I am using Swedge to model kinematically-possible surface wedges on the foundation of an arch dam. An important force that will act on these wedges is due to the 240-ft reservoir behind the dam, creating tremendous uplift. How do I model this uplift using your program? As there is no specific option to account for hydrostatic head acting on the wedge, how can I "fake it"?
Use the External Force option in the Forces tab of the Input Data Dialog. With this you can define a single force and its direction equivalent to the hydrostatic force.
Swedge computes average water pressures assuming that the pressure is always zero at the free faces (per page 8 in the manual). In the event that a pit lake develops after closure, this is no longer the case. What recommendations would you make for analyzing a wedge that daylights below the surface of a pit lake that has inundated a portion of the pit slope?
Determine what the average pressure should be for your geometry (see answer to next question) then adjust the unit weight of water to simulate this. You could also turn off water pressure and use a single external load that's equal to the resultant water force.
You most likely have the water pressure turned on. Go to the Forces tab in the Deterministic Input Data dialog and uncheck the water forces. The water forces exceed the normal force caused by the wedge. This is why the factor of safety is 0 (no normal force).
Yes, the location of the tension crack can be user-defined, by specifying a trace length. The trace length is the distance of the tension crack from the crest, measured along the trace of plane 1 on the Upper Slope Face.
See the Overview of Swedge Input Data for a figure which illustrates the tension crack trace length.
In Swedge version 3.0, only the Active model is used, by default, for all bolts.
In Swedge version 4.0, you may specify either Active or Passive bolts.
For a discussion of Active and Passive bolt models, see the Help topic: How Bolts are Implemented in Swedge.
Also see Evert Hoek's "Rock Slope Engineering" (Appendix 1) for the equations used.
Here are Evert Hoek's comments on the use of an Active or Passive bolt model in Swedge. He made these comments after he was asked whether a Passive model in Swedge would be useful and how it should be implemented:
This is a limit equilibrium analysis and so, for the moment, forget about everything else and think about the limit state. At the point of failure a tensioned rockbolt or a grouted dowel will be tensioned to their limit and all forces will be mobilized - both the upward restraining force and the normal forces across the surfaces. At this stage the rockbolt or dowel does not care how it got there and so the pretension or lack thereof is not an issue. Now we go back and look at the loading history. For a pretensioned rockbolt it is true that the forces applied to the rock are those of the pretensioning load and both the upward restraining force and the normal forces are active. In the case of an untensioned grouted dowel or cable, the restraining force is mobilized once the block displaces but, depending upon the sequence of installation (it is quite common to pre-reinforce rock slopes with untensioned grouted dowels), this displacement could be very small and I see no reason why you would discount the normal force generating the frictional resistance. The only distinction that a purist could make is that, because an untensioned grouted dowel permits greater displacement, one may consider using the residual shear strength of the rock surfaces rather than the peak strength that could be assumed for a pre-tensioned rockbolt or cable.
In practice rockbolts come in 20 ton steps and, for larges slopes, cables tend to come in 50 or 100 ton steps. Hence, too refined an approach is redundant. I recommend that you stay with the original model and just put in the ultimate capacity of the reinforcement and allow it to activate both the upward resisting force and the normal forces across the planes. From a practical point of view, I do not see the need for different rockbolt and dowel models.
I think it's up to you whether you use the current bolt model or not, I would. The program's Infoviewer contains all the information necessary for you to do some simple hand calculations with whatever model you consider appropriate. I think you might find that with a passive model, where you add the bolt force to the restraining forces and possibly ignore the increase in frictional resistance, you will require more support $$$. You better be sure.