You are on page 1of 6

Use of the Flat Dilatometer Test (DMT) in geotechnical design

G. Totani, S. Marchetti, P. Monaco & M. Calabrese
University of L'Aquila, Faculty of Engineering, L'Aquila, Italy

Preprint : IN SITU 2001, Intnl. Conf.
On In situ Measurement of Soil
Properties, Bali, Indonesia, May 2001

ABSTRACT: This paper presents an overview of the DMT and of its design applications, in the light of the
experience accumulated over the last 20 years. The following applications are discussed: determining
common soil parameters, predicting settlements of shallow foundations, compaction control, detecting slip
surfaces in clay slopes, predicting the behavior of laterally loaded piles, evaluating sand liquefiability,
estimating consolidation/flow coefficients, selecting soil parameters for FEM analyses. The basic differences
of the DMT compared to other penetration tests are also discussed.
The Flat Dilatometer Test (DMT), developed in Italy
in 1980, is currently used in over 40 countries both
for research and practical applications. The wide
diffusion of the DMT lies on the following reasons
(Lutenegger 1988): (a) Simple equipment and
operation. (b) High reproducibility. (c) Cost
effectiveness. (d) Variety of penetration equipment.
The test procedure and the original correlations
were described by Marchetti (1980). Subsequently,
the DMT has been extensively used and calibrated in
soil deposits all over the world. In addition to some
300 research publications, various standards (ASTM
Suggested Method 1986), regulations (Eurocode 7
1997) and manuals (US DOT 1992) are available
today. Design applications, recent findings and new
developments are described by Marchetti (1997) in a
state-of-the-art report.

The test starts by inserting the dilatometer into the
ground. Soon after penetration, the operator inflates
the membrane and takes, in about 1 min, two
readings: the A pressure, required to just begin to
move the membrane ("lift-off"), and the B pressure,
required to move the center of the membrane 1.1
mm against the soil. A third reading C ("closing
pressure") can also optionally be taken by slowly
deflating the membrane soon after B is reached. The
blade is then advanced into the ground of one depth
increment (typically 20 cm).
The pressure readings A, B are corrected by the
values ∆A, ∆B determined by calibration to take into
account the membrane stiffness (Marchetti 1999b)
and converted into p0, p1 as indicated in Table 1. The
data p0 and p1 are generally interpreted (in "normal"
soils) using the correlations reported in Table 1.
The DMT can test from extremely soft to very
stiff soils (clays with cu from 2 - 4 to 1000 kPa,
moduli M from 0.5 to 400 MPa).

The dilatometer consists of a steel blade having a
thin, expandable, circular steel membrane mounted
on one face. When at rest, the membrane is flush
with the surrounding flat surface of the blade. The
blade is connected, by an electro-pneumatic tube
running through the insertion rods, to a control unit
on the surface (Fig. 1). The control unit is equipped
with pressure gauges, an audio-visual signal, a valve
for regulating gas flow (provided by a tank) and vent
valves. The blade is advanced into the ground using
common field equipment, i.e. push rigs normally
used for the cone penetration test (CPT) or drill rigs.
(The DMT can also be driven, e.g. using the SPT
hammer and rods, but statical push is by far
preferable). Pushing the blade with a 20 ton
penetrometer truck is most effective (up to 100 m of
profile per day).

Figure 1. General layout of the dilatometer test


25 ϕsafe. Comparisons carried out at various National Research Sites by international research groups (see Figs 3-4) have shown quite good agreement between the profiles of cu DMT and MDMT and the profiles determined by other tests. First obtain MDMT = RM ED. Such factors are scarcely reflected e.85 u0 ≈ p2 ≈ C . since such tests measure soil responses instead of soil properties.2.5 KD) 1.g.6 log KD .DMT = (KD / 1. aging. ED should be used only AFTER combining it with KD (Stress History). then e. clay). A first glance at the KD profile is helpful to "understand" the deposit.g. while most of the penetration tests just provide one "primary" parameter (the penetration resistance) for the interpretation.∆B) p1 = B . deformability and (in sands) resistance to liquefaction. (b) The DMT provides two independent parameters.7 (p1 . e.0. Example of DMT results 36 the (variable) coefficients linking the 6 stress components to the 6 strain components. One measurement is a very small fraction of 36.47 .ZM . 2 . being Figure 2.6 RM = 0. − M (vertical drained constrained modulus) and cu (undrained shear strength) in the usual ways.RM. yet 100% more than one measurement.p0) / (p0 .56 c u. by cylindricalconical probes (Marchetti 1999a).5 KD) 1.15 (I D .14 + 0. The results are used as follows: − ID (Material Index) gives information on soil type (sand.0 = 0. − The profile of KD (Horizontal Stress Index) is similar in shape to the profile of the overconsolidation ratio OCR. by qc (cone penetration resistance from CPT) and by NSPT .g.0.22 σ'V0 (0. cementation.ZM + ∆A In freely-draining soils An example of DMT results is shown in Figure 2.18 log K D if RM < 0.1 Design using soil parameters In most cases the DMT is used to determine "commonly used" geotechnical design parameters.32 + 2.5 + 2 log K D if 0. E ≈ 0.2 for ID > 1. Yet such factors are of primary importance in determining some basic soil properties. It is known that in situ tests represent an "inverse boundary conditions problem".14 + 2.DMTA ≈ 7 cm2 / Tflex kh = c h γ w / Mh (Mh ≈ K0 MDMT ) (see Marchetti 1999a) MDMT = RM ED K0 OCR cu ϕ ch kh γ M u0 Vertical Drained Constrained Modulus Equilibrium Pore Pressure KD = (p0 .05 (A . an in situ test should be able to measure 36 responses. 4 DESIGN APPLICATIONS 4. also due to the arching phenomenon.0) log K D with RM. KD > 2 indicates overconsolidation.6 OCRDMT = (0.6 < I D < 3 RM = RM. silt. According to the theory.Table 1.5 .85 set RM = 0.2 for ID < 1. If ∆A & ∆B are measured with the same gage used for current readings A & B.u0) Coefficient Earth Pressure in Situ Overconsolidation Ratio Undrained Shear Strength Friction Angle Coefficient of Consolidation Coefficient of Permeability Unit Weight and Description K0. notably the undrained shear strength cu and the constrained modulus M.2 for ID < 1.1 log2 KD c h.∆B ID KD ED Material Index Horizontal Stress Index Dilatometer Modulus ID = (p1 . and in general.6) if K D > 10 RM = 0.0. Basic DMT reduction formulae SYMBOL DESCRIPTION BASIC DMT REDUCTION FORMULAE p0 p1 Corrected First Reading Corrected Second Reading p0 = 1. such as stress state/history.0 + (2.8 MDMT for ID < 1. set ZM = 0 (ZM is compensated) u0 = pre-insertion pore pressure σ'V0 = pre-insertion overburden stress ED is NOT a Young's modulus E. structure. Two measurements are also a very small fraction. 3 DMT vs OTHER PENETRATION TESTS (a) Comparative studies have indicated that the DMT results (in particular KD) are noticeably reactive to factors that are scarcely felt (especially in sands) by other tests.8 Tflex from A-log t DMTA-decay curve if I D ≤ 0. KD ≈ 2 indicates in clays OCR = 1.5)0.u0) / σ'v0 ED = 34.DMT = 28° + 14.ZM + ∆A) .ZM .p0) ZM = Gage reading when vented to atm.05 (B .DMT = 0.36 log K D if I D ≥ 3 RM = 0.

Schmertmann (1986a) reported 16 comparisons of observed vs DMT-calculated settlements at different sites and for various soil types.2 Settlement prediction Predicting settlements of shallow foundations is probably the application No. 1992). For this reason ED. The ratio calculated/observed settlement was 1. (a) Onsøy (Norway). The capability of taking into account σh is important. Japan). Tests performed by Kiso-Jiban Geotechnical Research Center. E can be derived from M via theory of elasticity (E ≈ 0. Since KD incorporates the effects of the horizontal stresses σh and stress history. The accuracy of settlements predicted by DMT has been confirmed by many investigators. Figure 5. If required. in general. such effects. Such correlation has also been confirmed theoretically by Finno (1993). since high σh dramatically reduce settlements (as observed e. Correlation KD-OCR for cohesive soils from various geographical areas (Kamei & Iwasaki 1995) 3 . through KD. Settlements are generally calculated by means of the one-dimensional formula: ∆σv ∆z (1) Z (m) 10 15 (a) 20 (b) Figure 3.3). DMT with ∆σv generally calculated according to Boussinesq and MDMT constrained modulus estimated by DMT using the correlation (see Table 1) MDMT = RM ED.8 M for ν = 0. (a) National Research Site of Bothkennar. leads to more realistic values of M. Comparison between M determined by DMT and by oedometer tests.0. then also MDMT incorporates. should not be used as such. (c) The availability of a second independent parameter KD. reflecting σh / stress history.75 to 1. Italy (AGI 1991). (b) The modulus obtained by expanding the DMT membrane (a "mini" load test) is physically more related to deformability than is the penetration resistance. since OCR governs many soil properties.30). M (MPa) 0 0 2 4 6 8 10 DMT 5 Oedometer 10 15 z (m) ∑M 5 20 z S= kPa kPa 0 Z (m) Figure 5 illustrates the good correlation between KD and OCR (note that KD ≈ 2 for OCR = 1). (b) National Research Site of Fucino. but should first be combined with KD to obtain M. by comparing the datapoints band amplitude (ratio between maximum and minimum) in Figure 6a (Hayes 1990) and in Figure 6b (Baldi et al.20 . Among the reasons: (a) Wedge shaped tips deform the soil considerably less than conical tips (Baligh & Scott 1975). Comparison between cu determined by DMT and by other tests.g. among many others. Similar agreement was observed. Note that ED (despite the symbol) should not be confused with the Young's modulus E. Several studies have indicated that the DMT reduces the uncertainty in settlement predictions by a factor of over 3 compared with predictions based on penetration resistance qc from CPT. with a narrow variation range (from 0. 25 30 35 (a) (b) 40 Figure 4. UK (Nash et al. because it lacks the stress history information contained in KD. OCR profiles are usually hard and costly to obtain. This can be observed e. (b) Komatsugawa (Tokyo. on the other hand. by Massarsch 1994). especially in sands. 1 of the DMT.4.18 on average. 1988). while. A comparison between profiles of OCR estimated by DMT and by other tests at the Bothkennar Research Site is presented by Marchetti (1997). The ability to estimate OCR is important. Tests performed by Norwegian Geotechnical Institute.g. where RM is a function primarily of KD. by Lacasse & Lunne (1986) and Sallfors (1988). where undisturbed sampling and estimating deformability parameters are particularly difficult.

Figure 8. Ratio MDMT /q c before and after compaction of a loose sand fill (Jendeby 1992) 4. with loss of structure. aging or cementation. Ratio E/qc as a function of Dr and OCR . In many cases. On the other hand.00 M DMT / qc 10 20 30 5 before compaction 10 Figure 6a. these layers are likely part of a slip surface (active or quiescent). The higher "sensitivity" of MDMT compared to qc was also observed. Note that the method involves searching for a specific numerical value (KD = 2) rather than for simply "weak zones". 8): − the process of sliding and reconsolidation generally creates a remolded zone of nearly normally consolidated (NC) clay. if an OC clay slope contains layers where KD ≈ 2. − since in NC clays KD ≈ 2. Several studies present comparisons of results of CPTs and DMTs performed before/after a compaction treatment. Moreover. The "KD method" provides a faster response than inclinometers in detecting slip surfaces (no need to wait for movements to occur). Figure 7. (1997). Similar results were observed by Jendeby (1992) in monitoring deep compaction of a loose sand fill by "vibrowing" (Fig. Schmertmann (1986b) observed that the increase in MDMT after dynamic compaction of a sandy soil was approximately twice the increase in qc (CPT). The method proposed by Totani et al. but even more in MDMT . is particularly suitable to monitor soil improvement. (1993) described the use of DMT performed before/after the installation of Atlas piles. after an excavation. unlike inclinometers. DMT and inclinometers can be used in combination (e. which could be detected just as easily also by other in situ tests. 1988) 4. the method itself.3 Compaction control The DMT.Ticino Sand (Baldi et al. The DMT has also been used to check the effects induced in the soil by the installation of various types of piles. which could reactivate e.g. the method enables to detect even possible quiescent surfaces (not revealed by inclinometers).4 Detecting slip surfaces in clay slopes The DMT permits to verify quickly if a slope in overconsolidated (OC) clays contains active or old slip surfaces. All the above observations indicate that the DMT results are noticeably reactive even to slight changes of σh (or relative density) in the soil. based on inspection of the KD profiles. does not permit to establish if the slope is moving at present and what the movements are. Comparison between observed and DMT-calculated settlements (data by Hayes 1990) Figure 6b. as shown by the large increase of the ratio MDMT /q c . KD method for detecting slip surfaces in OC clay slopes 4 . by Pasqualini & Rosi (1993). is founded on the following basis (Fig. the DMT is particularly suitable in cases where the expected stress variations are very small (e. due to its sensitivity to σh. in a vibroflotation case. Compaction increases both σh and Dr. Measurements "after" indicate an increase in q c . 7). and concluded that the DMT detects more clearly than the CPT the effects of the installation. Therefore.g. De Cock et al. relaxation upon "microboring"). use KD profiles to optimize location/depth of inclinometers).g.

Dr Figure 10.0 high seismicity areas (amax/g = 0. suggests that a clean sand (natural or sandfill) is adequately safe against liquefaction (M = 7. which. (The DMTA dissipation is perfectly analogous to the "holding test" by pressuremeter). while KD is fairly sensitive to the past stress-strain history.2 − medium seismicity areas(amax/g = 0. smearing. Case histories presented by Totani et al. coefficient of consolidation ch and permeability k h in clay by means of dissipation tests. KD 4. Robertson (1998) warns that the correlation qc . A number of independent validations (NGI.CRR. Figure 9. Various procedures have been formulated (DMTC. In fact. 4. 1988. while for medium to high risk projects he recommends to estimate CRR by more than one method. Sladen (1989) has shown that ignoring the non-unicity of the correlation qc .35): KD > 5. Georgia Tech and tests in Virginia sediments) have indicated that the two methods provide similar predictions. The non-unicity of the correlation qc .7 Consolidation and flow parameters The DMT permits to estimate the horizontal KD Figure 9.5 RECOMMENDED CURVE Kemigawa Ko ≈ 0. Once CRR has been evaluated from Figure 9.Dr for NC sands (Reyna & Chameau 1991). Marchetti & Totani 1989). 1997) that the relation qc . (b) absence of problems of saturation. Determining ch and k h from DMT dissipations presents various advantages over the piezocone: (a) lower distorsion induced in the soil by the penetration of the blade. hence large errors in CRR estimated from qc. the ones recommended by the authors are those by Robertson et al. All methods are based on the decay with time of σh total against the membrane after stopping the blade at a given depth. where Dr was determined on high quality frozen samples. Figure 9 permits to estimate CRR as an alternative to the methods which derive CRR from NSPT or qc.CRR may be adequate for low risk.6 Liquefaction Figure 9 summarizes the available knowledge for evaluating sand liquefiability by DMT.SP (SP = state parameter) is not unique. Curves for estimating the cyclic resistance ratio CRR from KD (Reyna & Chameau 1991) Horizontal stress index.SP.7 − low seismicity areas (amax/g = 0. Yu et al. The coefficient of permeability k h is then determined from ch and MDMT (see formulae in Table 1). Moreover. ch is estimated from the time Tflex at which the S-shaped decay curves A . based in part on their curve KD Dr (relative to NC sands) in Figure 10. DMTA. since it has been recently demonstrated (Sladen 1989. 5 . Correlation KD . The shaded areas represent datapoints obtained by Tanaka & Tanaka (1998) on frozen samples. on the other hand.log t exhibit a contraflexure point. (1991). This correlation has recently been confirmed by additional datapoints KD . The curve recommended to estimate the cyclic resistance ratio (CRR) from the parameter KD is the curve by Reyna & Chameau (1991). (1998) have indicated that ch values from DMTA are generally 1 to 3 times smaller than ch back-calculated from "real life" observations.5 earthquakes) for the following KD values: − non seismic areas: KD > 1.15): KD > 4.4.5 Laterally loaded piles Of the various methods developed for deriving P-y curves from DMT results. in combination with the available experience.45 in all cases Ohgishima Relative density. The possibility of obtaining independent evaluations of CRR is of great interest. it is used in liquefaction analysis with the methods developed by Seed. (c) "integral" rather than "punctual" . (1987) and Marchetti et al.SP in design can lead to catastrophic consequences.measurement. experimental work over the last 20 years (an overview is presented by Marchetti 1999c) has shown that.Dr obtained by Tanaka & Tanaka (1998) at the sites of Ohgishima and Kemigawa. The DMTA method (probably more used than DMTC) consists of taking a timed sequence of A readings until stabilization. due to the strong link SP CRR (SP governs the attitude of a sand to increase or decrease in volume when sheared) involves large scatter in the correlation qc .25): KD > 5. but strongly dependent on the stress level. qc is scarcely reactive to this factor. in quite good agreement with the observed behavior. greatly increases resistance to liquefaction. filter clogging. Robertson et al. small scale projects.

123. Conf. R. Geotechnical Engineering Conference. Dilatometer Based Liquefaction Potential of Sites in the Imperial Valley. Soils and Foundations.S. Vol.. De Cock. DFI. Atlanta. Proc. Atlas screw piles and tube screw piles in stiff tertiary clays. No. F. & Tanaka. ASCE Jnl GE. Final edition (ENV 1999-3). K. P. No. Massarsch. This approach has the shortcoming of involving. M. E. Penetration Testing in the UK .10 .K. S. XIII ICSMFE. Calabrese. S. 5 CONCLUSIONS Based on the available experience. Totani.. T. & Campanella. Robertson. The Flat Dilatometer: Design Applications. Tanaka.F. & Scott. Ghionna. P. of Birmingham. Current status of the Marchetti dilatometer test.. Soc. Proc. Stresa. 1: 281-286. Calabrese. 106. Rome. Géotechnique 42. Jul.. adjusting FEM input parameters to improve the matching of MFEM vs MDMT . Validity of compression modulus determined by dilatometer tests. Excess Pore Pressure and the Flat Dilatometer Test. J. 1993. 985-1001. Robertson. Keynote Lecture. P.G. Calabrese. The model is elementary. (f) (OCR and K0 in sand).M. Evaluation of undrained shear strength of cohesive soils using a Flat Dilatometer. at the same time. No.. Geotechnical characterization of Fucino clay. on Recent Advances in Geot. Testing Jnl. M. Proc. R.. F. & Totani. & Collins. yet it could help avoid gross mistakes in the FEM analysis. Conf.. Cairo University. Nash. 1998. Blacksburg. Proc. FHWA -SA-91-044. 1995. S. Vol. 4th Int. 1988. Vol. 1986b. J. 9: 886-888. Schmertmann. 1989. 86 pp. Pasqualini. VA Tech. Marchetti. No. Vol. No. 1994. (c) Monitoring soil improvement. Baldi. 2: 883-888. No. Design of Laterally Loaded Driven Piles Using the Flat Dilatometer. 1: 19-24. The Flat Dilatometer and its Applications to Geotechnical Design. Kamey. No. Reyna. Proc. No. Schnaid. Earthquake Engrg. 42: 299-305. Closure to discussion on "Analysis of Cone Pressuremeter Tests in Sands". XI Afr. (b) Settlement prediction. H.J. 1995. Proc. Campanella.K. Rio de Janeiro. P. 2: 93-101.. Proc. L'Aquila University. F. D. Marchetti. 241-254. Vol. Dilatometer to compute foundation settlement. Vol. P. Vol. Vol. L. (e) Distinguish freelydraining from non freely-draining layers. 1993. 303-321. Section 9: Flat Dilatometer Test (DMT). VA Tech. & Iwasaki. 1: 137-155. Sladen. (c) P-y curves for laterally loaded piles. M.P.. & Chameau. SMFE. Hamburg. Proc. Schmertmann. Vol. X ECSMFE. Vol. 1986a. 421-448.. Southern Ont. Hamza. Baligh. 1991. Vol. K. Special Lecture. 1992. Geot. Eurocode 7: Geotechnical design.F. M.Y. Deep Compaction by Vibrowing.M. Sallfors. I. In Situ '86 ASCE Spec. The Flat Dilatometer Test. Second Int. S. Univ. J. In Situ '86 ASCE Spec. ISOPT-1. can provide fairly accurate solutions. & Monaco. Conf.L.. Ghent. Vol. 1: 607-610. S. ASCE Jnl GGE. R. Orlando. 1988. Cairo. 1989. Vol. Seminar on "In Situ Testing and Monitoring". Washington. In situ determination of c h by flat dilatometer (DMT). Finno. Japanese Geot. 35.L.g. G. Two-day Seminar at NGI on Calibration of In Situ Tests. R. REFERENCES AGI (Associazione Geotecnica Italiana) . 21 pp. A. 1: 325328. & Miran. Stiffness of Sands from CPT. Davies. ISOPT-I. Characterization of Sandy Soils using CPT and DMT. 2: pp.8 MDMT . VA Tech.H. Society. & Jamiolkowski.. Paper No. Lacasse. 1992. S.J. J. Conf.K. 2: 323-332. 1986. 1: 567-576. Sept. Soils and Foundations. Proc. 1999. Marchetti. G. P. I. ASTM Subcommittee D 18. J. 1: 30-38. Proc. Unpublished report. P-y curves from DMT data for piles driven in clay. 686-699. Lecture notes at the International Seminar on the DMT held at the Japanese Geotechnical Society... S. Quasi Static Deep Penetration in Clays.R. Tokyo. CEN (European Committee for Standardization). Marchetti. J. Marchetti. Totani. 35. Proc. with a judicious choice of the parameters. 1992. J. 1999a.02. Third Int. SPT and DMT. Section Canad. Van Impe. 1988. Proc. The Marchetti Dilatometer and Compressibility. Vol. 12 Feb 1999.. In Situ '86 ASCE Spec. 3: 299-321. Vol. Gillespie. Correlations of DMT.F. GT11: 1119-1133. 2: 163-181. Canad Geot. No. L'Aquila University. No. Totani. 1999c. CPT and SPT in Nile Basin Sediment.G. J.A. In Situ Tests by Flat Dilatometer. S. G.8 Use of DMT for FEM input parameters Two approaches have been considered so far. V. many additional (unknown) parameters. Proc. (d) Recognizing soil type. Conf. Blacksburg. J. 1991. 1997.J. (e) Friction angle in sand. (a) Model the dilatometer test by a FEM computer program by adjusting the input parameters until the DMT results are correctly "predicted". No. H. On the calibration of the DMT membrane. Testing Journal. Bellotti. Experiences on a vibroflotation treatment. 237-240 (in italian). (b) Coefficient of consolidation and permeability. XII ICSMFE. but often even simple models. T. 437-446. Proc. G. 3: 442-459. 1975.A. Suggested Method for Performing the Flat Dilatometer Test. 102 pp. Dilatometer Tests in Sand. (f) Verify if a clay slope contains active/old slip surfaces. Marchetti. J. give an initial "tentative" set of input FEM parameters. 1999b. 1999.4. 9. Unpublished report. G. Proc. Blacksburg. Sand liquefiability assessment by DMT. (Chair) 1986. ICE.Various Authors. 1988. C. Part 3: Design assisted by field tests. 1980. & By. Settlement Analysis of Compacted Granular Fill. Problems with interpretation of sand state from cone penetration test. R. Geotéchnique 43. Robertson. & Lunne. Publ.F. 1: 263-272.H. CPT/DMT Quality Control of Ground Modification at a Power Plant. and Soil Dyn. This approach is less ambitious. Proc. 1997.. DC. 38. 1993. 3: 55-65.M. on Piling and Deep Foundations. Proc. oedometer). & Monaco. 101. 1990. the DMT best applications are: (a) M and cu profiles. An example of use of deformation parameters determined by DMT in design of underground structures (Cairo metro tunnels) is illustrated by Hamza & Richards (1995). New Delhi... ASCE Jnl GED. G. First International Conference on Site Characterization ISC '98. 1998. No. ASTM Geotechn. XIV ICSMFE. & Rosi. Yu. Evaluating cyclic liquefaction potential using the cone penetration test. Geotéchnique 39. Geotechnical Engineering. M. 1: 27-40. M.H. 1997. Proc. Marchetti. (d) Sand liquefiability. 1987. Louis. Conf. Florence. & Peiffer H. 12.P. S. Annual Meeting CNR Res. BAP II. Use of in situ flat dilatometer (DMT) for ground characterization in the stability analysis of slopes. & Richards. (b) Based on the soil information available. Geot. Orlando. W.. Hayes. Powell. 2: 111-116.. The DMT also gives useful information on: (a) OCR and K0 in clay. T. 1998.Briaud. Then simulate by FEM a simple laboratory test (e. Lutenegger. Marchetti. M. Analytical Interpretation of Dilatometer Penetration Through Saturated Cohesive Soils. Ch Evaluations from DMTA Dissipation Curves. 6 . D. Their numerical analyses adopted the simplest possible model (linear elastic). 359-367. Proc.Schmertmann. Vol. M. D. Jendeby. Vol. Initial investigations of the soft clay test site at Bothkennar. St. Nordic Geotechnical Meeting NGM -92. & Monaco. 1991. & Lloyd. US DOT for FHWA . with E ≈ 0.