Protection Pliling to resist ground creep

[EireChch]

Hi all,

Our client is proposing to construct a new house on a sloping site. House is to be set back from a 20 degree slope by approximately 3m. We have a hand auger borehole drilled in the footprint. Loess soils (silty sand) to at least 4.5m depth. (Refer attached sketch).

I haven’t done a stability analysis yet. Given the dense soils and absence of any groundwater (or potentiall for saturation) I assume a FOS will be above 1.5. This will be confirmed.

Ground creep is common in Loess soils in our area. Typically to 1-1.5m depth. Given the dense soils we are assuming the upper 1m may be prone to creep. To mitigate this, it is proposed to incorporate protection piling on the downslope perimeter to 3m depth. Upslope piles will reduce in length to provide a smooth transition to shallow foundations.

The structural was querying me on what forces should he be designing for. I know (well at least I think I know) ground creep is a difficult thing to assess and my only be 50mm of movement over 10 years for example. Would it be inappropriate to provide the structural engineer with an amount of displacment so he can design is piles to be suitably stiff to resist the bending moment of 50mm horizontal displacement over 1m height.

From discussion with some engineers in our office I understand we normally recommend the structural engineer to design the pile as a laterally loaded pile i.e. the pile is retaining 1m of soil on the upslope side while there is no passive resistance on the downslope side to 1m depth. The soils from 1m depth to 3m depth would be providing passive resistance to resist movement. The piles would be spaced at 3D centres with soil arching effect supporting the soils between piles.

What are your thoughts on this? Any help is greatly appreciated.

 

[tsek]

1. What are the loads (both axial and lateral) exerted on the top of the pile from the house?

2. I think designing as a retaining wall is too conservative. I suggest instead to analyze using LPile using the sloping ground parameter. The equivalent slope would probably be less than 20 degrees (on the order of 10 to 15 degrees) due to the offset from the top of the hill. However, if you want to be more conservative, you may use 20 degrees as the slope. You may also consider the top of the pile to be 1m above grade. You should ask the structural engineer how much deflection the house can take before it becomes unserviceable.

3. When using LPile, remember that you have to use a P-multiplier. If you only have one row of piles (or the next row of piles is at least 6.5D away), and they are spaced 3D apart within the row, you may use Pm = 0.64*3^0.34 = 0.93. If the next row of piles is less than 6.5D away, you must also factor the P-multiplier by [0.26Ln(S/D)+0.5] for the first row, 0.52Ln(S/D) for the second row, and [0.60Ln(S/D)-0.25] for any other row.

4. In addition to the lateral deflection, the house would also require protection against settlement. The first 1m of soil is susceptible to settlement and therefore downdrag (negative skin friction). You must rely on the underlying soils to keep your pile from settling.

These are just my 2 cents.
Kevin

 

[msquared48]

No sketch attached...

Mike McCann, PE, SE (WA)

 

[EireChch]

Sorry for the late response. Sketch now attached.

Thanks Kevin for your input. The dwelling is realtively light weight, its a two story strucutre, weatherboard cladding with corrugated tiled roof. I would expect looads on the perimeter footing to be <5kPa. I am not sure on the lateral loads, for this analysis i think i would just be assuming the lateral load is due to creep movement, unless you recommend i should be accounting for lateral loads from the structure, wind loading maybe?

Unfortunatly I am not farmiliar with the L Pile software. I have heard about it and read threads on this forum of what it can do. Given it is a residential dwelling, i feel that designing the pile as if it is retaining 1m of soil woudl be ok. 3m deep piles at ~1.8 c/c woudl only result in approximately 7 piles. Wont exactly break the bank!?

I have spoken to my reviewer and he was happy that the downslope periemter pile did not need to be specifically designed. He envisages a 3m deep 300mm dia RC piles at ~1.8c/c are sufficient. I agree that specific design is not reuqired. For 1m retained height a ~1.5m embedment woudl be sufficient and we are providing 2m embedment. The soils are very stiff as you can see from the shear vane readings.

I would be interested if any one could provide some guidance on a simple hand calc method to check the loads on the pile for 1m of retained height. I know how to calculate the load on the pile for 1m of soil (i.e. 0.5 * Ko * Gamma soil * H^2), however i am unsure as to what effect spacing has on this. A laterally loaded pile typically retains approximatley 4D of soils from its centre point, as i have shown on my sketch. How do i account for the reduction in load on each pile if spacing is reduced to say 2D where the 'retaining bulbs' over lap.