From Research to Design in European Practice, Bratislava, Slovak Republic, on June 2 – 4.

2010

THE PRESSUREMETER: SOME CONTRIBUTIONS

TO FOUNDATION ENGINEERING
Jean-Louis Briaud

President of the International Society for Soil Mechanics and Geotechnical Engineering,
Professor and Holder of the Buchanan Chair, Dpt. of Civil Engineering, Texas A&M
University, College Station, Texas, 77843-3136, USA, (briaud@tamu.edu)

ABSTRACT: This article summarizes some contributions by the author and his students at
Texas A&M University in the applications of the preboring pressuremeter (PMT) in
geotechnical engineering. The topics are mostly related to foundation engineering. They
include correlations to other soil properties, predicting spread footings behavior by the load
settlement curve method, predicting laterally loaded piles behavior including non traditional
cases, and predicting downdrag loads using a simple movement based approach. In each
case, the objective is presented, the project is outlined, and the essence of the results is
described. References are given for further details.

1. Introduction

Einstein said: ”Everything should be made as simple as possible but not one bit simpler than
that” (Safir, Safire, 1982). This should be the goal of any researcher when proposing a
solution to a problem. If the solution is too complicated, it is unlikely that it will be of much
use to the engineer; yet if it is too simple, it is unlikely that the solution will be able to capture
the complete process. Finding the threshold of optimum simplicity (Fig. 1) is at the same time
desirable and difficult. To find this threshold, it is much easier to simplify an over-complicated
solution than to complicate an over-simplified solution.

Fig.1 - The Einstein Threshold of Optimum Simplicity

Foundation engineering problems tend to be complicated. Some 100 years ago the tendency
for foundation engineering solutions was to resort to experimental correlations because
theoretical soil mechanics was in its infancy. In the last 50 years, soil mechanics theory and
the associated numerical simulations have made remarkable progress and have more than
caught up with sampling and testing developments. There is a need to aim for solutions with
a proper balance between theoretical and experimental considerations. In foundation
engineering, this appears to give the best potential for reaching the threshold of optimum
simplicity.
2. TEXAM and Menard Pressuremeters

The Menard pressuremeter is well known and goes back to Menard’s original work in the
early 60’s. It has a tri-cell probe to minimize end effects and the test is conducted in equal
pressure steps. The TEXAM pressuremeter (Fig. 2, Briaud 1992) was developed in 1981 at
Texas A&M University to simplify and make safer (no pressurized gas bottle) the operation
and the repairing of the Menard pressuremeter while allowing for more versatility in the types
of possible PMT tests (e.g.: cyclic tests). The TEXAM probe is monocellular with a 6.5 length
to diameter ratio to minimize end effects and the test is performed in equal volume
increments. The work reported in this article was performed using the TEXAM
pressuremeter.

Fig.2 – The TEXAM Pressuremeter

Fig.3 – Comparison of the TEXAM & Menard Pressuremeters in Clay and Sand

A comparison between the two types of PMT was performed to try to quantify any difference
between the two instruments. Several boreholes were tested at the National Geotechnical
Experiment Sites at Texas A&M University (Briaud, 1997). One site is made of an
overconsolidated high plasticity clay with an undrained shear strength (UC tests) averaging
120 kPa. The other site is made of a silty overconsolidated sand with an SPT blow count
averaging 20 blows/0.3 m. The tests in the boreholes were performed by alternating the two
PMT tests as depth increased. Fig. 3 shows the results. Further references on this work are
in Briaud (1992)

3. Comparison with other tests results

A database of PMT test results and other soil test results was organized. The data came
from 82 pressuremeter borings located in the South, Southwest, West, and Central United
States with 36 sand sites, 44 clay sites, and 2 silt sites. Next to the PMT borings, other
borings were performed leading to data on undrained shear strength su, effective stress
friction angle φ’, SPT blow count N, CPT point and friction resistance qc and fs. At each depth
in a boring, a record was created which consisted of all parameters available at that depth. A
total of 426 records were accumulated. The clay deposits had su values varying from 9.6 kPa
to 2490 kPa and averaging 150 kPa. The sand deposits had blow counts varying from 1 to
100 and averaging 43. Best fit regressions were performed for the entire database and lead
to Tables 1 and 2. The scatter in the correlations is very large as exemplified by Fig.4 and
makes those correlations essentially useless in design. Indeed such approach would fall
below Einstein’s threshold of optimum simplicity. Further references on this work are in
Briaud (1992).

Table 1 – Correlation Coefficients for Sand: column A parameter equals number in table
times row B parameter

Table 2 – Correlation Coefficients for Clay: column A parameter equals number in table times
row B parameter
 Fig.4 – Example of Scatter for the Database

4. Shallow foundations: is there a scale and embedment effect?

The case considered is the one of a square footing in sand subjected to a vertical load
applied at the center of the footing surrounded by a flat and horizontal ground surface. The
average pressure under the footing is pf, the settlement is ρ, the footing width is B, the depth
of embedment is D, and the average soil strength within the zone of influence of the footing
is sa. The question raised is: is there a scale effect and a depth of embedment effect on the
load-settlement curve in this case?

Fig.5 - Results of the Large Footing Tests at Texas A&M University

The answer is based on theoretical considerations and on the results of experiments. The
theoretical considerations included the bearing capacity equation and the theory of elasticity.
The experiments performed for the study included five large footing load tests and over 30
plate load tests at the National Geotechnical Experimentation Site at Texas A&M University
(Briaud, 1997). The footings were approximately 3x3m, 3x3m, 2.5x2.5m, 1.5x1.5m, and
1x1m; they were all embedded 0.75m into a medium dense silty sand. The results are
presented in Briaud and Gibbens (1994, 1997, 1999). Fig. 5 gives an example. The plates
were square, the sizes were 0.1m, 0.2m, 0.3m, and 0.4m, and the embedment varied from
0m to 0.8m. The results are presented in Barfknecht and Briaud (1999). A review of the
literature on this topic yielded four additional studies with footing load tests (Ismael, 1985,
Pu and Ko, 1988, Khebib, Canepa, and Magnan, 1997, Lutenegger, 1995).

The answer to the question posed is limited by the evidence mentioned above and used to
reach the following conclusions.
1. There is no scale effect and no embedment effect for the curve pf/sa vs. ρ/B regardless of
the soil profile.
2. There is no scale effect and no embedment effect for the curve pf vs. ρ/B when the soil
strength profile is constant with depth.
3. There is a scale effect and an embedment effect for the curve p vs. ρ/B when the soil
strength profile is not constant with depth. The effect is an increase or a decrease
depending on whether the strength increases or decreases with depth. This effect
disappears if the curve is normalized as pf/sa vs. ρ/B.
4. The general bearing capacity equation for sands assumes a soil strength profile which
increases linearly with depth because φ and γ are constant. In this case, Nγ and Nq are
constant and the equation gives the right influence of B and D. For any other strength
profile (the majority of cases), this equation does not represent the true variation of the
bearing capacity because the assumptions no longer correspond to the strength profile.

5. Shallow foundations: the load settlement curve method

The load-settlement curve method is used to generate the complete load settlement curve for
a footing. The Load Settlement Curve Method (LSCM) replaces the calculations of bearing
capacity and settlement which were done separately in the past. The LSCM was proposed
by Briaud and Jeanjean (1994) for square footings in sand resting on a flat ground surface
and subjected to a centered vertical load. This method is based on the point-by-point
transformation of the pressuremeter curve into the load-settlement curve for the footing
through the use of two equations.
ρ/B = 0.24 ΔR/Ro (1)
pf = Γ pp (2)
where ΔR/Ro is the relative increase in pressuremeter cavity radius, pp is the pressure on the
cavity wall applied by the pressuremeter probe, and Γ is a transformation function obtained
experimentally and theoretically (Jeanjean, 1995) (Fig. 6).

As discussed in the previous section, the scale and embedment effects are directly tied to
the soil strength profile within the zone of influence of the footing, and the p/sa vs. ρ/B curve
is independent of the foundation width B and the depth of embedment D. This indicates that
the Γ function is independent of B and D.
 Fig.6 - The Γ Function for the Load-Settlement Curve Method

This method was extended to the case of a rectangular footing (B wide, L long) subjected to
an eccentric inclined load (eccentricity e and angle of inclination δ) at a distance d from the
crest of a slope (Hossain, 1996). A number of correction factors are proposed based on
numerical simulations calibrated against the large footing tests mentioned in the previous
section. These factors are as follows:

Influence of the shape f L/B = 0.8 + 0.2(B/L) (3)

Influence of eccentricity fe = 1 – 0.33(e/B) center (4)
fe = 1 – (e/B)0.5 edge (5)
Influence of inclination fδ = 1 – (δ/90)2 center (6)
fδ = 1 – (δ/360)0.5 edge (7)
Influence of a slope fβ,d = 0.8 (1 + d/B)0.1 slope at 3 to 1 (8)
fβ,d = 0.7 (1 + d/B)0.15 slope at 2 to 1 (9)

For the time being, the superposition of cases is taken into account by multiplying the
influence factors as is common practice. There is some evidence that this approach is
conservative (Hossain, 1996). More research is needed in this area. The load-settlement
curve method consists of the following steps and Fig. 7 is an example:

1. Perform preboring pressuremeter tests (PMT) within the depth of influence of the
footing; usually at depths equal to 0.5B, 1B, and 2B.
2. Prepare the PMT curves and obtain the average curve (Briaud, Jeanjean, 1994).
3. Transform the average PMT curve point by point into the footing pressure vs. relative
settlement curve:
ρ/B = 0.24 ΔR/Ro (10)
pf = fL/B fe fδ fβ,d Γ pp (11)
where ρ is the settlement of the footing, B is the footing width, ΔR et Ro the increase
in radius and the initial radius of the cavity in the PMT test respectively, pf the footing
pressure corresponding to ρ/B, fL/B, fe, fδ, et fβ,d the influence factors for shape,
eccentricity, inclination, and proximity of a slope given by equations (3) to (9) above,
Γ the function given in Fig.6 and which already includes the scale and depth of
embedment effect, and pp the pressure in the PMT test corresponding to ΔR/Ro.
4. Prepare the load settlement curve for the footing once the pf vs. ρ/B curve is known.
 Fig.7 - An Example of the Load-Settlement Curve Method.

6. Deep foundations under horizontal loads: reference case

The problem is the one of a single pile subjected to a horizontal static load. This problem is
often solved by assuming that the pile is an elastic member and that the soil can be
represented by a series of non-linear horizontal springs called P-y curves. A method was
developed at Texas A&M University to obtain the P-y curve directly from the pressuremeter
curve. This method was simplified and lead to the Simple Analysis for Lateral Load On Piles
(S.A.L.L.O.P.) (Briaud, 1997).

The following observation is the basis for the simplification. A conceptual plot of the soil
resistance P per unit length of pile as a function of depth z is shown in Fig. 8a. The
sinusoidal nature of the P-z profile is such that the soil resistance alternates direction and
essentially cancels itself out except for a shallow zone close to the ground surface which
contributes most to the lateral resistance. More specifically there is a depth Dv where the
shear force in the pile is zero (Fig. 8b).
 Fig. 8 - (a) Soil Resistance versus Depth, (b) Free Body Diagram of Shallow Segment

The horizontal equilibrium of this shallow segment of pile is the basis of the SALLOP method.
The method consists of obtaining the lateral capacity Hou, the horizontal movement yo at 1/3
of that load, and the maximum bending moment Mmax under Hou/3. The lateral capacity Hou is
defined as the horizontal load corresponding to a horizontal movement at the pile head equal
to B/10 where B is the pile diameter. The method was developed on a theoretical basis but
was adjusted after studying a database of 20 full-scale pile load tests (e.g.: Fig. 9).

Fig. 9 - Example of Full Scale load Test in the Database

The steps of the method are as follows:

1. Perform preboring PMT tests within a depth corresponding to 2Dv.
2. Reduce the data and obtain the profile of limit pressures pl and the profile of first load
modulus Eo. Select a design pl value and a design Eo value from the profiles within the
depth Dv. Use 1.5m if Dv is not known.
3. Calculate the zero shear depth Dv by using:
Dv = (π/4) lo for L > 3lo (12)
Dv = L/3 for L < lo (13)
where lo = (4EI / K)1/4 (14)
and L is the pile length, E the pile material modulus, I the moment of inertia of the pile,
and K the horizontal soil modulus taken as 2.3Eo after studying the pile database. If the
pile length is between lo and 3lo use linear interpolation.
4. Calculate the lateral capacity Hou by using:
Hou = ¾ pl B Dv (15)
5. Estimate the horizontal deflection yo under Ho (a safe fraction of Hou) by
yo = 2 Ho / lo K for L > 3lo (16)
yo = 4 Ho / L K for L > 3lo (17)
These equations are theoretically based and were used to find the best fitting value of K
(K = 2.3 Eo).

The accuracy and precision of the SALLOP method can be evaluated on Fig.10. If a moment
is also applied to the pile head, the reader is referred to Briaud (1997).

Fig.10 - Predicted vs. Measured Results for Horizontal Capacity and Movement.

7. Deep foundations under horizontal loads: time effects

Piles under horizontal loading can be subjected to long term loading (e.g.: retaining walls) or
short term loading (e.g.: wave loading, guard rail posts). The method presented in the
previous section gives the ultimate load Hou and the deflection yo at Hou/3 for loading times
corresponding to the loading times of the load tests used to calibrate the method. These
loading times are of the order of hours. If the time of application of the load is very different
than that, there is a need to take this difference into consideration. Briaud and Garland
(1985) developed a time effect model as follows:
Hou(t)/Hou(to) = (t/to)-n (18)
y0(t)/yo(to) = (t/to)n (19)
where to is the reference time (say one hour) corresponding to Hou(to) or yo(to), and t is the
time for which Hou(t) or yo(t) are required. The rate effect exponent is n; it is determined by
running a pressuremeter test where the pressure is held while recording the volume increase
as a function of time. In this case equation 18 is applied to the volume of the PMT cavity and
n is back calculated since all other parameters are known. The n values accumulated by
Briaud and Garland (1985) ranged from 0.02 to 0.08 in clays and 0.01 to 0.03 in sands. The
model has been verified in the case where the soil behavior for t and to is undrained and in
the case where the soil is above the ground water level. An example of application is given
next. Fig. 11 shows the behavior of a bored pile loaded in 30 days increments. Also shown
on the figure is the envelope of the 30 minute readings at the beginning of the 30 days
increments. The second part of Fig. 11 is a pressuremeter test performed with 30 minutes
pressure holding steps. The back calculated n values for the pile and the PMT were very
close to each other. See Briaud (1992) for more details and for the database of 7 full scale
piles subjected to two different rates of horizontal loading.

Fig. 11 – Rate Effect Comparison for a Horizontal Pile Load Test and a PMT.

8. Deep foundations under horizontal loads: effect of cycles

If the horizontal load is repeated a number of times (e.g.: wave loading), the pile response is
affected and the reference procedure needs to be modified. One way cyclic loading refers to
the case where the load is increased and then decreased back to zero at most. Two way
cyclic loading refers to the case where the pile is pushed in one direction and then pushed in
the opposite direction past the point of zero deflection. For symmetric two way loading the
pile is pushed to the same deflection or load in both directions. The pressuremeter can
produce one way cyclic loading but not two way cyclic loading. Therefore the application of
the PMT is limited to the case of one way cyclic lateral loading of piles or to two way cyclic
lateral loading of piles in soils where two way loading gives the same results as one way
loading. A database of cyclic horizontal load tests on piles in clays (Makarim, Briaud, 1986)
and on piles in sands (Little, Briaud, 1986) was accumulated. Fig. 12 shows an example. It
was found that in clays, there is little difference between one way cyclic loading and two way
cyclic loading. However in sands there can be significant difference between one way cyclic
loading and two way cyclic loading. If yo is the deflection at the pile top obtained from the
reference procedure, the following model was used to predict the deflection yN at the pile top
after N cycles.
yN = yo Na (20)
where a is the cyclic exponent. The value of a is obtained by performing a cyclic PMT test
with a minimum of 10 cycles and using equation 20 with the PMT secant modulus. A
database of 28 cyclic horizontal pile load tests was accumulated (Briaud, 1992) and the
back-calculated a values were as follows. For clays a varied from 0.01 to 0.35 with an
average of 0.094 and no noticeable difference between one way and two way cyclic loading.
For sands a varied from 0.005 to 0.26 for one way cyclic loading with an average of 0.076
and from -0.14 to 0.06 for two way tests with an average of 0.002. Therefore the above PMT
procedure can be used to predict the behavior of piles subjected to one way cyclic horizontal
loading in clays and in sands, two way cyclic horizontal loading in clays, but not two way
cyclic lateral loading in sands. In that last case cyclic degradation appears to be limited
according to the database accumulated.

9. Deep foundations under horizontal loads: next to an excavation

A trench may have to be open next to a horizontally loaded piles. The first step of course is
to check that the trench is stable; if not, the trench must be properly shored. The presence of
the trench weakens the response of the pile and a correction factor must be applied to the
reference method. A method was proposed by Briaud, Tucker (1987) on the basis of PMT
tests performed next to a trench and of finite element simulations. Fig.13 shows the results of
the PMT tests and the progressive loss of resistance as the PMT test is performed closer and
closer to the trench. Note that at the sand site the loss of strength is more dramatic than at
the clay site. A simplified version of the Briaud-Tucker 1987 method consists of reducing the
working load at the top of the pile.

Htrench = λ Hno trench (21)

Where λ is the reduction factor given in Fig.14 taking into account the proximity of the trench.
For example, a 1m diameter, 6m long drilled shaft is designed to carry 100 kN for 5 mm
deflection without an trench. A stable trench (shored if necessary) is dug; the nearest edge of
the trench is 3 m from the front of the pile and the depth of the trench is 3m. The value of λ =
0.625 is found on Fig.14 for H/B = 3, D/L = 0.5. Therefore the allowable load for the pile at
the same deflection of 5 mm is:
Htrench = 0.625 x 100 = 62.5 kN
Fig.12 – Example of cyclic horizontal load test and cyclic PMT test.
 Fig.13 – PMT Tests Close to a Trench

Fig.14 – Reduction Factor λ to Take into Account the Presence of a Trench

10. Deep foundations: downdrag and bitumen coating

The project was aimed at developing a methodology to select bitumen coatings to reduce
downdrag. It lead to publications outlining the procedure for uncoated piles and for bitumen-
coated piles (Briaud, Tucker, 1997, Briaud, 1997), to a computer program called PILNEG
(http://ceprofs.tamu.edu/briaud/pileneg.htm), and to a videotape on how best to coat piles
with bitumen. The reason this method is mentioned here in connection with the PMT test is
that one of the important parts of the design is the behavior of the pile point and that the PMT
test gives one of the best prediction of the pile point load transfer curve. The method to
calculate the downdrag load and the allowable top load for an uncoated pile is outlined by
working through an example by hand in this article. The computer program, the case of a pile
group, and the bitumen selection process can be found in the publications mentioned above.
The example is the one of a single pile (Fig.15) driven in a soil deposit that will experience
the settlement profile shown on the figure. The maximum shear stress that the soil can exert
on the pile is taken as a constant equal to 25 kN/m2 to simplify the calculations; it is assumed
that the movement will be large enough to mobilize the full friction load in all cases. The point
resistance is given by a load transfer curve as shown with a maximum point load of 1000 kN.
The question is: how much load can be placed on top of the pile if the top settlement must be
less than 14 mm; the problem is solved first for the uncoated pile and then for the coated
pile.
Uncoated Pile. The first step is to calculate the ultimate capacity of the pile in positive friction.
Qu = (25 kN/m2 x 1.2 m x 30 m) + 1000 kN = 1900 kN
Lets try a top load Qt of 500 kN. The neutral point is found at a depth where the movement of
the pile is equal to the movement of the soil. The calculations advance as a trial and error
process.
If the neutral point is at a depth of 20 m, then according to the soil profile, the movement of
the soil at that depth is w NP(soil) = 50 mm. The pile point carries a load Qp of:
Qp = 500 kN + (20 m x 1.2 m x 25 kN/m2) – (10 m x 1.2 m x 25 kN/m2) = 800 kN
For a point load of 800 kN, the point movement is given by the point load transfer curve as 4
mm. Now it is possible to calculate the pile movement at the neutral point by adding the pile
compression between a depth of 30m and a depth of 20m (depth of NP) to the 4 mm
movement at the point.
w NP(pile) = 4 mm + (950 kN x 104 mm / 0.09 m2 x 2x107 kN/m2) = 9.3 mm
Since w NP(soil) ≠ w NP(pile) the initial guess of 20 m for the depth of the neutral point is incorrect
and a new guess is required.
If the neutral point is at a depth of 29 m, then according to the soil profile, w NP(soil) = 5 mm
and Qp = 500 kN +870 kN – 30 kN = 1340 kN. This is not possible since the maximum point
load is 1000 kN. The pile point will therefore reach the maximum load of 1000 kN and vertical
equilibrium of the pile gives
500 kN + X = 1000 kN + (900 kN – X) or X = 700 kN
This downdrag value corresponds to 23.3 m of friction.
If the neutral point is at a depth of 23.3 m, then w NP(soil) = 35 mm and the movement at the
top of the pile is:
wtop = 35 mm + (850 kN x 23300 mm / 0.09 m2 x 2x107 kN/m2) = 46 mm
This is more than the allowable movement of 14 mm. Therefore the top load must be
reduced. Fig.16a gives the load distribution in the pile for a top load of 500 kN.
Lets try a top load Qt of 100 kN, the same approach is taken. The final iteration is shown
here.
If the neutral point is at a depth of 29 m, then according to the soil profile, w NP(soil) = 5 mm
and Qp = 100 kN +870 kN – 30 kN = 940 kN. The movement at the pile point is therefore wp
= 4.7 mm
w NP(pile) = 4.7 mm + (955 kN x 103 mm / 0.09 m2 x 2x107 kN/m2) = 5.2 mm
In this case, w NP(soil) ≈ w NP(pile) and the neutral point position is indeed at a depth of 29 m.
The top movement can now be calculated:
wtop = 5 mm + (535 kN x 29000 mm / 0.09 m2 x 2x107 kN/m2) = 13.6 mm
 Fig.15 - Example Problem for Downdrag Calculations

Fig.16 - Load distribution in the pile for (a) the uncoated pile and a top load of 500 kN, (b) the
uncoated pile and a top load of 100 kN, (c) the bitumen coated pile and a top load of 500 kN.

This settlement is acceptable. The load distribution in the pile is shown on Figure 16(b). The
distribution indicates that this pile which has a capacity of 1900 kN, can only be allowed to
carry 100 kN (Fig.16b) because of the settlement criterion and of the downdrag. In addition
this pile has a point resistance under working load which has a very low factor of safety
against plunging. In this case, it becomes very advantageous to coat the pile with bitumen or
other bond breakers as shown in the following.

Bitumen-Coated Pile. A bitumen coating which reduces the maximum shear stress that the
soil can exert on the pile from 25 kN/m2 to 2.5 kN/m2 is selected. The coating however must
only be applied to the part of the pile which will be subjected to downdrag. The neutral point
is found in the same fashion as previously.
Lets try a top load Qt of 500 kN.
If the neutral point is at a depth of 29 m, then according to the soil profile, w NP(soil) = 5 mm
and Qp = 500 kN +87 kN – 30 kN = 557 kN (Fig.16c). This corresponds to a point movement
of 2.8 mm and a pile movement at the neutral point w NP(pile) close to w NP(soil). Therefore the
neutral point is at a depth of 29 m. The settlement of the pile top is:
wtop = 5 mm + (543.5 kN x 29000 mm / 0.09 m2 x 2x107 kN/m2) = 13.8 mm
This is acceptable. The capacity of the pile is:
Qu = (2.5 kN/m2 x 1.2m x 29 m) + (25 kN/m2 x 1.2 m x 1 m) + 1000 kN = 1117 kN
The factor of safety against plunging failure is therefore 1117 / 500 = 2.23. Fig.16(c) shows
the load distribution in the pile at working loads. By coating the pile with bitumen, the
allowable load has been increased from 100 to 500 kN; coating the pile is estimated to
increase the pile cost by 20 to 30 %.

11. Future work

The application of the PMT test to the design of retaining walls needs to be developed. The
present practice for retaining walls is dominated by a pressure diagram approach which
assumes that the wall displacement will generate the assumed pressure diagram. There is a
need to develop a displacement based approach to the design of retaining walls. The PMT
offers a way to obtain the relationship between the lateral movement of the soil and the
lateral pressure. The problem is that the PMT curve is an axisymmetric curve while the
retaining wall curve is a plane strain curve. It is this transformation which needs to be
developed. Such transformation will allow the PMT curve to serve as an input to Winkler
model based programs for the design of retaining walls. Such measured (Fig.17) and
numerically simulated curves have already been generated (Briaud, Kim, 1999) and have
lead to Fig.18 and 19. The transformation of the PMT curve remains to be done.

Fig.17 - Full Scale Instrumented Tieback Wall at Texas A&M University

 Fig.18 - Mean Earth Pressure Coefficient vs. Normalized Top Wall Deflection

Fig.19 - Mean Earth Pressure Coefficient vs. Normalized Mean Wall Deflection

12. Conclusions

From the testing point of view, the disadvantages of the PMT are that the quality of the
borehole influences the test results, the probe dimensions are not standardized, some soils
are difficult to test, and drainage is not controlled. From the design point of view, the PMT is
of limited help in the design of retaining walls and slope stability.

From the testing point of view, the advantages of the preboring pressuremeter (PMT) test are
that the test can be performed and data obtained in many soil types, the test gives an in situ
stress strain curve from which many important soil parameters can be calculated, the test
also represents an in situ load test, the PMT equipment is relatively inexpensive, and the
quality of the test can be judged from the shape of the curve. From the design point of view,
the best application of the PMT is the design of laterally loaded piles (including sustained
loading and cyclic loading) because of the close analogy, the next best application is the
design of spread footings, the third application is vertically loaded piles particularly end
bearing piles.

REFERENCES:

1. Barfknecht J., Briaud J.-L., (1999), “Effect of Scale and Depth of Embedment for Footings
in Sand,” Research Report, Texas A&M University, Dpt. of Civil Engineering, College
Station, Texas, USA.(can be obtained from: briaud@tamu.edu)
2. Briaud J.-L., (1992), “The Pressuremeter”, Taylor and Francis Group (A.A. Balkema),
London, UK. (http://www.tandf.co.uk/books/)
3. Briaud J.L., (1997) “Bitumen Selection for Downdrag on Piles,” Journal of Geotechnical
and Geoenvironmental Engineering, Vol. 123, No. 12, ASCE, New York, December 1997.
4. Briaud J.L., (1997) “SALLOP: Simple Approach for Lateral Loads on Piles,” Journal of
Geotechnical and Geoenvironmental Engineering, Vol. 123, No. 10, pp. 958-964, ASCE,
New York, October 1997.
5. Briaud J.-L., (1997), “The National Geotechnical Experimentation Sites at Texas A&M
University: Clay and Sand, A Summary”, Report NGES-TAMU-007-September 1997.
Civil Engineering, Texas A&M University.
6. Briaud J.-L., Chen H.-C., Kwak K., Han S., Ting F., (2001b), “Multiflood and Multilayer
Method for Scour Rate Prediction at Bridge Piers”, Journal of Geotechnical and
Geoenvironmental Engineering, Vol. 127, No.2, pp. 114-125, Feb. 2001, ASCE, Reston,
Virginia.
7. Briaud J.-L., Garland E., 1985, “Loading Rate Method for Pile Response in Clays”,
Journal of Geotechnical Engineering, Vol.111 No.3, ASCE Washington, March 1985.
8. Briaud J.-L., Gibbens R. M. (1997), “Large Scale Load Tests and Data Base of Spread
Footings on Sand,” Federal Highway Administration, Publication No. FHWA-RD-97-068,
McLean, Virginia, USA, pp. 217.
9. Briaud J.-L., Gibbens R. M., (1999), “Behavior of Five Large Spread Footings in Sand,”
Journal of Geotechnical and Geoenvironmental Engineering, Sept. 1999, Vol. 125, No. 9,
American Society of Civil Engineers, Reston, Virginia, USA.
10. Briaud J.-L., Gibbens R. M., Editors (1994), “Predicted and Measured Behavior of Five
Spread Footings on Sand,” Geotechnical Special Publication No. 41, American Society of
Civil Engineers, Reston, Virginia, USA, pp. 255.
11. Briaud J.-L., Jeanjean, Ph., (1994), “Load Settlement Curve Method for Spread Footings
on Sand”, Settlement ‘94 Specialty Conference, ASCE Specialty Publication No. 40,
ASCE, Vol. 2, 1774-1804.
12. Briaud J.L., Kim, N.K., (1999), Beam Column Method for Tieback Walls, Journal of
Geotechnical and Geoenvironmental Engineering, Vol. 124, No. 1, ASCE, New York,
January 1998.
13. Briaud J.-L., Lim Y., (1999), “Tieback Walls in Sand: Numerical Simulation and Design
Implications,” Journal of Geotechnical and Geoenvironmental Engineering, Vol. 125, No.
2, pp. 101-111, February 1999, ASCE, Reston, Virginia.
14. Briaud J.-L., Ting F., Chen H.C., Cao Y., Han S.-W., Kwak K., (2001a), “Erosion Function
Apparatus for Scour Rate Predictions”, Journal of Geotechnical and Geoenvironmental
Engineering, Vol. 127, No.2, pp. 105-113, Feb. 2001, ASCE, Reston, Virginia.
15. Briaud J.-L., Ting, F.C.K., Chen, H.C., Gudavalli, R., Perugu, S., Wei, G., (1999),
“SRICOS: Prediction of Scour Rate in Cohesive Soils at Bridge Piers”, Journal of
Geotechnical and Geoenvironmental Engineering, Vol. 125, No. 4, April 1999, pp. 237-
246, American Society of Civil Engineers, Reston, Virginia, USA.
16. Briaud J.-L., Tucker L.M., (1987), “Horizontally Loaded Piles Net to a Trench”,
Foundations for Transmission Line Towers, ASCE Geotechnical Special Publication
No.8, Atlantic City, April 1987.
17. Briaud J.L., Tucker, L.M., (1997) “Design and Construction Guidelines for Downdrag on
Uncoated and Bitumen-Coated Piles”, NCHRP Report 393, TransportationResearch
Board, National Academy Press, Washigton DC, 1997.
18. Canepa Y., Despresles, D. (1990), “Catalogue des essais de chargement de fondations
superficielles réalises sur sites par les L.P.C. (1978-1990), F.A.E.R.1.17.020, L.R.E.P.
Melun, 31/12/90, France.
19. Hossain K. M., (1996), “Load Settlement Curve Method for Footings in Sand at Various
Depths, under Eccentric or Inclined Loads, and Near Slopes,” Ph.D. Dissertation, Texas
A&M University, Dpt. of Civil Engineering, College Station, Texas, USA.
20. Ismael N. F. (1985), “Allowable Pressure from Loading Tests on Kuwaiti Soils,” Canadian
Geotechnical Journal, No. 22, Canadian Geotechnical Society, Ottawa, Canada, pp. 151-
157.
21. Jeanjean P., (1995), “Load Settlement Curve Method for Spread Footings on Sand From
the Pressuremeter Test”, Ph.D. Dissertation, Texas A&M University, Dpt. of Civil
Engineering, College Station, Texas, USA.
22. Khebib Y., Canepa, Y., and Magnan, M.-P. (1997), “Base de Donnees de Fondations
Superficielles SHALDB: Essais des Laboratoires des Ponts et Chaussées,” June 1997,
Laboratoire Central des Ponts et Chaussées, Paris, France.
23. Kwak K., Briaud J.-L., Chen H.-C., 2001, “SRICOS: Computer Program for Bridge Pier
Scour”, Proceedings of the 15th International Conference on Soil Mechanics and
Geotechnical Engineering, Vol. 3, pp 2235-2238, A.A. Balkema Publishers, Rotterdam,
The Netherlands.
24. Lim Y., (1996), “Three Dimensional Non Linear Finte Element Analysis of Tieback Walls
and Soil nailed Walls under piled Bridge Abutment”, Ph.D. Dissertation, Texas A&M
University, Dpt. of Civil Engineering, College Station, Texas, USA.
25. Little R.L., Briaud J.-L., (1987), “A Pressuremeter Method for Single Piles Subjected to
Cyclic Lateral Loads in Sand”, Research Report 5357 to the US Army Engineers
Waterways Experiment Station, Civil Engineering, Texas A&M University, April 1987.
26. Lutenegger A. (1995), personal communication by fax dated 9-27-95.
27. Makarim C.A., Briaud J.-L., (1986), “Pressuremeter Method for Single Piles Subjected to
Cyclic Lateral loads in Overconsolidated Clays”, Research Report to various Oil
Companies, Texas A&M University, December 1986.
28. Moody L.F., (1944) “ Friction Factors for Pipe Flow”, Transaction of the American Society
of Civil Engineers, Vol. 66, Reston, Virginia, USA.
29. Pu J.-L., Ko, H.-Y. (1988), “Experimental Determination of Bearing Capacity in Sand by
Centrifuge Footing Tests,” Proceedings of Centrifuge ‘88, J.-P. Corte Ed., A. A. Balkema
Publishers, Rotterdam, Netherlands, p. 293-299.
30. Safir L., Safire W. (1982), “Good Advice”, Times Books, New York, N.Y., USA