Vulnerability Assessment of Arizona's Critical Infrastructure



|[pic] |10th International Conference on Short and Medium Span Bridges | |

| |Québec City, Québec, Canada, |[pic] |

| |July 31 – August 3, 2018 | |

DESIGN OF THE STEEL SUPERSTRUCTURE FOR THE LACHINE CANAL BRIDGE IN MONTRÉAL

Gao, Yulin1,2, Billah, Muntasir1, Schueller, Matthias1 and Taylor, Ryan1

1 Parsons, Canada

2 Yulin.Gao@

Abstract: The Lachine Canal Bridge is one of the signature bridges for the Turcot project in Montréal, Québec. The bridge is a five-span continuous steel girder bridge with a total length of 350 m. It carries four traffic lanes each carriageway with a total deck width about 45 m. The structural system consists of six steel box girders with a composite concrete deck. There is a center cable stay system with a single pylon at pier 3 to provide additional support to the bridge. This paper will present the design of steel box girders, which have vertical webs and a varying girder depth. The slenderest girder has a span-to-depth ratio of 44. It will discuss the challenges encountered in the design and the decisions made to allow the successful delivery of the bridge design. The consideration of construction staging and the effect on the bridge design will be described. The aspects for the design of steel cross frame and steel bearing stool are also included.

1. INTRODUCTION

The Turcot Interchange is a hub for road traffic in the Montréal area. The Turcot project is the largest in history for the Ministère des Transports de la Mobilité durable et de l’Électrification du transport du Québec (MTMDET) with a budget at $3.7 billion (CDN). In December 2014, MTQ awarded the Turcot Interchange Project to KPH Turcot, a joint venture between Kiewit, Parsons and CRH Canada. This design-build contract is the largest portion of the program at $1.58 billion. The existing interchange was commissioned in 1967. The reconstruction work of the Turcot Project will include rebuilding four interchanges, which include 56 structures with a total deck area of 167,520 square meters. The project provides for rebuilding of the new interchange structures essentially underneath or parallel to the existing bridges. The Lachine Canal Bridge is the signature bridge for the Turcot Project. The concept development of the bridge is the result of a rigorous design process in an accelerated Design-Build-Project environment. This paper will describe the design of steel superstructure(s) for the bridge, which includes steel box girder, steel cross frame, and steel bearing stools.

2. BRIDGE CONCEPTUAL DESIGN

The proposed Lachine Canal Bridge must be built underneath the existing ramp bridge in the Turcot Interchange. The site conditions, including existing roads and the Lachine Canal, and the redevelopment plans requires a span arrangement for the bridge at (55+88+80+73+65) meters from west to east along its control line. To accommodate the existing traffic and minimize the impact on traffic, the bridge needs to be built in two halves, namely the north carriageway carrying westbound traffic (“F” bridge) and the south carriageway carrying eastbound traffic (“E” bridge) respectively. A physical gap about 6 m is provided between the two carriageways to allow the natural light to flow into the space below the bridge. Light and rain will allow vegetation to grow and help to rebuild the natural habitat under this wide bridge [Schueller, 2017].

The reference design included in the MTMDET’s Request For Proposal (RFP) documents specified that the new bridge over the Lachine Canal shall be designed as a signature cable-stayed bridge with a single legged pylon next to the Canal and a constant depth superstructure type. It is understood that the bridge site is in a former industrial neighborhood that will undergo significant urban improvement in the next decade.

The new bridge is located in a curved road alignment to suit the road geometry requirement with a 6% constant cross slope. The vertical grade is constrained by the clearance above the Canal at bottom and the clearance under the existing bridge above. These site conditions put a limitation on the available girder depth.

The following challenges and considerations have to be undertaken for the conceptual design of the bridge:

1. Meeting Owner’s specific requirement regarding to the cable-stayed span: MTMDET specified a signature bridge with the cable-stayed span next to the Canal. The new bridge is designed as an extradosed bridge, a system that equips a traditional girder superstructure type with an integrated stay system to reduce the negative moment demands in the primary girder system [Mermigas, 2008]. The distinguished feature of an extradosed bridge directly influenced the design of the Lachine Canal Bridge because it allows the engineer to select the degree of which the stay system participates in carrying loads. The stay system was designed to carry loads that match maximum serviceability limit state (SLS) live loads in the cable supported spans [Schueller, 2017]. In this way, the owner’s requirement is satisfied and it also provided an additional support to the girder system and made the shallow girders possible. One plane of stay cable in the middle of two carriageways was designed for the Lachine Canal Bridge.

2. Large span (L) over depth ratio (H): The constraint on the girder depth required the girder to have a span over depth ratio up to 44, which made traditional steel plate girders not feasible with a constant depth. The steel box girders provide both bending and torsional stiffness to allow the shallow girders to be designed to meet bridge code requirements. The multiple box girders together with the properly designed cross frames allow the forces to be transferred and shared between girders and provide a sufficient and robust load path for the design loads.

3. Curved alignment: As previously mentioned, the bridge is on a curved road alignment. A constant 6% cross slope is required for the road design. Due to their inherent torsional stiffness, steel box girders can accommodate and carry the loads very efficiently for the curved structure. Multiple steel box girders take advantage of this effect and increase structural redundancy and resilience.

4. Phased construction: The footprint of the existing bridge covered a substantial section of the new bridge. To maintain traffic during construction, a phased construction is a mandatory requirement. The north bridge will be partially built and commissioned to traffic before the existing bridge is allowed to be removed and the new south bridge to be built. The steel girders in the north bridge need to be designed to take all dead and live loads before the transverse tie beam connecting both carriageways and the stay cables are participating in sharing loads. The south bridge will be built after the existing bridge has been demolished and the steel girders in the south bridge will be designed with the stay cable in place to provide additional support to the girder system.

5. Architectural considerations: The owner has stipulated high expectations on bridge aesthetics and put forward some very specific requirements as to the overall appearance of the bridge, including bottom and side of the superstructure shall be enclosed, and the deck overhang shall follow a smooth and curved profile, etc.

6. Constructability consideration: The bridge needs to be designed in a way that the construction sequence can be optimized and repetitive construction activities are maximized and flexibility & alternate methods are permitted. Smaller steel box girders shall be used for their lighter weight and easier handling. Full depth precast deck panel shall be considered for the concrete deck to have a composite section with the steel girder and to minimize the cast-in-place concrete pour on site. Additional consideration shall be given to the steel detailing to include feedback from the steel fabricator and erector.

A steel box girder system acting compositely with full depth precast deck panel is designed for the superstructure. Multiple small box girder (three boxes for each carriageway) with a width of 3 m is utilized together with steel cross frames to form a multi-load path structural system. A center pylon at pier 3 with one plane of stay cables is designed to provide additional support to the bridge. Figures 1,2 and 3 present the plan, elevation of the cross section of the new Lachine Canal Bridge.

[pic]

Figure 1: Plan View of Lachine Canal Bridge

[pic]

Figure 2 Elevation along South Carriageway

[pic]

Figure 3: Typical Cross Section (Section C-C in Figure 1, taken at Pylon)

A rendering to illustrate the constructed Lachine Canal Bridge is shown in Figure 4 with the redeveloped surrounding area.

[pic]

Figure 4: Rendering of Lachine Canal Bridge

3. DESIGN OF STEEL BOX GIRDER

1. Design Approach

The design of steel box girders was undertaken in accordance with the Canadian Highway Bridge Design Code, CHBDC, (CSA 2006) and MTMDET’s project specific requirements. In certain conditions, the latest version of CHBDC (CSA 2014) was followed with regarding to the curved steel bridge behavior.

At the project initiation stage, the project design team faced an important and difficult question: how to model the bridge to get the design forces? Shall a simple grillage analysis model be created for computing the design forces? Or shall a complicated and sophisticated finite element model be developed to get the design forces? The grillage model has its limitations and many simplifications and assumptions must be made. But it’s simple, straight forward with the time (schedule) advantage. When modeled correctly, the grillage model can quickly execute the analysis and show the results. In addition, the simple model allows to update and revise the design in relative short time. But it has its shortcomings, such as approximation and assumption required for the torsional stiffness calculation, modeling of one steel box with two different web heights as one girder with an average web height, the approximation for the cross-frame modeling, etc. On the other hand, the detailed finite element model is able to model all structural components precisely, including different web heights, individual top & bottom flange plate and exact member size & configuration for the cross frame, etc. It can model the girder torsional stiffness accurately. But it is time consuming to build and test a sophisticated finite element model and the output results are not straightforward. An engineer’s interpretation is required to use the analysis results correctly. After careful consideration and weighing the pros and cons of both modeling methods, the decision was made to use the simple grillage model to undertake the analysis and compute design forces.

A post-processing design spreadsheet was developed to link the output raw forces from analysis model to the design forces required for the steel girder design. The spreadsheet has the capability to account for load combinations and identify critical design forces for each structural component. It also plots charts for bending moment, shear force and torsion, which provides a clear and graphical representation of the design forces. Certain design sections in accordance with the bridge code were built in the spreadsheet as well. It overcomes the shortfalls of the grillage model and supplements the analysis model with calculations that address the special features of a curved steel box girder, such as the additional stress in the girder flange due to torsional warping which the grillage model does not capture. The stress in the girder due to different web height is accounted by the spreadsheet as well.

The design spreadsheet provides great value at the early stage of the design when the road geometry is not finalized and the girder depth is varied. It allowed the design engineer to quickly update the girder design and re-calculate the steel tonnage for the project team. It also made possible to investigate and compare different structural options in relative short time.

The spreadsheet calculates and transfers the design forces into steel stresses and provides a direct comparison to the design values permitted by code. It also computes the girder capacity and presents the demand over capacity ratio in accordance with the CHBDC. It saves the design engineer valuable time and can easily accommodate any force formats that are outputted from the analysis model.

2. Phased Construction

The new bridge needs to be built underneath the existing bridge. Phased construction, which allows the new bridge to be built while maintaining the traffic during construction, has important impacts on the steel girder design. The proposed construction sequence is presented as follows: (1) Build the north bridge first and shift the traffic to the partially constructed north bridge; (2) Demolish and remove the old bridge; (3) Build the south bridge and install the transverse tie beam connecting both carriageways and pylon with stay cables; (4) Complete and commission the entire bridge.

The steel box girders in the north bridge are designed for all dead and live loads without the stay cable in place to accommodate MTMDET’s request to open the bridge before completion of the south bridge construction. Different live load combinations in terms of number of traffic lanes are considered. The north bridge girders are essentially able to carry all loads by the girder system alone. The girders in south bridge are designed to take all dead loads, including steel weight, deck weight and superimposed dead load. However, the stay cables are required to help the girders to carry the live load. Various scenarios are analyzed and compared to obtain the most critical design forces for the steel girder design and to verify that the girders have adequate strength for all design loads.

When the north bridge is being built, the existing bridge is still in operation with one existing pier in the way of steel box girder B. Girder B is locally cut with an opening for the existing pier. Special diaphragm and splice details are implemented to transfer the girder force around the opening and provide a reliable load path.

3. Design Detail

There are three steel box girders for each carriageway generally. Girder A, B, and C are for the girders in the north bridge while the south bridge has girders D, E, and F. The girder height is obtained from the maximum allowable girder depth available using the given road profile. All girder bottom flanges are in one horizontal plane except girder F, which will be discussed in detail below. Vertical webs are used with a different height at each side to accommodate the 6% deck cross fall. Each girder is typically 3 m wide with a girder spacing of 3 m as well. Two steel plate type stiffeners are used to stiffen the bottom flange plate. Bottom flanges are typically at 3600 mm wide. Thickness varies from 16 mm at mid span to 30 mm at pier. Top flanges have a constant width of 600 mm and their thickness changes from a minimum 30 mm to a maximum 57 mm. Web thickness varies from 19 mm to 22 mm. The web plate is stiffened with transverse stiffeners but no longitudinal stiffeners are considered. The maximum web depth is 3.2 m, which allows the girder segments to be fabricated without horizontal splices in the web and transported in vertical position. There is a top lateral bracing for all box girders to ensure stability during construction, especially when the steel grillage system is loaded with the deck panels.

At the northwest corner, the north bridge needs to be widened to accommodate an additional lane. Two smaller steel box girders replacing one typical box girder are utilized to gain extra deck width. Between the normal deck width (three box girders) and this extra deck width (four box girders), one typical box girder is changed to two steel plate girders with varying girder spacing to accommodate the changing deck width. Lateral bracings are added both at bottom and at top to simulate a box girder behavior when subject to applied loads. To make sure an adequate load transfer from the box girder to two plate girders, special transfer diaphragms are designed. A typical box girder cross section is shown in Figure 5.

[pic]

Figure 5: Cross Section Showing A Typical Box Girder

4. Girder F Design

Box girder F has a sloped bottom flange to satisfy the architectural requirement of a curved and smooth configuration at the external side of the girder envelop. With the sloped bottom flange, there is a considerable web height difference. One web is at 2.0 m deep while the other one is 0.8 m deeper at 2.8 m total depth. As the analysis model still simulates the girder F in the same way as that of the typical box girder which has only smaller web depth variations, there is a concern that the girder F design forces computed from the model might not be accurate. The forces need to be adjusted for the real situation where there is considerable web depth difference and the bottom flange might carry some shear forces from the web. A conservative design approach is adopted to calculate girder design forces and ensure adequate strength of the girder when subjected to different loads.

5. Girder Deflection

The calculated maximum live load deflection due to one CL-625 truck on the deck is 53 mm for the largest span. This corresponds to 1/1660 of the governing span length and demonstrates the effectiveness of the steel grillage system with its outstanding load sharing performance. The static deflection limit is 90 mm in accordance with CHBDC S6-06. The maximum girder deflection under live load with CL-625 truck at all traffic lanes is obtained from the analysis model and the ratio of deflection over girder span length is calculated. The deflection value and ratio is shown in Table 1 (Gao, 2016). It can be seen that the maximum ratio is 1/1135, which indicates that the girder is relatively stiff and the girder deflection satisfies the MTQ requirement of less than span/800.

Table 1: Girder Maximum Live Load Deflection (Span 2)

|Girder Line |Maximum Deflection (mm)|Span Length (mm) |Deflection/Span Length |

|A |74 |84000 |1/1135 |

|B |63 |82000 |1/1302 |

|C |68 |84000 |1/1235 |

|D |58 |85000 |1/1466 |

|E |62 |86000 |1/1387 |

|F |65 |87000 |1/1338 |

4. DESIGN OF STEEL CROSS FRAME

Cross frames (CF) are designed along the girder span at a nominal 6 m spacing to connect girders in the transverse direction. All cross frames are arranged to be perpendicular to the girder axis. They are integral parts of the steel grillage system and play important roles in load transferring and sharing between box girders. Per CHBDC S6-06 requirement, the CF members are designed as the primary members to resist the applied loads.

As mentioned earlier, a grillage analysis model was developed with girders modeled as beam elements. The CF members were not explicitly modeled instead an equivalent beam element was used to simulate the effect of cross-frame members [NCHRP-725, 2012] as shown in Figure 6. The cross-frame equivalent beam stiffnesses were calculated considering a Timoshenko Beam Element. This approach involves the calculation of an equivalent moment of inertia, Ieq, as well as an equivalent shear area Aseq for a shear-deformable (Timoshenko) beam element representation of the cross-frame. According to NCHRP-725, Timoshenko beam provides a closer approximation of the actual cross frame as this equivalent beam is able to represent both flexure and shear deformations.

[pic]

Figure 6: Equivalent beam representation of cross frames

From the analysis result, the moment and shear force from the equivalent beams are obtained. In order to get design forces for the CF members, the moment and shear force in the beam are calculated and transferred to the axial forces in the CF members. Fig. 7 shows how the shear force and bending moments are transferred to axial forces in the CF members between box girders. The design forces for the internal CF members are calculated in accordance with FHWA Steel Design Handbook on Bracing System Design (Helwig and Yura, 2012). Using the forces in the beam elements representing box girders and CF, as well the section properties, the design forces in the internal CF members were calculated.

[pic]

Figure 7: Design axial force in CF members between two box girders

Built-up I sections are used for the CF members based on the construction requirement. A built-up section is selected on the basis that a thicker web plate must be used to allow adequate strength to transfer the load from the entire member section to the connection plate. The member sizes are designed based on the force demands. Two different types of CFs are required based on the consideration of standardization and minimizing member size types. Both Type-I and Type -II members have the same top and bottom flange plate width of 150 mm but slight different flange thickness of 12 mm and 16 mm respectively. The same web plate of 200 mm wide and 19 mm thick is used to simplify the connection to the girder web stiffener, which has a similar 19 mm thickness. All CF members are connected to the girder webs using 4 or 6 numbers of M25 A325M high strength bolts. Bolt slip resistance govern the connection design and 16mm splice plate was found to be adequate. The built-up I-section members are placed vertically to allow web connection to the girder stiffener plates. To increase the connection efficiency, two connection plates were designed to connect the cross-frame member web plate to the girder stiffener plate so the bolts double plane shear capacity can be utilized. The design of CF members sizes and connection details also take account of the fatigue consideration. Considerable effort is spent on designing the connection detail for the constructability while minimizing the fatigue risk.

5. DESIGN OF STEEL BEARING STOOL

Given the inclined bottom flange of Girder F, a stool was required to have a horizontal bearing surface for the bearings. The stool is a cellular steel element that is attached to the inclined bottom flange of the girder. For aesthetic reasons, the stool was designed as trapezoidal piece in the longitudinal and transverse directions. This also provided a technical benefit in the longitudinal direction as the tapered profile promoted better stress flow from the girder into the stool. The cellular makeup of the stool consisted of 25 mm steel plates fabricated perpendicular to each other at approximately 300 mm centres with 20 mm cover plates. The stool was oriented parallel and square to the girder even though the pier and diaphragm lines were skewed. The tight cellular configuration provided solid platform for the girder diaphragms to bear on. In fact, the bearing stiffeners on the girder diaphragms line up directly with the interior stiffener plates of the stool. Figure 8a shows the cross section of the bearing stool.

[pic]

Figure 8: (a) Bearing stool cross section and (b) FE model of bearing stool (internal details)

The stool was modeled using Midas Civil using plate elements separate from the global bridge model (Figure 8b). Because of the skewed geometry, the girders experience primary and secondary effects including a raking effect on the stool. To capture this effect, dummy outrigger elements were placed at a close spacing along the girder spine element within global modal at the pier locations. The tip deflection values of the dummy elements were rationalized and applied to the bearing stool model.

Plate stresses were assessed for various loading conditions and it was determined the plate sizes mentioned above were sufficient for the application. The stool was designed so that simple fillet welds could be utilized. Feathered ends on the longitudinal plates were used to mitigate fatigue issues.

6. CONCLUSION

The Lachine Canal Bridge is a large size and complex structure. An innovative and aesthetic pleasing bridge has been designed. Many people have contributed to the completion of the bridge design, which was completed in the spring of 2016. The bridge has been under construction since 2016. By the end of 2017, the north bridge construction has essentially been completed and the traffic has been shifted to the new bridge deck. In 2018, the existing bridge will be removed and the new south bridge will be constructed and the tie beam and stay cables will be installed. The entire bridge is expected to open to traffic in 2020. A landmark bridge structure will become a reality in Montréal in the near future.

Acknowledgements

Owner: Ministère des Transports, de la Mobilité durable et de l’Électrification des transports(MTMDET)

Owner’s Engineer: Gestion AECOM BPR

Independent Certifier: SMi and Arup

Design/Build Contractor: KPH, a joint venture of Kiewit, Parsons and CRH Canada

Bridge Designer: Parsons

Steel Fabricator: Cherubini, Halifax

Steel Girder Erector: St. Lawrence Erectors, Canam Group

References

CSA. 2006. CAN/CSA S6–06, 2006. Canadian Highway Bridge Design Code, Canadian Standards Association, Mississauga, Ontario, Canada.

CSA 2014. CAN/CSA S6–06, 2014. Canadian Highway Bridge Design Code, Canadian Standards Association, Mississauga, Ontario, Canada.

Gao, Y. 2016. Canal Lachine Bridge 100% Design Submittal, Superstructure Design. Parsons, Burnaby, Canada.

Mermigas, K. 2008. Behavior and Design of Extradosed Bridges. Master’s Thesis, University of Toronto, Ontario, Canada.

NCHRP Report 725. 2012. Guidelines for Analysis Methods and Construction Engineering of Curved and Skewed Steel Girder Bridges. Transportation Research Board, Washington, D.C.

Schueller M, New Dynamic Center Piece for Montréal Delivers on All Challenges, 39th IABSE Symposium, Vancouver, Canada, September 2017.

Helwig, T. and Yura, J. 2012. Steel Bridge Design Handbook: Bracing System Design. Report No: FHWA-IF-12-052 - Vol. 13, Federal Highway Administration (FHWA), USA

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download