![Pile design in weak carbonate rocks – τmax = f(σc)
11
0
100
200
300
400
500
600
700
800
900
1000
0 1 2 3 4 5 6 7 8 9 10
AASHTO LRFD
0
100
200
300
400
500
600
700
800
900
1000
0 1 2 3 4 5 6 7 8 9 10
Design Value for Burj Khalifa
Mall of the Emirates Dubai
Pentominium Tower Dubai
AASHTO LRFD
0
100
200
300
400
500
600
700
800
900
1000
0 1 2 3 4 5 6 7 8 9 10
Design Value for Burj Khalifa
Mall of the Emirates Dubai
Pentominium Tower Dubai
AASHTO LRFD
Zhang and Einstein (Rough)
Rowe and Armitage (Rough)
0
100
200
300
400
500
600
700
800
900
1000
0 1 2 3 4 5 6 7 8 9 10
Design Value for Burj Khalifa
Mall of the Emirates Dubai
Pentominium Tower Dubai
AASHTO LRFD
Abbs and Needham
Zhang and Einstein (Rough)
Rowe and Armitage (Rough)
RQD<50%
RQD=100%
Smooth
RegularRough
UltimateUnitSkinFrictionτmax[kPa]
Unconfined Compressive Strength σc [MPa]
Design values improved by
preliminary pile testing](https://image.slidesharecdn.com/eddfcae7-3ec3-49ce-8ac4-210f84547cee-160127050819/85/2-Benoit-Latapie-Foundations-for-infrastructure-projects-in-MENA-11-320.jpg)
![Osterberg Cell (O-Cell) test – Typical results
Load [MN]
Displacement[mm]
• Independent measurements of side shear and end bearing
• The test results are easier to incorporate in the design
• Helps identify improper construction techniques
• Conventional reaction system not required
19](https://image.slidesharecdn.com/eddfcae7-3ec3-49ce-8ac4-210f84547cee-160127050819/85/2-Benoit-Latapie-Foundations-for-infrastructure-projects-in-MENA-19-320.jpg)
2. Benoit Latapie - Foundations for infrastructure projects in MENA
- 1. Foundations for infrastructure projects in MENA an approach to reduce risk and construction costs Geotechnica ME - 4th & 5th December 2013 Benoît Latapie Senior Geotechnical Engineer WS Atkins & Partners Overseas
- 2. Foundations for infrastructure projects in MENA Table of contents 2 • Some critical geotechnical issues in MENA • Common geotechnical approach in MENA • An approach to reduce risk and construction costs • Benefit of GI specification and supervision by specialist consultant • Limitations of AASHTO for pile design in MENA • Pile design in weak carbonate rocks – τmax = f(σc) & cost savings • Can we account for pile end bearing resistance? • The importance of preliminary pile testing for design • Osterberg Cell (O-Cell) test to avoid overdesign • BS EN 1997-1 & Pile foundations • Pile design to EC7 and τmax = f(σc) – cost savings
- 3. Some critical geotechnical issues in MENA Geotechnical Investigation Practices: • Clients often see GI as mandatory without appreciating its benefits • Quality of investigation is very variable • GI interpretation (GIR) is often produced by GI contractors • GI contractors often provide the only supervision themselves • GI standards and methods used on projects are often very low Geotechnical Design issues: • Contract specifications often demand use a mixture of codes • Approval bodies often misunderstand the difference between soil and rock • Approval bodies staff often is not specialised in geotechnical engineering • Geotechnical design is often done by the GI contractor GI = Ground Investigation 3
- 4. Common geotechnical approach in MENA 4 • Early stage project involvement • Desk study • Design/scoping and specification of high quality GI • Carry out GI • Full time GI supervision • Factual reporting by contractor • Review of factual reporting • Interpretation • Ground model + parameters + design STANDALONE GROUND INVESTIGATION CONTRACTOR CONSERVATIVE / UN-ECONOMIC / RISK ADVERSE SOLUTION
- 5. An approach to reduce risk and construction costs 5 • Early stage project involvement • Desk study • Design/scoping and specification of high quality GI • Carry out GI • Full time GI supervision • Factual reporting by contractor • Review of factual reporting • Interpretation • Ground model + parameters + design SPECIALISTGEOTECHNICAL CONSULTANT ECONOMIC RISK CONTROLLED FOUNDATION SOLUTION By GI Contractor
- 6. • D&B contracts are becoming more common • Field data used in back-analysis to validate geotechnical parameters • GI contractors are getting more experienced and invest in newer/better plant • The benefits of GI supervision by a specialist consultant is more accepted and even mandated in some cases • Clients are now better informed and have more experienced staff • Unforeseen ground conditions are getting recognised as a latent condition MENA geotechnical practices are getting better 6
- 7. Benefit of GI specification and supervision by specialist consultant 7 Weak to medium strong, off-white to pale yellow Limestone with inclusions of silt/clay Non intact core recovered as medium to coarse gravel of weak dolomitic limestone Conventional single tube core barrel Rotary coring with double tube core barrel (and plastic lining) Two sites a few kilometres away from each other, at similar depth
- 8. Benefit of specialist consultant in GI specification and supervision 8 Weak to medium strong, off-white to pale yellow Limestone with inclusions of silt/clay Non intact core recovered as medium to coarse gravel of weak dolomitic limestone Conventional single tube core barrel Rotary coring with double tube core barrel (and plastic lining) Rock modelled as soil Conservative Design Pile design often to AASHTO Rock modelled as rock Cost Efficient Design Pile design for Carbonate rock Low strength/stiffness High strength/stiffness
- 9. Benefit of specialist consultant in GI specification and supervision 9
- 10. • Based on a 35 years old correlation • It does not take account of the recent research and fully instrumented pile load tests • Not suitable for carbonate rocks that are predominant in MENA • AASHTO τmax correlation it is the most conservative published method when compared with correlations for carbonate rocks, • Gives artificially low τmax for RQD<50% 10 Table 10.4.6.5-1—Estimation of Em Based on RQD (after O’Neill and Reese, 1999) Table 10.8.3.5.4b-1—Estimation of αE (O’Neill and Reese, 1999) Pile design in weak carbonate rocks The limitations of AASHTO for pile design in MENA
- 11. Pile design in weak carbonate rocks – τmax = f(σc) 11 0 100 200 300 400 500 600 700 800 900 1000 0 1 2 3 4 5 6 7 8 9 10 AASHTO LRFD 0 100 200 300 400 500 600 700 800 900 1000 0 1 2 3 4 5 6 7 8 9 10 Design Value for Burj Khalifa Mall of the Emirates Dubai Pentominium Tower Dubai AASHTO LRFD 0 100 200 300 400 500 600 700 800 900 1000 0 1 2 3 4 5 6 7 8 9 10 Design Value for Burj Khalifa Mall of the Emirates Dubai Pentominium Tower Dubai AASHTO LRFD Zhang and Einstein (Rough) Rowe and Armitage (Rough) 0 100 200 300 400 500 600 700 800 900 1000 0 1 2 3 4 5 6 7 8 9 10 Design Value for Burj Khalifa Mall of the Emirates Dubai Pentominium Tower Dubai AASHTO LRFD Abbs and Needham Zhang and Einstein (Rough) Rowe and Armitage (Rough) RQD<50% RQD=100% Smooth RegularRough UltimateUnitSkinFrictionτmax[kPa] Unconfined Compressive Strength σc [MPa] Design values improved by preliminary pile testing
- 12. Pile design in weak carbonate rocks - τmax=f(σc) Year Name Recommended τmax Comments 1985 Abbs and Needham 0.375*σc σc<1MPa 0.375+0.1875*(σc-1) σc=1-3MPa 0.750 σc>3MPa Weak carbonate rock Calcarenite / Calcisiltite 1987 Rowe and Armitage 0.45*(σc)0.5 Regular 0.60*(σc)0.5 Rough Based on large number of field tests on weak rocks with no open discontinuities 1997 Zhang & Einstein 0.40*(σc)0.5 Smooth 0.80*(σc)0.5 Rough Recommendation base on a review of numerous available relationships 2010 AASHTO 0.65*αE*pa*(σc/pa)0.5 Based on Horvath & Kenney, 1979 for Shale and Mudstone Legend: σc Unconfined Compressive Strength (UCS) αE Reduction factor to account for jointing in rock pa Atmospheric Pressure (0.101 MPa) 12
- 13. Pile design in weak carbonate rocks = cost savings Soil Rock UCS=2.0MPa Conventional AASHTO Carbonate rocks method confirmed by preliminary testing Skin friction ignored End bearing ignored Socket Length 10m Socket Length 4m 131kPa<τmax<161kPa τmax=350kPa FULS=4,400kN FULS=4,400kN Ø=1.0m Ø=1.0m Concrete saving ≈4.7m3 per pile 13
- 14. Can we account for end bearing resistance? 14 • YES! • There are many methods available to estimate the end bearing resistance of a pile socketed into weak rock, for example: • AASHTO LRFD: qp = 2.5*σc • Zhang & Einstein, 1998: qp = (3.0 to 6.6)*(σc)0.5 For σc = 2.0MPa 4,200kPa < qp < 9,300kPa Supplementary ultimate load bearing capacity for a 1.0m diameter pile: 3,300kN to 7,300kN Same order of magnitude than skin friction contribution
- 15. Can we account for end bearing resistance? 15 • Necessary precautions to account for end bearing AASHTO LRFD No base cleaning = GAP End bearing ignored Base cleaning = NO GAP End bearing considered Often, programme constraints do not allow this procedure ACCURATE SKIN FRICTION ESTIMATION IS PARAMOUNT
- 16. The importance of preliminary pile testing for design Conventional approach in MENA Verification pile test & Construction Conservative skin friction estimate Conservative pile design Conservative design is built No contingency measures if the verification test fails Specialist Consultant Low quality GI Preliminary pile test to failure (O-Cell) Realistic skin friction estimate Refined skin friction & design update Cost-efficient design is built Design assumptions and installation methods verified before construction Construction 16 High quality GI
- 17. Pile testing solutions Monodirectionaltest Bi-directionaltest Conventional Test Osterberg (O-Cell) Test 17
- 18. Pile testing solutions Monodirectionaltest Bi-directionaltest Conventional Test Osterberg (O-Cell) Test 18 Reinforcement cage O-Cell
- 19. Osterberg Cell (O-Cell) test – Typical results Load [MN] Displacement[mm] • Independent measurements of side shear and end bearing • The test results are easier to incorporate in the design • Helps identify improper construction techniques • Conventional reaction system not required 19
- 20. 20 Osterberg Cell (O-Cell) test to avoid overdesign Ratio of Measured to Estimated ultimate loads – M/E M/E Ultimate capacity not reached Low result due to poor construction M/E=2.0
- 21. 21 Osterberg Cell (O-Cell) test to avoid overdesign Ratio of Measured to Estimated ultimate loads – M/E M/E Ultimate capacity not reached Low result due to poor construction M/E=2.0 COST/MN TEST LOAD – MN
- 22. 22 BS EN 1997-1 & Pile foundations Eurocode 7 prescribes the following (Clause 7.6.2.1): “To demonstrate that the pile foundation will support the design load with adequate safety against compressive failure, the following inequality shall be satisfied for all ultimate limit state load cases and load combinations:” Fc;d ≤ Rc;d Where: Fc;d is the design axial compression load on a pile or a group of piles Rc;d is the design value of Rc, the compressive resistance of the ground against a pile, at the ultimate limit state ULS shaft resistance ULS base resistance γRd = Model Factor Shaft Factor Base Factor The design resistance in compression is given by:
- 23. 23 BS EN 1997-1 & Pile foundations For the design of pile foundations, the UK National Annex to BS EN 1997-1 proposes different set of partial factors that relate to the amount of in situ pile testing carried out. Clause A.3.3.2 recommends the following model factors: • Without preliminary load test γRd = 1.4 • With preliminary load test γRd = 1.2 “If serviceability is verified by load tests (preliminary and/or working) carried out on more than 1% of the constructed piles to loads not less than 1.5 times the representative loads for which they are designed, the resistance factors can be reduced:” For bored piles γs γb • Without testing 1.6 2.0 • With testing 1.4 1.7
- 24. 24 BS EN 1997-1 & Pile foundations in MENA • In MENA: • end bearing is often ignored • Bored piles are often used • Using BS EN 1997-1 and the UK National Annex leads to savings: γRd x γs Saving FOS without testing 2.24 N/A FOS with testing to UK NA 1.68 -25%
- 25. Pile design to EC7 and τmax=f(σc) – cost savings Soil Rock UCS=2.0MPa Conventional AASHTO Carbonate rocks method + pile testing to EC7 Skin friction ignored End bearing ignored Socket Length ≈10m Socket Length 4m 131kPa<τmax<161kPa τmax = 350 kPa FULS = 4,400kN Ø=1.0m Ø=1.0m Concrete saving ≈4.7m3 per pile Socket Length 3m Ø=1.0m Concrete saving ≈5.5m3 per pile OR FSLS = 1,960kN FULS = 4,400kN FSLS = 2,620kN FULS = 3,290kN FSLS = 1,960kN FOS=2.24 FOS=1.68 FOS=1.68 25
- 26. An approach to reduce risk and construction costs 26 • Early stage project involvement • Desk study • Design/scoping and specification of high quality GI • Carry out GI • Full time GI supervision • Factual reporting by contractor • Review of factual reporting • Interpretation • Ground model + parameters + design SPECIALISTGEOTECHNICAL CONSULTANT ECONOMIC RISK CONTROLLED FOUNDATION SOLUTION By GI Contractor
- 27. Study based on highway construction Totalincreaseinconstructioncost(%) Investment in high quality GI saves on construction cost and reduces risk Geotechnical study cost / Construction tender cost (%) 15% 5% 27
- 28. References 28 • AASHTO LRFD Bridge Design Specification 2010 • Foundation design for the Burj Dubai - the world's tallest building - Poulos & Bunce, 2008 • Foundation design for the Pentominium tower in Dubai - Ibrahim, 2009 • Rock socket piles at mall of the emirates, Dubai – Alrifai, 2007 • Grouted Piles in Weak Carbonate Rocks – Abbs & Needham, 1985 • End bearing capacity of drilled shafts in rock – Zhang & Einstein, 1998 • A new design method for drilled piers in soft rock - Implications relating to three published case histories – Rowe and Armitage, 1987 • The landmark Osterberg Cell test, the deep foundations institute, nov/dec 2012. • The Osterberg Cell and bored pile testing – a symbiosis – Schmertmann and Hayes, 1997 • Bi-directional static load testing – State of the Art – England, 2003. • BS EN 199-1:2004 Geotechnical design – Part 1: General rules
- 29. Foundations for infrastructure projects in MENA an approach to reduce risk and construction costs Geotechnica ME - 4th & 5th December 2013 Benoît Latapie Senior Geotechnical Engineer WS Atkins & Partners Overseas +971 55 300 3797 benoit.latapie@atkinsglobal.com QUESTIONS?