Software for rigid inclusions / settlement reducing pile design 2

[geotechguy1]

Just curious if anyone know of any software packages specific for design of rigid inclusions or settlement reducing piles (i.e. piles not connected directly to the slab / foundation but sitting under a gravel load transfer platform). As opposed to doing it with a general-purpose FE program or hand calc / spreadsheet

 

[DanielSgeo]

Check out geogridbridge from I think Virginia tech/vdot

 

[BNGeo]

What aspect of the design are you referring to? The elements themselves or the footing-pier interaction? In my experience the rigid inclusions foe geotechnical design are designed using conventional pile design (ie for settlement calculate settlement based on an equivalent sized footing transferred to a depth of 2/3 the length down the RI group for friction elements, etc).

 

[geotechguy1]

The piled part of it is easy, the harder part is justifying a settlement reduction.

The main issue I have is the whole system - designing the LTP and setting up the thickness of the LTP when a concrete slab is on top.

Almost all of the published literature is primarily about piled embankments (ala VirginaTech / geogridbridge). With piled embankments you need a minimum embankment height to avoid differential settlement at the surface (assuming you've got a pavement on top) that is much larger than the LTP. ASIRI is the only reference that considers explicitly the case of a slab or footing sitting on the LTP instead of an embankment; they seem to use 0.4-0.6m thick LTP and include some vague statements about the slab contributing to the arching and forcing equivalent settlement (i.e. preventing the surface from becoming all bumpy as it would in a piled embankment with H < Hcrit) but there's not really a way to calculate it.

 

[iandig]

There are a couple of options which you can consider, dependent upon what is trying to be achieved and what is being supported.
Option 1: The piles are there to support the element above, and any settlement is limited to the contribution from piles, not the surrounding ground. For the load transfer platform, I would be including high strength geosynthetic reinforcement layer(s) and then follow BS 8006-1:2010+A1:2016 and BS EN 14475:2006 (I'm based in UK). This would then be a basally reinforced platform, with the thickness being a function of the clear span between piles/pile caps. This would typically be in the order of platform thickness = 0.7 x span. This would mean the whole of the load above would be transferred to the piles, and if the ground settled between the rigid inclusions, then the combination of geogrid and platform thickness would transfer load to the piles (in theory a void could form below the lower layer of reinforcement without affecting the ground above the platform).
Option 2: The piles are there to reduce the total contribution of settlement but will settle along with the ground. Similar to use of VSC / VCC / CMC (Priebe design approach, improvement factor etc) where the higher load required to cause the rigid inclusion to settle, reduces load on the surrounding ground and then the overall settlement is reduced. In this instance the platform performance is difficult to quantify, so I would generally go back to first principles and run this through Plaxis FEM 2D and run a sensitivity analysis on column spacing and load transfer platform thickness. In practice, the optimum platform thickness tends to be in the order of 0.6m with a balance on column spacing. Whilst not necessarily required in the FEM analysis, I would also specify an appropriate geosynthetic reinforcement within the LTP to maintain its continuity following column/pile installation and re-working of the LTP before the next phase of construction (this is of more importance with VSC and CMC, but the principle still stands)

 

[geotechguy1]

Interesting, thanks for the feedback Ian. Yes, designing as a basally reinforced embankment with a structure on top was one method I had considered (using either BS8006 or the German EBGEO or the Dutch CUR226 or French ASIRI National project or one of the Virgina Tech / DOT research projects (CGPR #77).

Where I get hung up is that in the embankment approach, you have some critical height of fill thickness above the LTP where the settlement at the surface is 'equal'. (i.e., it might be that the LTP needs to be 0.6m thick but the total embankment height is 1.5 to 2.0m to end up with 'equal' settlement at the surface and end up with an egg carton effect - at least with a pavement. With a concrete slab / structure on top I assume the slab would probably force the settlement to be equal by re-distributing the loads to the more rigid areas of the ground...but not sure how to model this or prove this in the absence of FE. The ASIRI national project seems to be the only one of the piled embankment research programs to have considered a concrete slab or footing from a structure on an LTP directly.

Having said that, as I've been doing this I've realised the concept is not actually any different from older concepts like 'creep piles' or 'settlement reducing piles' - except for the LTP between the foundation and the ground.