Page images
PDF
EPUB
[graphic]

Figure 39.--View of excavation showing great variation of materials. ED-7-27, June 5, 1947.

[graphic]

Figure 40.--Placing fill on cofferdam. ED-7-87, May 8, 1947.

The

overlying, uncemented materials varied from impervious to semipervious in character and were quite irregular in occurrence. They were utilized in the impervious zone, the semipervious materials being placed in the outer edges of the zone. Soils of this type were also placed in the upper 5 feet of the dam embankment throughout all zones. rocky layers were excavated separately and were placed in the rockfill. In pit G-2, the rocky layer was comparatively thin; and after being loosened by a ripper, the material could be removed by a shovel. The rocky layer in pit A-1 was thicker and required drilling and blasting before removal. The sand in both pits was excavated partly by shovel and partly by scraper. Neither pit was preirrigated, the geography and character of the materials making this impracticable. The Ogallala material, however, mixed readily with water on the fill.

95. Placing Equipment. The adequacy of the embankment as well as the contractor's costs are affected greatly by the type of placing equipment. The principal units used at Enders Dam were as follows:

(1) Crawler-type tractors, approximately 120 horsepower. Six were acquired new in April 1947, all equipped with cable controls. A seventh was acquired new in September 1947 with no equipment. Normally, one of these tractors was used with the scraper; one equipped with push blocks was used in conjunction with the wheeltype tractor scraper units; two or three were used with rollers; and the remaining two or three were equipped with dozer blades. One of the dozer blades was equipped with a scarifier. Occasionally, when a scraper or roller was not in use, the motive tractor would be rigged with a dozer blade.

(2) Crawler-type tractor, approximately 40 horsepower, equipped with dozer blade. One was acquired in March 1948.

(3) Tamping roller, Bureau design, double drum. Two were acquired new in April 1947, and one was acquired new in September 1947.

(4) Tamping roller, single drum, Corps of Engineers designation "air-borne equipment, roller, towed, model 10-S." The roller was acquired in May 1948.

(5) Pneumatic roller, nine wheels with rubber tires. The roller was acquired in August 1948.

(6) Trucks, diesel powered, equipped with 2,000-gallon water tanks. Two were acquired in March 1947. One was disposed of in March 1948, and the other was used throughout the job.

(7) Soil manipulator. One was acquired new in October 1947. This device was custom made and was essentially a rigid-toothed harrow made so that its frame could be mounted by trunnion pins on the rear of a crawler-type tractor of approximately 120 horsepower and so that the entire assembly could be raised or lowered by means of a cable control. It did not interfere with operation of the dozer blade on the front of the tractor.

(8) Vibrator-type soil compactor, gasoline-electric powered. One was acquired new in June 1948.

(9) Rammer-type soil compactor, gasoline-powered. Two were acquired new in September 1948 and one was acquired new in September 1949.

96. Earthfill Placing and Compacting Methods. Earthfill placing methods on Enders Dam and dike, in general, followed standard patterns. The placing of fill on the cofferdam is shown in figure 40. Zone 1 and zone 2 earth materials were placed so that the fills would be free of lenses, pockets, streaks, or layers of material differing substantially from the surrounding material. No brush, roots, sod or other perishable or unsuitable material was allowed to be mixed with fill materials. Fill materials were not placed when the materials or the surfaces upon which they were to be placed were frozen. The soils were placed so that the permeability increased towards the outer slopes. To bond layers to each other, fill surfaces that were too dry or too smooth were moistened or scarified or both.

After being dumped on the fill by trucks, scraper, or scraper units, piles of loose material were spread in uniform lifts usually by dozers but sometimes by motor patrols. Six-inch compacted lifts were obtained by spreading loose lifts of impervious material to about 9 inches in thickness and of pervious material to about 8 inches in thickness. Before rolling, moisture was added to the loose lifts, if required, by one or two trucks mounted with 2,000-gallon water tanks equipped with rear-end gravity spreader bars. Moisture was mixed into the loose lifts by means of a rigid-tooth harrow built especially for construction work and designed to mount permanently on the rear of a dozer. The harrow could be raised off the ground when not in use and did not interfere with normal operation of the dozer blade. Results obtained by use of this piece of equipment, while not ideal, were satisfactory; and its flexibility of operation made it a valuable tool on the embankment. It was intended to place the material slightly dry of laboratory optimum. Actually, moisture content for zone 1 was 1.3 percent dry of laboratory optimum and that for zone 2 was 3.6 percent dry. Moisture was controlled in zone 1 according to instructions now contained in the Bureau's Earth Manual, tentative edition.

The three rollers used on the embankment proper were built according to standard Bureau design and were ballasted to a minimum of 10,000 pounds per linear foot of drum width. All impervious fill was compacted by 12 passes of the standard roller, and the pervious fill by about four passes. Suitable compaction of some of the pervious fill was achieved by the treads of a crawler-type tractor. None of the borrow areas contained oversize rock. Oversize rock was, however, encountered in material from the required excavation. In these cases, when the fragments would not break under the roller, the larger oversize rocks were either dozed or bladed onto the rockfills. Smaller fragments were usually soft enough to break up under roller action, and any rock pockets which did form were scattered by dragging with the harrow or by dozing.

Power tamping was necessary in locations where the earthfill could not be compacted properly by the use of rolling equipment. Initial power tamping operations were carried on by the use of pneumatic equipment. Both standard backfill tampers of two or three makes and paving breakers with tamping feet were used. The standard backfill tampers proved to be slow, and the required compaction was difficult to obtain. Until the middle of the 1948 construction season, most of the hand tamping was accomplished by the use of the heavy paving breakers with tamping feet.

About the middle of the 1948 construction season, the contractor purchased and put into use two portable gasoline-powered rammers. These machines weighed 210 pounds, were slightly over 4 feet high, and had a foot diameter of 9-1/2 inches. They did not depend on a high-speed, short-stroke actuation of the tamping foot; but rather, with each explosion of the self-contained gasoline engine they leaped about 14 inches into the air, and the impact of their fall provided compacting action. These new machines proved to be very effective power tampers when in proper operating condition. Faulty operation of the magnetos, however, caused difficulties and delays, and during the winter of 1948 the machines were returned to the manufacturer for remodeling, after which they were much more dependable. Through the 1949 construction season, these two machines plus a third which had been purchased were used to place all of the power-tamped embankment around the right and left spillway inlet walls. When operating properly, they were capable of imparting the required compaction to a 10-inch loose lift of fine sandy silt containing near optimum moisture with one pass.

To reduce the quantity of power tamping necessary, the contractor acquired a small, single-drum sheepsfoot roller. This roller was of the type used by the Army airborne engineering units during the war. It had one drum of 55-inch length and 44-inch diameter. The tamping feet were 7 inches long and of rectangular shape. The weight of the roller and ballast divided by the total area of all feet gave a unit pressure of 23.6 pounds per square inch. The contractor powered this unit with a crawler-type tractor of approximately 40 horsepower and used it in confined areas too small to permit the operation of a larger crawler-type tractor and a standard roller. With 18 passes, the unit was capable of imparting required compaction to a 6-inch loose lift of impervious material within moisture limits.

It was at times necessary to resort to puddling to obtain a tight bond between the embankments and rocky contacts on the abutments. A narrow ditch approximately 18 inches wide and 8 to 14 inches deep was maintained against the contact. Enough water was kept in the ditch to develop a plastic, oozy mud, which worked into all the irregularities in the rock when the roller operated along the contact and overlapped the ditch. A thorough puddling job was not easy to obtain and required the constant vigilance of the inspector. Shallow excavations a few feet deep made periodically down the contacts verified the effectiveness of the method.

97. Control Tests. Embankment control methods used at Enders Dam followed standard patterns. Moisture control in the field was based on the use of the Proctor needle, which proved to be a very satisfactory means of moisture control. Most of the embankment materials contained only a negligible quantity of plus 1/4-inch sizes, and the various sources of material were quite uniform within themselves. The field inspector was equipped with a field kit consisting of a 1/4-inch screen, a lightweight compaction cylinder, a standard hammer, a Proctor needle stock and set of needles, and auxiliary tools. The penetration resistance of the fill material was checked as it was dumped on the fill; and if necessary, the underlying layer was also checked. The values obtained were compared with moisture penetration-resistance limits furnished by the laboratory for the various sources of material. No standards were set to require any fixed number of moisture checks; and, after the inspectors gained experience, they were able to estimate how many checks they needed to maintain proper control. Moisture checks on impervious material, for which the Proctor needle was useless, were made by the laboratory either as a part of the density test procedure or as requested by the inspector.

The quality of the embankment in place was checked by field density tests run by the sand volume method and by subsequent laboratory testing of samples taken in conjunction with the field densities. In general, the standards now presented in the Bureau's Earth Manual, tentative edition, were followed as to the number of tests to be taken. A running graphical summary was kept of the field density test results. The graphical summary had the advantage of indicating trends in embankment control and was a decided help to inspection and laboratory personnel in early detection of any general variation in density results.

Except for tests made at special installations, all densities were taken in the second layer down, that is, in the layer immediately below the top lift which had just been rolled. This method was followed for a specific reason in preference to a procedure in which the test might be started at a lower point or excavated more deeply. Roller action is known to have an effect for a considerable distance below the surface of the fill, but its greatest effect is in the second and third layers. Therefore, if the second lift from the surface was required to be within the allowable density limits before additional lifts were placed, any further compaction obtained by rolling in succeeding lifts would be in the nature of a safety factor.

98. Rockfill. The first rockfill was placed in the rockfill zone, zone 3, on May 23, 1947. The rock was part of the Ogallala formation obtained from the outlet works excavation and later from the spillway excavation. It soon became evident that the character of the rocky material available precluded any possibility of maintaining the rockfill as an entirely free-draining zone. In order to secure any appreciable quantity of rockfill, it was necessary to utilize all of the rockier Ogallala material even though it contained comparatively large amounts of fines.

Late in the year the rockfill zone reached approximately elevation 3075, which was slightly higher than the adjacent earthfill. Through the winter, material obtained from the required excavation was stored in a stockpile on the downstream toe of the dam immediately adjacent to the embankment already in place, for later placement in the rockfill. A small amount of rockfill was placed in the summer of 1948 as it was excavated, but the rockfill zone lagged far behind the earthfill. In September, dozers began moving the stockpiled material into zone 3. After the November 1948 shutdown on earthfill, the 3-1/2 cubic-yard shovel and the bottom-dump trucks were used to place the remainder of the stockpiled material in zone 3. At this time, also, some rocky Ogallala material

excavated in the spillway approach area was placed; and by the middle of December when all work ceased, the rockfill zone was at an average elevation of 3112. The rockfill zone was completed in 1949, using material from borrow pit A-1 and material excavated from the spillway approach area.

All rockfill was placed by dozing it out in 3-foot layers. As placed, the zone 3 probably contained at least 30 to 50 percent of Ogallala fines smaller than one-half inch. Some of the bigger fragments were as large as one-half cubic yard, but probably not over 20 percent exceeded 2 cubic feet in size. The finer material was placed next to the

earthfill and the coarser material on the outer slopes.

99. Gravel Blanket. - Specifications No. 1410 required a 12-inch gravel blanket to be placed on the upstream face of the dam, beneath the riprap. The material was to be reasonably sound and reasonably well graded from 1/4 inch to 2-1/2 inches. No such gravel was available locally, and transportation costs for hauling gravel in from outside sources were high. The contractor located a ledge of some of the harder Ogallala formation on the right abutment a short distance upstream from the dam and requested that he be allowed to crush this material for use in the blanket. Samples were submitted to the Denver laboratories, and after they were tested, permission was granted by the Denver office to use the crushed Ogallala material for the upstream blanket with the provision that the crushed rock should be graded to a 3-1/2-inch maximum size rather than the 2-1/2-inch maximum given in the specifications for the gravel. The hard ledge which the contractor proposed to quarry was overlain by topsoil and softer Ogallala materials, and he began stripping operations early in April 1949.

The contractor obtained a hammer-mill crusher, aggregate hoppers, and a shaker screen assembly, and by using a section of conveyor belts and frame which he had on hand, assembled a crusher. By early July the plant was ready, and the first crushed Ogallala blanket material was placed on the upstream face of the dam on July 7, 1949. A schematic diagram of the crusher plant is shown on figure 41. Production and placing of the crushed Ogallala material continued until November 28, 1949, at which time the upstream blanket was completed. Some delays in the production of blanket material were caused by mechanical failure of the crushing plant, and other factors affected adversely the efficiency and cost of operation. As first constructed, the plant was not equipped with a pan feeder, and production was slow. At the suggestion of Bureau personnel, a pan feeder was installed and although it was a homemade device constructed from an old crawler-type tractor track, its use almost doubled production.

The crushed rock was dumped by truck first on the toe and later over the crest of the dam and spread on the slopes by dozers. Very little segregation resulted from this operation, probably less than would have been caused by any other method. However, when placing material down the slope from the crest, the material was moved diagonally across the slope rather than straight down as segregation occurred when the latter system was followed. A total of 11, 733 cubic yards of the crushed Ogallala materials was placed in the upstream blanket on the dam, which amount included about 1,000 cubic yards of rock hauled in from Ogallala, Nebr. In addition to the above volume, there were 276 cubic yards of the local material placed on the downstream slope of the dam next to the spillway left channel wall.

The quality of the crushed rock was good, although the waste in crushing was high. Surprisingly little breakage of the fragments took place under repeated tractor traffic. The revised specifications for gradation called for a reasonably well-graded material 1/4 inch to 3-1/2 inches in size. The gradation of the crushed rock fell within these specifications, except that some undersize was usually present. Normally there was not enough undersize to be harmful; and, as its presence resulted from insufficient screening rather than breakage, the amount could be reduced, if it became excessive, by reducing the rate of feed to the vibratory screen.

The percentage of waste in the Ogallala material that was quarried and crushed was seldom estimated to be less than 40 percent and was sometimes as high as 80 percent. This large variation in the amount of waste resulted from the irregular quality of the Ogallala formation.

« PreviousContinue »