Horizontal seismic loads

[adidasmayo]

My question is when do we use the horizontal seismic load for base shear in a slope stability problem and when do we not. I believe using it generally decreases your factor of safety. There is an option in SLOPE/W that asks whether I want to use it or not. Please respond asap!!

 

[adidasmayo]

I think I got it. We use kh for a pseudo static analysis for short term stability and we don't use it for a long term analysis. Correct?

 

[adidasmayo]

Anyone?

 

[FixedEarth]

mnoorzay;

Long term and short term analysis is for static factor of safety. It is based on the shear strength of soil when material is undrained (short term) and drained (long term). A F.S. of 1.5 or greater is considered adequate for the static conditions.

Kh is the horizontal acceleration component, which is 1/3rd to 1/2 of the Peak Ground Acceleration. This kh is only used for dynamic or pseudostatic analysis. A F.S. of 1.1 or greater is considered adequate for the dynamic ( psedostatic) conditions. There is no short term or long term associated with dynamic safety factor. It is shortly after an earthquake event.

 

[Mccoy]

Actually, a seismic event, or the significant part of it takes place in a short or relatively short time.

In the long term, dynamic stresses no longer exist, although pore pressure will build up up to a few days after the event, depending upon the soil permeability.

As far as i know the suggested practice (see the Duncan and the Towatha books) is to calculate FS in dynamic conditions with Kh and no overpressures and Fs in static conditions with no Kh but with post-seismic hydraulic overpressures.

 

[dgillette]

Hello McCoy. I think you are saying what I am thinking, but let me make a small clarification: The analysis of stability during the earthquake should be made with the monotonic peak UNDRAINED shear strength, i.e., the resistance that must be overcome for the slope to yield and deform significantly under rapid loading. If the slope involves normally consolidated to lightly overconsolidated material, there will be some effect of excess pore pressure in the monotonic strength, like any other clay.

We also insist that the slope needs to be stable immediately after the earthquake. If there is a strong reduction in strength due to liquefaction, clay sensitivity, or migration of pore pressure from looser material, we would analyze with the reduced strength to verify that it is stable until all the excess pore pressure can dissipate.

Best regards,
Dave

 

[Mccoy]

Thanks dgillette, that was well put.

One aspect which I would like to underline, even because I've had some undecision about it in the past, is that to apply indiscriminately both effects together in NC clays (the inertial seismic effect-kh component- and strenght reduction caused by the excess pore pressure) is usually regarded as overconservative, barred particular cases.

Also, the term 'immediately' after the earthquake is to be meant in relative terms, it may involve seconds-minutes timespans up to a few weeks to dissipate excess pore pressures in low permeability clays. This latter case is particularly interesting and was observed in Italy during the 1980 'Irpinia' earthquake, when some landslides occurred many days after the mainshock and with no concurrent aftershock.

 

[KVgeo]

Just like everything we do a pseudostatic (or should it be pseudodynamic) analyses are a tool and the key is interpreting the results.

Some will tell you that a pseudostatic analysis is a good indicator of potential seismic damage to a structure. A high factor of safety will mean that the structure can withstand seismic loads without significant damage. A low FOS indicates some potential damage may take place.

Some will tell you that the analysis is a nonsense. During an earthquake event the acceleration is very rarely (if ever)aligned in a the same horizontal direction along the failure plane at any given time. A dynamic analysis is required to understand the strains that may occur under seismic loading conditions.

To remain consistent with previous advice I have given on this forum, if you don't understand the fundamentals behind the analysis you shouldn't be doing the analysis without senior review and input.

Other than that, my advice is to try and work out what you want to perform the analysis for. Why are you performing a pseudostatic analysis? What are you interpreting the analysis for?

Then review the literature and see if the analysis is appropriate for your purposes. The literature will also provide guidelines on appropriate seismic load coefficients and strength reduction factors for low permeable and granular materials.

 

[adidasmayo]

Ok so the question is that when should we use the kh to lower the factor of safety and when can it be applied to increase the factor of safety? And what type of material should it be used on?

 

[dgillette]

Back to you mnoorzay, it would be easier to answer your question if you would tell us what it is you are working on. As your question has been presented, it's like if I were to ask you "When should I use a hammer?" It would be much easier for you to answer my question if it was "I am building a wooden chair. How should I fasten the pieces together?"

Pseudostatic analysis with Kh (and properly selected material properties) is for deciding whether a slope would deform during the earthquake. Deformation analysis is for determining whether it moves enough to be a problem.

(KVgeo - "pseudodynamic" probably makes more sense semantically.

 

[Mccoy]

My question is when do we use the horizontal seismic load for base shear in a slope stability problem and when do we not

Excuse me if the answer to your original question may seem dumb, but I would use the horizontal seismic load when you are designing in a seismic area. I cannot think of instances when the application of it increases stability, so in seismic areas it's the most unfavourable condition, other parameters remaining unchanged. Other parameters may change though, hence Dgilette's very legitimate remarks on what exactly are your conditions and purposes.
Often the question is : 'Which value of kh to use', which is not granted at all, fixed earth has given a range but research shows those values are often overestimates, so regulations and good practice govern if you have no other technical sensible clues.
You do not use as a rule a kh when you are calculating slope stability in post-seismic conditions, that is: excess overpressure (again, it depends on soil conditions) and no earthquake stresses.

 

[KVgeo]

@dgillette

"When should I use a hammer" = priceless!

Does anybody else worry about the structure of questions on this board sometimes?

@mnoorzay

Mccoy answered your question. Horizontal seismic loads are typically applied in the direction of the slide. As such, they tend to decrease overall FOS. At its simplest the analysis assumes that the earthquake gives a horizontal push on the slide elements.

Did you perform a literature review yet? Because these questions should have been answered by the literature. You can even do a google search for "pseudo static slope stability" which should give you some preliminary reading material. Then you can go to your textbooks.

Why would you rely on the advice from some faceless people from the internet with god knows what kind of qualifications when the questions you are asking are answered in detail in published papers and text books.

For a decent paper check out:

US Corp of Engineers (1984) Rationalising the seismic coefficient method

The question you should be asking is: Is a pseudo static analysis appropriate for my structure? Will it give me the tools I need to make an engineering judgement regarding the seismic stability of the structure? I think that would be an interesting discussion.

 

[ItsOnlyDirt]

Guys,

All great posts but I think you have all missed the key question mnoorzay was asking.

The option in Slope/W to turn off the base shear, and when to use it.

I was looking at this issue a couple of weeks back and it threw me initially. What the option tries to model is a kind of 'half way house' between undrained conditions and drained conditions.

If you select the option to keep the shear strength unaltered due to seismic forces, Slope/W does an analysis first without seismic to establish the available shear strenght anlong the failure surface. The shear strength is then converted to an equivalent cohesion, (or undrained strength), and thus is not affected by normal stress), and the run is repeated using the pseudo static loading. It usually bumps up the FoS by a small amount.

The synopsis however, for me, is why bother !, do a proper dynamic analysis if you are worried about it.

My two penneth

 

[ItsOnlyDirt]

Oh I forgot

KVGeo and DGillette, you should check this paper out.

http://www.igu.in/10-4/1deepankar.pdf

It may help your respective semantics

Cheers

 

[dgillette]

"Pseudo-dynamic" seems like an odd choice of terms. That looks to me like an actual dynamic analysis, albeit a simplified one.

 

[ItsOnlyDirt]

Ha ha, I guess its always 'pseudo' unless its a real time experiment