C h a r les-Au gu st in C O U LO MB - A g e ot e chn ical t r ib u t e
P a r is , se p t e m b er 25 & 26, 2023
Workshop #1: Design methods for retaining walls
Assessment of active and passive earth pressure:
what do we owe to Boussinesq and Caquot?
S.Burlon
Te r r a s o l , S e t e c
Introduction
Values used for the design of
both:
gravity walls with limit
equilibrium methods
retaining walls with
subgrade reaction
methods
1 9 4 8 1966 1 9 7 3 1 9 9 0
1 stedition 2 ndedition 3 rdedition
SYMPOSIUM COULOMB 2
P ARIS, SEPTEMBER 25 & 26, 2023
Introduction
Two main issues:
what are the main
contributions for the
elaboration of these
tables?
How have these values
been determined?
1 9 4 8 1966 1 9 7 3 1 9 9 0
1 stedition 2 ndedition 3 rdedition
SYMPOSIUM COULOMB 3
P ARIS, SEPTEMBER 25 & 26, 2023
Boussinesq (1874)
Introduction (1874)
Conclusion (1874)
1874
1876
Strong criticism of the Rankine theory.
New equations are proposed but it is not possible to provide closed-form solutions.
SYMPOSIUM COULOMB 4
P ARIS, SEPTEMBER 25 & 26, 2023
Résal (1903)
Active earth pressures: first values
(horizontal component)
1903
SYMPOSIUM COULOMB 5
P ARIS, SEPTEMBER 25 & 26, 2023
Caquot (1934)
𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑
= 3𝑡𝑡 − 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 (1) = 𝑛𝑛𝑛𝑛 − 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 (2)
𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑 The stress state ensures the
𝑚𝑚 = 2𝑘𝑘 − 1 equilibrium of the system
2
4 that is fully plastic.
𝑚𝑚 = 1 + 4𝑡𝑡𝑡𝑡𝑡𝑡 𝜑𝜑 ± 𝑡𝑡𝑡𝑡𝑡𝑡2 𝜑𝜑 − 𝑡𝑡𝑡𝑡𝑡𝑡2 𝛼𝛼
𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐
The parameter k ensures that
the stress state with its
inclination α is on the Mohr
circle (‘+’ for active state and
‘–’ for passive state)
A first method is proposed
but the numerical values are
still those calculated by Résal
in 1903.
1934
SYMPOSIUM COULOMB 6
P ARIS, SEPTEMBER 25 & 26, 2023
Ravizé (1945)
The calculation principles are clearly
established with the distinction of
two parts: the first one is related to
the Rankine equilibrium while the
second one is related to the
Boussinesq equilibrium.
The equations are undetermined: for
any stress inclination on the wall, it
is possible to find an earth pressure Rankine
equilibrium
on the wall that ensures the Boussinesq
equilibrium with the Rankine state on equilibrium
t h e l i n e O M 1.
1945 Some values are proposed.
SYMPOSIUM COULOMB 7
7
P ARIS, SEPTEMBER 25 & 26, 2023
Caquot and Kerisel (1948 and 1949)
𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑
= 3𝑡𝑡 − 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 (1) = 𝑛𝑛𝑛𝑛 − 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 (2)
𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑
𝑚𝑚 = 2𝑘𝑘 − 1
2
4
𝑚𝑚 = 1 + 4𝑡𝑡𝑡𝑡𝑡𝑡 𝜑𝜑 ± 𝑡𝑡𝑡𝑡𝑡𝑡2 𝜑𝜑 − 𝑡𝑡𝑡𝑡𝑡𝑡2 𝛼𝛼
𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐
Active and passive earth
pressure coefficients are
calculated for various cases
using a systematic procedure:
the first tables proposed by
Caquot and Kerisel are
presented.
1949
SYMPOSIUM COULOMB 8
P ARIS, SEPTEMBER 25 & 26, 2023
Active and passive earth pressures
for weighted ground conditions
Active earth pressures: the values are the same since 1948.
𝑐𝑐𝑐𝑐𝑐𝑐 2 𝜆𝜆 − 𝜑𝜑 𝑡𝑡𝑡𝑡𝑡𝑡2 𝛽𝛽 + 𝑡𝑡𝑡𝑡𝑡𝑡2 𝛿𝛿
𝐾𝐾𝑎𝑎𝑎𝑎 = 𝜌𝜌𝑘𝑘𝑎𝑎 𝑘𝑘𝑎𝑎 = 2 𝑙𝑙𝑙𝑙 𝜌𝜌 = − 2 − 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑙𝑙𝑙𝑙 1 − 0,9𝜁𝜁 2 − 0.1𝜁𝜁 4 1 − 0,3𝜁𝜁 3
2𝑡𝑡𝑡𝑡𝑡𝑡2 𝜑𝜑
𝑠𝑠𝑠𝑠𝑠𝑠 𝜑𝜑 + 𝛿𝛿 𝑠𝑠𝑠𝑠𝑠𝑠 𝜑𝜑 − 𝛽𝛽 𝜌𝜌 > 1,0
𝑐𝑐𝑐𝑐𝑐𝑐 𝜆𝜆 + 𝛿𝛿 1+
𝑐𝑐𝑐𝑐𝑐𝑐 𝜆𝜆 + 𝛿𝛿 𝑐𝑐𝑐𝑐𝑐𝑐 𝛽𝛽 − 𝜆𝜆
Coulomb-Poncelet coefficient
SYMPOSIUM COULOMB 9
P ARIS, SEPTEMBER 25 & 26, 2023
Active and passive earth pressures
for weighted ground conditions
Active earth pressures: the values are the same since 1948.
𝑐𝑐𝑐𝑐𝑐𝑐 2 𝜆𝜆 − 𝜑𝜑 𝑡𝑡𝑡𝑡𝑡𝑡2 𝛽𝛽 + 𝑡𝑡𝑡𝑡𝑡𝑡2 𝛿𝛿
𝐾𝐾𝑎𝑎𝑎𝑎 = 𝜌𝜌𝑘𝑘𝑎𝑎 𝑘𝑘𝑎𝑎 = 2 𝑙𝑙𝑙𝑙 𝜌𝜌 = − 2 − 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑙𝑙𝑙𝑙 1 − 0,9𝜁𝜁 2 − 0.1𝜁𝜁 4 1 − 0,3𝜁𝜁 3
2𝑡𝑡𝑡𝑡𝑡𝑡2 𝜑𝜑
𝑠𝑠𝑠𝑠𝑠𝑠 𝜑𝜑 + 𝛿𝛿 𝑠𝑠𝑠𝑠𝑠𝑠 𝜑𝜑 − 𝛽𝛽 𝜌𝜌 > 1,0
𝑐𝑐𝑐𝑐𝑐𝑐 𝜆𝜆 + 𝛿𝛿 1+
𝑐𝑐𝑐𝑐𝑐𝑐 𝜆𝜆 + 𝛿𝛿 𝑐𝑐𝑐𝑐𝑐𝑐 𝛽𝛽 − 𝜆𝜆
Coulomb-Poncelet coefficient
Passive earth pressures: the values have varied.
Example:
c=0, ϕ=30° and δ/ϕ =-1
Caquot, Kerisel and Absi:
Sokolowski (1965):
K p= 6 , 4 2 ( 1 9 4 8 ) , K p= 6 , 5 6 ( 1 9 6 6 ) ,
K p= 6 , 5 5
K p= 6 , 5 0 ( 1 9 9 0 )
The differences are negligible for the practice but it may be interesting to better
understand how these values were determined.
SYMPOSIUM COULOMB 10
P ARIS, SEPTEMBER 25 & 26, 2023
An example of numerical integration
𝑑𝑑𝑑𝑑
= 3𝑡𝑡 − 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 (1)
𝑑𝑑𝑑𝑑
𝑑𝑑𝑑𝑑
= 𝑛𝑛𝑛𝑛 − 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 (2)
𝑑𝑑𝑑𝑑
Runge Kutta (RK4) numerical
integration method
SYMPOSIUM COULOMB 11
P ARIS, SEPTEMBER 25 & 26, 2023
An example of numerical integration
𝑑𝑑𝑑𝑑
= 3𝑡𝑡 − 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 (1)
𝑑𝑑𝑑𝑑
𝑑𝑑𝑑𝑑
= 𝑛𝑛𝑛𝑛 − 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 (2)
𝑑𝑑𝑑𝑑
Runge Kutta (RK4) numerical
integration method
SYMPOSIUM COULOMB 12
P ARIS, SEPTEMBER 25 & 26, 2023
An example of numerical integration
Method presented:
K p= 6 , 5 4 8
Caquot, Kerisel, Absi:
K p= 6 , 4 2 ( 1 9 4 8 ) , K p= 6 , 5 6
( 1 9 6 6 ) , K p= 6 , 5 0 ( 1 9 9 0 )
𝑑𝑑𝑑𝑑
𝑑𝑑𝑑𝑑
= 3𝑡𝑡 − 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 (1) Sokolowski (1965):
𝑑𝑑𝑑𝑑 K p= 6 , 5 5
= 𝑛𝑛𝑛𝑛 − 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 (2)
𝑑𝑑𝑑𝑑
Runge Kutta (RK4) numerical
integration method
SYMPOSIUM COULOMB 13
P ARIS, SEPTEMBER 25 & 26, 2023
Other examples
Active earth pressure – δ/ϕ
Relative
-1 -2/3 0 2/3 1
inclination
Presented
0,886 0,477 0,333 0,301 0,307
method
Caquot
0,981 0,476 0,333 0,300 0,308
Kerisel
Coulomb 0,866 0,469 0,333 0,297 0,297
Passive earth pressure – δ/ϕ
Relative
-1 -2/3 0 2/3 1
inclination
Presented
6,55 5,26 3,0 1,46 ---
method
Caquot
6,50 5,30 3,0 1,46 ---
Kerisel
SYMPOSIUM COULOMB 14
P ARIS, SEPTEMBER 25 & 26, 2023
Other examples
Active earth pressure – δ/ϕ
Relative
-1 -2/3 0 2/3 1
inclination
Presented
0,792 0,533 0,405 0,362 0,356
method
Caquot
0,869 0,533 0,405 0,358 0,355
Kerisel
Coulomb 0,766 0,533 0,403 0,349 0,334
Passive earth pressure – δ/ϕ
Relative
-1 -2/3 0 2/3 1
inclination
Presented
5,39 4,72 3,22 1,86 1,03
method
Caquot
5,40 4,70 3,20 1,86 ---
Kerisel
SYMPOSIUM COULOMB 15
P ARIS, SEPTEMBER 25 & 26, 2023
Failure mechanism – Slip lines
Case 1 Case 2
Active earth
pressures
𝑟𝑟 = 𝑒𝑒 − ∫ 𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 𝜔𝜔 .𝑑𝑑𝑑𝑑 with:
𝜋𝜋 𝜑𝜑 𝜔𝜔𝛼𝛼 −𝛼𝛼 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠
𝜉𝜉 𝜔𝜔 = + −
4 2 2
and 𝜔𝜔𝛼𝛼 = 𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠
SYMPOSIUM COULOMB 16
P ARIS, SEPTEMBER 25 & 26, 2023
Failure mechanism – Slip lines
Case 1 Case 2
Active earth
pressures
𝑟𝑟 = 𝑒𝑒 − ∫ 𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 𝜔𝜔 .𝑑𝑑𝑑𝑑 with:
𝜋𝜋 𝜑𝜑 𝜔𝜔𝛼𝛼 −𝛼𝛼 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠
𝜉𝜉 𝜔𝜔 = + −
4 2
and 𝜔𝜔𝛼𝛼 = 𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎
2 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠
Case 1 Case 2
Passive earth
pressures
𝑟𝑟 = 𝑒𝑒 − ∫ 𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 𝜔𝜔 .𝑑𝑑𝑑𝑑 with:
𝜋𝜋 𝜑𝜑 𝜔𝜔𝛼𝛼 +𝛼𝛼 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠
𝜉𝜉 𝜔𝜔 = − +
4 2 2
and 𝜔𝜔𝛼𝛼 = 𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠
SYMPOSIUM COULOMB 17
P ARIS, SEPTEMBER 25 & 26, 2023
A last example – Comparison with the
kinematical approach of the yield analysis
P r e s e n t e d m e t h o d : K p= 7 , 7 7
C a q u o t - K e r i s e l : K p= 8 , 0
The difference is significant
even if it remains very low
and negligible for the
practice.
SYMPOSIUM COULOMB 18
P ARIS, SEPTEMBER 25 & 26, 2023
A last example – Comparison with the
kinematical approach of the yield analysis
Simon (2002)
Rendulic (1935)
Talren (2022)
A k i n e m a t i c a l a p p r o a c h p r o v i d e s : K p= 7 , 9 5 The two mechanisms are very similar.
The « true » value should be between 7,77 and 7,95.
SYMPOSIUM COULOMB 19
P ARIS, SEPTEMBER 25 & 26, 2023
Conclusions
Around 175 years after Coulomb, Caquot and Kerisel developped another approach to
assess active and passive earth pressures for weighted ground conditions from the
contributions of Bousinesq, Résal and Ravizé.
Failure mechanisms can also be obtained and compared to those considered by Coulomb
explaining why Coulomb approach is not on the safe side.
Today, Bousinesq, Caquot and Kerisel approaches can be compared to some other
solutions obtained by the kinematical approach of the yield analysis.
New values of active and passive earth pressure coefficients might be thus obtained by
comparing both Bousinesq, Caquot and Kerisel approaches and the kinematical approach
of the yield analysis.
SYMPOSIUM COULOMB 20
P ARIS, SEPTEMBER 25 & 26, 2023
Thank you for your attention
SYMPOSIUM COULOMB
P ARIS, SEPTEMBER 25 & 26, 2023