Piles to be driven through obstructions to bed- rock with the least driving effort and soil displacement would favor a steel H-pile or open-end pipe pile. Foundations
13 pages

215 KB – 13 Pages

PAGE – 1 ============
r Types of Piles: Their Characteristics and General Use BERNARD A. GRAND, Hardesty and Hanover This paper presents a review of the current practice and usage of the numerous types of pile in general construction. Information on this ject was obtained from a review of existing literature and from field perience. The paper reviews the purpose of pile foundations and the various factors involved in the selection of a type of pile.. Emphasis is placed on the general, physical, and structural characteristics of the piles as well as durability and fabrication. Data are presented on the inherent advantages and disadvantages of the various types of piles and on sponding optimum pile length and load range. Information and data are presented on the field problems of pile installations and the proper od of handling and treabnent to avoid damage or failure of critical pile sections. The fundamental information is supplemented by case histories. ŁPILE FOUNDATIONS of timber were in use in ancient times. In its earliest form, a pile foundation consisted of rows of timber stakes driven into the ground. Pile tions such as these were used by the ancient Aztecs in North America. The Romans made frequent use of pile foundations as recorded by Vitruvius in 59 AD. Pile tions for ancient Roman dwellings have been found in Lake Lucerne. It is reported that during the rule of Julius Caesar a pile-supported bridge was constructed across the Rhine River. The durability of timber piles is illustrated ill the report of the reconstruction of an ancient bridge in Venice in 1902. The submerged timber piles of this bridge, which were driven in 900 AD, were found in good condition and were reused. In the years immediately preceding the turn of the twentieth century, several types of concrete piles were devised. These early concrete piles were the cast-in-place type. Further development of the concrete pile led to the precast pile and, relatively recently, to the prestressed concrete pile. The need for extremely long piles with high bearing capacity led to the use of concrete-filled steel-pipe piles about 50 to 60 years ago. More recently, steel H-piles have come into common usage. Their ease of handling, fabrication, splicing, and relatively easy penetration hastened their ceptability in foundation construction. THE PURPOSE OF A PILE FOUNDATION The primary function of a pile foundation is (a) to transmit the load of a structure through a material or stratum of poor bearing capacity to one of adequate bearing pacity; (b) in some instances, to improve the load-bearing capacity of the soil; and (c) to resist lateral loads and to function as a fender to absorb wear and sbcick. In dition, piles are also used in special situations (a) to eliminate objectionable ment; (b) to transfer loads from a structure through easily er0ded soils in a scour zone to a stable underlying bearing stratum; (c) to anchor structures subjected to drostatic uplift or overturning; and (d) to serve as a retaining structure when hlstalled in groups or in a series of overlapping (cast-in-place) piles. Paper sponsored by Committee on Substructures, Retaining Walls and Foundations and presented at the 49th Annual Meeting. 3

PAGE – 2 ============
4 NEED FOR SUBSURFACE INVESTIGATIONS The length of the pile and the method of pile installation are dependent on the nature of the subsurface conditions. Thorough subsurface explorations are necessary to termine the stratification of the foundation elements, including the depth to bedrock and the density of granular materials measured by the number of blows recorded on a standard split spoon sampler, and to obtain undisturbed samples of cohesive strata to evaluate the shearing strength and compressibility characteristics by laboratory ing. The desirable number of exploratory borings depends on the size of the tion area and the degree of uniformity of the foundation materials. In areas of glacial deposits, the foundation materials tend to be nonuniform, whereas the soil conditions are generally more uniform in marine or alluvial deposits. Ideally, subsurface explorations should extend to a depth of 100 ft or to a depth of 1 times the width of the sb.·ucture, unless bedrock is encountered at a shallower depth. Gi·oundwater conditions are pertinent in a pile foundation project from the point of the probable permanency of the groundwater level, which is relevant to serving the permanency of untreated timber piles. The condition of the groundwater is also relevant to steel and concrete piles where acid, alkali, or other injurious tions may be present. CHOICE OF PILE TYPE The initial and primary consideration is the evaluation of the foundation materials and the selection of the substratum that will provide the best pile foundation support. In certain situatioDB involving cohesive subsoils, the pile lengths will be dictated by the necessity to minimize settlement of the foundations rather than the need to develop load capacity. The selection of o. type of pile for a given foundation should be made on the basis of a comparative study of cost, permanency, stability under vertical and a horizontal loading, long-term settlement, if any, of the foundation, required method of pile installation, and length of pile required to develop sufficient point bearing and frictional resistance assuming that there is a great depth to bedrock or other hard bottom. The selection of a pile type and its appurtenances is dependent on environmental factors as, for example, piles in seawater. Environmental factors to be considered are the possibility of marine borer attack, wave action causing alternate wetting and ing and ultimate deterioration, and abrasion due to moving debris or ice. Piles cated in strong water currents could be subject to gradual erosion of the pile material due to scouring by abrasive river sediment. Strong chemicals in rivers or streams or alkali soils could adversely affect concrete piles. Steel piles in an electrolytic environment near stray electrical currents could suffer serious electrolysis ration. Foundation materials consisting of loose to medium-dense granular soils would favor a tapered displacement pile for efficient transfer of load along the surface of the pile by friction. If the granular soils were in a very compact state, the piles would probably have to be installed with the aid of water jets. Foundation materials consisting of cohesive soil underlaid by a granular stratum would favor a side.d pile to develop the greatest possible skin friction area along the pile and point bearing area at the base of the pile. Piles to be driven through obstructions to rock with the least driving effort and soil displacement would favor a steel H-pile or open-end pipe pile. Foundations subject to large lateral forces such as pier bents in either deep or swiftly moVing water or both require piles that can sustain large ing forces. P rl:!cast, prestressed concrete piles are suitable for such load conditions. The large-diameter Raymond cylindrical prestressed piles have large vertical load and bending moment capacity and are frequently used in such installations. TIMBER PILES Timber piles have a wide range of sizes and strengths. The usual timber pile is a tree with a straight trunk and trimmed of branches. The butt diameter ranges in size

PAGE – 3 ============
I. 5 from 12 to 20 in. and the tip diameter from 5 to 10 in. Their availability depends on transportation facilities and distance from lumbering regions. In North America the most commonly used trees for piles are southern yellow pine, Douglas fir, spruce, and oak. Southern cypress from the Atlantic and Gulf coasts are also extensively used in piling. Cedar piles, although decay resistant, do not find extensive use because of their relatively low strength. From Central America, some greenheart and angelique are used. They are hardwoods and have considerable resistance to marine borers. Physical Characteristics The maximum obtainable length of timber piles is of the order of 110 ft, but lengths over 80 ft are scarce. The normal length of available timber piles is 30 to 60 ft. The elasticity of timber makes wooden piles easy to handle. Timber is well adapted for use in dolphins and fenders for the protection of structures in water because of its silience, wearing qualities, and ease of replacement. Timber piles are comparatively light for their strength, and they can absorb normal driving stresses to d evelop their design load. However, they are vulnerable to damage in hard driving. Timber piles are also vulnerable to deterioration and to destruction by marine organisms as scribed later. Durability of Timber Piles Timber piles are subject to deterioration caused by decay, insect attack, marine borer attack, and abrasive wear. Decay is caused by growth of fungi that need ture, air, favorable temperature, and food. Decay can be prevented if wood can be kept dry, rendered unsuitable for food, or entirely embedded in earth and cut off low groundwater level or submerged in fresh water. Thus, untreated timber piles are subject to decay and insect attack where they project above the water table or above the ground surface, and to marine borer attack where they project above channel tom in saltwater. Reasonable protection against decay and insect attack, such as termites, can be attained by poisoning the pile by impregnating the wood with pentachlorophenal or with creosote. Treatment with pentachlorophenal is not recommended for marine piles. Creosote treatment by a pressure process is the most effective method of poisoning wood piles for long-term protection. However, this treatment will not prevent mate damage by certain species of marine borers, notably the liminora. Mechanical protection of wood piles in waterfront structures has been used fully to protect new piles and to repair piles damaged by abrasion or by marine borers. Mechanical devices include Gunite encasements and precast concrete jackets grouted to the piles. Intrusion-Prepakt concrete placed inside of forms fitted to timber piles has also been used. Such encasements generally extend from a few feet below the mud line to some distance above the high water level. Fabrication It is the general practice to remove the bark from wood because timber piles erally carry load by skin friction. A decomposed weak film ultimately develops tween the bark and the wood creating a plane of weakness. The b utts of timber piles are cut square and the edges chamfered. The chamfering tends to reduce the tendency to split during pile-driving. When piles are to be driven without the aid of water jets, it is standard practice to trim the pile tips to about a 4-in. diameter when driving through relatively firm foundation materials. In driving through gravelly soils, it is frequently the practice to point the pile tips and clad them with steel shoes to prevent brooming. Timber piles can be spliced when long piles are unavailable; however, it is consuming and rather difficult. Sleeve joint splices have been fabricated with 8-in. and 10-in. diameter pipe, 3 to 4 ft long. Bolted splices have been made by using timber and steel splice bars. Gunite splices 6 ft long have been made by utilizing spiral reinforcement surrounding %-in. diameter longitudinal reinforcing bars covered

PAGE – 4 ============
6 with a 5-in. thick mortar section. In current practice, splicing of timber piles is an infrequent occurrence. Structural Characteristics The normal design load for a timber pile is 15 to 25 tons with a maximum sible load of 30 tons. A number of load tests on timber piles embedded for their full length have indicated a safe load capacity of 40 tons. Timber piles are vulnerable to damage in hard driving, and a water jet is frequently utilized in the installation of piles in dense granular materials. A single jet pipe strapped to the pile is generally used to install the pile to within 2 to 3 ft of the desired tip elevation, and the pile is driven to its final position to the prescribed driving resistance. Timber piles, designed to develop their load by end bearing, are sometimes driven butt down to utilize llie larger end bearing area. Timber piles installed as dolphins are occasionally driven butt down to take advantage of the larger pile section in the zone of maximum bending produced by lateral loads. In fender pile systems, it is good practice to avoid the use of bolted connections tween piles, sheeting, bracking, and struts, because such fixed restraints tend to be destroyed when deflected by lateral impact. STEEL PILES Durability of Steel Piles Steel piles embedded in relatively impervious earth, at least 2 ft below ground face, will generally be free of corrosive effects because of insufficient atmospheric oxygen. Embedded steel piles may be subject to corrosion if the surrounding medium consists of coal, alkaline soils, cinder fills, or wastes from mines or manufacturing plants. Steel piles protruding from the ground are subject to rusting at and somewhat below the ground line. Steel piles protruding into fresh water are generally subject to little deterioration but usually experience severe deterioration in seawater. rosion is severest in the splash zone. Corrosion of steel piles by electrolytic action is uncommon. Local electrolytic action and subsequent corrosion may occur in a saltwater environment where the steel pile forms one pole of a battery with the other pole in a dissimilar metal in close proximity. However, when steel piles are embedded in a conc1·ete footing, and by insulated from stray electric currents from the superstructure, electrolysis is generally not a problem. Electrolytic deterioration of steel piles can be minimized or prevented by the application of a protective coating such as epoxy coal tar paint or by positive cathodic protection using either electrolytic or galvanic anodes. Steel piles can be protected against corrosion failure at critical zones by an crease in the steel cross section, or by encasements. Steel pile encasements have be1:1u made of poured-in-place concrete, precast concrete jackets, or Gunite applied before or after pile-driving. Steel H-Pile Steel H-piles are rolled steel sections with wide flanges so that the depth of the section and width of the flanges are of about equal dimension. The cross-sectional area and volume displacement of the H-pile are relatively small; consequently, they are well adapted to driving through compacted granular materials and into soft rock. Steel H-piles, of their small volume displacement, have little or no effect in causing ground swelling or rising of adjacent piles. The maximum length of steel H-piles is relatively unlimited. Unspliced pile lengths of 140 ft and spliced lengths of more than 230 ft have been driven. The optimum pile length is 40 to 100 ft. The recommended design stress for fully supported piles is 9, 000 psi. The normal load range is 40 to 120 tons. Piles with heavy reinforced flanged sections have been driven to design loads of 200 tons and test loaded to 400 tomi.

PAGE – 5 ============
I. 7 Steel H-piles are easy to splice. Splices can be either riveted, bolted, or welded, the latter being the most common procedure followed. It is desirable to keep splice material on the inner faces to avoid creating a hole in the ground larger than the pile section. This may result in a loss of frictionai resistance. For bard driving tions, splices should develop one-third the full strength of the section. Splices, in long piles with no lateral support, should develop the full strength of the section. Caps are not usually required for steel H-piles embedded in concrete. hensive tests by the Ohio Department of Highways in 1947 indicated that capped H-piles embedded for only 6 in. into concrete proved as effective in ring load as H-piles with cap plates. The points of steel H-piles are sometimes tapered and generally reinforced when hard driving is anticipated or when they are to be driven to bedrock. Points are ally reinforced by welding plates to increase the thickness of the original section by a factor of 21/z to 3. Devices can be attached to a steel H-section to increase the bearing capacity of the pile to be driven into firm materials. Some devices that have been used consist of short sections of straight or wedge-shaped H-piling welded to the sides of the pile to increase the cross-sectional area at or just above the point. Steel Rail Pile Old rails have been used as piles by welding 3 rails together at heads or bases. The usual length of rail piles is about 30 ft. Sections of these rail piles have been butt-welded to fabricate a pile 90 ft in length. Rail piles are generally made of doned steel rails and are not considered normal steel production piles. Steel Box Pile Box piles have been fabricated from sections of steel sheeting in the form of a closed rectangular section. Because of their relatively large exterior dimensions, such piles can sustain large lateral loads and have been used to stabilize sliding banks. Box piles can be cleaned out and filled with concrete for additional bending strength. Disk Pile Diak piles have been fabricated of cast-iron pipe with a plate or casting of enlarged size connected to the base of the pipe. A disk pile has been fabricated with a pipe size of 9 in. and a disk diameter of 36 in. Such piles are usually jetted into position for end bearing on a firm stratum. Disk piles are rarely used today. Screw Pile Screw piles were used more extensively in the past than they are at present. The pile consiSts of an open-end pipe section to which is attached a number of turns of a helical shaft or screw at the base of the pipe. The pile is screwed or augered into the ground. water jets are generally used to facilitate the advancement of the screw pile into the ground. A relatively recent screw pile installation involved a 42-in. diameter and ‘ls-in. thick shell to which was attached an 8-ft diameter helix at the tip of the pile. The steel shell was fitted with a conical point. Such piles were installed mechanically in 20 to 65 ft lengths. Screw piles can be installed with little or no disturbance to existing structures. CONCRETE PILES Concrete piles fall into 2 basic categories: precast and cast-in-place. Precast piles can be divided into the 2 general classes of normally reinforced piles and stressed piles. Cast-in-place piles can be further subdivided into piles with casing and piles without casing. There are a number of variations of both of these basic types including a variation of cross-sectional area and longitudinal shape. Concrete piles are essentially unaffected by biological organisms or decay as are timber piles. They

PAGE – 6 ============
8 are thus used in foundations where the piles extend above groundwater or are mersed in river water or seawater. Depending on the foundation conditions and the type of concrete pile selected, the load carrying ability of the pile can be developed in either skin friction or point bearing or a combination of the two. Concrete place piles, and more particularly prestressed concrete piles, can sustain high ing stresses and are frequently used in viaducts and trestle type of structures with the pile extending above ground or channel bottom level. Durability of Concrete Piles Plain or reinforced concrete piles embedded in earth are generally considered not subject to deterioration. The water table, if free from deleterious substances, does not affect their durability. In extremely infrequent situations, there is the possibility that concrete piles embedded in permeable soils may be damaged by groundwater rated by either acids, alkalies, or chemical salts. These commercial agents can sult from wastes discharging from manufacturing plants, sewer leakage, leaching from alkali soils, or leaching of acidic compounds from coal or cinder fill. The use of dense rich concrete with sulfate-resisting cement is a m eans of minimizing the effects of a deleterious environment. Concrete piles should not be used where severe detrioration could possibly result. Concrete piles extending above the surface of a body of water are subject to damage from the abrasive action of floating objects, from ice where such exists, and from sand scouring. Damage can also result from frost action, particularly in the splash zone, and from internal corrosion of the reinforcement causing spalling of the crete. The principal factors involved in these frequent types of failures are (a) position and density of the concrete, (b) porosity of the aggregates, and (c) concrete cover over the reinforcing steel. Normally reinforced concrete piles are more nerable to spalling failure than prestressed piles because of inherent fine cracks in the concrete that develop from shrinkage, from handling of the piles, and from tension and shear loads. . The deterioration of concrete piles can be minimized by careful formulation of the concrete mix, use of sound, hard aggregates, and proper mixing, placing, ing, and curing to achieve hard dense concrete. The reinforcing steel should have a minimum cover of 2 in., and the use of galvanized reinforcing is advisible where nomically permissible. Prestressing reduces cracks in concrete and should be used whenever possible. Piles can be protected against some agents of deterioration by use of coatings and jackets applied to vulnerable areas. On a project under way in New Jersey where stressed cylinder piles will be exposed to seawater, the interior and exterior surface of the piles are to be coated with an epoxy bonding compound immediately following sandblasting of the surface. The epoxy bonding compound is to provide a tight seal on the pile surfaces. On a recently completed project in Long Island involving the use of prestressed crete pile bents, the pile surfaces in the tide zone were protected by wrought-iron pile jackets grouted to the piles. Right-angle sections of ’11-in. thick wrought iron were bolted together to form a square jacket, and the 1 ‘fz-in. annular space between the jacket and pile was filled with grout put into place by a tremie. CAST-IN-PLACE CONCRETE PILES In general foundation work, the cast-in-place pile is more commonly used than the precast pile. Cast-in-place concrete piles generally need no storage space, are made in place to correct length, do not require special handling, and are not subject to age from handling. Cast-in-place piles can be subdivided into 2 basic types: those that are formed in a steel shell in the ground and those that are uncased. Cased piles are the more positive type in that they permit an inspection of the pile prior to placing concrete, and allow for more accurate control in placing concrete. Uncased piles are generally more economical; however, they bear a great inherent risk in their installation.

PAGE – 8 ============
10 places where it will deform because of soil displaced while adjacent piles are driven. The cased piles have an advantage in that the pile can be examined before it is filled with concrete. The placement of concrete, particularly in tapered piles, should be carefully inspected and controlled. There have been cases where such piles have been improperly filled, resulting in intermittent voids along the pile. Deformities or tortions in the pile shell could constrict the flow of concrete into the pile leading to the formation of intermittent voids. The following is a description of the various mandrel driven cast-in-place cased piles in general use. 1. Raymond pile (standard type)-This tapered pile with a thin corrugated shell is driven with a solid mandrel bearing on the boot of the pile. The shell sections come in 8-ft lengths, with available shell thicknesses varying from 10 to 24 gage depending on the nature of the foundation materials. The optimum pile length is 35 ft, and the optimum load range varies from 30 to 60 tons. The pile is best suited as a friction pile. 2. Raymond step-tapered pile-This pile embodies the same fundamentals as the standard type with the exception that sections of the pile increase in diameter forming a series of steps. This permits increasing the length of the pile up to a maximum of about 100 ft without an excessive increase in the butt diameter. The piles are driven with a stepped solid mandrel. 3. Monotube pile-This is a fluted pile with a tapered steel shell and is best suited for friction piles of medium length. The shells are furnished in gages ranging from 3 to 11, and sustain direct driving with hammers of comparable size to those used for driving timber piles. The optimum length of these piles ranges from 30 to 80 ft with a load range of 50 to 70 tons. This type of pile cannot take excessive driving because the relatively light-gage steel shell will deform at the head. However, these piles can sustain lateral pressures from adjacent driving considerably better than the thin-shell piles. 4. Cobi pile-The casing of this pile is a thin-gage corrugated shell of uniform ameter driven in lengths up to 60 ft. This pile is driven with a Cobi pneumatic mandrel, which, when inflated, expands to a diameter slightly larger than that of the shell. The mandrel and shell are driven as a unit to the desired depth without a tendency to curve during driving. In this respect, the Cobi pneumatic mandrel-driven pile has an tage over the standard mandrel-driven pile. 5. West’s Rotinoff shell pile-The pile consists of a series of precast reinforced concrete shell sections joined together by steel bands and connected to a concrete shoe. The pile is driven by means of a mandrel bearing on the concrete shoe. Following moval of the mandrel, the pile is inspected and filled with concrete. The pile can be driven at locations with restricted headroom because the pile is assembled in sections. Piles of this type have been installed in lengU1s up to 100 ft. This pile was developed in Great Britain and has been used primarily in Europe. 6. Button-bottom cased concrete pile-This pile is installed by driving a walled steel casing, usually 14 in. in diameter, plugged with a heavy concrete button having a diameter 1 in. larger than that of the casing, to a stratum of firm bearing. A corrugated shell with a flat plate at its base is placed inside the pipe. The flat plate has a center hole that fits over a bolt cast into the concrete button. To prevent ing or heaving during placement of concrete, the shell is anchored to the bottom by threading a nut over the bolt by means of a long socket wrench. The steel casing is withdrawn, and the shell is filled with concrete. These piles can take heavy driving through obstructions and derive their support primarily by point bearing. Piles 76 ft in length have been installed with design loads in the order of 50 tons. 7. Swage pile-This pile is formed by forcing a light steel casing, usually 11 in. in diameter and 1/a in. thick, over a tapered precast concrete plug so that the pipe is swaged out by the taper of the plug, forming a watertight joint. The pile is driven by means of a ram bearing on the plug inside the pipe, pulling the swaged pipe with the concrete plug. Following the removal of the ram, the casing is filled with concrete. These piles have been found to be advantageous in extremely hard driving conditions.

PAGE – 9 ============
11 8. Closed-end steel pipe pile-This pile is formed by driving a steel pipe into the ground to the desired penetration, and filling it with concrete. The cylindrical steel pipe is of relatively heavy-gage wall thickness, generally ranging from 5/16 to Ya in. The pipe diameter ranges from 8 to 36 in. Seamless pipe is furnished in diameters up to 24 in., with spiral welded pipe available in larger diameters. Lap-welded pipe is sometimes used but is not recommended in driving through obstructions. The piles are driven with a flat plate or with a tapered cast-iron or steel point welded to the bottom of the pipe. Additional sections of pipe can be added by means of a cast-steel drive sleeve, permitting easy installation of piles of variable length. The optimum pile length ranges from 40 to 120 ft. The optimum load range is usually 80 to 120 tons. The piles are structurally capable of carrying large loads above ground level; the shell participates in carrying the load. The piles also provide high bending resistance under lateral loading. Pipe piles provide alignment control during installation and are capable of hard driving. This type of pile is used extensively in underpinning work cause it can be installed in short sections by jacking. The advantages of this type of pile are offset by its relatively high cost. 9. Open-end steel pipe pile-These piles are similar to the closed-end pipe piles except that no closure is used at the tip of the pile. These piles are capable of being extended through obstructions because interferences can be broken or removed through the open pipe. The piles are used where soil displacements would be objectionable or where 215 KB – 13 Pages