You are on page 1of 8

The Biaxial Load-Strain Behaviour of Biaxial Geogrids

J. Kupec & A. McGown Department of Civil Engineering, University of Strathclyde, Glasgow, UK. geotech@strath.ac.uk

ABSTRACT: Technical assessments of structures reinforced with biaxial geogrids with integral or welded junctions in the field or laboratory, have consistently indicated much lower strain development in the geogrids than can be reasonably predicted by conventional or finite element analysis. It appears that biaxial geogrids exhibit much higher stiffness when acting under biaxial operational load/strain conditions than under uniaxial laboratory testing conditions. For this reason, in-isolation biaxial testing of these materials has been undertaken on three biaxial geogrids with integral or welded junctions. Test results are presented.

1 INTRODUCTION Polymeric geogrids are being used in an increasing number of civil engineering and environmental protection applications. They are produced in a variety of geometrical forms using a wide range of polymeric materials and numerous manufacturing techniques. However, two classes of geogrid reinforcements may be identified; uniaxial geogrids, which develop tensile stiffness and strength primarily in one direction, and biaxial geogrids which, develop tensile stiffness and strength in two orthogonal directions. the test methodologies employed to To date, characterise the load-strain-time properties of geogrids have involved the application of uniaxial loading, BS 6906 (1987), GRI-GG4 (1987), ISO 13431 (1999), ISO 10319 (1999), ASTM D5262 (2002). For biaxial geogrids this has generally involved undertaking two separate tests in orthogonal directions. Attempts to combine the measured properties in these two directions in order to obtain the overall biaxial properties/behaviour of the geogrids have proven to be problematic, McGown & Kupec (2004). the basic behavioural differences of In this paper, various forms of geogrids are related to their applications. Details are given of the development of an inisolation, (i.e. tested in-air and not in-soil), biaxial loading method for biaxial geogrids with welded and integral junctions. Test data obtained from short-term sustained biaxial loading tests are presented.

2 GEOGRID TYPES The load-strain behaviour of geogrids may vary significantly in the two axes of principal stiffness. This is often associated with variations in the material and geometrical properties of bars and junctions in these directions, Fig. 1. 2.1 Uniaxial Geogrids Uniaxial geogrids usually exhibit a high stiffness in the machine direction [MD] with a very low to negligible stiffness in cross-machine direction [XMD], however, it should be noted that there are products which have their maximum properties in the XMD. The main functions of the secondary cross-members and junctions are principally to provide geometrical stability during transport and installation, but they may also provide the possibility of interlock with the soil in which they are placed. These uniaxial geogrids are intended for use in plane strain applications, where the secondary direction has little or no tensile loading, i.e. plane strain applications. Some design codes or methods suggest/specify that two layers of uniaxial geogrids should be laid orthogonally in order to resist biaxial loading, e.g. BS 8006 (1995), EBGEO (1997).

349

2.2 Biaxial Geogrids Biaxial geogrids exhibit significant stiffness and strength in two orthogonal directions. In these materials, the bars and junctions provide geometrical stability during transport and installation and almost always provide interlock with the soil in which they are placed. 2.2.1 Anisotropic Biaxial Geogrids Anisotropic geogrids exhibit dissimilar stiffnesses in the two principal directions. They are used in anisotropic loading conditions, i.e. where there is both a primary and secondary degree of loading /strain. 2.2.2 Isotropic Biaxial Geogrids Isotropic biaxial geogrids exhibit very similar stiffnesses and strengths in two orthogonal directions. They are used in isotropic loading conditions, i.e. where there is almost an equal degree of loading/strain in two orthogonal directions. 3 GEOGRID FUNCTIONS It is important to relate the material properties of reinforcing products to their performance requirements under operational conditions, i.e. to identify both the Serviceability Limit State and the Ultimate Limit State requirements, in-soil behaviours. Test methods for uniaxial geogrids are well established and are currently employed to test biaxial geogrids. Short-term properties may be determined by a wide range of Index tests, which produce a broad range of quality control properties, e.g. Constant Rate of Strain [CRS] tests provide short-term stiffness and strength properties. Long-term properties may be determined by established Performance tests and are used in designs, e.g. sustained loading (creep) tests and stress relaxation tests provide longterm stiffness and strength properties. It has to be highlighted that the above properties are determined in a uniaxial manner, hence for biaxial geogrids two specimens are tested in orthogonal directions. Thus at the present time, uniaxial geogrids and anisotropic or isotropic biaxial geogrids are all tested in a very similar manner in-isolation or in-soil as test conditions for biaxial geogrids are identical to those employed in uniaxial geogrid testing, the biaxial nature of biaxial geogrids is largely being ignored. It is therefore suggested that a new approach to the testing of biaxial geogrids is required in order to determine their biaxial load-strain behaviour.

4 LOAD-STRAIN TESTING OF BIAXIAL GEOGRIDS Various forms of biaxial short-term and long-term load testing of geosynthetics have been undertaken previously and most have suffered specimen distortion into non-orthogonal shapes, non-uniform stressing particularly at the junctions, premature specimen rupture and/or problems with clamping of the specimens, Bhmert (1981), Bush & Bhmert (1992) and Nimmesgern (1994), Saathof (1997). However, in other engineering disciplines, biaxial loading test methods have been developed and proven to be capable of establishing the properties of biaxial products without the difficulties experienced in the testing of geosynthetics, e.g. Krause & Bartolotta (2001) and Bridgens & Gosling (2003) as illustrated in Fig. 2. Using the knowledge gained from biaxial testing in other disciplines, a new approach to testing biaxial geogrids is presented in this paper. In particular details are given of test specimen sizes and shapes, clamping arrangements, and test methodology. 4.1 Material used for Biaxial Load Testing of Biaxial Geogrids The properties and geometry of the biaxial geogrids tested are described in Table 1.

Table 1. Properties of tested geogrids


Property Mass Per Unit Area Polymer Composition Molecular State A 580 PP Varies from random to oriented (Amorph to SemiCrystalline) Integral Biaxially stretched punched sheet 40 / 40 Biaxial Geogrid B 370 PP Highly oriented (SemiCrystalline) C 560 PET Highly oriented (SemiCrystalline)

Junctions Macro Structure

Welded Welded Welded pre- Welded prestretched stretched monolithic monolithic flat bars flat bars 60 / 60 60 / 60

Short-term Nominal Strength*


* Manufacturer's specification

350

GEOGRIDS UNIAXIAL ANISOTROPIC Essentially uniaxial loadstrain-time behaviour Secondary direction stiffness very low or negligible Anisotropic load-straintime behaviour MD XMD MD and XMD exhibit dissimilar stiffnesses BIAXIAL ISOTROPIC Isotropic load-strain-time behaviour MD XMD MD and XMD exhibit very similar stiffnesses

UNIAXIAL LOAD/ STRAIN BEHAVIOUR

ISOTROPIC LOAD/ STRAIN BEHAVIOUR

Figure. 1 Types of geogrids

Figure. 1 Types of geogrids

(a) Geogrid with integral junctions

(b) Geogrid with welded junctions


Figure 2. Biaxial test apparatus for high-strength fabrics, after NASA Krause & Bartolotta (2001) Figure 3. Typical cruciform biaxial test specimens for biaxial geogrids

351

(a) Biaxial creep test apparatus with load arrangements

(b) Clamped specimen

Figure 4. Biaxial sustained loading (creep) test apparatus for geosynthetics testing 4.2 Specimen Sizes Cruciform shaped test specimens, similar to those employed in other disciplines were used in order to avoid problems reported in earlier work with square or rectangular specimens. The cruciform shape prevents the distortion of the test specimen at higher strain levels and avoids the problem associated with non-uniform stress conditions. Representative geogrid sample sizes incorporating at least 5 bars in each direction and so 25 junctions were selected to represent the grid behaviour. The central biaxial test specimen size was between 100 and 220 mm square, with an overall cruciform specimen dimensions of some 500500 mm, Fig. 3. 4.3 Test Methodology Test results from biaxial in-isolation CRS Index tests were reported by McGown & Kupec (2004). This paper seeks to determine long-term material properties that may be used in designs. It is suggested that the in-isolation and in-soil behaviour of the materials reported in this paper are very similar, therefore the uniaxial sustained loading tests were carried out strictly according to existing standards, i.e. BS6906-Part 5 (1991), ISO 13431 (1999), in the biaxial creep test apparatus to determine if differences between test apparatuses existed. In fact, the differences measured were negligible; hence all subsequent sustained loading tests were conducted in the biaxial creep test rig, Fig. 4. The biaxial sustained loading (creep) testing was undertaken with the identical clamping arrangements used in standard uniaxial sustained loading (creep) tests. The sustained loads in the two orthogonal directions, were applied in an isotropic manner, i.e. the loads in both directions were equal. This is not a requirement of the test procedure, anisotropic loads may be applied. However, no matter what loads are applied they must be applied in both directions at the same time. The sustained loads were thus applied simultaneously at the beginning of the test. Subsequently they were removed at the end of the test, again simultaneously, in order to observe the elastic (time independent) strain upon unloading and then the time dependent recovery strains. The sustained loads in each direction were varied between 10 and 50 percent of the nominal short-term tensile strength obtained from uniaxial (Index) testing conducted according to BS 6906 (1991). At least three tests were conducted at sustained load levels equal to 30 percent of shortterm loading to determine the specimen variation. The specimen variation was within 95% confidence limits and all subsequent tests were carried out once only. All loads were applied for at least 100 hours and after unloading the specimen was observed for further 100 hours to determine any recovery over time. The tests were carried out under a controlled atmosphere of 20 2C and 65% relative humidity. Deformation measurement was conducted by using a 2MPixel digital camera which achieved a postprocessing accuracy of 100m using software developed for the camera and National Instruments IMAQ Vision. 5 TEST RESULTS Uniaxial and biaxial sustained loading test results for geogrids A, B and C at load levels from 10 to 50 percent of nominal strength are presented in Figs. 5 to 7. The test results obtained at 30 and 50 percent nominal strength are highlighted to ease the interpretation of test data and enable comparisons to be made.

352

10 8 Strain [%] 6 4 2 0 -3 10
Geogrid A Creep uniaxial % of Nominal Strength 50% 40% 30% 20% 10%

MD

10 8 6 4 2 0 -3 10

XMD
Geogrid A Creep uniaxial % of Nominal Strength 50% 40% 30% 20% 10%

10

-2

10

-1

1 10 Time [hrs]

10

10

10

10

-2

10

-1

1 10 Time [hrs]

10

10

10

a) Uniaxial sustained loading (creep) test results


10 8 Strain [%] 6 4 2 0 -3 10

MD
Geogrid A Creep biaxial % of Nominal Strength 50% 40% 30% 20% 10% not tested

10 8 6 4 2 0 -3 10

XMD
Geogrid A Creep biaxial % of Nominal Strength 50% 40% 30% 20% 10% not tested

10

-2

10

-1

1 10 Time [hrs]

10

10

10

10

-2

10

-1

1 10 Time [hrs]

10

10

10

b) Biaxial sustained loading (creep) test results


10 8 Strain [%] 6 4 2 0 -3 10

MD
Geogrid A % of Nominal Strength Creep biaxial Creep uniaxial 50% 50% 30% 30%
uniax @50%

10 8

XMD
Geogrid A % of Nominal Strength Creep biaxial Creep uniaxial 50% 50% 30% 30%
uniax @50%

biax @50% uniax @30% biax @30%

6 4 2 0 -3 10

biax @50% uniax @30% biax @30%

10

-2

10

-1

1 10 Time [hrs]

10

10

10

10

-2

10

-1

1 10 Time [hrs]

10

10

10

c) Comparison of uniaxial and biaxial sustained loading test results at 30 and 50% nominal strength Figure 5. Uniaxial and biaxial sustained loading (creep) test results geogrid A [PP]

353

10 8 Strain [%] 6 4 2 0 -3 10
Geogrid B Creep uniaxial % of Nominal Strength 50% 40% 30% 20% 10%

MD

10 8 6 4 2 0 -3 10

XMD
Geogrid B Creep uniaxial % of Nominal Strength 50% 40% 30% 20% 10%

10

-2

10

-1

1 10 Time [hrs]

10

10

10

10

-2

10

-1

1 10 Time [hrs]

10

10

10

a) Uniaxial sustained loading (creep) test results


10 8 Strain [%] 6 4 2 0 -3 10

MD
Geogrid B Creep biaxial % of Nominal Strength 50% 40% 30% 20% 10%

10 8 6 4 2 0 -3 10

XMD
Geogrid B Creep biaxial % of Nominal Strength 50% 40% 30% 20% 10%

10

-2

10

-1

1 10 Time [hrs]

10

10

10

10

-2

10

-1

1 10 Time [hrs]

10

10

10

b) Biaxial sustained loading (creep) test results


10 8 Strain [%] 6 4 2 0 -3 10

MD
Geogrid B % of Nominal Strength Creep biaxial Creep uniaxial 50% 50% 30% 30%
uniax @50% biax @50%

10 8 6

XMD
Geogrid B % of Nominal Strength Creep biaxial Creep uniaxial 50% 50% 30% 30%
uniax @50%

biax @50% uniax @30%

uniax @30% biax @30%

4 2 0 -3 10

biax @30%

10

-2

10

-1

1 10 Time [hrs]

10

10

10

10

-2

10

-1

1 10 Time [hrs]

10

10

10

c) Comparison of uniaxial and biaxial sustained loading test results at 30 and 50% nominal strength Figure 6. Uniaxial and biaxial sustained loading (creep) test results geogrid B [PP]

354

10 8 Strain [%] 6 4 2 0 -3 10
Geogrid C Creep uniaxial % of Nominal Strength 50% 40% 30% 20% 10%

MD

10 8 6 4 2 0 -3 10

XMD
Geogrid C Creep uniaxial % of Nominal Strength 50% 40% 30% 20% 10%

10

-2

10

-1

1 10 Time [hrs]

10

10

10

10

-2

10

-1

1 10 Time [hrs]

10

10

10

a) Uniaxial sustained loading (creep) test results


10 8 Strain [%] 6 4 2 0 -3 10

MD
Geogrid C Creep biaxial % of Nominal Strength 50% 40% 30% 20% 10%

10 8 6 4 2 0 -3 10

XMD
Geogrid C Creep biaxial % of Nominal Strength 50% 40% 30% 20% 10%

10

-2

10

-1

1 10 Time [hrs]

10

10

10

10

-2

10

-1

1 10 Time [hrs]

10

10

10

b) Biaxial sustained loading (creep) test results


10 8 Strain [%] 6 4 2 0 -3 10

MD
Geogrid C % of Nominal Strength Creep biaxial Creep uniaxial 50% 50% 30% 30%

10 8 6
uniax @50% biax @50% uniax @30% biax @30%

XMD
Geogrid C % of Nominal Strength Creep biaxial Creep uniaxial 50% 50% 30% 30%

4 2 0 -3 10

uniax @50% biax @50% uniax @30% biax @30%

10

-2

10

-1

1 10 Time [hrs]

10

10

10

10

-2

10

-1

1 10 Time [hrs]

10

10

10

c) Comparison of uniaxial and biaxial sustained loading test results at 30 and 50% nominal strength Figure 7. Uniaxial and biaxial sustained loading (creep) test results geogrid C [PET]

355

6 INTERPRETATION OF TEST RESULTS The strain against log.time plots for geogrids A, B and C with integral or welded junctions indicated increased stiffness for specimens tested biaxially over those tested uniaxially. What can be observed is that there are significant changes in the initial, i.e. elastic, strain upon loading. Furthermore, geogrid A exhibits changes in the time dependent strains developed under sustained loading due to creep of the material. It is therefore suggested that changes in behaviour under biaxial loading may be twofold; firstly there may be changes in the initial strains and secondly there may be changes in the creep rate developed with time. In each case the biaxially loaded behaviour of biaxial geogrids is characterised by significantly greater stiffness at any time after loading. 7 DISCUSSION & CONCLUSIONS The tests show that geogrids with integral or welded junctions develop significantly higher stiffnesses under in-isolation biaxial loadings. These changes in stiffness may be expressed as ratios of biaxial to uniaxial initial strains and ratios of biaxial to uniaxial strains developed during sustained loading. It is suggested that the Isochronous Strain Energy approach may be used to combine the test data obtained from biaxial testing, as strain energy is a scalar quantity and not a vector quantity. The applications of this approach are many, as indicated in previous work by McGown (2000). Wovens and geonets are likely to reflect similar differences if they have entangled junctions and are tested in-soil. It is therefore required to undertake in-soil biaxial testing to determine any differences for such woven and geonet materials with entangled junctions. Lastly, it is suggested that the increase in biaxial stiffness may be directly related to the behaviour of junctions under tensile stresses or strains in the principal, orthogonal directions. The apparent increase in stiffness may be ascribed, to some degree, to the effect of Poissons Ratio and to the re-orientation of the amorph molecules in the junction areas. This effect is discussed in more detail by Kupec et al (2004).

REFERENCES ASTM D5262-02a 2002: Standard Test Method for Evaluating the Unconfined Tension Creep Behaviour of Geosynthetics. American Society for Testing and Materials, USA.

Bhmert, W. 1981: Prfverfahren und Einrichtungen zur Ermittlung des Kraft-Dehnungsverhaltens zweiachsikg beanspruchter beschichterer Gewebe fr Membran-konstruktionen., Vol.3, p. 159-176. Bridgens, B.N. & Gosling, P.D. 2003: Biaxial fabric testing to determine in-situ material properties. Proceedings of the Int. Conf. on Structural Membranes 2003, Barcelona, Spain. BS 6906 1987. Determination of tensile properties of geosynthetics. Part 1 & 5, British Standard Institution, London, UK. BS 8006 1995: Code of Practice for reinforced/strengthen soils and other fills. British Standard Institution, London, UK. Bush, D.I. & Bhmert, W. 1992: Geosynthetic reinforcements subject to bi-directional stress An experimental study. Earth Reinforcement Practice, eds. Ochiai, Hayashi & Otani, Balkema, Rotterdam, The Netherlands, p. 45-50. EBGEO 1997: Empfehlungen fr Bewehrungen aus Geokunststoffen. Deutsche Gesellschaft fr Geotechnik (Hrsg.) Verlag Ernst&Sohn, Berlin, Germany. GRI-GG4 1987: Determination of the Long-Term Design Strength of Stiff and Flexible Geogrids. Geotextile Research Institute, USA. ISO 10319 1993: Geotextiles - Wide-width tensile test. International Standard Organisation. Switzerland. ISO 13431 1999: Geotextiles and geotextile-related products - Determination of tensile creep and creep rupture behaviour. International Standard Organisation. Switzerland. Krause, D. & Bartolotta, A. 2001: Biaxial Testing of High-Strength Fabric Improves Design of Inflatable Radar Domes. NASA Glenn's Research & Technology reports, www.grc.nasa.gov. Kupec, J., McGown, A. & Ruiken, A. 2004: Index Junction of Geogrids. Proceedings of the 3rd Asian Regional Conf. on Geosynthetics, Seoul, Korea, to be published. McGown, A. 2000: 4th Mercer Lecture: The behaviour of geosynthetic reinforced soil systems in various geotechnical applications. EuroGeo 2000, Bologna, Italy: 3-26 McGown, A. & Kupec, J. 2004: A new approach to the assessment of the behaviour of geogrids subject to biaxial loading. Proceedings of 3rd European Conf. on Geosynthetics EuroGeo3, Mnchen, Germany, Vol. 2, p.643-648. Nimmersgern, M. 1994: Large scale biaxial creep tests on a Polypropylene geogrid. Otto-Graf Journal, Germany, p. 137-147. Saathof, F. 1997: Effects of stretched geotextiles in contact with soil. Proceedings 4th Int. Conf. on Soil Mechanics and Foundation Engineering, Hamburg, Germany, Vol. 3, p.1781-17848.

356

You might also like