Accelerated flow testing of geosynthetic drains

By Sam Allen

Introduction

Most planar geosynthetic drains consist of a polymeric core with or without a geotextile bonded to one or both sides. The core, whether used by itself or in association with a geotextile, forms the main fluid transmission medium. Synthetic cores used in North America typically take the form of geonets that, when laminated with geotextiles, are referred to as geocomposites.

Cores are also manufactured from geomembranes having nubs or three-dimensional drain patterns, or drainage sheets. Properties of these sheet drains, geocomposites, and geonets, such as compression strength, flow rate, roll dimensions, and interface shear strength determine the suitability of any product for a particular application.

One common concern with all of these products is the time-dependent loss of flow under load that results from the compressive creep of the core. Creep refers to a time-dependent deformation when a material is subjected to load for a prolonged period of time. For geosynthetic planar drains the type of load is compressive, hence creep leads to a decrease in thickness.

Fannin et. al. (1998) measured flow rate for geonets for 120 hours. Important conclusions of this work include that

• (a) the relationship between hydraulic performance and logarithm of time is linear;
• (b) the creep of a product is specific to the structure of the core of that product; and
• (c) significantly longer test durations than those covered by the authors were necessary for the data to be of any practical value.

With these limitations and acknowledgement of the variability in compressive creep properties and associated observed flow in geosynthetic drains, a focus on time-dependent flow measurements was established.

One result was the development of GRI–GC8, Determination of the Allowable Flow Rate of a Drainage Geocomposite (2001). This guide, still widely used today, presented a methodology for determining the allowable flow rate of a candidate drainage geocomposite. The resulting value could be used directly in a hydraulics-related design to arrive at a site-specific factor of safety. The procedure is to first determine the candidate drainage composite’s flow rate for 100 hours under site-specific conditions, and then to modify this value by means of creep reduction and clogging reduction factors as represented below in Equation 1.

q_allow = allowable flow rate
q₁₀₀ = initial flow rate determined under simulated conditions for 100-hour duration
RF_CR = reduction factor for creep to account for long-term behavior
RF_CC = reduction factor for chemical clogging
RF_BC = reduction factor for biological clogging

While this guide provides suggested ranges of clogging reduction factors, there was none provided for compressive creep given the readily accepted procedures to measure this property for each product and its structure. Indeed, ASTM D7406–Standard Test Method for Time-Dependent (Creep) Deformation Under Constant Pressure for Geosynthetic Drainage Products was published based on a 10,000-hour required measurement criteria.

Significantly, Narejo and Allen (2004) outlined an accelerated procedure to measure compressive creep properties using the Stepped Isothermal Method (SIM). They demonstrated in this work that accelerated SIM creep tests could be performed on drainage products and that such tests could be accomplished within a few hours as opposed to months required for conventional creep testing at room temperature.

The SIM application for compressive creep was standardized through ASTM by Allen and is published today as ASTM D7361–Standard Test Method for Accelerated Compressive Creep of Geosynthetic Materials Based on Time-Temperature Superposition Using the Stepped Isothermal Method. The test resulted in a full understanding of the thickness vs. time relationship for the tested product under a given compressive load.

Still, the 100-hour flow measurement prescribed by GRI–GC8 has contributed to many product design and conformance/verification testing programs. The flow is measured in accordance with ASTM D4716–Standard Test Method for Determining the (In-plane) Flow Rate per Unit Width and Hydraulic Transmissivity of a Geosynthetic Using a Constant Head.

The benefits and challenges realized by this approach are outlined in Table 1 below.

Thickness dependent transmissivity

In an effort to accelerate the 100-hour flow test prescribed by GRI–GC8 while taking advantage of a predetermined thickness vs. time relationship for a given geosynthetic drain under a given load, a thickness-dependent flow test was developed. Paramount to this approach was the measurement of a product thickness when the given product was sandwiched between adjacent layers in a site-specific drainage layer during a flow test.

The capture of product core thickness when adjacent to GCLs, geomembranes, and aggregates was necessary to correlate a specific product thickness to a given flow.

A hydraulic transmissivity test apparatus was modified to accomplish this goal. This was facilitated with the addition of several thickness monitoring devices in the sample loading tray to track changing thickness of the core during flow testing. The thickness monitoring devices consisted of 1in.-diameter by 1/16in.-thick stainless steel top plates that connected to a 1/16in.-diameter stainless steel, solid rod.

The bottom plate was also 1in. in diameter and 1/16in. thick, had a 1/16in.-diameter center hole and was connected to a hollow 1/8in.-diameter tube. The upper plate was positioned directly above the bottom plate so that the solid rod attached to the upper plate passed through the hole in the bottom plate and through the hollow rod.

Figure 1 shows the device described.

Five of the monitoring devices were used: one in the center and the other four located 1in. in from the corners of the 12in. × 12-n. ASTM D4716 test loading tray. A 1/16in.-deep × 1in.-diameter seat was machined into the base plate of the loading tray at each of the five monitoring locations. The loading platen above the specimen was also machined with five circular 1/16in.-deep seats located directly above the upper plates. These seats were used to accommodate the monitoring plate thickness during the compression of the tested geocomposite.

Testing programs

Thickness monitoring test

To assess the success of a thickness-dependent transmissivity approach, an initial test was performed to record the thickness of the geonet layer of a geocomposite during a 100-hour performance transmissivity test, complete with underlying geomembrane and overburden drainage sand.

The test was performed under a compressive stress of 10,000psf and a hydraulic gradient of 0.1. The tested geocomposite was outfitted with the thickness monitoring devices before hydraulic testing. The bottom textile of the geocomposite was removed from the specimen and placed on the loading tray base. Holes were punched in the textile at the five monitoring locations and the bottom plates were inserted through the textile and the transmissivity loading tray base. The five tubes extended approximately 4in. below the table bottom.

Then the geonet and upper geotextile were placed on the bottom plates and bottom textile. The upper geotextile was carefully cut and peeled back to allow the upper steel pins to be inserted through the geonet openings, through the bottom plates and tubes, and extended through the bottom tubes so they protruded approximately 1in. below the ends of the hollow bottom tubes.

Test thickness monitoring was accomplished by recording the distance between the end of the hollow tube connected to the plate under the geonet and the end of the solid rod connected to the plate on top of the geonet with digital calipers. Figure 2 shows this measurement.Core geonet thickness and system flow were recorded at 0.25, 1, 66, and 100 hours.

Thickness dependent test

A new sample of geocomposite, aligned in the same machine direction location as the previously tested specimen, with the same test profile with geomembrane and sand, was tested using the thickness dependent flow approach.

The geocomposite was equipped with the same thickness monitoring devices and then tested for hydraulic flow. However, instead of flow measurements secured at 0.25, 1, 66, and 100 hours, test data was collected at each of the approximate geonet core thicknesses at these test times, recorded during the previous specimen flow test.

Test results

Table 2 and Figure 3 present collected measurements.

100-hour creep test using SIM

In accordance with ASTM D7361, an additional specimen of the geocomposite, aligned in the same machine direction location as the previously tested specimens, was stripped of its laminating geotextile of each side of the geonet core, and tested for compressive creep properties under 10,000psf loading.

To start the test, the test specimen was placed within flat platens inside an environmental chamber as shown in Figure 4.The cross-head was adjusted so that platens were touching the specimen but there was minimal load on the specimen. Then the specimen was loaded at the rate of 2mm per minute to the 10,000psf load, which was then maintained for 1,000 seconds.

The first increment of test temperature was room temperature of 20 C (68 F). While compressive load was kept constant, distance between the plates was measured via LVDTs attached to the platens and recorded via computer. At the end of 1,000 seconds, the temperature was raised by 7 degrees to 27 C (81 F).

This process was continued until all six temperature increments were applied. The test data was then processed using the methodology of shifting data for each temperature increment to obtain a master curve reaching an accelerated test time of 10,000 plus hours.

Because we were interested in thickness dependent flow, however, the master compressive creep curve was used to select thickness data at 0.25, 0.5, 1, 66, and 100 hours only. This data is presented in Table 3 and shown and compared to time dependent flow test thickness data in Figure 5.

Discussion

Thicknesses from the SIM test matched well with thicknesses observed during the flow test. This is not surprising because the specimen selection was carefully performed to match the exact locations in the machine and transverse direction to minimize initial specimen thickness and geometry.

This excellent comparison between in-flow-test thicknesses and compressive-creep thicknesses, as well as the demonstrated ability to perform a representative thickness-dependent transmissivity flow test, suggest that quick load, thicknesses-dependent transmissivity tests may be used to investigate and qualify geocomposite products for flow properties. This is contingent, of course, upon knowing the product thickness vs. time relationship.

Several key advantages are realized by this approach:

• In multiple sample 100-hour transmissivity test regimens, which are often specified in conformance/verification testing programs, an intial thickness monitoring 100-hour test may serve to invite other quick load tests using initial thickness vs. time data collected, saving significant time and resources.
• Quicker testing, using principles of thickness-dependent flow testing, invite the user to test multiple specimens, avoiding the single-specimen test protocol often burdening typical 100-hour test programs.
• Employment of SIM, and longer-term conventional creep test data to establish thickness vs. time relationships for geonet core products, may be used successfully in transmissivity or flow test evaluations. This data is already needed for development of creep reduction factors needed for allowable flow determination, and may also be used for thickness dependent transmissivity tests.

The thickness-dependent transmissivity test does not come without challenges. Key among these is the higher loads necessary to affect a predetermined thickness in the short term. To affect a given geonet core thickness observed under 10,000psf during 100-hour flow testing, significantly higher loads were required during quick load, thickness-dependent flow testing.

Table 4 below lists the required applied loads to achieve the required thickness for thickness dependent flow testing.

While these loads were all well below the geocomposite’s short-term compressive strength of 44,000psf, one can predict that this will not always be the case for other applications. Key awareness and care need to be exercised in this regard.

Additional concerns are the in-transmissivity test frame thickness devices themselves. Great care must be taken to ensure that they do not interfere with measured flow events. The fragility that results for minimizing size and related impact of the devices must be appreciated, and associated care and operation of the data collection are mandatory for success.

Conclusion

While the data presented here is limited and provides only a small representation of this new measurement technique, it does suggest that thickness-dependent transmissivity tests may play an important role in defining time-dependent flow of geosynthetic drainage systems and qualifying materials in a significantly faster and more economical way.

In concert with SIM to establish time-dependent thickness retention of the geocomposite core, and associated use of this information in thickness-dependent flow tests, may assist in more robust characterization of these important products. Robust research continues in this regard.

Sam Allen is vice president and division manager at TRI Environmental/Geosynthetics Services, Austin, Texas, USA; +1 512 263 2101, sallen@tri-env.com.
This paper was originally presented at Geo-Frontiers, March 14, 2011, in Dallas, Texas, USA.
Reprinted from the “Proceedings of Geo-Frontiers 2011”; edited for Geosynthetics magazine style and format.

References

Fannin, R., Choy, H., Atwater, J. (1998); “Interpretation of Transmissivity Test Data for Geonets,” Geosynthetics International, Vol. 5, No. 3, pp. 265-285.

Narejo, D., Allen, Sam.R. (2004); “Using the Stepped Isothermal Method for Geonet Creep Evaluation,” EuroGeo 3 Conference 2004.