Graham Horgan looks at the ongoing revision to Section 8 of BS 8006 code of practice for reinforced soils and other fills.
The problems associated with building infrastructure embankments or development platforms over soft, weak compressible soils is not a new one and has challenged many generations of civil engineers around the world.
Traditionally, when the construction program permits, embankment construction is staged and includes monitoring of the dissipation of excess pore water pressures and settlement beneath the embankment.
With this information, an assessment of the relative strength gain of the sub-soils due to their ongoing consolidation can be made, and once sufficient strength is deemed to have occurred, the next stage of the embankment can be commenced.
The use of geosynthetic basal reinforcement can also provide additional tensile resistance at the base of the embankment in the shortterm, prior to the subsoil gaining sufficient strength. Additionally the inclusion of band drains and a basal drainage layer can also be included, all of which may speed up the primary consolidation process of the soft subsoil (Figure 1).
However, even after primary consolidation has finished and all the excess pore water pressure dissipated, secondary settlements can still be signifi cant, especially in soil types such as soft organics and alluvial deposits.
Nowadays, the demands of more rapid construction programs, (often driven by returns on initial capital expenditure), coupled with more stringent surface settlement performance criteria have increasingly led to alternative forms of embankment construction.
Coupled with increased time and performance pressures has been the transfer of risk on DBFO type contracts.
The responsibility for ongoing and post construction settlements has now generally been transferred from the client to the concessionaire or consortium with the long term maintenance liability.
For sites with short construction programmes or where the risk of secondary settlements would be too great, alternative options for embankment construction are desired.
Generally these alternative solutions fall in two categories; ground improvement of the soft soils beneath the embankment and additional structural support, in the form of piles, under the embankment.
These options allow the rapid construction of embankments to full height with limited and controlled post construction settlements.
In the UK, the design of embankment basal reinforcement and Basal Reinforced Platforms, (BRPs), over piles should be carried out in accordance with BS 8006: 1995, Code of practice for strengthened/reinforced soils and other fi ls, which is currently undergoing its 10 year review.
Designers have used the code successfully, both in the UK and abroad, on a large number of projects, for safe yet cost effective strengthened and reinforced fill structures. The code has however, received some criticism particularly in relation to the design of BRPs.
One of the goals of the current review is therefore to address some of these criticisms and hopefully provide clearer more concise guidance to help keep the code at the forefront of good practice worldwide.
With respect to BRPs one particular criticism relates to the degree of arching, or stress redistribution, within the embankment fill.
Due to the significant difference in the deformation characteristics that exists between the piles and the surrounding soft soil, the vertical stress distribution at the base of the piled embankments is non-uniform.
Soil arching between adjacent piles increases the stress acting on the pile caps and conversely reduces the stress acting across the geosynthetic reinforcement spanning between pile caps.
The greater the stress reduction, the lower the stress acting across the geosynthetic reinforcement and hence the lower the strength - and cost - of reinforcement required.
The British Standard uses Marston's ratio to determine the amount of arching within the embankment.
However, various authors have suggested several alternative arching theories over the years, and the merits of each have been discussed in the geotechnical fraternity.
Some authors have proposed stress reduction ratio to determine the efficacy of different arching theories.
This particular issue has been discussed within the pages of Ground Engineering, (GE November 1997), with subsequent discussions on the original article (GE June 1998).
An alternative method to that outlined in BS 8006 has historically been advocated by some manufacturers with respect to vertical load shedding or arching.
This has been promoted as an enhanced arching approach, sometimes referred to as the 'Guido' or 'Collin' method.
This approach results in very low stress reduction ratios, (less than 0.3), hence lower vertical stresses acting across the reinforcement, and typically produces a design of multiple layers of low strength bi-axial geogrids.
This alternative away from BS 8006 has been heavily criticised and a number of research projects have shown it can substantially underestimate the design strengths required in the geosynthetic reinforcement.
As with the current BS, the revision will still be unlikely to recognise the 'enhanced arching' methodology but is likely to allow other alternative arching theories to be considered. However, it may suggest a minimum stress reduction ratio be defi ned, (typically > 0.4), irrespective of the pile spacing and embankment height.
Further criticisms of the code have related to the lack of guidance on appropriate geosynthetic strain limits for the reinforcement spanning between piles, particularly with respect to shallow or low height embankments.
BS 8006 does give guidance on a 'practical upper limit of 6% strain' however it also warns 'with shallow embankments this upper strain limit may have to be reduced to prevent differential movement at the surface of the embankment' (Figure 2).
These proposed upper strain limits were perhaps more indicative of the polymers available to produce geosynthetic reinforcement at the time the code was written and published.
Polyester has been one of the main polymers traditionally used for geosynthetic reinforcement. Its relatively high initial stiffness and low creep characteristics made it a suitable choice for soil reinforcement geosynthetics.
But, recent advances in polymer technology have resulted in a greater choice of polymers suitable for the production of soil reinforcement geosynthetics. Aramid and Poly Vinyl Alcohol, (PVA), are two polymers that can now be used in their production.
These polymers exhibit a considerably stiffer response when compared to traditional polyester reinforcement (Figure 3). This is important for true material compatibility with the soil which it is reinforcing and can prove particularly cost effective where low strain reinforcement is required. This is often the situation when limiting the deection between widely spaced piles beneath shallow embankments.
It is interesting to note that the code gives no guidance in relation to lower geosynthetic strain limits nor does it provide the direct formulation to enable the magnitude of the maximum geosynthetic deection, (which occurs mid-pile span), to be determined.
The review of the code will aim to address these issues by including formulation to determine the maximum mid-span deection for a given geosynthetic strain, hence allowing the designer to design for an appropriate lower strain limit to satisfy the serviceability limit conditions for shallow embankments.
The revised code may therefore have to give advice to designers on the use and interpretation of isochronous curve data which illustrates the unique stress/strain relationship applicable to individual reinforcement polymers and products.
Another consideration with respect to thin embankments is that during the initial stages of embankment construction, the ground between the piles may offer some form of partial support and hence the full design load may not act across the reinforcement until the soil beneath it has settled slightly.
This loss of ground support can occur very quickly, (perhaps due to a rapid draw down of the water table in soft soils), or, more gradually as the initial increase in vertical stress and corresponding pore pressure dissipation results in consolidation settlement (Figure 4).
This effect can again have longterm serviceability issues particularly with shallow embankments.
The review is also likely to make some recommendations regarding construction issues relating to shallow embankments.
These recommendations would try to ensure that the tension and deformation in the geosynthetic reinforcement occurs during the construction stage, (rather than post construction), of the embankment. These measures could possibly include:
The option to pre-tension the reinforcement across the pile caps prior to placing embankment ll.
Thin fill layers placed immediately above the secured reinforcement layers, combined with heavy compaction.
Consideration of a thin layer of soft/loose compressible soil between the pile caps (Figure 5).
The use of dilatant granular fills within the 'zone of influence' to mitigate against differential settlement.
In addition to providing further guidance on anchorage and periphery detailing, the code will also try to encourage greater collaboration between the respective designers during the design process.
The current tendency is for the embankment's designer to optimise pile spacing with respect to the loadings imposed from the overlying embankment.
Generally the higher the embankment the closer the pile spacing and conversely the lower the embankment the wider the pile spacing, (assuming the same pile capacity).
Yet, an increased pile spacing is likely to result in higher geosynthetic reinforcement strengths and hence cost, and more importantly, could result in larger strains in the reinforcement spanning between the piles.
This in turn could result in excessive undulation developing at the surface of shallow embankments, which could ultimately affect the serviceability of the running surface.
By encouraging earlier collaboration between the main scheme designer and the specialists providing advice on the pile design and geosynthetic reinforcement, a more optimized, cost effective piled embankment solution will result.
(See Talking Point; GE September 2006).
The Federation of Piling Specialists (FPS) has raised another critisism of BS 8006. Speci. cally, the check on pile spacing considered within section 188.8.131.52 under the heading 'Pile group capacity'. This states 'the load carrying capacity of the pile group should be designed according to BS 8004, and should include any reduction in pile capacity due to group action.
If piles are to be installed on a square grid, the maximum pile spacing s is:
p is the allowable load carrying capacity of each pile in the pile group;
ffs is the partial factor for soil unit weight, (1.3 for ULS) . is the unit weight of the embankment . ll;
H is the height of the embankment;
f q is the partial load factor for external applied loads, (1.3 for ULS) ws is the external surcharge loading.
This check in BS 8006 has been the cause of debate with members of the FPS. In determining the allowable load-carrying capacity of a pile, Qp, factors of safety are already applied within BS 8004.
These consider both individual pile and group action. BS 8006 is then advocating applying an additional load factor of 1.3 to the dead and live loadings to determine the maximum pile spacing.
This effectively reduces the allowable pile spacing for a given pile capacity. Certain FPS members feel that these load factors, ffs and f q within BS 8006 should therefore be equal to 1.0.
Perhaps more importantly there is concern that a code of practice speci. c to reinforced soil design should be imposing prescriptive checks on areas, some FPS members consider to be outside the scope of BS 8006.
These additional load factors, and their validity, will also be considered as part of the BS 8006 review.
The final issue relates to the suitability of a particular geosynthetic reinforcement to ensure it is fit for purpose for its intended use, over the service life of the infrastructure.
Certain clients such as the Highways Agency (HA) insist that third party evaluation and certification, (British Board of Agrément, BBA), is a prerequisite for soil reinforcement products used for reinforced soil slopes and walls on the HA network.
Interestingly, no such criteria currently exist for the BRP reinforcement that may itself be supporting the reinforced soil wall or slopes above it.
Hopefully this anomaly will also be addressed within the revision.
In addition to the review of Section 8 of BS 8006, several other task groups have been formed to review all of the main sections of the document including the design of reinforced slopes, walls, and soil nailing.
It is expected that many new developments in these various . elds will be included in the revised Standard thus ensuring it remains at the forefront of good practice.
A draft for public comment is anticipated late 2006.
Graham Horgan is applications engineer at Huesker, International Geosynthetics Society, (IGS), representative on BS 526/4 committee reviewing BS 8006 and convenor of the task group reviewing Section 8.
1 British Standard Institution (1995). BS 8006.
Code of Practice for Strengthened/reinforced soils and other fills. British Standards Institute. London.
2 Russell, D. , Pierpoint, N.D. , (1997). An assessment of design methods for piled embankments.
Ground Engineering. November pp.39-44.
3 Tonks, D. , Hillier, R. (1998) Assessment revisited. Further discussion on 'An assessment of design methods for piled embankments' by D Russell and ND Pierpoint, Ground Engineering November 1997. Ground Engineering. June pp.46-50.