Chapter 4
Properties of Prefabricated Vertical Drain and Quality Control Tests

4.1 INTRODUCTION
4.2 TERMINOLOGY
4.3 PROPERTIES OF PVD
4.4 FACTORS CONTROLLING PVD SELECTION
4.5 QUALITY CONTROL TESTS
4.6 SPECIFIC ASPECTS
4.7 CONCLUDING REMARKS

4.1 INTRODUCTION

While designing a soil improvement scheme involving prefabricated vertical drain (PVD), the discharge capacity of the drain is normally assumed to be sufficiently large so that the well resistance can be ignored. During construction, it is important to ensure that the drains used meet this design assumption, that is, the drain should provide sufficient discharge capacity so that the effect of well resistance will be insignificant. As the performance of PVD is affected by many factors, some basic understanding of the properties of drain and the factors affecting the effective use of vertical drains will be essential for engineers working on soil improvement projects that use vertical drains. The major properties of vertical drains that need to be specified are:

1. The discharge capacity of the drain, q_w, and its variation with stress applied in the lateral direction, time, and hydraulic gradient.
2. The apparent pore opening (AOP) of the filter and the permeability of the filter.
3. The tensile strength of the entire drain and the filter.

Being the only element that provides the drainage for consolidation of clay, the property of PVD affects considerably the effectiveness of a soil improvement scheme that uses vertical drains.

Nowadays, there are many brands of PVD to choose from. However, not all the products are similar. While selecting the drain, the engineer should consider not only its price but also whether its properties will suit the soil and the need of the project. Normally, the drain manufacturer will provide a specification sheet indicating the values of the main properties, such as the discharge capacity, the tensile strength of the drain, and the pore opening of the filter. However, it is the engineer's responsibility to check whether the real properties of the drain meet the specifications. He should also know that the value for the same parameter could be different depending on the methods of determination. Different manufacturers can use different methods to determine a specific property. Therefore, the specifications given by the manufacturers cannot be compared directly. Engineers should check the specifications by conducting independent quality control tests. Only by adopting the same testing method can the properties of the drain be compared directly.

Furthermore, it is not uncommon for an ordinary soil improvement project to use thousands or even millions of drains that are delivered in several batches over several months or even years. The quality of the drains supplied over a long period may not be consistent, and so the quality in every batch should be checked before the drains are used. The engineer should not assume that the quality is always the same even if the drain comes from the same manufacturer or even the same factory. Therefore, conducting quality control tests on vertical drains is one of the topics discussed in this chapter.

4.2 TERMINOLOGY

Before we discuss the properties of vertical drains, let us make sure by going through each of the terms used, that we have the common language. The terminology for vertical drains is confusing because some of the terms are not defined by geotechnical engineers, but by chemical or material specialists who make vertical drains. We need some cross-disciplinary knowledge in dealing with vertical drains.

4.2.1 Permittivity of Filter

It defines the permeability of the filter in terms of per unit thickness in the direction perpendicular to its surface, as shown in Fig. 4.1, that is,

where Ψ is the permittivity in s⁻¹, k_n is the cross-plane permeability in m/s, and t is the thickness of the filter in m.

A geotechnical engineer will probably just use the cross-plane permeability of the filter instead.

4.2.2 Transmissivity of Filter

This term is used to describe the in-plane flow as shown in Fig. 4.2 and is linked to the in-plane permeability by:

Where Θ is the transmissivity in m²/sec, k_d is the in-plane permeability in m/sec, and t is the thickness in m.

4.2.3 Discharge Capacity of Drain

The discharge capacity of drain is one of the most commonly used parameters for PVD. It measures the rate of flow along the drain (Fig. 4.3) and is defined as:

where q_w is the discharge capacity in m³/s, Q is the discharge velocity in m³/s and i is the hydraulic gradient.

As discharge velocity can be calculated as Q = k_diA, where k_d is the in-plane permeability of the drain and A is the cross-section area of the drain, q_w, can also be calculated as:

4.2.4 Discharge Factor

As the efficiency of vertical drain in discharging water is controlled not only by the discharge capacity but also by the permeability of the soil and the length of the drain, another dimensionless parameter, the so-called discharge factor is defined:

where D is the discharge factor, k_h is the horizontal permeability of the soil, and l_m is the maximum discharge length. From Eq. (4.4), it can be seen that the larger the permeability of soil and the longer the drain, the higher the discharge capacity required. Therefore, D reflects the efficiency of the drain in discharging water during consolidation.

4.2.5 Apparent Opening Size (AOS)

AOS is a measure of the size of the fabric pore opening of filter. Like the size of soil particles, the sizes of the pore opening of filter fall within a certain range. AOS is usually defined as the size that is larger than 95 or 90% of the fabric pores, denoted as O₉₅ or O_90. In Europe and Canada, the term filtration opening size (FOS) may be used as an alternative. The difference between AOS and FOS is in the methods of determination. AOS is measured by a dry-sieving method and FOS by wet and hydrodynamic sieving.

4.2.6 D₈₅ of Soils

D₈₅ is used to measure the particle size of soil. It is defined as the size that is larger than 85% of soil particles as measured from the particle size distribution curve. Sometimes, D₉₀, D₅₀,or D₁₅ may be used.

4.3 PROPERTIES OF PVD

PVD normally consists of a core and a filter sleeve. Both are made of polymers. The dimension of the drain has been standardized to

be a normal 100 mm wide and 3–4 mm thick. The only differences in the different types of drain are the materials used for the core and the filter. Some typical types of drains are shown in Table 4.1. More examples are given in Ch. 5.

The performance of the vertical drain is affected not only by the drain itself but also by the type of soil and the installation method. For the former, the soil interacts with the filter. The properties of the filter control the entry of water into the drains. For the latter, the drain has to have a certain strength that can sustain the tensile stress applied to it during the installation. The main properties of vertical drains that need to be specified are discussed as follows.

4.3.1 Discharge Capacity

The purpose of using PVD is to release the excess pore water pressure in soil and discharge water. Therefore, the higher the discharge capacity, the better the drain. When we assess the value of discharge capacity, the following factors that influence the discharge capacity need to be considered:

1. Consolidation stress. The discharge capacity of drain decreases with increasing consolidation stress. This is mainly due to the reduction in the cross-section of the drain as the drain is compressed under pressure. The variations of discharge capacity with applied vertical pressure for a typical type of PVD under both straight and buckled conditions are shown in Fig. 4.4. It can be seen that the discharge capacity of the drain decreases with the increase in vertical pressure. This is mainly due to the reduction in the cross-section of the drain as a result of the compressibility of the drain and the penetration of the filter into the drain groove, as explained by Broms et al. (1994).
2. Deformation of drain. With the consolidation of soil, the drain will deform or buckle inside the soil. The discharge capacity of the deformed drain should also be measured. The buckled drain will normally have a smaller discharge capacity as compared to the straight drain, as shown in Fig. 4.4.
   
   

   Table 4.1 Common types of PVD.
   Core		Filter	Method of Assembly
   Cross-section	Description	Type	Separated filter and sleeve	Filter joined to core
   	Corrugated groove	Nonwoven fabric	Mebradrain MD 7007
   	Ribbed groove	Synthetic fiber	Mebradrain MD88
   	Monofilament	Needle punched nonwoven fabric		Colbond CX1000
   	Double Cuspated	Nonwoven fabric	Flodrain
   	Studded (one side)	Nonwoven fabric	Alidrain ST
   	Studded (two sides)	Nonwoven fabric	Alidrain DC

   
   
   
3. Time. The discharge capacity of drain may also change with time. This can be partly due to the creep deformation of the drain material, particularly the filter, which causes the effective cross-section of the drain to reduce. An example is given in Chu and Choa (1995).
4. Clogging of drain. When the pores of the filter are too large, the fines may go into the drain and clog the drain. To prevent clogging, the filter must meet a certain requirement. This will be discussed later.
5. Hydraulic gradient. The discharge capacity measured at different hydraulic gradients is different and is smaller when the hydraulic gradient is higher.
6. Temperature. The higher the temperature, the faster the flow and the larger the discharge capacity.

4.3.1.1 Requirement of discharge capacity

When the discharge capacity of the vertical drain is smaller than the amount of water that needs to be discharged, there will be well

resistance. Ideally, the discharge capacity of the drain, q_w, should be sufficiently large so that the well resistance can be ignored in the design. Consolidation theories that consider the well resistance have been formulated by several researchers (Yoshikuni and Nakanado, 1974; Hansbo, 1981; Xie, 1987; Zeng and Xie, 1989). The effect of well resistance can be evaluated using a well resistance factor. This factor is defined differently in different theories. Nevertheless, it can be generally written as ,where α is a coefficient which varies with the theories used. According to Xie (1987) and Wang and Chen (1996), the following condition must be met in order to hold the well resistance at an insignificant level:

This requires the discharge factor, D,to be:

Therefore, the required discharge capacity after applying a deduction factor to consider all the influencing factors on discharge capacity becomes:

where q_req = the required discharge capacity, F_s = deduction factor, F_s =4 ~ 6, k_h= the horizontal permeability of the soil, and l_m = the length of the drain.

The requirement stated in Inequality (4.6) is consistent with the threshold discharge factor of 5 specified by Mesri and Lo (1991). Inequality (4.6) defines the dependence of q_w on k_s and l_m.The relationship among q_w, k_s,and l_m for a F_s = 5 for negligible well resistance has been plotted in Fig. 4.5. The discharge capacity mobilized

as compared to the discharge capacity required for negligible well resistance for four cases are shown in Fig. 4.6 (Mesri and Lo, 1991). On the basis of Fig. 4.6, the minimum discharge capacity should be no less than 100 m³/yr, or 3 × 10⁻⁶ m³/s.

It should be pointed out that the use of an excessively high reduction factor for discharge capacity might not be necessary. This is because, although the discharge capacity reduces with the deformation of vertical drain and time, the permeability of soil reduces with consolidation (as discussed in Ch. 3), so the amount of water to be discharged also reduces with time.

4.3.2 Properties of Filter

Similar to the design for any other filters, two basic filter design criteria have to be met. The first is that AOS has to be sufficiently small so that it can prevent the clay particles from entering the drain. The second is that the permeability (or the permittivity) of

the filter has to be sufficiently large. The criteria adopted for vertical drain are:

4.3.2.1 Soil retention ability

According to Carroll (1983), the conditions for meeting the soil retention ability are:

and

Where O₉₅ is the AOS of filter, O₅₀ is the size which is larger than 50% of the fabric pores, and D₈₅ and D₅₀ refer to the sizes for 85 and 50% of passing of soil particle. O₉₅ =0.75 mm,or 75m,is often specified for prefabricated vertical drains.

The criteria adopted by Bergado et al. (1993) for Bangkok clay are:

and

It should be noted that the requirement for O₅₀ in Inequality (4.9b) is much relaxed than that in Inequality (4.8).

4.3.2.2 Permeability

As is normally required in filter design, the permeability of the filter should be at least one order of magnitude higher than that of soil:

where k_f = permeability of the filter and k_s = permeability of the soil.

As the soil to be treated with PVD has normally very low permeability, this requirement can be met easily in most, if not in all, cases.

4.3.2.3 Clogging resistance

Depending on the particle grading, the soil particle may sometimes be trapped in the filter and clog it. To preventing clogging, Wang and Chen (1996) recommended the following:

where, n is the porosity of filter; O₁₅ is the size which is larger than 15% of the fabric pores and D₁₅ and D₁₀ refer to the sizes for 15 and 10% of passing of soil particle.

In addition, Wang and Chen (1996) suggested that the permeability of filter should be smaller than 100 m/s, but no less than 100 times the permeability of soil, that is,

4.3.3 Tensile Strengths

PVDs should have adequate strength so that it can sustain the tensile load applied to it during installation. Therefore, the strength of the core, the strength of the filter, the strength of the entire drain, and the strength of the spliced drain need to be specified, normally in both wet and dry conditions. According to Kremer et al. (1983), a drain must be able to withstand at least 0.5 kN of tensile force without exceeding 10% in elongation. It is quite common nowadays to specify the tensile strength of the whole drain at either dry or wet condition as larger than 1 kN at a tensile strain of 10%. The spliced drain should also have a strength comparable to that of the drain.

4.4 FACTORS CONTROLLING PVD SELECTION

The quality and suitability of the drains play a key role in the soil improvement scheme involving vertical drains. Different design situations require different types of vertical drains. For example, it will not be necessary to use a vertical drain with a large discharge capacity value if the drain length is small. The drain filter should also match the soil type. The unit price of vertical drain is another important consideration besides meeting the design requirements. A considerable saving can be effected without sacrificing the performance of the drain if the control factors for a vertical drain can be identified and the design requirements are specified accordingly. The factors that control the selection of vertical drain, apart from the cost, are as follows.

1. Compatibility of the filter with the soil to be improved. The pore size or the AOS of the filter should meet the filter design criteria. On the one hand, the filter should have an AOS small enough to prevent the fine particles of the soil from entering the filter and the drain. On the other hand, the AOS cannot be too small because the filter has to provide sufficient permeability. The two key parameters that indicate the quality of the filter are the AOS and the cross-plane permeability of the filter. Some criteria for AOS, denoted by either O₉₅ or O₅₀, are discussed
   
   in Sec. 4.3.2. However, these requirements can be too stringent sometimes. For example, for Singapore marine clay D₈₅ is in the order of 0.01 mm. The O₉₅ of the filter used ranges from 0.04 to 0.06 mm, that is, O₉₅ = (4–6)D₈₅. Even though the O₉₅ is higher than what is required in inequality (4.7), the inner side of the drain is quite clean after even long-term tests in clay in the laboratory (Chu and Choa, 1995). Therefore, the above filter criterion appears to be conservative for Singapore marine clay. The permeability of the filter is normally required to be at least one order of magnitude higher than that of the soil, as discussed earlier. For the land reclamation project at Changi, a permeability of 10⁻⁴ m/s is required for the filter. Thisis 10⁴–10⁵ times larger than the permeability of the clay that is in the order of 10⁻⁹–10⁻¹⁰ m/s. The thickness of the filter is another consideration. Normally, the thicker the filter, the better it becomes, given that all the other conditions are the same. According to Wang and Chen (1996), the mass to area ratio should be generally larger than 90g/m².
2. Discharge capacity, q_w. As the discharge capacity required is controlled by the permeability of the soil and the length of the vertical drain, in theory, the discharge capacity required varies from point to point even within the same soil. Therefore, a higher discharge capacity does not always mean a better design. As a drain with a higher discharge capacity is normally more expensive, one should avoid specifying an unnecessarily high discharge capacity so that the design can be made cost-effective.
   An adequate q_w can be chosen based on Inequality (4.6). l_m has the most significant influence on the discharge capacity. If we take F_s =5, k_s =10⁻¹⁰ m/s, and l_m =25 m, then q_w = 2.45 × 10⁻⁶ m³/s, or 82 m³/yr. If l_m = 50 m instead of 25 m, then q_w =9.81 × 10⁻⁶ m³/s, or 327 m³/yr. Nowadays, most of the drains can provide such a q_w value even in a buckled condition. On the other hand, the permeability of soil can also have a great effect when permeability is not determined accurately. Take the previous case for example: if k_s =10⁻⁹ m/s instead of 10⁻¹⁰ m/s, then q_w =98.1 × 10⁻⁶ m³/s, or 3, 270m³/yr. In this case, some drains will not be able to meet the requirement.
   
3. Tensile strength of drain. During the installation, PVD has to sustain a large tensile stress and undergo some elongation. Therefore, the PVD must have some minimum tensile strength so that it will not be torn apart after installation. A 1 kN (or 100 kg) force for a 100 mm wide drain is normally specified, as mentioned before. However, one factor which is often neglected is that some drains can have permanent necking once it is stretched. Such a necking reduces the discharge capacity. Therefore, the amount of elongation and necking should also be observed and reported when the tensile strength of drain is measured.

It should also be pointed out that various methods and equipment have been used in determining the properties of PVD, and it may not be possible to compare directly the values provided by suppliers. Even within the same method, the values measured can still vary depending on the testing procedure. As the discharge capacity is dependent on hydraulic gradient, one can compare the discharge capacity only when the values are measured using the same method and at the same hydraulic gradient. Unfortunately, this seldom happens. So, it will be necessary to conduct one's own tests to determine the discharge capacity under standardised conditions.

4.5 QUALITY CONTROL TESTS

Quality control tests are tests used to assess the drain properties, including the discharge capacity, the tensile strength, the permeability, the AOS, etc. All these tests can be done at a site laboratory because the testing facilities required and the testing procedures are simple. In conjunction with the land reclamation projects at Changi East, Singapore, a set of PVD testing equipment and methods have been established, and these have been used for land reclamation projects for the past eight years (Broms et al., 1994; Chu and Choa, 1995; Bo et al., 2000b). These facilities and methods have been improved over the years. They are simple and yet serve their purpose well. Some of them are discussed below.

4.5.1 Discharge Capacity Tests

As the discharge capacity is an important parameter for assessing the quality of vertical drain, a test to determine the discharge capacity is often required. Although the ASTM4716 (ASTM, 1996) is often referred to, it should be pointed out that this standard is for determining the transmissivity of geotextile and not discharge capacity. Therefore, a standard method for measuring the discharge capacity of vertical drain has not yet been established. As such, various devices and methods have been developed for measuring the discharge capacity of vertical drain (Hansbo,1983; Kamon et al., 1984; Guido and Ludewig, 1986; Suits et al., 1986; Bergado et al., 1996). Some of these devices have been described and compared by Holtz et al. (1991), Akagi (1994), and Wang and Chen (1996). These can be classified as follows:

1. With the drain tested within the soil or without soil. For the latter case, a rubber membrane may be used. Test data (Liu et al., 1993; Lee and Kang, 1996) have shown that the discharge capacity measurement is affected by the medium in which the drain is embedded. The discharge capacity measured for the drain in clay is smaller than that on the drain that is tested inside a rubber membrane.
2. Test is conducted on straight drain or deformed (i.e., buckled) drain. Generally the discharge capacities measured on straight and deformed drains are different. Both values are often measured.
3. Short-term (1–2 days) or long-term (2–4 weeks) tests are conducted. Normally, a discharge capacity test is completed within a day or in a few days (short-term test). However, the effect of creep on discharge capacity can be studied by measuring the discharge capacity of the drain after it has been loaded for several weeks.

We recommend the tests in which the drains are embedded inside the soil and tested under both straight and deformed conditions. It is better to conduct short-term tests. However, correlation can be

made between the q_w measured for short-term and long-term tests. Some specific testing methods are given below:

4.5.1.1 Straight drain tester

As already mentioned, ASTM4716 may be used as a reference in conducting this test. However, the method has to be modified to suit the vertical drain. For example, the measurement will reflect the in-situ behavior of the drain better if the drain to be tested is embedded in soil rather than between two rigid platens. A new straight drain tester has been developed and used in Singapore by Broms et al. (1994) and Chu and Choa (1995). It was designed to comply with ASTM4716, but with necessary modification. A cross-section of the straight drain tester is shown in Fig. 4.7. It can accommodate a piece of 100-mm (or 300-mm)-long and 100-mm-wide

drain specimen. It consists of a base and a loading plate. The base plate contains a thin layer of soil and the drain specimen to be tested. It also provides channels for water to flow in and out of the drain. The top plate contains soil and the loading platen. The straight drain tester under testing is shown in Fig. 4.8. The drain specimen after testing is shown in Fig. 4.9.

Grooves have been cut in the base plate at the two ends of the drain to improve the flow along the drain and to facilitate the measurement of head loss. The water heads at the inlet and outlet are measured by standpipes. In this way, the true hydraulic gradient can be measured.

The dimensions of the tester have been chosen to fit the loading frame for standard oedometer test so that the tester can be used in any soil laboratory for routine quality control tests for PVD.

The discharge capacity of drain is determined by the constant head method, that is, by measuring the discharge rate at a certain constant hydraulic gradient. It should be noted that the discharge capacity can be affected by the hydraulic gradient used. Therefore, the discharge capacity should be measured at a hydraulic gradient similar to that in-situ, which is normally between 0.1 and 1. A hydraulic gradient of 0.5 is recommended. The discharge capacity is

calculated on the basis of Darcy's law. Therefore, it is important to maintain a steady flow during the tests and to remove all the air bubbles in the system.

As the discharge capacity of drain changes with the applied vertical load, the discharge capacity is measured at different vertical loads. Vertical pressures of 50, 100, 200, and 300, and 350 kPa are normally used. A load of 350 kPa corresponds roughly to 50m of effective overburden stress, which should be sufficient for most of the land reclamation projects. The discharge capacity measured for two typical types of drain under different pressures are shown in Fig. 4.10.

4.5.1.2 Buckled drain tester

When a drain deforms together with the deformation of clay, its discharge capacity will get reduced. Therefore, it is necessary to measure the discharge capacity of the deformed drain. Generally, the deformation will take one of the following forms as shown in Fig. 4.11, depending on the soil condition. Usually, the drain will buckle inside the soil. The extent of buckling is controlled by the stiffness of the soil, the depth of the soft clay, and the amount of deformation. The softer the soil and the larger the deformation, the greater the buckling.

For discharge capacity measurement, two methods have been adopted to deform the drain before measuring the discharge capacity. The first is to bend or kink the drain manually to a shape desirable, normally to the worst possible condition. The second is to allow the drain to deform naturally in clay to a strain comparable to the field situation. The clay used for the test is prepared to the

consistency of the weakest soil layer at the site. The former, although more stringent, may not be representative of the general condition and thus could underestimate unnecessarily the discharge capacity. Therefore, the latter approach is considered more representative of the in-situ conditions.

The buckled drain tester used at the Nanyang Technological university, Singapore is shown in Fig. 4.12. It was designed with an inner diameter of 150 mm to accommodate a 400-mm-long drain specimen. The cylinder with a drain specimen in the center is filled with reconstituted clay. The clay can be consolidated with compressed air via a piston. During the consolidation, the drain will buckle inside the clay. The vertical strain experienced can be between 20 and 50%. Once consolidation is completed, the discharge capacity can be measured by allowing water to flow from the bottom to the top of the drain. The total heads at both the top and the bottom are measured by standpipes. Pore water pressure at different distance from the drain can also be measured if required. The buckled drain after testing is shown in Fig. 4.13.

4.5.1.3 Kinked drain tester

In case the discharge capacity of kinked drain needs to be measured, the straight drain tester can be modified to test a kinked drain as shown in Fig. 4.14. The profile of the kinking is predesigned with a 15° bend. However, other angles can be easily accommodated. In between the drain, soil or a special waterproof sponge (as shown in Fig. 4.14) cut to fit the kinking profile can be used. A vertical stress can still be applied so that the discharge capacity under different pressures can still be measured.

4.5.1.4 Discussions on discharge capacity measurement

It needs to be pointed out that the discharge capacity measurement is affected by the hydraulic gradient used (Kamon et al., 1984; Broms et al., 1994; Park and Miura, 1998). As such, the discharge capacity should be measured at its in-situ hydraulic gradient. However, the in-situ hydraulic gradient is difficult to estimate. Reports on the field hydraulic gradient are also rare, except for one case reported by Nakanado et al. (1992) in which the in-situ hydraulic gradient was estimated to be between 0.03 and 0.8. From the testing point of view, Akagi (1994) pointed out that when the hydraulic gradient is

higher than 0.5, the flow inside the vertical drain may not be laminar any more. He suggested that the discharge capacity be measured at a hydraulic gradient ranging from 0.2 to 0.5. After analyzing the flow behavior in the drain under different hydraulic gradient, Park and Miura (1998) also suggested that a hydraulic gradient ranging from 0.2 to 0.5 should be used. The data presented in Wang and Chen (1996) also indicate that a steady flow can be difficult to achieve for vertical drain when i> 0.5. Based on the above studies, an i ≤ 0.5 should be used for discharge capacity measurement. However, the testing errors become higher when the hydraulic gradient i is small. Therefore, i = 0.5 appears to be the most suitable value. The recommended approach to discharge capacity measurement is to measure q_w at different i values ranging from 0.1 to 1, to establish the relationship between q_w and i for each pressure used, and then determine the discharge capacity at i =0.5 with reference to the relationship established.

As discharge capacity is a measure of the in-plane flow capacity of the drain, the measurement should not be affected by the smear effect. That is, the drain should be embedded in clay rather than being inserted in clay. The smear effect will affect the cross-plane permeability of the filter. However, this is a different issue. For the same reason, the discharge capacity measurement should not be affected by the size of the soil media used to embed the drain as long as the drain to be tested can be completely embedded in the soil.

4.5.2 Tensile Strength Measurements

To measure the tensile strength of PVD or the filter of PVD, a tensile strength testing machine is required. This type of machine is normally too expensive for a site laboratory to acquire just for testing the tensille strenght of drain. On the other hand, a compression machine used for triaxial tests is commonly available in a geotechnical laboratory, and it can be modified to conduct tensile strength tests for vertical drains.

To use a compression machine for tensile strength test, a pair of clamps must be made to hold the drain specimen in the machine. The clamps should be designed according to ASTM D4632-91.

A drain specimen of 200–400 mm in gauged length can be used. The test setup is shown in Fig. 4.15. The load is applied under constant-rate-of-extension (CRE). ASTM D4533 is often specified as the method to measure the tensile strength of drain in which the pull rate specified is 300 mm/min. A compression machine does not provide such a high pulling rate. However, this is not a problem because a slower rate will yield a smaller tensile strength. Therefore, when a compression machine is used, the tensile strength measurement will be on the conservative side.

Normally, the entire drain and the filter in both wet and dry conditions are tested. For wet conditions, the specimen is immersed

in water for 48 h before testing. As an example, the test results conducted on two typical types of PVD are shown in Fig. 4.16.

4.5.3 Cross-Plane Permeability of Filter

The cross-plane permeability controls the rate of water flow into a vertical drain. It can be measured using a constant head method. A permeameter that has been designed particularly for this purpose is shown in Fig. 4.17. It has an inner diameter of 62 mm to suit a drain of 100 mm in width. It was designed in accordance with ASTM D4491-96. The test setup and procedures also follow ASTM D4491-96.

A layer of filter with a thickness of δ is cut in a circular disk and placed in the permeameter. The test is conducted using a constant head method. A laminar flow condition is first established before permeability measurement. Only after a laminar flow is achieved was the flow quantity, Q, measured over a time interval, t, under a constant head, ∆h. The cross-plane permeability of the filter is calculated by:

where: k = permeability of the filter at a temperature of 20◦C, m/s; Q =quantity of flow, m³; δ = thickness of the filter tested, m; A = effective cross-section area of the filter, m²; ∆h = the head difference, m; t = time interval for flow to take place, s; R_t = temperature correction factor.

For each specimen, the permeability is measured under different water heads. As specified in ASTM D4491-96, the permeability is taken as the value corresponding to a 50-mm water head. As the filter is thin, 3–5 layers can be used. In this case, δ should be the total thickness of the filter.

4.5.4 AOS of Filter

The common method used to measure the AOS is to conduct sieve analysis using standard beads. This method is applicable to AOS

larger than 40m. A sieve shown in Fig. 4.18 has been specially made to anchor the filter to form a sieve. ASTM D4751-87 is followed in conducting the tests, except that the diameters of the glass beads used in the test range from 40to 170 µm, instead of from 75 µmto 170 µm only. In complying with the ASTM standard, the tests were conducted under a relative humidity of 60% and a temperature of 23°C. An electromagnetic shaker is used. The shaking duration is 10min.

In conducting an AOS test a single layer of filter, 100 mm wide and 100 mm long, is firmly placed in a specially designed sieve frame to form a sieve. Fifty grams of glass beads with diameters covering the 40–170 µm range is placed on the filter in the sieve. The sieve with its cover and pan is then placed on a shaker and shaken for 10min. The total mass of the glass beads in the pan, P, is measured and used in calculating the beads passing through the specimen:

where: B = beads passing through specimen, %; P =mass of glass beads in the pan, g; and T = total mass of glass beads used, g.

The B value is then used to correlate with the cumulative size distribution percentage of spheres with diameters less than the size as shown in Table 4.2, which is provided by the manufacturer.

4.6 SPECIFIC ASPECTS

4.6.1 Smear Effect

As the costs of vertical drain and installation have drastically declined in recent years, there is a tendency to use closer drain spacing to reduce consolidation time. However, when the drains are installed too close to one another, the smear effect may become too large to be ignored. The ‘smear’ effects come from the compressibility of soil and the disturbance to the soil structure. As discussed in Ch. 3, the smear zone can be as large as four times the size of the mandrel or five to eight times the equivalent diameter of drain. The dimension of a small mandrel is 120mm. The equivalent diameter of a drain is 66 mm. Therefore, the size of the smear zone can be as large as 480–528 mm. If the drain spacing is 1 m and the smear zone is 500 mm, then the soil is disturbed almost everywhere. As smear can cause a significant reduction in the permeability and the coefficient of consolidation of soil (see Ch. 3), it can be counterproductive when the drain spacing is too close. This is particularly so when the soil is sensitive. Back calculations using field settlement data obtained from some reclamation projects also show that the back-calculated c_h value, which reflects the smear effect, can be smaller than the c_v value measured by oedometer tests. If there is no smear effect, the c_h value should be at least two times higher than c_v. Therefore, when

the drain spacing is small (say, less than 2 m), the smear effect can become quite significant.

As discussed in the preceding section, the smear effect can be quite large if the drain spacing is small. So it becomes important to choose a mandrel that causes the least smear effect. For the same reason, the anchor plate used should also be small. The mandrels used generally come with four types of cross-sections: (a) rhombic, (b) rectangular, (c) square, and (d) circular as will be shown in Ch. 5. According to Bo et al. (2000c), the rhombic mandrel causes the least disturbance. In terms of driving mechanisms, the vertical drain rigs with static pushing should cause less disturbance than the rigs that use vibration. On the other hand, the smear effect is normally more severe in soft clay than in stiff clay. The vibrating type of rigs should not be used in soft clay. The speed of penetration and withdrawal may also play an important part. However, this factor has not been studied in detail so far. More aspects of the vertical installation will be discussed in Ch. 5.

Other factors to be considered during the installation of vertical drain are the verticality and the joints of the drain. For the former, one needs to check whether the drain rig is vertical. For the latter, only the methods that do not reduce the tensile strength of the drain should be used. For this, tensile strength tests on the joints should be conducted.

4.6.2 Standardization

As discussed in the preceding sections, a number of ASTM D-series standards have often been used as the standard for testing of vertical drains. However, some of those ASTM standards are not specifically written for vertical drains. Therefore, the testing procedures stipulated in the standards may not be the most suitable method. Some of the ASTM standards can be complied with by using different testing systems. This is why there are so many testing methods published in the literature and used by different organizations. As the performance of vertical drain can be affected considerably by the quality of the drain used and the control in the construction procedures, it will be highly desirable to set up some regulations

or codes of practice to govern the selection of vertical drain and to regulate the construction activities. Two such codes, JTJ/T256-96 and JTJ/T257-96, have been put into practice in China since 1996. JTJ/T256-96 (1996) is used as a design code to control the practice for installation of PVDs. JTJ/T257-96 (1996) stipulates the quality inspection standard for PVDs. As of now, China is probably the only country that uses codes to guide and control construction activities related to the use of PVDs.

4.7 CONCLUDING REMARKS

The properties of PVD, the factors that control the performance of PVD and the quality control tests, are discussed in this chapter. Other factors that affect the performance of vertical drains, such as the quality of vertical drain and the smear effect, are also mentioned. In conclusion, the selection of suitable PVDs and checks on the quality of the drains to be used are two important steps in achieving a cost-effective design for a soil improvement project using PVDs.