صفحات

۱۳۸۶ خرداد ۱۳, یکشنبه

cradle of invention


A new method of
conveying cable stays through bridge pylons without permitting them to interact
with one another is giving rise to bridges with longer spans and more inspiring
designs that nonetheless respect budget constraints as well as maintenance and
life-span requirements.
By W. Denney Pate, P.E., M.ASCE, and W. Jay
Rohleder, Jr., P.E., S.E., M.ASCE
,
Photos: FIGG,
all



The Penobscot Narrows Bridge and Observatory—which
houses the tallest public bridge observatory in the world in one of its
pylons—carries U.S. Route 1 along the Maine coastline. The cradles for this
bridge were positioned within framework to secure them in their exact positions
on the ground; the framework was then lifted into place within each
cast-in-place pylon.




t is said that necessity is the mother of all
invention. If that is the case, then the need for striking, elegant cable-stayed
bridges that can span greater lengths and be more easily


ادامه مطلب ......................................................


 


 



A new method of conveying cable stays through bridge pylons without permitting them to interact with one another is giving rise to bridges with longer spans and more inspiring designs that nonetheless respect budget constraints as well as maintenance and life-span requirements.
By W. Denney Pate, P.E., M.ASCE, and W. Jay Rohleder, Jr., P.E., S.E., M.ASCE,
Photos: FIGG, all



The Penobscot Narrows Bridge and Observatory—which houses the tallest public bridge observatory in the world in one of its pylons—carries U.S. Route 1 along the Maine coastline. The cradles for this bridge were positioned within framework to secure them in their exact positions on the ground; the framework was then lifted into place within each cast-in-place pylon.




t is said that necessity is the mother of all invention. If that is the case, then the need for striking, elegant cable-stayed bridges that can span greater lengths and be more easily maintained over the course of longer lives has yielded a significant new invention that may well benefit bridge designers the world over.


Driven by a desire to push the design of single-plane cable-stayed bridges beyond their current limits—both in span length and in aesthetic appeal—while still delivering an economic solution that is easy to construct and maintain, FIGG, an engineering firm based in Tallahassee, Florida, has created a novel system for routing stay cables from one end of a bridge deck, through the bridge’s pylon, and then down to the other end of the deck in a way that precludes the possibility of cable-to-cable interactions.


The innovation has been employed on two new bridges—the Veterans’ Glass City Skyway Bridge, which carries Interstate 280 across the Maumee River in Toledo, Ohio (see “Ohio dot Endorses Design for Maumee River Crossing,” Civil Engineering, September 2000, page 12), and the Penobscot Narrows Bridge and Observatory, which carries U.S. Route 1 over the Penobscot River in southeastern (“down east”) Maine near the coastline (see “Observatory to Cap Maine Crossing,” Civil Engineering, April 2004, pages 15–17).


The cradle offers benefits both during construction and over the life of the bridges. Most important of all, it permits the use of the largest number of strands within a single stay cable in the world: 156, an increase of more than 70 percent over the second-highest number known to have been used in the United States. It also makes it possible to increase the distance between stay cables by approximately 50 percent when used with precast delta frames to facilitate a single plane of stay cables, resulting in aesthetically superior designs. This is accomplished by having all the strands parallel as they travel from the anchors at the deck level through the cradle in the pylon and back to the deck. Individual sleeve pipes located within the cradle system enable each strand to act independently of adjacent strands. This also permits the cables to be much larger and to be spaced farther apart, which translates into longer spans.


Furthermore, the cradle system lowers initial costs by reducing the amount of materials and labor needed because no anchorages in the pylon are required. This simplifies construction operations by allowing all of the cable-stressing operations to be performed at the bridge deck level rather than within the (often restricted) confines of the pylon, high above the bridge deck. The system also includes removable “reference” strands in each stay cable that provide a simple, reliable method for verifying the condition of the stay cables in the future. Because the cradle system does not require strands to be grouted, they can be individually removed, inspected, and replaced, even when there is traffic on the bridge. Bridge owners can thus safely and accurately assess the conditions of the stay cables at any time over the course of the bridge’s life. In the case of the Maine bridge, new strand materials can be tested side-by-side with traditional materials in an actual setting, as opposed to a process of computer simulation.


What is more, the cradle system makes possible the use of a wider variety of pylon designs—some with much smaller and more elegant cross-sectional shapes—that can be constructed more economically. Engineers can thus design pylons that are far more unusual and aesthetically pleasing than has been the case in the past.


The cradle design works with the natural flow of forces because the forces transmitted through the cradle naturally compress the pylon in an efficient manner, the stresses being applied radially along the curve of the cradle (see the illustrations). In traditional systems, anchorages within the pylon required large tension ties to resist the high splitting forces that would be generated. Using the cradle system eliminates this requirement and further enhances the elegance of the pylon shapes.


he development of this new technique began in early 2000. The Ohio Department of Transportation (ODOT) and the Toledo Metropolitan Area Council of Governments had formed what was called the Maumee River Crossing Task Force Design Committee to assess and communicate the communities’ perspective on the developing design. The task force chose glass as the theme for the new cable-stayed bridge. Meetings involving a diverse cross section of the community had made the public’s views clear: the bridge was to feature glass in a very visible and striking way in recognition of Toledo’s industrial heritage as a leader in the glass industry. Many families in the area had worked in the local glass industry for generations, and it was important to them that the crossing be a symbol of the “Glass City,” as Toledo has come to be known. The public was also of the opinion that the bridge design should champion a product that had been of the utmost importance to the area’s economy. The task force further determined that residents wanted the bridge design to be light, simple, and elegant.


FIGG led two community workshops, or design charettes, during which the public was presented with a variety of aesthetic options. Community voting showed a clear preference for a single-pylon design. The consensus was that by creating a single tall pylon—403 ft (123 m)—using glass on the top 196 ft (60 m) and on all four sides, the new bridge would be visually stunning. In keeping with this directive, the top of the pylon takes on a prismatic shape, its panels of treated glass reflecting sunlight during the day on all four sides. Behind the glass are light-emitting diodes (LEDS) that enable the pylon to stand as a beacon at night. (The LED fixtures are controlled remotely and capable of literally millions of color combinations. In fact, various color schemes have been preprogrammed, some schemes marking major holidays and others exhibiting team colors for statewide sporting events.)


The ODOT had determined that the bridge was to have a precast segmental superstructure and that its cable-stayed main span was to provide 120 ft (36.5 m) of vertical clearance and 400 ft (122 m) of horizontal clearance for the shipping channel. Since that channel runs along the north side of the river, the pylon could be placed in the center without disrupting shipping operations. The pylon, which rises 440 ft (134 m) above the river, is now the second- tallest structure in Toledo, just slightly shorter than a nearby landmark skyscraper that for many years served as the headquarters of Owens-Illinois, a glass and plastics company.


The community also voted by a three-to-one margin for a single plane of stay cables, reflecting their desire for a structure that would be visually arresting but would not unduly interfere with views of the river and would clearly separate northbound and southbound traffic on the highway itself. Additional voting revealed a preference for a fan arrangement in the stay cables, which would draw the eye upward, dovetailing nicely with the theme adopted by the task force for the project: “Look up, Toledo!” The use of stainless steel sheathing on the stay cables—which would reflect daylight in much the same way as the top of the pylon—also was valued. Delivering a design to achieve this vision meant using large, widely spaced stay cables. As the design developed, FIGG also focused on ensuring that the design would be efficient and that the bridge would have a service life of more than 125 years.


Some of the cable-stayed bridges in the Untied States designed by FIGG have featured saddles in their pylons to carry the stay cables. Examples include Florida’s Sunshine Skyway Bridge (a maximum of 82 strands per stay cable), which crosses Tampa Bay (see “Landmarks in American Civil Engineering History: Sunshine Skyway Bridge,” Civil Engineering, November/December 2002, pages 162–163); the Varina-Enon Bridge, near Richmond, Virginia (a maximum of 90 strands per stay cable); and the Chesapeake and Delaware Canal Bridge, near St. Georges, Delaware (a maximum of 85 strands per stay cable). However, given the record number of strands used in the cables of the Toledo bridge—156—and the fact that it might be necessary to use even more strands in future designs, a new system was clearly needed. Consideration was given to placing anchorages in the pylon, but this would have meant increasing the pylon cross section by one third—from a width of 23 ft (7 m) to one of 31 ft (9.5 m)—which in turn would have increased the overall quantity of materials and the cost of construction. Such an increase would also have complicated construction, because the contractors would have been required to perform portions of their stressing operations approximately 230 ft (70 m) above the bridge deck.







Twenty cradles convey the largest known stay cables in the world through a single pylon on the Veterans’ Glass City Skyway Bridge, in Toledo, Ohio. With cables arranged in a single plane, the elegant bridge pays tribute to a material that has contributed significantly to the economic vitality of the region: glass.

In analyzing the increased stay cable size with the traditional use of a saddle, the engineers determined that by isolating the strands from one another, strand-to-strand interaction could be eliminated. In early 2000 FIGG engineers introduced its new solution, which provided each strand with its own curved stainless steel tube within a steel cradle. Here each strand could pass through the cradle independently. Avoiding the use of anchorages in the pylon made it possible for the design to reflect the communities’ preferences and produce a structure with a sleek and angular single pylon, a single plane of widely spaced stay cables, and extensive use of glass in the pylon.


Each epoxy-coated steel strand runs independently through its own 1 in. (25 mm) diameter sleeve pipe within the cradle. The spaces between the sleeve pipes are grouted, under controlled conditions, before the cradle is set into place inside the pylon formwork. The cradle will then be cast into the pylon as one piece. “Cheese plates” (named for the distinctive holes of Swiss cheese) located at each end of the cradle and centering plates located in the curved section of the cradle are used to maintain the relative positions of the sleeve pipes. The ends of the individual sleeve pipes within the cradle are widened to ease strand installation. Along the free length of the stay cables—that is, from the edge of the cradle to the anchorage at the bridge deck—each of the strands is housed only within the sheathing for the stay cables. The strands are kept parallel to one another by the cheese plates and the anchorages.


In the Toledo bridge, the cradle extends beyond the pylon in the longitudinal direction. Since the bridge design utilizes 18 in. (457 mm) and 20 in. (508 mm) diameter stainless steel sheathing (believed to be the largest in the world), the cradle itself also was manufactured from stainless steel for a unified look.


efore this new system could be used, however, it had to be tested by the ODOT. To complete the test, the department prepurchased the entire system of stay cables, including the saddle, in early 2001. This removed testing from the contractor’s schedule and assured the owner that the cable system components would be available as needed. Acceptance testing included the following:




  • An axial fatigue and ultimate static test was carried out on an 82-strand specimen fully representative of all materials, details, fabrication processes, and assembly procedures proposed for the production anchorages. Each specimen consisted of two anchorages and had a clear space of approximately 180 in. (4,572 mm) between anchor faces.


  • An axial fatigue and leak test was carried out on a 119-strand specimen. As part of the corrosion protection qualification process for the anchorage assembly, the anchorage specimen for the stay cables was tested as a system to detect any leaks. The tested specimen included the entire transition zone; a minimum of 3.5 ft (1 m) of free length; and all seals, coatings, and coverings that were to be installed in the actual application.


  • An axial fatigue and ultimate static test was carried out on a 156-strand specimen that was fully representative of all materials, details, fabrication processes, and assembly procedures proposed for the production anchorages. Each specimen consisted of two anchorages and featured a clear space of approximately 180 in. (4,572 mm) between anchor faces.


  • Single-strand cradle testing also was performed. Before conducting a test of the full-size cradle specimen for combined axial/flexural fatigue, three similar tests were conducted on single-strand specimens, each curved to a different radius through the cable. The purpose of these single-strand tests was twofold. First, the test provided a value for the friction coefficient between the epoxy-coated strand of the stay cable and the stainless steel sleeve within the cradle. Second, it provided an initial indication of the fatigue behavior to be expected from the interaction between the epoxy-coated strand of the stay cable and the stainless steel sleeve.


  • An axial/flexural test, or “cradle” test, was carried out on a 119-strand specimen that, again, fully represented all materials and processes to be used on the actual bridge. The specimen consisted of two anchorages and one complete cradle assembly for the stay cables.
    The acceptance testing was performed by ctl Group, of Skokie, Illinois, in accordance with the reference work Recommendations for Stay Cable Design, Testing and Installation (Phoenix: Post-Tensioning Institute, 1993). The leak test was adapted from the 2000 edition of this publication and completed successfully in December 2001. The construction work for the Toledo bridge was put out for bidding on January 15, 2002, and the low bid of $220 million was offered by the Fru-Con Construction Corporation, of St. Louis. Construction of the 2.75 mi (4.4 km) of ramps, roadways, and cable-stayed crossing is nearing completion, and opening is anticipated this spring.


s construction was proceeding in Toledo, the Maine Department of Transportation (MaineDOT) was challenged by the need to effect an emergency replacement of the bridge that carries U.S. Route 1 over the Penobscot River to link Waldo and Hancock counties. The existing suspension bridge was scheduled for renovation, but when deterioration of the main suspension cables was found to be much more pronounced than anticipated, a decision was made in July 2003 to replace the crossing with a cable-stayed bridge. FIGG was selected to design the bridge, which was constructed by a joint venture of Cianbro/Reed & Reed, LLC—which operated from a location adjacent to the bridge in Verona—at a cost of $68 million. The pressing nature of the project and an owner-facilitated design/build process moved the effort along rapidly and made it possible for the new bridge to open on December 30, 2006, just 40 months after the decision was made to replace the structure.











Figure 1
Figure 2


Figure 3                                                                                     
The strands run continuously from one end of the bridge deck, through the curved cradle that has been placed within the pylon, and back to the other end of the bridge deck (figure 1). Centering plates and “cheese plates” (figure 2) are located at each end of the cradle to position the individual stainless steel pipes or sleeves parallel to one another within the cradle (figure 3). Grout is used between the individual sleeves. The ends of the individual sleeves were flared to prevent the sleeve edges from coming into contact with the epoxy coating on the strands (figure 4).Figure 4

Those living in nearby communities wanted the replacement to complement its context. The bridge is located adjacent to a historically important military establishment, Fort Knox, the first military post in Maine to have been constructed of granite rather than wood. General Henry Knox, for whom the fort is named, served with distinction as President Washington’s commander of artillery and was the country’s first secretary of war. Adding to the primacy of granite is the fact that it was used in the core of the Washington Monument. The communities thus feel particularly connected to the material, as seen in the theme they developed for the bridge project: “Granite—simple and elegant.” The bridge was therefore designed to be constructed of cast-in-place concrete formed in lifts that in size and shape would suggest large blocks that blend with the rugged rock strata of the region. Additionally, granite was used in the bases of the pylons, and the pylons took the form of obelisks in a nod to the Washington Monument.







In 2005, as the pylon of the bridge in Toledo, Ohio, was cast, each cradle was lifted into place by crane and precisely positioned to meet the required geometric specifications.

To streamline environmental approvals and deliver the completed bridge as quickly as possible, the pylon foundations for the 1,161 ft (354 m) cable-stayed main span were placed on the riverbanks. As the design was being developed, the communities voiced misgivings regarding the pylons, which were to be more than 400 ft (122 m) tall to accommodate the main span. In response, the design team turned that height into an advantage by creating a three-story public observatory atop one pylon—the tallest public bridge observatory in the world—with views of both Fort Knox and the Maine coast.


The state’s heightened awareness of cable corrosion LED MaineDOT to explore all possible methods for minimizing the potential for corrosion in the cable-stayed bridge. Multiple layers of protection for the strands were provided, including epoxy coating on the strands, a pressurized nitrogen gas system that creates a corrosion-inhibiting environment in each stay cable, reservoirs within each stay to automatically recharge the gas system to 2 psi (13.8 kPa), readily accessible monitors for reading and recording fluctuations in gas pressure, and exterior sheathing that works with the cradle system to create an airtight environment. MaineDOT and the participants in the design workshops preferred light gray high-density polyethylene (HDPE) for the stay cable sheathing, in keeping with the aesthetic theme of granite. Because of its flexural characteristics, HDPE also provided an ideal material for creating airtight closures at the cradles and anchorages.


To accommodate the contractor’s preferred means and methods, the cradles were slightly inset from the pylon faces. With each cradle wholly within the confines of the pylons, it was economical to specify unpainted, uncoated carbon steel (“black steel”) for cradle fabrication.


One back span of the bridge curves in two directions, which created geometric challenges during construction. The stay cables originate in the centerline of the bridge deck and are conveyed through alternating sides of the pylons. The result is a unique V shape in the system of stay cables. This provides an opening in the center of one pylon to accommodate an elevator that takes visitors to the public observatory at the top. Constructability was enhanced by positioning the cradles within a frame during fabrication at ground level, surveying their position to a high degree of accuracy, and then securing the cradles to the frame. Once the framework was correctly positioned in the pylon formwork, the geometry for the side-by-side cradles was verified and the entire system cast into the pylon.


The flexibility created by enabling each strand to act independently allowed the incorporation of a force-monitoring system on each of the 40 stay cables in the bridge, a system that is believed to be the world’s first. By using a portable device, the owner can measure the forces in each stay cable to an accuracy of 1 percent at any given time. Regularly recording the stay cable forces and comparing them with the predicted values allows MaineDOT to easily assess the condition of the stay cables and their strands without additional expense, special equipment, or interruption to traffic.


As in the Veterans’ Glass City Skyway Bridge, a reference strand can be isolated for research and analytical purposes. Because of this, the designers were able to assist both MaineDOT and the Federal Highway Administration in their goal of developing new materials for bridges by replacing two strands in one short, one medium, and one long stay cable with carbon fiber composite strands. The strands will be monitored and inspected over time in this real-world service environment to evaluate the material for future use in cable-stayed bridge designs across America.


Necessity may have led to this cradle of innovation, but the benefits and advantages it confers have gone far beyond meeting those initial requirements. The cable-stayed cradle provides a new level of flexibility in the design of long-span bridges. By reducing costs and making long-term maintenance of stay cables simpler, this technology enables bridges to achieve a longer life at a lower initial cost. For Toledo and such communities as those in downeast Maine, this means distinctive bridges that not only serve their purposes but also enhance their communities.





W. Denney Pate, P.E., M.ASCE, is a senior vice president and principal bridge engineer for FIGG, of Tallahassee, Florida. W. Jay Rohleder, Jr., P.E., S.E., M.ASCE, is a senior vice president of project development for the firm and works from its Exton, Pennsylvania, office.

  • Digg
  • Sphinn
  • del.icio.us
  • Facebook
  • Google
  • Furl
  • Reddit
  • StumbleUpon
  • Donbaleh
  • Technorati
  • Balatarin
  • twitthis

4 نظرات:

بیننده گفت...

من یک نظر فنی بدهم.... تجربه 8 سال به من می گه مثل خاله زری عمل کردن بهتره....... موفقیتت صد چندانه......... نظر شما چیه اقا مهدی؟؟؟؟
بعد شما که راهنمایی خواستید برای وبلاگتان ..... به نظر من شاخه بندی اگر نمایید بهتر است ..... بخش اجتماعی بخش زنان بخش علمی بخش ادبیات و زبان شناسی بخش زبان بخش عمران و سازندگی انچه که بابایی ...... علاقمند است.... بخش کامیوتر بخش پزشکی........ نظر شما اقا مهدی چیه؟؟؟؟

نظر بده گفت...

سلام خسته نباشی....... واقعا نیاز مادر اختراع است مخصوصا اگر با ترس شدید که شبیه شوک باشد هر چی بخواهید می شود... البته یک کم بیشترش زهره ترک شدنه........... راستی چرا حرف های ما در سایتتان درج نمی شود... چه بهتر بود که با طرفداران ر و پا قرصتان مکاتبه یا ایمیل می کردید.....

ارش گفت...

سلام...وبلاگ خوبی داری. از اینکه به وبلاگ ما سر زدید. با تبادل لینک موافقم.لطفا بگویید نام لینکتان را در وبلاگم چی بذارم؟ ممنون

رایحه گفت...

سلام
وبلاگتون قشنگ بود اها علمی هم بود من هم توی وبلاگم در مورد نانو مطلب مینویسم خوشخال می شم همدیگرو لینک کنیم
7

ارسال یک نظر

 
ساخت سال 1388 مهندسی عمران - سازه شرقی.قدرت گرفته با بلاگر تبدیل شده به سیستم بلاگر توسط Deluxe Templates. طراحی شده بوسیله Masterplan. . بهینه شده برای سیستم فارسی مجتبی ستوده