Auger penetrometer development for the compaction control of sand

Summary

The Auger Penetrometer (AP) is a continuous flight auger which penetrates the soil by applying a constant static vertical load during rotary advancement. Auger resistance index (AI), is measured as the number of turns of the auger required for 100 mm depth of penetration. This paper describes an AP with a hinged handle and presents the results of experiments made to study the relation between AI and density of sand(rd), and, possible effects of operator technique and particle size on this relation. It is concluded that: (i) linear relations with strong correlations exist between log(AI) and rd of sands tested, (ii) the AP is capable of producing reproducible results for compaction control of sands, and (iii) different calibration line should be established for sands with considerably different particle size distribution.

Introduction

The use of direct or indirect density measurements for compaction control of sand fills is commonly adopted all over the world. Numerous methods have been developed for this purpose, including the sand replacement, rubber balloon and nuclear density meter methods. These methods disturb the soil surface and require additional laboratory testing or recalibration, but are generally considered to give reasonable to fairly accurate results.

The Auger Penetrometer (AP) was developed by Mohajerani (1982, 1985) at the University of Western Australia as an alternative approach for the determination of density of sand fills. It consists of a 22 mm diameter continuous-flight auger which cuts into the soil by applying a constant static vertical load (total weight of auger, shaft and handle = 43 Newtons) during rotary advancement. Auger resistance index (AI) is defined as the number of revolutions required for a depth of 100 mm of penetration,

i.e. AI = 100 NT/L

where NT is the number of turns of the auger and L (mm) is the measured depth of penetration (about 100 150 mm).

Strong linear correlations have been found between logarithm of AI and dry density of compacted sands (Mohajerani 1982, 1985; Erceg 1982), (Fig.1). Erceg carried out parallel testing with the AP, sand replacement method and dynamic Perth Sand Penetrometer (Glick and Clegg 1965) on a compacted sand fill and found that the results from the AP method compared very favourably with the results obtained using the sand replacement method (Fig. 2). Chivers (1982) used the AP to measure the dry density of a subgrade sand and found a strong linear relationship between the modulus of subgrade reaction and the logarithm of AI (Mohajerani 1985/2).

The APs used in the previous studies were equipped with fixed handle, making the penetration results dependent on the individual operators. This could explain the variation shown in Fig. 1, between the results of the two previous independent studies. It has been suggested (Mohajerani, 1985) that a more appropriate handle should be used with the AP to make it independent of operator technique.

Preparation for testing involved mixing a batch of sand of weighing approximately 15 kg with water to various predetermined moisture contents. Sand samples with moisture contents ranging from 1% to 12% were used to study the influence of changing moisture content on AI-density relationship. After mixing the sand thoroughly, a test specimen was prepared in a C.B.R. mould (height 177 mm, diameter = 152 mm). It was possible to vary the density by changing the number of layers in the mould and by varying the number of blows of light or heavy standard compaction hammer per layer. The specimen was placed under the auger penetrometer and the auger lowered until the tip reached the sand surface. The shaft was released to settle under its own weight. When the auger had come to rest an initial-settlement reading was taken. The auger was then rotated slowly (at different rates of approximately 0.5 to 1.5 revolution per second) in a clockwise direction and the number of revolutions were counted until 140 mm penetration from the auger tip (127.5 mm from the cutting edge) was reached. It should be noted that, there is relatively very little disturbance around the penetrometer, with apparently no influence of the mould-edge on the penetration resistance. This is because, AP is a quasi-static method and as it penetrates, it transfers the soil through the flights.

Experiments were made by four operators: A,B,C and D, using sand A (Fig.4), and their results were compared. Operators B, C and D had not used the AP before. They were instructed to place the auger tip on the sand surface and release it slowly and when the auger has come to rest to rotate the auger slowly in a clockwise direction and to count the number of revolutions it takes to penetrate 140 mm.

Operators A,B,C and D turned the auger at different rates (rotary velocities) of approximately 1, 1.5, 1 and 0.5 revolutions per second, respectively.

In order to examine the influence of auger condition on AI three augers were used; an auger in good condition, a rusty auger and a damaged auger. The rusty auger was prepared by etching off the lacquered surface. The auger was then left out in the weather for some months. After testing, the auger tip was then damaged by continual dropping of the auger onto a concrete floor, and the cutting edge and flight were damaged by repeated blows with a hammer.

Augers used in this study were steel augers (suitable for hardwood) with a diameter of 22.225 mm, lead of helix 50.8 mm and length of helix 300 mm (Fig. 3).

In order to study the influence of soil particle size on AI, three fine to coarse-grained, subrounded to subangular, quartzite sands were used. The grading curves are shown in Fig. 4).