Correlations between CPT and PMT at a Dynamic

Compaction Project
B. Hamidi & H. Nikraz
Curtin University of Technology, Perth, Australia
S. Varaksin
Menard, Paris, France

ABSTRACT: In many ground improvement projects the preliminary field testing

method is different from the tests that are carried out later. Hence, it may be
necessary to correlate the two tests for comparative purposes.
Dynamic Compaction, using pounders weighing up to 35 tons, has been used to
treat 175,000 m2 of hydraulically reclaimed carbonate sands with a thickness of about
16 m for a port expansion project in Ras Laffan, Qatar. In this project CPT was used
before and after ground improvement works. A number of Pressuremeter tests (PMT),
some in the immediate vicinity of the CPT locations, were also carried out. This paper
shall briefly describe the project and the ground improvement works. Previously
published literature correlating CPT to PMT will be reviewed and the results of the
two testing methods that have been carried out in Ras Laffan shall be correlated, and
conclusions will be made based on the findings.

1 THE PROJECT: NAKILAT SHIP REPAIR YARD

Ras Laffan, located about 70 km north of Qatar’s capital city, Doha, is one of the
world’s largest and fastest developing gas hubs. As shown in Figure 1 and as part of
Port of Ras Laffan’s expansion program it was decided to construct Nakilat Ship
Repair Yard by hydraulically reclaiming its site from the Persian Gulf.
The land was reclaimed using the carbonate sand and gravel that was dredged for
deepening the port. The material’s grain size was understood to be generally less than
75 mm, but it was anticipated that stones as large as 500 mm in diameter could also
be present. The maximum fines content (passing 63 μm sieve) of the fill was
generally less than 10%.
The dense layer of seabed in the reclaimed area was variable from -9.1 m to -13.2
m CD (chart datum). Design (final platform) level was specified to be at +3.5 m CD.
While it was understood that other less sensitive areas of the project would require
lesser treatment, three areas in the dry docks designated as DDR4, DDR5 and DDR 6
and shown in green in Figure 1 were deemed to require treatment by Dynamic
Compaction (Menard, 1972-74). The surface areas of these regions were respectively
57,064 m2, 35,643 m2 and 82,962 m2. Consequently, the total area was 175,669 m2.
Figure 1. Plan of Nakilat Ship Repair Yard (Dynamic Compaction areas shaded in green)

With the assumption that the reclaimed ground would settle about 5% due to
Dynamic Compaction (DC), reclamation was carried out 0.6 to 0.8 m higher than the
design levels. Hence working levels varied from +4.1 to +4.3 m CD.

1.1 Design and Acceptance Criteria

It was the primary requirement of the project for the ground to possess a relative
density, Dr, of 60% based on Baldi et al. (1986):

(1)

where qc= CPT cone resistance (kPa) and σ’v= effective vertical stress (kPa).
The correction factor to be applied to the cone resistance for carbonate sand was
stipulated to be 1.94.
For the purpose of calculations it was specified that the saturated density, γsat, and
unsaturated density, γunsat, were respectively 18.7 and 15.2 kN/m3. Average
groundwater level was assumed to be at +0.5 m CD.
As past experience indicated that this requirement would most probably not be met
without ground improvement, it was foreseen that Dynamic Compaction would be
carried out if the geotechnical investigation proved that this expectation was correct.
At the same time it was well understood that by implementing Dynamic
Compaction the improvement of the upper layers of reclaimed ground would be much
more than what was required. Consequently, the functionality of the project could
have been achieved more affordably by also envisaging alternative criteria based on
the design needs.
The alternative criteria specified that for an isolated footing carrying a 4000 kN
load:
 Allowable bearing capacity: 200 kPa
 Maximum settlement: 50 mm
1.2 Initial Ground Conditions
The CPT tests that were carried out as part of the geotechnical investigation
consistently demonstrated that the reclamation would not meet the criteria. In areas
DDR4, DDR5 and DDR6 the soil in the upper 3 to 5 m was medium to very dense
with qc ranging from as low as 5 to more than 20 MPa. Then the soil became loose to
medium dense with qc fluctuating between 1 to 7 MPa. Dense seabed was
encountered at depths of 13 to 17 m. The CPT friction ratio was generally well below
1%.

2 GROUND IMPROVEMENT BY DYNAMIC COMPACTION

The project’s schedule stipulated that mobilization, ground improvement and testing
had to be completed according to the below milestones:
 DDR4: 154 days after issuance of notice to proceed.
 DDR5: 63 days after issuance of notice to proceed.
 DDR6: 91 days after issuance of notice to proceed.
Two specially equipped cranes were mobilized and utilized to meet the schedule.
Dynamic Compaction was carried out in areas DDR4, DDR5 and DDR6 using
pounders weighing 35 tons, 28 tons, 25 tons and 15 tons. The 35 ton pounder was
dropped in free fall from 25 m and without engagement to the winch and cabling
system using the innovative MARS (Menard Automatic Release System) device that
was used for the first time in 2004 during the ground improvement works of a loose
fill with a maximum thickness of 28 m in Al Quoa’a, UAE (Varaksin, Hamidi and
Aubert, 2005). Figure 2 shows the two Dynamic Compaction rigs. The front crane is
dropping a 28 ton pounder and the back crane is dropping the 35 ton pounder using
MARS.

Figure 2. Dynamic Compaction using 28 ton and 35 ton pounders

2.1 Testing
In order to verify that the project requirements had been satisfied 1 CPT was carried
out per every 600 m2 of improved ground. A number of PMT (pressuremeter) tests
were also carried out for comparative purposes. This provided an opportunity to
perform a number of CPT and PMT tests in the same locations and to correlate the
results for carbonate sands.

3 CPT-PMT CORRELATION

3.1 Literature Review

Briaud et al (1985) collected 82 PMT borings data from various projects from 1978 to
1985. The result of this study proposed that the magnitude of correlation between qc
and PMT parameters Pl, limit pressure, and Ep, pressuremeter modulus, is shown in
Table 1.

Table 1. Correlation between PMT and CPT according to Briaud et al. (1985)
Soil type PMT parameter Correlation to CPT
Sand Pl 0.2qc
Ep 2.5 qc
Clay Pl 0.11 qc
Ep 1.15 qc

Baguelin, Jezequel and Shields (1978) have reviewed and interpreted a number of
CPT-PMT correlations such as Van Wameke (1962), Cassan (1968-69), Jezequel et
al. (1968) and Nazaret (1972) that were originally printed in French publications.
Baguelin et al. note that although the correlation between CPT and PMT in most
technical publications is based on the ratio of qc/Pl, the ratio qc*/Pl* would be more
representative (qc* and Pl* are respectively the net cone resistance and net limit
pressure and can be calculated from Equation 2 and Equation 3).

qc*= qc- qo (2)

Pl*= Pl- Po (3)

where qo and Po are respectively the in-situ vertical and horizontal total stresses. In
general the ratios qc*/Pl* and qc/Pl are close because qo and Po are small compared to
qc and Pl, but can be quite different at depth in soft clays.
The influence of depth on qc*/Pl* can be studied from the works of Jezequel et al.
(1968) on the dikes of a tidal power project in Rance. The dikes were hydraulic fills
composed of clean sand with dry density equal to 1.5 t/m3. The ratio qc*/Pl* in the
upper 1.5 m layer of fill was from 9.11 to 12.03. Even though qc varied from 2 to 10
MPa, qc*/Pl* was about 6.7 throughout the remainder of the 20 m thick fill.
Nazaret (1972) did not observe the same independency of qc*/Pl* from qc* in his
study on Loire sand, and reports a tendency of the ratio to increase with the increase
of qc*.
Baguelin et al. interpret that the high ratio values near the ground surface are due
to the differences between shallow and deep failure conditions. CPT has a small
diameter and rapidly reaches its critical depth. However, PMT has to reach a depth of
embedment of about 1.5 m (1 m in clays, 2 m in sands) before the test is no longer
influenced by the surface of the ground.
According to Baguelin et al. the soil type is the parameter that has the greatest
effect on the ratio of qc*/Pl*, and for depths of about 5 to 20 m there seems to be a
narrow correlation between qc* and Pl*. Baguelin et al. summarize the qc*/Pl* ratio to
be on average within the ranges shown in Table 2.

Table 2. qc*/Pl* for different soil types according to Baguelin et al. (1978)
Soil Description qc*/Pl*
Very soft to soft clays close to 1 or 2.5 to 3.5
Firm to very stiff clay 2.5 to 3.5
Very stiff to hard clay 3 to 4
Very loose to loose sand and compressible silt 1 to 1.5 and 3 to 4
Compact silt 3 to 5
Sand and gravel 5 to 12

Baguelin et al. understand that it is very likely that dilatancy is a key factor and
qc*/Pl* could prove to be a reliable indicator of the importance of dilatancy in the
resistance of a particular soil. If qc*/Pl* is about 5 to 6 then the soil is probably non-
dilatant or slightly dilatant. A ratio of 8 to 12 probably indicates a dilatant soil.

3.2 CPT – PMT Correlation for Carbonate Sand in Port of Ras Laffan
After execution of Dynamic Compaction in DDR5 using a maximum pounder weight
of 28 tons (without ironing) it was decided to perform a DC trial to study the
improvement effects using a 35 ton pounder that was dropped by MARS. This
process included 3 deep compaction phases and an ironing phase.
3 PMT tests were carried out next to 3 CPT tests in the below order:
 Before phase 1: PMT-007 and CPT-551 (in between impact points)
 After phase 1: PMT-009 and CPT-576 (in between impact points)
 After phase 3: PM-010 and CPT-595 (in impact point)
The cone resistances of the CPTs are shown in Figure 3. Likewise, the limit
pressure and pressuremeter modulus of the PMT are respectively shown in Figure 4
and Figure 5.

Figure 3. Cone resistance values used in the correlation

Figure 4. PMT limit pressure values used in the correlation

Figure 5. PMT modulus values used in the correlation

Figure 6. Ep/qc ratios for Ras Laffan carbonate sand

Figure 7. qc/Pl ratios for Ras Laffan carbonate sand

Figure 8. qc*/Pl* ratios for Ras Laffan carbonate sand

As shown in Figure 6, the Ep/qc ratios are substantially less than Briaud et al.’s
(1985) correlation. Here, the uppermost shallow points do not seem to correlate
differently due to the differences between the shallow and deep failure modes. The
average Ep/qc value for the 23 test points is 1.35. The average Ep/qc ratios for the
three comparisons on Ras Laffan carbonate sand are 1.5, 1.3 and 1.1. The maximum
and minimum ratios for the points were respectively 0.3 and 1.91.
Briaud et al. have proposed an average value of 2.5 for Ep/qc. This value is 1.85
times more than what has been measured for Ras Laffan carbonate sand, and is
basically within the range (95%) of the correction factor of 1.94 that was applied to
Baldi et al.’s equation (1986) to calculate relative density for carbonate sand.
The ratios of qc/Pl have been shown in Figure 7. It can be observed that the
average qc/Pl ratios for 21 tests points are equal to 4.54. That value is 90% of what
has been proposed by Briaud et al. (1985). The 21 test points do not include the
uppermost test points of PMT007 and PMT010 due to the differences in between the
shallow and deep failure modes. The average qc/Pl ratios for the three correlations on
Ras Laffan carbonate sand are 4.1, 5 and 5.3 excluding the mentioned uppermost
points. The minimum and maximum ratios were respectively 2.9 and 9.1 for the 21
points.
As shown in Figure 8, the qc*/Pl* ratios are identical in shape and very close in
value to the qc/Pl ratios. The average qc*/Pl* ratios for the 21 tests points are equal to
4.82 which is just below the range of 5 to 12 that has been proposed by Baguelin et
al. (1978). The average qc*/Pl* ratios for the three correlations are 4.3, 5.4 and 5.2
excluding the mentioned uppermost points. The minimum and maximum ratios for
the 21 points were respectively 3 and 9.3.

4 CONCLUSION

The correlation results of CPT and PMT for densified Ras Laffan carbonate sand did
not demonstrate any advantages of qc*/Pl* over qc/Pl. Both ratios were slightly less
than what Bageulin et al. and Briaud et al. have proposed, but for preliminary
purposes the ratios may be assumed to be 5. Ep/qc ratios appear to be less than what
Briaud et al. have proposed, and the same correction factor that is been used for
calculating relative density from Baldi’s formula may relate the relation between
carbonate and non-carbonate Ep/qc ratios as well.

5 ACKNOWLEDGEMENT

The authors would like to thank Menard for providing the project information and
data that has been presented in this paper.

6 REFERENCES

Baguelin, F., Jezequel, J. F. & Shields, D. H. (1978) The Pressuremeter and Foundation Engineering,
Aedermannsdorf, Trans Tech Publications,617.
Baldi, G., Belloti, V., Ghionn, N., Jamiolkowski, M. & Pasqualini, E. (1986) Interpretation of CPT's
and CPTU's - 2nd Part: Drained Penetration of Sands. 4th International Geotechnical Seminar
Field Instrumentation and In-Situ Measurements, Nanyang Technological Institute, Singapore, 25-
27 November 1986, 143- 156.
Briaud, J. L., Noubani, A., Kilgore, J. & Tucker, L. M. (1985) Correlation Between Pressuremeter
Data and Other Paramaters, Research Report. Texas A&M University.
Cassan, M. (1968) Les essais in situ en mécanique des sols. Construction, 10 (October), 337-347.
Cassan, M. (1969) Les essais in situ en mécanique des sols. Construction, 7-8 (July-August), 337-347.
Jezequel, J. F., Lemasson, H. & Touze, J. (1968) Le pressiomètre Louis Ménard: Quelques problèmes
de mise en oeuvre et leur influence sur les valeurs pressiométriques. Bulletin de Liaison des
Laboratoires des Ponts et Chaussées, 32 (June - July), 97-120.
Menard, L (1972) "La consolidation dynamique des remblais recents et sols compressibles," Travaux,
No. November, pp 56-60.
Menard, L (1974) "La consolidation dynamique des sols de fondation," Revue des Sols et Fondations,
p 320.
Nazaret, M. (1972) Influence du mode de mise en oeuvre de la sond pressiométrique. Rapport de
Recherche du Laboratoire Régional des Ponts et Chaussées d'Angers, F.A.E.R 1.05.11.1.
Van Wameke, A. (1962) Méthodes d'investigation des sols en place - Etude d'une campagne d'essais
comparatifs. Sols Soils, 2, 9-18.