How Credible is Consoldation Settlement Calculation using Ch values. 3

[ONENGINEER]

I am doing settlement analysis and have Cv from Oedometer Test as well as Ch from CPT. The Cv obtained through Ch from CPT is noticeably different from Cv of Oedometer test. Are there any professionals who have practically derived Cv values from Ch values. Most of the research I have done shows papers that err on the academic side with the intention to promote CPT procedure. Calculation of the 90% time for settlement using Cv derived from Ch gives considerably shorter times (5 to 50 times shorter depending on the Kh/Kv value). That makes me hesitant to derive Cv from Ch instead of uemploying Cv from Oedometer test, although it would offer an excellent opportunity to decrease the tiome of settlement and avoid ground improvement options.

One concern is that the Ch was obtained under certain overburden pressure during the CPT operation. When we load the clay (change the effectice stress) one cannot guarantee that the Ch value will remain the same, can he/she? However during osedometer test we allow changes in the effectiive stress. Of course one may say that in Oed test we assume that the permeablity is constant throughout the test.

 

[moe333]

I prefer CPT based values since they better account for the macro-permeability of the soil profile as opposed to a small lab sample.

 

[ONENGINEER]

I agree moe333. However to derive the Cv values from Ch value it requires to know Kh/Kv. It is believed that the K of soil is one of the most inaccurate parameters that one can obtain in soil mechanics. It would be appreciated if someone advises what would be the Kh/Kv for e.g. clayey silts.

 

[moe333]

kv/kh
No macrofabric or slightly developed macrofabric (homogeneous deposit)0.67 to 1
Fairly well to well developed macrofabric (eg. sedimentary clays with discontinuous lenses and layers of more permeable material)0.25 to 0.5
Varved clays and other deposits containing embedded and more or less continuous permeable layers
0.07 to 0.33


Jamiolkowski, M. Ladd, C.C., Germaine, J.T. and Lancelotta, R. 1985. New developments in field and laboratory testing of soils. State-of-the-art report. Proceedings of the 11th International Conference on Soil Mechanics and Foundation Engineering, San Francisco, 1, pp.57-153, Balkema Pub., Rotterdam.

 

[ONENGINEER]

The clayey silt (by gradation testing or silty clay with PI classification) is near the coast in a deltaic marine evironment and is often underlain by sand and gravel. It is not varved but CPT indicates some sandy seems towards the bottom half of the clayey silt layer, which are likely to be continous. Could one consider it as a faily well to well developed macrofabric one

 

[BigHarvey]

If you are using preloading and vertical drains, CPT will give you Ch directly and it's the most important parameter ( vertical consolidation can be neglected in this case as far as time is concerned ). As you ponted it, the ratio Ch/Cv is not obvious to determine. Generally considered to be not less than 1.5, it's generally around 10 but can be exceptionally as high as 30. As a first approach the value of 10 should not be too far. Personally I use 10 unless the Cv value seems optimistic to me, in this case I reduce the coefficient.

 

[ONENGINEER]

BigHarvey, the 1.5, 10 or 30 would affect on time of consolidation greatly (20 times ? From the site specific data currently present, if a comparison is made between the Cv (obtained from oedometer test) and Ch, the Ch/Cv varies from 1 and 67. So I am trying to converge into a narrower range and stablish some optimum values.

One may assume that during the dessipation test in CPT operation water may travel vertically near the cone location and it would be desirable to know, if this is the case, how much it could affect the Ch values. Thank you for your time.

 

[moe333]

I agree with Big Harvey that 10 is generally a good number for the material desribed. Might want to bracket it between 5 and 15.

 

[BigHarvey]

The influence of k is not that big : for Cv = 10-8 m2/s and a square grid for the drains of 1 m with k = 5 you will get 88% consolidation in 3 mths and in 6 for k = 10. As I said, if you are designing a vertical drains network ( otherwise you don't care about Ch ) go for CPTU and you will get a direct value for Ch and it will be quicker than a consolidation test. I have never done the reverse way ( deriving Cv from a Ch value ).

 

[ONENGINEER]

BigHarvey: at this stage I am not designing a drain but validating my oedometer Cv data by the Ch values from CPT with the belief that oedometer results are conservative and do not take into account the soil fabric, while the Ch obtained from CPT are more accurate. Then the question becomes what would be the Ch/Cv or Kh/Kv values so that I estimate a new Cv from Ch and compare it with the oedometer test results. Thank you.

 

[BigHarvey]

In this case I would go for a ratio of 10 for Ch/Cv. It could be an intersting research program !