Regional Seminar Papers 1997

Tools and Equipment — Design, Development and Specifications

Lessons from the Compaction of Some Local Soils for Feeder Road Construction in Ghana

A comparative technical quality study of roads built with labour-based and conventional equipment-intensive technologies in Ghana, showed that on the average the level of compaction achieved on labour-based projects were 10% lower than those achieved on equipment-intensive projects. Using data from this and other studies, this paper discusses some of the factors which might account for the lower levels of compaction achieved on labour-based roads.

Introduction

In Ghana, the most common types of soils used as subgrade and base material for highway pavement construction and as the wearing course on unsurfaced roads, are the soils formed from the decomposition of rocks known as granites and phyllites. In almost all the situations in which these soils are used in road construction, it becomes necessary to compact the material to very high levels of compaction. The normal practice is to attempt to attain the Modified AASHTO level of compaction. The specification adopted by the Ministry of Road and Transport in Ghana requires that subgrades and bases be compacted to a minimum of 95% and 98% respectively of the maximum dry density (MDD) obtained with the Modified AASHTO standard. The rationale behind this is that at the MDD, strength is known to be high and compressibility low. However, this general observation has not been found to be true for tropical soils, some of which are known to be very unstable.

Over the past few years, the ILO has assisted the Civil Engineering Department of UST in building its capacity in the area of labour-based road engineering. Part of the programme includes comparative studies into the technical quality of roads being built with conventional capital-intensive technology and those being built with labour-based technology. This paper looks at the issues that the studies have raised concerning some of the technical factors that affect the compaction of these local soils

Brief Description of Methodology

A number of feeder roads in the country constructed using conventional equipment-intensive method and those using labour-based methods were selected for the technical quality studies. On each road, up to three newly gravelled sections were selected and on each section, the level of compaction achieved was determined using the sand replacement method. Other tests included the determination of the thickness of the gravel layer and the measurement of the camber of the road section. In addition, samples of material were recovered from the gravel pits and subjected to routine laboratory identification and classification tests. The laboratory compaction characteristics of each sample were also determined. In this paper, only the results of the compaction measurements are presented and discussed briefly. The details of the study and the results of other measurements are reported elsewhere.

Discussion of Compaction Results

The level of compaction achieved

The levels of compaction achieved on eight capital-intensive and on six labour-based projects are shown in Table 1. It can be seen that even though neither the equipment-intensive nor labour-based projects achieved the minimum DFR specification of 98%, the field densities achieved on the labour-based roads are on the average 10% lower than those achieved on the equipment-intensive roads.

Two issues that arise from these results are :

1. the appropriateness of the compaction plant being used for labour-based compaction in Ghana, and
2. the effectiveness of moisture conditioning during compaction, which is related to the method of compaction being used in the field.

The appropriateness of labour-based compaction plant

The method of compaction in the field is usually described by the mechanical plant used. Compaction may be achieved by static pressure, by impact loading, by vibration, or by a combination of these. Table 2 is a summary of the types of plant generally available for compaction.

Generally, for compaction of feeder roads, vibratory rollers are used. Vibratory rollers send pressure waves through the soil at a wide range of frequencies depending on the size of the plant. In normal soils, compaction is achieved by a combination of vibration and pressure. Thus, in addition to the dynamic pressure, the static weight of the roller is also an important factor in its performance. Vibratory compactors range from a 135 kg vibrating plate to 13,000 kg towed vibratory roller and may be used for compacting a wide range of material from sands and gravels to boulders.

The standard compaction plant which labour-based contractors use in Ghana is the "Bomag 65S" pedestrian vibratory roller with an operating weight of 600 kg. This compaction plant gives a static linear load of 4.6 kg/cm and operates at a frequency of 58 Hz. On equipment-intensive projects, a range of compaction plant including 10,000 kg (10-tonne) rollers was observed on site. For a 10 tonne SAKAI SV100 vibratory compaction plant, the equivalent static linear load is of the order of 24 kg/cm. It can be seen that the capacity of labour-based compaction plant may be only a fifth in terms of static linear load or only 6% in terms of the gross static weight of the compaction plant being used on equipment-intensive projects. The adequacy of such low capacity labour-based compaction plant for compacting local soils, needs to be further evaluated.

Effect of Compactive Effort on Level of Compaction

In order to throw more light on the effect of compactive effort on the compaction characteristics of local soils, the laboratory compaction characteristics of the gravel material obtained using 10 blows, 25 blows and 55 blows of the Modified AASHTO specification were determined. In addition to these tests, the laboratory compaction characteristics using the Standard Proctor specification was also determined. The results of two such tests are shown in Figs 1(a) and 1(b). The index properties and the AASHTO classification of the material are also shown in the figures. In Fig. 1(b), only the curves for the modified AASHTO at 55 blows and the Proctor Standard are available. The figures show the well known results of increasing MDD (with a corresponding decrease in the OMC) as the number of blows increases. It may be noted, however, that the increase in MDD from the Proctor Standard to the Modified AASHTO (55 blows) standard for test LB-BA-TT is larger than that for LB-WR-DL1.

In order to have a common basis for comparing the compactive effort from different standards, the concept of specific energy was applied. The specific energy is the energy applied to the soil per unit volume during the compaction process. The values of the levels of compaction using the MDD value at a specific energy of 1,727kNm/m³ as a base are shown plotted against the input specific energy in Figure 3. The figure clearly shows that the effect of specific energy on the level of compaction depends on the type of soil. Whereas the material classified as A-2-6(2) in the AASHTO system gave a 96% level of compaction for a specific energy of 542 kNm/m³, the same specific energy gave a level of compaction of only 80% for a A-7-5(5) material

Where N= number of blows on each layer of

soil in the mould; n= the number of layers required to fill the mould, W= the weight of the tamper; h= the height from which the rammer falls and = the total volume of the mould.

In the field, when rollers are used the specific energy depends on the pressure and the contact area between the roller and the soil, the thickness of the layers being compacted and the number of passes. This laboratory result suggests that in general for good quality road base material, relatively high levels of compaction can be achieved with relatively lower levels of energy.

Moisture Conditioning During Compaction

The MDD and the corresponding OMC values in Table 2 are shown plotted in Figure 3. It may be seen that the OMC values for the material used on the road varied between 6% and 14%. However, the bulk of the data fell between 9 and 11%. This suggests that in order to achieve high levels of compaction, the water content of the material must be controlled within 9-11%. Studies done in Ghana (GHA, 1970) showed that whereas most of the OMC values fell between the relatively narrow range of 6-12% confirming the results of our studies, the in situ moisture content of lateritic gravel in the moist sub humid climatic zone ranged from as low as 2% to 28%. This suggests that it is not likely that the in situ water content would fall within the desired 2% of the OMC. From considerations of strength reduction on soaking, it has been suggested elsewhere (Gidigasu, 1991) that the concept of OMC and MDD be abandoned as a necessary placement condition in favour of placing at least sub-grade layer at a stable in situ moisture content.

In practice, however, when the in situ moisture content of the gravel material is low, it may be necessary to add water to the material before compacting. On the other hand, where the material is wetter than the optimum, it may be necessary to reduce the moisture content by aeration. However, it appears the direction of water content change has some effect on the compaction characteristics. Figure 3 is a summary of the results of a study to investigate the effect of direction of water content change on the compaction characteristics of residual soils derived from granites and phyllite (Ampadu, 1997). The results suggest that the effect of the direction of water content change on the compaction characteristics depends on the type of material. For materials derived from granites, the effect on the MDD is negligible, but for those derived from phyllites, the change can be as high as 4% of MDD.

Effect of Compaction on Particle Breakage

Figures 5(a) and 5(b) show the effect of laboratory compaction on the grading curves for granites and phyllites respectively. The figures are for re-used samples but a similar trend was observed even when fresh samples were used. The figures show clearly that compaction leads to particle breakage for both granites and phyllites, but the effect is much larger on phyllites than on granites.

The issue of whether higher or lower densities would be obtained in a particular compaction method depends on whether the breakdown of particles during compaction leads to an improvement or a deterioration of the grading. These results suggest that high compactive pressure may lead to particle breakage, which can affect not only the level of compaction, but also other properties of the pavement as well.

Field Compaction Control

Finally, the results of a field compaction control for the subgrade in decomposed phyllites for a construction project are shown in Fig. 6. The compaction plant was a 10-ton HAMM smooth drum vibratory roller and several passes were applied to the imported material. The laboratory compaction curves obtained by the Modified AASHTO for two samples of the material are superimposed. This figure serves to illustrate some of the problems encountered in the compaction of local soils.

1. With the possible exception of one point, all the field dry density points fell below the laboratory compaction curve even for water contents close to the optimum (for Sample No. 2).
2. Except for one point, the placing water content was higher than the optimum moisture content for the particular material illustrating the importance of moisture conditioning.
3. Despite the high capacity compaction plant utilised (10 ton vibratory roller) the levels of compaction achieved ranged from a low 86% to 95% of modified AASHTO
4. The compaction characteristics of the material from the gravel pit changed with depth as evidenced by the different compaction curves for Samples No. 1 and No. 2. Such changes have been observed over profiles of phyllites.

It is suggested that the lower levels of compaction achieved on labour-based roads in Ghana may be attributed partly to the rather low static weight of compaction plant being used. On the other hand, the availability of high capacity compaction plant does not necessarily guarantee a high level of compaction.

Conclusion

From the results of our studies into the levels of compaction achieved on equipment-intensive and on labour-based projects on local soils, the following preliminary conclusions may be drawn:

1. The 600-kg BOMAG 65S currently being used on labour-based projects appears too light to achieve the required levels of compaction under field conditions. There is a need for further study into the optimum capacity of compaction plant for compacting gravel on labour-based roads.
2. The use of high capacity plant by itself does not guarantee a high level of compaction. Moisture conditioning during compaction of local soils is an important factor.
3. There is the possibility of particle breakage arising from high pressures and this can lead to difficulties with achieving high levels of compaction. This should be taken into account in selection of plant for compaction.

References

Ampadu, S. K. 1997. The compaction characteristics of residual soils. Proceedings of Fourth International Conference on Structural Engineering Analysis and Modelling, Accra, Sept 9-11, 1997

Ampadu, S. K. 1996. Effect of soaking on strength and deformation of a local soil. Proceedings of Annual General Meeting of Ghana Institution of Engineers, 1996

Gidigasu, M. D. 1991 Characterization and use of tropical gravels for pavement construction in West Africa

Gidigasu, M. D. and Mate-Korley E. N. 1980. Highway geotechnical characterisation of residual micaceous soils over granite. 7th Regional Conference for Africa on Soil Mechanics and Foundation Engineering, Accra, 1980.

Ghana Highway Authority (Public Works Department) 1970. Detailed engineering and design materials report, Contract 2. Report prepared by Ingeroute Engineers (France), Ghana Highway Authority, Accra.