hashash (2004) "Soil damping formulation in nonlinear time domain site response analysis,"
Journal of Earthquake Engineering, Vol. 8, No. 2, pp 249-274.
Park, D., and Y.M.A. Hashash (2005) "Estimation of seismic factors in the Mississippi Embayment: I. Estimation of
dynamic properties," Soil Dynamics and Earthquake Engineering, Vol. 25, pp. 133-144.
Park, D., and Y.M.A. Hashash (2005) "Estimation of seismic factors in the Mississippi Embayment: II. Probabilistic
seismic hazard with nonlinear site effects," Soil Dynamics and Earthquake Engineering, Vol. 25, pp. 145-156.
Tsai, Chi-Chin (2007) "Seismic Site Response and Interpretation of Dynamic Soil Behavior from Downhole Array
Measurements," Ph.D. Thesis. Department of Civil and Environmental Engineering. Urbana: University of Illinois
at Urbana-Champaign.
Tsai, Chi-Chin, and Y. M. A. Hashash (2008) "A novel framework integrating downhole array data and site
response analysis to extract dynamic soil behavior," Soil Dynamics and Earthquake Engineering, Vol. Volume 28,
No. Issue 3, pp 181-197.
Tsai, Chi-Chin, and Youssef M.A. Hashash (2009) "Learning of dynamic soil behavior from downhole arrays,"
Journal of geotechnical and geoenvironmental engineering, Vol. in press.
Phillips, Camilo, and Youssef M. A. Hashash (2008) "A new simplified constitutive model to simultaneously match
modulus reduction and damping soil curves for nonlinear site response analysis," Geotechnical Earthquake
Engineering & Soil Dynamics IV (GEESD IV). Sacramento, California.
Phillips, C. and Hashash, Y. (2009) “Damping formulation for non-linear 1D site response analyses” Soil
Dynamics and Earthquake Engineering, Vol. 29, No. 7, pp 1143-1158.
• DEEPSOIL v1.0: First version of DEEPSOIL with both an equivalent linear analysis
capability and a new pressure dependent hyperbolic model in nonlinear analysis:
• The equivalent linear capability was based on the pioneering work of Idriss and Seed
(1968), and Seed and Idriss (1970) as employed in the widely used program SHAKE
(Schnabel, et al., 1972) and its more current version SHAKE91 (Idriss and Sun, 1992).
• The new pressure dependent hyperbolic model introduced by Park and Hashash (2001)
is employed in nonlinear analysis. This model extended the hyperbolic model
introduced by Matasovic (1992) and employed in the nonlinear site response code D-
MOD, which was in turn a modification of the Konder and Zelasko (1963) hyperbolic
model. The hyperbolic model had been employed with Masing criteria earlier in the
program DESRA by Lee and Finn (1975, 1978). The hyperbolic model was
originally proposed by Duncan and Chang (1970), with numerous modifications in
other works such as Hardin and Drnevich (1972) and Finn et al. (1977).
• DEEPSOIL v2.0-2.6:
• Full and extended Rayleigh damping is introduced in DEEPSOIL (Hashash and Park,
2002; Park and Hashash, 2004) with a user interface. This was in part based on
Clough and Penzein (1993) and the findings of Hudson et al. (1994) as implemented
in the program QUAD4-M.
• DEEPSOIL v3.0-3.7: Additional enhancements are made to the user interface as well as
inclusion of pore water pressure generation/dissipation capability.
• Current pore water pressure models employed include the same model introduced by
Matasovic (1992), Matasovic and Vucetic (1993, 1995) and employed in the program
D_MOD.
• DEEPSOIL v3.5: A new soil constitutive model is introduced to allow for significantly
enhanced matching of both the target modulus reduction and damping curves (Phillips
and Hashash, 2008).
• A new functionality in the user interface is implemented that allows the user to
automatically generate hyperbolic model parameters using a variety of methods
(Phillips and Hashash, 2008).
• DEEPSOIL v3.7: A new pore water pressure generation model for sands is added –
the GMP Model (Green et al., 2000), in addition to various improvements in the user
interface, as well as the capability to export output data to a Microsoft Excel file.
• DEEPSOIL was made multi-core aware, leading to much faster completion of batch-
mode analyses.
• An update manager was added to notify the user when updated versions of
DEEPSOIL were available.
Hardware Requirements
2 GHz or faster processor*
2 GB or more available RAM
200 MB available on hard drive for installation
Software Requirements
Windows 7 or later
Microsoft .NET Framework 4 (Client Profile)
Administrator privileges are required for installation
Installation
Run “DEEPSOIL Installer.exe”
The DEEPSOIL installer will automatically detect if you system supports 64-bit installations and
install the appropriate libraries
2 Program Organization
The DEEPSOIL graphical user interface is composed of several steps to guide the user
throughout the site response analysis process as illustrated in the Navigation box shown in Figure
1 presented to the user upon starting DEEPSOIL.
At the top left, the user has the option of choosing the “Analysis,” “Motions,” or “Profiles” tab.
These tabs are discussed in the following section.
Figure 2 shows the Options window. This window can be accessed by clicking on the “Options”
menu. The window allows the user to set the default working directory, the directory containing
input motions for use in analyses, the default directory in which to save profiles, the default units,
the analysis priority, and enable or disable multi-core support.
DEEPSOIL contains a motion tab which can be used to view/process input motions. To
view/process a motion, simply select it from the list and press the View button. A new window
will open (Figure 3) and DEEPSOIL will generate acceleration, velocity, and displacement and
Arias intensity time histories, as well as the response spectrum and Fourier amplitude spectrum
for the selected motion. The relative size of the plots can be adjusted by clicking on the gray
vertical line and dragging to the left or right. Double-clicking on the response spectrum and
Fourier amplitude spectrum plots will cause the axes to alternate between linear and log scales
on the axes (each plot supports 3 different views). The calculated data is also provided for the
user in data tables which can be accessed by selecting the “Time History Data” or “Spectral Data”
tabs at the top of the window (Figure 4).
This window also provides the user the option to linearly scale the selected input motion. The
user is provided two options for scaling: scale the original motion by a specified factor (scale by)
or scale the original motion to a specified maximum acceleration (scale to). The desired method
can be selected using the drop-down list in the upper right corner of the window. Press the
Apply button to scale the motion and recalculate the other data. After scaling, the user can save
the new motion by pressing the Save As button.
DEEPSOIL can perform baseline correction for any input motion (Figure 5). By selecting an
input motion and pressing the Baseline Correction button, a new window appears which shows
the acceleration, velocity, and displacement time-histories corresponding to the motion. Motions
which exhibit non-zero displacement time-histories for the latter part of the motion should be
corrected. The corrected time-histories are also calculated and presented to the user. The
response spectra and Fourier amplitude spectra for the original motion and baseline-corrected
motion are also provided for the user. The spectra should be carefully examined by the user to
ensure the baseline correction process did not greatly alter the input motion. The baseline-
corrected motion can then be stored as a file defined by the user. The relative size of the plots
can be adjusted by clicking on the gray vertical line and dragging it to the left or right. Dragging
to the left causes the response spectra and Fourier amplitude spectra plots to increase in size,
while dragging to the right causes the time-histories plots to increase in size.
The baseline correction routine in DEEPSOIL is adapted from the baseline correction routine
included in the USGS motion processing program BAP (USGS Open File Report 92-296A). The
baseline correction is accomplished using the following steps:
1. Truncate both ends of the motion using the first and last zero-crossings as bounds.
The frequency-domain solution, the Newmark β method and Duhamel integral solutions are the
three most common methods employed to estimate the response of Single Degree of Freedom
(SDOF) systems and therefore to calculate the response spectra. A brief description is presented
for each method to calculate the response of SDOF systems and to solve the dynamic
equilibrium equation defined as (Chopra, 1995; Newmark, 1959):
where m, c and k are the mass, the viscous damping and the system stiffness of SDOF system
respectively. ü , u̇ and u are the nodal relative accelerations, relative velocities and relative
displacements respectively and ü g is the exciting acceleration at the base of SDOF.
Frequency-domain solution
In the frequency-domain solution, the Fourier Amplitude Spectra (FAS) input motion is modified
−𝑓𝑓𝑛𝑛2
H(f) = 2
(f − 𝑓𝑓𝑛𝑛2 ) − 2𝑖𝑖𝑖𝑖𝑓𝑓𝑓𝑓𝑛𝑛
1
where fn is the natural frequency of the oscillator calculated as fn = 2π �k⁄m and ξ is the
c
damping ratio calculated as ξ = 2√km. Use of the frequency-domain solution requires FFTs (Fast
Fourier Transforms) to move between the frequency-domain, where the oscillator transfer
function is applied, and the time-domain, where the peak oscillator response is estimated. Over
the frequency range of the ground motion, the frequency-domain solution is exact.
where:
𝜉𝜉
𝐴𝐴 = 𝑒𝑒 −𝜉𝜉𝜔𝜔𝑛𝑛Δ𝑡𝑡 � 𝑠𝑠𝑠𝑠𝑠𝑠(𝜔𝜔𝐷𝐷 Δ𝑡𝑡) + 𝑐𝑐𝑐𝑐𝑐𝑐(𝜔𝜔𝐷𝐷 Δ𝑡𝑡)�
�1 − 𝜉𝜉 2
1
𝐵𝐵 = 𝑒𝑒 −𝜉𝜉𝜔𝜔𝑛𝑛Δ𝑡𝑡 � 𝑠𝑠𝑠𝑠𝑠𝑠(𝜔𝜔𝐷𝐷 Δ𝑡𝑡)�
𝜔𝜔𝐷𝐷
𝜔𝜔𝑛𝑛
𝐴𝐴′ = −𝑒𝑒 −𝜉𝜉𝜔𝜔𝑛𝑛Δ𝑡𝑡 � 𝑠𝑠𝑠𝑠𝑠𝑠(𝜔𝜔𝐷𝐷 Δ𝑡𝑡)�
�1 − 𝜉𝜉 2
𝜉𝜉
𝐵𝐵 ′ = −𝑒𝑒 −𝜉𝜉𝜔𝜔𝑛𝑛Δ𝑡𝑡 �𝑐𝑐𝑐𝑐𝑐𝑐(𝜔𝜔𝐷𝐷 Δ𝑡𝑡) − 𝑠𝑠𝑠𝑠𝑠𝑠(𝜔𝜔𝐷𝐷 Δ𝑡𝑡)�
�1 − 𝜉𝜉 2
1 1 𝜔𝜔𝑛𝑛 𝜉𝜉 1
𝐶𝐶 ′ = �− + 𝑒𝑒 −𝜉𝜉𝜔𝜔𝑛𝑛Δ𝑡𝑡 �� + � 𝑠𝑠𝑠𝑠𝑠𝑠(𝜔𝜔𝐷𝐷 Δ𝑡𝑡) + 𝑐𝑐𝑐𝑐𝑐𝑐(𝜔𝜔𝐷𝐷 Δ𝑡𝑡)��
𝑘𝑘 Δ𝑡𝑡 �1 − 𝜉𝜉 2 Δ𝑡𝑡�1 − 𝜉𝜉 2 Δ𝑡𝑡
1 𝜉𝜉
𝐷𝐷′ = �1 − 𝑒𝑒 −𝜉𝜉𝜔𝜔𝑛𝑛Δ𝑡𝑡 � 𝑠𝑠𝑠𝑠𝑠𝑠(𝜔𝜔𝐷𝐷 Δ𝑡𝑡) + 𝑐𝑐𝑐𝑐𝑐𝑐(𝜔𝜔𝐷𝐷 Δ𝑡𝑡)��
𝑘𝑘Δ𝑡𝑡 �1 − 𝜉𝜉 2
The parameters β and γ define the assumption of the acceleration variation over a time step (Δt)
and determine the stability and accuracy of the integration of the method. A unique
characteristic of the assumption of average acceleration (β = 0.5 and γ = 0.25) is that the
integration is unconditionally stable for any Δt with no numerical damping. For this reason, the
Newmark β method with average acceleration is commonly used to model the dynamic response
of single and multiple degree of freedom systems.
The Newmark β method has inherent numerical errors associated with time step of the input
motion (Chopra, 1995; Mugan and Hulbe, 2001). These errors generate inaccuracy in the
solution resulting in miss-prediction of the high-frequency response. To determine if a motion’s
time step is too large to be used directly, the response spectrum calculated with the Newmark β
method can be compared with the response spectra calculated by other means and with and
without a time step correction in the motion viewer/processor (see section 2.2).
i
fi =
time step ∗ n
where fi is the i-th frequency, n is the number of points in the FFT, |F|i is the Fourier amplitude
at the i-th frequency, and Ci is the i-th amplitude and phase (in complex number representation)
of the FFT. The maximum frequency that can be contained in the motion is dictated by the
motion’s time step. This maximum frequency is called the Nyquest frequency and is calculated
using the following equation:
1
fNyquest =
2 ∗ time step
DEEPSOIL can also smooth the calculated Fourier amplitude spectrum to make interpretation
easier by providing a clearer view of the overall frequency content. DEEPSOIL uses a triangle
smoother in log space (also called a log-triangle smoother). The smoothing routine in
DEEPSOIL uses a sliding triangular smoothing window in log-space and is adapted from a
routine developed by David Boore. The weights assigned to each point are based on the log
distance from the point of interest. We currently have our maximum smoothing width set to 0.2.
At each frequency of the spectrum the weights of the smoothing window are calculated as
follows:
Wi = 1
where the upper and lower bound indices are determined using the desired window width and
index of the current frequency.
𝑡𝑡
𝜋𝜋
𝐼𝐼𝑎𝑎 (𝑡𝑡) = �[𝑎𝑎(𝑡𝑡)]2 𝑑𝑑𝑑𝑑
2𝑔𝑔
0
where 𝑔𝑔 is the acceleration due to gravity and 𝑎𝑎(𝑡𝑡) is the acceleration time history.
𝑡𝑡 2.5
where T is the period and 𝜉𝜉 is the damping ratio. In DEEPSOIL, the Housner intensity is
calculated assuming a damping ratio of 5%.
By clicking Convert Input Motion you will be able to convert the motion from NEHRP Site
Class A to Site Class B/C boundary conditions and vice versa (Figure 6). This option is
particularly useful in using the generated ground motions from the USGS website. The USGS
website generates motions at Site class B/C boundary, which have to be converted to Site class A
to be imposed at the bottom of the bedrock. The converted input motion is then plotted for
comparison to the original motion (Figure 7). The user is provided with the option to save the
new motion.
USGS hazard maps are developed for a Site Class B/C boundary (according to 1997 NEHRP
Provisions) that represent a weak rock condition. The USGS website allows generation of
representative ground motions anywhere in the U.S. The generated motions represent motions at
a Site Class B/C boundary. The motions can be used as input motions imposed at the bottom of
the soil column. However, the motions cannot be used in the original form. The motion has to be
converted to Site Class A condition, which represents a hard rock condition. DEEPSOIL allows
converting of a Site Class A motion to Site Class B/C motion and vice versa.
Motions may be added to DEEPSOIL by using the built-in Add Motion window. To access this
tool, click on the Motions tab of the main DEEPSOIL window and press the Add button.
Alternatively, click on the File menu and select New and then Motion. This tool is designed to
convert motions from the PEER “.AT2” format to the DEEPSOIL format. This process is fully
automated. DEEPSOIL will read through the PEER file and determine the number of data points
and the time step. Additional options are provided for reading non-PEER motions and should be
set as needed. If DEEPSOIL cannot complete the conversion, a message box is used to notify
the user of the failure. Upon successful conversion, the user is notified by a message box and the
motion is added to the Motion Library.
Motions can also be added manually. This is done using a text editor capable of producing .TXT
files. To add an input motion, enter the necessary data in the format described below and save as
a .TXT file in the “Input Motion” directory. The default input motion directory is:
C:\Users\[User Name]\Documents\DEEPSOIL\Input Motions\. If the user has specified a
different directory, the input motion file should be placed in the user-specified directory. If this
method is used, DEEPSOIL must be closed and reopened before the input motion is available for
analyses.
The analysis tab options are discussed in detail in the next section.
3 Analysis Flow
Before creating a new profile, or opening an existing profile, it is recommended to verify the
“Current Workspace Directory” at the bottom of the page. The DEEPSOIL “Working” directory
is chosen by default as the default working directory specified using the Options window (Figure
2). If a different directory is preferred, press the “Change” button to bring up a folder browser
and select the preferred directory.
To create a new analysis, the user must specify the type of analysis before proceeding to the next
stage of analysis. The user must specify:
6. The porewater pressure boundary condition at the bottom of the soil profile (for
analysis with PWP generation and dissipation)
Permeable
Impermeable
The pore water pressure generation and dissipation options are only available for nonlinear (time
domain) analyses. Note that (2) and (3) can also be changed in the next stage.
The option of defining the soil curves using discrete points is only applicable for the Equivalent
Linear analysis. For this option, the G/Gmax (-) and damping ratio (%) are defined as functions
of shear strain (%).
Deconvolution requires definition of a soil profile. The following properties need to be defined
for each layer:
• Thickness
• Shear Wave Velocity (𝑉𝑉𝑠𝑠 ) or Initial Shear Modulus (𝐺𝐺𝑚𝑚𝑚𝑚𝑚𝑚 )
• Damping Ratio (%)
• Unit Weight
Note: Deconvolution cannot be performed in the time domain analysis. Finding the motion at the
bottom of the soil profile given the motion at the ground surface is an inverse problem in
nonlinear analysis that is complex to solve and is not amenable to a simple deconvolution
computation.
Several soil models are available for user to select from. The analysis can be with or without
porewater pressure generation.
The user has the option of obtaining the site response results using the equivalent linear method
automatically whenever nonlinear site response analysis is conducted. It is highly recommended
that EL results be always examined whenever a NL analysis is conducted.
This stage is divided into three partitions. The first partition to be considered requires the user to
define the soil profile and specify the soil properties of each layer (Figure 9). The type of input
required depends on the analysis parameters selected in Step 1.
The entire form is broken up into three sections. The section located at the left is a visual display
of the soil profile. The section at the right is the table where the values for required input
parameters must be entered. The section at the bottom contains information about the soil
column, options for adding/removing layers, water table settings, and conversion functions.
If a total stress analysis is selected, the user must specify the typical soil properties of each layer
based on the type of analysis that was selected (Linear, Nonlinear, etc).
If an effective stress analysis is selected, the user must specify additional parameters including
the model to be used (Sand/Clay) and their respective parameters. The models are identified as
Sand (S) or Clay (C), and by the initials of the model developer (e.g. M for Matasovic, D for
Dobry, GMP for Green, Mitchell, Polito):
• F/A/ FC (Define F for Sand model, A for Clay model, FC (%) for GMP model)
• s/B/- (Define s for Sand model, b for Clay model, leave blank for GMP model)
• g/C/- (Define g for Sand model, C for Clay model, leave blank for GMP model)
• v/D/v (Define v for Sand model, Define D for Clay model, v for GMP model)
• -/g/- (Leave blank for Sand model, Define g for Clay model, leave blank for GMP)
If an effective stress analysis is selected with the option to Include PWP Dissipation, the user
must also specify:
*For v, Matasovic (1993) recommends a value ranging from 3.5 – 5.0, with an average value of
3.8.
The GMP model parameters, which can be used for sands, are:
• α = Scale Factor
• Dr (%) = Relative density
• FC (%) = Fines Content
• v = Curve fitting parameter* (same as used in the Matasovic (1993) Sand model)
For “Effective Stress Analysis” with the “Include PWP Dissipation” option:
• Cv = Coefficient of consolidation
Units
b. Convert Units: Convert all units from English to Metric or vice versa.
c. Convert Shear: Convert shear modulus to shear wave velocity or vice versa. All layers
require a unit weight to perform this conversion.
d. Water Table: Choose the depth of the water table by clicking the drop-down menu. The
layers appear in ascending order, so click the layer that the water table will be above. The
Graphical soil column display responds to this by changing the background color of every
layer beneath the water table to blue. The location of the water table is only of influence
when introducing the pressure dependent soil parameters or performing an effective stress
analysis. The location of the water table does not influence the frequency domain solution.
3.2.2 Maximum Frequency (for Time Domain Analysis only) (Step 2b)
Upon completing the definition of the soil and model properties, the user is shown a plot of the
maximum frequency versus depth for each layer (Figure 10). A plot of maximum frequencies
(Hz) versus depths of all layers are displayed. The maximum frequency is the highest frequency
that the layer can propagate and is calculated as: fmax = Vs/4H, where Vs is the shear wave
velocity of the layer, and H is the thickness of the layer. To increase the maximum frequency, the
thickness of the layer should be decreased. This check is performed solely for time domain
analyses. It is recommended that the layers have the same maximum frequency throughout the
soil profile, though this is not required. For all layers, the maximum frequency should generally
be a minimum of 30 Hz.
Upon completing the definition of the soil and model properties, the user is shown a plot of the
implied strength of the soil profile. The window provides three plots for the user to view:
implied shear strength versus depth, normalized implied shear strength (shear strength divided by
effective vertical stress) versus depth, and implied friction angle versus depth (Figure 10). The
shear strength and friction angle are also provided in the table to the right for closer inspection.
The implied shear strength is calculated from the modulus reduction curves entered as part of
step 2a. At each point on the curve, the shear stress is calculated using the following equation:
𝐺𝐺
𝜏𝜏 = 𝜌𝜌𝑉𝑉𝑠𝑠2 𝛾𝛾
𝐺𝐺0
The maximum value of shear stress for the given layer is then plotted at the depth corresponding
to that layer. Using this maximum value, the implied friction angle is then calculated using the
following equation:
𝜏𝜏𝑚𝑚𝑚𝑚𝑚𝑚
𝜙𝜙 = 𝑡𝑡𝑡𝑡𝑡𝑡−1 � ′ �
𝜎𝜎𝑣𝑣
The user is encouraged to carefully check the provided plots. If the implied strength or friction
angle of particular layer is deemed unreasonable, the user should consider modifying the
modulus reduction curve for the layer to provide a more realistic implied strength or friction
angle.
After defining the soil and model properties, the user must now define the rock / half-space
properties of the bottom of the profile (Figure 11).
The user has the option of selecting either a Rigid Half-Space or an Elastic Half-Space. An
informational display makes the user aware that a rigid half-space should be chosen if a within
motion will be used, and an elastic half-space should be selected if an outcrop motion is being
used. If a rigid half-space is being used, no input parameters are required. If an elastic half-
space is being used, the user must supply the shear wave velocity (or modulus), unit weight, and
damping ratio of the half-space. In general, the shear wave velocity of the bedrock should be
greater than that of the overlying soil profile.
Bedrock properties may be saved by giving the bedrock a name and pressing the Save Bedrock
button. The new bedrock will appear in the list of saved bedrocks below. To use a saved
bedrock, select the file from the list box and press the Load button.
The motion and output selection stage allows the user to select layers for time-history output and
specify the input motion(s) to be used in the analysis.
The layers at which output data is needed may be selected by checking the appropriate checkbox
in the first column of the window. All layers can be selected or deselected using the Select All
button located at the bottom of the layer list. Note that requesting time-history output for
additional layers will increase the time required for analyses to complete. Maximum PGA, stress,
strain and pore pressure (if applicable) profiles will be generated regardless of the layer output
selection. Therefore, it is recommended that the user only request time-history output for layers
of interest.
The input motion(s) must be selected from the current input motion library (to which the user
may add additional motions, see section 2.2.7). The motions may be selected by checking the
appropriate checkbox in the second column of the window. All motions can be selected or
deselected by using the Select All button at the bottom of the motion list. Once a motion is
selected, DEEPSOIL will calculate and plot the acceleration, velocity, displacement, and Arias
intensity time histories (Figure 12) and the response and Fourier amplitude spectrum (Figure 13).
If multiple motions are selected, a single motion can be highlighted in the plots by clicking on it
in the motion list or clinking in its column in the table below the plots. The table also allow for
control of which motions are displayed in the plots. Buttons are available at the bottom of the
window to change the colors of the plots.
The user should also enter the damping ratio for the calculated response spectra. The response
spectra are calculated using the frequency domain method (see section 2.2.2) and the default
damping ratio is 5%. This value may be adjusted at the user’s discretion.
Figure 12. Step 3/6: Input Motion and Output Layer(s) (Time History Plots Tab)
Figure 13. Step 3/6: Input Motion and Output Layer(s) (Spectral Plots Tab)
This stage will only appear for time domain analyses. This step allows the user to set the viscous
damping formulation and select the optimum modes/frequencies for the analysis (Figure 14).
This window is unique to DEEPSOIL. This window will help control the introduction of
numerical damping through frequency dependent nature of the viscous damping formulation.
Note that when multiple input motions are selected for an analysis, the viscous damping
formulation and selected modes/frequencies are the same for all selected input motions.
• Graph Lin. Freq. Domain – Graphs the linear frequency domain for specified options
above
• Check with Lin. Time Domain – Graphs corresponding linear time domain
• Clear Time Plots – Clears the time domain graphs
• Show Rayleigh Damping – Graphs the Rayleigh damping, not available for frequency
independent formulation
For more details on this stage, please refer to Example 6 in the tutorial.
Viscous damping formulation is used to model small strain damping. The viscous damping
formulation results in frequency dependent damping and can introduce significant artificial
damping. It is therefore important to select an appropriate viscous damping formulation and
corresponding coefficients to reduce the numerical damping (Hashash and Park, 2002; Park and
Hashash, 2004). There are three types of Rayleigh damping formulations in DEEPSOIL, as listed
below. It is, however, recommended that the frequency independent damping formulation be
selected for most analyses.
This procedure solves for the eigenvalues and eigenvectors of the damping matrix and requires
no specification of modes or frequencies. This formulation removes many of the limitations of
Rayleigh Damping and does not greatly increase the required analysis time in most situations. A
complete explanation of the damping formulation is presented in Phillips and Hashash, 2009.
A complete explanation of the extended Rayleigh damping formulation is presented in Park and
Hashash, 2004.
Modes/frequencies selection
There are two options available for selecting modes. The first option is choosing the natural
modes (e.g. 1st and 2nd modes). The second option is choosing the frequencies for Rayleigh
damping directly. The resulting Rayleigh damping curve can be displayed by pressing Show
Rayleigh Damping and the curve will be displayed at the right bottom window. Note again that
the viscous damping is frequency dependent. The goal in time domain analysis is to make the
viscous damping as constant as possible at significant frequencies.
The time domain solution uses the frequency dependent Rayleigh damping formulation, whereas
actual viscous damping of soils is known to be fairly frequency independent. The frequency
domain solution uses frequency independent viscous damping. The appropriateness of the chosen
modes/frequencies should be therefore verified with the linear frequency domain solution.
Press Graph Lin. Freq. Domain. The results of the linear frequency domain solution
(Frequency ratio vs. Freq. and Response spectrum plots) will be displayed as blue curves. The
goal is to choose the appropriate modes/frequencies that compare well with the linear frequency
domain solution.
Enter the desired modes/frequencies as input. Then press the Check with Lin. Time Domain
button. The results (in the same window as frequency domain solution) will be displayed as pink
curves. Choose the modes/frequencies that agree well with the linear frequency domain solution.
This is an iterative procedure and optimum modes/frequencies should be chosen by trial and
error.
This option is only applicable for nonlinear solutions. During the excitation, soil stiffness and the
frequencies corresponding to the natural modes of the profile change at each time step. The
natural modes selected are recalculated at each time step to incorporate the change in stiffness
and the damping matrix is recalculated.
This feature is enabled by clicking the Yes button in the Damping Matrix Update selection
window. Note that using this feature may significantly increase the time required to complete an
analysis.
𝐺𝐺 ∗ = 𝐺𝐺(1 + 𝑖𝑖2𝜉𝜉)
𝐺𝐺 ∗ = 𝐺𝐺 �1 − 2𝜉𝜉 2 + 𝑖𝑖2�1 − 𝜉𝜉 2 �
𝐺𝐺 ∗ = 𝐺𝐺(1 − 𝜉𝜉 2 + 𝑖𝑖2𝜉𝜉)
Figure 15. Step 5/6: Analysis Options for Frequency Domain or Time Domain Analysis.
• Number of Iterations
• Effective Shear Strain Ratio
• Complex Shear Modulus
o Frequency Independent
o Frequency Dependent
o Simplified
1 − 𝑀𝑀
𝑆𝑆𝑆𝑆𝑆𝑆 =
10
• Step Control
o Flexible
o Fixed
• Maximum Strain Increment
• Number of Sub-Increments
The accuracy of the time domain solution depends on the time step selected. There are two
options in choosing the time step (Hashash and Park, 2001).
The procedure is the same as that for the Fixed Step above, except the Flexible option is chosen.
Type the desired Maximum Strain Increment into the text box. The default and recommended
value is 0.005 (%).
Linear (time-domain) interpolation is the classical approach in which the change in acceleration
is simply divided into equal increments. This method has been shown to fundamentally alter the
motion by adding energy to the signal at frequencies above the Nyquest frequency of the original
signal. This can potentially add high frequency noise to the output signal.
Upon completion of analysis, the following output for each selected layer will be directly
exported to a text file “Results - motion.txt” in the working directory specified in step 1.
If multiple motions were selected for analysis, the output can be found in the user’s working
directory in a folder named “Batch Output”. Within this folder, there will be a folder
corresponding to each collection of batch analyses (ie. Batch0, Batch1, …etc). These folders
will contain the results from each motion.
If a single motion was selected for analysis, the results can be found in the user’s working
directory.
After analysis is complete, the user may immediately view the following output visually (Figure
16) by selecting the appropriate tab for the selected layer:
Output data for each layer analyzed is automatically exported to “Results – motion.txt” in the
user’s working directory.
DEEPSOIL also provides the option to export the analysis results to a Microsoft Excel file. This
is done by clicking the Export to Excel button on the results form. Note that this feature
requires Microsoft Excel be installed on the system.
To view the PGA profile click the command button labeled Summary Profiles in the lower left-
hand side of the window.
The Summary Profiles Window shows the PGA, maximum strain, and maximum shear stress
ratio for each layer. If an analysis with porewater pressure generation was conducted, this
window will also show the maximum excess porewater pressure ratio (excess/effective vertical)
for each layer. Note that the PGA is calculated at the top of each layer, while all other values are
calculated at the midpoint of each layer. To view the layers in the plots, check Show Layers.
To change the color of the plotted layer lines, click the color box and select a new color. When
you are finished, press Back to return to the output plots.
To view the displacement profile and animation click the command button labeled Column
Displacement Animation in the lower left-hand side of the window.
The Column Displacement Animation Window allows the user to adjust the speed of the
animation as well as to stop the animation and show the displacement at a given time. These
options can be adjusted using the scroll bars below the plot. Click Start to start the animation or
click Back to return to the output plots.
To view the convergence click the command button labeled Check Convergence in the lower
left-hand side of the window.
This option enables checking whether the solution has converged in an equivalent linear analysis.
Plots of maximum strain profiles for each iteration are displayed (Figure 19). To view the layers
in the plots, check Show Layers. To change the color of the plotted layer lines, click the color
box and select a new color. When you are finished, press Back to return to the output plots.
To review the input parameters, click the View menu and select Input Summary. The input
summary window (Figure 20) may be viewed any time after completing step 1. Note: tabs will
only appear after the corresponding parameters have been input. Use the Save button to create a
text file of the input parameters.
4 Soil Models
A variety of models are available for DEEPSOIL analyses. These models include: a) Equivalent
Linear, b) Hyperbolic (MR, MRD, DC), c) a Non-Masing Hyperbolic model (MRDF), and d)
Porewater Pressure Generation and Dissipation.
where Gmo = initial shear modulus, τmo = shear strength, γ = shear strain. Beta, s, and γ r are
model parameters. There is no coupling between the confining pressure and shear stress.
DEEPSOIL extends the model to allow coupling by making γ r confining pressure dependent as
follows (Hashash and Park, 2001):
b
σv'
γ r = REF . strain
REF . stress
where σv’ is the effective vertical stress. Ref. stress is the vertical effective stress at which γ r =
Ref. stress. This model is termed as the “pressure-dependent hyperbolic model.”
The pressure-dependent modified hyperbolic model is almost linear at small strains and results in
zero hysteretic damping at small strains. Small strain damping has to be added separately to
simulate actual soil behavior which exhibits damping even at very small strains (Hashash and
Park, 2001). The small strain damping is defined as
d
1
ξ = Small strain damping
σv '
d can be set to zero in case a pressure independent small strain damping is desired.
In summary, the parameters to be defined in addition to the layer properties are:
• Reference Strain
• Stress-strain curve parameter, Beta
• Stress-strain curve parameter, s
MR: Procedure to find the parameters that provide the best fit for the modulus
reduction curve with potentially significant mismatch of the damping curve.
MRD: Procedure to find the parameters that provide the best fit for both the modulus
reduction and damping curve
DC: Procedure to find the parameters that provide the best fit for the damping curve
with potentially significant mismatch of the backbone curve.
where DF(γm) is the reduction factor calculated as a function of γm,, the maximum shear strain
experienced by the soil at any given time, and ξMasing is the hysteretic damping calculated using
the Masing rules based on the modulus reduction curve. Two formulations for DF(γm) are
implemented in DEEPSOIL and are discussed in the following sections.
4.2.2.1 MRDF-UIUC
The MRDF Pressure-Dependent Hyperbolic (Phillips and Hashash, 2009) model available in
DEEPSOIL allows the user to introduce a reduction factor into the hyperbolic model. The
reduction factor has the form:
where 𝛾𝛾𝑚𝑚 is the maximum shear strain experienced at any given time, 𝐺𝐺(𝛾𝛾𝑚𝑚 ) is the shear
modulus at 𝛾𝛾𝑚𝑚 , and P1, P2, and P3 are fitting parameters.
By setting P1= 1 and P2= 0, the reduction factor is equal to 1 (regardless of the value of P3), and
the model is reduced to the Extended Masing criteria.
4.2.2.2 MRDF-Darendeli
The MRDF Pressure-Dependent Hyperbolic (Phillips and Hashash, 2009) model can also be
used with alternative formulations for the reduction factor. One alternative is the formulation
proposed by Darendeli, 2001. This formulation is an empirically-based modified hyperbolic
model to predict the nonlinear dynamic responses of different soil types. The developed model is
implemented as a reduction factor with the form:
where 𝛾𝛾𝑚𝑚 is the maximum shear strain experienced at any given time, 𝐺𝐺(𝛾𝛾𝑚𝑚 ) is the shear
modulus at 𝛾𝛾𝑚𝑚 , and P1 and P2 are fitting parameters.
By setting P1= 1 and P2= 0, the reduction factor is equal to 1, and the model is reduced to the
Extended Masing criteria.
𝐺𝐺0 ((𝛾𝛾 − 𝛾𝛾𝑟𝑟𝑟𝑟𝑟𝑟 )⁄2) 𝐺𝐺0 (𝛾𝛾 − 𝛾𝛾𝑟𝑟𝑟𝑟𝑟𝑟 ) 𝐺𝐺0 (𝛾𝛾 − 𝛾𝛾𝑟𝑟𝑟𝑟𝑟𝑟 )
𝜏𝜏 = 𝐷𝐷𝐷𝐷(𝛾𝛾𝑚𝑚 ) �2 − � + + 𝜏𝜏𝑟𝑟𝑟𝑟𝑟𝑟
(𝛾𝛾 − 𝛾𝛾𝑟𝑟𝑟𝑟𝑟𝑟 ) 𝑠𝑠 1 + 𝛽𝛽(𝛾𝛾𝑚𝑚 − 𝛾𝛾𝑟𝑟 )𝑠𝑠 1 + 𝛽𝛽(𝛾𝛾𝑚𝑚 − 𝛾𝛾𝑟𝑟 )𝑠𝑠
1 + 𝛽𝛽 � 2𝛾𝛾 �
𝑟𝑟
p ⋅ f ⋅ N c ⋅ F ⋅ (γ ct − γ tvp ) s
uN =
1 + f ⋅ N c ⋅ F ⋅ (γ ct − γ tvp ) s
which no significant pore water pressure is generated. γ tup is between 0.01% and 0.02%
for most of sands and is represented by the parameter “g” in DEEPSOIL
• γ ct is the most recent reversal strain.
• f is 1 or 2 depending on 1-D or 2-D directional generation of water pressure respectively
• p, s, and F are curve fitting parameters
The GMP model (Green et al. 2000) is an energy-based pore pressure generation model. The
excess pore pressure is calculated as follows:
𝑊𝑊𝑠𝑠
𝑟𝑟𝑢𝑢 = 𝛼𝛼�
𝑃𝑃𝑃𝑃𝑃𝑃
Where α is a scale factor, 𝑊𝑊𝑠𝑠 is the dissipated energy per unit volume of soil divided by the
initial effective confining pressure, and 𝑃𝑃𝑃𝑃𝑃𝑃 is the “pseudo energy capacity” (a calibration
parameter).
The dissipated energy, 𝑊𝑊𝑠𝑠 , is calculated as the area beneath the current stress-strain path and has
the form:
𝑛𝑛
1
𝑊𝑊𝑠𝑠 = ′ �(𝜏𝜏𝑖𝑖+1 + 𝜏𝜏𝑖𝑖 ) ∗ (𝛾𝛾𝑖𝑖+1 − 𝛾𝛾𝑖𝑖 )
2𝜎𝜎0
𝑖𝑖=1
The determination of the 𝑃𝑃𝑃𝑃𝑃𝑃 calibration parameter can be conducted either via graphical
procedure or by use of an empirical relationship. The graphical procedure is described in detail
by Green et al. (2000). However, this causes an interruption in analysis as it requires the
construction of the graphical procedure outside of site response analysis software.
Polito et al. (2008) derived an empirical relationship between 𝑃𝑃𝑃𝑃𝑃𝑃, relative density (𝐷𝐷𝑟𝑟 ), and
fines content (FC) from a large database of laboratory data on non-plastic silt-sand mixtures
ranging from clean sands to pure silts. The use of this empirical relationship allows the use of
the GMP model directly in nonlinear site response analysis software by removing the need to
find the value of 𝑃𝑃𝑃𝑃𝑃𝑃 through graphical procedures. The empirical relationship is defined as:
The pore water pressure dissipation model is based on Terzaghi 1-D consolidation theory:
∂u ∂ 2u
= Cv ( 2 )
∂t ∂z
where Cv is the consolidation coefficient.
Dissipation of the excess pore water pressure is assumed to occur in the vertical direction only.
Porewater pressure generation and dissipation occur simultaneously during ground shaking.
Example 1 B:
Example_1B_DS-TL0.dp
5.1.3 Results:
Example 1 A:
400
200
Spectral Acceleration (g)
0.01 0.1 1 10
Period (sec)
Example 1 B:
2
Spectral Acceleration (g)
01 0.1 1 10
Period (sec)
Example 2 B:
Example_2B_DS-TL0.dp
5.2.3 Results:
Example 2 A:
0.4
0.2
Spectral Acceleration (g)
0.01 0.1 1 10
Period (sec)
Example 2 B:
0.4
0.2
Spectral Acceleration (g)
0.01 0.1 1 10
Period (sec)
Example 2 B:
Example_3B_DS-TL0.dp
5.3.3 Results:
Example 3 A:
0.4
0.2
Spectral Acceleration (g)
0.01 0.1 1 10
Period (sec)
Example 3 B:
0.4
0.2
Spectral Acceleration (g)
0.01 0.1 1 10
Period (sec)
5.4.3 Results:
0.4
0.2
0
Spectral Acceleration (g)
0.01 0.1 1 10
Period (sec)
Example 5 B:
Example 5 C:
5.5.3 Results:
Input Motion
Nonlinear MR Fitting
Equivalent Linear MR Fitting
Nonlinear MRD Fitting
Equivalent Linear MRD Fitting
Nonlinear DC Fitting
Equivalent Linear DC Fitting
1.6 1
0.1
0.01
Fourier Amplitude (g-sec)
1.2 0.001
Spectral Acceleration (g)
0.0001
1E-005
0.8 1E-006
1E-007
1E-008
0.4 1E-009
1E-010
1E-011
0 1E-012
0.01 0.1 1 10 0.01 0.1 1 10 100
Period (sec) Frequency (Hz)
5.6.3 Results:
0.8
0.6
0.4
0.2
Spectral Acceleration (g)
0.01 0.1 1 10
Period (sec)
5.7.3 Results:
0.8
0.6
0.4
0.2
0
-0.2
Shear Strain (%)
-0.4
-0.6
-0.8
4
-1
-2
Stress Ratio (Shear/Eff.Vert.)
-3
-4
6 References
Chopra, Anil K. (1995) Dynamic of Structures, Theory and applications to Earthquake
Engineering Englewood Cliffs, New Jersey: Prentice Hall.
Clough, Ray W., and Joseph Penzien (1993) Dynamics of structures, New York: McGraw-Hill.
Duncan, James M., and Chin-Yung Chang (1970) "Nonlinear analysis of stress and strain in
soils," Journal of the Soil Mechanics and Foundations Division, Vol. 96, No. SM5, pp 1629-
1653.
Finn, W. D. L., Lee, K. L, and Martin, G. R. (1977) “An effective stress model for liquefaction.”
Journal of the Geotechnical Engineering Division, ASCE, Vol. 103, No. GT6, 517-533.
Green, R.A., Mitchell, J.K. and Polito, C.P. (2000). "An Energy-Based Pore Pressure Generation
Model for Cohesionless Soils", Proceedings: John Booker Memorial Symposium, Melbourne,
Australia, November 16-17, 2000.
Hardin, B. O. and Drnevich, V. P. (1972) “Shear modulus and damping in soils: Design
equations and curves.” Journal of the Soil Mechanics and Foundations Division, ASCE, Vol. 98,
No. SM7, 667-692.
Hashash, Y. M. A., and D. Park (2002) "Viscous damping formulation and high frequency
motion propagation in nonlinear site response analysis," Soil Dynamics and Earthquake
Engineering, Vol. 22, No. 7, pp. 611-624.
Hashash, Youssef M. A., and Duhee Park (2001) "Non-linear one-dimensional seismic ground
motion propagation in the Mississippi embayment," Engineering Geology, Vol. 62, No. 1-3, pp
185-206.
Hashash, Y.M.A., Phillips, C. and Groholski, D. (2010). "Recent advances in non-linear site
response analysis", Fifth International Conference on Recent Advances in Geotechnical
Earthquake Engineering and Soil Dynamics, Paper no. OSP 4.
Hudson, M., Idriss, I.M., and Beikae, M. 1994. (1994) "QUAD4M - A computer program to
evaluate the seismic response of soil structures using finite element procedures and incorporating
a compliant base." Davis, CA: Center for Geotechnical Modeling, Department of Civil and
Environmental Engineering, University of California, Davis, CA.
Idriss, I. M. and Seed, H. B. (1968) “Seismic response of horizontal soil layers.” Journal of the
Soil Mechanics and Foundations Division, ASCE, Vol. 94, No. SM4, pp 1003-1031.
Idriss, I. M. and Sun, J. I. (1992) “User’s Manual for SHAKE91, A Computer Program for
Conducting Equivalent Linear Seismic Response Analyses of Horizontally Layered Soil
Deposits”
Kramer, Steven Lawrence (1996) Geotechnical earthquake engineering, Upper Saddle River,
N.J.: Prentice Hall.
Lee, M. K. W. and Finn, W. D. L (1975) “DESRA-1, Program for the dynamic effective stress
response analysis of soil deposits including liquefaction evaluation.” Soil Mechanics Series No.
36, Department of Civil Engineering, University of British Columbia, Vancouver, Canada.
Lee, M. K. W. and Finn, W. D. L (1978) “DESRA-2, Dynamic effective stress response analysis
of soil deposits with energy transmitting boundary including assessment of liquefaction potential.”
Soil Mechanics Series No. 36, Department of Civil Engineering, University of British Columbia,
Vancouver, Canada.
Matasovic, Neven, and M. Vucetic (1993) "Cyclic Characterization of Liquefiable Sands," ASCE
Journal of Geotechnical and Geoenvironmental Engineering, Vol. 119, No. 11, pp 1805-1822.
Menq, Farn-Yuh (2003). Dynamic Properties of Sandy and Gravelly Soils, Department of Civil,
Architectural and Environmental Engineering, The University of Texas, Austin, Texas.
Newmark, Nathan M. (1959) "A method of computation for structural dynamics," Journal of the
Engineering Mechanics Division, Vol. EM 3, pp 67-94.
Park, D., and Y. M. A. Hashash (2004) "Soil damping formulation in nonlinear time domain site
response analysis," Journal of Earthquake Engineering, Vol. 8, No. 2, pp 249-274.
Phillips, Camilo, and Youssef M. A. Hashash (2008) "A new simplified constitutive model to
simultaneously match modulus reduction and damping soil curves for nonlinear site response
analysis," Geotechnical Earthquake Engineering & Soil Dynamics IV (GEESD IV). Sacramento,
California.
Phillips, C. and Hashash, Y. (2009) “Damping formulation for non-linear 1D site response
analyses” Soil Dynamics and Earthquake Engineering, v. 29, pp 1143–1158.
Phillips, C., Kottke, A.R., Hashash, Y.M.A., and Rathje, E.M. (2012) “Significance of ground
motion timestep in one dimensional site response analysis” Soil Dynamics and Earthquake
Engineering, v. 43, pp 202–217.
Richart, F.E., Hall, J.R., and Woods, R.D. (1970). “Vibrations of Soils and Foundations,”
Prentice Hall, Englewood Cliffs, New Jersey, 401 pp.
Roblee, Cliff and Chiou, Brian. (2004) “A proposed geoindex model for design selection of non-
linear properties for site response analyses.” Caltrans Geo-Research Group. Sacramento, CA.
Sanderson, Conrad. (2010) “Armadillo: An Open Source C++ Linear Algebra Library for Fast
Prototyping and Computationally Intensive Experiments.” Technical Report, NICTA.
Schnabel, P. B., Lysmer, J. and Seed, H. B. (1972) “SHAKE: A computer program for
earthquake response analysis of horizontally layered sites.” Report No. EERC 72-12,
Earthquake Engineering Research Center, University of California, Berkeley, California.
Seed, H. B. and Idriss, I. M. (1970) “Soil moduli and damping factors for dynamic response
analyses.” Report No. EERC 70-10, Earthquake Engineering Research Center, University of
California, Berkeley, California, 40p.
Distance
Record Date of to Fault USGS
Motion Name Magnitude PGA (g)
Number Event Rupture Site Class
(km)
*Hypocentral distance
** Geomatrix Site Class
The first example considers a simple linear frequency domain analysis. The profile for Example
1 (“Ex1_Lin_Freq_Undamped_Rigid.dp”) is shown below.
The profile consists of a 70-ft thick soil column overlying rigid bedrock. The soil layer is
assumed to be undamped (zero damping) and linear elastic.
STEP 1/6
For Step 1/6, first choose the method of analysis by selecting Frequency Domain - “Linear
Analysis.”
For this example, the number of layers will be 1. Check that the value in the “# of Layers” input
box is 1.
Now we must choose whether to define the stiffness of the layer in shear wave velocity or shear
modulus. Select “Wave Velocity.”
Finally, the analysis stress type will be “Total Stress Analysis.” Check that “Total Stress
Analysis” is selected and press the Next button.
STEP 2/6
In Step 2/6, the user must define the soil column and soil properties. The figure below shows the
window that displays the soil properties.
IMPLIED STRENGTH
This step uses the material properties specified in step 2a to calculate the implied strength of the
profile. Because this example in an idealized soil column, the values will seem very large. In a
real analysis, the soil properties should be modified to reflect realistic strengths. For now,
simply press Next to continue to step 2b.
STEP 2b/6
In Step 2b/6, the properties of the bedrock are specified. In this case, the analysis considers rigid
bedrock.
STEP 3/6
In Step 3/6, the options for the Frequency Domain analysis must be specified.
First select the Fourier Transform Type you wish to use for analysis. There are two options
which are the Fast Fourier Transform (FFT) and Discrete Fourier Transform (DFT). It is
generally recommended that FFT be used for analysis. (Note: FFT and DFT will give the same
results, but FFT is faster)
You’ll notice that the “Effective Shear Strain” is disabled. This is because the effective shear
strain is irrelevant for a linear analysis. Similarly, the “Number of Iterations” for a linear
analysis is also irrelevant, so the default value of “1” does not need to be changed.
The final selection in this step is selecting the complex shear modulus. There are three options:
1. Frequency Independent
2. Frequency Dependent
3. Simplified (Kramer, 1996)
It is recommended that the “Frequency Independent” complex shear modulus be used for all
analyses. The “Simplified” modulus is based on the “Frequency Independent” modulus, but
modified to result in a simpler form (Kramer, 1996). The “Frequency Dependent” modulus is
equivalent to the modulus used in SHAKE91.
Select the “Frequency Independent” modulus and press the Next button.
STEP 4/6
Step 4/6 involves the selection of a) input motion and b) layers for output.
A motion library is provided which will automatically plot the selected motion for the user’s
inspection. Select the input motion “Kobe.txt” from the motion library.
In the frequency domain analysis, the number of points for the FFT must be defined. The number
of points is a power of 2. DEEPSOIL will calculate the minimum number of points needed for
the input motion and automatically sets the number of points to be used in the FFT to this
minimum value. Note that the number of points for FFT should not be smaller than the
minimum value recommended by DEEPSOIL.
After selecting the input motion and associated parameters, select the layer(s) for output (shown
in the left column). Layer 1 is selected by default.
Finally, select the calculation method to use for the response spectra, and enter a damping ratio
for the output response spectrum (shown in the lower left corner). The recommended method is
“Frequency Domain” and the recommended damping ratio is 5%.
In case the user wishes to define new motions, the format of the ground motion file should be as
follows:
The first number is the total number of data points. The second number is the time step. The
actual time history should be written in two columns, the first column is the time and the second
column is the acceleration. The time should be in units of seconds, and the acceleration should be
in units of g.
STEP 6/6
Upon completion of analysis, the user will be presented with the output window. The output
window displays acceleration, strain, and stress time histories, in addition to stress vs. strain
curves, Fourier amplitude spectrum, Fourier amplification ratio, and response spectra.
Compare your results with the figures shown below. The results should be exactly the same
(note the scales).
The output data has been automatically exported to “Results - Motion.txt” in the user-specified
working directory. To view the output text file, simply click the “Show analysis results in folder
view…” link located above the Close button. This will open the user-defined working directory,
which should contain the output text file.
Note that resonance occurs at natural frequencies and therefore results in significant
amplification of the motion at such frequencies.
STEP 1/6
All options are the same as in Example 1. Press the Next button to proceed to the soil profile
window.
STEP 2/6
Enter “80” for the thickness of the layer in the soil properties spreadsheet. All other values are
the same as given in Example 1. Press the Next button to continue.
STEP 2b/6
In this step, we will define the elastic properties of the bedrock. Select the “Elastic Half-Space”
option to define the elastic bedrock properties. Enter the input for the Shear Velocity, Unit
Weight, and Damping Ratio as 5000 ft/sec, 160 pcf, and 2% respectively. You can also save the
bedrock properties by giving the bedrock a name and then clicking the Save Bedrock. Press the
Next button to proceed to Step 3/6.
For the remaining steps, all options should be selected to be the same as in Example 1 (Input
Motion”Kobe.txt”; Frequency Independent Complex Shear Modulus; FFT).
After you have checked that all options are the same as in Example 1, click the Analyze button
to begin the analysis.
Check your analysis results with the figures shown on the following page. The first figure shows
the calculated surface response spectrum. The elastic bedrock absorbs a significant amount of
energy compared to the rigid bedrock and results in lower resonance.
Examples 1 and 2 assume that the soil layer has zero damping. This assumption is unrealistic
because soils are known to exhibit damping even at very small strains. Example 3
(“Ex3_Lin_Freq_Damped_Elastic.dp”) is similar to Example 2; the only difference being that
the soil is damped instead of undamped. As such, the steps of the analysis are the same as those
outlined in Example 2 except where noted below.
STEP 1/6
All options for Step 1/6 are exactly the same as those in Example 2.
STEP 2/6
Damping of 5% is imposed on the soil layer. Enter “5” into the “Damping Ratio” column of the
soil properties spreadsheet. Press the Next button to proceed to Step 2b/6.
Select all other options to be the same as Example 2 (Input MotionKobe.txt; Frequency
Independent Complex Shear Modulus; FFT). After you have checked that all options are the
same as in Example 2, click the Analyze button to begin the analysis.
The calculated surface response spectrum is shown in the figure on the following page. Note
how the damping imposed on the soil results in lower resonance.
STEP 1/6
The input for Step 1/6 is similar to Example 3, with the following exceptions:
For “Analysis Type,” select the Frequency Domain – “Equivalent Linear” analysis. This will
enable the “Equivalent Linear” options.
For an equivalent linear analysis, the G/GMAX and damping ratio curves can be defined using
either a) Discrete Points or b) the Modified Hyperbolic Model.
If discrete points are selected, the G/GMAX and damping ratio will be defined in discrete points at
various strain levels. It is also possible to define the G/GMAX and damping curve using the
modified hyperbolic model. In that case, the user needs to define the nonlinear parameters for the
soil model. DEEPSOIL will automatically develop corresponding G/GMAX and damping ratio
curves.
For this example, select “Discrete Points” and then press the Next button.
STEP 2/6
The user can go directly to the spreadsheet, the graphical soil column, or use the “Material
Properties” button to define the soil curves. From the spreadsheet, left-click any cell of the layer
for which you want to define the soil curve to select that layer, and then press the Material
Properties button. The user can also double-click any cell in the spreadsheet to open the
Material Properties window for that layer. Similarly, double-clicking a layer in the graphical soil
column will open the Material Properties window for that layer.
The user can define the G/GMAX and damping properties by first defining the number of data
points. Note that the number of data points should be identical for G/GMAX and damping. The
strain and damping values should be entered as a percent [%].
To save the data points, type a name to identify the properties and press Save Material. Once
saved, the newly saved file will appear in the “Use Saved Material Properties” listbox.
The user can also use saved material properties by selecting the appropriate file from the listbox
and pressing the Use Saved Material button. We will use this method in this example. Select
the saved material named “S&I_Mean.dsm” and click the Use Saved Material button. The
discrete point data for this material should now be loaded in the spreadsheet as shown below.
To compare the selected material to a material from the material library, the user must define a)
the Material Type, and b) the Target Curve.
Click on the Material Type drop-down menu and select “Sand”. Two new items will appear:
Basic Parameters and Target Curve. The Basic Parameters for this case simply displays the
vertical stress at the midpoint of the layer. Now we must define the Target Curve.
Click on the Target Curve drop-down menu. A list of various models for sand will appear.
Select the “Seed & Idriss, 1991 (Mean Limit)” item. The model soil curves will be plotted in
pink for your reference. In addition, a new item appears labeled: “Data Points to Fit.” These are
the points that define the model curves. To use this model data, click the “Use Material Data”
button. The discrete points of your soil model will be updated to match these points. Click
“Calculate Curves” to verify that the models are the same.
Once you are satisfied with your soil curves, press the Apply button to apply the properties and
return to the profile spreadsheet.
When you have finished checking the data, press the Next button to proceed.
STEP 2b/6
The entries for this step are the same as those specified in Example 3.
STEP 3/6
Equivalent linear analyses require a number of iterations to obtain more accurate results. The
recommended number of iterations is 15. For the sake of accuracy, you should not choose less
than 10 iterations. For this example, choose (at least) 10 iterations.
The next step is selecting the effective shear strain ratio. The equivalent linear analysis selects
shear modulus and damping ratio at a representative shear strain at an effective strain as a ratio
of maximum shear strain. Enter an effective shear strain ratio of 0.65.
Select the Frequency Independent Complex Shear Modulus for use in this analysis.
Finally, press the Next button to proceed to the input motion and output layer(s) selection
window (Step 4/6).
STEP 4/6
Similar to the previous examples, select “Kobe.txt” as the input motion and select the desired
layers for output. Layer 1 is automatically selected by default. Press the Analyze button to
begin the analysis.
STEP 6/6
The figures below show the computed response spectrum at the surface. Check that your results
match those presented in the figures.
STEP 1/6
Press the Open Existing Profile button and browse for Example 4. It should be located in the
“Examples” directory. Once you find the appropriate directory, open Example 4
(“Ex4_EQL_Single_Layer.dp”).
STEP 2/6
As you can see, all of the information for Layer 1 corresponds to Example 4. We will now
modify this data and add two additional layers to the profile. First, change the Thickness and
Shear Wave Velocity of Layer 1 to 10 ft and 1000 ft/sec, respectively.
There are two methods of adding a layer to the profile. We will use the first method to add the
first layer, and the second method to add the second layer.
To add a layer to the profile by the first method, first select Layer 1 by left-clicking any of the
cells in that row. Now, right-click to bring up the soil properties pop-up menu and select “Add
Layer” from the list of commands. A new “Add Layer” window will appear.
In the “Add Layer” window, select the “After Layer” option and select Layer 1 from the drop-
down list. After pressing Add, the new soil layer should be visible in the spreadsheet.
Enter the thickness (30 ft), unit weight (125 pcf), and shear wave velocity (1500 ft/sec) of the
soil layer. Also apply the “Seed & Idriss, 1991 (Mean Limit)” curves for the layer as was done
in Example 4.
To add the third layer, left-click one of the cells in the spreadsheet. Now click the Add Layer
button in the Soil Profile group located in the middle of the form. Again select the “After Layer”
option and select 2 using the drop-down box and press the Add button. Repeat the same process
outlined above, but using a thickness of 40 ft and a shear wave velocity of 2000 ft/sec. Be sure
that you check your input in the spreadsheet to confirm that it matches the one shown below.
For Step 2b/6 and Step 3/6, keep all other options the same as Example 4.
STEP 4/6
Keep all other selected options the same as in Example 4, including the input motion
(“Kobe.txt”). If you like, you may select to analyze Layers 2 and 3 (Layer 1 is selected by
default) by checking (double-clicking) each layer’s corresponding checkbox located to the left of
the input motion plot. Once you have checked your input and specified which layers are to be
analyzed, press the Analyze button to run the analysis.
STEP 6/6
The figure below shows the computed surface acceleration. Check that your results match with
those shown.
DEEPSOIL also allows checking the convergence of the equivalent linear analysis. You may do
so by pressing the Check Convergence button located near the lower left corner of the form.
In a non-linear analysis, the thickness of each layer has to be changed. This is because the
thickness controls the maximum frequency that can be propagated by the layer. The greater is
the thickness of the layer, the lower the maximum frequency that can be propagated by the same
layer.
The equation that correlates the maximum frequency with soil thickness is as follows:
h = Vs/4fmax
Where h = thickness of the soil layer, Vs=shear wave velocity of the layer, and fmax is the
maximum frequency that can be propagated.
Simple calculations reveal that h for the layers should be 10 ft, 15 ft, and 20 ft for layers 1, 2, and
3 respectively. The first layer does not need to be changed, whereas the subsequent layers need
to be subdivided into 2 thinner layers.
Now let's actually develop the input file for this example.
STEP 1/6
Change the “Analysis Type” from “Equivalent Linear” to “Non-Linear.” Then select “Pressure-
Dependent Hyperbolic Model : Masing Criteria” in the “Nonlinear” section. Press the Next
button to proceed to the soil properties input form.
STEP 2/6
Note that the basic properties of the layers (Thickness, unit weight, and shear velocity) are
preserved.
Subdivide Layers 2 and 3 into 2 thinner layers with each having a thickness equal to half of the
original layer by adding layers as described in the previous example.
For each layer, bring up the soil properties window using the “Soil Properties” button or by
double-clicking the layer.
The default non-linear parameters are given as “S&I_M_NL.dsm.” Find the file in the “Saved
Materials” list box and press Use Saved Material to apply the material data to the layer.
Find the “Seed & Idriss, 1991 (Mean Limit)” curves in the Material Library as was done in
previous examples. Now, press Calculate Curves to display the soil curves. Compare the
calculated curves to the Seed and Idriss mean cohesionless curves. The Seed and Idriss curves,
which are the reference curves, will be shown in pink.
To match the Seed and Idriss curves, the material constants need to be changed. The soil model
incorporated in DEEPSOIL is the extended modified hyperbolic model:
b
σv' Damping ratio
γ r = REF . strain ξ=
REF . stress (σ v ')d
The parameters that control the shape of the backbone curve are β (beta), s, and γr.
The curve can be made confining pressure dependent by selecting the reference stress and the
“b”-parameter. Select b = 0 to make the curve pressure independent. Note that γr = reference
effective strain for b = 0 or σv’ = reference stress.
The small strain damping properties can also be made pressure dependent by introducing the “d”
parameter. The “d” parameter in the equation is the small strain damping in the user interface.
Select d = 0 to make the curve pressure independent.
Try various combinations to get a good match with the Seed and Idriss reference curves.
Once a satisfactory match is obtained, save the material in the material library. Then assign the
selected parameters for all other layers.
For the purposes of this example, use the “S&I_M_NL.dsm” saved material for all layers.
After all of the input parameters have been specified, the spreadsheet should look like this:
For nonlinear analyses, DEEPSOIL will automatically check the maximum frequency of each
layer. The Maximum Frequency vs. Depth will be plotted with a table of corresponding values
given on the right. This check is to ensure that the maximum cut-off frequency is always greater
than or equal to 25 Hz.
After checking the results, press the Next button to continue to Step 2b/6 of the analysis.
STEP 2b/6
STEP 3/6
The third stage of the analysis is the analysis control stage. In a time domain analysis, the user
must specify a step control scheme. Choose either a “Flexible” (default) or “Fixed” sub-
incrementation scheme. The “Flexible” sub-incrementation scheme subdivides a time interval
into small steps if the calculated strain increment is higher than the user-defined maximum strain
increment. The “Fixed” scheme sub-divides all time intervals into user-defined sub-increments.
For the purpose of this tutorial, select the “Flexible” sub-incrementation scheme and use the
default value of 0.005. Press the Next button to continue.
STEP 4/6
This stage of analysis requires selection of the input ground motion and layers to be analyzed for
output. As in previous examples, select “Kobe.txt” as the input motion. You may select the
additional layers to be analyzed as well. Layer 1 is selected by default.
STEP 5/6
The fifth stage of analysis requires selection of the appropriate Rayleigh damping coefficients.
The purpose of this stage of analysis is to reduce frequency dependent damping introduced due
to the viscous damping formulation. This stage allows selection of optimum coefficients by
comparing the linear time domain solution with the linear frequency domain solution (Note: the
linear frequency domain solution uses frequency independent damping).
First, click the Graph Lin. Freq. Domain button. DEEPSOIL will display the transfer function
values and response spectrum plots corresponding to the linear frequency domain solution.
The default selections using 2 modes/frequencies are the 1st and 8th modes. Click the Check
with Lin. Time Domain button to view the linear time domain solution. Using the default
modes, a good match is obtained with the linear frequency domain solution.
Finally, select the “Frequency Independent” option for the analysis. We have now optimized this
analysis. Press the Analyze button to continue.
STEP 6/6
The figure shown below is the calculated surface response spectrum for Layer 1. Check that
your results match those shown in this tutorial.
In a non-linear analysis, it is also possible to animate the column displacement time histories.
You can do so by clicking the Column Displacement Animation button. The Column
Displacement Animation Window allows the user to adjust the speed of the animation as well as
to stop the animation and show the displacement at a given time. These options can be adjusted
using the scroll bars below the plot. Click Start to start the animation or click Close to return to
the output plots.
The PGA profile can also be displayed by clicking the PGA Profile button. The PGA Profile
Window shows the PGA for each layer. Note that the PGA is calculated at the top of each layer,
not the midpoint. To view the layers in the PGA plot, check “Show Layers.” To change the
color of the plotted layer lines, click the color box and select a new color. When you are finished,
press Close to return to the output plots.
STEP 1/6
Change the “Analysis Type” from “Total Stress Analysis” to “Effective Stress Analysis.” This
will enable the option to “Include PWP Dissipation.” Check the checkbox next to “Include PWP
Dissipation” to allow for both pore water pressure generation and dissipation in the analysis.
When the “Include PWP Dissipation” option is selected, a new item appears labeled: “Boundary
Conditions for Bottom of Profile.” These options are used to specify whether the bottom of the
profile is a permeable or impermeable boundary. For the purposes of this example, select the
“Permeable” option.
Press the Next button to continue to the soil properties input form.
STEP 2/6
Using the horizontal scroll bar, we see that there are new parameters which must be defined for
the pore water pressure generation and dissipation model. If the spreadsheet is too large for your
window, press Expand Soil Properties Spreadsheet to open the spreadsheet in full-screen
mode.
The first parameter that needs to be defined for each layer is the “PWP Model.” The models that
can be used in analysis are Sand (1), Clay (2), or GMP (3) which is another model that can be
used for sands. Each layer may use a different PWP Model. For the purpose of this example, set
each layer to use the Sand Model by entering 1 into each layer’s corresponding cell.
The next parameter is “f/s/Dr (%).” The notation for the parameters including a “/” is that the
first listed parameter is for the Sand Model, the second listed parameter is for the Clay Model,
and the third listed parameter is for the GMP Model. So, in the case of “f/s/Dr (%),” “f” must be
defined if the Sand Model is selected, “s” must be defined if the Clay Model is selected, or “Dr
(%)” must be defined if the GMP Model is selected. (Note that the parameters are defined in
Section 4.2)
Dashed parameters such as “-/g/-” indicate that a certain model has no input for this column. In
the case of “-/g/-“, the Sand and GMP Models have no input for this column. You may leave the
cell blank for the Sand and GMP Models.
f/s/f f = 1
p/r/Dr (%) p = 1
F/A/FC(%) F = 0.73
s/B/- s = 1
g/C/- g = 0.02
v/D/v v = 3.8
Cv = 0.1
The PWP section of the spreadsheet should look like the following figure.
After checking your input, press the Next button to continue to the third stage of analysis.
The remaining steps of the analysis are exactly the same as in Example 6. Check that your input
for Steps 3/6 – 5/6 are the same as in Example 6. In Step 4/6, be sure to select the “Kobe.txt”
input motion for analysis.
STEP 6/6
The figure shown on the following page is the calculated surface response spectrum for Layer 1.
Check that your results match those shown in this tutorial.
Now let’s take a look to see if any pore water pressure was generated in Layer 1 due to the input
motion. You can do this by selecting the “PWP vs Time” tab for a quick visualization. For the
purposes of this example, let’s examine the exported output data. Use Windows Explorer to
navigate to the folder you specified as your working directory when you started DEEPSOIL or
press the “Show results in folder view…” link shown above the close button. If you kept the
default directory suggested by DEEPSOIL, then navigate to the “Working” folder of the
DEEPSOIL program path. The current working directory can also be found using the input
summary. To view the input summary, click on the “View” menu and select “Input Summary.”
The working directory will be listed on the “Analysis Selection” tab of the form.
Open “Results – Kobe.txt.” If you have completed other analyses with the Kobe motion, the
results file will be “Results – Kobe#.txt,” where # is simply an index referring to the most recent
analysis.
“Results – Kobe.txt” contains all of the output data produced by DEEPSOIL. As can be seen
from the figure above, the last column of data contains the pore water pressure in the layer at a
given time.
Scroll down to the very bottom of “Results – Kobe.txt.” Here you will find data regarding the
PGA, Maximum Strain, Maximum Stress Ratio, and Maximum Pore Water Pressure Ratio
Profiles.
As you can see from the results, almost no pore water pressure was generated in Layer 1, and the
largest pressures were generated in Layer 5.
Using “Results – Kobe.txt,” we can determine the generation of pore water pressures with time,
and also quickly identify which layer experiences the maximum generation of pore water
pressure.
If you would prefer to view these results in the form of a Microsoft Excel file, simply click the
Export Output to Excel button on the results form. This will create an Excel file that contains
all of the data contained in “Results – Kobe.txt” in an easy to read and manipulate spreadsheet.
It will also contain plots of the profile data. The user will be prompted to provide a file name
and location for the Excel file. The output file will be in .XLSX format, which requires Excel
2007 or greater to open.
f/s/f f = 2
F/A/FC(%) FC = 15
v/D/v v = 3.8
Cv = 0.1
The PWP section of the spreadsheet should look like the following figure.
After checking your input, press the “Next” button to continue to the third stage of analysis.
The remaining steps of the analysis are exactly the same as in Example 6. Check that your input
for Steps 3/6 – 5/6 are the same as in Example 7. In Step 4/6, be sure to select the “Kobe.txt”
input motion for analysis.
The response spectra and excess pore pressure plots are shown below for your comparison.