FOUNDATION DESIGN

Recent years have seen the development of unusual features at civil works projects in the design of foundations for major hydraulic structures. Principal among these were the use of anchors to reinforce the foundation, the design of major drainage facilities, and the initiation of an extensive program for evaluating the shear strength of relatively weak foundation rock.

Foundation anchors.-Relatively short steel bars have been used extensively in the past to anchor spillway chute and stilling basin slabs to foundation rock to resist hydrostatic uplift forces. However, the recent use of long bars for reinforcing structural foundations was an important innovation at civil works projects. At one project such anchors were used in controlling differential rebound in a faulted, jointed soft rock. The differential rebound became evident upon completion of rock excavation. If this phenomenon had continued after construction of the structures, severe structural damage would have occurred. The design utilized long anchor bars grouted into holes drilled deep into the foundation and bonded at their tops to the concrete structure. The anchors retarded expansion of the top foundation layers and have effectively prevented differential movements along fault planes. This solution to the foundation problem was much more economical then the alternative of providing sufficient concrete weight (which would also require additional excavation) to restrict the foundation movements. In the future more frequent use of anchors can be expected to solve structural stability problems where otherwise costly excavation and concrete quantities would be needed.

Drainage. A recent change in design criteria for hydraulic structures will provide greater construction savings. The change is the result of measurements made at existing structures and provides for increased reliance on foundation drainage to reduce hydrostatic uplift pressures against the structures, which results in lighter, more inexpensive structures. Foundation drainage has also been used to reduce hydrostatic buckling pressures around steel liners in tunnels and to improve the stability of earth dam abutments.

Foundation shear evaluation.-A program is presently being prosecuted to develop more reliable criteria for determining the true in situ shear strength of relatively weak rock foundations.

CONCRETE EMBEDDED PENSTOCKS

Design methods for penstocks embedded in concrete have not been well established. Designs have varied for different projects with art progressing as experience is accumulated. To obtain the economical advantage of this type of composite construction, the corps developed applicable design methods which involve consideration of both internal pressures (distributed to the steel and reinforced concrete backing) and external buckling pressures. It was estimated that this composite construction in lieu of penstocks standing free in the concrete tunnel saved over $10 million at one project.

DEVELOPMENT OF IMPROVED DESIGN CONCEPTS

Flood control channels.-An extensive study, which was initiated in 1959 and scheduled for completion during 1963, is being conducted to determine optimum reinforcement and concrete slab joint arrangements for the control of cracking and reduction in deterioration of reinforced concrete in high-velocity flood control channels. Some of the results to date have been used to make improved economical designs, and at the conclusion of this study, new criteria will be distributed for application at all pertinent projects.

Lock design. In the design of large U-shaped drydock-type locks certain assumptions must be made concerning the distribution of foundation pressure beneath the structure, or otherwise the design is based upon a theoretical analysis of support from an elastic foundation. An instrumentation program on two lock structures is currently being conducted to provide engineering data for the future design of similar structures. Pressure cells, stress meters, and other devices were incorporated into the construction to measure foundation pressures, compressive stresses, settlements, and deflections.

Reinforcing steel working stresses. Another investigation, which concerns the possibility of increasing working stresses for concrete reinforcing steel, is being conducted at the Treat Island, Maine, Exposure Station. It consists of concrete beam exposure tests which were initiated in 1951. The purpose of the test is to determine the effect of tensile cracking upon the durability of the concrete beams. Steel stresses were varied over a wide range so as to obtain a comparison between beams with different amounts of tensile cracking. Durability results to date have led to an increase in allowable working stresses for hydraulic structures from 18,000 to 20,000 pounds per square inch. Upon completion of the tests, it is hoped that results will be conclusive with respect to the advisability of increasing working stresses for rail or high-strength reinforcing steel.

Reinforced concrete. The Reinforced Concrete Research Council maintains a program of sponsorship and financial support to research projects which promise to develop answers to reinforced_concrete problems leading to improved practice. The Corps of Engineers maintains an active participation in the program through membership on the council. The results of sponsored projects are used to improve design criteria and concepts in our civil works construction.

Revised AISC specification. The new revised AISC specification will be applied to the design of structural steel for permanent buildings. Applicable parts such as the more exact column formulas and formulas for combined stresses will be used in designing other steel structures such as gates and bulkheads. Published corps design manuals will be revised to incorporate the applicable new design rules of the AISC specification.

ELECTRONIC DATA PROCESSING

Where available, use is being made of electrical data processing devices to speed up routine computations relative to the stability computations of embankments and foundations of dams and levees, and foundation stability of floodwalls. Electronic data processing is also being used in solving a variety of other design problems.

CONSTRUCTION MATERIALS FOR CONCRETE

CEMENTING MATERIALS

General. Prior to 1930, with a few exceptions, portland cement was the only cementing material used in portland cement concrete in the United States. Also it was not until after 1930 that specifications were developed for different types of portland cement and the different types became generally available. Slag cement was first used in dam construction in a public utility dam in the early 1920's. Natural cement was first used in dam construction by the corps about 1946. Pozzuolanas were used as early as about 1927 or 1928 in Big Dalton Dam built by the Los Angeles Flood Control District.

Portland blast-furnace slag cement.-This material is the product obtained by intimately intergrinding a mixture of portland-cement clinker and granulated blast-furnace slag. It is not a new material having been used in Europe for sometime past, however, until about 1954 or 1955 there was only one producer in the United States. About 1954 when portland cement was in short supply several (five or six) other manufacturers started production of PBFS cement. The main purpose was to provide a greater amount of cement from the available portland-cement clinker. There was little precedent for use of PBFS in this country and no substantial basis for judging the quality or performance even though there was available a Federal specification promulgated in 1952. The corps undertook a comprehensive investigation of PBFS in 1955 which was completed in 1956. Tests were made on samples of PBFS from all known producers. The producers cooperated by furnishing samples of PBFS, portland-cement clinker and slag. Largely as a result of these investigations the Federal specification for PBFS was subsequently revised to provide the PBFS cement equivalent in behavior and properties to types I, IA, II, and IIA portland cement.

During the period 1954-58 when portland cement was in short supply the price of PBFS was the same as portland cement. After 1959 at least one producer was offering PBFS at 20 cents per barrel less than portland cement. Since 1959 when portland cement supply began to exceed demand several of the producers have discontinued regular production of PBFS but are presumably in position to produce PBFS when ordered. During the period 1954-58 there were six to eight mills regularly producing PBFS. At present there seems to be only one mill regularly producing PBFS. PBFS has been used in only one corps project in the last 5 or 6 years.

Pozzolan. Pozzolan is a silicious or silicious and aluminous material which alone possesses little or no cementitious value but which will, in finely divided form and in the presence of moisture, react with calcium hydroxide at ordinary temperatures to form compounds having cementitious properties. Pozzolan is one of the oldest, if not the oldest of cementing materials, having been used by the Romans as early as 12 A.D. and by the Greeks at even earlier dates. Pozzolan

has been in use in Europe in modern times and was used in the United States as early as 1927. Pozzolan was first used in corps work in the construction of Bonneville Dam (1937-42). At Bonneville a portland-pozzolan cement was used. It was not until 1957 however that the Corps of Engineers developed a satisfactory basis for widespread use of pozzolan. Specifications and criteria were developed from a comprehensive investigation of pozzolanic materials available in the United States which was started in 1951 and completed about 1957. Pozzolan has been used in more than 20 major corps structures started since 1957. Generally, pozzolan is used in mass concrete as a volume replacement for portland cement at a rate of as much as 30 percent.

Currently, investigations are underway to determine the maximum amount of pozzolan which can be technically and economically used. The investigation is being conducted using concrete made with 94 pounds (one bag) of portland cement and enough pozzolan to give required plasticity and strength. The investigation is not yet complete but some of the early results are encouraging. Usually concrete made with gravel coarse aggregate, natural sand and a 30-percent volume replacement of fly ash (an artificial pozzolan) requires about 135 to 145 pounds of mixing water per cubic yard for lean interior (2.00 to 2.25 bags of cementing material per cubic yard) concrete. In the investigation underway with this combination of materials, using 94 pounds of portland cement and as little as 100 pounds of fly ash workable concrete of ample strength for concrete dams and moderate height has been made with less than 85 pounds of water per cubic yard. The most abundantly available pozzolan in the eastern half of the United States is fly ash, an artificial pozzolan. Fly ash is the finely divided residue that results from the combustion of powdered coal which is transported from the combustion chamber by exhaust gases. Each thermal power station using pulverized coal fuel and electrostatic precipitation of the fly ash is a potential source. Essentially all of the fly ash used up to now on corps projects has been supplied from four plants.

Natural pozzolans are available from at least five commercial sources west of the Mississippi River. The five known sources are located in Oklahoma, Texas, Wyoming, Oregon, and California.

Price of fly ash has ranged from $3.75 to about $17 per ton in the concrete. The f.o.b. source price is understood to be about $2 per ton. Freight, unloading, storage, and reclaiming make up the major portion of the cost. Most of the fly ash used has cost between $6 and $12 per ton in the concrete. Fly ash has been used as far west as Ice Harbor Dam in the State of Washington. Material for this project was supplied from Chicago and the price in concrete was $15.25 per ton. Price of natural pozzolan produced in Portland, Oreg., which is being used at John Day Dam is $16 per ton. Available information indicates that natural pozzolans such as the calcined shales are priced at about $12 per ton f.o.b. source. As a basis of comparison with the price of pozzolans, portland cement costs from $22 to $28 per ton on mass concrete.

Slag and natural cements.-Neither of these materials are new, but not until 1954 and 1957, respectively, were there Federal specifications covering these materials. Both had been used by the corps before

these dates. The current Federal specifications for these materials were largely developed from the corps research on and experience with these materials.

Slag cement is a finely divided material consisting essentially of water-quenched granulated blast-furnace slag and hydrated lime. The corps has used slag cement as a 20-percent replacement by volume for portland cement in lean mass concrete.

Natural cement is finely pulverized calcined argillaceous limestone. The corps has used natural cement as a 25- to 30-percent replacement by volume for portland cement in lean mass concrete.

Slag cement is available from only one source (in Alabama) in the United States. Natural cement is available from three sources (in Indiana, Kansas, and New York) in the United States. Slag cement in large volumes has usually been priced at about $0.40 to $0.50 per barrel (1.91 cubic feet solid volume) less than portland cement. Natural cement has usually been priced at about $0.20 per barrel less than portland cement.

Since pozzolans have come into widespread use both these materials have lost their competitive position in mass concrete construction.

Trief cement. This is a cementing material made by a process patented in Belgium. The material may consist of either of the following:

(1) Sixty-eight to seventy percent pulverized granulated blastfurnace slag cement, 28 to 30 percent portland cement, and 2 percent sodium chloride.

(2) Ninety-seven percent pulverized granulated blast-furnace slag, 1.5 percent sodium chloride, and 1.5 percent sodium hydroxide. A principal feature of the process is the wet grinding and storage of the slag at the point of use. Another feature is the use of sodium chloride

(common salt) as a catalyst.

The combination described in (1) above resembles portland blastfurnace slag cement, but contains a greater proportion of slag than is allowed for PBFS.

The process has never been used in this country so no costs are available. It is understood also that it has had very limited use in Europe. It is unlikely that the process would be economical in our economy unless its use became widespread.

Ready-to-use readily available dry materials such as portland cement, slag and natural cements, and pozzolans would seem to have an economic advantage over the Trief process because they fit into the practices and plant of most contractors better than the Trief process material. In order to use the Trief process contractors would have to add materially to their capital investment in plant and make major modifications in their practices.

Supersulfated cement. This material is essentially granulated blastfurnace slag, calcium sulfate and a small amount of portland cement. It is a European cement which as far as is known is not available nor has it ever been used in the United States. The corps cooperated with the National Bureau of Standards in a limited investigation of this material several years ago. Strengths were comparable to strengths obtained with portland cements manufactured in the United States, but concrete made with this material demonstrated poor resistance in both laboratory freezing and thawing tests and in natural weathering tests.