BEHAVIOUR OF CEMENT BOUND SOIL AGGREGATE MIXTURES FOR BASE COURSES

Abstract

In India the need for constructing large mileage of of low cost roads to cater for medium to heavy traffic is well established. The conventional specifications require the use of hard stone for base courses, which involves its haulage from long distances in most cases, thereby increasing the cost of construction exorbitantly. Use of locally available mater ials for construction of roads is gaining considerable popularity these days in order to bring down the cost without sacrificing the strength of pavement structure. In localities where bricks are cheaply available, use of overburnt brick ballast against conventional crushed aggregate will further economise the road construction. This study, therefore, aims at exploring the possibility of the use of Cement Bound Soil-Aggregate (CBSA^) mixtures for base courses with overburnt brick ballast as coarse aggregate and low proportions of cement to bind the soil and aggregate. Investigations are carried out to establish suitable proport ions of various ingradients in the CBSA mixture and to suggest a method for designing the base courses. The structural responses of loads and thermal effects have been considered for the design. The investigation has been broadly divided into six parts. The FIRST PART deals with the determination of the engineer ing properties of the CBSA mixtures to obtain strength parameters & Cement Bound Soil Aggregate Mixture will be referred to as CBSA mixture for brevity throughout this study. -iifor their use in subsequent studies on pavements. To start with, an optimum proportion of well graded coarse aggregate and soil was established. Degradation tests were conducted on coarse aggregate and the soil-aggregate mixture to study their effect on the gradings adopted. Effect of three main factors viz. , cement content, period of curing and temperature of curing were studied on the unconfined compressive strength, indirect tensile strength and flexural strength of CBSA mixtures. Equivalent cube strength tests were also carried out. Analysis of variance indicated that all the three factors and their interactions significantly affected the strength values. Rastrup's maturity combining the curing parameters has been mathematically correlated with the observed compressive strength of the mix. Multiple linear regression analysis has been utilized to express the compressive strength by a single equation involving independent variables as comont content and maturity. Relationships developed between various strength parameters show that one strength value of CBSA mix can be estimated from the other with a fair degree of accuracy. The pulse velocity measured across the cubes has been correlated with the compressive strength. Tests were carried out to evaluate the static modulus of elasticity in flexure and the dynamic modulus of elasticity in the longitudinal mode. The latter values exhibited a good correlation with compressive strength* flexural strength and the static / -iiimodulus of elasticity. Durability characteristics of CBSA mixes were also studied. The mix exhibited adequate resistance to the weathering cycles. Considering the flexural strength and the unit cost of the mix, an optimum cement content of 5 per cent was established, which was utilized for further investigation. Other tests on CBSA specimens included the evaluation of Poisson's ratios thermal expansion coefficients drying shrinkage, moisture movement and autogenous healing. The various properties determined above clearly indicate that hardened CBSA mixtures possess rigid characteristics. The SECOND PART of the study deals with the prediction of the fatigue life of the CBSA mix by the fracture mechanics approach, since this approach is capable of giving a quanti tative description of the fatigue process. For this, notched beams of varying notch to depth ratios were tested for static and repeated load cycles to establish the crack propagation law, which has been used to compute the fatigue life of the paving material. The accuracy of the established law was checked by testing additional beams. A generalized mathematical relationship has been suggested for evaluating fatigue life for different values of bending moments and notch to depth ratios. In the THIRD PART, an attempt was made to study the Ay 1/ -IVstructural behaviour of CBSA mixtures by application of the beam action hypothesis. A theoretical study exhibited an excellent correlation between the maximum stresses and deflect ions in beams and the corresponding values In slabs under edge loading condition. Large beams of varying depths were cast on subgrade and tested upto failure under the action of load. Comparison of observed values of maximum deflections and the subgrade reactive pressures with their corresponding values showed good agreement. Statistical analysis for goodness of fit by Chi-square test between the experimental and the corresponding theoretical values of maximum stresses revealed that the curves fit within 90 per cent degree of confidence. The FOURTH PART of the study has been devoted to the analytical formulation for the structural analysis of the CBSA base courses. Westergaard's conventional procedure and the finite element method for computation of theoretical values of stresses and deflections have been discussed. The finite element programme of solving plates on elastic foundation uses four noded fully compatible plate bending element with sixteen degrees of freedom, and is capable of solving the pavement system for any applied load on base as well as on composite pavement with base and overlay having different properties. A simplified design approach based on Westergaard1s equations has been discussed for the design of CBSA pavement with cement concrete overlay. A rational design procedure requires the evaluation of warping stresses induced in the base -Vcourse slab due to environmental temperature variations along with load stresses. Theoretical computations of the combined stresses for a wide range of base course thicknesses and panel geometry established the edge region to be the most critical. A simplified approach for computation of warping stresses in CBSA composite pavement having cement concrete overlay has been discussed. The FIFTH part deals with experimental study on the load response characteristics of the CBSA model base course slabs of varying thicknesses. Tests were carried out for stress-deformation and pressure transmission response under 'no warping' conditions. The loads were applied through rigid plates of different sizes at the three critical locations. Tests were conducted, first within the elastic zone and then in the plastic zone up to the ultimate failure. Since CBSA base shows poor resistance to abrasion, it should be protected by an overlay. A cement concrete overlay was employed in one CBSA base course to form a composite system and testing was carried out. In base course slabs as well as in composite pavement, the experimental values of stresses and deflotions within the elastic limits have been analysed on the basis of the theoretical work discussed in the preceding part. A statistical analysis of goodness of fit by Chi-square test between the experimental and theoretical values of maximum stresses -virevealed similar trends as in the case of large beams. For the design of CBSA base courses 3 design equations have been developed based on the observed values of maximum stresses. The observed ultimate loads were compared with the corres ponding theoretical values. The crack patterns observed over CBSA base slabs under ultimate loads were similar to those generally found in plain cement concrete pavements indicat ing resemblance in the load distributing phenomena of the two types of pavements. A field study on CBSA base course is also included in this part. A CBSA track was laid in the field with a view to investigating the stress-deformation characteristics of CBSA base courses under prototype conditions. The results of the static load tests on the test track slab under 'warped down' position indicated that the design equations developed under controlled conditions were applicable to the prevalent conditions in the field. The observed values were somewhat on the conservative side. Study of temperature gradients developed across the base slab thickness was carried out in order to compute the warping stresses. The trend of temperat ure variation in this case is observed to be similar to that of a plain cement concrete pavement. The SIXTH PART of this study has been devoted to the applications of results of the present investigation for the design of CBSA pavements. In order to reduce the computation time required in the use of the developed design equation, a -Vllhomogram for the wheel load stresses has been developed. This part of the study also includes economic analysis. Comparisons of the relative costs of CBSA pavements with bituminous concrete overlay and cement concrete overlay vis-a-vis the conventional cement concrete pavement and the flexible pavement have been made. Thus justification in selecting the CBSA pavement for highway construction jobs has been examined. The present study indicates that the CBSA mixture is undoubtedly a suitable material for base course construc tion with respect to stability, durability and economy. It is therefore concluded that CBSA mixes can be employed with confidence for constructing large mileage of roads to cater for medium to heavy traffic under the prevalent Indian conditions.