Geotechnical Testing Journal, Vol. 31, No. 1 Paper ID GTJ101007 Available online at: www.astm.

org

Prsio L. A. Barros1 and Paulo R. O. Pinto1

Oedometer Consolidation Test Analysis by Nonlinear Regression

ABSTRACT: A numerical method based on least squares nonlinear regression for the evaluation of the consolidation parameters of soils from consolidation tests is presented. A model which includes the initial compression, the primary consolidation, and the secondary compression is used in the regression. This approach allows the resulting regression curve to better t the experimental data. The method takes the settlement-time readings from the oedometer step-loading consolidation test and calculates automatically the magnitudes of the coefcients of consolidation and of secondary compression. The performance of the proposed method is accessed through consolidation tests executed on four different clay soils, which are analyzed by nonlinear regression and by the usual graphical methods. It is concluded that the proposed method gives results that are close to those obtained by the standard methods of analysis. KEYWORDS: consolidation test, nonlinear regression, least squares, coefcient of consolidation, secondary compression

Introduction
The oedometer step-loading consolidation test is one of the most widely used tests in the soil mechanics laboratory. Introduced by Terzaghi as an experimental support for his one-dimensional consolidation theory (Terzaghi 1943), the oedometer test has remained essentially unchanged since then. The main objective of the consolidation test is to access the consolidation characteristics of a soil from the measured settlementtime curve. From the consolidation theory, those characteristics are expressed by the soil coefcient of consolidation cv and total consolidation settlement 100. The evaluation of cv and 100 from the measured consolidation curve is generally performed by hand-draft curve tting, being the t method (Taylor 1948) and the Casagrandes Taylors log t method (Casagrande and Fadum 1940), the most used. Those are regarded as standard methods of cv evaluation. Other methods were developed later for this same purpose, such as the rectangular hyperbola method (Shidharan and Prakash 1985) and the early stage log t method (Robinson and Allam 1996). But all the curve tting methods above require graphical constructions that introduce undesirable subjective interpretations in the process. The various curve tting methods lead, in general, to results that are different from each other because each one focuses on different portions of the consolidation curve. The main cause of these discrepancies is the experimental consolidation curve departure from theory, which is mainly caused by: There is an initial settlement 0 just after the load application, which is due to incomplete sample saturation, conning ring expansion, and deformation of the loading apparatus. The settlement continues after the theoretical end of consolidation in a process known as secondary compression.
Manuscript received January 15, 2007; accepted for publication July 16, 2007; published online September 2007. 1 Associate Professor, Department of Geotechnics and Transportation, and Graduate Student of Civil Engineering, respectively; State University of Campinas, Brazil.

The initial settlement is easily treated by the various graphical methods and poses no special difculties to the analysis. On the other hand, secondary compression interpretation is much more challenging for there is no well established theoretical model for it. Even the point in time when secondary compression may rst be detected is subject to controversy (see, e.g., Robinson 2003). Added to those discrepancies, the observed consolidation curve may also differ from the theoretical model due to the nonlinear behavior of the soil compressibility during the load increment and even due to limitations of the test equipment and instrumentation. It is worth noting that if the soil behavior during the consolidation followed Terzaghis consolidation theory then all those tting methods would lead to the very same results. But the deviations from theory make the result dependent on the particular aspect of the theoretical behavior each tting method arbitrarily takes into account. In this way, a less arbitrary procedure, with a sounder statistical base, like those based on least squares regression, is desirable. The automatic interpretation of the consolidation curve through least squares regression was implemented in recent works (Robinson and Allam 1998; Chan 2003; Day and Morris 2006). The proposed method adds to those the inclusion of secondary compression in the regression model. As a consequence, the resulting regression curve ts better to the nal points of the experimental consolidation curve. Thus, the effects of the secondary compression on the resulting cv value can be controlled. Furthermore, the regression formulation developed here can be easily implemented in spreadsheet programs which are available in most soil mechanics laboratories. The main benets of the proposed method are: The need for human intervention in the interpretation process is kept to a minimum. In this way, manual calculation errors are avoided. The interpretation process takes much less time, so the laboratory technicians can focus their attention on the test execution procedures. The values of the consolidation parameters obtained are essentially free of subjectivity, which makes correlations between them and other soil parameters more consistent.

Copyright by ASTM Int'l International, 100 Mon Aug 1 Drive, PO Box C700, Copyright 2008 by ASTM(all rights reserved);Barr Harbor 20:01:13 EDT 2011 West Conshohocken, PA 19428-2959. Downloaded/printed by Universidade Federal do Rio Grande do Sul pursuant to License Agreement. No further reproductions authorized.

2 GEOTECHNICAL TESTING JOURNAL

All measured points are considered in the analysis. The secondary compression characteristics of the soil are also obtained in the process. It is possible to couple an automatic data-logging system to the evaluation method, resulting in a completely integrated system for logging, calculating, and interpreting the test results. pressure measurements (Robinson 2003), showed secondary compression beginning, as indicated by the pore pressure measurements, from about 65 % of primary consolidation for organic clays to about 95 % for inorganic clays. Unfortunately, it is not feasible to identify the beginning of secondary compression from the standard oedometer test results. Thus, the value of t0 should be arbitrarily xed. If t0 is taken as a function of U T , then a relation between t0 and cv can be established: t0 = TBoSH2 d cv (6)

Regression Model and Evaluation of Parameters

The nonlinear regression model considered here comprises three parts: 1. The initial instantaneous settlement 0. 2. The primary consolidation p. 3. The secondary compression s. Then, the settlement at any given time t is given by: t =
0

t +

(1)

The primary consolidation is given by Terzaghis onedimensional consolidation theory (Terzaghi 1943), expressed by:
p

t =

100U

(2)

where 100 is the total primary consolidation settlement and U T is the degree of primary consolidation, given by: U T =1 and T is the time factor: T= c vt H2 d (4) 8
2 m=0

1 e 2m + 1 2

2m + 1 2 2 T 4

(3)

where TBoS is the time factor corresponding to the model beginning of secondary compression. The choice of t0 (or of TBoS) do have some inuence on the resulting cv and 100. Smaller values of t0 lead to larger values of cv and to smaller values of 100. This inuence depends on the amount of secondary compression exhibited by the soil. This limitation is common to all curve tting methods. But for the method proposed here, since the model beginning of secondary compression is explicitly chosen by the user, the inuence of the secondary compression on the calculated cv values can be assessed. Furthermore, the application of the results to eld problems can take this choice into account. It is worth noting that Taylors method implicitly assumes that no secondary compression occurs before T90 (the time factor corresponding to 90 % of primary consolidation), and the graphical construction used for the 100 evaluation in Casagrandes method implies that the secondary compression, if present, should be linear with log t and should begin at around T95. The complete model resulting from the superposition of the three parts is expressed by: t = +
100U

where cv is the coefcient of consolidation and Hd is the drainage height. For the secondary compression part, a number of alternatives are available. Rheological models (Gibson and Lo 1961; Whals 1962), can describe very well the secondary compression, but require long-term tests to fully evaluate the model parameters. For the standard short-term tests, which takes 24 hours for each load step, empirical models are preferred. Among the empirical models, the linear-log t model (Buisman 1936) is the older, simpler, and most used. According to the linear-log t model, the secondary compression can be expressed by:
s

c vt H t log10 max 1, 2 +c 1 + e0 t0 Hd

(7)

with model parameters 0, 100, c , and cv. The model expressed by Eq 7 is linear with respect to 0, 100, and c , and nonlinear only with respect to cv. This fact allows for a specialized nonlinear least squares regression procedure. The least squares procedure is expressed by the minimization:
N 2

min
0, 100,c ,cv i=1

ti

(8)

t =c

H t log10 1 + e0 t0

(5)

where c is the coefcient of secondary compression, H is the sample height, e0 is the initial void ratio, and t0 is the time when the secondary compression is supposed to begin. The previous equation is valid only for t t0; for t t0 no secondary compression is supposed to occur. It is important to note that t0 may differ substantially from the actual beginning of secondary compression; t0 is the time when the linear-log t secondary compression model is assumed to begin. Therefore, t0 is called the model beginning of secondary compression. The time of beginning of secondary compression t0 is usually dened in terms of U T . Special consolidation tests with pore

where ti , i are the time-settlement pairs measured during the consolidation test and N is the number of readings in the load increment stage being analyzed. Since the model is linear with respect to all but one parameter, the problem reduces, through the use of derivatives with respect to each parameter, to a system of three linear simultaneous equations coupled to one nonlinear equation. Then, the solution of the complete system of equations can be obtained using a simple 3 3 system of linear equations solver coupled to a numerical root nding method. The proposed regression method can be implemented in common spreadsheet programs in a very straightforward way, either through macro-programming or entirely with spreadsheet built in functions. In the last case, most spreadsheet programs do incorporate in their standard conguration a solver for systems of linear equations and also a root nding facility called goal seek, which can be used in the procedure. Derivation details for the proposed method are given in the Appendix.

Copyright by ASTM Int'l (all rights reserved); Mon Aug 1 20:01:13 EDT 2011 Downloaded/printed by Universidade Federal do Rio Grande do Sul pursuant to License Agreement. No further reproductions authorized.

BARROS AND PINTO ON OEDOMETER CONSOLIDATION TEST ANALYSIS TABLE 1Clay samples characteristics. Sample 1 2 3 Description Bluish-gray clay Commercial bentonite Organic clay Gs 2.63 2.75 1.95 LL 46 % 237 % 230 % PI 23 % 188 % 46 %

As an alternative, one can use Eq 8 directly as input to a general minimization routine, also found in some spreadsheet programs. This approach has some drawbacks when compared to the procedure used here, though. An initial guess for all four model parameters is required when using a general minimization routine, whereas for the specialized procedure, only an initial guess for cv is required. Moreover, general minimization routines are more illconvergence prone. That is, the minimization procedure can fail to converge, or converge to unsatisfactory values. The proposed procedure is, by contrast, much more stable and robust, because the nonlinear minimization is performed in only one direction. In the other three directions the minimization is accomplished by the solution of a simple system of linear equations.

FIG. 1Consolidation curve for load increment from 100 kPa to 200 kPa, for Specimen 1.

Consolidation Tests
In order to access the performance of the proposed regression method, a series of consolidation tests on different types of clay soils were executed. Three clay samples were used. The rst one is a low plasticity, bluish-gray clay from a deposit near the city of Santa Gertrudes, in the State of So Paulo, Brazil. The second clay is a commercial bentonite sample, from the State of Paraiba, Brazil. The last sample is a highly organic, black, marine clay, from the city of Cubato, State of So Paulo, Brazil. The main characteristics of those samples are given in Table 1, where Gs is the specic gravity of soil solids, LL is the liquid limit, and PI is the plasticity index. Four remolded soil specimens for the consolidation tests were prepared with those three samples. The rst one was prepared with Sample 1 only. The second and third specimens were prepared from mixtures of Sample 1 and Sample 2. Finally, Specimen 4 was prepared with Sample 3 only. The specimens composition is shown in Table 2. The specimens, measuring 71.4 mm in diameter and 20 mm in height, were prepared for the tests by putting the soil mixture at a water content in the plastic range and then gently molding by hand a small block from which the specimen was cut with the test ring. In this way, very soft, saturated specimens for which the magnitude order of the coefcient of consolidation remained constant for all load increments could be molded. The specimens initial water content w0 and void ratio e0 are also shown in Table 2. The consolidation cell uses a xed-type ring setup with drainage on both top and bottom ends of the specimen. The consolidation tests were conducted in the standard way, with load increment ratio LIR= 1, with rst load q1 = 12.5 kPa and last load q8 = 1600 kPa.
TABLE 2Specimens composition and initial data. Specimen 1 2 3 4 Bluish-gray clay 100 % 75 % 50 % 0% Bentonite 0% 25 % 50 % 0% Organic Clay 0% 0% 0% 100 % w0 36.9 % 63.4 % 79.6 % 177.1 % e0 1.00 1.72 2.23 3.33

After each load increment, settlement measurements at times t in a geometric sequence with ratio ti / ti1 = 2 were taken. For those measurements, a digital dial gage with 0.001 mm resolution, linked to a data logging program in a computer was used. Each load increment was left on the soil specimen for 24 hours, or until the secondary compression was clearly dened. After the tests, the datasets were processed both by hand calculation and by the proposed regression method. The hand calculation was carried out independently by two experienced laboratory technicians; one of them using Taylors square root of time method and the other using Casagrandes logarithm of time method.

Results and Discussion

The plots in Figs. 14 show the measured points along with the regression curves obtained with the proposed method, for the load increment from v = 100 kPa to v = 200 kPa, for the four specimens. The agreement between the data points and the regression curves is excellent in all cases. The diversity in the curve characteristics for each soils is worth noting. Specimen 1 curve shows small deformability and high coefcient of consolidation cv. As the content of bentonite in the soil is increased in Specimens 2 and 3, the deformability increases and the magnitude of cv decreases. The coefcient of secondary compression c also increases with the bentonite content.

FIG. 2Consolidation curve for load increment from 100 kPa to 200 kPa, for Specimen 2.

Copyright by ASTM Int'l (all rights reserved); Mon Aug 1 20:01:13 EDT 2011 Downloaded/printed by Universidade Federal do Rio Grande do Sul pursuant to License Agreement. No further reproductions authorized.

4 GEOTECHNICAL TESTING JOURNAL

FIG. 3Consolidation curve for load increment from 100 kPa to 200 kPa, for Specimen 3.

For the organic clay (Specimen 4), the consolidation curve shows high deformability and also high values for both cv and c . The almost absence of the inexion point in this last curve should be noted. This fact makes the application of the standard graphical methods to the analysis very difcult, specially the Casagrandes method. But the regression method is capable of dealing with this type of situation without much trouble. The application of the proposed method to the analysis of the other load increments, not shown here for brevity, presented similar results. Figures 58 show the variation of cv as a function of the applied vertical effective stress v for the four specimens, calculated for all the load increments by the two standard manual methods and by the automatic least squares regression method. It can be seen that the results obtained by the three methods are not far from each other, with exception of the cv values obtained for the organic clay. For that soil, the differences between the cv values obtained with the three methods as well as the cv variation with the applied vertical stress is larger than for the other specimens. In all cases, however, the regression method gives results that are in the range of variation of the values obtained by the standard methods. The magnitude of the coefcient of secondary compression c is also obtained with the proposed method. Figure 9 shows the variation of c with the vertical effective stress v for the four specimens. The plot in Fig. 9 shows that c is heavily dependent on the soil type, but is less dependent on the applied load magnitude. As expected, the larger values of c were obtained for the organic clay,

FIG. 5Coefcient of consolidation cv for Specimen 1 evaluated by different methods.

whereas the smaller values were obtained for the low plasticity clay. Also, the effect of the bentonite content on c can be clearly observed in the plot, increasing c by a large amount. It should be

FIG. 6Coefcient of consolidation cv for Specimen 2 evaluated by different methods.

FIG. 4Consolidation curve for load increment from 100 kPa to 200 kPa, for Specimen 4.

FIG. 7Coefcient of consolidation cv for Specimen 3 evaluated by different methods.

Copyright by ASTM Int'l (all rights reserved); Mon Aug 1 20:01:13 EDT 2011 Downloaded/printed by Universidade Federal do Rio Grande do Sul pursuant to License Agreement. No further reproductions authorized.

BARROS AND PINTO ON OEDOMETER CONSOLIDATION TEST ANALYSIS

FIG. 8Coefcient of consolidation cv for Specimen 4 evaluated by different methods.

noted, however, that the obtained c values, specially those values for Sample 3, which has the lowest coefcient of consolidation, are not completely trustful, since they are based on a small number of readings in the secondary compression range of the consolidation curve. Long-term consolidation tests would be necessary to get higher condence on the c values. Anyway, the obtained c values for all soils are in the range reported in the literature for those types of soil (Mesri and Godlewski 1977). As stated in the previous section, the model beginning of secondary compression should be arbitrarily set in the proposed method. For the numerical results shown hereto, the model beginning of secondary compression is set to 95 % of the primary consolidation TBoS = 1.129 . This setting has some effect on the obtained values of cv, depending on the amount of secondary compression in the soil behavior. Figure 10 shows the variation of the cv values obtained for the load increment from v = 100 kPa to 200 kPa with the setting of the beginning of secondary compression, for all samples. The increment of the calculated cv values with the reduction of t0, shown in Fig. 10, is much larger for the organic clay than for the other three samples, due to the larger amount of secondary compression in the organic clay settlement. But the earlier values of t0 used to get the results in Fig. 10 are hardly justied, without addi-

FIG. 10Variation of cv with the model beginning of secondary compression as a percentage of the primary consolidation (load increment from v = 100 kPa to 200 kPa).

tional information on the ongoing secondary compression, even taking the reports of beginning of secondary compression, based on experimental observations, at points prior to 75 % of primary consolidation for peat (Robinson 2003), into account. The linear-log t model used here probably does not apply to those reported, extreme cases, which require a more rened secondary compression model to match the experimental observations. Also, those early beginning of secondary compression times were observed in consolidation tests with load increment ratio much lower than 1. Thus, in order to stay consistent with the standard graphical methods, values between 90 and 95 % are suggested for the model beginning of secondary compression, for all soil types, when LIR= 1 and the linearlog t model is used, unless more information on the secondary compression for that soil type is available. It is interesting to note, also, that the regression curves resulting from different settings of the model beginning of secondary compression are close to each other. Figure 11 shows the regression curves for Specimen 4, obtained by setting t0 at 50 %, 75 %, and at

FIG. 9Variation of c with

for the four clay specimens.

FIG. 11Regression curves obtained with the beginning of secondary compression set at 50 %, 75 %, and 90 % of the primary consolidation (Specimen 4, load increment from v = 100 kPa to 200 kPa).

Copyright by ASTM Int'l (all rights reserved); Mon Aug 1 20:01:13 EDT 2011 Downloaded/printed by Universidade Federal do Rio Grande do Sul pursuant to License Agreement. No further reproductions authorized.

6 GEOTECHNICAL TESTING JOURNAL

TABLE 3Coefcient of consolidation values for original, reduced, and reduced-with-noise datasets (load increment from v = 100 kPa to v = 200 kPa). 103 cv mm2 / s Set with Noise Specimen 1 2 3 4 Original Set 73.9 5.72 1.81 33.2 Reduced Set 75.6 5.80 1.82 31.5 Mean 76.6 5.94 1.79 32.2 Std. Dev. 13.3 17.3 % 0.87 14.6 % 0.20 10.9 % 3.97 12.3 %

95 % of the primary consolidation. The plot in Fig. 11 shows the three regression curves almost coincident. Thus, it is not feasible to use a regression method to estimate an optimal point for the model beginning of secondary compression. It should be noted that the linear-log t secondary compression model introduces a small kink in the regression curve at t0. This kink, which is an artifact of the regression model, is apparent only when the coefcient of secondary compression is high, as for soft organic clays. A consequence of this discontinuity is that in some rare cases the numerical procedure may fail to converge. This may happen when a test reading point is too close to t0 and the quality of the data is poor. As a workaround, one can try small changes in the TBoS, so t0 is moved away from that reading point. As a nal note on this topic, it should pointed out that the secondary compression model can be replaced by another one more appropriate, if it is available, with only minor modications in the proposed method. The test data used in this research were obtained with an equipment setup that may be of higher quality than the equipment available in some laboratories, specially the apparatus setup for settlement measuring, which employs a high precision electronic dial gage coupled to an automatic data logging system. In many cases, mechanical dial gage extensometers with 0.01 mm precision and manual data logging are used in practice. In those cases, a much reduced number of settlement readings are taken, usually at times 8, 15, and 30 s, 1, 2, 4, 8, 15, and 30 min, 1, 2, 4, 8, and 24 h. Moreover, the manual data logging can lead to even less precision in the readings, specially for the initial ones, when the dial pointer is moving faster. In order to test the proposed method behavior in face of those adverse conditions, a number of numerical simulations were made. The simulation process was applied in two steps. First, the extra readings between those taken at times close to those listed above were deleted from the datasets and the kept settlement values were rounded to two decimal gures. New values of cv were then obtained with the regression method from these reduced datasets. Next, in the second step, a Gaussian random noise with a pessimistic standard deviation of 0.02 mm (two divisions on the mechanical dial gage face) was added to that reduced settlement values and the regression was applied to the resulting datasets. This second step was repeated ten times for each reduced dataset. Then, mean and standard deviation values of the cv were calculated. Table 3 shows the results obtained with those simulations. The numerical results in Table 3 show that the regression method works very well with the datasets with reduced number of points and lower precision, since the cv values obtained with both the original and reduced datasets are very close to each other for all four specimens. Also, the mean cv values obtained with the reduced set with added noise are close to the original cv values. The ob-

served relative standard deviation magnitude can be considered low, given the magnitude of the Gaussian noise introduced. As expected, the relative standard deviation magnitude is higher for the stiffer specimens, specially for Specimen 1, because the noise amplitude relative to the total settlement is higher for those specimens. It is concluded, from the simulations made, that the proposed regression method is very stable and gives consistent results. Additional extensions to the regression method proposed herein can be devised. One can assign different weights to the readings, thus favoring some of them over the others. For example, larger weights can be assigned to the initial readings, which are less affected by secondary compression. In this way the resulting cv will be less affected by the secondary compression. This weighted regression method can be implemented by replacing the least square procedure in Eq 8 with a weighted least square procedure:
N

min
0, 100,c ,cv i=1

wi

ti

(9)

where wi are the weights assigned to each reading. Another possible development is to perform a compound simultaneous regression with datasets from more than one load increment at once, sharing the same cv (and possibly the same c ) (Bowen and Jerman 1995). This procedure will result in an average value for cv (and c ) that is appropriate for a load range wider than one single load increment.

Concluding Remarks
A nonlinear regression method based on the least squares procedure for the analysis of the oedometer consolidation test data was presented. The method can deal with test data from different types of clay soils and calculates the values of the coefcient of consolidation cv and of secondary compression c without the need of hand drawing the consolidation curve and of other graphical constructions. The resulting cv and c obtained with this method have statistical signicance since the least squares procedure is used. The regression method can be implemented in spreadsheet computer programs that is commonly found in most soil laboratories. This implementation can be done with macro-programming or with the spreadsheet built in functions. In the last case the goal seek facility found in those computer programs can be used. Numerical values for cv, from consolidation tests executed on different types of clays, show that the values obtained by the proposed method are in good agreement with those obtained by the standard graphical methods. Also, the values of c obtained with the proposed method are in the range of values reported in the literature. Simulations through variation of data quality showed that the numerical regression procedure is very stable and leads to consistent results for most cases. For the rare cases that have convergence problems, a simple workaround can be applied. The main limitation of the proposed method is the time of model beginning of secondary compression which has to be arbitrarily set at some degree of the primary consolidation. This limitation is common to all analysis methods based only on settlement measurements. Possible extensions of the proposed method include the use of different models for the secondary compression, the use of a weighted least squares procedure, and the compound simultaneous

Copyright by ASTM Int'l (all rights reserved); Mon Aug 1 20:01:13 EDT 2011 Downloaded/printed by Universidade Federal do Rio Grande do Sul pursuant to License Agreement. No further reproductions authorized.

BARROS AND PINTO ON OEDOMETER CONSOLIDATION TEST ANALYSIS

regression, in which more than one load increment dataset is processed at once.

N U cv,ti S t0,ti

U cv,ti 2 U cv,ti U cv,ti S t0,ti

S t0,ti U cv,ti S t0,ti S2 t0,ti

0 100

iU

cB

cv,ti S t0,ti i (16)

Acknowledgments
The authors want to thank to the laboratory technicians, L. E. Meyer and J. B. Cipriano, for the graphical calculations, and to the technicians, E. Jurgensen and R. B. L. Silva, for the help with the consolidation tests. The second author was partially supported by the Conselho Nacional de Desenvolvimento Cientco e Tecnolgico CNPq, Brazil. This research has received partial support from Maccaferri do Brasil Ltda.

with implied summation over i. A fourth equation is obtained by taking the derivative of D rela tive to cv and equating it to zero: F= where U cv,t = and log10 max 1, S cv,t = cv ti t0 0 1 cv ln 10 t t t0 t0 (19) U cv,t = 2t e cv m=0
2m + 1 2 2 cvt 4

D =2 cv i=1

100U

cv,ti + cBS cv,ti Ri = 0

(17)

Appendix: Algorithm Derivation for the Regression Method

The model assumed for the soil sample settlement in the oedometer, under a stress increment, as a function of the time is: =
0+

(18)

100U cv,t + cBS t0,t

100

(10) is the total

where 0 is the initial instantaneous displacement, primary consolidation displacement, U cv,t = 1 8

2 m=0

1 e 2m + 1 2

2m + 1 2 2 cvt 4

(11)

is the percentage of primary consolidation from Terzaghis theory, cv = cv / H2 is the modied coefcient of consolidation, Hd is the d height of drainage, cB = cBH = c H / 1 + e0 is the modied coefcient of secondary compression, H is the sample height, e0 is the initial void ratio, and S t0,t = log10 max 1, t t0 (12)

The solution of the set of four equations is obtained by an iterative procedure based on a simple solution for the 3 3 linear system and on the Newton-Raphson root-nding method for the last equation. v First, an initial guess c0 is estimated. The initial t0 is then calcu lated from Eq 13 and 0, 100, and cB are calculated from Eq 16. Next, a better approximation for cv can be obtained from: cvk+1 = cvk where F =2 F = cv i=1 + cBS cv,ti and
2 n n 100U 2

F F

(20)
cvk

where t0 is the time of the model beginning of secondary compression. The parameters for the model above, to be determined from the measured settlement i at time ti during the consolidation test, are 0, 100, cv, and cB. The time t0 is supposed to be related to cv by: TBoS t0 = cv (13)

cv,ti + cBS cv,ti Ri + 2

i=1

100U

cv,ti (21)

U cv,t =

where TBoS is the time factor which corresponds to the model beginning of secondary compression, which is supposed to be known. The regression algorithm is based on a least-squares t procedure. The sum of the squared differences between the model displacements at ti and the measured displacement di is:
N

U cv,t 2m + 1 2e = t2 2 m=0 2 cv log10 max 1, v c2 ti t0 0 1 v c2

2m + 1 2 2 cvt 4

(22)

t t

t0 t0 (23) is the

S cv,t =

= cvk+1

ln 10

D=
i=1

R2 i

(14)

where Ri =
0

The procedure is repeated until required precision. The initial guess cv0 can be taken as: cv0 =

cvk

, where

100U

cv,ti + cBS t0,ti

(15)

is the residue at the ith point. Taking the derivatives of D relative to 0, 100, and cB, and equating each one to zero in order to get a minimum, a linear system of equations is obtained. This system, in matricial form, is:

T50 0.192 = 50 50 t t

(24)

where 50 is the estimated time for 50 % of the consolidation, which t can be taken as the time that corresponds to half of the total settlement, excluding the initial settlement, due the load increment, 50:

Copyright by ASTM Int'l (all rights reserved); Mon Aug 1 20:01:13 EDT 2011 Downloaded/printed by Universidade Federal do Rio Grande do Sul pursuant to License Agreement. No further reproductions authorized.

8 GEOTECHNICAL TESTING JOURNAL

= 50
1

+ 2

(25)

t After calculating 50, 50 may be estimated by linear interpolation from the two available readings closest to 50.

References
Bowen, W. P. and Jerman, J. C., 1995, Nonlinear Regression Using Spreadsheets, Trends Pharmacol. Sci., Elsevier, Vol. 16, No. 12, pp. 413417. Buisman, A. S. K., 1936, Results of Long Duration Tests, Proceedings of the First International Conference on Soil Mechanics and Foundation Engineering, Cambridge, Massachusetts, Vol. 1, pp. 103105. Casagrande, A. and Fadum, R. E., 1940, Notes on Soil Testing for Engineering Purposes, Harvard University Graduate School of Engineering Publication, n. 8. Chan, A. H. C., 2003, Determination of the Coefcient of Consolidation Using a Least Squares Method, Geotechnique, Vol. 53, No. 7, pp. 673678. Day, R. A. and Morris, P. H., 2006, Discussion on Determination of the Coefcient of Consolidation Using a Least Squares Method, by Chan, A. H. C., Geotechnique, Vol. 56, No. 1, pp. 7376.

Gibson, R. E. and Lo, K. Y., 1961, A Theory of Consolidation for Soils Exhibiting Secondary Compression, Norwegian Geotechnical Institute Publication 41, pp. 116. Mesri, G. and Godlewski, P. M., 1977, Time- and StressCompressibility Interrelationship, J. Geotech. Engrg. Div., ASCE, Vol. 103, No. GT5, pp. 417430. Robinson, R. G., 2003, A Study on the Beginning of Secondary Compression of Soils, J. Test. Eval., ASTM, Vol. 31, No. 5, pp. 110. Robinson, R. G. and Allam, M. M., 1996, Determination of the Coefcient of Consolidation from Early Stage of Log t Plot, Geotech. Test. J., ASTM, Vol. 19, No. 3, pp. 316320. Robinson, R. G. and Allam, M. M., 1998, Analysis of Consolidation Data by a Non-Graphical Matching Method, Geotech. Test. J., ASTM, Vol. 21, No. 2, pp. 140143. Sridharan, A. and Prakash, K., 1985, Improved Rectangular Hyperbola Method for Determination of Coefcient of Consolidation, Geotech. Test. J., ASTM, Vol. 8, No. 1, pp. 3740. Taylor, D. W., 1948, Fundamentals of Soil Mechanics, John Wiley and Sons, New York. Terzaghi, K., 1943, Theoretical Soil Mechanics, John Wiley and Sons, New York. Wahls, H. E., 1962, Analysis of Primary and Secondary Consolidation, J. Soil Mech. and Found. Div., ASCE, Vol. 88, No. SM6, pp. 207231.

Copyright by ASTM Int'l (all rights reserved); Mon Aug 1 20:01:13 EDT 2011 Downloaded/printed by Universidade Federal do Rio Grande do Sul pursuant to License Agreement. No further reproductions authorized.