Home

Safe Oasys GEO Suite for Windows

1. 68 5 3 68 Ene OAS a 69 Pressure 69 45 4 RestralMlS ace 70 1 5 5 eeepc 72 A6 RUN PCIE 73 4 6 1 Analysis Options Gud PER OS buf 74 4 6 2 Time Dependent Consolidation 76 Specialis ik 01 9 RN 78 AiG S Wel S d 79 Data Heailred Glo Dall ob de ee o ie Co te ez eae ca QU eee ee ee 79 Data leguireg H rn SpeolflO 4 a 80 81 4 7 1 General Material cox eda in aeu pe eoe ei 82 COUN SIESS cnc atia oos rad oit bts ep tna va vetare 84 Relaxation is EM 85 LNECLABL aII bi gee 86 1 6 ial MOGONS eee m 87 181 Stress StralDdla eerie aa ca 8
2. _ 11 32 20 03 09 11 36 20 03 09 sert Tunne 11 48 20 03 09 288 1 Long Term for Tunnel 11 50 20 03 09 5 4 4 11 52 20 03 08 Bg 5 J diEscaeat Stage 1 11 55 20 03 08 0 Insert Prop and Excayate Stage 2 11 20 03 09 8B a8 7 D End of Construction 15 11 20 03 09 5 3 8 ee Drained Following Construction 00 23 20 03 09 oy S wol 3 Rela Wall for Long Term 00 26 20 03 09 gt Events Sequences Page Events A 1 The Events tab at base of the table lists the event number and then the run for that event The previous run and increment from which it continues and for consolidation calculations the end time in days Once a run has been analysed the date and time number of increments iterations are also listed For analysis see Analysis and Data Checking Eagle sfd Events an r1 a mi l Ta L L l 4 Events Seque Page Sequences A 1 The Sequences tab lists horizontally the available alternative runs for each event and vertically the order that the runs follow to create the calculation sequences A B C etc Any number of Oasys Ltd 2009 Input Data 102 sequences can be added as Safe adds a new column as each one is created Frew Safe Link Overview Frew Flexible
3. PLSTAN 0 D 20 20 0 0 60 User Specified 13 0 0 80 Specified _ Elastic Mohr Coulornb PES Tan PLSTRN Draned FL STRN PLSTRN Drai L STRN lastic Mohr Coulornb PL STAN Elastic Mohr Coulomb PL STAN Drained PL STRN ies PL STRN EXE Oasys Ltd 2009 113 Safe Oasys GEO Suite for Windows 5 3 7 Groundwater Frew and Safe use different approaches for modeling pore pressure distribution Following are a couple of important differences Piecewise linear interpolation of pore Radius of influence approach pressure All soil zone materials share the same pore Each material has its own pore pressure pressure distribution data distribution In Safe each data point is characterized by pore pressure value its gradient and a radius of influence The net pore pressure at a given point is the weighted average of the pore pressures calculated at the given point using the existing pore pressure data points The weights for points which lie within a square defined by this radius of influence the weights are typically much higher compared to the weights for the points located outside the square In Frew we can have different pore pressure gradients above and below a particular pore pressure data point However this situation is not possible in Safe for a given data point as only a single pore
4. W man Safet std Tests ests FREY man sfd 4 C Program Filesi Sare3 sFd Exit Combinations QUEUES window Help A standard File Open dialog opens up which prompts the user to specify the name of the Gwa file from which to import the data After the user specifies the file the data is transferred from the Gwa file into the current Safe file Oasys Ltd 2009 Frew Safe Link 104 5 3 5 3 1 Import GWA Look ir FHEw man a lt 2 E HEL Recent 5 FEEwWman gwa Documents mrin Desktop Documents vikram gadicher aon File name FR Ewrman qwa wt Files of type Text Tab delimited qwa tet ww Data Conversion As mentioned before this feature enables the user to create the equivalent Safe model from Frew model There are some approximations and limitations in this data conversion from Frew to Safe The following topics detail the assumptions limitations and approximations involved in this process Stages Runs Geometry Restraints Surcharges Struts Materials Groundwater Unsupported Features Stages Runs Stages in Frew are roughly equivalent to Runs in Safe All stages in a Frew model translate to a sequence of runs following each other in Safe without any branching Frew Stages Oasys Ltd 2009 105 Safe Oasys GEO Suite for
5. Figure a Correction for Tensile Stress b If the Mohr Coulomb criterion is violated the shear stress is reduced and if dilation is specified the mean normal stress is increased as indicated in section on Methods of Analysis Material Types Figure b Reduction of Shear Stress Oasys Ltd 2009 137 Safe Oasys GEO Suite for Windows Figure Increase of Mean Normal Stress These corrections to the stresses lead to non equilibrium with the surrounding stresses The errors are applied as nodal equivalent forces in the next iteration of the initial stress process To accelerate convergence over relaxation factors are applied to the components of stress error in co ordinates Notes e n the process of correcting the stresses principal directions are unchanged e Dilation can only be used if the elastic properties are isotropic 7 4 3 1 Dilation in Elastic Mohr Coulomb Materials Plastic strains 6Vp and are derived such that with no change of total strain stresses will return to the Mohr Coulomb envelope The additional elastic strains are dYp The changes of mean normal stress 5 and shear stress are therefore OS KVe S 01 02 2 t 02 2 For plane strain As Ave E 2 1 v 2v2 The dilation rule is Oasys Ltd 2009 Detailed Processes and Models in SAFE 138 7 5
6. 3909 738 50 65 68 E a 6753 214 79 30 10 35 1G 352 aes 77 96 11 018 333 76 65 11 68 3 11 676 974 75 35 1 12 324 TOTAL FLOW ACROSS THIS L s I7 BJSE GO6 m75 m m 4g woh pP c m m gs r e ua Test Switches Test switches can be set in Safe to use superseded methods for comparison with old analyses to use soil models still in development for example strain softening and for more detailed output of various intermediate results to the log file Many of the test switches should only be used under supervision Please contact Oasys Ltd for a full list of switches available General Loading of Axisymmetric Structures Although only two dimensional elements are available in Safe full three dimensional loading of axisymmetric structures can be modelled using Fourier coefficients Wilson 1965 The following restrictions apply The structure or soil mass must be axisymmetric in both geometry and material properties The material properties must all be linear elastic so that super position can be used The user must specify coefficients of load and restraint for a restricted Fourier series of the Oasys Ltd 2009 Detailed Processes and Models in SAFE 168 angular co ordinate A separate computation is carried out for each frequency in the series The results for stress strain and displacement are also Fourier coefficients which mus
7. Oasys Ltd 2009 49 Safe Oasys GEO Suite for Windows The Post Processing options allow for x and or y restraints to be applied to all nodes generated along the line To delete lines from the mesh 1 During drawing selecting Edit Undo will delete lines in the reverse order of drawing or 2 Select the Pick Lines tool Ady click near the middle of the line then press Delete or 3 Open the Lines table highlight the line to be deleted and press the Delete button Crossing lines lines are drawn crossing each other these will generally be split by the program during drawing The option to Split crossing lines is available on the right click menu in case data has been imported from another source or complex sets of lines and curves have been created 4 4 1 3 2 Creating Areas The whole region to be meshed must be defined as Areas prior to generating the mesh as any lines not within areas will cause the mesh generator to return an error message Areas can be automatically or manually created a To automatically create areas select the Generate Areas button This will only be available if no areas already exist Complex areas may not be generated by this method To quickly create areas in any remaining xi gaps click the Find Polygon button then click within the required area To make areas manually 1 Select the Pick Lines tool 2 Use the left mouse button to select the lines which make up t
8. Frew Manual fwd Struts A Node Prestress 2 arm number kN Am t _ Im Safe Nodal al Springs Spring constants kN m per m Oasys Ltd 2009 109 Safe Oasys GEO Suite for Windows 5 3 6 Safel Node Loads Materials In Frew material data is specified only for Soil strata In the Safe model the excavations are represented using void material the wall is modeled using a linear elastic material and the soil is modeled using Mohr Coulomb materials Safe void material data Safel Water or air properties Seles EI IE a E2 NUI2 612 Ta cM kPa Safel Linear elastic material properties ae o9 Te 5 T E T E T 5 T D cum cL UT 1e 007 0 3 The Frew material model has all the data required for Safe Mohr Coulomb model except for the angle of friction 2 and Poisson s ratio v Frew material data zz Frew Manual fwd Materials Rector _ co Lol eis S ee Eme A SECs EE EO Unit Earth Cw c0 Drai Material no fo Description weight pressure Ic m ained effective stress KN Zm kN coefficient dio F ratio kN Zm undrained parameters Calculated amm Dead 50000 200 0 60 User Specifi
9. Position Ho m Defaults 1 2 3 4 H B 9 1 4 0000 2 4000 6000 14 0000 24 0000 1 7 0000 1 2 2000 6 1000 2 4000 4 0000 Quick mesh settings Generate Areas and elements Global element size m The gridlines will be drawn on the Graphical Input view Clicking the Simple mesh icon HE in the Graphical toolbar gives the following dialog The Lines only method will create lines along all gridlines with the entered Global element size Oasys Ltd 2009 44 Input Data This is useful if you need to add more detail before generating the mesh 10 00 4 000 5 000 10 00 15 00 20 00 5 000 5 000 15 00 25 00 35 00 15 00 Scale x 1 864 y 1 564 X m If Areas and Elements are selected the program will create the complete mesh 10 00 5 000 e a e a e a e a a T T 30 00 20 00 10 00 10 00 5 x 1 533 1 533 m Oasys Ltd 2009 45 Safe Oasys GEO Suite for Windows 4 4 1 3 Generating mesh data by drawing First draw lines to define boundaries between material zones edges of structural elements excavation profiles etc See Mesh Generation Pointers for guidance on where to place lines Individual points within the problem can be added if required Then create areas either manual or automatically If material
10. QUEUES Graphical Display Options ete lt WIEST QDIOMS utu tubis inten TSI LARS pL 6 25 NE C r rr SATII Sa GRAD dst redit hte A le E 56 3 GOMBINATIONS E E E AE 7 Detailed Processes and Models in SAFE ECT lt 1 ee ee oe E SE E EEE EEE E 7 2 Mesh Generation Pointers eere ern nen 7 3 Gamma w Values eren ener hene nun nenne nasus nene R enun urs 7 4 Stress Strain Models esee ee ee enne nennen nnn 7 4 1 Linear Elastic Materials 7 4 2 Water or Void Materials 7 4 3 Elastic Mohr Coulomb Materials Dilation in Elastic Mohr Coulomb 7 5 Duncan and CHANG iioi eoo deserere Oasys Ltd 2009 Contents IV 7 91 Chang Use liv SAPE na ERE rna na EP ERES 139 7 5 2 Duncan and Chang in Axisymmetric Analyses 140 7
11. 12 00 Y m 10 00 3 000 6 000 15 00 5 000 5 000 15 00 Seale x 1 370 y 1 370 x m 4 lt i Highlight Node or Element allows you to select a node or element number and it will be highlighted with a purple cross 2 For node operations Select the Pick Nodes tool Ts Left click close to each node until all required nodes are selected or draw a box round the required nodes Note Use the Zoom options V Zoom in on an area and R to restore to enlarge dense areas of nodes Selected nodes are indicated with purple crosses Any number of individual nodes or groups of nodes can be added to the selection To clear the selected nodes use Edit Clear selection or click 2 Right click anywhere in the graphics area to bring up the Nodes pop up menu Move Label Restraints applied displacements Springs Line loads Seepage restraints Create list This allows many attributes to be applied to individual nodes for each run by clicking on each item These options are only applied to the current run data Create list creates a globally available list of the selected nodes This can be used later for producing output for the node list Oasys Ltd 2009 Input Data 56 3 For element operations click the Select Elements button BE Left click close to each element centre or draw a box around the required elements Selected elements are indicated with purple crosses Rig
12. 140 7 61 Accelerated Convergent eisioes m 142 1 02 142 e Date e M 142 FA Modified Cam erm 143 7 8 e M 144 A N pip 145 7 8 2 Details of Data for Reinforcement Elements 145 TBS IMGUNOG enm PR 145 y o er 146 BUICK 146 7 9 1 Special Features of the BRICK Model in SAFE 146 1292 Data tetto AREE 147 9 3 EXA MPE TM 149 7 9 4 Deriving BRICK Parameters From Soil Test Results 149 re ea BRICK REGEN eS e TN 150 1 10 U drained m 150 7 10 1 Simpson amp Naylor Method for Undrained Computations 153 7 11 Consolidatiori and Seepage oerte rea aaa a 154 11 1 Tamestep Data orbe nd curva do 154 7 11 2 Steady State Seepage RUNS Aa ae
13. 4 000 2 000 0 65000 0 5000 1 5 160 Cumulative Total Stress Component v 210 20 210 20 Scalex 1 79 vT 78 Interval 20 Scale factors tr 00 2 500 3 500 1 4 500 Satel Event 1 Run 1 Increment 11 21 140406 Oasys Ltd 2009 161 Safe Oasys GEO Suite for Windows Cumulative Pore Wiater Pressure 12 00 10 00 6 000 Y rn rn 4 000 2 000 0 5000 0 5000 1 500 2 500 3 800 1 4 500 1 79 v 1 78 Interval 5 Scale factar Safel Event Run Increment 1 14 00 19 04 06 Figure 3 Cumulative Total Stress Component 12 00 10 00 5 000 6 000 4 000 2 000 0 5000 0 5000 1 500 2 500 3 400 4 500 Stale x 1 78 v 1 78 Interval 20 Scale factors tn Safel Event Run Increment 1 14 00 19 04 06 Oasys Ltd 2009 Detailed Processes and Models in SAFE 162 Cumulative Pore Water Pressure Cumulative Total Stress Component 12 n 12 00 10 00 10 00 o 000 i 6 onn 5 000 pam 4 000 4 000 a 2 000 2 000 0 0 Scalex 1 78 y 1 78 acalex1 8 1 78 Interval 20 Scale factorx tn Safe Event2 Run 5 Increment 1 12 56 19 04 06 Figure 4 Oasys Ltd 2009 163 Safe Oasys GEO Suite for Windows Cumulative Pore Water Pres sure Cumulative Total Stress Component 12 00 12 00 10 00 10 00 6 000 8 00 6 000
14. Safe Oasys GEO Suite for Windows 7 10 1 238 Computed best avoided Possible ned ned to ensure E ue use in cases where the limiting no volume stress stress strength of the soil will not be change gt parameter paramete approached This is unlikely for s linear rs soft clays Possible for OC clays elastic and to give an indication of pore pressure changes suctions close to the surface of an excavation Table X1 Definitions and limitations of Methods A to D a These drainage types are short hand terms which are generally descriptive but not precise definitions In some cases they may be misleading b In some programs the pore pressures are computed at the integration points by including the stiffness of water in the global stiffness matrix then multiplying the very small computed volumetric strains by the very large bulk stiffness of water This can lead to a discOontinuous distribution of pore pressures across the mesh which may be smoothed for the purpose of presentation This approach with some modification is available in SAFE using test 32 However the standard approach in SAFE 1s to find the pore pressures at the nodes iteratively so as to reduce the volumetric strain to near zero Gauss point values are then interpolated from the nodes This process 1s essentially similar to that used for consolidation an
15. 200 2 Run 3 Run 4 Run 5 Run 6 Run 7 Run 8 Run 8 Oasys Ltd 2009 Step by Step Guide 32 Thrust Run 2 Run 3 Run 4 Run 5 Run 6 Run 7 Run 8 Run 9 12 30m y coordinate x 220 0 0 20 0 40 0 60 0 80 0 100 120 140 160 180 Thrust kN m 4 Input Data 4 1 Opening new or existing files To start a new project file select the Create a new data file option on the opening screen or New from the File menu Oasys Ltd 2009 33 Safe Oasys GEO Suite for Windows 4 2 Welcome to Safe Oasys Safe 19 0 build 15 beta Create a new 2 Open an existing file Select recent file Safez Im sfd Pile Analysis sfd Pile Analysis vz sFd Pile sfd Show this welcome screen startup Click OK If creating a new file the program will show a dialog allowing the choice between a full analysis or a steady state seepage analysis Choose the required option and click OK Run 1 is created at this point The Graphical Input window is opened For a new file follow the steps in Creating a Mesh to generate mesh data for the problem The node coordinates element topology and spring data are global data but are initially defined in Run 1 All other geometrical data is run specific Data navigation On opening Safe and choosing to create a new data file the program will creat
16. For problems with time dependent consolidation three sets of data are required consolidation data permeabilities and seepage restraints Consolidation data mainly control the magnitude of timesteps In principle the magnitude of the timesteps for which accurate results can be obtained are determined by the smallest value of he for any element This could result a requirement for many thousands of very small time increments A specially developed iterative approach has been used to avoid this Nevertheless it should be noted that the lengths of maximum time steps will be roughly inversely proportional to the square of the element sizes Very small elements must therefore be avoided as much as possible Materials with high coefficients of consolidation also require short time steps but where precise results in these are of minor importance to the physical problem the techniques described below may be used to prevent timesteps being too small Oasys Ltd 2009 77 Safe Oasys GEO Suite for Windows TT Safe1 sfd Consolidation Data Loading Stepped Ramped Unit weight of water Timestep contral Percent Time at end of run daus Constant increment days Tolerances First increment days Time increment Minimum increment Amplitude of pore Maximum increment mm pressure change kPa Steady state analysis Specialist options Loa
17. See Timestep data for more information on manual settings for timestep data Steady state This box should be checked if steady state groundwater conditions are to analysis be applied In this event only the target end time and the unit weight of Oasys Ltd 2009 Input Data 78 water are required See Detailed Processes Steady State Seepage Runs for a discussion of modelling undrained and consolidating materials in a steady state run Tolerable Allowable magnitude of change of effective stress caused by volume effective stress change The result is within tolerance if the stress increments caused by increment volume change lie within either of these two criteria not necessarily both Amplitude of This prevents rapid changes between iterations of pore pressures and 15 pore pressure necessary when large timesteps are in use Default value is 5 kPa change Tolerance on The accuracy to which the time increment should have settled before the time increment results are accepted Default value is 1 4 6 2 1 Specialist Options Selecting Specialist options for consolidation opens a second dialog box for consolidation data Consolidation specialist options NOTE These values should nat narmally be changed Explicit implicit weighting Factor Maximum ratio between successive time increments Damping factor for pore pressure iteration Ratio of tolerance on pore pressure to that on effective stress 1 Redu
18. Structural analysis of axisymmetric shells AIAA Journal 3 12 Zienkiewicz and Cheung 1967 The finite element method in structural and continuous mechanics McGraw Hill Zienkiewicz OC Chan AHC Schrefler BA amp Shiomi T 1999 Computational geomechanics Wiley Oasys Ltd 2009 169 Safe Oasys GEO Suite for Windows Brief Technical Description SAFE Safe is a computer program designed to perform finite element computations for geotechnical problems The main features of Safe are listed below e Analysis of plane stress plane strain or axially symmetric problems General loading of linear elastic axisymmetric structures can be carried out using a Fourier series technique e Linear or non linear behaviour e Gravity loading e Pore pressures and effective stresses are identified separately allowing computation for drained or undrained conditions and time dependent consolidation e Incremental loading and changes of material properties permitting the formation of excavations embankments etc e Intermediate results can be stored and inspected before the problem is continued further e Pressure distributed and line loading e Fixed or spring restraints e 8 noded quadrilateral elements These can be curvilinear Consolidation problems e Wells e Seepage and flow e Extensive graphical input options and in built mesh generation facility e Wide range of graphical output and printin
19. d 6s ot dK G Let the final stresses be 51 and initial unacceptable stress So to S1 S0 05 ti to ot 1 COS 51 SIN Mohr Coulomb Therefore to dt c s dS SIND Therefore ot cosp so Sind to 1 kd G sino 6s kd amp Duncan and Chang A version of the stress strain model published by Duncan and Chang ASCE SM5 Sept 1970 is available in Safe The Duncan and Chang model is an incremental elastic model of soil behaviour It represents soil by elastic type parameters and varies these as straining proceeds The version in the ASCE paper has some severe limitations and it is necessary to extend it in order to represent the range of stress paths that will be encountered in finite element analyses of substantial problems The model assumes that soil stiffness is generally proportional where is the confining pressure in a triaxial test and n is a material constant The constant of proportionality and n may vary depending whether the soil is loading or unloading reloading During initial loading stiffness is also proportional to 1 R S where is a constant close to 1 and S is the proportion of the soil s shear strength which is currently mobilised Incremental elastic models have never been popular with British soil mechanicians because they cannot properly represent the beha
20. If this is not done the net effect is to give the wrong value of K This is because the effective stresses computed initially before Run 1 are incorrect and are used to compute horizontal effective stresses K Hence the initial horizontal effective stresses are incorrect During Run 1 the vertical effective stresses are corrected automatically by the requirement of equilibrium but the horizontal effective stresses are not changed in proportion by K Rather typically Ao x v 1 v If an initial equilibrium check 15 specified so that Run becomes Event there is no problem In this case vertical effective stress is calculated from equilibrium over ruling Equation A above before it is multiplied by K to get horizontal effective stress So if initial water pressures are not hydrostatic it is best to require an initial equilibrium check making Run 1 become Event 0 The analysis progress view will give a warning message if y and y values 2 and 3 above not equal Stress Strain Models The following stress strain models are available in Safe Linear elastic materials Water or Void materials Elastic materials with a Mohr Coulomb failure criterion An incremental elastic model as proposed by Duncan and Chang 1970 An elastic plastic brittle model with stress magnitude restriction Multi Linear model Modified Cam clay based on Simpson 1973 Soil reinforcement Strain hist
21. Run 4 Run 5 Run 6 Run 8 Rung coordinate x 12 30 300 25 0 20 0 15 0 10 0 5 00 0 Cumulative Displacement x mm In order to plot bending moments shear forces and thrusts in the wall it is necessary to define a list of the elements that comprise the wall This can be done in the Graphical Input window by clicking the Select Elements toolbar button u and clicking on each of the elements that make up the diaphragm wall Then right click and select Create List Now click the Graph toolbar button select the pre defined radio switch and select the appropriate element list for the wall from the Select component from list dropdown list Set the coordinates of the line as before or specify the start and end nodes of the required line Click OK It should now be possible to plot bending moments shear forces and thrusts by clicking the T appropriate buttons respectively Oasys Ltd 2009 Step by Step Guide 30 Bending Moment Run 2 Run3 Run4 Runs 6 Rung Rung 12 30m y coordinate x 350 300 250 200 150 100 50 0 0 50 0 100 150 Bending Moment kMrm m Oasys Ltd 2009 31 Safe Oasys GEO Suite for Windows Shear Force 12 30m y coordinate x 150 100 50 0 0 50 0 Shear Force kNm 100 150
22. With the elements selected as shown the option to split into two along edges 2 and 4 was selected The edges are numbered according to the node listing in the Element Topology table However there is no need to check which edge to set Use the Preview button to see what the result will be and if elements will be split the wrong way simply change the numbers in the Split into cells and press Preview again Oasys Ltd 2009 Input Data 62 Sculpt Elements Translate By x jo rio ads Create copy elements Reflect About axis We Split into Split alona Ed iand 1 Edges 2 and 4 Refine at corner 3 Final result is as follows 00 600 znD D 20D0 1000 Scale x 1 341 w 1 34 __ Er Of course the resulting split elements can be further split to obtain a much finer area of mesh Take care to carry out appropriate refining of the neighbouring elements to ensure there are no Oasys Ltd 2009 63 Safe Oasys GEO Suite for Windows discontinuities in the overall mesh 4 4 1 5 5 4 Sculpt Examples Refining Example refine elements to connect two areas of differing element size This example starts with a simple mesh which has already had elements split successively 10 00 86 6000 2 000 2000 X 10 00 Scale x 1 341 w 1 344 The aim is to refine the el
23. and a ring about the y axis in axisymmetric problems Their units are e For plane stress and plane strain load units are force unit thickness of mesh default kN m e For axisymmetric and Fourier problems load units are force radian of mesh default kN rad Positive components of line loads are in the x y positive directions Line loads can be added in either graphical and tabular input Graphical Input 1 Select the Pick Nodes tool click on individual nodes or draw a box round a group of nodes 3 Right Click to display the Nodes popup menu and select Line loads 4 Enter the required values in the boxes and click OK to accept or Cancel to quit without saving the new data Note The Z Hoop load box only becomes active for Fourier problems Tabular Input Run Data Node Tables Loads Enter the list of node numbers at which to apply load select the required direction and type in the load magnitude Pressure Loads Pressure loads are loads applied along one side of one or more elements For each loaded element the node numbers along the required element side must be entered as list The sign convention for pressure loads is as follows e Tangential pressures are positive in the direction in which the nodes along the element edge are numbered e Normal pressures are positive 90 anticlockwise from that direction e Fourier hoop pressure coefficients are positive anticlockwise in plan Oasys Ltd
24. and deeper materials For a given state as time increment is increased so volume changes will increase and consequently changes of effective stress caused by volume change The program will iterate to find time increments which for all elements give changes of effective stress caused by volume change within one of the two criteria expressed as a change of effective stress generally in kPa or as a percentage of current mean normal effective stress p Iteration will continue until the time increment has stabilised to specified Time increment tolerance 1 596 in the example shown To stabilise the iterative process pore pressures at nodes are not allowed to change between consecutive iterations by more than the specified Amplitude of pore pressure change 5 kPa in the example Typical settings for the parameters controlling time increments could be Constant time increment unset First time increment probably unset unless an initial drained increment is required Minimum time increment unset Maximum time increment unset Tolerable effective stress increment 2 kPa and 195 The following restrictions apply e Where a constant time increment is set minimum and maximum values cannot be set nor can stress tolerances For the first increment of the run a specified first time increment overrules a constant time increment setting e Minimum and maximum increments are never mandatory e otress tolerances must always be set In some case
25. displacements to node coordinates 604711 5 5 Reset iterationz MITER B B Continue from previous increment n OBSOLETE T Apply horizontal seismic acceleration B Gausz point coordinates to a file called ident GFS g 9 Set Fourier frequency 10 10 S AFSD Write RRL to log 11 11 Report flags vv etc screen and log file 12 12 3 rite CSTRS and CSTRHM to log 13 Write element stiffness matrix to log file 14 14 time and node displacements to log file 15 15 BRICK incremental output for element m Leave Apply column blank to apply to ALL A number of different tests can be selected by placing a tick in the Select box Many of the switches write extra detailed results to the log file Some cause the calculation program to read extra data for features under development For more detail see Test Switches 6 1 2 Results Manager The Results Manager is located at the base of the Analysis menu This allows the user to delete selected results thereby reducing the size of saved files particularly where are large number of progressive increments have been used Oasys Ltd 2009 117 Safe Oasys GEO Suite for Windows 6 2 6 2 1 mc8 std Results Manager Increment Creation time Increment 1 13 14 17 10 01 Change ta undraimed and relax Increment 1 12 02 07 11 01 Install Tunnel Lining Increment 1 12 04 0
26. normal consolidation Sega In s Yo Modified Cam Clay material 2 Lambda and kappa see above figure the voids ratio on the critical state line when s 1kPa where S o 1 0 2 2 i sin at the critical state V poisson s ratio for elastic behaviour Scs0 initial value of Scs i e s at the intersection of the current yield surface with the critical state line at level Yo yo level of yo see Scso dSes dy the gradient of Scs with depth The sign is chosen so that this parameter is positive when Scs increases with depth The parameter will be constant for layers with constant OCR including normally consolidated layers Tys Tolerance on convergence of iterative process to find intersection of stress path with yield surface expressed as a proportion of Scs Oasys Ltd 2009 Detailed Processes and Models in SAFE 144 7 8 Recommended value 0 01 i e 1 Size of sub increment of plastic strain to be used in tracing plastic stress strain path Recommended value 0 002 i e 0 2 strain The strain increment magnitude used for this is the root sum square of the X y and Yxy components HF Factor by which the dummy elastic matrix used to form the global stiffness matrix may be softened relative to the standard elastic matrix governed by y if continuous plastic deformation is expected This parameter may be used to speed up convergence but requires a
27. the relevant confining pressure is assumed to be For vertical reinforcement including piles the confining pressure is assumed to be oN in the plane of deformation The stress in the third dimension is not considered The shear between reinforcement and soil is assumed to be entirely frictional not cohesive The value of Parameter 14 must not exceed 10 If it does no check is made on the continuity of stress in the reinforcement Oasys Ltd 2009 145 Safe Oasys GEO Suite for Windows 7 8 1 7 8 2 7 8 3 Restrictions The following restrictions apply e Reinforced elements must be 8 noded rectangles with 4 Gauss points having vertical and horizontal sides e The reinforcement must be vertical or horizontal e The element numbers of all corresponding soil elements must be smaller than those of all reinforcement elements So generate the soil elements first then overlay the reinforcement elements Details of Data for Reinforcement Elements The General data for the soil reinforcement material should be entered in the Run Data Materials General table with the material type set to Soil Reinforcement The specific parameters for the material can then be set in the Run Data Materials Soil Reinforcement table The following points should be noted e The reinforcement should be treated as drained e The density assigned to the reinforcement will act together with that of the soil
28. y and z components of normal stress or strain the xy shear component of stress or engineering shear strain and the major and minor principal stresses and strains Mean volumetric stress or strain shear and ratio of horizontal to vertical stress or strain are also given Sign convention Oasys Ltd 2009 Analysis and Results 120 C Convention of positive stresses T Cy T bi Mahr s circle Major principle direction of angle Is anti clockwise from x direction Note For London Clay the plastic strain is also reported and for Cam Clay the voids ratio is given The element number and Gauss point reference numbers are shown on every page The Error Tab gives the stress errors for non linear elements defined as the difference between the stress compatible with the strains and that required for equilibrium 6 2 3 File Organisation The files for Safe are organised in the following manner Main folder Automatic Description Sub folder Name As specified by user File sfd Main data file Filebak sfd Back up data file Safefile Folder name automatically taken from Oasys Ltd 2009 121 Safe Oasys GEO Suite for Windows file with Safe prefix Created during analysis filer1 sfl Log file can be opened in Notepad or equivalent filer1i1 sfr Results file r1 run number i1 increment number 6 2 4 Graphical Results To obtain a plot of
29. 36 00 40 00 4 000 8 000 12 00 16 00 20 00 Run 2 Insert wall and excavate to 2m To add Run 2 click on the Runs item in the Gateway Click Add New Run and enter the above title when the dialog appears Click OK and the new run should be added In this run the clay materials will be switched to undrained to model the short term behaviour of the materials as the excavation progresses This is done using an effective stress approach whereby stiffness and Poisson s Ratio are specified in terms of effective stress and strength parameters are specified in terms of total stress i e cohesion cu and phi 0 for an undrained Oasys Ltd 2009 Step by Step Guide 14 material During undrained behaviour clays can typically sustain negative pore pressures In this case it has been assumed that the weaker less stiff Clay 1 can sustain 50 kPa of negative pore pressure while undrained and the stronger stiffer Clay 2 can sustain 100 kPa of negative pore pressure This is entered using the Min PWP column in the permeabilities dialog Since the clays are modelled as undrained materials in this run and subsequent runs it is necessary to delete the Clay 1 and Clay 2 rows from the pore pressure table Safe will calculate pore pressure automatically for undrained materials Material parameters are also required for the retaining wall the void occupying the excavation and the slip elements in
30. Cancel Force 001 per M Length level 1 perm Mass 1 kg Stress 0 001 per Time 1 pers Reset Units SI kip ft kip in Default options are the Systeme Internationale SI units KN m The drop down menus provide alternative units with their respective conversion factors to metric Standard sets of units may be set by selecting any of the buttons 51 KN m kip ft kip in Once the correct units have been selected then click to continue SI units have been used as the default standard throughout this document Oasys Ltd 2009 Input Data 36 4 3 3 Preferences Automatic file backup This option allows a time interval to be set for automatically saving the data file Automatic saving can be disabled if required by clearing the Save file check box Note It is good practice to save your file after every new entry of data in each dialog box or table is completed The Save icon on the Standard Toolbar or Ctrl S quick keys can be used to save the file The company information button Company Info in the preferences dialog box allows external companies to specify the bitmap they would like to appear the top of the printed output The company names are as provided in the license file sent to clients by Oasys Ltd These cannot be edited within the program Company Information Enter the full path of the bitmap that you would like to appear on
31. Data GONVEFSION 2 2 a 5 3 1 Stages 5 32 GIGO Ie ly 5 93 39 PROS URANUS ANDES a AEA 5 34 HUE CMAN OOS E t E Lud uu Mee Di 3207 MALOU AIS RENE DID IU IC S FirstStage tua ola ct alie i cher eee ea teats GEDUDOWOIOL E Eo Loo EIS DDR PER ERE 5 3 8 Unsupported Features 6 Analysis and Results 6 1 larl cic MM NEI ES I e A 6 1 2 Results Manager iso ree E 62 PROS UNS cid eed eee Mn Dc LESE 6 21 Tabulated HesullS soap oic noa anoo eu n tabs bs eost 6 2 2 Node Element Result Tables Node a dump fa t t tea Element BeSUIS eu ap dits eine dae e hs em 6 23 Fll OrganiSatl lis iode EE ee dos 6 2 4 Graphical Results Graphical
32. GlobalData RunData Analysis Combinations Queues Window Help oel Axar z 4 M i B E3 Ei OK dem TT Bowles 95a sfd Runs and Events El Input Er Global o Titles Units and Preferences General Events Geometry Gridlines 0 Nodes 1438 Lines 4 Areas 1 Elem Topology 441 E Data for run 1 E Materials 1 o Permeabilities 1 Analysis Options Consolidation B MadeT ables Restraints 1 o Seep Rest 48 Elem Topology 441 Graphical Input l Butput for run 1 T abulated Results NodeElement Result T ables i Graphical View For Help press F1 EX TT Bowles 95a sfd General Data Number of nodes per element Number of Gauss points per element Default deformation mode DLE Calculate initial stresses Plane stress by equilibrium check Event 0 Plane strain Fourier problem Asisymmetric Seepage analysis v Unda Cleanup deletes results fram all increments except the last in the run Delete results deletes all the results of the currently selected run and its dependencies Ca EditingEventiRunit 2 Material permeabilities are entered in the Permeability table which is adjusted to allow input of new materials displacement analysis the Permeabilities table only shows materials which have already been entered in the Material General Data table Fo
33. Initialization Event 1 Aun 2 Install wall and excavate to 2m Event 2 Run 3 Excavate to 4m Event 3 Aun 4 Place UDL H Aun 7 Excavate to 6m and Install Prop at 2m Scenario 2 Event 4 H Aun 4 Excavate to 6m and Install Prop at 2m Aun d Place UDL Scenario 2 Event 5 Aun G Consolidation Steady State E Aun 9 Consolidation Scenario 2 Delete results Cleanup Edit run Delete run Cleanup deletes results from all increments except the last in the run Delete results deletes all the results of the currently selected run and its dependencies Ld This completes the setup of the construction sequence sensitivity check 3 9 Analysis Analysis can be carried out for a single run at a time or for the whole construction sequence Clicking the Analyse toolbar button will bring up the Analysis Progress dialog with the default to analyse the current run It can be useful to analyse each run and check the results before proceeding to enter data for the next run especially during the early stages of an analysis It is a good idea to do this for this example While calculating each run it should be noted that Run 1 will converge within 2 iterations At the end of all other runs however a message is displayed stating Problem has not converged IFAIL The IFAIL value is the number of stress components in the mesh for which the iterative errors are outside tol
34. Input Data 88 dilation ratio may be quoted An incremental elastic model as proposed by Duncan and Chang 1970 An elastic plastic brittle model with stress magnitude restriction Multi Linear model Modified Cam clay based on Simpson 1973 Soil reinforcement The BRICK model Simpson 1992a b Details of data for the first three models are given in the Input Data sections For information about the remaining models consult Detailed Processes General 4 8 4 Stress Strain Data The stress strain parameters required for the material models available in Safe are tabulated below In most applications of Safe all these parameters refer to effective stress behaviour Material Type More information about these models is presented in Stress Strain Models Material types for models 1 2 and 3 1 2 V13 V12 NU13 NU12 G12 G12 1 Linear Elastic E1 V13 2 V12 G12 08 2 Water or Void E1 V13 Eo V12 G12 O1 Ta E1 V13 Eo V12 G12 Qu d 3 Elastic Mohr Co 201 ulomb Cam Clay 3A lt Tsy Agmax Tt First principal Young s Modulus representing stiffness with respect to direct strains parallel to the plane of a soil stratum e g a bedding plane v13 Second principal Young s Modulus for strains perpendicular to the soil stratum Default setting
35. Oasys GEO Suite for Windows Level y Effective Vertical y Ground level stress oy zone 1 7 Water table zone 2 Increase of density x zone When a soil stratum is partially submerged the 9 parameter differs for the wet and dry part of the stratum even though all other parameters are identical Hence two materials are needed to model the partially submerged material in the first stage These extra materials are generated and the appropriate g values for all the strata are calculated during the export process The following figures illustrate this situation Frew first stage material data Oasys Ltd 2009 Frew Safe Link 112 l Frew Manual fwd Data Entry View EIE Level 25 hom um um um cmo um 5 000 3 000 Scale x 1 160 y 1 275 STAGE 0 Inital condition 3 000 Defaults d Maen Equivalent Safe first stage material data Safe1 General material properties Description Type mode aterial rst Stage Material 1 rst Stage Material 2 rst Stage Material 3 rst Stage Material 4 Stage Material 5 rst Stage Material Defaults i lastic Mohr Coulomb E Elastic Coulomb Elastic Coulomb E PL STAN 1 F __ B __ E TOE ete PAREN kN m g TT Been 32 Linear
36. Windows FrewSample fwd Stage Operations Stage 0 Initial Condition No change Stage 1 D ewater AHS amp dig to 27 6 mUD Stage 2 Stress anchor Change this stage Stage 3 Lock anchor and dig to 24 6 ml Stage 4 Install slabs and remove anc Delete stage Add stage Mew stage will be added after the current stage as shown an the Status bar Safe Runs d Frew Safe Link sfd Runs and Events Event 1 Bun Initial Condition Event 2 Aun 2 Dewater AHS amp dig to 27 5 mOD Event 3 Aun 3 Stress anchor Event 4 Aun 4 Lock anchor and dig to 24 5 mOD Event 5 Aun 4 Install floor slabs and remove anchor Delete results Edit run Cleanup deletes results from all increments except the last in the run Delete results deletes all the results af the currently selected run and its dependencies Oasys Ltd 2009 Frew Safe Link 106 5 3 2 Geometry For a given Frew model points are generated at locations corresponding to Material layer boundaries Groundwater levels Surcharge levels Strut locations Wall end points Rigid boundary intersections A series of lines connect these points in the form of a grid Areas are formed from these lines ee ee Critical Points Area n Once this geometry is created mesh generator is called if required by the user The node spacing is dense towards excavation levels Unlike in Frew
37. Windows DXF file select the file and choose Open The DXF Import Options dialog appears DXF Import Options Number of segments Minimum node spacing m Generate areas after import i This allows setting of a uniform number of segments for imported lines and an option to generate areas after import The minimum node spacing should be set to correspond with the data if lines are not exactly joined at their ends in the DXF file this setting allows almost coincident points to be joined together as one node in the imported data Check the DXF file carefully if the result is not as expected complex areas are not created during the import process they can be created manually see Creating Areas Exporting a mesh may be useful if the problem requires areas of mesh to be specified with different global settings The first area can be generated and exported then deleted and the second area generated and finally the first area imported to adjoin the second although a similar operation can now be done using mesh generation options for each individual area or group of areas The mesh can be exported in two ways e By choosing Export mesh from the Mesh Data menu This allows the creation of a tab separated text file which contains the node coordinates and element topology data in the required fixed format e By opening the node coordinate and element topology tables Global Data Node Coordinates and Globa
38. analysis of Retaining Walls is another Oasys program which predicts the soil structure interaction due to construction of a retaining wall Frew calculates the bending moments shear forces and displacements down the wall and the earth pressures on either side of the wall However it does not calculate movements of soil behind or in front of the wall The Frew Safe link feature enables the user to create a Safe model which is nearly equivalent to the Frew model and obtain all the output provided by Safe This feature involves creation of a text format file with the suffix gwa The following steps are involved in the process 1 Validation of Frew data 2 Entry of wall data 3 Export of data from Frew file to Gwa file 4 Import of data from Gwa file to Safe file The first three steps are discussed in detail in the Frew help manual In this manual the fourth step alone is discussed along with the assumptions made in creating an equivalent Safe model from a Frew model Oasys Ltd 2009 103 Safe Oasys GEO Suite for Windows 5 2 Data Import The Frew Safe Link feature is invoked in Safe by clicking on the Import item in the file menu w Edit Mew View alobal Data Run Data Analysis Open Close Save Save 5 Save Run 1 As Save in old Format Print Setup Export GA Import History Version C TesEs JFREWman sfd Tests
39. be re specified for this run subtracting the amount of loading already achieved This setting is repeated in the Consolidation data dialog box Number of equal load The number of equal increments in which to apply the defined applied increments loads and any non zero nodal displacements In many cases this will be 1 Maximum number of For non linear problems a limit must be set on the number of iterations iterations per required This over rides the convergence tolerances set for each increment material In many cases a value of 50 is found to be adequate Save results n times in The results files created by Safe may occupy a large amount of disk this run space In order to limit this the results of some intermediate increments may be discarded Results which have been saved can also be discarded later see Results Manager If this is done no Oasys Ltd 2009 75 Safe Oasys GEO Suite for Windows Convergence Tolerance stress units Estimated ratio of plastic strains Set existing strains and displacements to zero Large Deformation Theory Seismic accelerations Proportion of g output can subsequenily be obtained from those increments For non linear materials a convergence tolerance is usually required as a proportion of the shear strength See for example elastic Mohr Coulomb materials on Elastic Mohr Coulomb The tolerance required here is in absolute units of stress and provides an alter
40. bit of experimentation Try setting it equal to if continuous plastic straining is expected Ti Convergence tolerance for effective stress as a proportion of shear strength tano Soil Reinforcement Safe has a special facility for modelling reinforced soil in which the reinforcement is represented as an averaged or smeared effect Slip between the reinforcement and soil is modelled but the elastic distortion of the soil relative to the reinforcement is neglected Potentially the same technique might be used to model dense groups of piles The reinforced soil model is formed by overlaying elements representing soil and reinforcement The reinforcement element has the same nodes as the corresponding soil element but a separate element number The data required for the reinforcement material are the same as for Elastic Mohr Coulomb materials with the addition of one extra parameter This parameter parameter 14 governs the shear resistance between reinforcement and soil and is equal to where N number of reinforcement strands per unit area of soil perpendicular to the direction of the strands perimeter of each strand perpendicular to the direction of the strand u coefficient of friction between soil and strand Thus for flat reinforcement width b at spacing Sv vertically and Sh horizontally 2b n 1 Sv Sh and Parameter 14 2bu Sv Sh Note For horizontally placed reinforcement
41. easier than C1 in that the initial water pressures are retained eg possibly hydrostatic pressures but no change of water pressure is computed Hence the final water pressures displayed are not correct for the undrained material The total stresses computed are correct but the changes in water pressure are included in what are presented as change of effective stress Method B is probably the most commonly used method The computed changes of water Oasys Ltd 2009 Detailed Processes and Models in SAFE 152 pressure are not correct though the changes of total stress are correct within the linear Mohr Coulomb assumptions of the model Hence the computed apparent changes of effective stress actually include part of the change of water pressure Both computed water pressures and effective stresses should therefore be ignored only total stresses are meaningful For this reason undrained Method B should generally not be used as the starting point for a consolidation analysis unless carefully justified in particular circumstances Method D is in principle the most justifiable and correct method However it relies on having an excellent model of stress strain behaviour which is rarely available and likely to be difficult to use requiring a high level of expertise and understanding Method A should generally not be used unless carefully justified in particular circumstances Meth Draina od Undrai Undrai ned
42. for more details Runs and Events On starting a new file the first run Run 1 is created automatically This is used for entry of the mesh and creation of the base data for the problem To create additional runs once the mesh and run 1 data has been finalised select Runs and Events from the Global Data menu or the Gateway The Runs and Events option opens two windows e Runs and Events showing a tree diagram specifying the links between all the different runs and events in the file e Events and Runs which provides the tabulated equivalent plus a Sequences page which shows all alternative construction sequences Note The number of the first event is dependent on whether an equilibrium check is to be carried out This is specified in General Data If the option to carry out an equilibrium check is selected then the program performs a special initialising Event 0 ie Run 1 is Event 0 This is used to obtain the initial stresses and strains at equilibrium in the ground These are not added to those of later runs If this is unchecked then no equilibrium check is carried out and Run 1 is Event 1 Runs and Events tree diagram This option allows the various Events and Runs within the Safe file to be created and linked On starting a new file Event 0 or 1 and Run 1 will appear in the diagram Oasys Ltd 2009 39 Safe Oasys GEO Suite for Windows Safe1 Runs and Events Sele Evert 1 Rund Delete
43. for pore pressures is derived from Young s modulus divided by a factor this factored value being treated as if it were a Shear strength for this purpose The default value is 1000 Generally it is recommended that no changes be made Well Data Rows of wells can be modelled in steady state seepage conditions If they are used in time dependent consolidation no allowance is made for the consolidation caused by out of plane convergence on the wells That is the behaviour is as if this convergence took place in rigid material with the specified permeability This feature has not been tested to a great extent convergence of the iterative procedure is obtained and the results look reasonable it is very likely that they are correct The wells are involved in the water flow part of the program but take no part in the equilibrium formulation since they effectively represent behaviour out of plane of the 2D calculation They have water pressure but no effective stresses The only density allocated to them is the weight density of the pore fluid but even this does not feature in the equilibrium calculations Forming wells is part of the mesh generation and should be done before restraints and loads are applied or data created for more than one run Node numbering will be changed Data Required Global Rows of wells can be created as in the previous Oasys program SEEP The input is done initially via the graphical input To add a row of
44. for the first Run Note The analysis for Run 1 is influenced by the use of the initial stress theory see General Material Properties Oasys Ltd 2009 Input Data 74 4 6 1 Analysis Options The Analysis Options dialog box allows specification of the following overall parameters for analysis of the current run E Safe Analysis Options o TEN Ramped of equal load increments 1 of iterations per increment 1 00 Convergence tolerance stress unis 1 Increments to save t AJ Specify 1 Estimated ratio of plastic strains Fourier frequency Large deformation Seismic accelerations proportion of g Horizontal 0 Vertical 0 Switches set Far run 1 None aw e Loading Stepped or Ramped Stepped loading is applied in a single instantaneous step at the start of a run Ramped loading adds load incrementally in proportion to increment number if the run is not time dependent and in proportion to time if the run is time dependent The ramping is also applied to changes in fixed heads at seepage restraints see also Test Switches Switch 44 can be used to ramp pore pressure loading on selected nodes only In some cases the target time period will not be achieved within the current run in which case only part of the ramped load will have been applied In this case a further run or runs will be needed to complete the time period and the loads must
45. has been added the user should check the information that has been carried forward from the previous run and cancel out any repeated incremental loading or displacements as required see Restraints Loads Pressures and Spring The user should then work down the Run Data menu in the same way as for Run 1 completing all the information required to create the data for the new run Deleting results e Click on the Delete results button to delete the results from the currently selected run and all its dependencies Deleting Runs Note Once a run has been created it can not be deleted If a run is no longer needed it should be by passed in the calculation sequence by the addition of new runs Cleanup Cleanup deletes results from all increments except the last in the run Editing a run To edit the title or preceding linked run number and increment e Click on the Edit run button e Amend Title run and increment as required Run details Details for run 4 Event 3 Follows Run Increment 0 for last o Oasys Ltd 2009 101 Safe Oasys GEO Suite for Windows 4 11 1 2 Events and Runs Table The Events and Runs table is a non editable form of the tree diagram which details the sequencing between all events and runs Course File vi3 sfd Events and Runs Ff s m rgj NS cid Description Analysed a Time Uate Ines
46. if the dilation of the material is correctly modelled This will affect the strength of the material if strength is specified in terms of effective stress Computations Effective stress approach In the effective stress approach as from February 2002 undrained materials are treated as a special case of consolidating materials with two particular features e their permeabilities are set very low to 1 E 20 m s e nodes at interfaces with drained materials are treated as user specified no flow nodes That is their pore pressures do not constitute boundary conditions to the undrained consolidating zones When undrained materials are used all data normally required for consolidating materials are required except permeability The Consolidation data module must also be completed with a Oasys Ltd 2009 95 Safe Oasys GEO Suite for Windows 4 8 5 2 4 9 target time greater than that of the previous run but not so large that significant consolidation could take place for materials with permeability 1 E 20 m s Undrained materials may be used in the same run as consolidating materials Soil stiffnesses should be specified in terms of effective stresses Before February 2002 computations for undrained behaviour were carried out using a modified form of the method of Simpson 1973 and Naylor 1973 This is described in Simpson amp Naylor method for undrained computations and is still available using s
47. little meaning in this element falling linearly across the element to zero at the end of the reinforcement Brick The BRICK model of soil behaviour has been described by Simpson 1992 a b Information contained in these publications is not repeated here and this appendix should be read in conjunction with Simpson 1992b Reference should also be made to the manual of Oasys BRICK This contains details of recent developments to the theory and further explanation of the parameters required It describes the differences between four versions of the theory all of which are available in Safe as follows BRICK Description Theory Number 1 Original Rankine lecture plane strain model 2 As 1 but held constant during swelling 3 As 1 but 3D behaviour 4 As 2 but 3D behaviour Material types BRICK Theory 1 and 2 may be used for axisymmetric runs as well as plane strain Axisymmetric models are under test and should be used only after reference to Arup Geotechnics Generally the use of tyoe BRICK Theory 2 is recommended Special Features of the BRICK Model in SAFE In Safe the use of the BRICK model differs from other material models in the following respects 1 The whole of the geological history of the soil from initial deposition as a slurry is modelled in Safe This is not problematic or difficult either for the user or for the computer but requires slightly different data The main computation is carried out once for each str
48. material Stress strain parameters can be omitted if only steady state seepage is being modelled The fields in the Permeability table are required for all materials but those directly related to principal permeability k should be omitted for undrained materials For further information see Permeabilities The requirements for pore pressure information is dependent on the chosen state of the material For further information see Modelling Total and Effective Stresses and Pore Pressures General Material Properties All the materials which will be required during the course of the analysis can be added here The materials are then allocated to the different elements to create zones of material types for each run see Element Material Zones New materials can also be added to the list and allocated to elements on the creation of each new run 1 Dewer Place v10 sfd General material properties Sele ae ee es ee Col D ipti Material L in Total weight Relaxation cM L density factors Se Seen aaa Comp Material Linear Elastic Isotropic PL STRN No water Elastic Mohr Coulomb Isotropic PL STRN No water Alluvium Elastic Mohr Coulomb Isotropic PL STRN Undrained Mudstone Elastic Mohr Coulomb Isotropic PL STRN No water Air Water or Void Isotropic PL STRN No water LS Old Masonry Elastic Mohr Coulomb Isotropic PL STRN No water i Volume Loss for North Tu Linear Elastic Isotropic P
49. n3 Poisson s ratio such that Agi Ao1 E1 v12 2 E2 v13 Note that v13 v31 If the default setting for E2 i e E1 is used v12 is set equal to v13 Shear modulus compatible with E1 and E2 Oasys Ltd 2009 89 Safe Oasys GEO Suite for Windows e Alphat Phi Tt yo Beta d Dilation ratio Gradient of dc dy cohesion Ta Default setting E1 2 1 v13 Inclination in degrees of E1 measured anti clockwise from the x direction For horizontal bedding E is horizontal and a 0 Intercept of the Mohr Coulomb envelope with the shear stress axis on the plot This may also be treated as a total stress parameter if there is no pore pressure where Cu Angle of shearing resistance for effective stresses in degrees This may also be used as a total stress parameter if there is no pore pressure Convergence tolerance for effective stress as a proportion of shear strength Convergence is accepted when the difference in the values of stresses between successive iterations is less than Tt coso S SiNd where s Where E2 major principal effective stress and 03 minor principal effective stress See Modelling Total and Effective Stresses with regard to tolerance on undrained pore pressures Ordinate at which specified soil layer stiffness applies This is normally the top for horizontal strata At other dep
50. ned No Non p water Orous d Undrai ned eg BRICK Water pressures Computed to ensure no volume change Not considered set to Zero Not changed from initial data Computed to ensure no volume change Stiffness stress parameter s linear elastic E Va undrained parameter linear elastic Boxe undrained parameter linear elastic Special parameter s of the particular soil Strength Applicability C undraine d strength total stress undraine d strength total stress undraine d strength total stress Gives correct results for undrained analyses in terms of total stresses and displacements Computes pore pressures and effective stresses which in general are not correct and should not be used explicitly eg for consolidation analyses following the undrained stage Gives correct results for undrained analyses in terms of total stresses and displacements Computes only total stresses Does not compute pore pressures and effective stresses Gives correct results for undrained analyses in terms of total stresses and displacements Computed pore pressures and effective stresses remain constant and are not correct Gives correct results for undrained analyses in terms of total stresses and displacements Computed pore pressures and effective stresses are more Oasys Ltd 2009 153
51. or the stress magnitudes involved The results will be very sensitive to this rather questionable assumption and it will increase the sensitivity to KO This feature can be switched off so that the soil is treated as loading whenever shear stress is increasing C he soil will not attain states outside the Mohr Coulomb envelope it fails in shear However the model gives way of calculating the behaviour as the envelope 15 approached unless both shear stress and strain are continuously increasing In states where mean stress is reducing such as the passive wedge beneath an excavation shear failure therefore occurs very suddenly d All increments of stress and strain are automatically divided into sub increments to get reasonable accuracy Two versions are available called Original and Revised in the input table In the revised version the following additional changes are incorporated Duncan and Chang assumed that besides its dependence on approach to shear failure stiffness is proportional to minor principal stress constant Version has the more common assumption that stiffness is proportional to s 5 mean normal stress in plane strain soil approaches limiting shear strength only shear stiffness in the direction of current shearing reduces Bulk stiffness retains its full value and the effective Poisson s ratio therefore changes automatically g The proportion of soil stren
52. pressure gradient is specified In order to generate a roughly equivalent pore pressure distribution in Safe the pore pressures are calculated at locations midway between the Frew pore pressure data points Frew piezometric data Safe piezometric data points points Pore pressure profile oafe plezometric points midway between Frew points Oasys Ltd 2009 Frew Safe Link 114 5 3 8 Unsupported Features The following features are currently not supported by the Frew Safe Link feature Free soil wall interaction Passive softening Generated Young s modulus profiles Minimum equivalent fluid pressure Wall relaxation Mindlin model Sub grade reaction model Analysis and Results Analysis To analyse select Analyse Current Run from the Analysis menu or click the Analyse button 3 on the Safe toolbar File Edit Global Data Run Data Analysis Combinations Queues Window Help Examine results Analvse current run Results Manager view log Files This opens the Analysis Progress dialog Select whether to analyse the current run only all runs in the current sequence or all runs in the file If mutliple runs are selected for analysis the program will only analyse those without current valid results Test switches can be set at this stage but will not be saved with the data for any subsequent analyses Click Advanced Options to select different solver options and request d
53. rather than voids ratio e as used in original Cam clay work and can be derived from oedometer tests provided that the stress range is taken high enough to produce the normally consolidated behaviour represented by A They are related to the conventional coefficients of consolidation Cc and swelling Cs as follows Cc In 10 1 e Cs In 10 1 e Since voids ratio e is a variable A and Cc cannot both be constants similarly and Cs The BRICK model assumes that and are better constants than Cc and Cs and the conversion represented by the equations shown above should be carried out for the a representative value of e The value of Poisson s ratio v can be found from small strain laboratory tests The value required refers to purely elastic behaviour which only occurs in isolation at very small strains following a reversal in the stress strain path A value of about 0 2 is thought to be suitable for many soils The derivation of parameter values from a series of high quality laboratory tests was described by oimpson 1992b When possible a similar process could be followed for other materials Two tests found to be particularly valuable were ashear test in which the direction of shearing was reversed following which stiffness was recorded for a range of strains including very small strains b tests using bender elements to measure the shear stiffness of a specimen at very small strains for a range
54. results Cleanup deletes results from all increments except the last in the run Delete results deletes all the results of the currently selected run and its dependencies Note It is inadvisable to add a new run until all of the data for Run 1 has been completed This is because the new run is created by copying and editing a previous run Adding editing the title of a run To edit the title or preceding linked run number and increment e Click on the Edit run button e Amend Title run and increment as required Run details Details for run 4 Event 2 H3 but rough footing Follows Increment 0 for last Oasys Ltd 2009 Input Data 40 4 4 Geometry 4 4 1 Creating a Mesh Creating a mesh is done from the Graphical Input view The mesh consists of one or more polygons which together cover the whole area to be modelled Each polygon must be defined as an Area prior to mesh generation This can be done automatically or manually Area boundaries can be e straight lines defined by end points or e anticlockwise curves defined by three points which lie on a circle The lines or curves will be split by the mesh generator into a number of divisions to represent the sides of elements The number of divisions initial default 6 can be edited for new lines on the bar at the top of the Graphical Input view Click the Apply button to register any change 11 00 y 2300 Otho Gr
55. second 1 2 Fici 0 22 F2cc 0 5 F2ci OF2ii 0 5 Fourier Analysis Oasys Ltd 2009 Analysis and Results 132 7 2 Combinations may be used to combine results from Fourier analyses to combine Fourier and non Fourier results or to form a results file for a particular orientation from a Fourier results file which contains only Fourier coefficients When a Fourier file is to be combined with a non Fourier the Fourier file must be the first in the combination The combination produces a Fourier file in which pore water pressure is not stored The user chooses to store either Total or Effective stress from the non Fourier file Detailed Processes and Models in SAFE General This section provides a detailed description of the assumptions and calculation methods carried out by Safe It is important to gain a good understanding of these methods in order to be able to use and interpret the program to 15 full extent Mesh Generation Pointers e Model the region to achieve the desired mesh this is not necessarily the same as a physical representation Excessive detail may lead to very large numbers of elements e The co ordinate system requires x horizontally to the right and y vertically upwards For axisymmetric problems the axis of symmetry forming the left boundary must be at x 0 and other x co ordinates must be gt 0 e Element sizes should be reduced in areas likely to compute high strai
56. selection with the node numbers 4 4 1 5 3 Create Node List Lists of nodes can be created for printing Select the nodes required and then Create list Type in a relevant name for the list of nodes The list will be saved and can be used later for convenient printing of node results for that set of nodes 4 4 1 5 4 Label Elements This option labels the current element selection with the element numbers or material zone number 4 4 1 5 5 Sculpt Sculpt allows selected elements to be translated reflected split or refined The operations can be carried out on the original elements or copies can be created This is a very powerful and flexible command and some examples are given below Note that these are small examples for illustration only and may not follow the usual desirable guidelines for size of neighbouring elements aspect ratio of elements etc Tips e When splitting or refining elements use the Preview command to ensure the split direction or refinement corner is as required e Check the resulting mesh is still properly connected using the Highlight Edges command e Any operation can be undone immediately by selecting the Edit Undo command Oasys Ltd 2009 Input Data 4 4 1 5 5 1 Sculpt Examples Reflection Example Copy Reflection about the y axis 58 20 00 10 00 0 10 00 Scale x 1 564 y 1 564 Sculpt Elements Translate Bv x Create
57. strength limit to the stress that could occur at these strains but it is determined by the elastic behaviour It will normally be sensible to start the envelope at strains positive and negative very close to zero The operation of the model in Safe is as follows a For each iteration compute the cumulative principal strains b Assuming elastic behaviour compute the cumulative stresses in the directions of the principal strains For anisotropic materials these may not necessarily be principal stresses User beware Ifthese stresses lie outside the specified envelope reduce them to lie on the envelope Oasys Ltd 2009 Detailed Processes and Models in SAFE 7 6 1 7 6 2 7 6 3 Accelerated Convergence For elements which already lie on the envelope the stiffness entered into the global stiffness 142 matrix may be reduced in subsequent runs This will greatly help convergence provided loading is monotonic ie there is no unloading and the material is isotropic Note that an element may unload because of the brittle behaviour of an adjacent element with no reduction in the applied loading specified in the data Consider for example an element next to one which cracks and loses its strength Problems caused by this might be avoided by applying loading in reasonably small increments Lack of Fit The Multi Linear model contains a feature intended to allow for lack of fit between two adjoining o
58. test sfd the results will be written to the subfolder Safetest The filenaming conventions for the results files are described in File Organisation If the analysis is successful for a single run then the option to save the updated file will be given If the detail pane is opened at the bottom of the dialog by clicking Show detail the following convergence information is shown This is always stored in the log file Event Increment Iteration Maximum At IFAIL Stress Errors Number Number Incremental Node Displacement Oasys Ltd 2009 Analysis and Results 116 Max ElementRMS PWP Element Error Error Error The maximum incremental displacement and the node at which it occurs IFAIL The number of stress components in the mesh for which the iterative errors are outside tolerance Each Gauss point has five stress components including pore pressure The maximum and RMS Root Mean Square stress error in the mesh and the element in which the maximum occurs The maximum pore pressure error and the element in which it occurs 6 1 1 Test Switches Selection of the Test Switches option in the Solution Progress dialog allows the application of various test switches during the analysis Test Switches st Description Select Apply to Value 1 Write times for each subroutine ta log file 2 List all error messages and switches to log file at end of ru 3 3 Logoff at end af run 4 4
59. that has been tested is the use of individual plane stress elements in axi symmetric runs Note For Fourier problems select Axisymmetric If this is selected then the program performs a special initialising Event O ie Run 1 is Event 0 This is used to obtain the initial stresses and strains at equilibrium in the ground Note These are not added to those of later runs If this is unchecked then no equilibrium check is carried out and Run 1 is Event 1 The initial stresses are taken as those specified by the user by the parameter g see General Material Properties In non hydrostatic conditions it may be advisable to specify an equilibrium check See Gamma w Values Oasys Ltd 2009 Input Data 38 4 3 5 for an explanation of this Fourier problem If Fourier theory is to be used in the solution then the Fourier frequency must also be entered in each run s Analysis Options dialog Note For Fourier problems self weight must be zero stress strain properties must be linear and there must be no pore water pressure Total stress analysis may be used for undrained behaviour Seepage analysis the problem is for steady state seepage only with no stress strain analysis at any stage check this box Material stress strain data will not be required and the data input menus Gateway items and available output will be filtered to only show the options relevant to seepage analysis See Steady State Seepage
60. the wall in Safe is made up of Quad 8 elements Oasys Ltd 2009 107 Safe Oasys GEO Suite for Windows 5 3 3 5 3 4 Vikram sfd Graphical Input x 2200 2800 C Orthogonallines Snap interval 2 00 Number of divisions for new lines New line data E oo GA SS X m Restraints The following points should be noted regarding the export of data related to restraints 1 All restraints in the Frew model are of pin type and are applied at the rigid boundaries 2 The restraints are constant across all stages Variable boundary distances across different stages not supported Surcharges The surcharges are applied as element edge loads in the corresponding stages Frew surcharge data Surcharges m i 30 00 Strip Equivalent Safe element load data Oasys Ltd 2009 Frew Safe Link 108 E MERE Leui D G 358155 Noma 25 00 2 665852 Normal 4 5 3 5 Struts Struts in Frew modeled as springs in Safe and act at nodes located on the wall axis Lever arm information is not exported from Frew to Safe Pre stress 15 applied as a node load in Safe The prestress along the spring axis direction 15 resolved into components along the X and Y axes and applied as a pair of node loads in Safe Frew strut data
61. to deformation are not identified This approach should only be used for the linear elastic water void or elastic Mohr Coulomb material types with 0 Undrained stiffnesses Eu etc should be specified with Poisson s ratio close to 0 5 for isotropic materials All parameters including stiffness should be specified in terms of total stress but the material should be declared as drained that is the program is not required to alter water pressures to prevent volume change A Poisson s ratio of about 0 49 not 0 5 should be given to model undrained isotropic materials Strength will be governed by parameters It is generally sensible to set Cu for this purpose making strength independent of stress level All computed stresses are therefore given in terms of total stress Pore pressures should generally be set to zero and will remain unchanged although account will be taken of any non zero or pore pressures which have been specified Undrained behaviour Effective stress approach The more common approach is to use effective stresses and pore pressures In this approach stiffness parameters are given in terms of effective stress including Poisson s ratio and strength parameters in terms either of total stress or effective stress The material should be declared as undrained Initial pore pressures must be specified unless they are zero The values of computed pore pressure will only be strictly correct
62. value is used in the calculations as follows The permeabilities of elements in the column adjacent to the original nodes are of adjacent material 2 nW S In S nD k1 1 e6 inclination of normal to the line of nodes which specify the well so typically 0 meaning K1 is horizontal If there is a conflict i e the same well element is attached to elements of two material zones a warning is given and the higher of the two permeabilities adopted Permeability of the second column of elements representing the well K1 K2 permeability specified by user Kw n D SWw Because the well mesh exists for every run the user must declare wells active or inactive in individual runs When they are inactive their permeabilities are set to very low non zero value in the calculations Oasys Ltd 2009 81 Safe Oasys GEO Suite for Windows 4 7 The run specific well data is set by selecting Run Data Well Data which opens the following dialog box Well Settings 1 Select well welll Well active CO Well inactive Permeability 20 Well spacing m Well diameter m The Well Data menu command is only available if a well has already been created in the Graphical Input view Materials Access to the various material types is via Run Data Materials or the Gateway Adding material data 1 Open the General materials table 2 Add the
63. wells open the graphical input view select a line of nodes then press the well button m This opens the Well parameters dialog box Oasys Ltd 2009 Input Data 80 4 6 3 2 Well parameters Description wat ell Well spacing m Well diameter m Permeability Enter the spacing 5 and diameter of the wells D and the permeability kw of material within the wells usually very large The program then creates two columns of elements attached down one of their edges to the string of specified nodes These are vertical columns if the line of nodes defining the well is vertical as will be common The column of elements attached to the original nodes represents the ground through which water converges on the wells its width W is set equal to 2 S where S is the spacing between wells This is an arbitrary choice since what matters is the product of width and permeability W 2S looks sensible on graphics The second column represents the wells themselves its width Ww D the diameter D of the well When invited to renumber the mesh select the method which minimises bandwidth see Renumbering All the new nodes created in this process are restrained to have no movement in any direction The information given up to this point is global and applies to every run Data Required Run Specific For each run the user must specify the permeability of the material in the wells This
64. which a seepage restraint is required is recorded individually in the Seepage Restraints table but can be specified in groups via the Graphical Input view Restraint types can be no flow fixed head fixed flow or phreatic Oasys Ltd 2009 71 Safe Oasys GEO Suite for Windows For a fixed head node the piezometric water level at the node must be entered For a fixed flow node the flow rate must be entered units m3 s m for plane strain or plane stress m3 s radian for axisymmetric Phreatic nodes are defined as those at which water may leave the mesh and the water pressure at these nodes will be maintained at atmospheric pressure No flow nodes allow no net flow of water into or out of the finite element mesh Internal nodes are taken to be no flow nodes by default and these do not need to be entered in the table Nodes at an interface with drained elements are normally treated as fixed head nodes in the computations if this is not required they may be declared here to be no flow Graphical Input selection by line 1 Select the Pick Lines tool a and click near the centre of the required line 2 Click the Add Seepage Restraint button H on the toolbar 3 Enter the required type of seepage restraint and any relevant values in the dialog Click OK to accept or cancel to quit Graphical Input selection by node 1 Select the Pick Nodes tool and click on individual nodes or draw a box round group of
65. with pore water pressures This specification may vary between material zones For soils the latter approach is usually preferable and for models of frictional behaviour it is almost essential In the following the term drained is taken to mean that pore pressures remain constant during deformation Undrained means that pore pressures adjust to prevent volume change Consolidating means that the pore pressures are determined by seepage and permeability either in time dependent consolidation or in steady state seepage Note Safe normally allows any sequence of drained undrained consolidating steady state Oasys Ltd 2009 Input Data 94 4 8 5 1 materials and runs The only exception is that steady state runs may not include undrained zones of elements A run without equilibrium will allow no displacements and no change of effective stress However equilibrium may be restored in a succeeding run No oads should be imposed during non equilibrium runs and material densities should not be changed from the previous run They may be changed in the succeeding run Undrained Behaviour Undrained behaviour may be represented by either a total stress approach or an effective stress approach More discussion of the applicability of these approaches is available in Detailed Processes Undrained Modelling Undrained behaviour Total stress approach In the total stress approach changes of pore pressure due
66. 0 39 00 Scale 1 238 v 1 238 Interval 0 2 Scale factor 1 m Excavation_Example Event 5 Run Increment 14 07 19 06 09 The equipotentials shown in the diaphragm wall and Fill can be turned off by clicking the View Settings Toolbar button and deselecting the appropriate checkboxes from the Select material zones to display table 3 11 Creating graphs Graphs are plotted by clicking the Graph toolbar button _ which will display the Graph Options dialog Line graphs for displacements and stresses can be plotted by prescribing the end coordinates of a straight line and clicking OK Oasys Ltd 2009 Step by Step Guide 28 Graph Options Use a pre defined element group Advanced Graph Options Pre defined element group Select component From list Definition of axis Enter nodes along required line to plot or a node list name Two nodes define a straight line More than two should be split into series of straight lines CR Specify an axis along which to plot The plotted Point x 12 3 0 parameter can be changed once the graph is displayed 1 2 123 Manual select Multiple runs on same graph Scrolling through the runs will add lines to the graph for each run Oasys Ltd 2009 29 Safe Oasys GEO Suite for Windows Cumulative Displacement x Run 1 Run 2 Run 3
67. 00 Scale 1 238 y 1 238 Interval 1 Scale factor 1 A m Excavation Example Event 2 Run 3 Increment 1 10 05 19 06 09 Oasys Ltd 2009 25 Safe Oasys GEO Suite for Windows Cumulative Displacement Component resultant Units mi 2 000 Oto 210 0 410 B to 2 000 ato 10 te 4 000 1 te 14 tc 000 16 te 18 tc 8 000 201 22 tc E 10 00 12 00 14 00 16 00 18 00 20 00 5 000 5 000 15 00 25 00 35 00 45 00 Scale x 1 238 y 1 238 Interval 2 Scale factor 1 A m Excavatian Example Event 3 Run 4 Increment 1 10 06 19 06 09 Cumulative Displacement Component resultant E 5 000 5 000 15 00 25 00 3500 45 00 Scale 1 238 y 1 238 Interval 2 Scale factor 1 A m Excavation Example Event 4 Run 5 Increment 10 07 19 06 09 Oasys Ltd 2009 Step by Step Guide 26 Cumulative Displacement Component resultant 4 000 2 000 2 000 4 000 6 000 8 000 Y rn 10 00 12 00 14 00 16 00 18 00 20 00 5 000 5 000 1500 25 00 35 00 45 00 Scale 1 291 v 1 291 Interval 2 Scale factor 1 A m Excavation Example Event 5 Run Incerement 1 14 11 19 06 09 The steady state analysis calculates groundwater pressure profiles and flows that will occur in the long term The total flow into the excavation can be obtained by opening the Log file for appropriate runs by selecting Analysis View Log files from the menu bar and opening the appropriat
68. 009 47 Safe Oasys GEO Suite for Windows Safe1 Node Coordinates bs __ A B Coordinates Hode m Defaults FEE 12 0000 18 0000 14 0000 18 0000 14 0000 2 0000 12 0000 2 0000 4 0000 18 0000 4 0000 0000 14 0000 3 7100 4 0000 12 5700 Points table showing coordinates ii Safe1 Lines D E Subdivision Humber of Stark End 3rd segments Points Ho Defaults Segments Segments Segments Segments Segments Segments 11 Segments 12 Segments Lines table showing line properties L ll C C c ca G 1 00 1 00 1 00 1 00 1 00 1 00 1 00 1 00 The properties of each line can be inspected and edited in the Graphical Input view by clicking the Pick Lines button dy left clicking on the required line to highlight it then right clicking The Line Properties dialog box will appear Oasys Ltd 2009 Input Data 48 Line Properties Mame Line 3 Start point Finish paint Third point Change this line only Change all lines at this point Subdivision type Ma af segments CO Constant distance Humber of divizions B Distance of element sizes Set ta gt 1 0 if smaller elements required at START of line and 41 0 if smaller elements required at END of line Post Processing Uptions Restrained direction Rest
69. 13 G12 G12 Shear modulus compatible with E1 and E2 Default setting E1 2 1 v13 Oasys Ltd 2009 Input Data 92 4 8 3 4 8 4 ai Alpha1 Inclination in degrees of E1 measured anti clockwise from the x direction For horizontal bedding E is horizontal and a 0 Note If values for E2 NU12 or G12 are not entered then they are assumed to be equal to E1 NU13 and E1 2 1 NU13 respectively Water Void The parameters required for water or a void are the same as for linear materials with the addition of convergence tolerance for effective stresses in units of stress A small positive value must be supplied for Use of 0 0 for Poisson s ratios is recommended Ta Convergence tolerance for effective stresses in units of stress Elastic Mohr Coulomb The parameters required for Elastic Mohr Coulomb materials are the same as for linear materials with the addition of the following Intercept of the Mohr Coulomb envelope with the shear stress axis the plot This may also be treated as a total stress parameter if there is no pore pressure where C Cu Phi Angle of shearing resistance for effective stresses in degrees This may also be used as a total stress parameter if there is no pore pressure Convergence tolerance for effective stress as a proportion of shear strength Convergence is accepted when the difference in the values of stresses between successive iter
70. 1387 0 00753 409 0 0075 1 3561 417 0 01371 0 00880 409 0 0075 1 4460 417 0 01364 0 01236 410 11100 00075 S8 1 1 3673 417 0 01349 0 00724 321 Problem has not converged IFAlL 1 Results for run increment 1 saved buffer deallacated lt Scroll Convergence be achieved in most runs by opening Analysis Options dialog from the Gateway and increasing the max no of iterations per increment to 200 or 250 and re analysing the model Examining the results Results can be primarily viewed by clicking the Graphical Output toolbar button 4 Results can be displayed in various ways by using the graphical toolbar Ground movements can be assessed by inspecting the graphical output some examples of which Oasys Ltd 2009 Step by Step Guide 24 are shown below Cumulative Displacement Component resultant nits 2 000 Oto 0 5 to 0 1 0 to 1 5 to 2 000 2 0 to 2 50 4 000 3 0 tn 3 5 10 6 000 4 0 to 4 5 tn 8 000 5 0 to 4 5 to 6 0 to 10 00 6 5 to 14 00 16 00 18 00 20 00 5 000 5 000 14 00 25 00 34 00 45 00 Scale x 1 238 w 1 238 Interval 5 Scale 1 A m Excavation Example Event Run 2 Increment 1 10 04 19 06 09 Cumulative Displacement Component resultant LInite 2 000 0 1 0 2 3 2 000 4 5 4 000 ri 6 000 8 J 5 000 1 y 10 00 1 12 00 14 00 16 00 18 00 20 00 5 000 5 000 15 00 25 00 35 00 45
71. 2009 Input Data 70 4 5 4 Arrows show direction of positive loads if nodes i j are numbered in that order Pressure loads can be added by graphical or tabular input Graphical Input Click the Add Pressure Loads button on toolbar A dialog appears allowing entry of the required normal and tangential pressure load at each end of the load Enter the values and click OK Then left click on the left and right ends of the required load These should be near the corner nodes of elements as a pressure load can only be applied along a complete element edge The view will be redrawn to show the new load Tabular Input Run Data Pressure Loads Enter the node numbers along the edge of a single element and provide a direction and load for the first and third nodes along the edge of the element Note that each element must be entered on a separate line of the table If the pressures at the first and third nodes are different the program will apply a linearly varying load along the element edge V Safel sfd Pressure Loads A 8 amp 8 E D jg Pressure Defaults 1 81 74 61 10 00 15 00 z 101 34 8 10 00 5 00 3 Enter the list of nodes for this pressure load must be along one side of 1 element Seepage Restraints Seepage restraints are used to set boundary conditions for consolidation and steady state seepage problems Each node at
72. 6 000 E z z 4000 4000 2 000 2 000 L 0 0 0 5000 0 6000 1 500 2 400 3 500 1 4 500 0 5000 0 5000 1 500 2 500 3 800 4 500 Scalex 1 78 v T 78 Scalex 1 78 v T 78 Interval 5 Scale factar im Interval 20 Scalefactor tn Safed Event 2 Run 8 Increment 14 17 140406 Event 2 Run 8 Increment 1 4 17 140406 Figure 5 Oasys Ltd 2009 Detailed Processes and Models in SAFE 164 Cumulative Pore Water Pres sure 12 00 10 00 000 6 000 im Y rrj 4 000 2 000 0 5000 0 5000 1 500 2 500 3 500 4 500 Scalex 1 78 v 1 78 Interval 5 Scale factor aate Event 3Run 8 Increment 14 19 140408 Figure 6 7 11 3 Steady State Seepage Cumulative Total Stress Component vy 12 00 10 00 6 000 6 000 4 000 2 000 0 5000 0 5000 1 500 2 500 3 500 4 500 Scalex 1 79 vT 78 Interval 20 Scalefactor tn aatel Event 3Run 8 Increment 1 14 18 140406 If the check box in General Data is ticked to define the entire problem as a steady state seepage run only then material stress strain data is not required and only permeabilities and seepage restraints need to be entered In addition one physical restraint must be provided The Gateway view and the Run Data menu are adjusted to remove the material and load tables Oasys Ltd 2009 165 Safe Oasys GEO Suite for Windows SAFE 18 3 Bowles 95a sfd Seles File Edit View
73. 7 11 01 Excavate soil inside tunnel Increment 1 12 06 07 11 01 Apply permanent loads Increment 1 12 02 07 11 01 Change to drained Increment 1 12 08 07 11 01 Delete selected results Delete all results To delete selected results Highlight the required run numbers and press the Delete selected results button To delete all results Select the Delete all results button Results The results are output in both tabular and graphical form Results can be obtained in several different ways and many graphical output options are available using results derived from nodal variables and element stresses and strains Tabulated Results Safe produces large quantities of output and frequently only results for particular sets of nodes and elements are required for printing node and or elements lists have been created see Create Node List and Create Element List the output for those nodes and elements can be written to a text view by clicking the Tabulated Results option on the Gateway Alternatively open the Lists view from the Gateway and select which lists to print then choose Print on the File menu Note Condensed print output is not available for the Results Output view Oasys Ltd 2009 Analysis and Results 118 6 2 1 1 6 2 2 6 2 2 1 List Syntax The following syntax can be used to create list of nodes or elements in the list table obtained using File Print Selection Tabular results
74. 8 4 0 2 Linear rl a 91 4 9 3 Water VOId 5 el 92 8 4 pcne 92 Bloei E Ee N E 93 4 8 5 Modelling Total and Effective Stresses 93 undraned a 94 c ELM RU C 95 1 9 POLE Pressure S 95 4 9 1 Interpolation Of Pressures oui an xanh Eu Cc ac aen eK CER RA 96 4 10 Element Material Zones n uo ua Ead cod vu E cut vau wo UPC ats 97 4 TI rurtnerdH ns RUNS and EVENTIS usen udi SO zu 98 411 1 RUNS ANG Events an das ta Ms tes du aate dat de 98 uns and Events Tree Blagdtalitsuesddssapatuee ope E enean Deo FR CH PE REDE 99 Evens and Runs RN UO E a RENT 101 Oasys Ltd 2009 Safe Oasys GEO Suite for Windows 5 Frew Safe Link Meri a Aa AEE aaa Sa 5 2 Data IMPON a a a a 53
75. Gridlines 0 Event 6 Nodes 5497 B Run6 51 3 0 1 8 Lines 0 Event Areas 0 Run 52 1 0 5 8 Elem Topology 1776 Event 8 4650 1 5 004 1 8 004 1 007 5000 4650 1 5 004 1 8 004 ce Spring Properties 9 Run 8 3 First basement s amp Data for run 12 Bun 11 Same as run 8 but amp Materials 24 5 Eagle sfd Graphical Output 12 Gener 24 Addn _ Si Lo Lom 20 A Total Stress Cumulative JE Permeabilities 24 Linear Elastic 10 Water Void 2 M Elastic Mohr Coulomb 1 sfd inj Graphical Output Ubiquitous joints 0 Analysis Options Consolidation Porepressures 0 amp NodeT ables Cumulative Total Stress Component Restraints 3 App Disp 0 Loads 0 Seep Rest 71 Elem Topology 1776 Pressure Loads 0 Active Springs 3 Graphical Input 2 0007 amp Output for run 12 o Tabulated Results Node Element Result Tables 2000 Graphical View 4 000 6 000 5 000 Scale x 1 557 y 1 557 Interval 20 Scale factor 0 1 Eagle Event 9 Run 12 Increment 1 15 46 13 02 07 For Help press F1 Cell H 9 NUM Editing Event 1 Run 1 Initial equilibrium Analysis Methods Safe is a computer program designed to perform two dimensional finite element c
76. It will often be appropriate to give the reinforcement zero density e The smeared stiffness of the reinforcement ie stiffness per unit cross sectional area of soil is represented by the Young s modulus E1 e The inclination of the reinforcement and of E1 to the horizontal must be 0 or 7 2 1 5708 not 909 e The strength of the reinforcement may be limited using C 5 Note that these are applied to the smeared stresses in the reinforcement ie averaged over unit cross sectional area of soil Limited tensile strength may be represented using 1 5533 ie 89 but in radians and C where T is the required smeared tensile strength THIS FEATURE HAS NOT BEEN ADEQUATELY TESTED e The Poisson s ratios Parameters 2 and 4 should be set to zero and the perpendicular Young s modulus E2 and the shear modulus G12 should be negligible compared with E1 e The convergence tolerance should be a small fraction of the shear strength such that stress errors of smaller magnitude would not be significant This is a bit tricky if p 89 Method Initially the reinforcement is assumed to deform with the ground and stresses are computed from strains accordingly The following corrections are then applied iteratively maintaining equilibrium e The gradient of stress in the reinforcement is limited by the shear that can be supplied by the soil This is determined by Parameter 14 and the normal effective stress in the correspo
77. L STRN Drained IL New Masonry Elastic Mohr Coulomb Isotropic PL STRN No water Volume Loss for Piles Linear Elastic Isotropic PL STRN Drained Short Term Concrete Linear Elastic Isotropic PL STRN No water MG Sleeving Elastic Mohr Coulomb Isotropic PL STRN No water Alluvium Sleeving Elastic Mohr Coulomb Isotropic PL STRN Undrained Mudstone Sleeving Elastic Mohr Coulomb Isotropic PL STRN water Mudstone Large Piles Inte Elastic Mohr Coulomb Isotropic PL STRN No water Mudstone Small Piles Inte Elastic Mohr Coulomb Isotropic PL STRN No water cc ccc Cc IM Material number The material numbers correspond to the zones used to represent the materials in the mesh The user therefore needs to be aware of the number assigned to each material This is now helped by the Materials tab in the Gateway which allows drag and drop of material properties on to areas or elements Colour Generates a colour which will be used to identify this material in the Graphical Input view This can be edited by clicking in the cell Water or void materials default to white if the weight density is zero or blue otherwise Description Add a short but recognisable description This will be copied through to the other tables associated with the selected material type Type of material Select the material type from the drop down list Oasys Ltd 2009 83 Safe Oasys GEO Sui
78. Ltd 2009
79. Mesh Generation Pointers for details of error messages and recommended action to correct them Note Mesh can be generated for one or more areas at a time and reviewed before generating mesh for the remaining areas The program will not create duplicate nodes along the area boundaries However if the mesh settings are different particularly if the varying step definition is set then this process may not generate consistent meshes for neighbouring areas Importing and Exporting Meshes A mesh can be imported in three ways e By creating a DXF file from a drawing package e g AutoCAD then choosing Import DXF from the graphical input view global right click menu e By creating a tab separated text file of a fixed format then choosing Import TXT from the Graphical Input view global right click menu e By pasting suitable data from the clipboard into the Node Coordinates and Element Topology tables The DXF file import process should recognise lines polylines arcs and circles On importing a mesh if there are existing nodes at the same location the node numbers in the imported section will be set to the corresponding existing node numbers and the element topologies revised accordingly To import DXF file right click in the Graphical Input window and choose Import DXF from the popup menu A standard File Open dialog will appear Navigate to the location of the required Oasys Ltd 2009 53 Safe Oasys GEO Suite for
80. O Suite for Windows for convenience If Kw is positive it represents the bulk modulus of pore water usually taken to be 2200MN m For Water or Void materials an appropriate value must always be quoted Permeabilities Orthogonal principal permeabilities k1 k2 k1 horizontal if inclination is zero k2 permeability at right angles to k1 Inclination Inclination of the first principal permeability measured anticlockwise from the x axis specified in degrees Level y0 oet these values if the permeability varies with depth within the material Beta yO is the reference level at which k1 and k2 have been defined and Beta describes the relationship between that and permeability at other depths dk dy K yo where K yo is K at yo Desensitising factor F This must be set to 1 0 for all materials for which an accurate consolidation analysis is required However it will sometimes be the case that the problem involves low cv materials for which accurate consolidation is required and high cv materials which provide boundary conditions to the others and for which some error in pore pressure could be tolerated In this case it is desirable that the time steps are suited to the low permeability materials and the desensitising factor is used to keep the high permeability materials in a stable state In effect for each iteration the coefficient of consolidation of the high cv materials is decreased by this factor ho
81. OWS seo Ptole 50 Generating The MES EL 51 Oasys Ltd 2009 Contents Importing and Expo ning Mesh E Secere ciie sorier Cor Ve err ii dieses 52 Melanie ce 54 Node ard Element ODeralloriS tene deer e b t bor eo EG OE a aeuo den dedica di viae 54 mc 56 amp DTE 57 Greate Node ANE E 57 Label EIEMENMS carpet tado vao a aE 57 siet cT 57 Sculpt Examples ioi iita ti entiers tu 58 Sculpt tE amples TrarislatloHt d iu io ore gt SOS ias aee rendra 59 Sculpt Examples pee uta 61 Sculpl Examples RENNING sess asin roots eric p 63 Modify IRR rS PROP ren Foe Elea x Peut Ebr Dr EXE Pr XXIV vel Pla VERUS UOS 65 Create Element a MAD D M E D cM DE DUE 65 65 4 5 Restraints Loads Pressures and Springs 66 AS RESTANTS D ETT 66 45 2 Apphed DISPIACEMEIUS e
82. SAFE 139 Linear elastic 91 Duncan and Chang in axisymmetric analyses 140 Linear Elastic Materials 135 List of References 168 List syntax 118 Oasys Ltd 2009 171 Safe Oasys GEO Suite for Windows Loads 68 Seepage restraints 70 Simpson amp Naylor method for undrained computations 153 Soil Reinforcement 144 Material Models 87 Special features of the BRICK modelin SAFE 146 Materials 81 Specialist options 78 Matrix Solution Technique 4 Springs 72 Mesh Options 126 otress Strain Data 88 Method 145 Stress Strain Models 134 Modelling Total and Effective Stresses 93 T Modified 143 Moving nodes 56 Multi Linear Model 140 Templates 127 Test Switches 116 167 N Time dependent Consolidation Data 76 Timestep 154 Node results 118 Tips 146 Nodes view 54 Titles 34 Non linearity 4 U p Undrained behaviour 94 Permeabilities 86 Units and preferences 35 Plotting queues 124 V Pore Pressures 95 Pressure loads 69 Print tabular results 117 View Settings 123 Viewing Tabulated results 118 W Relaxation Factors 85 Renumbering 65 Water or Void Materials 135 Restraints 66 Water Void 92 Pressures and Springs 66 Welldata 79 Restrictions 145 Results Manager 116 Results Output 117 Run Data 73 Runs and Events Run 1 38 Runs and Events Option 98 Runs and Events tree diagram 99 5 SAFE 169 Oasys Ltd 2009 Endnotes 2 after index 172 Oasys
83. STRM Fil Interface Elastic Mohr Coulomb sotropic PL STRN Sand Interface Elastic Mohr Coulomb Isotropic PL S TAN Clay 1 Interface Elastic Isotropic PL STRM 10 Clay 2 Interface Elastic Mohr Coulomb Isotropic PL S TAN water water Consolidating Consolidating Consolidating water water water Consolidating Consolidating Consolidating 0 28 0 28 p uw 2 Defaults 1 35 00 01 0 oe to keep the table readable some columns with zero entries have been omitted from the screen shot below 1 MI one Description Ey a T Yo P Dilation E jme rp j meal ri Im a 025 0 2 25 4 004 0 0 35 3 3 004 0 25 0 0 30 4 3 102 004 0 25 0 0 30 5 1 5 004 0 25 0 1 16 67 4 004 0 2 0 0 23 3 n 3 004 0 25 0 0 20 B 3 102 004 0 25 0 0 20 0 05 0 05 0 05 0 05 0 05 0 05 0 05 0 05 f co 0 0 0 0 10839 0 017 0 0 017 0 0 0 0 0 1099 0 017 0 0 017 0 In Safe groundwater conditions during consolidation analyses defined using seepage restraints at the boundaries of the mesh Safe will calculate appropriate pore pressures based on the seepage restraints and variations in permeabilities of the mate
84. Safe Version 19 0 Oasys Oasys Ltd 13 Fitzroy Street London W1T 4BQ Central Square Forth Street Newcastle Upon Tyne NE1 3PL Telephone 44 0 191 238 7559 Facsimile 44 0 191 238 7555 e mail oasys arup com Website http www oasys software com Oasys Ltd 2009 Safe Oasys GEO Suite for Windows Oasys Ltd 2009 All rights reserved No parts of this work may be reproduced in any form or by any means graphic electronic or mechanical including photocopying recording taping or information storage and retrieval systems without the written permission of the publisher Products that are referred to in this document may be either trademarks and or registered trademarks of the respective owners The publisher and the author make no claim to these trademarks While every precaution has been taken in the preparation of this document the publisher and the author assume no responsibility for errors or omissions or for damages resulting from the use of information contained in this document or from the use of programs and source code that may accompany it In no event shall the publisher and the author be liable for any loss of profit or any other commercial damage caused or alleged to have been caused directly or indirectly by this document This document has been created to provide a guide for the use of the software It does not provide engineering advice nor is it a substitute for the use of standard ref
85. a E 3 x Number of nodes per element Number of Gauss points per element Default deformation made Plane stress Plane strain Fourier problem Asisummetric Seepage analysis o See General Data for more details of non default options The Lists Runs and Events items are populated as data is generated so we will return to those later The remainder of the general data relates to problem geometry and mesh generation Problem Geometry The next stage in this example is to generate the mesh This will be done using gridlines to help Oasys Ltd 2009 Safe Oasys GEO Suite for Windows with drawing geometry lines which will in turn be used to generate areas and ultimately the mesh To generate the gridlines open the Gridlines dialog from the Gateway and enter the following gridlines Position Eu 0 000 5 000 10 000 11 900 12 000 12 600 12 700 14 600 19 600 40 000 20 000 17 000 14 000 12 000 5 000 5 000 4 000 3 000 2 000 1 000 0 000 F 4 Press lt gt to start new record 5 4 Mesh Gr dl Bl xl The extents of the graphical view can be changed by clicking the View Settings button on the 41 menu bar and entering the following data Oasys Ltd 2009 Step by Step Guide 8 View settings EB x Boundaries of v
86. aaaee Ea oai 156 EATS Steady State SCC PAGE MEDIA a FIORI ok 164 TAZ TeSt SWICK CS t 167 7 13 General Loading of Axisymmetric Structures nnmnnn 167 8 List of References 168 Sel LIST OMMEICRCNCES EO CUT EE LL LLL DET UN 168 9 Brief Technical Description 169 Hu eE E 169 Index 170 Oasys Ltd 2009 Safe Oasys GEO Suite for Windows 1 2 About Safe General Program Description Safe is a program designed to carry out finite element computations for a wide range of geotechnical problems The program enables the user to study the soil stresses and strains and deformations through one or more sequences of events The events may include excavation or filling insertion or removal of nodal restraints application or removal of loads or changes in material properties Stress relaxation can be specified for any material Time dependent consolidation and steady state seepage can be modelled A range of material models is available to represent soils and structures including linear elastic elastic Mohr Coulomb modified Cam Clay and Brick The program calculates displacements total and effective stresses strains pore water pressures and residual errors for each event or specified increment of a event The results can be obtained in tabular format
87. alysis and gives a smooth distribution of pore pressures across the mesh C The undrained Poisson s ratio v has the value 0 5 for isotropic materials For numerical reasons this is generally entered into the programs as a value very slightly less than 0 5 The undrained stiffness parameter Young s modulus E is derived from effective stress Young s modulus by E E 1 v 1 v 1 5E 1 v d Correct here means that the results comply with the effective stress strain model eg linear elastic Mohr Coulomb and the undrained shear strengths comply with the values tabulated in the input to the program Simpson amp Naylor Method for Undrained Computations This method was used in Safe versions prior to February 2002 and is still available if required using test switch 58 A large and realistic bulk modulus is assigned to the water and its stiffness is added to the effective stress stiffness of the soil Soil stiffnesses should be specified in terms of effective stresses It is found however that in its simplest form this method leads to large fluctuations in pore pressure within each element This has been overcome by changing the formulation of the elements so that a single pore pressure is calculated for each element given by the bulk modulus of water times the mean volumetric strain of the element At individual Gauss points significant volumetric strains may occur but the total volume change of the element is kep
88. and graphically Graphical output options include plotting of vectors and contours of a wide range of derived results and graph plots of parameter variation along a user defined line or the midline of a set of elements Note It is important to realise that Safe is an advanced program which can be used to analyse complex problems The user should be fully aware of the various material models and the finite element method of analysis before use This manual is not intended to teach the basic principles of the finite element method and the user is referred to the List of References for useful texts Note At least 64 kB of RAM is required to run Safe Windows 95 and 98 are not recommended operating systems Program Features The main features of Safe are summarised below Analysis features e Analysis of plane stress plane strain or axially symmetric problems by discretization into 8 noded elements with 4 or 9 Gauss points e General loading of linear elastic axisymmetric structures can be carried out using a Fourier series technique e Linear or non linear behaviour e Pore pressures and effective stresses are identified separately allowing computation for drained or undrained conditions and time dependent consolidation e Incremental loading and changes of material properties permitting the formation of excavations embankments etc e Steady stage seepage with a feature to model flows to a row of wells and the facility to calculate the po
89. and plane strain e Force unit thickness of mesh unit of displacement kN m m For axisymmetric and Fourier problems e Force radian of mesh unit of displacement kN rad m Positive spring stiffnesses act so as to restrain movement Nodes may also be restrained by springs with stiffnesses specified in the directions of co ordinate axes Tabular Input Global data Spring Properties Oasys Ltd 2009 7 Safe Oasys GEO Suite for Windows 4 6 ij Safe1 sfd Nodal Springs Ee B p x y inclination Spring constants M z g Hode from horizontal kN m per m x y 0 0 Defaults 0 43 0 0000 E To activate springs in each run choose Run Data Active Springs and set the Active field to Yes Graphical Input 1 Select the Pick Nodes tool click on individual nodes or draw a box round a group of nodes 2 Right click to display the Nodes popup menu and select Springs 3 Enter the required values in the boxes and click OK to accept or Cancel to quit without saving the new data Run Data Data for Run 1 must be completed before the creation of any additional runs All the possible data options for a run are listed in the Run Data menu Run 1 is created automatically by the Safe program to allow creation of the mesh the setting of the Global Data information which is carried forward to all runs and to allow the setting of the initial data
90. aphical input toolbar The line can also be set to have elements of a fixed size To draw curves Click on the curves button J This will open the curves dialog box Oasys Ltd 2009 Input Data 46 Create Arc Circle Arce Circle Vert split Horz split Semicircle Bottom Right Quadrant SE OSW conrd Y coord Centre Radius Start point End point Third point Arc An arc requires a start and end point and a Third point which is any point on the arc between the start and end points Circle circle requires a centre and a radius It needs to be split either horizontally or vertically for the generation of the mesh Semicircle Semicircles can be defined at top bottom left or right of their centre Each requires a centre or a radius Quadrant The four quadrants also require a centre and a radius A curve will by default be assigned 16 segments This is to obtain better element shapes around the curve As the mesh data is drawn on the screen the points and lines are added to the Grid Nodes and Lines tables These can be viewed and edited by double clicking the item on the Gateway or selecting the menu options from the Mesh Data menu Editing should be done with care Note that any type of curve is converted to a general arc with 3 points defined circles semicircles quadrants are not saved in terms of their centre and radii Oasys Ltd 2
91. ations is less than T coso S SinQ where Where major principal effective stress and o3 minor principal effective stress See with regard to tolerance on undrained pore pressures yo Ordinate at which specified soil layer stiffness applies This is normally the top for horizontal strata At other depths Eryyo 1 B yo y E2 and G12 are found similarly but v13 and v12 are always constant with depth p dis dE dy Etyo Where Etyo is E at yo D is defined so as to be positive when stiffness increases with depth for which dE dy lt 0 and is applied to E1 E2 and G12 i e add negative sign to entered data for a material whose stiffness decreases with depth Oasys Ltd 2009 93 Safe Oasys GEO Suite for Windows 4 8 4 1 4 8 5 d Dilation Dilation constant such that the ratio of plastic volumetric strain ratio increment to plastic shear strain increment is d compression positive i e add negative sign to entered data for a material which compresses during shear d 2 sin v where v is the angle of dilation Milligan 1983 quotes Rowe 1962 who relates to A relationship can also be found in Bolton 1986 Gradient of dc dy positive values indicate cohesion increasing with depth cohesion generally kPa m Ubiquitous Joints For Mohr Coulomb materials the option to specify ubiquitous joints
92. atum of BRICK material and required variables eg KO and brick positions are stored as functions of OCR This calculation is correct only for materials which have a simple history of consolidated from a slurry followed by swelling More complicated histories eg any form of reloading after erosion including the effects of groundwater lowering must be represented in the first Events of the finite element run itself When this is done it is Oasys Ltd 2009 147 Safe Oasys GEO Suite for Windows 7 9 2 sensible to set existing displacements and strains to zero in Solution Parameter data before starting the main computation 2 Hence the value of KO adopted in Safe is not taken from General Material Data in the usual way but is computed by the program from the geological history The value in General Material Data is disregarded The value of g set in General Material Data should relate to the state at the end of swelling before any reloading takes place Reloading may be specified by pressure loads or elements overlying the BRICK layers in Run 1 which may be Event 0 but this is not included in g 3 After Event 1 of a computation BRICK material may be replaced by other material in the usual way eg soil being replaced by concrete but BRICK material cannot replace another material Data Required The following data are required for the BRICK model Lambda Kappa Kappa must be lt lambda 1
93. be selected from pull down menus in the Graphical Output toolbar and Scale Factors and Contour intervals can be reset from the same bar Some parameters whose values are calculated at the Gauss points are extrapolated to nodal positions then the set of nodal values derived from each neighbouring element are averaged where the element edges coincide This helps to smooth out contour lines for example This averaging can be switched off and the parameters switched back to the Gauss point positions if required In addition many of the buttons on the main graphical toolbar operate on this view 41 Opens the View settings dialog see View Settings This is different from the Graphical Input View Settings dialog Turns the ruler guide on or off Turns the background grid on or off Change scale This shows a dialog with options for user specified best fit or engineering scale Best fit is the default Specify Scaling Select scaling User specified Best fit C3 Engineering Independent XY scaling Set scale to 1 scale to ilies zea Turns independent x y scaling on or off Oasys Ltd 2009 123 Safe Oasys GEO Suite for Windows ag Decrease or increase the font size on the view Zoom options Clicking the zoom button allows drawing AJ X a zoom window using the left mouse button and dragging the mouse until the required area is within the box Zooming out will return to
94. bered as shown below Oasys Ltd 2009 Analysis Methods 4 2 1 2 1 1 Gauss point Made Order of numbering node and Gauss points in 8 node element Matrix Solution Technique The overall stiffness matrix is assumed to be symmetric positive definite and banded The band width may be minimised by appropriate numbering of the nodes of the mesh This should usuallu be done after generating the mesh and the mesh generator will automatically carry out one variant of renumbering see Renumbering The matrix is partially inverted into an upper triangle form ready for back substitution This partially inverted matrix can be stored within the file so that further load cases increments or iterations can be carried out without re forming the stiffness matrix Non linearity Non linear behaviour of materials is modelled using the initial stress Zienkiewicz and Cheung 1967 or corrected residual technique Geometric non linearity buckling etc may be modelled approximately using the large strain feature An overall stiffness matrix is formed using approximate linear elastic properties and a set of displacement and strain increments is derived The difference between the true stress increments for the computed strains and the equivalent linear stress increment is then found and the errors in the stresses are applied as additional loads to a further iteration This process is continued until all computed stress
95. cal Input 1 Select the Pick Nodes tool click on individual nodes or draw box round group of nodes 2 Right Click to display the Nodes popup menu and select Restraints 3 Enter the required values in the boxes and click OK to accept or Cancel to quit without saving the new data Select x direction y direction or both to create a PIN restraint For axisymmetric Fourier runs restraint in the hoop direction is also available Restrained nodes will be marked with short purple lines indicating the restraint direction Oasys Ltd 2009 Input Data 68 4 5 2 Applied Displacements 4 5 3 The Applied Displacements table allows input and editing of specified non zero nodal displacements To open the table select Run Data Node tables Applied Displacements Jj Safe1 sfd Applied Displacements Ho Displacement Direction Value mm Defaults EA 2 Node number apply restraint to An applied displacement can be specified in either the x or the y direction In addition applied displacement in the hoop direction is available for Fourier runs Enter the list of nodes to which the displacement is applied and select X Y or HOOP from the drop down list To delete an applied displacement from the table 1 Place the cursor in the grey number column on the line of applied displacement to be deleted 2 When the arro
96. can be found on the manual examples folder in the installation area of the Oasys Geo programs The data in the example is taken from the data for London Clay used by Simpson 1992b Deriving BRICK Parameters From Soil Test Results BRICK parameter values suitable for London Clay have been derived by Simpson 1992b During derivation and refinement of these values a finite element analysis of an excavation in Singapore marine clay was also carried out and a set of parameters used for this work has also been reported Recent work has suggested that the basic parameters used for London Clay may be suitable for another European stiff clay provided the difference in consolidation history 15 modelled It appears therefore that these values form a useful starting point for derivation of BRICK parameter values for clays of marine origin and high plasticity It will normally be sensible to derive and refine parameter values by getting an overall best fit to a range of laboratory and field data Although the model describes a very wide range of soil behaviour particular attention should be paid to parameters which are most critical to the more limited range of behaviour likely to be encountered in an actual problem being considered The parameters and v can be derived from standard laboratory tests The values for A and are equivalent to A and as defined by Houlsby and Wroth 1991 since they are defined in terms of volumetric strain v
97. ch 94 below Figure shows mesh in which the left hand set of elements is undrained and the right hand zone is to be steady state with 5m excess head applied at the base Figure 2 shows the initial conditions drained before the excess head is applied In Figure 3 Run 2 the left zone has been declared undrained and the right zone consolidating the run is declared steady state It is clear that there is continuity of pore pressure between the two zones which should not be the case for the materials as specified More seriously the pore pressures in the undrained zone on the left have changed more or less as those in the steady state zone except at the top and bottom In fact both zones are treated by SAFE as steady state though with a much lower permeability imposed on the undrained zone and the difference between the two arises because the seepage boundary conditions are not applied at the top and bottom of the undrained zone Two ways of modelling a combination of undrained and steady state materials have been investigated a Consolidation analysis The first approach 15 to use a consolidation analysis taken to a point where the steady state material is fully consolidated but the undrained material has not really started to consolidate the program gives it an extremely low permeability This is a natural approach representing what really happens in practice Figure 4 Run 5 shows results Again there is continuity of pore
98. ch contain the same material type are termed the material zones Materials can be assigned to the different elements using either graphical or tabular input methods Graphical Data entry View Graphical Input Switch to the Materials tab on the Gateway To drop existing materials on to single elements left click on the material name and hold down the left mouse button Drag the coloured cursor to an element and release the left mouse button The same can be done for multiple elements by selecting the Pick Elements tool selecting the required elements then dragging and dropping a material Each selected element marked with crosses will be set to that material Alternatively select elements and then right click and choose Modify Properties Tick the Material Properties selection and enter the required material zone number The material zone number is the number from the first column of the table in General Material Properties On selection of the OK button the elements will be re drawn shaded in different colours for each material type Tabular Input Run Data Element Material Zones or Global Data Element Topology These two menu commands lead to the same table which allows the different material zones to be created or edited for each run Oasys Ltd 2009 Input Data 98 Course File v13 sfd Element node incidences and zones Sele sop AH te El mera B aterial Bob um Nodal connection poi
99. cipals the scale factor refers to the full width of the cross and not the distance from the centre For Numbers it is the numerical value to be displayed for the unit value of the parameter Vectors The maximum arrow length can be set If the user chooses to Show scale bars on arrows then should the arrow length required exceed the set length Safe will reduce the arrow by a factor of 10 and place a bar across the arrow Two bars means that the arrow has had to be reduced by a factor of 100 and so on Contours e Contour interval e Maximum number of contours This is set per element For elements that have high gradients of stress or strain and thereby are crossed by many contours the contour interval will be increased to avoid the element being filled with closely soaced contours e Number of intermediate points on contours This is the number of points created to interpolate the contour line across individual elements Number format This sets the number of significant figures to be used in showing number values on the plot Note for contour plots the number of significant figures is based on the required contour interval rather than this setting Toggle results display for material zone This facility allows the user to turn off the results for specific material zones The feature is particularly useful where there are high concentrations of stresses and strains in concrete elements that are causing distortion to the creation of s
100. clays 4 7 2 Permeabilities Consolidation or seepage analyses may be undertaken for materials with isotropic or anisotropic permeability Course File v13 sfd Permeabilities B E D E IT o Drainage Unit wt of Gradient EEN Perm Description pore fluid iu lt E Dee SERES pwp k1 kN m te kN m Defaults No water 2 2 006 1 00 1000 00 1 006 No water 0 0 00 0 00 0 00 0 00 0 00 0 Undrained 2 2e 006 0 00 0 00 0 00 1 00 1000 00 1 006 Undrained 2 2e 006 0 00 0 00 0 00 1 00 1000 00 1e 006 Drained 0 0 00 0 00 0 00 0 00 0 00 No water 0 00 0 00 0 00 0 00 0 00 No water 0 00 0 00 0 00 0 00 0 00 No water 0 00 0 00 0 00 0 00 0 00 No water 0 00 0 00 0 00 0 00 0 00 Undrained 0 00 0 00 0 00 0 00 0 00 oo ccc Cc 2 5 8 3 lt Material description not editable here Description and Repeated from General Material Properties for convenience drainage condition Unit weight of pore The unit weight of pore fluid with depth which is used to calculate pore fluid water pressures from specified data Only in hydrostatic conditions is this equal to the weight density of the water Vw yw should be set to zero for layers of soil above the water table which are unable to sustain suction Bulk mod Kw Used for undrained materials Repeated from the general materials table Oasys Ltd 2009 87 Safe Oasys GE
101. copy elements Reflect About Y axis w Split into Split along Ed 1 and 3 Edges 2 and 4 CO Refine at corner 20 00 w 4 With all elements selected the options above are set in the Sculpt dialog and this produces the following result Oasys Ltd 2009 59 Safe Oasys GEO Suite for Windows HT BERSEERBRERERERE Coe ee A us m Scale x 1 564 y 1 564 A m 4 4 1 5 5 2 Sculpt Examples Translation Example Translate elements along the x axis To Scale x 1 564 y 1 564 4 m iw iw Oasys Ltd 2009 Input Data 60 Sculpt Elements Translate x Create copy elements CO Reflect About Split into CO Split along Edges 1 and 3 Fe Edges 2 and 4 I CO Refine at corner With the rightmost 4 columns of elements selected options above are set the Sculpt dialog note the Create Copy Elements box is not ticked and this produces the following result 20 00 10 00 0 10 00 20 00 Scale x 1 564 1 564 TIT X m 3 Oasys Ltd 2009 61 Safe Oasys GEO Suite for Windows 4 4 1 5 5 3 Sculpt Examples Splitting Example Split elements into two 10 00 6 000 2000 X 1 D Scale x 1 341 w 1 344
102. ction factor Young s modulus for comparison with mean effective stress 1000 Explicit implicit seepage flow in the interval between time t1 and t2 is based weighting factor on an interpolation between the pore pressures 1 and u2 those times equal to 1 1 In the example shown 0 666 the default value Maximum ratio Between consecutive increments the increase of size of time between successive increment adopted is limited to a maximum ratio 1 5 by time increments default If this value is denoted by R then between iterations the time increment is allowed to increase by 1 R 1 10 or to decrease by 1 R 1 20 Thus for R 1 5 the maximum and minimum ratios between iterations are 1 05 and 0 975 Damping factor for Between iterations if the iterative change in the increment of Oasys Ltd 2009 79 Safe Oasys GEO Suite for Windows 4 6 3 4 6 3 1 pore pressure iteration pore pressure reverses its magnitude is limited to a proportion of the previous iterative change 0 5 by default Ratio of tolerance on In checking iterative convergence the tolerance for pore pore pressure pressure may be different from that applied to effective stress errors The default ratio is 1 For iterative pore pressures the tolerance is further reduced in early iterations Reduction factor on For linear elastic materials which have no strength limit the Young s Modulus tolerance
103. d for the limits of compressive and or tensile stress then the Oasys Ltd 2009 141 Safe Oasys GEO Suite for Windows effective stresses are restricted in the directions of the cumulative principal strains In this case the stress limits may vary as functions of cumulative strain a Single limits on compressive and tensile principal effective stress The model has an initial elastic phase which may be anisotropic though anisotropy is not tested and not recommended Principal effective stresses are limited to values specified in the data b Stress limit as function of strain The behaviour of the model is illustrated below It has an initial elastic phase which may be anisotropic with initial stresses ie at zero strain not necessarily zero An envelope to the stress strain behaviour is specified in terms of principal directions of strain by up to 8 stress strain data points of which some may represent compression conditions and others tension Iteratively the program will prevent stress states occurring outside this envelope S 2 Data Point rz 4 Initial e Stress strain stiffness envelope 1 ja pcc Tension Tension Model Behaviour Principal strain and stress in corresponding direction COMPRESSION POSITIVE For the data in the above figure the envelope is absent for strains greater than that of point 4 and less than point 1 there is no
104. data has already been entered material properties can be assigned to areas by dragging and dropping Then click the Generate Mesh button to produce the mesh 4 4 1 3 1 Drawing Lines To draw straight lines Click on the straight line button The cursor changes to a cross hair and as the mouse moves across the screen the current coordinates are shown at the top left of the view Move the cursor to the required first point and click the left mouse button If the point is within the Snap interval of an existing point the beginning of the line will snap to the existing point If the point is near an existing line the line will be split into two and a new point inserted at the required location Alternatively enter the coordinates of the start point of the line in the input bar at the bottom of the screen LINE Enter First point Move the cursor to the required second point then click the left button again While moving the mouse a dotted line will be drawn If the line endpoint is within the snap interval of an existing point the end of the line will snap to the existing point If the point is near an existing line the line will be split into two and a new point inserted at the required location Alternatively enter the coordinates of the second point into the input bar Any new lines will be split into the default number of subdivisions This is initially 6 but can be set to any even number using the edit box on the gr
105. de of pore pressure change kPa This concludes the data input required to analyse all anticipated stages of the excavation Oasys Ltd 2009 21 Safe Oasys GEO Suite for Windows 3 8 sequence However this example goes on to check the sensitivity of the design to construction sequence by including additional runs as follows Alternative Scenario An alternative sequence in which runs 4 and 5 are reversed is considered This is easily accommodated in Safe by adding new runs but copying the previously defined run data add Run 7 a replica of Run 5 but following Run simply select Run 5 from the Runs dialog from the Gateway Click Add Run and ensure that the New run follows Run textbox is set to Run 3 Details for new run Copy data from run na 5 Run Em and Install Prop at 2m Scenario 2 Hew run Follows Rur Increment All other parameters should be correct for this run To add the UDL stage in the alternative scenario copy Run 7 as described above and ensure it follows Run 7 Amend the run by adding the UDL as described in Run 4 To model the final consolidation to steady state Run 6 can be copied and made to follow Run 8 as described earlier The Runs dialog at this stage is shown below Oasys Ltd 2009 Step by Step Guide 22 jer Example sfd Runs and Events Run
106. ding Stepped loading is applied in a single instantaneous step at the start of a run Ramped loading adds load incrementally in proportion to increment number if the run is not time dependent and in proportion to time if the run is time dependent The ramping is also applied to changes in fixed heads at seepage restraints see also Test Switches Switch 44 may be used to ramp pore pressure loading on selected nodes only In some cases the target time period will not be achieved within the current run in which case only part of the ramped load will have been applied In this case a further run or runs will be needed to complete the time period and the loads must be re specified for this run subtracting the amount of loading already achieved Unit weight of This value is required to allow the program to convert pore pressures to water hydraulic potentials It is a constant for all material types unlike the value specified with the materials which should be regarded as the weight density of the pore fluid of the individual material See Gamma w Values for more detail on the difference between these values and how they are used in the program Timestep control The program will select suitable time increments on the basis of user specified criteria This is the recommended option as the program can then use relatively long timesteps when pore pressures are changing only slowly Alternatively selection of timesteps can be manually overridden
107. e 6 Run 9 shows the results of the subsequent run in which the left hand zone is still undrained but the right hand zone is now declared drained retaining the pore pressures from Run 8 Figure 5 The pore pressures and vertical stresses are now discontinuous between the two zones as they should be for ideal undrained and steady state zones Run 9 takes its pore pressures in the steady state drained zone from Run 8 but continues the effective stresses and displacements from Run 1 since they were not changed in Run 8 equilibrium is then re imposed in Run 9 with the left hand zone undrained giving the correct pore pressures for this zone Conclusion The results shown in Figures 4 Run 5 and 6 Run 9 are essentially similar Choice between the two approaches is a matter of preference Generally use of Switch 94 would be quicker because it does not involve a full consolidation run Oasys Ltd 2009 159 Safe Oasys GEO Suite for Windows 12 00 5 000 Y mi 4 000 0 2 000 0 2 000 4 000 6 000 Scale x 1 79 y 1 79 Fri Figure 1 Oasys Ltd 2009 Detailed Processes and Models in SAFE rn Cumulative Pare Water Pressure 12 00 10 00 8 000 6 000 rr 4 000 2 000 0 5000 0 6000 1 500 2 500 3 400 4 500 Scalex 1 78 v T 78 Interval 5 Scale factar amp 1m Safel Event 1 Run 1 Increment 1 11 21 19 04 06 Figure 2 12 00 10 00 5 000 6 000
108. e Run 1 initially containing no data Once data has been entered for Run 1 further runs can be added Run 1 may be Event 1 or Event 0 see General Data for details Runs are the units in which the data are prepared and analysed A whole family of runs can be created as illustrated below where alternative construction sequences are being considered Oasys Ltd 2009 Input Data 34 4 3 4 3 1 Event 1 2 Event 2 E ines Event 3 2 alternative rung in 5 alternative runs Event 5 Run 8 Run 11 3 This is represented in the program by the Runs and Events tree diagram Between Events material properties may be changed including drainage and density springs may be added or removed and incremental loads and restraints may be altered A new stiffness matrix is formed and inverted for each Event The current run is always shown at the bottom right of the status bar To move between Runs for editing either double click on the run title in the Runs and Events tree view or use the navigation buttons on the toolbar gt 2 of Each alternative path through the data is called a Sequence The arrow buttons go to the first run previous run next run selected run and last run in the current sequence respectively If there is only one sequence data navigation is straightforward as each run just follows from the previous one If there is more than one sequence the program wil
109. e actual pore fluid in the material It provides the connection between pore pressure in equilibrium with other stresses and hydraulic gradient needed for seepage and consolidation calculations 2 Weight density of pore fluid in Materials General y This is used in two ways a In Run 1 it is used to compute initial see note below b Hydraulic potential needed for seepage and consolidation calculations is computed as pore pressure y coordinate 3 Gradient of pore pressure in Pore pressures This is used only for calculation of initial pore pressures Special consideration for Run I only An anomaly occurs in Run 1 Event 1 if conditions are not hydrostatic Weight density y 2 above is used to derive initial vertical effective stresses using the total weight density y and the intercept g At level y Oasys Ltd 2009 Detailed Processes and Models in SAFE 134 7 4 o g y x y yyo A However from 3 above is used to compute pore pressures Hence if these two values of y are not consistent the sum of vertical effective stress and pore pressure will not give the correct total vertical stress To avoid this anomaly the value of y 2 above should be changed to equal y 3 above in Run 1 where it is Event 1 This causes no other problems provided that Run is not a consolidation or steady state seepage run which is not recommended
110. e log file e g Excavation Exampler6 sfl or Excavation Exampler9 sfl in this case The total inflow and outflow is shown near the bottom of the text file The results for Run 9 are shown below Total water inflow 2 0 1966E 11 m3 s m Total water outflow 0 1966E 11 m3 s m Groundwater flow across a particular line can be obtained by selecting nodes on the line along which flows are required form the Graphical Input screen When the nodes are selected create a list by right clicking somewhere on the Graphical Input screen and selecting Create List a name for the list will then be requested It will be necessary to close the Safe file and reopen it or re analyse the run in order to generate results for this list Once this has been done nodal flows can be inspected by opening the Tabulated Results table from the Gateway Equipotentials can be displayed for the steady state seepage runs viewing the run in the Graphical zu Output window clicking the Line Contours toolbar button and select Equipotentials from the dropdown list It is advisable to change the contour interval to 0 2 on the toolbar and refresh the graphical output view by clicking the Redraw toolbar button Oasys Ltd 2009 27 Safe Oasys GEO Suite for Windows Cumulative 2 000 0 2 000 4 000 6 000 6 000 10 00 Y rrj 12 00 14 00 16 00 18 00 20 00 5 000 5 000 15 00 25 0
111. e that to keep the table readable columns D to F have been removed from this screen shot jt M o Zune ERE Dilation Taa 5E kPa m mmu 0 1 1 5 004 0 25 0 2 25 0 05 0 0 0 0 0 2 4 004 0 2 0 0 35 0 05 2 0 0 1089 0 0 3 004 0 25 0 75 0 0 05 4 0 017 0 0 0 4 3 102 004 0 25 8 0 0 05 6 0 017 0 5 0 Since the groundwater table is situated 2m below ground level the following pore pressure information should be entered Oasys Ltd 2009 13 Safe Oasys GEO Suite for Windows 3 3 LZ off Pressure influence gradient IKH m pn ip i ext 100 10 00 2 53 200 200 800 100 10 0 3 200 000 100 100 a Each element must be assigned a material type This can be done by selecting each area E selecting groups of areas by clicking the Select Areas toolbar button and selecting relevant areas These areas can be assigned a material type by switching the Gateway to the Materials tab then clicking and dragging the appropriate material type name to the appropriate group of areas The stratigraphy is shown below Om to 2m Fill 2m to 4m Sand 4m to 6m Clay 1 6m to 20m Clay 2 8 000 4 000 0 4 000 8 000 12 00 16 00 20 00 24 00 28 00 32 00
112. e to to specify a range e not to specify exclusions e step to specify granularity of a range e all to specify all For example 1 to 10 not 8 1 to 20 step 2 1 3 5 7 etc all all nodes or elements depending on selected type in table Node Element Result Tables Two tables of results are available for all nodes and elements These are opened by selection of Node Element Result Tables on the Gateway Node Results For each node the output table shows e The x and y co ordinates e The cumulative and incremental displacements in x and y Incremental displacement is that which has taken place during one increment of the run and cumulative displacement is the sum of the incremental displacements for the run and all its preceding runs e The applied loads and residual forces in the x and y directions The applied loads include both point loads and pressure loads transformed into forces at the nodes The residual forces are the forces exerted by the finite element mesh on the fixed boundaries at the end of the run For internal nodes residual forces should ideally be zero Their deviation from zero gives a measure of the convergence achieved in non linear problems e The groundwater head and flow for corner nodes e A user variable This allows a user defined value to be pasted in for each node These results can be plotted for comparison purposes 6 2 2 2 Element Results For elements seven separate pages of outp
113. each method is presented For more detail and advanced topics the reader is referred to the tutorial movies available on the web site Creating the file and setting global data On opening Safe select New Data File from the start dialog Select Full analysis from the next popup dialog this only appears when creating a new file The main screen will open and the program automatically creates an empty set of data for Run 1 Note that the navigation view on the left called the Gateway gives access to the General data items in the top section and the run specific data for the current run in the lower section The data entered in the Titles dialog is reproduced in the title block at the head of all printed information for the calculations so it is useful to provide enough detail to help identify individual calculation runs The Units dialog allows switching between various standard sets of units For this example the first run needs to model initial equilibrium in the ground This can be done by opening the General dialog from the Gateway and placing a tick in the Calculate initial insitu stresses by equilibrium check checkbox If this checkbox is left unchecked then no initial equilibrium check is carried out and stresses are based solely on the g parameter entered for each material Click Apply to activate this change all other defaults are suitable for this analysis TT Excavation Example sfd General Dat
114. ectively adding or removing material between stages of loading Thus an embankment or excavation may be produced in stages The results of a computation may be stored and the problem continued adding further stages of load at a later time The problem area is discretized into a mesh of finite elements to create a model for analysis The elements available are 8 noded isoparametric elements These are quadrilateral with 3 nodes on each side The sides may be straight or they may have quadratic or cubic shapes Thus irregular boundaries may be dealt with adequately Quadratic Element with numbered Gauss points For an 8 node element the displacement variation within the element is quadratic This gives linear strain and stress variation For strain ei in coordinate directions Xj the 4 node elements assume OXj zero if i and constant if z otresses and strains are computed at Gauss points within the elements These are sampling points from which the program carries out integration over the area of the element To show the full variation of strains 8 node elements would have 9 Gauss points However it is found that a smoother distribution of stress is obtained if only 4 Gauss points are used for elements with 8 nodes The 8 node element with 4 Gauss points is particularly reliable and is used most frequently The positions of the Gauss points are computed as described by Zienkiewicz and Cheung 1967 and they are num
115. ed 10 20 5 00 0894 4 899 02 0 00 0 0 000 000 Drained Ol Material 2 25000 190 0 80 User Specified t 0 35 350 1 183 3 74203 10 00 0 0 000 0 00 Drained Oasys Ltd 2009 Frew Safe Link 110 5 3 6 1 Equivalent Safe Mohr Coulomb material data Elastic Mohr Coulomb material properties ed FN E 3 j G12 Alphal Phi Strength m HU13 kPa HU12 kPal kPal 7 m Beta Dilation UEM ae Petes Fe O04 0 1667 43 7 001 3 aH 2 5e 004 0 2308 10 31 27 7 7 4 Poisson s ratio is calculated from using the following equation Angle of friction is the average of values obtained from the following expressions for coefficients of active and passive pressure E d tun dq Mii 1 dq The angle of friction is limited to a maximum value of about 50 degrees First Stage Material In Safe as discussed earlier the initial ground stresses are calculated using two parameters KO and g The slope of the effective vertical soil stress profile is the effective unit weight of the material KO determines the slope of the effective horizontal stress profile and g is the reference level for the linear stress distribution i e the level at which the vertical or horizontal stresses are zero see figure below Oasys Ltd 2009 111 Safe
116. edure is followed for initial loading but v 0 4 is retained during unload reload Duncan and Chang in Axisymmetric Analyses Both versions A and B of the model may be used in axisymmetric analyses but the following limitations should be noted a Inversion A the stiffness in the circumferential direction is reduced along with other stiffnesses during initial loading and as failure is approached by shearing in the radial vertical plane In version B the stiffness is only reduced in the direction of maximum shear stress in the radial vertical plane b In both versions the effect of excessively high or low circumferential stresses is not checked Only the ratio of principal stresses in the radial vertical plane is checked this controls the stiffness of the model Hence shear failure can only take place in the radial vertical plane Multi Linear Model The Multi Linear model is intended to model elastic plastic brittle materials in which the magnitudes of stress are to be restricted The model is particularly suited to masonry and concrete It may be used in plane strain plane stress or axisymmetry but the transverse or hoop stress is not considered in checking non linear behaviour Stiffness may vary linearly with depth lf only single values are specified for both compressive and tensile stresses then the principal effective stresses of the material are constrained to lie within those values b multiple values are specifie
117. ement material zone until the relevant option is switched off or the selection is cleared using Edit Clear selection Node or element selection can be done by switching off the Zoom button and drawing a box around the area to select Definitions Ratio Inc Cum Gives the ratio of the radius Diameter of the Mohr s circle of stress strain increment to the mean volumetric cumulative stress strain Component ox or x y Component or y Xy Shear component or For plain strain or stress z component Oz or z For radial symmetry Hoop circumferential component Mean Mean stress s in the plane of deformation ox oy 2 Volumetric Volumetric strain in the plane of deformation x y Shear Radius for stress or diameter for strain of the Mohr s circle Modulus Wm For Stress V w 2 2 For Strain Yow Resultant NP For vectors V Y TF gt Ma 23 For stress 3 7 For strain gt Oasys Ltd 2009 127 Safe Oasys GEO Suite for Windows Major Principal Major principal stress or strain Minor Principal Minor principal stress or strain Maj Prin Inclin Inclination of major principal stress or strain degrees from Ox Min Prin Inclin Inclination of minor principal stress or strain degrees from Ox Strength Percentage of shear strength mobilised assuming proportionate mobilised variation of a
118. ements below the narrowest elements One of them is selected ready for this process Oasys Ltd 2009 Input Data 64 Sculpt Elements Translate x Ion Create elements y Split into O Split along i along Edges 1 and 3 Preview Edges 2 and 4 Refine at corner 2 ES The corners are numbered in accordance with the node topology list but there is no need to check this Just use the Preview button to view what the result will be and edit the corner number until the required result is obtained Finally the next element is refined and the narrow elements representing a wall for example are changed to a different material property 10 00 8 000 2000 2D0D 000 10 00 Scale x 1 341 w 1 344 X m II p Oasys Ltd 2009 65 Safe Oasys GEO Suite for Windows 4 4 1 5 6 Modify Properties This option can be used to set the material for the selected elements or to set a group number which acts as an identifier for those elements This can be used to filter some graphical views and speed up display of results later in the program Element Properties Eg Material zone E Group 4 4 1 5 7 Create Element List Lists of elements can be created for printing or creating graphs of bending moment shear force or thrust Select the elements required and then Create list Type in a rele
119. erance Each Gauss point has five stress components including pore pressure When an IFAIL value of 0 is obtained the run is considered to have converged Convergence can generally be achieved by increasing the number of iterations allowed or relaxing the tolerance required in the Analysis Options dialog from the Gateway If however a run fails to converge with a large number of iterations and still produces a high IFAIL value and or high stress errors this may indicate an error in the model or inherent instability in the problem being analysed Upon examining the results it may be decided that the errors obtained within the iterations carried out are acceptable In this case the IFAIL and error values should be relatively small and can probably be considered acceptable Oasys Ltd 2009 23 Safe Oasys GEO Suite for Windows 3 10 Analysis Progress Huns to analyse t Current run t Curent sequence f All runs Cancel Hun Information Advanced Options Aun 9 Initialization Increment 1 af 1 Target time days tests set Hun Progress lteration 100 of 100 Cumulative time 0 4000 Increment 0 0 IFAIL vs iteration show displ Calculation Completed Warming 9 Hide detail Inc At IFAIL Disp Node Elem AMS enor Pwperor Elem 0 0075 1 48557 412 0 01410 0 00618 411 0 0075 1 4051 417 0 01394 0 00747 321 0 0075 1 4662 412 0 0
120. erences The user is deemed to be conversant with standard engineering terms and codes of practice It is the users responsibility to validate the program for the proposed design use and to select suitable input data Printed August 2009 Safe Oasys GEO Suite for Windows Table of Contents 1 About Safe 1 15 General Program DescripllOn 1 UA deorum 1 1 3 COmponents or the User INEM ACC siaa a ose ava OE eue 2 2 Analysis Methods 2 2 1 Matrix Solution TECIMIGUG eee E 4 211 oue Eae a a rie ei deeds Es 4 3 Step by Step Guide 5 3 1 Creating the file and setting global data 6 3 2 Hun 1 Initial ConditlOD Catan 12 3 3 Run 2 Insert wall and excavate to 2m 13 3 4 Run 3 Excavate Io 4 eae iut e O E au AS d eR e vw Uwe o dn 16 3 5 Run 4 Place UDL behind 2 2 17 3 6 Run 5 Excavate to 6m and place prop at 2m 18 3 7 Run 6 Consolidation steady state 18 3 8 Alternative ce qr cuum lc
121. ery small amount related to the compressibility of water and the soil grains which is negligible for purposes considered here Hence undrained means constant volume If the soil is not saturated it is possible to have volume change in undrained deformation When soils are subject to undrained deformation their pore water pressure changes so as to prevent volume change which could occur for two reasons a because the total confining pressure on the soil may change b because the soil skeleton tends to dilate positively expansion or negatively compression but this is prevented by change of pore water pressure Loose soils including normally consolidated clays tend to compress as they are sheared leading to an increase of undrained pore pressure and reduction in effective stress in the absence of any change in total confining pressure Dense soils including overconsolidated clays tend to expand as they are sheared leading to an increase of undrained pore pressure and reduction in effective stress in the absence of any change in total confining pressure Modelling Several approaches are available for modelling undrained behaviour in finite element analysis using combinations of effective stress and total stress parameters Both stiffness and strength must be considered and in many cases it is not necessary to take the same approach to both of these In particular it may be convenient to use effective stress para
122. es Sign convention oy p Convention of positive stresses Oasys Ltd 2009 Analysis and Results 130 6 3 Cy T bi Mahr s circle Major principle direction of angle Is anti clockwise from x direction Combinations This option enables the user to factor add and subtract results and to find results for single orientations from Fourier runs It forms combination files with results which can be printed or plotted Existing combination files can be used to form further combinations Combination files are all identified as Run 0 and the increment number is replaced by a Combination number specified by the user The files sfd id is taken from the first file in the combination Mesh zone and restraint data are also all taken from this file To specify combinations choose Combinations Specify Combinations Run 1 must be from the current file The file for the second run defaults to the current file If results from a different file or a previous combination are required press on the Browse button and select the appropriate sfr file Note that results files are named lt FILE gt rX iY sfr and combinations are named lt FILE gt cZ Sfr When one or more combinations have been entered selecting Combinations Combinations Manager allows the combinations to be managed To return Safe to normal mode press the Cancel Combinations button or select Combinations Ex
123. es are within a specified tolerance of their correct values for the computed strain increments Residual errors may be reduced further in future increments since they are handed on as equivalent loads Hence errors are not cumulative Oasys Ltd 2009 Safe Oasys GEO Suite for Windows stress Equivalent linear behaviour i 4 m PO foe True Behaviour Strain The initial stress interactive process The process is illustrated for a single component of stress Convergence is improved if the equivalent elastic stiffness is close to and not less than the true stiffness In the program convergence is accelerated by over relaxation of each component of stress at each Gauss point The method of solution places virtually no restrictions on the non linear material properties which can be adopted For example there are no requirements of plastic normality The material types available are discussed in Stress Strain Models Step by Step Guide The purpose of this section is to enable new users to become quickly familiar with the main features commands and operation of Safe For the analysis of a simple retaining wall step by step instructions are given for entering and analysing the data Users should take time to understand the significance of each operation rather than merely input the commands by rote This help topic should be printed out before undertakin
124. front and behind the wall The diaphragm wall will be modelled as a liner elastic material with a short term Young s Modulus of 18 9GPa and a Poisson s Ratio of 0 1 The material will be assumed to have No Water drainage conditions no permeability information is required The slip elements will be modelled using similar parameters to the soil strata which they represent however cohesion values will be halved and friction angles multiplied by a factor of 0 67 The stiffness parameters can remain unchanged as it has been shown that such analyses are relatively insensitive to the stiffness of interface elements Potts amp Zdravkovic 1999 The material types should be assigned to the appropriate areas of the model using the same technique used for Run 1 12 00 8 000 4 000 0 4 000 8 000 12 00 16 00 20 00 24 00 28 00 32 00 36 00 40 00 44 00 4 000 8 000 Oasys Ltd 2009 15 Safe Oasys GEO Suite for Windows 1 000 0 1 000 2000 3 000 4 000 5 000 6 000 7 000 8 000 9 000 10 00 11 00 12 00 13 00 14 00 15 00 16 00 17 00 18 00 19 00 20 00 During Run 1 of this analysis it is not anticipated that any displacements should be generated hint if they are then something has gone wrong but it is good practice to reset displacements to zero after initialisation This is done by selecting Run 2 from the runs list and opening the Analysis Options dialog from the Gateway and selecting the Set exist
125. g facilities Note The Windows version of Safe requires at least 64KB of RAM to run Windows 95 and 98 are not recommended operating systems Oasys Ltd 2009 Index 170 Index Elastic Mohr Coulomb Materials 135 A Elastic Mohr Coulomb 92 Element Material Zones 97 Element results 118 Accelerated convergence 142 Events and Runs table 101 Analysis and Data Checking 114 Example 149 Analysis Options 74 Exporting and importing meshes 52 Applied displacements 68 File organisation 120 Brick 146 Further Runs Runs and Events 98 BRICK References 150 G C General 132 Combinations 130 General Data 37 Company info 36 General loading of axisymmetric structures 167 Consolidation 154 General Material Properties 82 Create element list 65 General Program Description 1 Create node list 57 Generating the mesh 51 Creating a graph 128 Graphical Display Options 124 Creating a Mesh 40 Graphical Toolbar 121 Creating sub regions 49 Graphs 127 D Data 142 Initial Ground Stress 84 Data required 147 Interpolation of pore pressures 96 Data required global 79 J Data required run specific 80 Definitions 126 Deriving BRICK parameters from soil test results Joining nodes 54 149 Details of data for reinforcement elements 145 L Dilation in Elastic Mohr Coulomb Materials 137 Drained behaviour 95 Label 457 Drawing mesh lines 45 Lackoffit 142 Duncan and Chang 138 Dimedads J89 Duncan and Chang Use in
126. g the example The printer should be set to the best graphics quality available It is inconvenient to keep this help box on the screen while the example is completed Consider the construction stages of a propped retaining wall The initial data run 1 defines the problem geometry and soil conditions including pore pressures before commencement of construction The wall is inserted followed by excavation and placement of a load behind the wall runs 2 to 4 Further excavation insertion of the prop run 5 and finally a switch to long term drained conditions run 6 complete the model An alternative sequence of events in which the excavation and propping is carried out before the application of the load is easily accommodated in the same data file The program records such alternatives as Sequence A Sequence B etc The sequence and run currently being edited or viewed is always shown at the bottom right of the main window in the Status bar The program is designed so that much of the data can be input by graphical means but tabular equivalents exist for most data items It can often be easier to delete items if this is required in Oasys Ltd 2009 Step by Step Guide 6 the tabular form e g nodal restraint lists The following sections describe the data input for general problem data and then for each run in turn As the problem mesh can be generated in a number of ways some introduction of the features of
127. general material data required for all materials and the material type required 3 Select each type specific material table in turn 4 Complete the list of parameters 5 Enter the correct material permeability parameters in the Permeability table for each completed material type 6 Assign the materials to a particular zone of elements in the mesh see Element Material Zones Note Material parameters can be altered for each run as required Assigning materials to the mesh Areas or zones of materials within the mesh are defined by assigning a particular material type to each element All the elements in one zone will have the same material parameters for any given run and may be given a prescribed regime of pore water pressure Elements may be allocated different material types or parameters for different Events thereby allowing the modelling of construction sequences or changes in material behaviour For example to model the removal of material representing excavation the material zone for the appropriate elements should be changed from soil to void The change from undrained to drained conditions can be modelled by altering the material properties between Events Oasys Ltd 2009 Input Data 82 4 7 1 For each material zone four types of data may be specified 1 General data 2 Stress Strain parameters 3 Information about pore pressures 4 Permeabilities General data must be entered for each
128. give the run the above title Tolerable eff stress increment Stress units 2 Percent Tolerances Time increment Amplitude of pore pressure change kPa The only changes to make in this run are to change the material type of the sand elements in front of the wall to Excavation void using the method described earlier and to increase the target time at the end of the stage to 0 2 6 000 4 000 2 000 0 2 000 4 000 6 000 8 000 10 00 14 00 16 00 While setting up Run 2 the existing strains and displacements were set to zero It should be Oasys Ltd 2009 17 Safe Oasys GEO Suite for Windows 3 5 confirmed that this option is unchecked for Run 3 and all subsequent runs Run 4 Place UDL behind wall In this run a uniformly distributed load of 20kPa will be placed on the surface immediately behind the diaphragm wall to x approximately 25m To do this first add the run and give it an appropriate title Zoom into an area around the surface reaching from the diaphragm wall to approximately x 25m Click on the Pressure Loads toolbar button A enter a normal pressure of 20 kPa in the Start and End textboxes and click OK Define Pressure Load Enter required pressure values Start End Tangential pressure pressure 20 20 0 E Cancel The pressure loads are added by selecting two nodes one at either end of the location where the load is
129. gth mobilised 5 which governs incremental stiffness 15 calculated by comparing the shear stress with the limiting shear stress at the current value of mean stress s rather than the current value of o as in the original model 7 5 1 Duncan and Chang Use in SAFE In Safe the model has the following parameters The values at the right of the table have been used for Frankfurt Clay 1 20 2 20 3 Poisson s ratio when remote from shear failure 0 5 4 constant of proportionality 225 5 K for unload reload 650 6 n index as above 0 6 7 Nar n for unload reload 0 6 Oasys Ltd 2009 Detailed Processes and Models in SAFE 140 7 5 2 7 6 8 R as above 0 9 9 pa atmospheric pressure eg 100 kPa used to define stress units 100kPa 10 maximum sub increment of strain Subject to parameter 13 5E 4 11 T convergence tolerance as a proportion of shear strength eg 0 05 0 05 means 5 12 0 1 for don t do observe yield surface see b above 1 13 Maximum number of sub increments Within this restriction the 30 number used depends on 10 and is also reassessed to ensure that the stress sub increments do not exceed 5 T shear strength 14 Original or revised model 1 If Parameter 3 Poisson s ratio is set to 1 in Version A its value is substituted as follows 5 lt 0 55 v 0 4 S gt 0 98 v 0 495 0 55 lt S lt 0 98 interpolated linearly If it is set to 2 the same proc
130. h Lines shorter Advanced option than the specified fraction of the longest line will be set to have 2 segments This is just to save the user having to explicitly set numbers of segments for example where lines are automatically split during or after drawing Refinement coefficient lf step size varies significantly this parameter may Advanced option become important It governs how large the areas of finer mesh should be The larger the value the more space will be occupied by fine mesh maximum value 0 35 Renumber during mesh This carries out node renumbering to reduce the problem generation bandwidth See Renumbering for more detail but note that because no node is selected at this stage only the bandwidth minimising form of renumbering is available Select OK to generate the mesh The program will report the number of nodes and elements and the bandwidth obtained If the mesh is not satisfactory selecting Edit Undo immediately will erase it Alternatively right click on the graphical view area and select Erase Generated Mesh or click on the toolbar After erasing the mesh the view will return to show the mesh generation data which can be edited until the required mesh is produced Selecting Erase All Mesh Data from the right click menu will erase all points lines and subregions as well as the nodes and elements the mesh generator cannot form the mesh using the data provided an error message will be given See
131. have to specify the materials first If multiple areas are selected the same number will be applied to all of them To delete an area 1 Select the Pick Areas tool r3 2 Click near the area s centre 3 Press the delete button Creating and Editing Points Points entered as part of the mesh generation data will have nodes created at their location These nodes are the same as other nodes created during line drawing or mesh generation but have two additional properties a desired element size adjacent to the node and a radius of influence over which the mesh generator will attempt to apply this desired element size If a generated mesh is deleted these individual points will not be deleted but will remain as part of the mesh data To create a point click the Add Node button onthe toolbar A dialog appears in which the point s coordinates desired adjacent element size and radius of influence should be entered On clicking the OK button the point is added to the mesh data and to the Grid Nodes table accessible via the Global Data menu and the Gateway Oasys Ltd 2009 51 Safe Oasys GEO Suite for Windows 4 4 1 3 4 Add All dimensions in m Coordinates Length of neighbouring elements Radius of influence To edit an existing point click the Select Nodes button click on a node to select it Right clicking then brings up a popup menu from which the Edit o
132. he sub region Each line will change to orange as it is selected When all the lines constituting an area ie a closed polygon ignoring any hanging internal a lines have been selected click the area button 4 f the area can be created successfully the Area Properties dialog will appear showing the lines which make up the area boundary This also allows entry of the material to assign to all elements generated within this area Properties Area AT Material Lines an boundary If the area has not been properly specified then an error message will be shown Oasys Ltd 2009 Input Data 50 4 4 1 3 3 A Boundary nat closed or nothing selected Auto generation of areas will not set the material property for the elements within the areas There are two ways to do this If material data has already been entered switch to the Materials tab on the Gateway To drop existing materials on to single areas left click on the material name and hold down the left mouse button Drag the coloured cursor to an area and release the left mouse button The same can be done for multiple areas by selecting the Pick Areas tool Pr clicking in each area then dragging and dropping a material Each selected area marked with crosses will be set to that material Alternatively select areas and then right click The Area Properties dialog will appear Enter the required material number you do not
133. ht click to bring up the Elements pop up menu Label ina Label zone Sculpt Modify properties Create list Select Zones This allows labels material group numbers and distributed loads to be assigned These options are only applied to the selected elements in the current run data Note to create distributed loads the element edge s along which to apply the load also have to be selected Create list creates a globally available list of the selected elements This can be used later for producing output for the element list The Sculpt option allows elements to be split or refined into successively smaller elements This obviously affects the mesh used throughout the model not just the current run 4 4 1 5 1 Moving Nodes This option allows the location of individual nodes or a line of nodes with the same x or y coordinates to be moved 1 Choose the required nodes using the Select Nodes tool Ady 2 Right click to obtain the node menu 3 Select the Move option to open the following Translate by direction 0 1 y direction 0 Enter the required translation distance s Note Extreme care should be used when moving nodes to make sure that they do not pass beyond adjacent nodes Midside nodes will be repositioned automatically to remain midway between corner nodes Oasys Ltd 2009 57 Safe Oasys GEO Suite for Windows 4 4 1 5 2 Label Nodes This option labels the current node
134. idinterval 0 500 Spit new lines Display Areas can also contain hanging lines and individual points of interest at which nodes will be generated Lines or curves need not be joined to other lines at both ends but do need to be inside a polygonal area in order to be meshed Individual points of interest at which nodes are required can also be specified These may lie on a line or may be internal to a subregion For more detail on defining mesh data please see Mesh Generation Pointers There are three methods of creating a mesh in Safe e by specifying Gridlines and generating the mesh based on those e by drawing lines on the screen and combining them into polygons either automatically or manually e by importing data from a DXF file or text file The first and second of these methods can be combined to generate simple data based on gridlines and add detail by drawing 4 4 1 1 Options in the Graphical Input View Object Snap The graphical input view defaults to snap newly drawn objects to existing objects New lines will snap to existing line endpoints midpoints or will be drawn perpendicular to existing lines if the object snap is ON This is shown in the Snap toolbar A on If the Object Snap is switched OFF the settings on the View Settings dialog will govern whether a new line is joined to an existing line View Settings Select from the toolbar to edit the view settings Oasy
135. iew Drawing settings direction Pen width in pixels 2 ik N Grid interval Y direction Min 25 Snap to grid fw Tolerance for node 0 05 coincidence Cursor made Crosshairs Arrow only Cancel Gridlines should be displayed as shown below 15 00 10 00 5 000 0 5 000 10 00 15 00 20 00 25 00 30 00 35 00 40 00 45 5 000 5 000 10 00 15 00 20 00 1 Clicking the Regular Mesh toolbar button will display the following dialog from which the Lines Only option should be selected This will generate the blue geometry lines that are required to define the mesh The default global element size should be accepted as this will be adjusted later Oasys Ltd 2009 Safe Oasys GEO Suite for Windows Quick mesh settings dH X Generate Areas and elements 15 00 10 00 5 000 RU 5 000 10 00 15 00 20 00 25 00 30 00 35 00 40 00 45 5 000 5 000 10 00 15 00 20 00 Thin interface elements 100mm wide will be placed on either side of the concrete wall elements in order to limit friction on the back and front of the wall In finite element analysis of excavations supported by retaining walls it is usual to include some refinement of the mesh around the wall in order to obtain more accurate results This is done by assigning each geometry line a particular subdivision type which will determine how many segments the lines a
136. increment starts with the assumption of an unnecessarily small time increment It is therefore recommended that the user should impose a reasonable first time increment in such cases For strictly undrained behaviour see Undrained behaviour Optionally the user may set the minimum and maximum time increments to be used If the minimum is not set by the user it is found by Safe using the smallest value of h2 6Cv for any element in the mesh If the maximum is not set it is assumed equal to 10000 times this smallest value It is recommended that the user should not normally set these limits Safe derives suitable time increments by limiting the magnitudes of those changes of effective stress which are caused by volume change This is achieved by computing separately the stress increments including volume change and those that would occur if the strains were changed by removing the volumetric component then taking the difference The user can specify what magnitude of change of effective stress caused by volume change is to be tolerated The specification is given both in stress units and as a percentage of the existing mean normal effective stress at each Gauss point The result is regarded as within tolerance if the stress increments caused by volume change lie within either of these two criteria not necessarily both This combination makes it possible to give reasonable criteria for both near surface materials where stresses may be very low
137. ing a drainage boundary Drained undrained and consolidating materials may be used together in the same run Drained elements provide additional fixed heads at their nodes Total weight density Saturated bulk moist weight density as appropriate Relaxation factors Added after Run 1 replacing Ko and g see Relaxation Factors Weight density of poreWeight density of pore fluid is used to calculate pore pressures from fluid specified data More precisely this is the gradient of pore pressure with depth Only in hydrostatic conditions is this equal to the weight density of the water oee Gamma w Values for more details Bulk modulus of waterUsed for undrained materials see Undrained behaviour Kw Bulk modulus of pore water is usually taken to be 2200MN mz except for material type 2 Water or air for which an appropriate value must always be quoted Run 1 only coefficient of earth pressure in initial condition o h Oasys Ltd 2009 Input Data 84 4 7 1 1 9 1 only g defines the profile of initial vertical effective stress with depth and is equal to yw beneath the water table and y o y y above the water table where y is the ordinate of any point in the zone g is a constant for each material zone provided the ground surface is horizontal and there are no variations of overburden pressure across the zone g is the intercept of
138. ing strains and displacements to zero checkbox Click Apply This dialog is also used to change the maximum number of iterations allowed during the calculation of each increment discussed further later T Excavation_Example sfd Analysis Options C Ramped Ma of equal load increments of iterations per increment 00 Convergence tolerance stress Increments to save AJ C Specify 0 Estimated ratio of lo plastic strains Fourier frequency Large deformation thea Stem loading Seismic accelerations proportion of g Horizontal 0 Vertical 0 Switches set Far run 2 None Switches Apply Undo When undrained materials are modelled it is necessary to increase the target time for the end of each run To do this open the Consolidation dialog form the Gateway and change the Time at end of Run days to 0 1 Click Apply This is usually increased in increments of 0 1 days for each construction stage Oasys Ltd 2009 Step by Step Guide 16 TT Excavation Example sfd Consolidation Data Loading Stepped Ramped Unit weight of water EN m3 1 0 Timestep control Time at end of run daps 0 1 Constant increment days First increment days Minimum increment aximum increment Steady state analysis Specialist options 3 4 Run 3 Excavate to 4m Add a new run in the way described earlier and
139. is available An additional item is added to the Gateway to enter the relevant data Up to 5 ubiquitous joint sets can be specified or Hoek Brown behaviour specified After the normal stress checks for the base material stress limits on the joints are checked Joints take no tension and optionally display hysteresis Hysteresis implies that when cracks have opened they are filled eg with mortar so that they do not need to close before taking compression If Hysteresis is set to No the default the cracks have to close reversing the tensile strains to zero before compression is taken Joints may be given Barton Bandis behaviour by setting cohesion to zero and adding the JRC and JCS values If Hoek Brown is specified then the Mohr Coulomb c phi are ignored and sigma ci m s and a are used instead Once the number of joints has been selected the correct number of lines is added to the table for completion of the data A sil Barton Bandi arton Bandis Params pier uad Tore ee ate Optional Hoek Brown Parameters 3E EN Um kN m m JRC JCS Sigmaci Defaults Fill dry Hoek Brown Grade V wet 3 joints 28 5 31 2 3 4 Grade wet 2 joints 6 Grout _ v Hoek Brown FEE 4 Cell A 6 Modelling Total and Effective Stresses For each material zone computations may be carried out in terms of either total stress or effective stress together
140. isplacement including zero displacement in the x and or y directions For Fourier problems nodes can also be restrained or have a specified displacement in the hoop direction Inclined sliding restraints are not available Note When a new run is created by copying data from an existing run any restraints and non zero applied displacements will be copied into the new run If they are no longer required they should be deleted Restraints and applied displacements can be specified in either tabular or graphical input The two Oasys Ltd 2009 67 Safe Oasys GEO Suite for Windows forms of data entry are fully interchangeable Tabular Input Run Data Node tables Restraints The Restraints table allows input and editing of the direction and node numbers of horizontal vertical or pinned restraints For Fourier problems nodes can also be restrained in the hoop direction To specify a restraint enter each list of nodes to be restrained and choose X Y or PIN from the drop down list Safe1 sfd Nodal Restraints 1 eee Mo List of nodes Restraint Defaults 5 i x 2 13 20 33 40 53 60 73 80 33 100 113 120 1 PIN 3 Made number to apply restraint bo To delete a restraint via the Nodal Restraints table 1 Place the cursor in the grey number column on the line of restraints to be deleted 2 When the arrow appears left click to highlight the line 3 Press the delete button Graphi
141. it Combinations Oasys Ltd 2009 131 Safe Oasys GEO Suite for Windows Specify Combination E 5 x First Hun File C datastest restraints std Second RunCombination File C datatest restraints sfd Increment um Increment i Irientatian NM Onentatior jo Factors represented as Fnjk F Factor n 1 or 2 for the first or second run in the combination Apply the factor to the cumulative or incremental results from the run Use this factored result as part of the cumulative or incremental value in the combination Runs from existing combinations should be designated as 0 and given their corresponding combination increment number Factors allow results from the runs to factored before being combined Both the original results and the combination will include cumulative and incremental values However in the combination it may be required that the incremental values are actually the difference between the cumulative values of two runs The factors make this possible Example 1 In the combination the cumulative results are to be those of the first run whilst the incremental results are to the difference between the cumulative results of the first run and the cumulative results of the second run Then Ficc 1 Fici 1 F1ii 0 F2cc 0 F2ci 1 F2ii 0 Example 2 Form a load case equivalent to 2 the first run minus half the
142. l Data Element Topology copying all data to the clipboard and pasting into Excel Format for TXT file for importing or exporting a mesh First line number of nodes n Lines 2 to 1 x and y coordinates of nodes separated by tabs 1 node per line Line n 2 number of elements Subsequent lines element material 0 if unassigned and the 8 nodes in each element ordered anticlockwise Oasys Ltd 2009 Input Data 54 4 4 1 4 1 Joining Nodes 4 4 1 5 two parts of a mesh have been formed in different ways for example one generated and the other imported and the nodes at the interface do not exactly coincide pairs of nodes can be joined or made coincident To do this select the two nodes to be joined and then the Join icon The highest numbered node will be deleted and all references to it will be converted to reference the lower node number Note the existence of two nodes at one location can be checked by labelling the nodes in the area of interest Care should be taken to maintain 8 noded elements with midside nodes Safe can not accommodate other types of elements Node and Element Operations Clicking the right mouse button produces a popup menu which varies according to whether Nodes Elements or neither are currently selected Selecting nodes or elements 1 The program will detect whether you are clicking close to an individual node element or drawing a box around a group of nodes or e
143. l step through the runs in the current sequence current sequence is always shown at the right of the status bar at the bottom of the main window The leftmost button allows the user to change sequence Sequences are given letters and there is always a sequence containing all runs The example shown above would have 6 alternative sequences plus All Global Data Titles The titles are reproduced in the title block at the head of all printed information for the calculations The fields in the Titles view can be used to provide as many details as required to identify individual calculation runs The field for notes can be reproduced at the start of the tabular data output The box at the right of the Titles window can be used to display a picture beside the file titles To add a picture place an image on to the clipboard This must be in RGB Red Green Blue Bitmap Oasys Ltd 2009 35 Safe Oasys GEO Suite for Windows format Select the Paste button to place the image in the box The image is purely for use as a prompt on screen and can not be copied into the output data Care should be taken not to copy large bitmaps which can dramatically increase the size of the file To remove a bitmap select the Remove bution 4 3 2 Units This option allows the user to specify the units for entering the data and reporting the results of the calculations Conversion factor Displacement 1000
144. lements If there are no nodes or elements currently selected the following menu will be shown Erase generated mesh Erase all mesh data Import DAF Import Split crossing lines Highlight Erase Generated Mesh will erase the nodes and elements but keep the lines and nodes at ends of lines and areas After erasing the mesh the view will return to Sketch mode and will show the mesh generation data which can be edited until the required mesh is produced Selecting Erase All Mesh Data from the right click menu will erase all points lines and areas as well as the nodes and elements Import DXF and Import TXT allow mesh data to be imported from previously prepared files See Exporting and Importing Meshes for details Split crossing lines will split any lines which cross each other and add a node at their intersection This is also done when a mesh is generated but this explicit option is sometimes to be preferred as there are some circumstances in which line splitting will not work for example arcs intersecting other arcs or very complex sets of intersecting lines and arcs Highlight shows a sub menu with Edges Node and Element listed Highlight Edges is a useful tool as it will show any discontinuities in the mesh e g produced by incremental mesh generation or an error in element sculpting Oasys Ltd 2009 55 Safe Oasys GEO Suite for Windows 20 00 18 00 16 00 14 00
145. lobal Data General Data This will ensure a Event 0 is carried out in which the mesh is brought into equilibrium 4 7 1 2 Relaxation Factors In runs after Run 1 the General material properties table is amended to allow specification of stress relaxation G Helax n factors Tension For any material relaxation factors may be specified with different values for compression and tension These are applied to the existing principal effective stresses They specify the amount by which the stresses will reduce if there is no further strain Relaxation may cause the element to be out of equilibrium with its surroundings and the mesh will then react using the incremental stiffnesses as specified for the current run Oasys Ltd 2009 Input Data 86 2 ry gt J E d Creep i Relaxation at CI ant ee Lone strain 2 Method of Stress Relaxation Where no relaxation is to occur the relaxation factor is 0 If relaxation is used to model the change from a short term secant stiffness E1 to a long term secant stiffness E2 as shown in the above figure then the value of the relaxation factor should be E1 E2 E1 In the same run and generally for subsequent runs the value of the incremental stiffness must be changed from E1 to E2 This stress relaxation approach may be used to model creep With a little thought it could also be used to model processes such as increase of due to ageing in soft
146. lota lota must be kappa U Poisson s ratio D G Beta G Beta G controls effect on stiffness of overconsolidation Beta phi Beta phi controls effect on strength of overconsolidation Exponent n Stiffness is proportional to stress n Only n 1 is allowed Gvh Ghh Anisotropy of shear stiffness use 1 if in doubt Consolidation type One dimensional or isotropic Consolidation y coordinate Yc Consolidation y Gd At level y pre consolidation ov Yc y Gd Yc is roughly the maximum previous ground level causing overconsolidation Gd is the mean between Yc and y at the time of preconsolidation Model normally consolidated soil as very lightly overconsolidated Mu Parameter used in the 3D models to control the shape of the yield surface in the plane Mu has no effect in the plane strain models Iteration tolerance Oasys Ltd 2009 Detailed Processes and Models in SAFE 148 Tolerance as a proportion of soil shear strength eg 0 01 means 1 Maximum first strain increment Maximum size of first sub increment of strain used in deriving stress from strain In order to calculate stress increment accurately from strain increment the strain increment is divided into sub increments which start small and gradually increase in size The results are slightly sensitive to the magnitude of the strain sub increments allowed becoming more accurate as this is made smaller However use of large numbers
147. lways be such that there is an even number of elements along a line minimum 2 The specified number will be increased or the element edge length reduced so as to achieve this e f the node spacing parameter is set to varying the number of elements generated along line will be increased automatically as necessary in even increments to achieve a feasible Oasys Ltd 2009 133 Safe Oasys GEO Suite for Windows 7 3 mesh and to reduce large discrepancies in the mesh Note that this will not happen if the node spacing parameter is set to constant e The ratio between last and first step sizes parameter is ignored i e assumed to be 1 0 when the node spacing parameter is set to varying Gamma w Values Something like appears three times in data 1 Weight density of water in Consolidation data y used to convert pressure to metres head 2 Weight density of pore fluid in Materials General in case pore fluid is water could be air for example 3 Gradient of pore pressure in Pore pressures Y in case non hydrostatic these parameters have units KN nr which might be thought of as kPa m in the case of Ypg three values are equal for soil saturated with water in hydrostatic conditions Their more detailed use is noted below 1 Weight density of water in Consolidation data This 15 the density of water irrespective of th
148. meters for stiffness eg E v Oasys Ltd 2009 151 Safe Oasys GEO Suite for Windows together with total stress parameters for strength c If effective stress parameters c are used for strength the model must compute undrained pore pressure in order to determine effective stresses and hence to find the undrained strength This relies on a good computation of dilation during shearing which is difficult to achieve in stress strain modelling Simple models such as elastic Mohr Coulomb cannot produce a good enough prediction of dilation hence get incorrect pore pressures and incorrect undrained strengths this was called Method A in the Nicoll Highway Inquiry More sophisticated models such as BRICK in SAFE or the soft clay model in Plaxis give a better prediction of dilation pore pressure and hence undrained strength The available methods are summarised in the table below using the Method A to D nomenclature of the Nicoll Highway Inquiry Method C has been divided into two methods Cl and C2 The SAFE manual refers to a Total stress approach and an Effective stress approach These terms relate to the stiffnesses used in the approaches rather than to the strengths Hence in the table below Methods A B and D would be included as effective stress approaches while Methods C1 and C2 are total stress approaches in the terms of the SAFE manual Comparison of approaches None
149. native Thus if an iterative stress error lies within either tolerance it is accepted That is it is accepted if it lies beneath the envelope shown in the figure below Talerance Stress units Erie lope Proportion of strength specified vurth material propertie specified E 9 nennen nnne here gt Shear strength This is the user s estimate of the ratio of plastic strain magnitudes in the first increment of the current run to those in the last increment of the previous run Its use is only to speed convergence If in doubt use 0 0 This box should be checked if values of cumulative displacement and strain are to be re started from zero This option is greyed out for Run 1 This is an experimental facility which should only be used under supervision oeismic loading may be modelled by the pseudo static method in which constant vertical and horizontal accelerations are applied to all masses i e D Allembert forces These are taken to be specified proportions horizontally and vertically of the weight of each element The seismic coefficients kh and kv specify the constant accelerations to be applied to all masses in the mesh in a pseudo static analysis of seismic effects The values of the coefficients are proportions not percentages of the acceleration due to gravity g Their signs show the direction of the accelerations with the same sign convention as the coordinate system positive to
150. nd tan For Brick models of critical state shear strength Shear Mean For stress o2 61 02 Angle of dilation For strain sin 1 e1 62 e1 2 Shear Vol e1 2 1 2 Hor Vert For stress ox oy For strain x y Hor Vert For stress oy For strain y 6 2 4 4 Templates Templates can be used to save the settings for graphical views The view settings saved in a template are e Scale e Colours e Scale factor e Axis on or off e Selected parameter and component e Contour interval e Plot centre and zoom To save a template create the graphical view you require in terms of scale parameter and component contour interval scale factor etc then select View Save in template file Provide a file name File1 sft and place the file in the required folder All following graphics can be created with the template file settings by selecting View Graphical output and then retrieving the template file using Graphics Load template file 6 2 5 Graphs Graphs may be plotted for numbers and contours When any graph is plotted the values on the graph can be output to a tab separated file using File Export graph data This file can then be picked up in Excel for further manipulation and plotting Oasys Ltd 2009 Analysis and Results 128 6 2 5 1 Creating a Graph mm To create a graph select the Graph button This will open the graph options dialog Graph Op
151. nding soil element e The stresses in the reinforcement are smoothed between adjacent elements to give a continuous distribution of stress varying linearly across each element This is similar to the technique used in Safe to smooth pore water pressures e Strains in the reinforcement are then computed for the stresses Displacements of the reinforcement are not computed by the program but can be calculated by comparing the strains in the reinforcement with those in the corresponding soil elements Oasys Ltd 2009 Detailed Processes and Models in SAFE 146 7 8 4 7 9 7 9 1 Tips Where reinforcement comes to an end at the edge of a zone of reinforcement elements the stress in the reinforcement will drop to zero This will be the case even if the nodes at the edge of the reinforcement are restrained However the program will only impose the drop to zero stress in the reinforcement iteratively and convergence may be slow if the nodes are restrained Reinforcement should therefore be stopped one element short of boundaries which are restrained in the direction of the reinforcement Where the ends of reinforcement are embedded in stiff material such as a concrete wall Parameter 14 of the reinforcement should be set to a very high value eg 100 Steps should be taken to ensure that there is positive effective stress normal to the reinforcement The stresses and strains computed for the reinforcement will have
152. ng meshes and are described in following sections A context sensitive menu is available by right clicking within the drawing area Erase generated mesh Erase all mesh data Import Import Split crossing lines Highlight Most of these items are self explanatory Split crossing lines will operate on all the data and split any lines or arcs which cross each other although straight lines are split automatically as they are drawn Note Arc arc intersections are not accommodated Highlight has a sub menu containing Edges Node or Element Highlight Edges will draw a Oasys Ltd 2009 43 Safe Oasys GEO Suite for Windows 4 4 1 2 Click the gridlines at significant x or y coordinates e g at locations of vertical structures piles retaining walls or horizontal features stratum changes basement slab levels Fu Generating mesh data using Gridlines purple line around any unconnected element edges In normal circumstances these should only be around the edge of the problem Internal edges usually indicate a mesh discontinuity Highlight Node or Element lets you select a node or element number which is then marked with a purple cross This is a very quick method for regular meshes but can also be used as the basis for more complicated data button or choose Gridlines from the Gateway to open the Gridlines table Add T Safe2 Mesh Gridlines ee
153. nodes 2 Right Click to display the Nodes popup menu and select Seepage Restraints 3 Enter the required values in the boxes and click OK to accept or Cancel to quit without saving the new data Even if the node list contains midside nodes the seepage restraints will only be set at the corner nodes Phreatic nodes are marked with blue circles fixed head nodes with blue squares and fixed flow nodes with blue triangles Tabular Input Run Data Node Tables Seepage restraints Oasys Ltd 2009 Input Data 72 T damflow sfd DE Restraints heal fal fll ps oe Restraint Piezometric Outflow level type m o s E Ma flew 2897 Fixed head 321 Fisted head 368 Fixed head 430 Fixed head BUS Fined head 563 Fired head Phreatic Phreatic Phreatic Fhreatic Phreatic Phreatic Phreatic Fhreatic 3 LEN EN Em 3 io a fz d4 15 dB A Each node should be entered on a separate line on the table Midside nodes do not require seepage restraints as the solution process only considers corner nodes 4 5 5 Springs Springs connect nodes to a fixed world outside the mesh Springs connecting nodes to each other are not available Spring stiffnesses are quoted as components in the x y and z coordinate directions of the mesh in the following units For plane stress
154. ns As a rough guide the differential displacements between adjacent nodes should be roughly constant across the mesh e The problem layout should be planned to ensure that the entered or generated mesh will accommodate all the proposed Events Thus horizontal area or element boundaries should exist at the correct level for excavation or dewatering levels in future Events e Place lines at material boundaries excavation levels groundwater levels and edges of structures It can also be useful to place vertical lines around the area of most interest so that the mesh can be made increasingly coarse towards the lateral boundaries e ry to avoid placing internal nodes or the ends of internal elements close to area boundaries This can dramatically increase the number of elements generated e Adjacent areas should model the same common line s rather than different but coincident lines The Graphical Input view will not create coincident lines and will split existing lines if coincident but shorter or longer lines are drawn over them e Think about the number of elements you want along a line New lines will have the default number of segments 6 assigned regardless of line length You can specify a desired number of segment or step size element edge length Specifying a large step size around the outer edge of the problem and finer step size only where it is needed can work well e The actual element edge lengths generated along a line will a
155. ntorceme 74 74 Specify stress in pile or reinforcement eg jacking Zone 75 75 erative output of reinforcement displ and stresses Leave Apply to column blank to apply to ALL Clear All Cancel Once the analysis is complete the tabular and graphical output will show only the following results and numbers contours and vectors derived from these Nodes flow and piezometric head Element Gauss points seepage velocities hydraulic gradients piezometric head and level and pore pressure Total inflow and outflow is shown in the log file In addition flow across a list of nodes for example phreatic nodes can be obtained by defining those nodes as a List in the Graphical Input view then choosing the Tabulated Results option Oasys Ltd 2009 167 Safe Oasys GEO Suite for Windows 7 12 7 13 Lex Bowles 95a Zera min pvrp sfd Results Output Analysis cared out by Safe version 18 2 NODAL SEEPAGE RESULTS Event f increment T Runt Node Coordinates Nodal Flow Piezometric Head x Y m m z m 133 100 00 344 7E 6 0 0 145 a TOLBE 6 9065664 166 96 80 6 amp z4 194 NI 3909E 6 _ 3873 ecl 93 65 15 1617 253 B5 93 3255 eae 90 64 658 6790 341 59 16 42 aos 57 69 16 SXLBETY 451 ae6 z4d 55 3790 525 54 51 58z26 586 53 41 20 2965 az u z
156. nts node difference is zz element 1 Note Re numbering may be carried out repeatedly but should be completed before any other nodal data restraints loads etc are entered Restraints Loads Pressures and Springs Restraints Standard restraints can be easily applied by clicking the is button on the toolbar This will add pinned restraints to all nodes on the lower edge of the problem if this is horizontal and x restraints on the left and right edges of the problem if they are vertical Any node may be restrained in the x and or y directions to have zero or specified displacements For planar analyses at least three restraints are essential for equilibrium one horizontal and two vertical or vice versa For axisymmetric analyses the minimum requirement is one vertical restraint though it is sensible to provide a horizontal restraint on the axis of symmetry Boundaries of the mesh which form planes or axes of symmetry should be restrained perpendicular to the plane or axis Boundaries remote from the main action should generally be restrained in both directions representing the fact that soil subject to small strains is usually very stiff For Fourier problems with frequency greater than zero it is not essential to have a vertical restraint In addition the axis of the model should not be restrained horizontally since displacement need not be symmetric about this axis Any node may be restrained to have a specified d
157. nts in anti clockwise direction E Pie 1 2 3 4 5 6 F 8 eade efaults 1320 1352 1566 1477 1677 1601 1530 1404 1203 1253 1320 1404 1530 1348 1228 1203 940 1057 1203 1209 1228 1073 959 950 B3 fol 940 950 959 603 711 533 B4B 835 B3 533 711 B08 FAL 422 S09 63 63 63 63 63 63 Ddl CO 63 630 63 63 1 a 3 3 a 3 3 3 3 3 3 3 a 3 5 a 3 Enter the material zone number Far this element Highlight the material zone box for each element and add the number of the material taken from the list created in General Material Properties 4 11 Further Runs Runs and Events Once the Global data has been added and the information completed for Run 1 the remainder of the runs can be added This is done using the Runs and Events tree diagram and table in the Global Data menu Global Data Runs and Events 4 11 1 Runs and Events Option The Runs and Events menu option opens two windows e Runs and Events showing a tree diagram specifying the links between the different runs and events in the file e Events and Runs the tabulated equivalent plus a Sequences page which shows all alternative construction sequences Oasys Ltd 2009 99 Safe Oasys GEO Suite for Windows 4 11 1 1 Runs and Events Tree Diagram This option allows the various Events and Runs within the Safe file to be created and linked Mult runs
158. of overconsolidation ratios Oasys Ltd 2009 Detailed Processes and Models in SAFE 150 7 9 5 7 10 Test a was used to derive the parameter 1 iota and to set up the step wise S shaped curve and this was then checked for tests involving other stress paths Tests 0 were used to confirm 1 and set the parameter which governs the gain of stiffness with overconsolidation Tips Generally convergence is very good The only exceptions found so far have been where a big increase in effective stress occurs in a single increment In this case convergence is obtained by applying Test 24 with parameter value 10 which prevents over relaxation of the iterations BRICK References Houlsby G T and Wroth C P 1991 The variation of shear modulus of a clay with pressure and overconsolidation ratio Soils and Foundations 31 3 138 143 Simpson B 1992a Development and application of a new soil model for prediction of ground movements Proc Wroth Mem Symp Oxford Simpson B 1992b Retaining structures displacement and design 32nd Rankine Lecture Geotechnique 42 No 4 Undrained Modelling from a note by Dr Brian Simpson 30 Mar 2006 Background This note deals with the undrained deformation of saturated soils means that no pore fluid enters or leaves the soil during deformation If the soil is saturated throughout the deformation its volume can only change by a v
159. of small increments increases run time appreciably especially for soft clays in which strains near surface level may be very large Some experimentation may be necessary to check sensitivity to this parameter to assist this the same process can be modelled in the BRICK program using the Safe option for increment type In Safe the ratio between successive sub increments increases from 1 0 to 1 3 as the ratio of volumetric to shear strain reduces from infinity to zero String data This data module specifies the steps in which the S shaped curve is to be modelled as shown below The data required are the string lengths and the resulting ratio of tangent shear modulus to its maximum value Gmax ie at zero strain Each line of data specifies the bottom of a step as shown in below Plastic Strain String length Step height proportion af material oe strain Strain The s shaped curve represented in Step Wise fashion The string lengths must be given in order of increasing magnitude and the final value of G Gmax must be 0 Strain is specified in absolute units not or log Oasys Ltd 2009 149 Safe Oasys GEO Suite for Windows 7 9 3 7 9 4 The critical state value of for plane strain is computed from this data and displayed as data check For similar data BRICK also gives crit for triaxial compression and extension Example An example of the Brick model
160. of the available methods is completely straightforward All require some thought and understanding Method provides an approach which is easy to understand but may be difficult to set up in the data This adopts undrained strength c undrained stiffness parameters E and v and makes no pretence at computing pore pressure so avoiding any confusion about the computed values This approach 15 relatively easy to use provided the material zone can be represented by No water condition at initialisation in Run 1 In SAFE however if the material is represented as having water pressures in Run 1 these will be lost when it is changed to No water changing total stresses which will put the mesh out of equilibrium effective stresses are not changed by the change to No water Alternatively if the zone is initially set up as No water then K must be considered in terms of total stresses which is unusual A change from drained with water pressures to No water causes no problems if it involves a change to a fairly stiff elastic or strong material such as a change from soil to concrete for a retaining wall or footing In this case the stiff material readily takes up the out of equilibrium stresses caused by the absence of water pressure with negligible deformation However if the new No water material is deformable and or of limited strength deformation would be required to take up the stresses Method C2 may be
161. omputations for geotechnical problems Plane stress plane strain or axially symmetric problems can be analysed For linear elastic materials analysis of axisymmetric structures subject to general loading is also possible The deformation mode selected may imply significant approximations the implications of which must be carefully thought out It is sometimes possible to check the validity of approximations by preliminary runs of simple problems which form components of the main problem to be analysed For example the lateral deflection of a pile group might be studied by a plane stress analysis in plan Oasys Ltd 2009 Safe Oasys GEO Suite for Windows Note Different material types within the same problem can be allocated different modes of deformation These can also be changed at different stages in the computation but some limitations apply for an axisymmetric environment individual materials may be set to use the plane stress mode of deformation to eliminate the effect of hoop action Plane stress could possibly be used in a plane strain environment Any other combination should not be used Pore pressures and effective stresses can be identified separately Gravitational loads and initial stresses are also accounted for Non linear problems are dealt with by the initial stress technique which may be used incrementally or otherwise and the facility is also available for changing material types or eff
162. orces by 8 to get displacements mE Use and D to get and 6 fie gt apply corrections f c De Finished amp output 7 11 2 Steady State Seepage Runs Combination of undrained and steady state seepage in the same run from a note by Dr Brian Simpson 19 April 2006 In SAFE materials for which steady state seepage are required are specified as consolidating as Oasys Ltd 2009 157 Safe Oasys GEO Suite for Windows they would also be for a consolidation analysis However the run itself is specified as steady state As of 19 Apr 06 using undrained and consolidating materials in a steady state run had problems The reason for this is that undrained materials are also treated by SAFE as consolidating There is no distinction between the two in the computations except that for undrained zones the specified permeability 15 over written by a very low value e seepage boundary conditions eg fixed heads are not applied and e where an undrained zone interfaces with a drained zone the drained zone does not impose a seepage boundary condition on the undrained zone as it normally does for consolidating materials However this does not apply where an undrained zone interfaces with a consolidating zone including steady state There is always continuity of pore pressure between undrained and consolidating steady state zones the only discontinuity 15 with drained zones the key to Swit
163. ory BRICK model Simpson 1992a b Oasys Ltd 2009 135 Safe Oasys GEO Suite for Windows 7 41 Linear Elastic Materials Materials of the linear elastic may be cross anisotropic Energy considerations impose some restrictions on the combinations of parameters which may be used Pickering 1970 the most important of which are ES 2 Fi s Thus if E2 lt lt E1 use v12 O lt lt 1 22 zu 1 2 Hence for isotropic materials v 0 5 Note The notation used here for v12 is defined in Linear Elastic and differs from that used by Pickering 7 4 2 Water or Void Materials If a material of type Water Void is specified as weightless it will also have zero stresses representing a void If it is given weight it will have pore pressures only which will either be constant drained or determined by constant volume undrained requirements Nodes which are solely in void material are constrained not to move Dummy elastic stiffnesses are required for this material type These should be non zero but negligible compared with the stiffness of other elements 7 4 3 Elastic Mohr Coulomb Materials The stresses calculated from linear elastic assumptions are modified by the following iterative procedure a If a principal stress is less than C COt itis increased to be equal to C Cot Oasys Ltd 2009 Detailed Processes and Models in SAFE 136
164. pressure across the interface and in this case steep gradients of pore pressure occur in the first element into the undrained zone third column of elements from the left of the mesh The pore pressures have changed very little in the two left hand elements however which 15 correct compare Figure 2 This pattern 15 largely a function of mesh gradation undrained elements adjacent to the interface will always have sharp gradients b New Switch 94 In the second approach a new switch has been added to SAFE so that when steady state 1s declared a only analysis is carried out without any check on equilibrium This condition can already be declared in Global Data General but in that case it applies to all runs In a subsequent run the pore pressures computed for the steady state zone are retained by declaring the zone drained without any specification of pore pressures for the zone Oasys Ltd 2009 Detailed Processes and Models in SAFE 158 Figure 5 Run 8 shows the results for the steady state no equilibrium run In this run the correct steady state pore pressures are obtained e effective stresses are unchanged so total stresses change and are no longer in equilibrium This can be seen because the total vertical stress at the base of the mesh is now about 240kPa compared with 200kPa derived from the weight of the material see Figure 2 e displacements are not incremented Figur
165. ption can be selected to display the node properties dialog in which the coordinates desired element size and radius of influence can be edited Multiple nodes can be selected by drawing a box around them However the only option available for multiple node selection before mesh generation is to delete them by pressing the Delete key Note specifying a node in this way is likely to lead to a finer area of mesh around the node Points of this type are not required at the endpoints of every line endpoints are created automatically on drawing lines Generating the Mesh Click the Generate Mesh button ME to produce the mesh This will show the Mesh Generation dialog with defaults set to the most commonly used options Some less common options are available on Advanced dialogs Mesh Generation spacing x bo reduce t Constant 4 elem size difference Advanced Renumbering Renumber during mesh generation Node spacing Constant will generate nodes at equal distances from each other along each area boundary unless the line s Ratio setting is not equal to 1 Varying will generate nodes with a linear increase or Oasys Ltd 2009 Input Data 52 4 4 1 4 decrease of the step size depending on neighbouring element sizes This can lead to large numbers of elements Adjust segment This checks that short lines have fewer numbers of numbers segments than longer lines in the mes
166. r overlaid elements As the nodes of the element begin to displace no strain is recorded in the element until the displacements exceed the specified lack of fit Data The material parameters required for the Multi Linear model are as follows 1 6 E1 v13 E2 v12 G12 and a Elastic parameters as material types 1 and 3 same defaults for E2 v12 and G12 if E2 left zero or blank Emin E1 where Emin is the minimum Young s modulus allowed in setting up the stiffness matrix and E1 is parameter 1 Suggested value 0 01 8 Yo y coordinate for which elastic constants apply as type 3 9 D proportional gradient of stiffness as type 3 10 Tolerance for iterations absolute stress 11 12 x y components of lack of fit Units metres If Model a is required set Parameters 21 to 25 as follows 21 Any value gt 0 22 The lining compressive stress O 23 the value 0 24 The lining tensile stress O 25 0 or blank If Model b is required set Parameters 21 to 36 as follows 21 36 Odd strain values on envelope Oasys Ltd 2009 143 Safe Oasys GEO Suite for Windows Even stress values on envelope Compression positive limits are given first followed by tension negative In both cases the strains are given in order of increasing absolute magnitude 7 7 Modified Cam Clay The program uses a two dimensional version of Modified Cam clay based on Simpson 1973
167. r currently plotted in the Graphical Output view or displacement in the y direction if the Graphical Output view is not open Checking Multiple runs on same graph will plot the results for each new run on the same graph if you move between runs with a graph view already open Selecting Advanced Graph Options opens a second dialog Oasys Ltd 2009 129 Safe Oasys GEO Suite for Windows Advanced graph options Plot parameter versus Distance Time Increment Time increment plot options Plot far nade s 2 Plat far element s End run Graphs can be plotted against distance or for variables other than bending moment shear force and thrust for a single node or element all 4 Gauss points or a list of nodes or elements in terms of time or increment For plots against time or increment specify the element or node number s and the last run to be included in the plot If a zero is entered as the last run results for all runs are plotted The latter is useful for creating a quick overview of results Plotting against increment is often good for time dependent problems because it gives something similar to log time as the time increments gradually increase For graphs plotted against distance you can enter a run list and the results for each run along the specified axis will be added to the graph To do this click Plot for run list and enter the runs required separated by spac
168. r flow is active A long term stiffness for the diaphragm wall of 13 5GPa will be enforced to model concrete s reduction in stiffness in the long term Safe can model creep in an analysis using Relaxation Factors which will be applied to the concrete material type in the General Material Properties table from the Gateway The relaxation factors are calculated as follows Relaxation E1 E2 E1 Where E1 Short Term Stiffness E2 Long Term Stiffness When carrying out a consolidation analysis all consolidating materials must be set to Consolidating in the General Material Properties table and strength parameters should be set to long term effective stress parameters for the clays Note that this example is for illustrative purposes only and the limitations of the Mohr Coulomb material as a basis for consolidation stages outlined in Undrained Modelling should be noted Oasys Ltd 2009 19 Safe Oasys GEO Suite for Windows Tension Col D T I 0 Total ami mour escription m ae factors Comp Batante am EE Elastic STRAN Fill Elastic Isotropic PL STRM 7 Elastic Mohr Coulomb lsotropic PL STAN Cla Elastic Mohr Coulomb Isotropic PL S TAN Elastic Mohr Coulomb sotrapic PL STRM Excavation Vaid Water or Vaid Isotropic PL STAN Linear Elastic sotrapic FL
169. r seepage analyses the drainage condition should be set to Consolidating If the analysis is only to consider saturated flow then the minimum pore pressure should be set to zero and Switch 68 should be used in the analysis This causes the permeability of elements through which the phreatic surface passes to be scaled according to the proportion of the element which is saturated This feature is important when one of the purposes of the analysis is to establish the position of the phreatic surface or of any seepage faces To set switch 68 choose the Switches button in the Analysis dialog and tick switch 68 Oasys Ltd 2009 Detailed Processes and Models in SAFE 166 Test Switches st Description Select Apply to Value 61 Write material properties PROP ORT to log file B2 Print stresses and strains to log file from satsh 53 B3 Specify stresses strains etc in one element 64 64 energy absorption and store as z strain 55 B5 Difference between two runs BB Adjust v to equal y x tant Br 67 Set minimum Cu for use with switch 66 or when gradient B8 Scale permeability when saturation surface lies within ele B3 B3 Two Zones of reinforced soil First zone ends at element r 70 Fix vert eff stress in M C zone Other eff stresses unchan 71 71 Extra output in new REFSOL for one node T2 72 Use new REF SOL ma 73 Acceleration deceleration if 1 0 factor for rei
170. rained in y direction The coordinates of the points can be edited if Change this line only is selected this will create a new point at the revised coordinates but other lines which previously connected to this line will not be affected if Change all lines at this point is selected the existing point will be moved to the new location and all connected lines will remain attached to that point The subdivision type defines the method of spacing the nodes along each side of a subregion The subdivisions can be specified as a number of segments or a constant distance between corner nodes of elements No of segments The line or curve will be split into the number of segments given in the Number of divisions field or the next highest even number if the input number is odd Constant distance The mesh generator will attempt to put corner nodes at this distance apart provided the result is an even number of subdivisions If this is not possible the step size will be reduced to obtain an even number of subdivisions The of element sizes setting provides a means of controlling the spacing of the nodes If this is 1 0 the nodes will be equally spaced along the line If it is less than 1 the nodes will be closer spaced towards the end of the line If it is greater than 1 the nodes will be closer spaced towards the beginning of the line This setting will be ignored if the overall node spacing setting is set to varying
171. re divided into or by specifying the size of elements along particular geometry lines In this example the mesh will be refined three stages To define the first set select all of the geometry lines by clicking the Select Lines toolbar button A and drawing a box around all of the lines they should all turn orange in colour Right click anywhere in the Graphical Input window to display the following dialog Oasys Ltd 2009 Step by Step Guide 10 Line Properties multiple selection E x Subdivision type C No of segments Constant distance D of divisions 2 Distance 1 af element sizes Set ta gt 1 0 smaller elements required at START of line and 41 0 if smaller elements required at END of line Post Processing Uptions Restrained in direction Restrained in y direction For this example set the Subdivision to Constant distance and change distance to 2 Next select the geometry lines in the first area of the mesh to be refined as shown below 10 00 Again set the Subdivision Type to Constant distance but this time set the distance to 1 to refine the mesh in this area Further refine the area shown below to a constant distance of 0 5 Oasys Ltd 2009 11 Safe Oasys GEO Suite for Windows 10 00 5 000 10 00 15 00 20 00 30 00 35 00 40 00 45 Finally click
172. rial types It is therefore extremely important that correct permeability values are entered at this stage Seepage restraints are applied by selecting appropriate nodes right clicking and selecting Seepage Restraints from the dropdown menu The dialog shown below is displayed where various types of seepage restraint can be specified Seepage Restraints a No flow Fixed flow Fixed head Phreatic Outflow m2 s Piezometric level m 0 Oasys Ltd 2009 Step by Step Guide 20 First select all nodes along the horizontal surface in front of the wall and apply a Phreatic restraint Then select all nodes on the right hand side of the wall at y 2m and apply a fixed head restraint with a Piezometric Level of 2m The final task for this run is to make it a steady state analysis This is done by opening the Consolidation dialog from the Gateway checking the Steady State Analysis checkbox and clicking Apply P Excavation Example sfd Consolidation Data Loading Stepped Ramped Unit weight of water n 0 Timestep control Time at end of run daps 0 1 Constant increment days po m First increment days o mo BA inira increment 7 hd asm increment ee Specialist options Tolerable eff stress increment Stress unis Percent Tolerances Time increment Amplitu
173. s b Forcompressional stresses which lie outside the Mohr Coulomb failure envelope the shear stress is reduced and if dilation is specified the mean normal stress is increased until it lies on the boundary of the failure envelope T Oasys Ltd 2009 91 Safe Oasys GEO Suite for Windows Figure b Reduction of Shear Stress Dilation Figure c Increase of Mean Normal Stress These corrections to the stresses lead to non equilibrium with the surrounding stresses The errors are applied as nodal equivalent forces in the next iteration of the initial stress process To accelerate convergence over relaxation factors are applied to the components of stress error in co ordinates Notes e In the process of correcting the stresses principal directions are unchanged e Dilation can only be used if the elastic properties are isotropic 4 8 2 Linear Elastic The properties of the linear elastic materials including isotropic materials E First principal Young s Modulus representing stiffness with respect to direct strains parallel to the plane of a soil stratum e g a bedding plane Coto Fo E2 Second principal Young s Modulus for strains perpendicular to the soil stratum Default setting E1 V13 V12 NU13 Poisson s ratio such that NU12 Ast E1 v12 Ao2 E2 v13 Note that v13 v31 If the default setting for E2 i e E1 is used v12 is set equal to v
174. s Ltd 2009 41 Safe Oasys GEO Suite for Windows View settings Boundaries of wiew Drawing settings direction Fen width in pixels Min Grid interval Y direction Snap to grid Tolerance for node 0 05 coincidence Cursor made p Crosshairs C2 Arow only This allows setting of the view boundaries the grid snap interval and a tolerance within which nodes will be the background grid will be drawn to show this interval and if Snap to grid is checked lines drawn on the screen will snap to the nearest grid point The pen width of line drawing can be revised here Important if Object Snap is switched off the tolerance for node coincidence in the specified units for Length influences whether new nodes are created or whether the program snaps to an existing node on clicking nearby In general this should be set to a value related to the overall size of the problem domain It may need revision if more detailed mesh data is required in a particular area If you see unexpected results as you draw lines check the value of this parameter Buttons Display options IR Turns the ruler guide on or off RA Turns the background grid on or off Turns the display of Gridlines on or off Turns mesh detail on or off If on this will show the number of divisions along each line or the positions at which the mesh generator will try to put nodes s Turns di
175. s their main significance will be overruled by other time settings However they are also used for other minor purposes within the program so should be given sensible values Oasys Ltd 2009 Detailed Processes and Models in SAFE 156 The iterative process aims to give a solution within a small tolerance on pore pressures The program varies this tolerance setting it very small for early iterations but gradually larger as iterations proceed approaching values equal to the allowable changes in effective stress see above times a factor R where R is the maximum ratio of time increments between successive increments The default value is 1 0 If the allowable effective stress changes are fairly small e g 2kPa 196 R may be set to 1 0 However if a quick rough run is wanted it may be appropriate to set the maximum allowed changes in effective stress fairly large e g 10kPa 10 whilst still requiring tight convergence of each increment in terms of pore pressures This can be achieved by setting R 1 The flowchart below shows the process used in Safe to solve consolidation problems In the flowchart B is strain displacement matrix D is stress strain matrix S is force displacement matrix Time dependent Reddm behavieur Form element stiffness matrices B D BTD B vol Add the stiffness matrices togethe to get S Also apply loads magos o O Compute new pore pressures Multiply the f
176. sfd Runs and Events Event 1 Bun Event 2 Run 2 Inc loading by 100 Run 3 Inc loading by 50 Event 3 Run 4 Inc loading by 50 Delete reslis Cleanup deletes results from all increments except the last in the run Delete results deletes all the results of the currently selected run and its dependencies The current run is always shown highlighted To move to a different run double click on its title or use the icons on the toolbar to move to adjacent runs Adding a run Click the Add Run button Details for new run data from run Run title Change to long term drained Hew run Follows Hun Increment e The new run will be numbered automatically by the program By default it is assumed to follow the highest numbered existing run and to copy the data from the currently highlighted Oasys Ltd 2009 Input Data 100 run These options can be changed if required Note Following a Run from the highest numbered Event will create a new Event Alternative runs can be created for a particular Event by following a Run or Runs from a previous event more than once e Fill in the run title being as descriptive as possible as these titles are used in the page heading for printed output The title can be copied from a previous run using the Copy Title button e Add the run number and increment it is to follow e Click on the OK button Once a new run
177. sition of the phreatic surface e Problems where wave action generates pore pressures storm loading feature Interface features Alternative sequences of events can be modelled in the same data file Intermediate results can be stored and inspected before the problem is continued further Extensive graphical input options Built mesh generation facility or import from DXF other data sources Wide range of graphical output and printing facilities including contour plots and line graphs Oasys Ltd 2009 About Safe 2 1 3 of most parameters bending moment diagrams plots of principal stresses and percentage strength mobilised Components of the User Interface The principal components of Safe s user interface are the Gateway Table Views Graphical Input Graphical Output Graph views toolbars menus and input dialogs These are illustrated below SAFE 18 2 Eagle sfd Main toolbar Safe toolbar File Edit GlobalData RunData Analysis Graphics Combinations Queues Window Help Safe toolbar NET eo 1 Input Eagle sfd Elastic Mohr Coulomb material Table View View d BGA pul Gatewa Lc D S ghi il ss KPa al e Titles Run 2 Model existing struc Units and Preferences Event 3 5000 7500 General Run 3 Diaphragm wall insta Lists 1 Event 4 Runs Run 4 Dw 34 0 Events Event 5 Geometry Run5 Ex 25
178. splay of seepage restraints on or off iH Toggles between display of cumulative or run loads If cumulative this will show loading up to and including the currently displayed run 4 Change scale This shows dialog with options for user specified best fit or engineering scale Best fit is the default Oasys Ltd 2009 Input Data 42 Specify Scaling Select scaling C3 User specified Best fit C3 Engineering Independent XY scaling scale to 1 Set scale to 1 zea The x and y scales should be the same for drawing The User specified option can be used to set the window to the correct size if a bitmap is added behind the grid to aid drawing Sometimes it is useful to set independent x y scaling to select groups of nodes more easily Turns independent x y scaling on or off Decrease or increase the font size on the view Zoom options AJ x Clicking the zoom button allows drawing a zoom window using the left mouse button and dragging the mouse until the required area is within the box Zooming out will return to the previous scale Resizes the graphical display to include all current data After clicking this button the left mouse button will Pan the view rather than starting drawing of a zoom window The same functionality is available from the middle mouse button The other buttons are for adding or selecting drawn objects or other data or generating deleti
179. t Hor Vert Gauss point values only Major Principal Minor Principal Maj Prin Min Prin Incln Strength Mob Phi Mob Shear Mean Hor Vert Pore pressure Nodal flows Piezometric head x Strain Nodal or Gauss point values X y Xy Z Hoop 2D Volumetric 3D Volumetric Shear Modulus Resultant Shear Vol Hor Vert Gauss point values only Major Principal Minor Principal Maj Prin IncIn Min Prin Incln Angle of Dilation Hor Vert Errors Nodal or Gauss point values X y Xy Z Hoop 2D Mean 3D Mean Shear Modulus Resultant Gauss point values only Major Principal Minor Principal Maj Prin Incin Min Prin Incln Hor Vert Seepage velocities x y Resultant Hydraulic gradients Note Principal strains Arrows are not shown for tension Oasys Ltd 2009 Analysis and Results 126 6 2 4 3 1 6 2 4 3 2 Mesh Options none of the other option buttons vectors contours etc are selected the right click mouse menu will show the Mesh Options available w Full Zones Label nodes Label elements Full will show the element boundaries of the deformed mesh with Incremental or Cumulative displacements Zones will show the material zone boundaries of the deformed mesh with Incremental or Cumulative displacements The labelling options will all label subsequently selected nodes or elements with the node number element number or el
180. t be summed appropriately to give the overall results This is an advanced facility which must be used cautiously after studying Wilson 1965 or similar texts List of References List of References Bolton M D 1986 The strength and dilatancy of sands Geotechnique 36 No 1 pg 65 78 Milligan G W E 1983 Soil deformations near anchored sheet pile walls G otechnique 33 1 41 55 Naylor D J 1974 Stresses in nearly incompressible materials by finite elements with application to the calculation of excess pore pressures Int J Num Meth Eng 8 443 460 Naylor D J Pande G N Simpson B and Tabb R 1981 Finite elements in geological engineering Pineridge Press Pickering D J 1970 Anisotropic elastic parameters for soil Geotechnique 20 271 6 Potts D M amp Zdravkovic L 1999 Finite Element Analysis in Geotechnical Engineering Vol 1 Theory Vol 2 Application Thomas Telford Publishing Rowe P W 1962 The stress dilatancy relation for static equilibrium of an assembly of particles in contact Proc Roy Soc Series A 269 500 527 Simpson 1973 Finite elements applied to problems of plane strain deformation in soils Ph D thesis University of Cambridge Simpson B O Riordan N J and Croft D D 1979 A computer model for the behaviour of London Clay G otechnique 29 2 149 176 Smith I M amp Griffiths D V 1998 Programming the finite element method Wiley Wilson L 1965
181. t very small by the large bulk modulus of the water It is considered that this is a reasonable model of the process which occurs in the ground and maintains smoothly distributed pore pressures For 8 node elements with 4 Gauss points a further refinement is used This requires iteration of the initial stress type and has two stages 1 The pore pressure at each Gauss point is corrected such that if total stresses remained constant the incremental volumetric strains would all return to zero Oasys Ltd 2009 Detailed Processes and Models in SAFE 154 7 11 7 11 1 2 From each element pore pressures at its corner nodes are extrapolated from the Gauss points The several values of pore pressure for each node are then averaged New pore pressures for the Gauss points are then interpolated from the corner nodes The iteration converges very rapidly but operation 2 largely cancels out the effects of 1 The result is a set of pore pressures which are smoothly distributed and conform with overall equilibrium Mean volumetric strain in each element is very small but significant volumetric strains may occur at individual Gauss points especially in areas of high stress gradient The iterations required are concurrent with initial stress iterations for non linear materials and so add little to the cost of the computation The smoothing process is only used on connected undrained elements Drained elements are not affec
182. tand is specified as Ui in the porepressure table If a gradient of suction is to be specified then yw is given a value which is then multiplied by Ui and depth Linear increase of pore pressure with depth at gradient yw is automatically allowed for but may be overruled by specifying several pore pressures within the zone In this case an additional parameter Ri is needed for each zone Where only one point is specified per zone it is often easiest to set yi at the level of the phreatic surface where Ui O In this case Xi is assigned an arbitrary value Where more than one point is specified per zone then R becomes important and more thought should be given to choice of Xi and yi Ri is a sort of radius of influence for the data used if the program is required to interpolate between specified pore pressures in the same material zone using the method described in Interpolation of Pore Pressures Oasys Ltd 2009 Input Data 96 The following data are required for each pore pressure point Material Co ordinates Pressure Radius of influence select the material from the drop down list of material descriptions Enter the x y co ordinates of the pore pressure point Where only one point is specified per material it is often easiest to set y at the level of the phreatic surface where the pressure is zero In this case x is assigned an arbitrary value Where more than one point is specified per zone
183. te for Windows Linear Elastic Water or Vaid Elastic Mahr L aulomb Modified Lam clay Duncan amp Chang For detailed information see Material Models Deformation mode Plane strain plane stress axisymmetric see Analysis Methods Note The assignment of these deformation modes overrides the general calculation method defined in General Data Drainage condition Models can be created to simulate the terms of Total and Effective stress For information on how to use these models and specify the correct drained and undrained parameters see Modelling Total and Effective stresses Drained means constant pore pressure Undrained means that no volume change will occur and pore pressures are computed Consolidating means that time dependent consolidation is to be computed Convergence will be assigned to the columns of elements created by the program adjacent to a well It should not normally be selected by the user Well will be assigned to well elements created by the program It should not normally be selected by the user No water means that there is no water pressure and no flow of water Thus it could be used to model for example an impermeable concrete wall The behaviour is essentially drained i e volume change can occur but unlike other drained zones a no water zone does not interact with consolidation or seepage That is its nodes do not become fixed head nodes form
184. ted and there may be discontinuities of pore pressure between them The interactive tolerance on pore pressures is 10 times that on strength If all materials in the mesh are linear elastic interaction should be forced by setting one of the materials to type 3 with large and very small tolerance Consolidation and Seepage Consolidation analyses perform two sets of iterative calculations one for the soil displacements and stresses and the other for the pore pressures See Time dependent Consolidation for more details For information on consolidation data settings see Timestep Data If a consolidation analysis is required to be run until a long term steady state a steady state seepage run can be defined in a coupled stress and consolidation problem Some considerations for this type of analysis are discussed in Steady State Seepage Runs If only steady state seepage analysis is required without any consideration of material displacement the steady state seepage option in General Data should be selected The data input menus Gateway items and tabular and graphical output options are then filtered so that only the relevant information for steady state seepage is shown This information is summarised in Steady otate Seepage Timestep Data For each run in a time dependent consolidation analysis the target time to be achieved at the end of the timestep is required This is a cumulative time measured from the start of consolidation in
185. the Generate Areas toolbar button to create individual areas of the analysis geometry Mesh Generation To generate the mesh click the Generate Mesh for all Areas toolbar button La which will display the following dialog Renumbering is carried out automatically to reduce the bandwidth of the problem maximum node umber difference in an element The default values are acceptable for this analysis The following mesh should be generated comprising 1222 elements 10 00 30 00 35 00 40 00 45 ee LL HS HM nee HHH T P EIE LET LLL LEUTE L e CERES HE ERE HE RC ALII I1 44 dE dd EEA A The bandwidth can be checked by clicking the Renumber toolbar button and further renumbering can be done at the users discretion in this case it may be required for larger more complex meshes Oasys Ltd 2009 Step by Step Guide 12 3 2 Restraints For this example standard restraints can be applied by clicking the Restraints toolbar button de This will restrain the vertical edges of the mesh in the x the x and y directions direction and the bottom edge in Run 1 Initial Condition The following material data should be entered for Run 1 by opening the General dialog from the Materials section of the Gateway It is good practice
186. the effective stress line on the y axis For further information see Initial Ground Stress Initial Ground Stress Initial stresses in the ground may be specified directly by the user or may be computed by the program from the weight densities and distribution of the materials The initial ratio of horizontal to vertical effective stress is also required If all the initial stresses including water pressure increase linearly with depth within each material zone initial stresses can be specified directly This usually occurs when soils are horizontally layered and the ground surface and water table are also horizontal In more complicated cases the program can perform a special initialising stage Event 0 in order to ensure equilibrium before starting to accumulate displacements and strains The computed initial stresses will lie within the strength limits of the materials and will conform as closely to the specified as equilibrium will allow movements in Event 0 are set to zero before further analyses proceed Oasys Ltd 2009 85 Safe Oasys GEO Suite for Windows Level ga Effective vertical E stress ch 3 Ground level Note If the ground surface stratum interfaces and water table not all level equilibrium require that Ko and g are not constants within a zone In this case average values should be given for Ko and g and an equilibrium check should be requested in G
187. the previous scale Resizes the graphical display to include all current data E After clicking this button the left mouse button will Pan the view rather than starting drawing of azoom window The same functionality is available from the middle mouse bution Allows a bitmap image to be inserted behind the grid Saves the image as a jpeg or Windows metafile for pasting into other applications bmp Jpg wmf 6 2 4 2 View Settings The settings for the graphical output can be reached through Graphics Edit Settings or the toolbar ICON P View Settings Scale factor Show scale bars on arrow length mm 15 Plot multiple graphs on same axes fone Select Vectors Tolerance for graphs element size Contours Interval Maz of contours Number of intermediate points on contours Humber format of significant Figures Scale Factor Select maternal zones to display Material Fill dry Fill et Cet AS vet Grade Vw wet Grade Ill vet Oasys Ltd 2009 Analysis and Results 124 6 2 4 2 1 6 2 4 3 This is particularly useful when one zone is undergoing large deformation and the information is masking detail from other areas The scale factor for the deformed length is the ratio of the plotted length of 1 unit of the parameter to the plotted length of 1m of the mesh geometry For Prin
188. the results select Graphical Output from the View menu or the Display graphical results icon from the Safe toolbar Edit View GlobalData Analysis Graphics Combinations Queues Window Help Toolbars w Status w Gateway Tabulated Results Node and Element Tables Graphical Output The graphical view shown will be that of the currently selected run It is only possible to view one set of results at a time apart from adding run results to an existing graph Once the graphical view is open a Graphics menu is added to the main menu selections Edit settings Change values Change component Save in template File Load template On opening the Graphical Output view defaults to show the deformed mesh Selection of other options is done via the graphical toolbar the right mouse button and options on the View menu 6 2 41 Graphical Toolbar The graphical toolbar in the graphical output view frame controls how the results are displayed in terms of Vectors E Principals X Characteristics Oasys Ltd 2009 Analysis and Results 122 20 zen r1 stata E i en Numbers Line Contours Filled contours Toggle between Gauss point values and nodal values Toggle between averaged and un averaged nodal values Note Not all options are available for all results see Graphical Display Options Parameters Values and Components see Graphical Display Options can
189. the right and upwards The seismic forces are applied once at the start of the run for which they are specified They are not increased incrementally if there are several increments in the run As with other applied forces the seismic forces remain applied until they are removed in a later run this is achieved by applying reversed seismic forces at that stage Oasys Ltd 2009 Input Data 76 4 6 2 Time Dependent Consolidation Data Time dependent consolidation may be modelled in Safe for all material types The element types are restricted to 8 node quadrilaterals which may be curvilinear Pore water pressures are computed at the four corner nodes only and are interpolated from these to give water pressures at Gauss points Boundary conditions include fixed head fixed flow and phreatic nodes at which water may exit but not enter the mesh Within the elements a limit may be set on any suction which develops oince Safe is used primarily for computations involving non linear materials which require iterative processes an iterative technique is also used to solve the consolidation equations These two sets of iterations progress simultaneously The time increment to be used for each step of the consolidation process is selected by the program during the iterative process and this makes it possible to allow large time increments at stages of the computation where the pore pressure regime is changing only slowly
190. the series of runs it is not merely an increment of time for the current run In consolidation analyses each increment within the run represents an increment of time The user may specify the magnitude of this increment in which case no further data are required under Consolidation options However this approach is not recommended for most purposes Optionally the user may specify the magnitude of the first time increment in the run independent of other specifications By making it very small this could be used to give an effectively undrained increment at the start of the run If the magnitude of the first time increment is set the results will always be retained for post processing Regardless of how small the first time increment is constant volume behaviour will not be achieved in elements immediately adjacent to a boundary where there has been a step change in fixed head Pore pressures are assumed to vary almost linearly across the elements so a sudden change at a boundary inevitably causes a sudden change of pore pressure in the adjacent elements This effect may be reduced by using smaller elements If this happens when the program is free to choose the first time increment of the run it will tend to use progressively Oasys Ltd 2009 155 Safe Oasys GEO Suite for Windows smaller increments as it iterates trying in vain to reduce the stress changes caused by volume change This then means that the second
191. then interpolation between the points becomes important and more thought should be given to choice of x and y values Gauge pressure at the point in units of stress relative to atmospheric pressure If this is set to zero for a single point in a material a hydrostatic gradient will be created below a level surface corresponding to the level of the specified point Safe will assume that this remains constant until specified otherwise This value is used if the program is required to interpolate between specified pore pressures in the same material 4 9 1 Interpolation of Pore Pressures Within each material zone if a number of pore pressures are specified each x y point is assumed to be surrounded by squares of side twice the radius of influence for each point ARI Interpolation method for pore pressures Point X Y lies within one or more squares Pore pressure at is calculated as Oasys Ltd 2009 97 Safe Oasys GEO Suite for Windows i k xl If X y lies outside the square and smaller af and 5 xj y Y Wi 0 0001 ri When this facility is used it is essential to examine the output carefully to check that the distribution of pore pressure is as required 4 10 Element Material Zones The different materials specified in Materials are allocated to the elements in the mesh to create the ground model The groups of elements whi
192. ths E Ero 1 y E2 and G12 are found similarly but v13 and v12 are always constant with depth dE dy Etyo Where Evyo is E at yo D is defined so as to be positive when stiffness increases with depth for which dE dy lt 0 and is applied to E1 E2 and G12 i e add negative sign to entered data for a material whose stiffness decreases with depth Dilation constant such that the ratio of plastic volumetric strain increment to plastic shear strain increment is d compression positive i e add negative sign to entered data for a material which compresses during shear d SIN v where V is the angle of dilation positive values indicate cohesion increasing with depth generally kPa m Convergence tolerance for effective stresses in units of stress Note If values for E2 NU12 or G12 are not entered then they are assumed to be equal to E1 NU13 and E1 2 1 NU13 respectively Oasys Ltd 2009 Input Data 90 For model 3 stresses computed assuming linear behaviour which lie outside the Mohr Coulomb failure envelope are corrected in two ways to bring them back within the envelope These corrections are illustrated below a For tensile failure where the minor principal stress is less than C COt it is increased to be equal to C COt the maximum tensile stress the soil is able to sustain Figure Correction for Tensile Stres
193. tions Use a pre defined element group Advanced Graph Options Pre defined element group Select component Fromm list Definition of axis Enter nodes along required line to plot or a node list name 4275 129 Two nodes define a straight line More than two should be split into series of straight lines OR Specify an axis along which to plot The plotted Point X parameter can be changed once the graph is displayed 1 2 Manual select Multiple runs on same graph If element lists have been defined and you wish to plot bending moment shear force and thrust select the Pre defined element group option If this is selected a drop down box listing the element lists will be enabled To define the x y axis for the graph either enter 2 or more nodes along the required axis or enter two coordinates in the table Two node numbers define a straight line three or more will be split into either straight line or 3 node curved segments depending on the option selected in the adjacent dropdown box If the Graphical Output view is open you may also select Graphical Select and then OK Draw the required x y axis on the Graphical Output view For all types of axis graphs of Displacement Total Stress Effective Stress and Pore Pressure are available In addition if the Graphical Output view is open many other parameters can be shown on a graph The graph will default to show the paramete
194. to be applied In this case first click on the node on the right hand side of the top of the diaphragm wall then click on a node at approximately x 25m The pressure load should be shown as a series of orange coloured arrows SRR Oasys Ltd 2009 Step by Step Guide 18 3 6 3 7 Run 5 Excavate to 6m and place prop at 2m Start by adding the run and changing the material type of the appropriate elements to the Excavation void material to model the excavation to 6m The prop will be modelled as a spring element with a stiffness of 600 000 KN m per m run This is added by zooming to left hand side of the wall at approximately 2m A node should already be located at this position Click the Select Node toolbar button and click on the left hand wall node at 2m x 12 y 2 Next right click and select Springs from the dropdown menu Enter the stiffness in the x y direction textbox and click Spring Properties R Spring constants Inclination 0 direction 0 2 direction 0 Cancel In order for the spring to be included in the run it must be activated by opening the Active Springs dialog from the Gateway and setting the Active in this Run option to Yes Run 6 Consolidation steady state This run is intended to assess the long term behaviour of the excavation when the negative pore pressures dissipate and steady state groundwate
195. to define all soil types as Drained or Water when initialising and carrying out an equilibrium check even when undrained materials are to be modelled NENNEN e Colour otra EN Dram E Tow Tue Initial DE Param EX _ kN m Defaults H Elastic aE STAR water JF Elastic Mohr DCoulomb sotrapic PL STAN water 18 0 5 0 Elastic Mohr Coulomb sotropic FL STHMN D rained 20 1 1 6 Elastic Coulomb Isotropic PL S TAA Drained 21 1 5 1 091 Elastic Mohr Coulomb PL STRH Drained 2 1 5 1 031 The following data should be entered into the Permeabilities dialog from the Gateway This is important because later in the analysis groundwater flow and consolidation will be considered Ea x welt Bulk modul EE son s e condition pore Fluid Permeabilities Incl 1 ee 1 A EA E C a uen l Defaults water 2 2e 006 1 00 1000 00 1e 006 1 Fil No water 10 2264006 0 0001 0 00 0 00 0 00 0 00 0 00 0 2 Drained 10 22 0 0001 0 0001 0 00 0 00 0 00 0 00 0 00 0 3 Drained 10 2 2 1e 010 1 010 0 00 0 00 0 00 0 00 0 00 0 4 Drained 10 Z2e 006 1602 1 012 0 00 0 00 0 00 0 00 0 00 0 The following data should be entered into the Elastic Mohr Coulomb dialog from the Gateway Not
196. u nur atl tm sumus Uu cu UU 21 3 9 SUS 22 3 OE XAG 23 3 11 Creating graphs d 27 4 Input Data 32 4 1 Opening new or existing files REGES e ERES 32 1 2 Br u griueri gt 33 4 39 Global Data Ds ew PUN Dc dura cun Ru Dura Go b Qu uode e Ud d ud e cie i de dd rn 34 AS WUC clc Rua ELM EM ME 34 432 UMS IE DG at p testet e dv Dr EINE UEM 35 4 3 3 uocis ep dL I MEE EM E e 36 4347 General CIE Tni ole ir peer P CR US 37 4 3 9 Burns and Soo euet e disease eMe a PEEL M iE MADE aro E ELE EE 38 n dicen R M 40 4 4 1 40 Options ihe GrapniGal Pere epa 40 Generating data using Giidlin 2 c eee de 43 Generating mesi data Dy beta deae di tassa 45 Drawing LISS e a ee ey ee 45 canere 49 Creating atid Edino R
197. urrounding contours in soil elements Turning off the concrete material zone allows fuller inspection of the soil contours Plotting Queues The current view can be plotted for a queue of runs in the current data file To do this select Queues Set Queues and specify the runs increments to be queued After clicking OK on the Set Queue dialog the print dialog box will open allowing selection of the printer Graphical Display Options Safe can produce a large variety of parameters in graphical form These are accessed through a combination of the View menu commands and a subsidiary menu accessed from the right mouse button The following provides a table of the various options available View Change Parameter select View Change component values using right mouse Oasys Ltd 2009 125 Safe Oasys GEO Suite for Windows Mesh when no option button selected Vectors m Principals ES Characteristics patio Inc Cum K Numbers 2 Contours Cumulative Incremental Cumulative Incremental Cumulative Incremental Cumulative Incremental Ratio Inc Cum button Displacements Applied loads Residual forces Applied loads and residual forces Seepage velocities Hydraulic gradient Total stress Effective stress Strain Errors Displacement x y Resultant Total stress Nodal or Gauss point values Effective stress x Z Hoop 2D Mean 3D Mean Shear Modulus Resultan
198. ut are provided These can be selected by clicking on the tabs at the base of the page ystiess A Effective stress incremental stress incr ett stress Sirain incremental strain A Error Oasys Ltd 2009 119 Safe Oasys GEO Suite for Windows If all the tabs are not visible maximise the size of the Window or drag the end of the base scroll bar out of the way Tabs Element Gauss x coord y coord Pore pressure X y z Hoop Major Minor Angle Mean vol Shear Ratio User variable Stress Effective Incrementalncrementa Strain Incrementa Error Stress Stress Effective Strain Stress Number Point 1 to 4 x Coordinate of Gauss point y coordinate of Gauss point Pore pressure at this Gauss point Not applicable Component ox or amp x Component or y Shear component or Yxy For plane strain or stress z component gt or ez For radial symmetry Hoop circumferential component Major principal stress or strain Minor principal stress or strain The inclination of the major principal stress anticlockwise from the x axis Mean stress oy 2 or volumetric strain x 2 in the plane of deformation Shear Stress or Strain Ratio of shear to mean stress or volumetric to shear strain This allows a user defined value to be pasted in for each Gauss point These results can be plotted for comparison purposes The columns give the x
199. vant name for the list of elements The list will be saved and can be used later for printing results for that set of elements 4 4 1 6 Renumbering L Once the mesh has been generated then the option on the graphical toolbar to renumber becomes active This is used to reduce the bandwidth of the problem which can improve solution speed Mesh Renumbering Preference to minimize node numbers C in x direction direction distance from C2 to minimise bandwidth Iterations ao Preference to renumber elements Re numbering based on x y or radial distances from a node is only available if one node is selected The selected node which should be a corner node becomes the new node no 1 Renumbering to minimise bandwidth does not require a selected node Oasys Ltd 2009 Input Data 66 4 9 4 5 1 It can be worthwhile to have several attempts at numbering to get a convenient numbering scheme with the best bandwidth if this proves critical If prefer x is the most suitable option do prefer first then prefer x to get numbering primarily preferring x but secondarily preferring y Minimise bandwidth has the effect of reducing the bandwidth This can be used more than once to progressively reduce the bandwidth On the selection of the OK button Safe will carry out the renumbering and report the new mesh statistics Generation Statistics 1 133 nodes and 36 eleme
200. viour of soil which is clearly non elastic Two particular problems occur in the Duncan and Chang model a Poisson s ratio is constant and the ratio of bulk stiffness to shear stiffness is therefore constant This is clearly wrong when soil approaches shear failure for which shear stiffness becomes very small with no reduction in bulk stiffness Duncan and Chang suggest that an average Poisson s ratio of 0 4 or so might be used b The model does not recognise that shear stress sometimes changes with normal stress even in the absence of shear strain For example if mean normal stress reduces the soil may approach an active or passive state in which shear stress becomes proportional to normal stress This does not appear to occur in the original model so the soil could attain states outside the Mohr Coulomb envelope Oasys Ltd 2009 139 Safe Oasys GEO Suite for Windows not clear what happens as principal axes rotate The original model is only defined for laboratory triaxial tests The model in Duncan and Chang s paper has been implemented in Safe with the following changes a The model operates in plane strain instead of the triaxial axes they used Plane strain could be extended to include axial symmetry if necessary b All behaviour is assumed to be unloading reloading if it occurs at a stress ratio shear mean which is lower than one previously attained irrespective of principal directions
201. w appears left click to highlight the line 3 Press the delete button Graphical Input 1 Select the Pick Nodes tool and click on individual nodes or draw a box round a group of nodes 2 Right Click to display the Nodes popup menu and select Applied Displacements 3 Enter the required values in the boxes and click OK to accept or Cancel to quit without saving the new data Loads Loads may be applied as concentrated forces line or ring loads or as pressures normal and shear Pressures are applied along sides of elements Line Load Pressure Load All loads may be applied incrementally A 50 kN load applied in 5 increments will be subdivided into 10 kN increments Note This is a change from the data format of the DOS version of Safe in which a 5OkN load applied in 5 increments would have been specified as 10kN in the data file DOS data files will be re interpreted correctly Oasys Ltd 2009 69 Safe Oasys GEO Suite for Windows 4 5 3 1 4 5 3 2 When runs are created by copying data from previous runs the loading data from the original run will be copied across to the new run The data should be checked and edited as necessary to ensure that only the required loading is applied to the new run Line Loads Line or ring loads are applied at individual nodes in the x y or hoop directions The plane of loading is in the z direction which forms a line in plane stress and plane strain problems
202. wever the overall effect is made less severe as iterations proceed within each time increment Users must inspect the results for materials with values of F greater than 1 0 and satisfy themselves that they are appropriate for the problem Min pwp Minimum pore water pressure attainable by the material This is the pressure at which the material becomes unsaturated it will generally be zero or negative representing suction For unlimited suction leave blank Permeability factor If the pore pressure in the element tends to fall even lower than that specified above it will be held at the specified minimum and the permeability of the element will also be reduced by this user specified factor This represents conditions in a non saturated zone In simple cases the factor should be set very low the default is 10E 6 meaning that the permeability of the unsaturated material is one millionth of the saturated value Other values could be used given a proper understanding of unsaturated flow 4 8 Material Models Safe can accommodate the following material types e Linear elastic material with homogeneous stiffness which may be anisotropic e Water or Voids very small non zero effective stiffnesses should be set for these materials e Linear elastic material with a Mohr Coulomb strength criterion Stiffness may vary similarly with depth and optionally shear failure defined by a Mohr Coulomb cut off A constant Oasys Ltd 2009
203. witch 58 Drained Behaviour The material should be declared as drained and all stiffness and strength parameters given in terms of effective stress Pore Pressures Pore water pressures may be specified for each material zone for any stage of the computation If they are not specified for Event 1 they are assumed to be zero If they are not specified for later Events they are assumed to be unchanged from the previous Event Pore pressures will be changed by the program for undrained or consolidating materials When required the pore pressure Ui for each material zone should be specified for at least one point Xi yi Only one point is needed to specify hydrostatic conditions more points may be used to specify more complex conditions The point Xi yi need not necessarily lie inside the material zone Note that Ui is gauge pressure in units of stress relative to atmospheric pressure For hydrostatic conditions yw is set to 10 see General Material Properties The phreatic surface can then be specified as a single point under Pore Water Pressures where Ui 0 This will create a hydrostatic gradient below a level surface corresponding to the single specified point Safe will assume that this remains constant until specified otherwise Soil suction If soil suction is to be constant throughout the material then yw is set to zero within General Material Properties with Kw 0 5 The maximum soil suction that the soil can withs
204. ynamic graphics during analysis The analysis will not proceed if results are not available from the previous run Oasys Ltd 2009 115 Safe Oasys GEO Suite for Windows Analysis Progress Huns ta analyse Hun Information Advanced Options Hun 1 Curent run C Curent sequence C All runs Increment 1 af 1 Target time days Ma tests set Hun Progress teratian IFAIL v iteration Show displ Calculation Warnings Show detail Press OK to start the analysis The program carries out a basic data check that all elements have a material property assigned to them and that there are sufficient physical restraints defined before starting analysis The dialog will be continually updated to show the progress of the calculations The embedded graph can be switched to show displacement versus iteration by clicking the Show displ button The maximum iterations can be changed during analysis or the analysis paused and resumed later if analysing one run at a time Those options are not available if analysing multiple runs Run details and warning or error messages are also written to a log file with the filename format data filename rR sfl where R is the Run number For example the log file for run 2 of datafile test sfd will be testr2 sfl All results files and log files are written to a sub folder which has the same name as the data file prefixed by Safe For datafile
205. your printed output The bitmap will be fitted into space approximately 4 cm by 1 em but its aspect ratio will be maintained Select the company name that you would like to appear an your printed output Note For internal Arup versions of the program the bitmap option is not available Oasys Ltd 2009 37 Safe Oasys GEO Suite for Windows 4 3 4 General Data The following general data is required T Safe1 General Data Number of nodes per element Number of Gauss points per element Default deformation made Plane stress Plane strain CO Asisummetric Calculate initial insitu stresses by equilibrium check Event 0 Fourier problem Seepage analysis Number of nodes per element Number The Windows version of Safe uses an 8 noded element of Gauss points per element Default deformation mode Insitu stresses by equilibrium check with 4 or 9 Gauss points Specify the default deformation mode of Plane Strain Plane Stress or Axisymmetric This mode will be applied initially to all elements but individual material zones may be given a different mode at a later point in the data entry using the General Material Properties table see General Material Properties For example to prevent a retaining wall modelled as circular behaving as a very stiff ring in axisymmetric computations set the zone representing the wall to plane stress The only combination

Safe Oasys GEO Suite for Windows

Contents

Download Pdf Manuals

Related Search

Related Contents