Using slope stability software to determine internal forces

[RFreund]

Is it possible/practical to use slope stability software to determine internal forces on a wall?

One question that always bothers me is how to "accurately" account for offset surcharges and / or the influence of a tired wall upper wall on a lower wall when determining internal pressures/forces. Having said that...

Can you set up a slope stability model with your wall geometry, for this example lets say it is a vertical wall with tiebacks. Input your soils stratums and slopes. Now Enter horizontal forces at your tieback locations, iterate the force until you reach the desired FS (usually between 1 and 1.3 in US). These are then anchor forces due to soil pressure, you could then apply a surcharge and find your anchor forces due to the surcharge (by again iterating).
You could then assume a soil pressure distribution and back solve for an equivalent Ka for soil pressure and again backsolve for a Ka for the surcharge, right? In the case of a tieback wall you might assume a trapazoidal shape, but you could use this same method but now for a cantilever concrete retaining wall, right? Except for a cantilever wall, or really any "gravity" type wall you would apply your horizontal force (in the slope stability model) at 2/3H from the top of the wall. Again iterate the force and backsolve for Ka?

Am I losing it here or is this reasonable?

I suppose this question leads to why not design all walls using the mechanics of slope stability to determine internal / external pressures... Probably leads to unnecessary over complication...


EIT

http://www.HowToEngineer.com

 

[Doctormo]

Yes, you can use slope stability methods but I would suggest that the trial wedge method would be more appropriate for earth pressure forces for odd loading conditions. It can be done in a Rankine or Coulomb format so that that basic theories are still used and you can match the values for simple wall design conditions.

There are some people who want to do the analysis in a slope stability manner but it is not as simple as a couple slope stability runs and the results are different than conventional theory. The FS would still have to be 1.5 like it is for the conventional calculations not 1.3.

There is merit in this approach when the procedures are refined but it will probably require specialized software or methods to do the analysis properly, probably more like finite element analysis. It will also require some changes in the way stresses and pullout are looked at. I think it is probably ok for the external stability analysis for obtaining magnitude of thrust on a wall but using it for the internal design may be going step too far at the current time.

Pretty complicated subject to discuss in a forum without all the figures, diagrams, equations, etc.

 

[RFreund]

Doctormo - Great points. I suppose the trial wedge is more appropriate. Can the trial wedge be applied with different soil stratums? I suppose so, right? I'm just not sure how to program that.

Pretty complicated subject to discuss in a forum without all the figures, diagrams, equations, etc.

I mean no disrespect when I say this but that statement always tugs me the wrong way...Discussion is what forums are for... Now having enough time to present the information is a different story and can completely understand that. I suppose the two go hand-in-hand.

EIT

http://www.HowToEngineer.com

 

[dcarr82775]

I have done similar things. Use the slope stability to find the restraining force and then back out how you want that load applied (triangular, apparent, etc).

 

[Doctormo]

RFreund - trial wedge gets rather comlicated for multiple soil strata (similar to slope stability) whereas it is fairly straight forward for one set of retained soil parameters. It is complicated enough when modeling tension cracks. I think there is a fundamental problem when trying to get too precise with soils and lateral earth pressures as it is an imprecise science at best. I favor trial wedge as it is a geometric model that applies basic physics to the analysis which are hard to argue with and fairly rational even though the model is not totally correct (failure planes are not a straight line).

As far as my comment about complexity goes, forums are a great place to discuss concepts but they are not most people's full time job. I typically do not have the time to prepare exhibits and papers for extensive review and discussion which is sometimes required with certain topics. A picture is worth a thousand words as they say and I wish I had the time to do this full time, maybe when I retire....

Dcarr82775 - Yup, sounds like what I have done on occasion. The problem is what to do with the answer and how to apply it. For example, the thrust on a vertical wall and battered wall will be about the same via slope stability but a retaining wall analysis would say that the battered wall has less thrust on it. So what do you do with these two answers?

 

[dcarr82775]

I agree the tricky bit is what to do with the number, but that is why we get paid the big bucks right?

I prefer trial wedge when the problem lends itself to it. I trust the spreadsheet I wrote more then the program I did not.

 

[FixedEarth]

You can set up a spreadsheet with different columns and say 30 nodes to represent wall height. Column 1 will static earth pressure (triangular, trapezoid or rectangle, etc), you decide. The static pressure column will also account for backfill slope angle and any wall-soil friction. Column 2 will be dynamic earth pressure (inverted or right side down triangle), you decide. Column 3 through 5, could be point, line and strip load surcharge pressure) and column 6 will be the sum of all columns.

If you are comfortable with equivalent beam method and you have a beam analysis spreadsheet, you can solve for anchor force(s) by entering the sum column as a distributed load on the beam and anchor force will be a support. No iterations to worry about.

http://www.soilstructure.com/

 

[RFreund]

@FE - yes, I was just thinking of different ways to find the pressure for odd geometries/slopes. I do want to setup a calculation as you have noted, however my issue becomes the mathematics of find the reactions and moments when applying the varying distributed load to the beam. I will need to spend some time to solve this mater.

Thanks!

EIT

http://www.HowToEngineer.com