6.1 GENERAL
There are no special methods for calculation of bearing capacity and settlements in silt. For sands, there are a number of methods for estimating the bearing capacity and settlements of shallow foundations based primarily on CPT tests or SPT tests.
Besides being restricted to sands, there are often also other restrictions relating, for example, to minimum depths of embedment. In some of the methods, a very cautious extrapolation to silty sand has been proposed, but none of them has been proposed to be valid for silt.
One of the aims of the current project was to investigate the extent to which methods normally used for sands can be applied also for silt. The aforementioned methods have therefore been tested in the silty soils, sometimes under conditions which are also far outside the geometrical restrictions normally applied. The results in the present investigations therefore do not reflect on the general usefulness of the methods within the areas for which they have been intended.
Other methods are more general and are intended to cover all types of soil and geometrical conditions. As its name implies, the "general equation for bearing capacity" is thus intended for all soils and conditions. In this equation, both empirically estimated shear strength parameters and parameters determined by more elaborate testing may be introduced. Also the Menard type of pressuremeter test and the related method of calculation of bearing capacity are intended to be general. The new interpretation and calculation method for pressuremeter tests presented by Briaud ( 1995) was, on the other hand, only proposed to be valid under certain conditions in sand.
For calculation of settlements, the methods using theory of elasticity to estimate the distribution of the stress increase and moduli are applicable to all types of soil.
The moduli may then be estimated empirically from sounding tests, evaluated semi-empirically from dilatometer tests or measured in oedometer tests, among
SGI Report No 54
194
other ways. As before, the calculation methods based on pressuremeter tests according to Menard are general, but the Briaud method has so far been restricted to sand.
6.2 PREDICTION OF SETTLEMENTS
Empirical methods
Estimation of settlements ofshallow foundations on sand is often made on the basis of results from CPT tests and empirical calculation methods elaborated specifical
ly for this purpose. Such methods have been proposed by Meyerhof ( 1965 and 1974), DeBeer (1967) and Schmertmann (1970 and 1978) among others. Of these methods, the DeBeer and Schmertmann methods are most commonly used in Sweden.
Ill According to the DeBeer method, the settlement of a foundation can be calculated as
S
= f
.L..J-23 f og(a'o+L!CYJ
, LlZ Ao c ao
where C=
a'o
z = depth of influence
cr' in situ effective vertical stress Licr= increase in vertical stress
The settlement thus calculated refers to the settlement after 10 years.
The increase in vertical stress is calculated by theory of elasticity and the calculations are performed over a depth z below the foundation level, which corresponds at least to the depth where the vertical stress is increased by 10 % or more. The stress increase below a rigid footing is calculated for the "characteristic point", i.e. the point at which calculated stresses and settlements are independent of the rigidity of the foundation, Fig. 6.1.
According to DeBeer, this method normally overestimates the settlements.
Investigations and load tets in silty soils 195
Characteristic point
Minimum depth 1,1 a~ for calculations
.c
...
0. Q)
0
Fig. 6. I Stresses contrib
uting to the settlements in the DeBeer method.
111111 By using the Schmertmann method, the settlement is calculated as
c
where
1 = 1-
o.s%
/q*c2
= 1 + 0.2log(l0t)t
=
time from load application, yearsIz
=
an influence factor varying with depth according to Fig. 6.2 Izp = 0.5+0.l~q *f
dvpq*
=
q q = net foundation pressure at the foundation level correspondingO
to the peak value of the influence factor
%
= a
'vo=
effective overburden pressure at the foundation levela
vp = net foundation pressure at the level corresponding to the peak value of the influence factor IzpE sz
=
equivalent Young's modulus varying from 2.Sqcz for a square footing (lib= 1) to 3.Sqcz for a long strip footing (lib~ 10)qcz= cone resistance at depth z below the foundation level in CPT tests
SGI Report No 54
196
0 0,1 0,2 0,5 fz ' '
-'=.!lb 1 0,5+ 0, 1 Vq*/O~p
b ' ' ' ' ,
T,:·
5,I \min 0,5 , ,
,,
7" , ,
.c
g- "'
2b~
!lb <! 10 ,,Cl , ,,,
, ; 3b , ,,,,
:
,
I
I Fig. 6.2 Variation of influence factor 12
I with depth.
I
4b
The two methods differ in principle. The DeBeer method assumes a distribution of stresses and settlements according to theory of elasticity and indirectly yields a modulus (dcr'/dc) which increases with stress and strain, and which in principle normally corresponds to the oedometer case. The Schmertmann method assumes an empirical stress (strain) distribution with depth and indirectly yields a modulus (dcr'/dc) which decreases with stress and strain and is more in line with what is normally observed in load tests on footings, Fig. 6.3.
Load 0
...
C:Q)
E
E 1--DeBeerJl :-o-Schmertmann
Fig. 6.3 Schematic relation between load-settlement curves calculated by the DeBeer and Schmertmann methods.
Investigations and load tets in silty soils 197
B A simpler method was proposed by Meyerhof (1974) in which the settlement is calculated from
s = - -qb 2 qc
where qc is the average cone resistance within depth b below the foundation level.
In principle, this corresponds to a constant modulus.
B Settlement calculations in sand can also be made using empirical methods elaborated for results from dynamic penetration tests. Such methods have been presented by Schultze and Sherif (1973) and Parry (1971). According to Parry, the settlement can be calculated from
0.3qb ,m
s =
-N30
where q
=
average foundation pressure, MPa b=
foundation width, m=
blows per 0.3 m of penetration in SPT tests (alternatively N20 NetN30
blows per 0.2 m of penetration in HfA dynamic probing tests can be used)
The calculated settlement refers to the settlement after 10 years.
The relevantN-value is calculated as a weighted average over a depth 2b below the foundation level, Fig. 6.4.
Fig. 6.4 Calculation of relevant N-value (Parry 1971 ).
SGI Report No 54
198