Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
CONCRETE AND MASONRY DAMS 183 6 Concrete And Masonry Dams GRAVITY DAMS Gravity dams (see Figure 6-1) are the most common of the concrete and masonry types and the simplest type to design and build. A gravity dam depends on its weight to withstand the forces imposed on it. It generally is constructed of unreinforced blocks of concrete with flexible seals in the joints between the blocks. The most common types of failure are overturning or sliding on the foundation. The foundation for a gravity dam must be capable of resisting the applied forces without overstressing of the dam or its foundation. The horizontal forces on the dam tend to make it slide in a downstream direction, which results in horizontal stresses at the base of the dam. These in turn may try to induce shear failure in the concrete at the base or along the concrete-rock contact or within the rock foundation. Uplift forces, in combination with other loads, tend to overturn the dam, which in turn may cause crushing of the rock along the toe of the dam. There are a number of older dams in existence constructed of rock and cement or concrete masonry. These generally have been relatively small and are usually of some form of gravity-type configuration. Their greatest weakness generally lies in the tendency for the masonry or cement between blocks to deteriorate with resultant leakage, deformation, and general disintegration.
CONCRETE AND MASONRY DAMS 184 Figure 6-1 Gravity clam. Source: Courtesy, U.S. Bureau of Reclamation. Buttress Dams This is a form of gravity (see Figure 6-2) dam so far as the force distribution is concerned. It consists of a sloping slab of concrete that rests on vertical buttresses. Because of its shape there are high unit loads underneath the buttresses; thus, the foundation must not undergo unacceptable settlement or shearing. In addition to the factors mentioned for gravity dams, particular attention must be paid to the quality and performance of the concrete in the face slab. Because of its relative thinness it cannot withstand excessive deterioration, pitting, or spalling that will decrease the strength of the slab and
CONCRETE AND MASONRY DAMS 185 increase its potential for seepage through the concrete. The buttresses also must be designed to withstand overturning forces. If their footings are too small, the resulting high unit loads can induce crushing in the rock. Because of their shape, buttress dams usually do not require extensive, if any, drainage systems, and drainage galleries within the dam would not be feasible. Arch Dams Arch dams (see Figure 6-3) are relatively thin compared with gravity dams. The forces imposed on such a dam are, for the most part, carried into the abutments, and the foundation is required only to carry the weight of the structure. The shape of the dam may resemble a portion of a circle, an Figure 6-2 Simple buttress dam. Source: Courtesy, U.S. Bureau of Reclamation.
CONCRETE AND MASONRY DAMS 186 ellipse, or some combination thereof. The dam usually is constructed of a series of relatively thin blocks that are keyed together (see Figure 6-4). The construction joints that result may be grouted during or after construction or left open. In the latter ease it is expected they will close under the reservoir load. Occasionally, flexible seals may be installed in the vertical joints between the blocks. Figure 6-3 Plan, profile, and section of a symmetrical arch dam. Source: Courtesy, U.S. Bureau of Reclamation. Because of the translation of imposed forces into the abutments, the design must consider the amount of deformation (modulus of deformation) that will occur in the abutments when the various loads are imposed on the dam. If the deformation exceeds design criteria, tension cracking can occur in the concrete. (See the section Abutment or Foundation Deformation.) Because the design is predicated on the flexibility of an arch, it is generally
CONCRETE AND MASONRY DAMS 187 desirable that the modulus of elasticity of the rock abutments be less than that of the dam concrete. Although controversial, some designs do consider the possibility of uplift. Thus, there may be drainage galleries and their appurtenant drain holes within the dam; drainage galleries and drain holes are generally installed in the abutments. Possible failure modes in an arch dam are overturning, excessive abutment movement causing tension cracks in the concrete and subsequent rupture of the dam, mass movement of the abutments causing dam failure or disruptive stresses in the dam, and excessive uplift in the foundation that causes movement of rock blocks in the foundation and/or overturning of the dam. Arch-Gravity Dams In arch-gravity dams imposed loads are carried partially by the foundation and partially by the abutments. These dams are of block construction and have a cross section that has a mass somewhere between that of an arch and Figure 6-4 Concrete arch dam under construction; shows keys between blocks.
CONCRETE AND MASONRY DAMS 188 that of a gravity dam. The comments made earlier for arch and gravity dams are applicable to this type of structure, too. Miscellaneous Types Various combinations of the types of dams described above may be designed for unique site situations. These include multiple arch (see Figure 6-5), multiple dome, compound arch, and gravity-buttress. The type of dam indicates the mode of distribution of the forces imposed on it. COMMON DEFECTS AND REMEDIES The following discussions are intended, first, to emphasize the defects and remedies that generally could be relevant to any type of concrete dam and, Figure 6-5 Double-wall buttress multiple-arch type. Source: Courtesy, U.S. Bureau of Reclamation.
CONCRETE AND MASONRY DAMS 189 second, to indicate those remedies that are applicable primarily to a specific type of dam. A summary of the discussions is presented in a matrix format in Table 6-1. Abutments Joints, Fractures, Faults, And Shear Zones The orientation of major discontinuities in abutments is critical in relation to the distribution of stresses from an arch dam but not as critical for a gravity structure. For an arch dam the main consideration is whether the direction of such discontinuities is parallel to or closely parallel to the directions of thrust from an arch (see Figure 6-6). If so, movements can occur that would result in weakening or possible loss of large blocks in the abutment. For a gravity dam the potential for sliding may be greatest when the foundation rock has horizontal bedding, particularly where combined with slick bedding planes. Consideration also must be given to a zone within the foundation rock that is peculiarly susceptible to the development of unacceptable uplift forces. The presence and behavior of large faults or shear zones in those abutment areas within the zone of stress influence of the structure is of potential concern. Mass abutment movement may occur because percolation of water through these zones or water-softening of the rock material may reduce the shearing strength or cause consolidation of the rock. If at the upstream side of the dam the zone is more pervious than at the downstream side, uplift or pressure buildup can occur. Seepage or Leakage Seepage developing in the abutments for any type of concrete dam can produce a critical condition. It usually is associated with fractures or shear zones. Of particular note is whether such seepage at the outlet is clear or contains silt or rock fragments. If the water is cloudy, silty, or muddy the water flow may be eroding the rock material itself or washing out clay or other impervious material that has been in the rock cracks. Continuation of this process (piping) can weaken the overall strength of the abutment or can produce increasingly large channels for water flow. If left untreated, the openings can enlarge sufficiently to cause abutment collapse or major movement of the abutment with the creation of unacceptable stresses in the body of the arch. Clear water leakage may be of concern if the quantity represents an unacceptable loss of reservoir storage, or the water may lubricate rock surfaces or reduce the strength of the rock element or discontinui
CONCRETE AND MASONRY DAMS 190
CONCRETE AND MASONRY DAMS 191
CONCRETE AND MASONRY DAMS 192
CONCRETE AND MASONRY DAMS 193
CONCRETE AND MASONRY DAMS 194
CONCRETE AND MASONRY DAMS 195
CONCRETE AND MASONRY DAMS 196 ties in the rock system. Clear or muddy water may indicate the development of uplift forces in the abutment that were not contemplated in the design of the dam. (See the section Uplift.) The Malpasset Dam failure in France (International Committee on Large Dams 1973) is an example of failure due to foundation water pressures and lubrication (see Figure 5-5). Figure 6-6 Relation of geologic structure to arch thrust. Various methods of leakage control have been used. ⢠Channelizing the seepage flow so increased head does not develop as a result of erosion. ⢠Installation of sand filters in the flow channels at the point of egress to prevent piping. ⢠Grouting with cement or other sealing materials to provide a barrier to the flow. Such barriers should be created near the upstream face of the dam, not downstream from the line of intersection between the dam axis and the abutment because it could result in unacceptable uplift in the dam or the abutment. ⢠Sealing the entrances to such cracks. Sealing materials would include bitumens, epoxies or other resins, cement grout, bentonite, concrete, and impervious soil blankets. Abutment or Foundation Deformation This is particularly critical for an arch dam because excessive deformation can produce unacceptable tensile stresses. Its effect may appear as tension
CONCRETE AND MASONRY DAMS 197 cracks in the concrete or as high tensile stress measurements if instruments are installed in the extrados of the arch. In addition, the dosing and possible crushing of faces of discontinuities, such as joints and fractures, might be observable. Also, there may be anomalous decreases in abutment seepage rates as the result of the closure of openings in the rock mass; such decreases may cause undesirable uplift pressures. Unusual movements in the rock mass of the abutment may result in loosening of large blocks or possible slides of both upstream and downstream abutments of the dam. In a gravity structure, incipient sliding motion may develop at the contact between the concrete and the rock. The evidence for this might be a zone of freshly exposed rock observable on the upstream face of the contact. Particularly critical are the slopes adjacent to spillways; slide blockage of an intake, chute, or spillway basin could be disastrous if it occurs during the operation of a spillway. In the case of an arch dam care must be taken to ensure that stresses developing from the dam have a sufficient mass of stable rock available to accept such stresses without undesirable displacements occurring in the rock mass. That is, there must be no topographic reentries immediately downstream from and within the influence of the abutment thrust area (see Figure 6-7). If such reentries occur, the possibility exists that the entire abutment mass may move in a downstream direction. The major area for the acceptance of thrust from an arch is often within an acute triangle that has its apex at the concrete- rock contact; the internal angle is about 15°, and the river side of the triangle is parallel to the thalweg of the river valley. In examining abutment areas consideration must be given to the fact that minimum safety may exist in the upper part of double curvature arches because the upper parts of valley walls are generally looser than the lower walls and earthquakes would induce stronger reactions in the upper area of both the dam and its abutments. Drainage galleries in the abutments should be examined carefully to determine if all drains are open and operating or whether some have been plugged by mineral deposits and/or silt. Pressure gages on drains should indicate if excessive uplift forces are developing. Abutment deformation can be recognized by fresh cracks in the rock surface, blocks falling from abutments, or displacement of vegetation. Recording instruments or surface survey markers may be installed in the abutment whereby both vertical and horizontal movement may be detected. Cracks in the dam concrete where it joins the rock also may be evidence of deformation. (It should be noted that such cracking or crushing may also be caused by excessive deformation of the arch caused by something other than abutment movement.)
CONCRETE AND MASONRY DAMS 198 Figure 6-7 Favorable topography for arch dams (A and B); unfavorable topography- unshaded parts of arrows indicate part of thrust that is daylighting (C and D). Deformation may be caused by reservoir load, earthquake forces, arch dam thrust, unfavorable orientation of the cracks in the rock with respect to the load directions imposed by the dam or reservoir, large ranges of temperature, freezing and thawing of water within the cracks in the rock, softening of the contact surfaces between rock blocks caused by water or other weathering forces, an abutment mass insufficient to withstand the overall thrust forces from the arch dam, presence of shear or fault zones that were not contemplated in the design, or excessive uplift forces. Any unforeseen movement in the abutment will induce stresses in an arch dam or buttress that may not have been considered in its design. Loosened blocks can endanger any structures in their potential fall path. If sufficient rock falls into the channel below the dam, it can cause blockage of the outlets to waterways. Rock falls into the reservoir immediately upstream from the dam may interfere with operation of outlets, penstocks, or
CONCRETE AND MASONRY DAMS 199 spillways. Rock falling onto the dam can endanger personnel and damage the concrete and roadways or other structures on top of the dam. Motion of the abutment can cause leakage with a potential for lubrication or pressure buildup and ultimately can cause complete failure of the dam. If ancillary structures are located on the abutment (or within it), the rock movement can interfere with their operation and/or cause damage to them. Remedies are highly dependent on the cause of the deformation or rock falls. Possibilities include any one or combination of the following: ⢠Deep rock anchors or rock bolts to tie together and strengthen the abutment and reduce or prevent further deformation. ⢠Horizontal or vertical concrete beams across the rock mass and anchored by rock bolts. Such anchors can be tensioned and grouted in or tensioned by future movement of the rock mass. ⢠If the rock falls are caused by abutment deformation and the deformation cannot be reduced, some relief can be achieved by extensive scaling of the abutment (removal of potentially loose blocks). If continuous loosening of blocks is expected, it may be desirable to build diverter walls. These are massive concrete walls located such that any loose rock falling from above will hit the wall and be diverted away from critical structures below the wall. This system was used successfully at Kortes Dam in Wyoming. ⢠If water is the causative agent, elimination of the flow of water into the abutment is needed. ⢠Extensive grouting to improve the modulus of deformation of the rock mass. ⢠Placing of buttressing rock, concrete, or gabions. Abutments are particularly susceptible to damage by earthquake forces. As previously noted, generally the upper part of valley walls are looser than the lower part of the walls; thus, earthquakes can induce much stronger reactions in the upper portion of the abutments. One California owner of a concrete gravity dam with a suspect abutment has used a number of measures to monitor and stabilize an abutment including regrading part of the abutment (upstream) and installing horizontal drains and inclinometers, installing drains under the dam to reduce uplift, and construction of concrete crib-wall backfilled with rock just downstream of the abutment. Stabilizing rock placed against the dam in the vicinity of the abutment and an elaborate survey control system and inclinometers have been installed. The foundation also has been heavily cement grouted. Other effective stabilizing measures are erosion control, rock bolting, deep anchorage, chemical grouting, installation of galleries and pries or other reinforcing members, and resloping of abutments. Con
CONCRETE AND MASONRY DAMS 200 siderable experience in grouting, both cement and chemical, is recorded in the ASCE (1982). Uplift Any estimate of uplift should be based on the current effectiveness of the foundation drainage system, as indicated by measurements, the quality of maintenance, and other influencing features, such as the possible presence of a silt layer on the reservoir floor that is more impervious than the rock foundation. The existence of a reliable foundation drainage system might justify relaxation of the design criterion published by several federal dam building agencies to the effect that full reservoir pressure should be assumed under the portion of the dam base not in compression. That criterion was originally adopted without the support of substantiating field measurements and without regard to specific characteristics of dam and foundation. However, it was considered appropriate for incorporation into conservative design criteria for new dams. Agencies promulgating such design criteria intend them to be guides to uniformly safe design rather than restrictions on the designer and intend to permit variations wherever warranted. Thus, it is only reasonable that the evaluation of existing dams consider the influence of actual site features and characteristics. In a recent stability evaluation for an overtopping flood condition (Clay-tot Dam, New River, Virginia), an effective foundation drainage system was considered as warranting deviation from the assumption of full reservoir pressure under any portion of the dam base not in compression. Papers describing the Claytor Dam studies were presented by representatives of the American Electric Power Service Corporation on October 28, 1982, at the annual ASCE Convention in New Orleans. A supporting paper (Goodman et al. 1982) was presented. Essentially, the studies found that the foundation drains would continue to function under loading conditions causing a lack of compression on a portion of the base and would preclude a buildup of uplift to full reservoir pressure under the dam. Experience at the U.S. Army Corps of Engineers' Dworshak Dam on the North Fork of the Clearwater River, Idaho, attests to the effectiveness of drains in reducing hydrostatic pressures in narrow cracks. During and shortly after reservoir filling, vertical cracks striking upstream-down-stream occurred in the center of 9 monoliths of the concrete gravity structure. They extended from the base of dam to heights ranging to almost 400 feet and propagated downstream past the drainage gallery. Drain holes (1.5 inches in diameter) were drilled into the cracks from the galleries and angled to intersect the cracks at 5-foot vertical intervals along a line about
CONCRETE AND MASONRY DAMS 201 30 feet from the upstream face. These drain holes successfully reduced hydrostatic pressures in the cracks, resulting in the reduction of crack widths and arresting of the downstream propagation of the cracks. Costly remedial measures might be avoided where foundation drainage, either existing or added, can be relied on to preclude the buildup of uplift pressures to full reservoir magnitudes in narrow cracks of the size comparable to those created by structure rotation under the increased flood loading. In such cases uplift can be represented by a linear distribution from tailwater pressure at the toe to tailwater pressure plus a percentage of the difference between the headwater and tailwater pressures at the line of drains and thence to headwater pressure at the heel of structure. The percentage of the difference factor that determines the uplift at the drains should be based on pressure measurements that can be obtained during normal pool levels. Generally, the most effective and economical solution to reduction of uplift forces is the installation of drains. Where they already exist, their monitoring and maintenance are essential. Regular drain flow observations must be part of any surveillance program. Accumulation of deposits in the drains is monitored by periodic probing to determine location and characteristics of the obstructing material. When uplift is steadily increasing or when seepage flows have decreased substantially, the need for cleaning drains or drilling new ones is indicated. When drains become so obstructed as to impair their function, and the deposits are relatively soft, they can be cleaned by washing. However, this is often only a temporary remedy. A better solution is to redrill the old drain or to drill new drains (Abraham and Lundin 1976). Where drains do not exist or are inadequate, new ones can often be drilled into the foundation from existing galleries or from the downstream face. At the California dam referred to earlier, some old drains leading to a gallery were cleaned, new drains were drilled from the gallery into the foundation, the drains under the spillway bucket were cleaned, and new drains were drilled at a flat angle underneath the spillway bucket. Concrete Quality For existing dams a great deal can be learned about concrete quality by visual observation. Surfaces subject to rapidly flowing water, such as spillways or outlet chutes, must be examined carefully. Silty or sandy water can erode concrete. Water moving rapidly past abrupt surface changes creates regions of negative pressure which may cause cavitation erosion as evidenced by increasingly deep holes in the concrete. Vibration also can result from pressure fluctuations and high-velocity impingement. Where small
CONCRETE AND MASONRY DAMS 202 amounts of material have been removed, simple repairs have been made at some dams with a very smooth epoxy coating. Large repairs have needed extra strong concrete, such as fiber-reinforced concrete. Steel plates have in some cases been installed on cavitated surfaces. Also, cavitation erosion often can be prevented by the introduction, under the water flow, of air through slots or other openings. Where strength is a question, nondestructive tests, such as the rebound hammer or sonic velocity measurements, are only qualitative. The most accurate evaluation of strength can be made by extracting cores of a diameter two to three times the size of the largest particles in the concrete. These can be tested for compressive and tensile strength, for modulus of elasticity and Poisson's ratio, and for density. All of these properties are needed for any analysis of the behavior of a dam. Careful attention should be paid to the appearance of weathered concrete surfaces. Pattern cracking may denote either drying shrinkage or, in extreme cases, alkali-aggregate reaction. Heavy surface scaling may indicate freeze- thaw effects or insufficient cement and, consequently, low strength. Experience with Deterioration at Drum Afterbay Dam The story of Drum Afterbay Dam is a good example of the detection and investigation of a dam with deteriorating concrete. Built in 1924, this dam was a thin arch structure, 95 feet high, situated at elevation 3,200 feet on the western slope of the Sierra Nevada Mountains in California. Aggregate for the concrete was crushed from the rock (schist) at the dam site, which turned out to be an unfortunate decision. Twenty years after construction the downstream face showed visible signs of deterioration due to frost action, with particularly noticeable deterioration in the horizontal joints between lifts. At that time some repairs were made by chipping out poor concrete and filling with gunite. After another 20 years it was apparent that a more thorough investigation should be made to pinpoint the causes of the worsening deterioration. This later study found, in addition to freeze-thaw action, visible signs of a possible alkali- aggregate reaction. At this time a more elaborate study was made, utilizing 6- inch and NX cores and sonic velocity measurements. From the cores, measurements were made of strength, modulus of elasticity, Poisson's ratio, density and thermal diffusivity; also, a careful petrographic examination was made. Correlations between pulse velocity measurements and strength were used to target the areas of generally deteriorated concrete, which by this time had reached strengths as low as 1,400 psi. The petrographic examinations showed that the principal culprit was pyrites in the aggregate, which in
CONCRETE AND MASONRY DAMS 203 combination with the lime from the hydrating cement set up new compounds of low strength. After this the prognosis for the concrete was more of the same or worse. The dam was deteriorating at an accelerating rate, and the decision was made to replace the dam entirely (Pirtz et al. 1970). Experience with Synthetic Materials For Concrete Repairs The strength and exceptional adhesive ability of certain synthetic materials have led to their application in repairing concrete both for surface treatment and for injection to seal cracks. Resins with low sensitivity to water have been used as bonding agents between old and new concrete. Epoxy-based and polyester- based resins have been widely used for facing on dams and other hydraulic structures. Epoxy-based resins of appropriate mix have been found to be more effective on damp concrete than polyester-based resins. The viscosity of resins used for injection can be varied from pump-able mortar to very thin grout. Careful workmanship is required to ensure lasting protection by resins. The Southern California Edison Company has made effective use of synthetics in sealing concrete surfaces. For example, the upstream face of Rush Meadows Dam, a concrete arch at high elevation in the Sierra Nevada, was coated in 1977 with a layer of gunite covered by two coats of polysulfide. The first layer of polysulfide was thin, placed over a primer, and was followed by a thicker final layer. The treated face effected a substantial reduction in seepage and has shown no signs of distress, neither peeling nor general deterioration. Edison has made such applications on other dams with comparable success. Pacific Gas and Electric Company also has used similar techniques successfully. Repair of concrete by injection of synthetics has a less extensive record but holds promise in special cases. At the Corbara Dam in Italy an experimental attempt was made to seal cracks in buttresses (due to thermal shrinkage) by application of epoxy resins. Remedial work was done in the winter to ensure the widest opening of the cracks. The work entailed drilling, chemical washing of cracks, blowing with air, placing small copper pipes to drain and control grouting, superficial mortaring, and grouting at about 60 to 70 psi. Some of the work was done by flowing warm air into the crack prior to the injection. Several difficulties were incurred at some cracks, such as only partial penetration due to excessive viscosity or inadequate adhesion because of moisture or unfavorable temperature. However, there was an appreciable improvement in shear strength along the cracks sufficiently treated with the resin. For internal remedy of general fine cracking in concrete structures, the potential for success can be enhanced by injection of resins into boreholes,
CONCRETE AND MASONRY DAMS 204 with careful temperature control, drying with hot air, and proper venting (Vallino and Forgano 1982). Experience with Steel-Fiber Concrete Where concrete is subjected to high impact or erosion or cavitation, improvement can be obtained by removing damaged material and replacing it with a mix containing randomly distributed steel fibers. This was successfully accomplished by the U.S. Army Corps of Engineers at Dworshak Dam in Idaho in the stilling basin and at a sluice. The fibrous concrete had a low water/cement ratio and a high cement factor and was placed in the more deeply damaged areas. Some surfaces were polymerized to improve durability. Fibrous concrete was used similarly for remedial work on the stilling basin at Libby Dam in Montana. Additionally, certain areas of floor slabs in the stilling basin were polymerized. Shallower repairs at Dworshak were done with epoxy mortar but did not prove satisfactory; most of it failed after a rather short period of service. Nonetheless, in other projects with less demanding service conditions, epoxy mortar has provided effective repair. STABILITY ANALYSES Concrete and masonry dams must interact with the rock foundation to withstand loads from the weight of the structure, forces from volume change due to temperature, internal water pressures (uplift), external water pressures, backfill, silt, ice, earthquake forces, and equipment (see Figure 6-8). Uplift pressures used in stability analyses should be compatible with drainage provisions and uplift measurements if available. Dams should be capable of resisting all appropriate load combinations and have adequate strength and stability with acceptable factors of safety. The factors of safety recommended for various loading combinations are given in U.S. Bureau of Reclamation (1976, 1977). The foundation has a significant influence in the stability evaluation of masonry structures. It must have adequate strength to support the heavy loads of the structure without excessive displacement. In addition, it must function as the water barrier with adequate provisions for drainage and relief of uplift. It should also be as free as practicable of such weaknesses as extensive weathering, faults, jointing, and clay seams. The existence of such defects at existing dams should be evaluated carefully to determine if they require treatment. Gravity dams can be analyzed by the gravity method, trial-load twist analysis, or the beam and cantilever method, depending on the configuration of the dam, the continuity between the blocks, and the degree of re
CONCRETE AND MASONRY DAMS 205 finement required. The gravity method is the most common and is applicable when the vertical joints between individual monoliths are not keyed or grouted. Trial-load twist analysis and the beam and cantilever method are appropriate when the monolith joints are keyed and grouted; however, the gravity method can be used in this situation for an approximate or preliminary analysis. Descriptions of these methods, together with safety factors and allowable stresses, can be found in U.S. Army Corps of Engineers (1958-1960) and U.S. Bureau of Reclamation (1976). Figure 6-8 Expected loads on a concrete dam. Arch dams are usually analyzed by the independent arch theory (limited to relatively small structures or analyses preliminary to more refined methods) or by trial-load methods. Both two-and three-dimensional finite element methods of analysis are available and can be used to perform trial-load analysis or other stress-determination methods. Details of some methods, with appropriate safety factors and allowable stresses, can be found in U.S. Bureau of Reclamation (1977). Flood Loading The evaluation of stability of gravity dams during a spillway design flood is necessary in deciding whether modifications, such as added spilling capac
CONCRETE AND MASONRY DAMS 206 ity or strengthening measures, should be accomplished. In most existing dams a probable maximum flood would overtop the dam. However, concrete gravity dams on firm rock foundations are inherently resistant to overtopping flows provided stability against overturning and sliding are ensured and that the groin and foundation downstream of the dam are capable of resisting erosion and disintegration resulting from impingement of the overtopping water. An analysis to determine stability during great Floods should be based on conservative estimates of headwater and tailwater elevations. The analysis must consider site-specific conditions, such as quality of materials in the dam and auxiliary structures, foundation permeability and competence, and overturning. However, extensive damage to the structural components may be acceptable in certain cases for this extreme event. In addition to estimates of headwater and tailwater elevations, it is necessary to estimate the possible increase in the uplift loading on the structure. Seismic Loading Ground Motions The ground motions to be used in an analysis of the seismic load conditions are discussed in Chapter 5. Concrete Dam Response The way in which a dam responds to an earthquake is complex and varies with the type of dam and its foundation. For example, at a concrete dam on a rock foundation the earthquake motion is first felt at the foundation as rapidly changing motions in all directions, and many motions per second. Usually the horizontal accelerations are stronger than the vertical components of the motion, but all are present. The vertical acceleration adds to or subtracts from the weight of the dam. The dam responds by deforming elastically and developing stress. For a given seismic record methods now exist for determining these stresses and deformations. The computed stresses developed by the earthquake are compared with the strength of concrete cores obtained from the dam. In the latter circumstance an allowance must be made for the rapidity of loading and the linearity of the analysis. Fresh cores must be used in these strength tests. Some concrete dams have been damaged by earthquakes; others have been left untouched. For example, Koyna Dam, a concrete gravity dam in India, suffered a number of major cracks near the top after the Koyna earthquake in 1967 (Chopra and Chakrabarti 1973). However, these
CONCRETE AND MASONRY DAMS 207 cracks are confined mainly to horizontal construction joints. For safety the dam was later buttressed with additional concrete. On the other hand, Pacoima Dam, a concrete arch dam in California, sitting practically on the epicenter of the San Fernando earthquake in 1971, was undamaged from a shock measured at over 1.2 acceleration due to gravity on one abutment. [The recording at the abutment may be of questionable validity. However, peak horizontal acceleration at the dam base may have been on the order of 0.75 acceleration due to gravity (Seed et al. 1973)]. Methodology Most existing concrete dams in potentially seismic zones were designed for seismic loads by using equivalent static forces. These forces were obtained by multiplying dam weight by a seismic coefficient. It is generally agreed that this method is adequate for structures located in seismic zones below 3 (see Figures 5-12 and 5-13). In zones 3 and 4, or in other locations where the proximity to active faults warrants, a dynamic analysis should be made using, at a minimum, a response spectrum analysis. A time history analysis should be made where stress variations with time are critical. Descriptions of these methods can be found in Chopra and Chakrabarti (1973), Chopra and Corns (1979), U.S. Army Corps of Engineers (1958-1960), and U.S. Bureau of Reclamation (1976). IMPROVING STABILITY General Measures An existing gravity dam that has questionable resistance against sliding or overturning may be improved in various ways, depending on the suspected cause of instability. If excessive uplift is a problem, foundation drainage can be improved by cleaning drains and/or adding more drains. Increased positive resistance has been accomplished by stressed tendons anchored in the foundation rock, addition of concrete mass, construction of concrete buttresses, or placing a buttressing embankment against the downstream face. Buttress and Multiple-Arch Dams Slab and buttress and multiple-arch dams built 50 to 70 years ago were designed on principles that may not meet modern standards. Many of these structures have been modified to overcome questionable stability, especially in resistance to lateral loading, such as earthquake acceleration. In
CONCRETE AND MASONRY DAMS 208 some cases the strength of the concrete also has been found to be low. Cracks in various elements have indicated serious overstressing, even under normal loading. The arches forming the faces of some old multiple-arch dams have small central angles, so that the arches impose considerable thrust on the buttresses normal to their center lines. Such forces must be resisted partly by the adjoining arches if the buttresses are insufficiently braced. Some dams of this type originally had no steel reinforcement in the buttresses. In such cases cracking has typically been observed to extend diagonally downward from the upstream to the downstream face, being open at the juncture with the arches but terminating in hairline cracks at the downstream extremity, suggesting a slight rotational movement of the buttress about its toe. Micrometer gage readings have generally not shown appreciable movement at such cracks after their initial formation. Stability analyses of slender buttressed concrete dams with minimal reinforcement and bracing have disclosed typically that, even with relatively low seismic accelerations, the buttresses could be unstable. These weaknesses can be overcome by reinforcing the buttresses in various ways. Successful methods of strengthening include posttensioning the buttresses and constructing bracing members between them. The addition of shear walls in alternate panels or bays has in some cases provided effective lateral resistance. Shear keys have typically been provided at the joints, and bolts have been extended through the buttresses, with large bearing plates on the back side to distribute the bolt load on the old concrete. Horizontal beams bolted between buttresses also have served effectively. A basic requisite is that the connections between bracing elements and buttresses be detailed in such a way that lateral loads are transferred safely. Otherwise, the struts might be of less benefit than intended. In an investigation of buttressed concrete dams in California, concrete strengths in some of these structures were found to average less than 2,000 psi. For example, the compressive strength of concrete cylinders taken from one old multiple-arch dam averaged 1,889 psi and varied from 1,225 psi to 3,185 psi. This wide variation was attributed primarily to deficiencies in quality control during construction. No evidence of alkali-aggregate reaction was found. Where such chemical activity has been involved, even broader ranges of strength have been observed, with the minimum being less than one-fourth the maximum in some cases. In such cases the principal emphasis must be on the low-strength zones of the dam. A complete determination of structural adequacy necessitates data on the whole strength envelope, including both the range and distribution of values. Lack of uniformity of strength may induce excessive stress concentrations in low-strength areas, particularly if the weak zones are large. A concrete dam may have the capability to bridge across defects of limited extent.
CONCRETE AND MASONRY DAMS 209 Rollcrete The threshold of a new technology was recently crossed by design and construction of Willow Creek Dam in Oregon, by the U.S. Army Corps of Engineers. This dam is made entirely of roller-compacted concrete, which is essentially a well-graded gravel fill containing cement. Other dams of this type are in the design phase. The costs of concrete placement and construction time are reduced substantially by using this method. Rollcrete differs from soil cement in several important respects, primarily related to the mix, although both are placed in layers and are compacted by rollers. The cement content of soil cement may range as high as 18% by weight. In contrast, the cement content of rollcrete may be between 2.5 and 7% by weight. Compared with regular concrete, rollcrete requires less cement to attain equal strength, and its mix demands less strict processing and gradation. Compaction ensures denser concrete. The promise of this new technology may be greatest in construction of large new structures, where the potential economies of scale are obvious. However, it would appear to offer advantages also in remedial work on existing dams. For example, it would have useful applications where mass concrete sections have to be enlarged or where spillways and other channels need to be extended. Experience at Condit Dam The 125-foot-high Condit Dam in the State of Washington was rehabilitated in 1972 by improved drainage facilities and by installation of steel anchors (deSousa 1973). This concrete gravity structure, 60 years old at that time, had been determined to have a marginal factor of safety under normal loading conditions and to have inadequate resistance to extreme loadings by flood or earthquake. The concrete was in satisfactory condition, but the drains were only partially effective. Nearly full uplift pressure occurred at the midpoint of the dam base. In one phase of the remedial program a series of new drain holes was drilled radially from two sluice pipes that pass through the dam at low level. This reduced the uplift pressure from a maximum of 33 psi to less than 8 psi. Concrete cores recovered from the drilling had test strengths varying from 2,760 psi to 6,690 psi, with an average of 4,470 psi. Under extreme loading the dam would have been stressed in tension at the heel up to unacceptable levels. Therefore, as an additional remedial measure, steel anchors were installed to limit tension to 20 psi. Twenty-two posttensioned anchors were installed in the dam, and 3 were placed in the spillway foundation. The anchors varied in length from 50 to 100 feet and, each was loaded to about 300 kips. The typical depth of embedment in the
CONCRETE AND MASONRY DAMS 210 basaltic foundation was 25 feet. Other corrective work at the Condit Dam included pumping of 470 cubic yards of concrete into a fissure under the spillway structure and drilling six drain holes radially upward from the diversion tunnel to the base of the dam. Experiences at Spaulding Dams Three separate concrete dams form Pacific Gas and Electric Co.'s Lake Spaulding in California. The main dam is a 276-foot-high arch-gravity dam. Dam 2 (the main spillway dam) is a 42-foot-high gated gravity structure. Dam 3 is a 91-foot-high gravity buttress dam. All three dams were built between 1912 and 1919, and the concrete had deteriorated significantly. The investigations and improvements to these dams illustrate some varied economical solutions to different problems. Investigation of the concrete in the main dam included determination of concrete strength, density, modulus of elasticity, and overall quality as determined by sonic velocity testing. Coring was done to determine depth of cracks and deterioration as well as the bond between lifts. Chemical analyses of reservoir and leakage water were made. Recording thermometers were installed in the dam to determine seasonal concrete temperatures. Stress/strain gages were applied to the surfaces to record these values for comparison to water loading and temperatures. Stress analyses were conducted using both two-and three-dimensional finite element methods of analysis. Various input parameters for concrete and foundation properties were used. For dams 2 and 3 conventional static stability analyses were made. Dam 3 was found to be marginally stable. Dam 2 was stable for existing loads but required anchors to accommodate loads resulting from increased flood loading. The main dam was improved by constructing a 12-inch-thick reinforced concrete membrane over most of the upstream face after the old deteriorated concrete was removed. Vertical drains were placed between new and old concrete at the vertical joints. This membrane reduced leakage and prevented further deterioration of old concrete. The dam crest was raised slightly to increase the spillway capacity at dam 2. At floods greater than the 1:500-year occurrence level the main dam will overtop, so protection for downstream appurtenances was provided. Two radial gates were added at dam 2 to increase spillway capacity. Posttensioned anchors were installed to improve stability under the increased water level conditions. An epoxy coating was applied to the concrete to prevent further deterioration. Dam 3 was partially reconstructed with a trippable flashboard type spillway at its lower end. In its higher reaches, where overtopping could
CONCRETE AND MASONRY DAMS 211 not be tolerated, the crest was raised slightly. A reinforced concrete membrane was constructed on the entire upstream face, and rockfill was placed against both upstream and downstream sides for part of their height, in order to increase stability. This is an example of a fully integrated approach to resolve deterioration, stability, and spillway capacity problems. REFERENCES Abraham, T. J., and Lundin, L. W. (1976) T. V.A. 's Design Practices and Experiences in Dam and Foundation Drainage Systems, Transactions, ICOLD. American Society of Civil Engineers (ASCE) (1982) Proceedings, Grouting in Geotechnical Engineering, W. H. Baker, ed., New Orleans. Chakrabarti, P., and Chopra, A. K. (1973) ''Earthquake Analysis of Gravity Dams Including Hydrodynamic Interaction,'' International Journal of Earthquake Engineering and Structural Dynamics, Vol. 2, No. 2 (October-December), pp. 143-160. Chopra, A. K., and Chakrabarti, P. (1973) "The Koyna Earthquake and the Damage to Koyna Dam," Bulletin of the Seismological Society of America, Vol. 63, No. 2, pp. 381-397. Chopra, A. K., and Corns, C. F. (1979) Dynamic Method for Earthquake Resistant Design and Safety Evaluation of Concrete Gravity Dams, Transactions of ICOLD Congress, New Delhi. deSousa, S. A. (1973) Rehabilitation of an Old Concrete Dam, Proceedings of Engineering Foundation Conference on Inspection, Maintenance and Rehabilitation of Old Dams, Pacific Grove, California. Goodman, R. E., Amadei, B., and Sitar, N. (in press) Analysis of Uplift Pressure in a Crack Below a Dam, paper given at ASCE Annual Convention, New Orleans. Pirtz, D., Strassburger, A. G., and Mielenz, R. C. (1970) "Investigation of Deteriorated Concrete Arch Dam," American Society of Civil Engineers, Power Division Journal, January. Seed, H. B., Lee, K. L., Idriss, I. M., and Makdisi, F. (1973) Analysis of the Slides in the San Fernando Dams During the Earthquake of February 9, 1971, Earthquake Engineering Research Center, Report No. EERC 73-2, University of California at Berkeley. U.S. Army Corps of Engineers (1958-1960) Gravity Dam Design, EM 1110-2-2200, Government Printing Office, Washington, D.C. U.S. Bureau of Reclamation (1976) Design of Gravity Dams, Design Manual for Concrete Gravity Dams, Government Printing Office, Washington, D.C. U.S. Bureau of Reclamation (1977) Design of Arch Dams, Design Manual for Concrete Arch Dams, Government Printing Office, Washington, D.C. U.S. Committee on Large Dams (1975) Lessons from Dam Incidents, USA , ASCE, New York. Vallino, G., and Forgano, G. (1982) Design Criteria for Improvement of the Concrete Buttresses of Corbara Dam, Transactions of 14th Congress, Rio de Janeiro, ICOLD . Recommended Reading Chopra, A. K. (1970) "Earthquake Response of Concrete Gravity Dams," Journal of the Engineering Mechanics Division, ASCE, Vol. 96, No. EM-4 (August), pp. 443-454. Dungar, R., and Severe, R. T. (1968) Dynamic Analysis of Arch Dams , Paper No. 7, Symposium on Arch Dams, Institution of Civil Engineers.
CONCRETE AND MASONRY DAMS 212 Golze, A. R., et al. (1977) Handbook of Dam Engineering, Van Nostrand Reinhold, New York. Howell, C. H., and Jaquith, A. C. (1928) "Analysis of Arch Dams by Trial-Load Method," ASCE Conference Proceedings. International Commission on Large Dams (1970) Proceedings, Xth Congress, Montreal, Recent Developments in the Design and Construction of Concrete Dams. International Commission on Large Dams (1979) Proceedings, XIII Congress, New Delhi, "Deterioration or Failures of Dams." Jansen, R. B. (1980) Dams and Public Safety, U.S. Department of the Interior, Denver, Colo. Proceedings of the Engineering Foundation Conference (1973) Inspection, Maintenance and Rehabilitation of Old Dams, Pacific Grove, Calif. Proceedings of the Engineering Foundation Conference (1974) Foundations for Dams, Pacific Grove, Calif. Proceedings of the Engineering Foundation Conference (1976) The Evaluation of Dam Safety, Pacific Grove, Calif. Severn, R. T. (1976) "The Aseismic Design of Concrete Dams," Water Power and Dam Construction, pp. 37-38 (January), pp. 41-46 (February). Structural Engineers Association of California (1975) Recommended Lateral Force Requirements and Commentary, San Francisco. Thomas, H. H. (1976) The Engineering of Large Dams, Vol. I, John Wiley & Sons, New York. U.S. Bureau of Reclamation (1977) Design of Small Dams, Government Printing Office, Washington, D.C. Westergaard, H. M. (1933) "Water Pressure on Dams During Earthquakes," Transactions ASCE, Vol. 98.
