 |
 |
Skin Friction between Peat and Silt Soils with Construction Materials
Doctor Candidate in Institut für Geotechnik - TU Bergakademie Freiberg From Engineering Faculty, Tanjungpura University, West of Kalimantan, Indonesia and TU Bergakademie Freiberg, Institut für Geotechnik, Gustav-Zeuner-Straße 1, 09596 Freiberg, Germany |
ABSTRACT
Until recently the values of skin friction used for design purposes were the average values obtained by field tests, with only qualitative reference to such factor influencing their magnitude as type of soil, type of construction material, and surface finish, moisture content of the soil, etc.
The modern trend is to establish skin friction coefficients through laboratory experiments in which the factors influencing the results may be controlled quantitatively.
Seventy-two experiments were carried out by the Author to determine the magnitude of skin friction, in which the following variables were considered:
(1) Various construction materials: steel, concrete, wood.
(2) For each material two surface conditions were used: smooth and rough which are described in such a way that they may be reproduced by anyone with a reasonable degree of accuracy.
(3) Variation of the normal load between the friction surfaces.
The test results show that for cohesive soils both cohesion and internal friction should be considered in evaluation of skin friction. The results include ratios of adhesion to cohesion, and of angle of skin friction to the angle of internal friction for definite types of soil, moisture content, various construction materials and their surface finishes; so that for practical application it is necessary to test the soil in shear and to make a sieve analysis. From the shear test data skin friction can then be evaluated by used of the given coefficients.
Keywords: Peat and silty soils, soil-structure interaction, skin friction, direct shear tests
INTRODUCTION
The classical laws of friction do not apply to footwear sliding on artificial and natural surface (Valiant, 1993; de Lange and Winkelmolan, 1995). They only apply to two dry solid metallic surfaces sliding over each other. This has been known for more than fifty years, since early work on rubber tires tested on concrete and varnished wood (Derieux, 1934), yet their application continues to be reported in the scientific and promotional literature.
Friction has been the subject of intensive study since the early investigation of da Vinci, Amontos, Coulomb and Euler (Dawson, 1979). At the molecular level, even smooth, solid surfaces have valleys and ridges or asperities, and at a given instant some of these asperities will be touching. How these asperities respond to each other when sliding depends on their respective material deformation properties. For inelastic materials like polymers, visco-elastic, visco-plastic, and relaxation effects, lead to a time-dependence of the contact area and hysteresis losses associated with the loading-unloading cycles (Czichos, 1986). Analytical research into friction attributed this to a complex molecular-mechanical interaction between the contacting surface. This complex interaction was though to be due to a multitude of factors, including, the combined effects of asperity deformation, plowing by hard surface asperities and wear particles, and adhesion between flat surface. Despite considerable experimental and analytical research, no “simple” theoretical model has been developed to calculate the friction between two given surface (Suh and Sin, 1981; Czichos, 1986). Outsoles have various pattern or cleat configurations that interact with either artificial turf or natural turf made of particles of soil and grass. While the mechanism of traction would be different, the mechanism used to explain dry friction could provide the basis to explain the mechanisms associated with field footwear-surface interaction.
Soils have a particular structure consisting of discrete particles that are not strongly bonded together and are relatively free to move with respect to each other. Natural surfaces are usually subjected to rain so the pore space between the soil particles can be partly filled with water and what space remained would be filled with air depending upon the degree of saturation of the pore space. When a load is applied to the soil surface through the outsole and cleats, the soil resists the applied loads by developing contact forces wherever they touch at their asperities. There are a large number of contacts within a soil mass – about five million contacts within one cubic centimeter of fine sand, for example. At each contact, the particle respond by deforming in three ways: compressing, bending, and sliding. Deformation due to sliding is usually the most significant, and is non-linear and irreversible, making the load-deformation behavior of soil non-linear and irreversible as well (Lambe and Whitman, 1979). Because sliding between particles predominates, the mechanisms used to explain dry friction can been applied to soils.
The external forces that cause sliding within soils are resisted by friction and bonding forces between the particles. If the applied forces become sufficiently large, failure of the soil mass may occur when the contact resistance (friction and bonding) reaches its limit and the soil mass as whole slides. The plane connecting all the particles where failure has occurred is known as the failure or slip plane, and unlike solid surface, where it corresponds to the plane between the surfaces, the failure plane in soils is not predefined.
Recent developments in civil engineering, especially in soil mechanics and foundations of structures, permit the designing engineer to take a great step forward from “design by experience” to design by a well-established theory verified by experiments. In the case of the question: What is the stress-strain relation if one starts to move in relation to the other? This mutual effect of soils and structures in the transmission of forces from the one to the other through the contact surface is called skin friction.
Until recently the values of skin friction were obtained from filed observations, or they were calculated from the resistance of pile driving, sinking of caisson, etc. These previous values were values averaged along the pile of caisson, and it was impossible to relate them to the behavior of soil layers. Furthermore, no relationship was given to the surface finish of different construction materials.
Many geotechnics problems involve estimation of stresses transferred along the interface between soils and solid surfaces (structures). While considerable work (Paikowsky et al. has listed several works of significance) has been done on the interfacial friction between cohesion-less soils (sands) and solid surfaces, The interfacial shear resistance between fine grained soils and solid surface depends on whether its mobilization takes place in the drained or in the un-drained condition. Accordingly, there are basically two approaches for the estimation of interface resistance. One is the total stress or un-drained strength approach in which the interface resistance is related to the un-drained shear strength by an empirical adhesion factor, a.
In the second approach, known as the effective stress approach, the interface resistance is related to the effective normal stress (s¢n) acting on the interface and the effective angle friction (f). In the case of piles, the shear deformation occurs within a relatively thin zone around the pile shaft, and drainage from this narrow zone take place rapidly during loading. Thus most pile loading situations tends towards the drained condition. The interface resistance (f) in this approach is expressed as
 |
(1) |
Several kinds of apparatus have been used to investigate the interfacial friction between fine-grained soil and solid surfaces, for example the direct shear apparatus and the simple shear apparatus. Model pile tests have also been used for this purpose. The results of the studies reported in the literature show that (’ depends on the roughness of the solid surface. It increases with the surface roughness, and when the roughness reaches a critical value it becomes constant and is equal to the angle of internal friction of the soil. As most of these studies are on normally consolidated soils, the influence of over-consolidation ratio on (’ values has received little attention.
Herein is examined the effective angle of interfacial friction obtaining between cohesive soils and solid surface as influenced by the roughness of the material surface, soil type.
The purpose of the investigation carried out by Author was to determine the values of skin friction between peat and silty soils and construction materials. The subject of this article is to make an analysis of these values and to give a relationship between skin friction and the strength of peat and silty soil and also between skin friction and the surface characteristics of various construction materials.
APPLICATION OF GENERAL FRICTION THEORY
The basic idea of friction seems simple, but it has constituted a problem for many decades. Every filed of engineering is concerned with friction, but so far in connection with soils and solids no theory has been available. The theory of skin friction between solid materials has had some development, and some of the conclusions may be applied to the friction between soils and solids.
For solid materials it was found that the magnitude of friction always depended on whether the surface was dry or moist, or completely lubricated by some liquid. This led to the division of friction into such groups as dry and liquid friction. Between them is semi-fluid or composite friction. The magnitude of friction is a very great extent dependent upon cleanliness, atmospheric dust and humidity, oxide and other films, surface finish, velocity of sliding, contact pressure, temperature, grain size, direction of grain, vibration and static loads, etc. One can see that the problem of friction between solids is more complicated than it may seem at first. The analysis and investigation of skin friction of soils, which are much less homogeneous materials, is obviously still more complicated. Bowden and Tabor (1950) showed that in the case of smooth steel plates the contact surface varied from 1/100,000 of the gross area at a low normal load to 1/400 at a high normal load. Merriman (1930) quotes a similar postulate for wood specimens; when the normal pressure reaches the allowable stress for wood the fibers bite into each other and the coefficient of friction increases. In the case of a lubricated solid surface, the liquid could only partly produce a lubricant film between the two surfaces because the two solid materials are in contact over a percentage of the gross area. Therefore, the friction force lies between the values that it would have for a solid-solid boundary and a solid-liquid boundary.
Soils are, according to their composition, in a state between solid and liquid materials. Therefore, the mechanics of friction will be partly like those of a solid and partly like those of a liquid. The skin resistance could not be higher than the ultimate shearing strength of soil, and it is important to find out the ratio between soil strength and skin resistance. Since the development of skin friction due to displacement, the stress-strain curves can be obtained only from the experimental results. For granular soils, the experimental stress-strain relation was expressed in a mathematical term by Kézdi (1959) and applied to determine the earth pressure and pile resistance. In the case of skin friction, the stress-strain curve was expressed as an exponential function of displacement:
 |
(2) |
where t= shearing stress which produces a displacement of “s”
s = normal stress
d = angle of skin friction
s = displacement
s0 = maximum displacement due to failure
k= constant for the soil.
DESCRIPTION OF TESTS
To determine the strength of peat soil, and skin friction between peat soil and construction materials, strain controlled shear box was used throughout this investigation.
For the determination of physical properties of soil well-known standard equipment was used.
The stress controlled shear box had a shearing area 40 cm2, and it was drained on both sides. For the measuring of horizontal movement a dial gauge was fixed to the base plate. The specimens of construction material were placed in the lower portion of the box, and the soil was placed in the upper half.
Selection of Construction Material and Soils
In the choice of construction materials for the tests, consideration was given to the application of the results in the field of civil engineering. In the present stage of engineering, important construction materials are steel, concrete, and wood, and thus it was logical to select them for the investigation of skin friction.
Steel; The quality of the steel was that of common (U-36; spl = 3600 kN/m2), commercial mild steel, which is widely used for piles, sheet piles, etc. For practical purposes, two kinds of surface finish were used in this investigation. The completely smooth surface polished by finer sandpaper which have value of crudity of surface N =3 (0.1 mm; N £ 4) represents one extreme case. The other case was produced by artificially rusting the specimen which have value of rough of surface N = 9 (6.3 mm; N ³ 8) and afterwards removing the loose rusted.
Wood; It was much harder to choose the type and quality of the wood specimens. In engineering practice – especially in West of Kalimantan, Belian –local name or Iron-wood is possibly the most commonly used for sub-structure construction, because it was class I. Sound Belian (iron wood) was used, and it was cleaned from any unnatural surface irregularities and defects. Each test piece was shaped by planning to minimize the effect of roughness attributed by other than the natural texture of the wood. The finish thus obtained was similar to the surface of the plywood sheets used in shuttering. Unfortunately, no test or references are available in connection with the effect of hardness of wood on the contact with differences soils. It may be supposed that granular materials produced a certain indentation into the surface of the wood, and this may be increased with the intensity of the normal load. The tests were carried out in two directions to the grain of the wood, parallel and at right angle.
Concrete; concrete type used at this research is quality of K-225 (sv = 225 kN/m2) which is commonly used for construction. For concrete specimens two different grain sizes of aggregate were used. For fine surface the maximum grain size was 4.75 mm and for a rough surface 9.52 mm. The first concrete was poured into a plywood form, and the second one flat rough ground. The first specimen represents the smooth concrete surface made in planed wood form, and the second specimen represents the rough concrete surface poured against the side of an excavation.
Two soils –peat soil, and silty soil– were used for the study. After the selection of the basic types of soils, the next step was to determine their general physical properties. This was an important procedure, because it was necessary to know these values for the purposes of later discussion and comparison with other results.
For silt soil, based on data in Table 1 and Unified Soil Classification System, when the plastic index and the liquid limit plot in the hatched portion of the plasticity chart, the soil given double symbol ML–OL.
According to AASHTO system, a soil is term fine-grained if more than 35% passes No. 200 (0.075 mm) sieve, 82.5 > 36; Liquid limit = 37.4 < 40; and also make an Plasticity index = 7.2 % > 10 %; is entering group classification of A-4 with most dominant material type is soil have silt. Become its conclusion [of] clay soil used at this research enter organic clay group with low plasticity (OL).
For peat soil, it can be described as follows:
The physical properties of these soils are presented in Table 1. To achieve a large variation in the value of roughness, six solid material round plates of size ( 71,5 x 6 mm consisting of mild steel smooth and rough (material 1 and 2) and Belian wood in the grain of the wood and parallel to right angle direction (material 3 and 4) and concrete smooth and rough surface (material 5 and 6) were adopted as materials representing the solid surface. The methods adopted to achieve various grades of roughness are detailed elsewhere. The average roughness or qualities of construction material roughness are given in Table 2.
Table 1. Properties of the soils used
| |||
| No. | Property | Soil | |
| Peat | Silty | ||
| |||
| 1 | Bulk unit weight, (g) gr/m3 | 0.965 | 1.572 |
| 2 | Dry unit weight, (gd) gr/m3 | 0.083 | 0.933 |
| 3 | Water content, (w) % | 1169 | 68.4 |
| 4 | Specific gravity, (Gs). | 1.27 | 2.41 |
| 5 | Void ratio, (e) | 15.8 | - |
| 6 | Dust content | 4.2 | - |
| 7 | Organic Carbon content, % | 95.8 | - |
| 8 | pH H2O | 3.6 | - |
| 9 | Average fibre content, % | 25.0 | - |
| - Coarse fibre | 13.1 | - | |
| - Medium fibre | 29.4 | - | |
| - Small fibre | 32.5 | - | |
| 10 | Consistency limit | ||
| - Liquid limit, (wL), % | - | 37.4 | |
| - Plastic limit, (wP), % | - | 30.2 | |
| - Plasticity index, (Ip), % | - | 7.2 | |
| 11 | Gradation Analysis | - | |
| - Sand, % | - | 17.5 | |
| - Silt and Clay, % | - | 82.5 | |
| |||
Table 2. Properties of materials used
| ||
| Material No. | Type of material | Surface condition |
| ||
| 1 | Mild steel smooth surface | N =3 (0.1 (m) |
| 2 | Mild steel rough surface | N = 9 (6.3 (m) |
| 3 | Wood in grain of wood | “smooth” |
| 4 | Wood in parallel to right angle | “rough” |
| 5 | Concrete smooth surface | “Fly-wood surface” |
| 6 | Concrete rough surface | “Excavation surface” |
| ||
Shear tests on soils were conducted in direct shear mode as per standard procedures in the conventional direct shear box (specimen size 40 cm2). The direct machine was strain controlled. For interface friction tests, the conventional direct shear box (40 cm2) was modified to conduct the interface friction tests. The lower half of the conventional direct shear apparatus was replaced by an Æ 71.5 x 22 mm mild steel plate. The test material of 6 mm thickness was mounted over this plate so that the total thickness was equal to the thickness of the bottom half of the box (28 mm), as can be seen in Figure 1. The area of the interface was 40 cm2 and remains unchanged throughout the test.
Figure 1. Schematic diagram of the apparatus used for the study (dimensions in mm)
The soils were undisturbed condition and prepared in tube sample having a diameter of 100 mm and a height of 400 mm. The sample thus obtained was cut by a wire, trimmed and transferred to the 40 cm2 direct shear box. The sample was further consolidated to the required pressure directly over the test material using monotonically increasing load increments. A sample thickness of 25 mm was adopted for the soil-soil test and 15 ± 1 mm was used for the soil-material interface test. The sample thickness adopted satisfied the recommendations of Jardine and Chow. Pairs of rubber strips coated with silicon grease were used to reduce the friction between the edge of the box and the solid surface (Figure 1).
Direct shear tests were conducted in normally consolidated states under drained conditions. The normal stress (s¢n) values adopted are 12.5 kN/m2, 25 kN/m2, and 50 kN/m2.
In the investigation of skin friction, the Author found that similar redistribution occurred, but the various construction materials had different effects on the moisture content of the contact surface. In order to avoid absorption of moisture by building materials, the specimens made of wood and concrete were saturated in water 48 hours, and just before the testing were surface dried. The average moisture content of wood was 41% and 3% of concrete.
(a) Between the soil and steel specimen the moisture was higher than the average the soil sample. In the case of rough steel this increment was 2-3% and in the case of smooth steel 4.5-5.5%.
(b) In the case of wood, in testing parallel to the grain, the increase was 0-1.5%, while testing at right angles to the grain there was no increment at all.
(c) In the case of concrete the moisture content on the contact surface decreased, on smooth concrete by 7-10%, and 5-8% on rough concrete.
For determination of moisture content on the contact surface a thin slice was taken from the soil, the thickness of which was about 2 mm.
ANALYSIS OF EXPERIMENTAL RESULTS
The investigation results are presented in Table 3 for peat soil and Table 4 for silt soil.
Table 3. Values of shear strength and skin friction for peat
Material
| ||||||
| Material | s'v (kN/m2) |
c or ca (kN/m2) |
 |
f or d |  |
 |
| ||||||
| Peat | 12.5 | 23.1 | 1 | 15°55’48 | 1 | 1 |
| 25 | 23.1 | 1 | 16°00’36 | 1 | 1 | |
| 50 | 23.6 | 1 | 16°25’12 | 1 | 1 | |
| Peat - peat | 12.5 | 18.4 | 0.797 | 10°54’00 | 0.684 | 0.675 |
| 25 | 17.2 | 0.745 | 11°33’36 | 0.722 | 0.713 | |
| 50 | 16.5 | 0.699 | 11°09’00 | 0.679 | 0.669 | |
| Smooth steel | 12.5 | 15.2 | 0.658 | 6°02’42 | 0.379 | 0.371 |
| 25 | 15.4 | 0.667 | 6°31’48 | 0.408 | 0.399 | |
| 50 | 16.1 | 0.682 | 6°10’48 | 0.376 | 0.367 | |
| Rough steel | 12.5 | 17.3 | 0.749 | 13°43’12 | 0.861 | 0.855 |
| 25 | 15.8 | 0.684 | 14°24’00 | 0.899 | 0.895 | |
| 50 | 14.9 | 0.631 | 14(24’07 | 0.877 | 0.871 | |
| Wood parallel to grain | 12.5 | 20.4 | 0.883 | 10°48’00 | 0.678 | 0.668 |
| 25 | 21.2 | 0.918 | 10°54’36 | 0.681 | 0.672 | |
| 50 | 18.6 | 0.788 | 10°32’24 | 0.642 | 0.631 | |
| Wood at right angles to grain | 12.5 | 17.3 | 0.749 | 12°38’24 | 0.793 | 0.786 |
| 50 | 21.2 | 0.918 | 12°04’12 | 0.754 | 0.745 | |
| 25 | 186 | 0.788 | 11°53’24 | 0.724 | 0.714 | |
| Smooth concrete | 12.5 | 22.4 | 0.970 | 11°18’36 | 0.710 | 0.701 |
| 25 | 23.0 | 0.996 | 11°49’48 | 0.739 | 0.730 | |
| 50 | 19.9 | 0.843 | 11°15’00 | 0.685 | 0.675 | |
| Rough concrete | 12.5 | 20.0 | 0.686 | 14°52’12 | 0.933 | 0.930 |
| 25 | 22.1 | 0.957 | 15°12’00 | 0.949 | 0.947 | |
| 50 | 20.5 | 0.869 | 15°19’48 | 0.934 | 0.930 | |
| ||||||
Table 4. Values of shear strength and skin friction for silt Material
| |||||
| Material | c or ca (kN/m2) |
|
f or d | |
|
| |||||
| Silt | 239 | 1 | 30°30’00 | 1 | 1 |
| 231 | 1 | 30°42’00 | 1 | 1 | |
| 246 | 1 | 29°18’00 | 1 | 1 | |
| Silt - Silt | 214 | 0.895 | 23°36’00 | 0.774 | 0.742 |
| 208 | 0.900 | 23°24’00 | 0.762 | 0.729 | |
| 158 | 0.642 | 24°54’00 | 0.850 | 0.827 | |
| Smooth steel | 152 | 0.636 | 22°48’00 | 0.748 | 0.714 |
| 154 | 0.667 | 21°54’00 | 0.713 | 0.677 | |
| 161 | 0.654 | 22°48’00 | 0.778 | 0.749 | |
| Rough steel | 173 | 0.724 | 30°00’00 | 0.984 | 0.980 |
| 158 | 0.684 | 28°54’00 | 0.941 | 0.930 | |
| 149 | 0.606 | 29°30’00 | 1.007 | 1.008 | |
| Wood parallel to grain | 204 | 0.854 | 22°06’00 | 0.725 | 0.689 |
| 212 | 0.918 | 23°00’00 | 0.749 | 0.715 | |
| 186 | 0.756 | 25°48’00 | 0.881 | 0.861 | |
| Wood at right angles to grain | 173 | 0.724 | 26°00’00 | 0.852 | 0.828 |
| 212 | 0.918 | 28°00’00 | 0.912 | 0.896 | |
| 186 | 0.756 | 25°18’00 | 0.863 | 0.842 | |
| Smooth concrete | 224 | 0.937 | 22°36’00 | 0.741 | 0.707 |
| 230 | 0.996 | 27°42’00 | 0.902 | 0.884 | |
| 199 | 0.809 | 26°30’00 | 0.904 | 0.888 | |
| Rough concrete | 200 | 0.837 | 29°54’00 | 0.980 | 0.976 |
| 221 | 0.957 | 30°00’00 | 0.977 | 0.972 | |
| 205 | 0.833 | 30°00’00 | 1.024 | 1.029 | |
| |||||
Skin Friction on Steel
(a) Peat. Table 3 shows the values of internal and skin friction of saturated peat. It can be seen that, when the normal load was increased from 12.5 kN/cm2 to 50 kN/cm2, the angle friction increased from 15° 55’ 48 to 16° 25’ 12. The relation between d¢ and f¢ shows the same value under different loads. For peat-peat interface/friction angle (d¢) to the angle of internal friction (f¢) are d¢ = 0.695f¢. It is interesting to note that the rough steel has a good skin resistance on peat and better than smooth steel. The values are d¢ = 0.879f¢ for rough steel, and d¢ = 0.388f¢ for smooth steel with peat.
(b) Silt. The skin friction between silt and steel was investigated with the silt in saturated conditions. The values of internal and skin friction of saturated silt. It can be seen that, when the normal load was increased from 12.5 kN/cm2 to 50 kN/cm2, the angle friction increased from 29° 18’ 00 to 30° 42’ 00. The relation between d¢ and f¢ shows the same value under different loads. For silt-silt interface/friction angle (d¢) to the angle of internal friction (f¢) are d¢ = 0.795f¢. In the case of saturated silt the stress controlled shear box was used with smooth and rough specimens. For saturated silt both of rough and smooth steel were applicable because it could be tested in the small strain controlled shearing box (0.30 mm/min). The test results are shown in Table 4. It is interesting to note that the rough steel has a good skin resistance on silt and better than smooth steel. The values are d¢ = 0.977f¢ for rough steel, and d¢ = 0.746f¢ for smooth steel with silt. In the case of rough steel and saturated silt the skin friction was close to the internal friction of the soil itself. The values of d¢ and f¢ increase somewhat with the increase of the normal load.
Skin Friction on Wood
Relative to the direction of shear force, two positions of grains of wood specimens were used; grain parallel and grain at right angles.
(a) Peat. Table 3 shows the values of internal and skin friction of saturated peat. The relation between d¢ and f¢ shows the same value under different loads. It is interesting to note that the right angle to grain wood has a good skin resistance on peat and better than parallel to grain. The values are d¢ = 0.757f¢ for right to angle to grain wood, and d¢ = 0.667f¢ for parallel to grain of wood with peat.
(b) Silt. Values of skin friction are shown in Table 4. The skin friction between silt and wood was investigated with the silt in saturated conditions. In the case of saturated silt the strain controlled shear box was used with parallel to grain and right angle to grain specimens. For saturated silt both of parallel to grain and right angle to grain wood were applicable because it could be tested in the small strain controlled shearing box (0.30 mm/min). The test results are shown in Table 4. It is interesting to note that the right angle to grain wood has a good skin resistance on silt and better than parallel to grain. The values are d¢ = 0.876f¢ for right to angle to grain wood, and d¢ = 0.785f¢ for parallel to grain of wood with silt. In the case of rough steel and dry silt the skin friction was close to the internal friction of the silt soil itself. The values of f and d increase somewhat with the increase of the normal load.
Skin Friction on Concrete
The concrete specimens, as was mentioned before, had two kinds of roughness.
(a) Peat. Table 3 shows the values of internal and skin friction of saturated peat. The relation between d¢ and f¢ shows the same value under different loads. It is interesting to note that the rough concrete has a good skin resistance on peat and better than smooth concrete. The values are d¢ = 0.939f¢ for rough concrete, and d¢ = 0.711f¢ for smooth concrete with peat.
(b) Silt. The skin friction between silt and concrete was investigated with the silt in saturated conditions. The values of internal and skin friction of saturated silt. The test results are shown in Table 4. It is interesting to note that the rough concrete has a good skin resistance on silt and better than smooth concrete. The values are d¢ = 0.994f¢ for rough concrete, and d¢ = 0.849f¢ for smooth concrete with silt. In the case of rough steel and saturated silt the skin friction was close to the internal friction of the soil itself. The values of d¢ and f¢ increase somewhat with the increase of the normal load.
CONCLUSIONS
The angle of internal friction (f¢) of silt soil (29° 18' 00 - 30° 42' 00) is bigger than the angle of internal friction (f¢) of peat soil (15° 55' 48 - 16° 25' 12), in spite of visual that fibers of peat soil are rougher than grain-size of silt soil.
The angle of friction (d¢) of silt soil (0.762f¢-0.850f¢) is bigger than the angle of friction (d¢) of peat soil (0.679f¢-0.722f¢), thus generally these values are agree with Bowles recommendation, for practical design d¢ = 2/3f¢.
The peak shear stress increases as the solid surface roughness increases. The shear deformation required to reach the peak value also increases with the surface roughness.
The modified direct shear apparatus used to generate the data reported in this study does not permit the establishment of residual friction angles. Therefore, the results and the relationships established in this paper are limited to peak values. It is suggested that in geotechnical applications where deformations at the interface are small, as in the case of gravity retaining walls, the peak value of interfacial friction is appropriate. When interfacial deformation are large, as in the case piles, a critical state value of this frictional resistance is more preferable, as has been observed from field and laboratory studies.
The Authors have analyzed the results of experimentation on the change of skin friction as a function of grain distribution of soils, moisture content, normal load, type of construction material, and difference of surface finish. In every case the skin friction was lower than the shearing strength of the soil. It is therefore important to determine the ratio between skin friction and shearing stress. However, because the two types of soil investigated behave in different ways, the Author believes that it is more practical to give the value of the “Coulomb line” for skin friction. In this equation the total skin resistance can be expressed in a similar form to that of Coulomb, if fc = ca/c and ff = d/f, the equation of skin friction becoming:
 |
(3) |
Because cohesive soils have constant cohesion at maximum density, other coefficient should be given for purely cohesive soils in the range where the shearing stress is independent of normal load. These fc, fa coefficients are given in Table 3 and 4.
For the designing engineer the safety against failure is essential. Therefore, in cases where the skin friction works as the bearing capacity of soils, the skin friction values should be reduced. Two basic theories could be applied, the plastic and elastic. In the ultimate strength design the factor of safety means the ratio of applied external load to the ultimate capacity of the structure. Therefore, in connection with skin friction, the ultimate skin resistance must be divided by the factor of safety.
In elastic theory this can be done in two ways:
(a) Plotting the stress-strain curve in regard to the Coulomb line. The yield point in the stress-strain curve (or reduced value of the yield point) projected back to the values of normal load (normal stress) will intersect the reduced safety value. In this case the sum of the cohesion and the friction part of the skin resistance is reduced by the safety factor.
(b) Approaching the allowable value of skin friction by reducing the adhesion and the angle of skin friction with safety factor separately.
Some authors suggest a comparison between the angle of skin friction and the angle of internal friction; others suggest a comparison of their tangents. The Author, however, made a quite wide range investigation of skin friction, which showed that it is more convenient to use the angle values of friction both in the designing and in field engineering. In some cases the angle value could be used directly (angle of inclination of earth pressure); in other cases and interpolation for ca, c and d, f values will be easier.
In summarizing it must be note that four major factors determine skin friction; the moisture content of soils, the roughness of surface, the composition of soils, and the intensity of normal load. By these the Author has shown, that in the case of cohesive soils the adhesion and friction part should be taken into account in evaluating skin friction.
REFERENCES
 | |
| © 2005 ejge | |