net load and settlement effect after site cut 1

[namron2]

Hoping someone will be able to help me clarify some soil behavoir.

I have a green field site to be developed with a treatment plant which includes tank structures up to ~70m in diameter and approx 5-7m high.

There is a site cut to create a level bench for construction of the tanks, which will be up to 8m depth.

The founding soil is an overconsolidated clay which is >30m in depth.

The site cut will significantly reduce the net load of the tanks, sometimes resulting in a no net increase in load case (although not always as the cut tapers across the site).

What is the logical process to follow regarding predicting settlements?

I am doubtful that after the cut is complete and the tank loading applied that a no net load can be taken as a no nett settlement.

I take it the soil will heave or rebound elastically somewhat on unloading once bulk earthworks are completed.

Should I then be applying a settlement calculation that uses the consolidation ratio in the pc range (overconsolidated) and applying the full structural load of the tank to the soil mass?

Or is there some effect in the swell range of the consolidation curve that I should consider that will reduce the settlement ?

Can someone enlighten me

Thanks

 

[fattdad]

Look at each of the following:

1) Elastic compression (i.e., based on load increase and soil modulus)

2) Use Cr and calculate the consolidation

Take the greatest value.

Just somethings to consider.

f-d

¡papá gordo ain’t no madre flaca!

 

[namron2]

thanks fot that.

Its basically where I am at at present, calculating the settlement using the consolidation ratio from oedometer testing and applied tank load (ignoring any soil removal from cutting).

I figuring that the excavated soil surcharge that was removed is only good for adding to determination of preconsolidation pressure. (i.e. ensures we dont slip over onto the virgin compression part of the consolidation curve, which will result in normally consolidated behaviour and a higher compression ratio).

I was curious to make sure I am on the right track or have I missed something that may lead to reduced settlement behaviour due to soil removal after site cutting.

 

[Mickney]

Thought in the dark here. Obtain an "undisturbed" sample of the clay in the zone of influence and set the sample up in a 1-d consolidometer. Run a consolidation test where you stress the sample up to the in-situ overburden pressure and then unload the sample back to zero. Then re-load the sample to several times over the pressure that will be applied when the tank is constructed. This may provide you with information on how much the soil will heave and then recompress.

 

[BigH]

Look up the idea of buoyancy rafts. If your net loading is at or very near zero - and you have no uplift pressures on the base of the tank, you will have very little settlement. Since your clay is overconsolidated - and I assume that this means OCR >2. You will excavate - this will move you back on the recompression curve (a bit of heave) - the Recompression ratio is usually in the order of 1/10 of the compression ratio. You will then construct the tank and load it. As this load is the same as the soil removed, you get no net loading. Swelling of a overconsolidated soil and its recompression to the level it was will have, in my view, such minor movements that you will not need to worry about it - so long as the clay is not a swelling clay (i.e., with montmorillionite or other swelling clay minerals). See Tomlinson's book on Foudnation Design and Construction for how to look at buoyancy rafts.

 

[Mickney]

Another thought that I had concerning the tanks:

Will these tanks have the same fluid level at all times? i.e. you may have significant stress changes if the fluid level fluctuates substantially.

 

[namron2]

Thanks for all the posts so far, they have been most helpful.

The soil is a residual clay of basalt origin. The liquid limit of this clay is typically around 80-100% and field moisture content is around 40-50%. Plots a CH to MH clay/silt around the atterberg A line.

The clay is considered to contain a fair proportion of montmorillionite (swelling clay).

For one large rigid rectangular slab of 'icecube' shaped clarifier tanks. I have elected to replace some of the clay with granular engineered fill under this structure to reduce possible seasonal differential movements.

Given the site cutting, my first thoughts were of relatively low net loads. I initially thought that the seasonal surface movement would be my primary concern for structures at this site, but then had trouble convincing myself of using the net load case for settlement calcs.

Big H, thanks for your input, I shall seek out Tomlinsons advice. Just to clarify, are you saying that in a large cutting (i.e. not a confined basement situation), it would be practical to say take the soil weight away from the strucutural load (i.e. to yield a net load case) and then do a settlement calc using this as the applied structural load? If I adopted this (which was my initial thought) many of my strucutures would only exert marginal net loadings. However as the cut is uneven the net load case at opposite edges would differ due to the tapering cut across the site. All of this was beginning play on my nerves a bit so I chose a conservative case of using an overconsolidated CR for settlement calc and ignored the site cut. Thought I needed to get my head right about this one though for the next time I run into such a problem.

thanks again

 

[fattdad]

For an overconsolidated clay above the water table (and considering that this is residual soil), I'd look at this soley from the elastic compression perspective. Net loading would not be a factor, rather the soil modulus and the loading to depth would effect a settlement response.

Soil modulus will vary with depth (confining pressure) and can be modeled using hyperbolic parameters, but for a starting point you can just use a Bousinesq stress distribution and Es based on 11N to estimate to what extent this may be a factor to your overall performance.

I don't think of this in terms of 1d consolidation.

Hope this helps.

f-d

¡papá gordo ain’t no madre flaca!