An Experimental Study of Band Drain

GeotextilesandGeomembranes13 (1994) 669~577
Elsevier Science Limited

Printed in Ireland. 0266-1144/94/$7.00
ELSEVIER

An Experimental Study of the Performance of Geosynthetic Band Drains

Y. Wasti & T. Hergtil
Department of Civil Engineering, Middle East Technical University, 06531 Ankara, Turkey (Received 10 March 1994; accepted 5 April 1994)

ABSTRACT The performance of two basic types of geosynthetic band-shaped/strip drains - - a composite drain (core surrounded by a geotextile filter) and a monolithic drain (without a geotextile filter) - - has been compared by means of radial consolidation tests. Tests were repeated employing the monolithic drain wrapped in the same filter fabric as the composite drain, a cylindrical porous stone and a sand drain. Two fine-grained soils of varying plasticity and gradation were used, initially at a water content equal to their respective liquid limits. The results were evaluated to compare the rate and amount of consolidation in each case and to assess the possible effect of clogging on the performance of geosynthetic drains. It was observed that the composite drain performed better than the monolithic drain, especially in the case of finer soil. Wrapping the monolithic drain with the geotextile filter significantly increased its performance.

INTRODUCTION Preloading used together with vertical drains is an effective ground improvement technique, accelerating settlement and gain in strength of soft cohesive soils. Geosynthetic strip or band-shaped vertical drains have now largely replaced sand drains due to ease and speed of installation, less soil disturbance, low cost, etc. Most synthetic drains have a composite
669

670

Y. Wasti, T. Hergfil

construction: a corrugated or studded inner core surrounded by a geotextile filter sleeve or jacket. The type which may be called 'monolithic' has a plastic fluted core without a filter sleeve, flow into the core taking place through the perforations on the outer surfaces of the vertical channels or tubes of the drain. The monolithic drain is claimed to be less expensive, lighter in weight and more constant in quality; being in one piece, it cannot be broken into any components and is considered to work even when folded at right angles (Anon., 1983). There are several records of field and laboratory experiments to assess the relative performance of sand drains and geosynthetic strip drains (Hansbo & Torstensson, 1977; Davies & Humpheson, 1981; Eriksson & Ekstr6m, 1983; Cheikhismailzada, 1991; Bergado et al., 1993) or various types of strip drains, filter sleeves and cores (den Hoedt, 1981; Guido & Ludewig, 1986; Faisal, 1991a). However there are only a few studies which include both composite and monolithic drains (Lawrance & Koerner, 1988; Faisal, 1991b). An investigation of these studies indicates poorer performance of monolithic drains, and it was suggested that the geotextile filter sleeve is essential to prevent the clogging of drainage paths (Faisal & Yong, 1988). This paper reports the findings of an experimental study carried out specifically to compare the efficiency of composite and monolithic geosynthetic drains by means of radial consolidation tests.

EXPERIMENTAL PROCEDURE The radial, equal-strain condition consolidation test apparatus consists of a cylindrical mould with dimensions to accommodate a sample diameter of 100 m m and height of 150 mm, top caps, a base plate, a loading frame and attachments for measuring the amount of water drained from the central drain. The load was applied by means of dead weights to exert selected pressure increments of 50, 100, 150 and 200 kPa. A schematic of the test apparatus is given in Fig. 1. Strip samples, 13 mm wide, were cut lengthwise from Amerdrain 407 composite and Desol monolithic drains for use as a central drain in the test. Tests were repeated employing the monolithic drain wrapped in the same nonwoven geotextile flter as the composite drain. A cylindrical porous stone and a sand drain (of circumference equal to the perimeter length of the strip drains in accordance with the commonly adopted equivalent diameter concept) were also tested in order to contrast the efficiency of various types of drains. A test was also performed using a sand drain with the same size and shape as the strip drains and results very close to those of the cylindrical sand drain were obtained.

Performance of geosynthetic band drains

671

,7
CLIP

\

TOP CAP -11
!

RUBBER

MEMBRANE

..uu

!

~
~

In
|

O-RING LOADING FRAME

STANI CENTRAL DRAIN
• ~

SOIL SAMPLE ,,~--MOULD

i

I

/ ~
SEALANT~

/ O-R ~ . ~

!

/ -

BASE PLATE STE EL

/

\

.jt.I
ADAPTER
o

RUBBER TUBE

I-1

d

uI
,5

I////,

~///A~...~

DEAD WEIGHTS

r////////A
k\X\\\\\\

K\\\"~

k\ \\",r" ~ld ",~\ \

rJ//,/////A
\ \ " \ \ ~r"----- Ol SC

Fig. 1. Test a p p a r a t u s .

Table 1 P r o p e r t i e s o f the Soils U s e d

G~
Soil 1 Soil 2 2.738 2.735

Dso (mm)
0.007 0.03

% Clay size
28 17

Liquid limit (%)
46 32

Plastic limit (%)
17 19

Plasticity index (%)
29 13

672

Y. Wasti, T. Hergfil

Two fine-grained soils of different plasticity and gradation were used in the experiments. The properties of the soils are given in Table 1. Suitability of the geotextile filter sleeve of the composite drain for the soils was checked, referring to the filter criteria adopted in American (Anon., 1990) and German practice (John, 1987), and found to be appropriate (Hergiil, 1994). The first step in the testing procedure was to fix the central drain on the base plate (Fig. 1). Then the inside of the mould was greased to reduce the side friction and mounted on the base plate. Dry soil was mixed with water into a homogeneous mass at a water content equal to its liquid limit and then placed in the mould. Care was taken not to entrap air and provisions were made to maintain the vertical position of the drain during the sample preparation and testing. Sand drains were formed inside a plastic netting and filter paper was put both around the sand drain and porous stone. Applied pressures were kept on the sample until measurable water release had ceased, which was 1 to 7 days depending on the type of soil and central drain. It was observed that at the start of the tests under the first loading increment of 50 kN/m 2 the discharged water was turbid for the first 10-15 minutes in tests with geosynthetic strip drains. This indicated a temporary piping situation which is also reported by den Hoedt (1981) and attributed to the relatively large discharge flow at initial stages that forced fines to enter the core. Both the core of the composite drain and vertical channels of the monolithic drain were checked at the end of the tests and no clogging or blocking was observed. Duplicate tests verified repeatability. The details of the experimental work are presented elsewhere (Hergiil, 1994).

E X P E R I M E N T A L RESULTS The results of consolidation tests performed are presented by plotting the amount of water drained through the central drain (volume change of the sample) against the logarithm of time. As an example, plots for the first and the last applied pressures in the case of Soil 1 are given for sand drain (SD), geosynthetic composite drain (CD), monolithic drain (MD) and monolithic drain wrapped with geotextile (GMD) (Figs 2(a) and (b)). On these plots 'theoretical curves' are also drawn by assuming the coefficient of radial consolidation Ch equal to the coefficient of vertical consolidation Cv (as often resorted to in the absence of radial consolidation test results) to illustrate the deviation due to this assumption. An examination of Fig. 2 and plots for other pressures and also for Soil 2, reveals that the greater proportion of the volume change takes place under the first pressure increment of 50 kN/m 2. For both soils, the sand drain

Performance of geosynthetic band drains
Time (rain)
a 1 i llllll

673

Jo

I

1

1111111

,oo

I

I

I II

,,o, po

I

I

I I II

,opo

3o. i¢ E
(3

6O
O8 f~ 0

~.
E

15o(o) Vertical Pressure=50 kPo o,
I I I Illlll

I0

/

I

i lllllJ

Time ~rnin) I00
i I

I000
I I

I I I IIAI

I T M

I0000
i I

E

so
"~ 15~>

\

CD
(b) Vertical Pressure = 2 0 0 kPa

\\

20

Fig. 2. Volumechange versus log time (Soil I): (a) vertical pressure = 50 kPa; (b) vertical pressure = 200 kPa. discharges the largest amount of water at the fastest rate. The monolithic drain diverges notably indicating the worst performance; that is, the least and the slowest compression. The composite and geotextile-wrapped monolithic drains perform equally well and are comparable to the sand drain. At larger applied pressures the monolithic drain continues to exhibit the worst performance, with the per cent compression under a given pressure becoming increasingly smaller, while the other drains perform equally efficiently. The poorer performance of the monolithic drain is attributed to the reduction in the amount of water drained due to the increased clogging of the holes of the drain as consolidation progresses under increased applied pressures. The above-mentioned observations are quantified in Fig. 3, where per cent volume change (defined as the volume change at the end of each

674
I00

Y. Wasti, T. Hergfil

80-

c o

o E
0

40-

", o - - o Vol. Ch. (SD)/Finol VoL Ch. (SD} A - - A VoI. Ch.(GMD)/Final Vol. Ch. (SD) O- -O VoLCh.(CD)/Finol Vol. Ch. (SD) x...--.x VoLCh.(MO)/Final Vol. Ch.(SD)

>

20

O'

0

so
Applied

a6o
Verticol Pressure ( k N / m 2 )

25O

Fig. 3. Volume change for different drains (Soil 1).

pressure increment for any drain divided by the final volume change for the sand drain test at the end of the last pressure increment of 200 kPa) is plotted against applied pressure increments. This shows that the sample with the monolithic drain compresses 17% less than the sand drain sample for Soil 1 which is more plastic and has a higher per cent clay size: The corresponding value for Soil 2 is 14%. The performance of composite and monolithic drains as consolidation progresses is illustrated in Figs 4(a)-(d). The ordinate represents the ratio of the time intervals (At) required by other drains to those required by the sand drain for successive chosen increments of compression: 1% (i.e. for a compression change of 0-1%, 1-2% and so on) for the case of 50 kPa loading, 0-25% for 100 kPa and 0.2% for subsequent loadings. The abscissa represents individual compression increments. As seen from the figures, this incremental time ratio remains around 1 for the composite and the geotextile-wrapped monolithic drain throughout consolidation for both soils indicating as efficient drainage as the sand drain and a steady filtration situation. On the other hand, for the monolithic drain and especially for the finer soil (Soil 1) this ratio is very large, asymptotically approaching infinity indicating clogging at the final stages of consolidation at each pressure increment. The curves for the monolithic drain are also erratic; occasionally the rate of flow is higher when the load is first applied and hydraulic gradients higher, followed by intermittently increased clogging. The rate of consolidation has also been compared by determining t90 values for each loading using the curve-fitting method due to Wharton (1966; after Berry & Wilkinson, 1969). The ratio of t90 of the geotextilewrapped monolithic drain and composite drain to that of sand drain

Performance of geosynthetic band drains
30
i

675

25O3
_-" 20-

1
MD (SoiJ i )

I_ Y _
x CD(Soil I) o CD(Soil 2)

15-

o
,o-

o

MD (Soil 2) .... /'-"

~
,,7

~ G M O ( S o i l I) O GMD(Soil2)

G

o',

I z-'3
J-z

I +'-5 I
3-4 ~

6'-7 I s'-s 1 ~'_,
7-s 9-=o Increment(%)

.~2

I

Compression

(o)Verficol Pressure=5OkPo
3. = 30-

Z

3 oo ~'zs.
~zo.
6 15 ¸

MD (Soil I)

/

~

l~o(3
~ ~
0.~5

~

,.

m

m.,
LIS-15 Incremnt

=°

~,

o-b.z5

1 a~.arnt

,-,~zs

(~ (b) Vertical Pressure = IO0 kPo

0.'~- I Compreslion

I

ts'~n I

t'/~- 2

z3.zn

Fig. 4. Comparison of rate of consolidation for different drains: (a) vertical pressure = 50 kPa; (b) vertical pressure = 100 kPa;

varies between 1 and 2 for Soil 1 and between about 1 and 1.5 for Soil 2; the larger values are associated with smaller applied pressures. On the other hand, the ratio for the monolithic drain to that of the sand drain is about 8 to 13.5 for Soil 1 and from 3 to 6.5 for Soil 2.

CONCLUSIONS On the basis of tl~e results presented in the previous sections, the following conclusions can be drawn:

676
35.

Y. Wasti, T. Hergfil
l

~') 25-

M

.E

lZ Q = o

~s-

MD (Soil 2)

x CD ( S o i l I) o CD ( S o i l 2 ) GMD(SoilI) 0 GMD (Soil 2 )

m

i
O.Z--O,4

B

~
O~=O.e

~

~,
I-I.Z

~
1.4-1.6

Compression Increment (%) (c) V e r t i c o l P r e s s u r e = I S 0 kPo 50
45-

,', 4 0 0')
35-

~
-~
m
i

i

MD (Soil P)

•~

zs20 °

MD (.Soil I )

J" ~

15I0-

0

i~1

BI

0-0.2

I

0,2-0.4

0.4 -0~

i

[]=
]

•

4
I

0.6-G8

0 -I

~

I ; 1.2

Compression increment (°/o) (d) Verticol Pressure = 2 0 0 k P o
Fig.

4.-contd. (c) vertical pressure = 150 kPa; (d) vertical pressure = 200 kPa.

(1) The composite geosynthetic strip drain clearly performed better than the monolithic drain. The difference in performance was accentuated in the case of finer soil. (2) The poorer performance of the monolithic drain is attributed to clogging of the perforations in its tubes rather than reduced core flow capacity. (3) Wrapping the monolithic drain with the geotextile filter increased its performance to the same level as the composite drain or beyond. (4) Consolidation curves drawn by assuming ch = Cv are inaccurate. (5) For further research, tests using larger samples and several different soils may be appropriate, but it is expected that the main findings of the present study will still be valid.

Performance of geosynthetic band drains REFERENCES

677

Anon. (1983). Desol band drain. Canal Publicite, Paris. Anon. (1990). Geotextile Design and Construction Guidelines, Publication No. FHWA-HI-90-001, US Federal Highway Administration, Washington, DC. Bergado, D.T., Alfaro, M.C. & Balasubramaniam, A.S. (1993). Improvement of Soft Bangkok Clay using vertical drains. Geotextiles and Geomembranes, 12, 615 63. Berry, P.L. & Wilkinson, W.B. (1969). The radial consolidation of clay soils. Geotechnique, 19(2), 253-84. Cheikhismailzada, M. (1991). Wick drains. PhD thesis, Middle East Technical University, Ankara. Davies, S.A. & Humpheson, C. (1981). A comparison between the performance of two types of vertical drains beneath a trial embankment in Belfast. Geotechnique, 31(1), 19-31. Den Heodt, G, (1981). Laboratory testing of vertical drains. In Proceedings o.1 Tenth International Conference on Soil Mechanics and Foundation Engineering. Balkema, Rotterdam, pp. 627 31. Eriksson, L. & Ekstr6m, A. (1983). The efficiency of three different types of vertical drains--results from a full scale test. In Proceedings of Eighth European Conference on Soil Mechanics and Foundation Engineering, Vol. 3, Helsinki, pp. 605-10. Faisal, H.A. (1991a). The influence of filter jacket and core geometry on the longitudinal permeability of a prefabricated drain. Soils and Foundations. 31(3), 120-6. Faisal, H.A. (1991b). The flow behaviour of deformed prefabricated drains. Geotextiles and Geomembranes, 10, 235-48. Faisal, H.A. & Yong, K.W. (1988). Performance of prefabricated vertical drains. In Proceedings of First Indian Geotextile Conference on Rein)Cbrced Soil and Geotextiles, pp. E9-13. Guido, V.A. & Ludewig, N.M. (1986). A comparative laboratory evaluation of band-shaped prefabricated drains. In Consolidation of Soils, Testing and Evaluation, ASTM, STP 892, Philadelphia, pp. 642-62. Hansbo, S. & Torstensson, B.A. (1977). Geodrain and other vertical drain behaviour. In Proceedings of Ninth International Conference on Soil Mechanics and Foundation Engineering, Vol. 1, Japan, pp. 533-40. Hergiil, T. (1994). An experimental evaluation of the performance of geosynthetic strip drains. MSc Thesis, Middle East Technical University, Ankara. John, N.W.M. (1987). Geotextiles. Chapman and Hall, New York. Lawrance, C.A. & Koerner, R.M. (1988). Flow Behaviour of Kinked Strip Drains. ASCE, Geotechnical Special Publication No. 18, pp. 22-39.