An Evaluation on Discharge Capacity of Perfabricated Vertical Drains

Using Large Scale Test Apparatus

Y.M. Park
School of Civil, Urban and Environmental Engineering, Yeungnam University, Gyongsan, Korea
(Rep.)
ympark@yumail.ac.kr
S.S. Kim
Department of Civil and Environmental Engineering, Hanyang University, Ansan, Korea(Rep.)
Kimss@hanyang.ac.kr
H-Y. Jeon
Faculty of Applied Chemical Engineering, Chonnam University, Gwangju, Korea (Rep.)
hyjeon@chonnam.ac.kr

ABSTRACT: Discharge capacity is the most important factor to evaluate the well resistance of prefabricated
vertical (PV) drains. In this study, large-scale test apparatus were devised to investigate the accurate discharge
capacity of plastic board drain (PBD) and fiber drain (FD). A series of laboratory flow test of PBD and FD
has been carried out to evaluate those discharge capacities with same condition in the field. Drain sample
length tested ranging from 0.7m to 30m by confining the drain in clay. It is found that discharge capacity qw of
PV drains decreases with increasing the confining pressure, the hydraulic gradient, and drain length. It is
strongly recommended that for determining the design value of qw and kw the discharge capacity test should be
performed using a half sample length of real field depth. It is found out that well resistance calculated from
Onoue’s equation is underestimated comparing to that of Hansbo’s one.

1 INTODUCTION

One of the largest changes in the last ten years of the use of prefabricated vertical (PV) drains and has been
the change in terminology and common acceptance of their use as a soft ground improvement tool throughout
the world. PV drains, in conjunction with embankment preloading and/or vacuum systems, have been widely
used to accelerate primary consolidation and to eliminate most of the anticipated secondary settlements of the
soft ground.
There are two kinds of PV drains; one is plastic board drains (PBD), the other is natural fiber drains (FD).
PBD is band shaped products consisting of a geosynthetic filter jacket surrounding a plastic core. The size of
a PBD is typically 10 ㎝(4in) wide by 3 ㎜(1/8 in) in thickness.
Lee et al. developed FD (1995), it has been used in several countries of Southeast Asia. FD is also band
shaped products consisting of 4coir strands of coconut fiber core enveloped by two layers of jute burlap filter.
The size of FD is 10 ㎝(4in) wide by 9 ㎜(3/8 in) in thickness.
Hansbo (1979) modified the equations developed by Barron (1948) for PV drain applications. The modified
general expression for average degree of consolidation is given as:

− 8Th
U h = 1 − exp[ ] (1)
F
and F = F (n) + Fs + Fr (2)

330
 Where, F is the factor which expresses the additive effect due to the spacing of the drains, F(n); smear
effect, Fs; and well resistance, Fr.
Since the prefabricated vertical drains have limited discharge capacities, Hansbo (1979) developed a well
resistance factor, Fr, assuming that Darcy’s law can be applied for flow along the vertical axis of the drain,
The well resistance factor is given as:

kh
Fr = πz ( L − z ) (3)
qw

Where z is the depth of the drain under consideration; L is the length of the drain having one-way drainage
or half this value for two-way drainage; kh is the coefficient of permeability in the horizontal direction in the
undisturbed soil; and qw is the discharge capacity of the drain.
The discharge capacity defined as qw=AwKw is a parameter of the longitudinal permeability which can be
determined by: qw=Q/i. Where Aw is hydraulic cross section of the drain; kw is longitudinal permeability of the
drain; Q is water flow through the drain; i is hydraulic gradient (∆h/L).
Ordinary case, the application of PV drains is ranged from 20m to 40m. Discharge capacity qw is one of the
most important factors to evaluate the drainage performance of PV drains.
Discharge capacity of PV drains considerably depends on its length. But, most of the laboratory tests have
been performed using sample length of 10 ㎝ to 100 ㎝. They may have some problem to evaluate accurate
discharge capacity. It is reasonable to perform the laboratory test using long sample, which is adjusted same
length in the field. Oostveen and Troost (1990) reported the discharge capacity of the drain, depending on the
drain condition, the rate of water flow, and the long discharge behavior. This paper describes variation of
discharge capacity due to sample length of PV drains using middle and large scales test apparatus. And this
paper is also discussed the rational sample length of PV drain in laboratory test to evaluate the discharge
capacity in the field.

2 TEST APPARATUS

In this study, middle scale test apparatus (main cylinder diameter: 30 ㎝, height: 100 ㎝) and large scale test
apparatus (diameter: 120 ㎝, height: 200 ㎝) as shown in Figure 1, 2 was used for evaluation of discharge
capacity due to change of drain length. Test apparatus consists of large-scale steel cylinder which is similar
type of triaxial device, miscellaneous equipment of outlet & inlet water tank, compressor etc.

3 TESTING METHOD

3.1 Test Procedure

The drain sample is set inside the cylinder similar to that of setting the triaxial test sample. The top and
bottom pedestals are fixed at the top and bottom of the cylinder and connected to outlet and inlet water flow
systems respectively.
Hansbo(1983) recognized the importance of confining the drain by clay on determining the discharge
capacity. Miura et al. (1997) recommended have performed laboratory test that and recommended for
determining the design value, the discharge capacity test should be conducted by confining the drain in clay.
So, all of drain sample tested were inserted inside flexible vinyl hose with diameter of 10cm by 0.8mm in
thickness.
The remolded clay with water content of 150% for easy filling in the hose is put into vinyl hose layer by
layer keeping the drain in center.

331
 Figure 1. Middle scale test apparatus Figure 2. Large scale test apparatus

One end of vinyl hose with the drain sample is fixed and sealed at the bottom pedestal, after the drain
sample is arranged in the cylinder; the other end of vinyl hose is fixed and sealed at the top pedestal. The
arrangement of the drain set spiral type in case of the drain length is longer than cylinder height to investigate
the discharge capacity of full size length in the field.
Constant head discharge capacity test has been performed. The position of inlet water tank was changed to
control the constant hydraulic gradient. But, it was very difficult to adjust the difference of water head in case
that sample length is longer than 2m. In such cases, motor pump and regulator system was used. For example,
drain sample length is 20m. The inlet water pressure, controlled by regulator, was adjusted 1kg/cm2 (100kpa)
to control the hydraulic gradient of 0.5. The water to measure the discharge capacity was mixed with seawater
and distilled to keep the same degree of salinity of the field located in the coastal area.
After settling of drain, steel cylinder is filled with water for applying the confined pressure. Measurement
of discharge capacity under desired hydraulic gradient is performed after 1 day for the finish of consolidation
of clay.
Figures 3, 4 show test results about variation of qw with elapsed time adopted two kinds of water supplying
system for the control of desired hydraulic gradient by keeping of constant head water level and controlling of
regulator and pump system. The difference of discharge capacity of both systems is less than 3% after 24
hours as shown in Figures 3 and 4. That means water-supplying system used in this study has no problem to
measure the discharge capacity.

6
6 Pump(FD3)
Pump(FD3)
Water level(FD3)
Water level(FD3)
Discharge Capacity, Qw(cm3/sec)
Discharge Capacity, Qw(cm3/sec)

Pump(FD1-D) Pump(FD1-D)
5
5 Water level (FD1-D)
Water level (FD1-D)

4
4

3
3 2
Confining Pressure(σ3) : 1kgf/cm 2
Confining Pressure(σ3) : 1kgf/cm
Hydraulic Gradient(i) : 0.4
2 2 Hydraulic Gradient(i) : 0.4
Sample Length : 200cm
Sample Length : 200cm

1 1

0 0
0 5 10 15 20 25 30 0 5 10 15 20 25 30

Elapsed time(hour) Elapsed time(hour)

Figure 3. Variation of qw with elapsed time Figure 4. Variation of qw with elapsed time
(σ3 = 1kgf/㎠) (σ3 = 4kgf/㎠)

332
4 MATERIALS USED

The PV drains adopted in this study are two kinds (PBD and FD). Figure 5 shows the picture of each
sample tested. The clay was sampled from site of Busan New Port. Its index properties are: specific gravity, ρs
of 2.68, liquid limit, wL of 27%, passing ratio of 200# sieving of 9.83% and classified CH by USCS.

(a) Shape of PBD(MD88) type (b) Shape of FD1-D type

(c) Shape of FD1-F type (d) Shape of FD3-C type

Figure 5. Shape of PBD and FD tested

5 TEST RESULTS AND DISCUSSIONS

5.1 Effect of confining pressure

The effect of magnitude of confining pressure on the discharge capacity of the drain (qw) is depicted in
Figures 6, 7. The results show that qw decreases with increasing confining pressure. From a practical
viewpoint, it is considered reasonable to assume a linear relationship between qw and σ3 over the range of
from the 1 kgf/cm2 (100kpa) to the 4 kgf/cm2 (400kpa). Comparing to Figures 6 and 7, qw of the longer drain
sample much more decrease due to increased of confining pressure than that of the shorter one. It may predict
that the long sample occurred the relatively large creep deformation of drain due to increase of confining
pressure.

FD(1)-D
1
FD(1)-3
FD(1)
Discharge Capacity, Q w(cm /sec)

Discharge Capacity, Q w(cm /sec)

Hydraulic Gradient(i) : 0.4 PBD(MD88) Hydraulic Gradient(i) : 0.4 FD(3)

FD(1)
100 FD(1)-1
Sample Length : 70cm Sample Length : 20m
3

FD(3)
3

FD(3)-1

10
0.1

0.1
0.01
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
2 2
Confining Pressure, σ3 (kgf/cm ) Confining Pressure, σ3(kgf/cm )

Figure 6. Variation of qw with confining Pressure Figure 7. Variation of qw with confining Pressure
(Drain length of 0.7m) (Drain length of 20m)

333
5.2 Effect of hydraulic gradient

Pradhan et al. (1991) reported that although the hydraulic gradient is the same, the resulting qw varies
depending on the hydraulic head difference. Hence, it is important to execute the laboratory tests adopted the
flow gradient suitable for the site condition.
To calculate the hydraulic gradient that is applicable at the site, the equation can be used:

∆H
i= (4)
(L / 2)

Where, ∆H is head of water (= ∆P / γ w ) .

Nakanodo et al. (1991), by varying the head distribution of the drain at the site, found out that the hydraulic
gradient ranges from 0.03 to 0.8. Figures 8, 9 show the variation of qw with the hydraulic gradient. qw
decreases with increasing hydraulic gradient. In the laboratory, the flow of water inside the drain may change
from laminar flow to turbulent flow. Hence, Darcy’s law may not be applied.

10
FD(1)
Discharge Capacity, Q w(cm3/sec)

60 2
Confining Pressure(σ ) : 1kgf/cm2
Discharge Capacity, Q w(cm3/sec)

Confining Pressure(σ3) : 1kgf/cm FD(3)

3
Sample Length : 70cm Sample Length : 20m FD(1)-1
50 FD(1)-D FD(3)-1
FD(1)-3
PBD(MD88) 1
40
FD(1)-S

30

20 0.1

10

0 0.01
0.0 0.2 0.4 0.6 0.8 1.0 1.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2

Hydraulic Gradient, i Hydraulic Gradient, i

Figure 8. Variation of qw with the hydraulic gradient Figure 9. Variation of qw with the hydraulic gradient
(L = 0.7m) (L = 20m)

Moreover, qw becomes almost independent of the hydraulic gradient. It is reasonable to assume that the flow
of water from the subsoil with low permeability into the drain in laminar flows. It is therefore recommended
to adopt hydraulic gradient ranges from 0.2 to 0.5 in the laboratory test. We can evidently consider from
typical field example as follow: In case of drain lengths are 40m and 20m, embanking height of 3m-unit
weight of embankment material of 1.8tf/m3. i=∆H / (L/2)=5.4/20=0.29. If the length of drain in 20m, we can
get hydraulic gradient of 0.54.

5.3 Effect of drain length.

Figures 10, 11 show the variation of qw with the drain length. qw decreases with increasing drain length. It may
come out the increase of well resistance with increasing drain length.

334
 FD1-B
FD1-E FD1-B
FD1-D FD1-E
FD3-A 100 FD1-D
100 FD3-B
Confining Pressure : 4kgf/cm2 FD3-A

Discharge Capacity, Qw(cm3/sec)

Confining Pressure : 1.0kgf/cm2 FD3-C
Discharge Capacity, Q w(cm3/sec)

FD3-B
FD3-D Hydraulic Gradient(i) : 0.4
Hydraulic Gradient(i) : 0.4 FD3-C
PBD(MD88) FD3-D
10
10 FD1-S(S) PBD(MD88)
FD1-S(T)
FD1-S(S)
FD1-S(T)

1
1

0.1 0.1

0.01 0.01
0 0.7 2 4 10 20 30 0 0.72 4 10 20 30

Length of FD, L(m) Length of FD, L(m)

Figure 10. Variation of qw with the drain length Figure 11. Variation of qw with the drain length
(σ3=1.0kgf/㎠) (σ3=1.0kgf/㎠)

6 DISCUSSION

Typical equations for calculation of well resistance are two proposed by Hansbo(1983) and Onoue(1988) as
follows:

kh
Fr = πz ( L − z ) (5)
qw
2
32 k ⎛ L ⎞
LCWR = 2 h ⎜⎜ ⎟⎟ (6)
π kw ⎝ dw ⎠

Where, LCWR is the coefficient of drain well resistance; kw(qw/Aw) and kh are the permeability coefficient of
drain and undisturbed soils respectively; dw is diameter of drain.
The Standard Drain Discharge Test is not defined the drain sample length. From this investigation, it shows
that the drain sample length is important factor affecting the discharge capacity. Table 1 shows the variation
of qw and kw with the drain sample length. Figures 12~15 show calculated well resistance; Fr and LCWR of
PBD and FD adopted typical clay (k h = 1 × 10 −7 cm / ses) and silty (k h = 1× 10 −5 cm / ses) ground. It seems that
the well resistance can vary with increasing of drain length.

Table 1. The variation of qw and kw with drain sample length

Sample PBD FD
length (m) qw (cm3/s) kw(cm/s) qw (cm3/s) kw(cm/s)
0.4 96 4.889 - -
0.7 18 0.916 3.4 0.245
2.0 8.5 0.432 1.8 0.129
4.0 7.4 0.376 1.6 0.115
10.0 5.5 0.280 1.4 0.101
20.0 5.0 0.254 1.2 0.086
30.0 - - 1.0 0.072

335
 P B D (σ 3=4kgf/cm3) F D (σ 3=4kgf/cm3)
100 1000

kh =1×10-5 cm/ s kh =1×10-5 cm/ s

10 100

1 10

0.1
1

Fr
Fr

0.01
0.1
0.001 Sample length k h =1×10-7 cm/ s
Sample length
kh =1×10-7 cm/ s 0.4m(MW) 0.7m 0.7m 2m
0.01
0.0001 2m 4m 4m 10m
10m 20m 20m 30m

0.00001
0.4m(MW) 0.7m 0.001 0.7m 2m
2m 4m 4m 10m
10m 20m 20m 30m
0.000001 0.0001
0 5 10 15 20 25 0 5 10 15 20 25 30 35
Length of PBD (m) Length of FD (m)

Figure 12. Variation of Fr with drain sample length Figure 13. Variation of Fr with drain sample length

P B D (σ 3=4kgf/cm )
3 F D (σ 3=4kgf/cm3)
100 1000

k h =1×10-5 cm/ s
10 100 -5
kh =1×10 cm/ s

1 10

0.1
1
L C WR
L CW R

0.01
0.1
0.001 Sample length Sample length
k h =1×10-7 cm/ s 0.4m(MW) 0.7m 0.7m 2m
0.01 k h =1×10-7 cm/ s
0.0001 2m 4m 4m 10m
10m 20m 20m 30m

0.00001
0.4m(MW) 0.7m 0.001 0.7m 2m
2m 4m 4m 10m
10m 20m 20m 30m
0.000001 0.0001
0 5 10 15 20 25 0 5 10 15 20 25 30 35
Length of PBD (m) Length of FD (m)

Figure 14. Variation of LCWR with drain sample length Figure 15. Variation of LCWR with drain sample length

In basis of a series of laboratory test, test result using short sample at the discharge capacity test may lead to
underestimate the well resistance. It is recommended that drain sample length should be adjust a half of
installed one is the site. It may consider that the accurate well resistance can calculate using long sample with
a half of real length in the field. Onoue (1988) modified Yoshikuni and Nakanodo (1974) equation for
axisymmetric flow the average degree of consolidation, U h on a horizontal place as follow:

⎛ − 8Th ⎞
U h = 1 − exp⎜⎜ ⎟⎟ (7)
⎝ ( F ( n ′) + 0 . 8 LCWR ) ⎠

Onoue (1988) considered 0.8LCWR of well resistance for calculation of average degree of consolidation with
radial flow as shown in equation (7).

100

80

60
Fr = 0.8L CW R

Fr
40
PBD (σ₃=1kg/㎠)

PBD (σ₃=4kg/㎠)
20
FD ( σ₃=1kg/㎠)

FD ( σ₃=4kg/㎠)
0
0 20 40 60 80 100

0.8LCWR

Figure 16. The comparison of Fr by Hansbo and 0.8LCWR by Onoue

336
 Figure 16 shows the comparison of Fr by Hansbo (1981) and 0.8LCWR by Onoue (1988). It seems that well
resistance calculated from Onoue’s equation is underestimated compare to that of Hansbo’s one.

7 CONCLUSION

From the discharge capacity test of PV drains in the laboratory, following conclusions can be drawn:

1) It is confirmed that large-scale test device is available to evaluate the discharge capacity of PV drains with
same condition of in-situ.

2) The discharge capacity of PV drains is largely depending on the drain sample length. It is strongly
recommended that for determining the design value of qw and kw the discharge capacity test should be
performed using a half sample length of real field depth.

3) Well resistance calculated from Onoue’s equation is underestimated compare to that of Hansbo’s one.

8 REFERENCES

Barron, R.A., “Consolidation of Fine-Grained Soils by Drain Wells”, Trans. ASCE, Vol. 113, No. 2346, pp.
718~742(1948).
Hansbo, S., “Consolidation of Clay by Band-shaped Prefabricated Drains" Ground Engineering, Vol. 12, No.
5, pp. 21∼25(1979).
Hansbo, S., “Consolidation of Fine Grained Soils by Prefabricated Drains", Proc. 10th ICSMFE, pp. 677~682
(1981).
Hansbo, S., “How to Evaluate the Properties of Prefabricated Drains”, proc. of 8th ECSMFE, Helsinki, Vol. 2,
pp. 621∼626 (1983).
Lee. S. L, Karunaratne. G. P. and Aziz. M. A., “An Environmentally Friendly Prefabricated Vertical Drain for
Soil Improventment”, Bengt B. Broms Symposium on Geotechnical Engineering. pp 243∼261(1995).
Miura, N., Chai, J.C., and Toyota, K., “Investigation on Some Factors Affecting Discharge Capacity of
Prefabricated Vertical Drain”, Proc. 6th Int. Conf. on Geosynthetics, Atlanta, U.S.A., Vol.2, pp.845~850
(1997).
Nakanodo, Y., Hanai, H., and Imai, W. “Permeability of Plastic Board Drain”, Hukken Technical Report,
Vol.17, pp.139~151 (in Japanese), (1991).
Onoue, A., “Consolidation by Vertical Drains Taking Well Resistance and Smear into Consideration”, Soils
and Foundations, Vol. 28, No. 4, pp. 165~174(1988).
Oostveen, J. P., and Troost., “Discharge Index Tests on Vertical Drains , Proc. 4th Int. Conf. Geotextiles,
Geomembranes and Related Products, The Hugue, Vol. 1. pp. 345∼350 (1990).
Pradhan, T. B. S., Kamon, M. and Suwa, S., “Design Method for The Evaluation of Discharge Capacity of
Prefabricated Band-shaped Drains”, 9th Asin Reg Conf. Bangkok, Vol. 1, pp. 523∼526 (1991).
Yoshikuni, H. & Nakanodo, H., “Consolidation of Fine-grained Soils by Drain Wells with Finite
Permeability”, Jap. Soc. Soil Mech. Fdn Engn 14, No. 2 (1974).

337