PROCEEDINGS FROM THE SECOND INTERNATIONAL FLAT DILATOMETER CONFERENCE

Interrelationships of DMT and CPT readings in soft clays

Paul W. Mayne
Civil & Environmental Engineering, Georgia Institute of Technology, Atlanta, GA, USA

Keywords: clays, cone penetration, dilatometer, in-situ tests, porewater readings, pressures

ABSTRACT: Interrelationships between the flat dilatometer readings (lift-off pressure, p0, and expansion
pressure, p1) and piezocone readings (cone tip stress, qt, and penetration porewater pressures, u2) are explored
for three soft clay sites. Within the intact regions, the p0 and u2 measurements are quite consistently similar in
magnitude, whereas qt is variably larger than both p0 and p1, perhaps somewhat dependent on the effective
friction angle of the clay. Companion sets of DMT and CPTU at a given site could be used to better define the
extent of the crustal zone, degree of fissuring, intact regions, and related permeability characteristics of these
substrata within a clay formation.

1 INTRODUCTION been developed (Campanella & Robertson, 1986;

Robertson, 1990). In these systems, conflicts can
The combined use of flat dilatometer tests (DMT) arise as paired readings or normalized parameters
together with piezocone penetration tests (CPTU) from the qt-fs and qt-u2 charts can provide different
can be a nice complement in defining sublayer zones evaluations for the same depths.
and general geostratigraphy within the subsurface For the DMT, the soil type is evaluated from the
environment. While many consider each of these in- material index: ID = (p1-p0)/(p0-u0) per the recom-
situ tests to be self-standing by themselves for de- mendations of Marchetti (1980), whereby clays are
tailing a soil layer profile, in some instances, the use indicated by ID < 0.6 and sands are identified by ID >
1.8. Further distinctions of silty to sandy subcatego-
of CPT soil behavioral charts (e.g., Robertson, 1990)
rizations are available too. The original relationship
can, in fact, give misleading or erroneous results
appears to solidly produce reasonable evaluations of
and/or miss changes in soil strata and substrata
soil types over two decades later (e.g., Marchetti, et
(Zhang & Tumay, 1999).
al. 2001). An advantage of the DMT over CPTU
The standard piezocone test provides three sepa-
profiling is the lack of worry over desaturation of a
rate readings with depth, including: cone tip stress
porous element and ability to detail geostratigraphy
(qt), sleeve friction (fs), and penetration porewater
pressure at the shoulder (u2), whereas the flat dila- at sites having a deep groundwater table.
tometer determines two readings: the lift-off or con-
tact pressure (p0) and expansion pressure (p1). For 2 INTRA- AND INTER-RELATIONSHIPS
the CPT, soil types are often distinguished by use of
2 of the 3 of the readings, as summarized by Kul- For each test with multiple measurements, intra-
hawy & Mayne (1990) and Fellenius & Eslami relationships between the individual readings can be
(2000). The earlier CPT classification methods util- sought to ascertain trends in the measurements, par-
ized qt and fs, yet some measurement difficulties can ticularly within a specific geologic formation or soil
be found with the sleeve friction because of rough- type. Within that given geotechnical unit, inter-
ness, wear, porewater presure corrections, and other relationships between different test data (lab or field)
factors (Lunne, et al. 1986). On the other hand, soil can be made to develop correlative and statistical
behavior type (SBT) using qt and u2 readings will trends. Herein, some interrelationships between the
undoubtably be weak in interpretations for situations DMT and CPT readings in soft clays have been ex-
involving deep water tables, as porewater readings plored.
will be zero or change with capillarity effects. Con- Intra-relationships between the two DMT readings
sequently, SBT methods utilizing all 3 readings have in different soils have been explored by Garcia

220
 PROCEEDINGS FROM THE SECOND INTERNATIONAL FLAT DILATOMETER CONFERENCE

(1991) based on compiled databases from field tests 3 CLAY SITES INVESTIGATED
and calibration chamber test series. The successful
evaluation of soil type using ID would corroborate Companion series of DMT and CPTU soundings
such findings. For the CPT in clays, intra-relations were obtained in two intact soft clays and one fis-
between tip stress (qt) and penetration porewater sured clay by GT field crews, as well data from as
pressures on the cone tip (u1) and shoulder (u2) have one very well-documented intact soft clay site re-
been produced (Mayne, Kulhawy, & Kay, 1990). ported in the literature. Table 1 lists the four sites
The presence of fissures, whether from crustal for- considered for this study.
mation and/or desiccation, or from mechanical over-
consolidation effects, was shown significant in the Table 1. Clay sites with DMT and CPT datafiles.
qt-u2 link, yet much less so in the qt-u1 trends. Site Soil Conditions Reference
Interrelationships between the DMT and CPTU Amherst, MA Soft varved clay Hegazy (1998)
readings have been investigated previously by Bothkennar UK Soft clay Nash et al. (1992)
Mayne & Bachus (1989) who showed that, as a first Ford Center, IL Soft glacial clay This study
approximation: I-10 & 42, LA Stiff fissured clay Chen-Mayne (1994)

p0 ≈ umax (1) Recently, tests were performed by the GT field crew

in soft clay deposits north of Chicago, Illinois. These
where umax = peak penetration porewater pressure in-situ tests were conducted as part of the geotechni-
given by u2 in intact clays and by u1 in fissured cal site investigation for the Ford Design Center lo-
clays, as shown by Figure 1. cated on the campus of Northwestern University, in
conjunction with an instrumented excavation pro-
ject. The project is located near the national geo-
technical experimentation site (NGES) next to Lake
Michigan (Finno, et al. 2000). Subsurface consists of
a shallow sandy fill overlying soft silty clays from
glacial freshwater lacustrine deposits and a ground-
water table located about 3 m deep.
Figure 2 shows the profiles of dilatometer expan-
sion pressure and measured cone tip resistance with
depth and Figure 3 presents the dilatometer contact
pressure with penetration porewater pressures from
two piezocone soundings. The region of intact clay
can be interpreted for depths below 9 m, as evi-
denced by the agreement & similarity of p0 and u2
profiles. Above 9 m, less consistency in the readings
are observed. For the same depth range, qt > p1.

Measured Stress (MPa)

Figure 1. Trend between CPTu porewater pressures and DMT 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4
4
contact pressures in clays (after Mayne & Bachus 1989).
6

The aforementioned trend was later found applicable 8

for residual silty soils of the Atlantic Piedmont geol-
Depth (m)

ogy by Mayne & Liao (2004). 10

Direct comparisons of the profiles of the measured 12

cone tip resistance (qt) with the DMT p0 and p1 pres-
sure readings in clays, as well as other readings, 14 F02 qt
have been made at sites in Northwestern Canada F03 qt
(Sully & Campanella, 1990; Sully 1994). Herein, 16

generalized trends are explored between the DMT DMT 01 p1

18
and CPT measurements at four clays sites tested fol-
lowing the 1989 correlations. These data were ob- 20
tained from 3 soft clays (two tested by the authors
team) and one fissured clay that was overconsoli- Figure 2. DMT p1 and CPT qt at Ford Center Design, IL.
dated by desiccation.

221
 PROCEEDINGS FROM THE SECOND INTERNATIONAL FLAT DILATOMETER CONFERENCE

Series of DMTs and CPTus were conducted by

Measured Pressure (kPa)
0 200 400 600 800
the GT field crew at the Amherst NGES (Martin &
4
Mayne 1997; Hegazy 1998). The soils consist of an
upper shallow clay fill and desiccated crust overly-
6 ing soft varved lacustrine clay. Groundwater lies
8
about 1 m deep. Full details on the testing program
and soil properties for the NGES are given by Lute-
10 negger (2000). Figure 4 shows the comparison pro-
Depth (m)

u2 (CPT02)
files of three sets of p1 with three sets of qt, indicat-
12
ing the intact varved clay below depths of 4 m. Here,
u2 (CPT03)
14 the cone tip resistance is just barely greater than the
u0 expansion pressures. There is also a parallel profil-
16
DMT po ing of p1 and qt in the upper clay fill and desiccated
18
crust, as well.
u0 In Figure 5, the DMT contact p0 pressures are
20 comparable to the CPT shoulder u2 porewater pres-
sures. However, it is also apparent that for two of the
Figure 3. DMT p0 and CPT u2 at Ford Design Center, IL. CPTs, either the porous elements were insufficiently
saturated prior to testing, or else became desaturated
during advancement through the crust. Only CPTu
Measured Stress (MPa) sounding 01 appears to have properly delineated the
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 transition into the soft intact region below 4 m. In
0 contrast, the p0 readings clearly and consistently
1 show the change in strata, as well as a relatively uni-
2 formity in the underlying soft clay. Thus, the DMT
3 offers an advantage in that the p0 measurements are
4 not subject to desaturation effects.
Depth (m)

5 In-situ test data from DMTs and CPTs obtained

6 CPT01 qt
in the soft clay at the British national experimenta-
7 CPT05 qt
tion test site at Bothkennar (Nash, et al. 1992) were
8 CPT07 qt
also reviewed and digitized. These data were utilized
9 DMT1 p1
to provide a reference benchmark in relative com-
DMT2 p1
10 parisons of the data from the Amherst and Evanston
DMT3 p1
11 sites.
12

Figure 4. DMT p1 and CPT qt at Amherst NGES, MA. 4 DMT-CPT TRENDS IN INTACT CLAYS

Interrelationships between the dilatometer pressures

Measured Pressure (MPa) and cone penetrometer measurements can be ap-
-0.1 0 0.1 0.2 0.3 0.4 0.5 0.6 proximately formulated in terms of cavity expansion
0
theory (e.g., Mayne & Bachus, 1989; Sully 1994).
1
The relationships can be established in terms of total
2 stress parameters: i.e., the undrained shear strength
3 (su) and rigidity index (IR = G/su), where G = shear
4 modulus. Alternatively, the relationships may be ob-
tained from more fundamental derivations using
Depth (m)

5
6 critical-state soil mechanics to utilize the effective
7
CPTu2 01 stress friction angle (φ') and stress history in terms of
CPTu2 05 overconsolidation ratio (OCR = σp'/σvo'), where σp'
8
CPTu2 07 = preconsolidation stress and σvo' = current effective
9
DMT1 po overburden stress (Mayne, 2001). In any event, the
10 DMT2 po expressions can only be approximate since neither
11 DMT3 po the flat dilatometer blade nor the cone penetrometer
12 with 60º apex tip are represented by an infinite cyl-
inder nor by a perfect sphere. Instead, empirical rela-
Figure 5. DMT p0 and CPT u2 at Amherst NGES, MA. tions can be explored.

222
 PROCEEDINGS FROM THE SECOND INTERNATIONAL FLAT DILATOMETER CONFERENCE

For the data corresponding to the intact regions of

the three soft clays, Figure 6 shows the direct rela- 1000
tionships between p1 and p0. Best fit lines from re- Evanston, IL
gression analyses with forced intercepts equal to
zero are shown for each (y = mx with b = 0). As the Amherst, MA

Contact Pressure, p 0 (kPa)

800
groundwater tables are rather shallow for these sites, Bothkennar UK
these regressions correspond directly with the indi-
y = 0.975x
vidual material indices for each site, including: the 600 2
R = 0.945
Ford Design Center at Evanston, Illinois (ID = 0.163 y = 0.986x
± 0.069), Amherst NGES in Massachusetts (ID = 2
R = 0.899
0.166 ± 0.044), and Bothkennar test site in Scotland 400
(ID = 0.291 ± 0.052). All three sites contain lightly
overconsolidated clays with 1 < OCRs < 2 in the soft
intact zones. Additional index parameters and prop- 200
y = 0.959x
erties of these clays are summarized in Table 2, in- 2
R = 0.987
cluding: natural water content (wn), liquid limit
(LL), plasticity index (PI), and effective stress fric- 0
tion angle (φ'). 0 200 400 600 800 1000
Porewater Pressure, ub (kPa)
Table 2. Mean values of index parameters for soft clay sites.
Clay Depth wn LL PI φ'
Site (m) (%) (%) (%) (deg) Figure 7. One-to-one relationship between DMT p0 and CPT u2
Amherst 6 to 12 62 51 21 22º readings in soft intact clays.
Evanston 10 to 18 32 33 17 26º
Bothkennar 2 to 16 65 70 45 37º
1500

Evanston, IL
Expansion Pressure, p 1 (kPa)

The notable trends between p0 and u2 at each of the

Amherst MA
sites are shown in Figure 7, substantiating the origi-
nal correlation represented by equation (1) based on 1000
Bothkennar UK
earlier data. Similarly, forced fit best lines (b = 0)
are shown with their associated coefficients of de-
termination (R2). The interrelationship of p0 and u2 y = 0.905x y = 0.706x
appears unique and applies to all three intact clays. 2
R = 0.891
2
R = 0.916
500

1000
y = 0.645x
y = 1.134x 2
Expansion Pressure, p 1 (kPa)

R = 0.992
2
R = 0.964
800 0
0 500 1000 1500

Cone Tip Stress, qt (kPa)

600
y = 1.168x
y = 1.218x 2
2 R = 0.918 Figure 8. Observed relationship between DMT p1 and CPT qt
400 R = 0.994
readings in soft intact clays.
Evanston, IL
200 Amherst MA

Bothkennar UK For the p1 trends with qt, Figure 8 shows that each of
0 the clays shows a distinct and unique interrelation-
0 200 400 600 800 1000 ship. In this case, the ratios p1/qt appear to decrease
Contact Pressure, p0 (kPa)
with the effective stress friction angle of the clay.

Figure 6. Interrelationships of p1 with p0 for intact clays.

223
 PROCEEDINGS FROM THE SECOND INTERNATIONAL FLAT DILATOMETER CONFERENCE

Measured Pressure (kPa) Measured Stress (kPa)

0 1000 2000 3000 4000 0 1000 2000 3000 4000 5000
0 0

DMT po
5 5
CPTu1 07
CPTu2 07
10 10
CPTu1 06
Depth (meters)

CPTu2 06
15 15

Depth (m)
20 20

25 25

30 30

DMT p1
35 35 CPT 07 qt
CPT 06 qt
40 40

Figure 9. Comparision of p0 with u1 and u2 readings in over- Figure 10. Comparision of p1 with qt profiles in OC clay at Ba-
consolidated clay at I-10 and Route 42 near Baton Rouge, LA ton Rouge, LA.

5 DISCUSSION FOR FISSURED CLAYS Interestingly, the relative profiles of DMT expan-
sion pressure and CPT tip resistance with depth ap-
In the case of fissured overconsolidated clays, the pear to behave similarly to that noted for the intact
piezocone shoulder porewater pressures tend to- clays and the p1 interrelationship with qt is not ap-
wards zero and even negative values (Mayne, et al. parently affected by the presence of fissuring.
1990). Thus, since face porewater pressures at the In the case of fissured crusts overlying soft clays,
tip or midface (u1 readings) will remain as positive the DMT can be used to help delineate the extent of
values, these will better correlate with DMT p0 read- the desiccation zone, without fear of desaturation of
ings. Yet, it is likely that u1 > p0, as shown previ- porous elements or poor element saturation practices
ously in Figure 1 by fissured London clay at Brent associated with piezocone deployment. In compan-
Cross and fissured Gault clay at Madingley (Lunne, ion sets of DMT and CPTU soundings, the results
et al. 1997). can be used together to better define the zone of in-
This facet is illustrated by DMT and CPTU data tact clays where permeability characteristics are
collected at the I-10 and state route 42 site near Ba- likely to be low. In the upper crustal regions with
ton Rouge, Louisiana (Chen & Mayne, 1994), as fissuring, the permeability will be higher and will
shown in Figure 9. Index parameters for the stiff also reduce the operational undrained shear strength.
clay are given in Table 3. At this site, a multi-
element piezocone was used and perhaps the water-
saturated porous elements were not as responsive as 6 CONCLUSIONS
those should glycerine or silicon oils have been used
for the saturation process. In any event, the p0 more Interrelationships between DMT pressures and CPT
closely parallels a profile with the measured face u1 readings are explored to discern general trends in
porewater pressures than with the u2 readings that soft clays. Data from three soft intact clays show
are normally used in practice because of the need for that the DMT contact pressure (p0) is about equal to
porewater corrections on the measured cone tip re- the CPT shoulder (u2) penetration porewater pres-
sistance (Campanella & Robertson, 1988; Lunne et sure and the CPT tip stress (qt) exceeds the expan-
al., 1997). sion pressure (p1) by 10 to 50 percent. Companion
sets of DMT and CPT can help better define the ex-
Table 3. Index parameters of stiff fissured clay from Louisiana tent of crustal & desiccated zones. In fissured clays,
Clay Depth wn LL PI φ' the profiles of qt and p1 appear similar, but p0 more
Site (m) (%) (%) (%) (deg) closely follows the CPT face (u1) porewater pres-
Baton Rouge 5 to 30 34 61 33 27º sures because u2 readings go negative.

224
 PROCEEDINGS FROM THE SECOND INTERNATIONAL FLAT DILATOMETER CONFERENCE

Mayne, P.W., Kulhawy, F.H., and Kay, J.N. (1990). Observa-

ACKNOWLEDGMENTS tions on the development of porewater pressures during
piezocone tests in clay. Canadian Geot. J. 27 (4): 418-428.
The author appreciates the support of the National Mayne, P.W. (2001). Stress-strain-strength-flow parameters
Science Foundation (Award No. CMS-0338445) for from enhanced in-Situ tests, Proceedings, International
which Dr. Rick Fragaszy is the program director. Conference on In-Situ Measurement of Soil Properties &
Case Histories, Bali, Indonesia: 27-47.
Thanks to Professor Richard Finno of Northwestern
Mayne, P.W. and Liao, T. (2004). CPT-DMT interrelationships
University who provided access to the Illinois site
in Piedmont residuum. Geotechnical & Geophysical Site
and to Professor Alan Lutenegger for his help at the
Characterization (2), Millpress, Rotterdam: 345-350.
Amherst NGES.
Nash, D.F.T., Powell, J.J.M. and Lloyd, I.M. (1992). Initial in-
vestigations of the soft clay test site at Bothkennar, U.K.
REFERENCES
Geotechnique 42 (2): 163-181.
Robertson, P.K. (1990). Soil classification using the cone pene-
Campanella, R.G. and Robertson, P.K. (1988). Current status
tration test. Canadian Geotechnical J. 27 (1): 141-158.
of the piezocone. Penetration Testing 1988, Vol. 1 (Proc.
Sully, J.P. (1994). Measurement of in-situ lateral stress during
ISOPT, Orlando), Balkema, Rotterdam: 93-116.
full-displacement penetration tests. PhD Dissertation, Dept.
Chen, B.S-Y. and Mayne, P.W. (1994). Profiling the overcon-
of Civil Engrg., Univ. British Columbia, 485 pages.
solidation ratio of clays by piezocone tests. Georgia Tech
Sully, J.P. and Campanella, R.G. (1990). Measurement of lat-
Research Corp. Report No.GIT-CEEGEO-94-1submitted to
eral stress in cohesive soils by full-displacement in-situ
National Science Foundation, 280 pages.
tests. Transportation Research Record 1278: 164-171.
Fellenius, B.H. and Eslami, A. (2000). Soil profile interpreted
Zhang, Z. and Tumay, M.T. (1999). Statistical to fuzzy ap-
from CPTu data. Proceedings Geotechnical Engineering
proach toward CPT soil classification. Journal of Geotech-
Conference, Asian Institute of Technology, Bangkok: 1-18.
nical & Geoenvironmental Engineering 125 (3): 179-186.
Finno, R.J., Gassman, S.L. and Calvello, M. (2000). NGES:
Northwestern Univ. National Geotechnical Experimenta-
tion Sites (GSP No. 93), ASCE, Reston/VA: 130-159.
Garcia, S.R. (1991). Interrelationship between the initial lift-off
and extended pressure readings of the Marchetti flat blade
dilatometer in soils. Special Research Project, MS in Civil
Engrg., Georgia Inst. of Technology: 119 pages.
Hegazy, Y.A. (1998). Delineating geostratigraphy by cluster
analysis. PhD Thesis, Civil & Env. Engrg., Georgia Insti-
tute of Technology, Atlanta, GA: 464 p.
Kulhawy, F.H. and Mayne, P.W. (1990). Manual on estimating
soil properties for foundation design. Report EL-6800,
Electric Power Research Institute, Palo Alto, 306 p.
Lunne, T., Eidsmoen, T., Gillespie, D. and Howland, J.D.
(1986). Laboratory and field evaluation of cone penetrome-
ters. Use of In-Situ Tests in Geotechnical Engineering (GSP
6), ASCE, Reston/VA: 714-729.
Lunne, T., Robertson, P.K. and Powell, J.J.M. (1997). Cone
Penetration Testing in Geotechnical Practice, EF
Spon/Blackie Academic, Routledge Publishers, London.
Lutenegger, A.J. (2000). NGES: Univ. of Massachusetts. Na-
tional Geotechnical Experimentation Sites (GSP No. 93),
ASCE, Reston/VA: 102-129.
Marchetti, S. (1980). In-situ tests by flat dilatometer. Journal
of Geotechnical Engrg. 106 (GT3): 299-324.
Marchetti, S., Monaco, P., Totani, G. and Calíbrese, M. (2001).
The flat dilatometer (DMT) in soil investigations (ISSMGE
TC 16). Proc. Intl. Conf. on In-Situ Measurement of Soil
Properties & Case Histories, Bali, Indonesia: 95-131.
Martin, G.K. and Mayne, P.W. (1997). Seismic flat dilatome-
ter tests in Connecticut valley varved clay", ASTM Geo-
technical Testing Journal 20 (3), 357-361.
Mayne, P.W. and Bachus, R.C. (1989). Penetration porewater
pressures in clays by CPTU, DMT, and SBP. Proc. 13th
Intl. Conf. Soil Mechanics & Fdn Engrg (1), Rio: 291-294.

225