SUMMARY

  • Stress history is important, as it significantly increases moduli and, in sand, liquefaction resistance.
  • If Stress History is not felt, its benefits are ignored, leading to an uneconomical design.
  • It is well established that the ED parameter by DMT cannot be used as such, but must be “leveraged” by the stress history index KD in order to estimate M. Similarly, for liquefaction, CRR predicted by CPT cannot be used as such, but needs to be leveraged by KD (Marchetti 2016).
  • In clay: the 1980 OCR correlation OCR = (0.5 KD)1.56 generally provides reasonable OCR estimates in “textbook” clays.
  • In sand: OCR can be evaluated if both Qc and KD are available. OCR cannot be evaluated from Qc alone or from KD alone.
  • In sand: the DMT parameter KD in sand is considerably more sensitive to Stress History than parameters from other in situ tools.

STRESS HISTORY-OCR IN SAND

Experience has shown that the Stress History parameter KD by DMT is considerably more sensitive to Stress History than other in situ tools. This sensitivity of KD is important, since stress history significantly reduces settlements and increases the liquefaction resistance. There are not many alternatives to KD for obtaining in situ information on stress history, especially in sand. On the other hand, in absence of adequate information on stress history, the benefits of stress history on stiffness and on liquefaction resistance are not felt and therefore ignored, leading to a more expensive design.

The first researcher to point out the higher sensitivity of DMT to stress history was Schmertmann (1984), who noted “the cone appears to destroy a large part of the modification of the soil structure caused by the overconsolidation and it therefore measures very little of the related increase in modulus. In contrast the lower strain penetration of the DMT preserves more of the effect of overconsolidation. Using the CPT to evaluate modulus changes after ground treatments may lead to a large overestimate of the settlement“.

The higher sensitivity to stress history (including aging) of KD, compared with the sensitivity of Qcn, has been confirmed by numerous researchers (e.g. Schmertmann et al. 1986, Jendeby 1992, Jamiolkowski and Lo Presti 1998, Monaco & Schmertmann 2007, Monaco and Marchetti 2007, Marchetti 2010, Kurek and Balachowsky 2015).

A comprehensive calibration chamber research project, specifically aimed at comparing the sensitivity of CPT and DMT to stress history, was carried out in Korea (Lee et al. 2011)  on Busan sand. Fig. 1 compares the effects of stress history on Qcn and on KD obtained by executing CPT and DMT on normally consolidated and on overconsolidated sand specimens. The different increase in Qcn and in KD shown in the two diagrams highlights the much higher sensitivity of KD to Stress History.

The higher sensitivity of the DMT to Stress History emerges also when monitoring compaction, which is a way of imposing stress history. Schmertmann et al. (1986) noted in a compaction job “MDMT increased much more than Qc after the ground modification work, with an average MDMT gain about 2.3 times the Qc gain“. A similar gain ratio, higher than 2, was found by Balachowski and Kurek (2015).

Experience has shown that compaction increases both Qc and MDMT, but MDMT  at a faster rate.

Fig.-1
Fig. 1. Sensitivity of CPT and DMT to Stress History (Lee et al. 2011).

It is noted:

  • qc reflects essentially Dr, and only to a minor extent OCR / stress history (see Fig. 1). (Though the qcDr relation is sand dependent).
  • KD reflects the total effect of Dr plus various stress history effects such as OCR, aging, Ko, structure and possibly cementation (more on the effects of cementation).
  • Fig. 1 clearly shows that it is impossible to estimate OCR in sand from CPT alone or from DMT alone, i.e. shows the necessity of both qc and KD to evaluate OCR in sand. In fact Fig. 1(a) shows the low sensitivity of Qcn to OCR and Fig. 1(b)  shows that if only KD is known and is entered in Fig. 1(b), its value could be due to a low relative density Dr and a high OCR or to a high Dr and a low OCR. In order to evaluate OCR, qc (well reflecting Dr) must also be available to provide an indication of Dr on the horizontal axis of Fig. 1(b). In conclusion both Qcn and KD are necessary to evaluate OCR in sand (Fig. 2).
  • Stress History (reflected by KD) produces a small increase in penetration resistance, but a significant increase in CRR and in stiffness of a cohesionless soil.
  • When OCR (or compaction) is applied, modulus increases much faster than strength. E.g. Yoshimi (1975): “Upon initial loading, the NC sand specimens were at least six times more compressible than the prestressed sand” or Lambrecht and Leonards (1978): “Prestressing increased the modulus by one order of magnitude, while Qc had only a slight increase“. Hence OCR (or compaction) cause an increase of the ratio MDMT/ Qc (Fig. 2). More information on estimating OCR in sand.
Fig.-2

OCR = 0.0344 (MDMT /qt)2 – 0.4174 (MDMT /qt) + 2.2914

Fig. 2. Correlation OCR = f(MDMT / Qc) for the Venice Lagoon Sandy layers (Monaco et al. 2014).

OCR IN CLAY

The Marchetti 1980 correlation for OCR in clays was OCR = (0.5 KD)1.56. It was obtained experimentally by interpolating a line through the then available high quality KD – OCR datapoints (Fig. 3a) . In 1995 that correlation was experimentally reconfirmed (Fig. 3a) by Kamey and Iwasaki (1995), based on KD – OCR datapoints from various clays they had investigated. In 1993 and in 2004 the correlation was independently confirmed theoretically by Finno (1993) and by Yu (2004), who used two different theoretical methods (Fig. 3b  and 3c). Thus the 1980 OCR correlation has “empirical” and “theoretical” roots, and appears a well founded average, generally able to provide reasonable estimates of OCR in average “textbook” clays.

 

finno-ecc-tris

 

Fig. 3. Correlations between OCR and KD in clay: (a) Experimental (Marchetti 1980 and Kamey and Iwasaki (1995); (b) Theoretical by Finno (1993);  (c) Theoretical by Yu (2004).

 

An instructive case history in clay: OCR from KD at Bothkennar

An instructive case history concerning the ability of DMT to predict OCR in clay is available for the UK National Research Site of Bothkennar. The results of a first careful investigation (high quality piston samples, Laval samples, Landva’s preparation technique, small increment oedometers etc..) were published in Geotechnique (Nash et al. June 1992). The published results (p.171, Fig. 8b – reproduced herein as Fig. 4a) included OCR estimated by various methods and also OCR estimated by DMT. On the base of all the results the Authors  concluded “it is apparent that OCR is approximately constant with depth” (see “Suggested Profile” in Fig. 4a). This was somewhat in contrast with the OCR profile predicted by DMT (also in Fig. 4a), exhibiting at 10 m depth a reentrance suggesting a “a mild buried crust”. After many additional tests, one year later (1993), the organizers sent to many investigators around the world updated data on Bothkennar, including a re-edited OCR profile (Fig. 4b). In this new edition, following a supplement of investigation, the “ reentrance and the mild buried crust” signalled by DMT was confirmed.

 

Conclusion: in this case details of the OCR profile were initially missed by a high quality investigation. An additional investigation confirmed one year later OCR details that DMT had predicted at a fraction of the time and cost.

 

bothkennar_1992_&_1993

Fig. 4 (a). Yield Stress Ratio from Incremental Load Consolidation Test (Nash et al. 1992); (b) OCR after additional investigations.

Print this pageEmail this to someoneShare on FacebookShare on Google+Tweet about this on TwitterShare on LinkedIn
Main menu