Liquefaction Analysis

[moe333]

When performing a liquefaction analysis with simplified methods (NCEER etc.) using either SPT or CPT data, the analysis generally assumes free field conditions, ie. it does not take into account chanaged loading from a foundation or from a cut or fill over the ground in which the insitu testing was performed. Does anyone have any good references on how to model the changed loads using the simplified methods?

I know the loading would change the SPT or CPT data due to the change in confining pressures.

Thanks

 

[damguy]

I haven't really seen any textbook approach to doing this. It should be done however. The following link is to a software company (don't have anything to do with it)that developed a computer program for liquefaction analysis. Their manual includes a section on how this program addresses this problem.

http://www.civiltechsoftware.com/downloads/li_manu.pdf

 

[BigH]

Good question - wouldn't you take the "new" in situ stresses due to the structures into account? Adjust your N values accordingly to see if the soils are liquefiable or not. I'll have to give some more thought to this.

 

[jdmm]

If it is simply the effect of a change in area load from say a cut or fill then you can adjust the effective stress of the ground before running your analysis of liquefaction. Generally, increasing the effective stress on the ground will decrease the susceptibility to liquefaction. If you introduce shear stresses there is a correction for sloping ground but to model a building foundation or a dam you need numerical modeling such as the program FLAC to account for these details.

 

[moe333]

Thanks all for the responses.

In my case, I have a relatively heavily loaded mat foundation (about 4,000 psf). I can't apply this load uniformly to come up with new effective stresses because the load distribution changes with depth as well as from the center to the edge of the slab. I guess I can look at different cases varying the load for edge/slab/depth.

JDMM: I can't think of a reason why I would need to use FLAC or some other numerical model if I can vary the effective stresses as discussed above.

Thanks again