Description of the Project:
The Johnson Street Bridge, constructed in 1922, is recognized as a key feature of the Victoria Harbor landscape. It is one of the only remaining operating bascule style bridges in North America designed by Strauss (the designer of Golden Gate Bridge). The bridge has three spans. Two approaches which are composite plate girders with floor beams spanning 73’ (22 m), 110’ (34 m) respectively, and the main bridge which is 60 m span single leaf, heel trunnion bascule bridge, fig (1) below. The main bridge comprises the 10m width highway bridge, and the 7 m width railway. Seismically, the main bridge could be divided into: Moving part which is the bascule Warren truss structure that rotate for navigational pass, and the fixed part which is a rigid triangle steel frame that carry the rotating counterweight (CW), operating motors and the relevant members. The members are primarily riveted lattice truss work, double angle sections, heavy H-shape built-up sections, and plate girders with stringers. The substructure comprises: the rest pier (P1), main trunnion pier (P2), the CW pier (P3), and two massive concrete abutments. They rested either on rock or on timber piles driven through soft soils to reach the rock. The bridge was inspected , and the client was concerned to get sound engineering judgment regarding the bridge’s capacity against earthquakes.
My Role in the Project (in Detail):
The Johnson Street Bridge, constructed in 1922, is recognized as a key feature of the Victoria Harbor landscape. It is one of the only remaining operating bascule style bridges in North America designed by Strauss (the designer of Golden Gate Bridge). The bridge has three spans. Two approaches which are composite plate girders with floor beams spanning 73’ (22 m), 110’ (34 m) respectively, and the main bridge which is 60 m span single leaf, heel trunnion bascule bridge, fig (1) below. The main bridge comprises the 10m width highway bridge, and the 7 m width railway. Seismically, the main bridge could be divided into: Moving part which is the bascule Warren truss structure that rotate for navigational pass, and the fixed part which is a rigid triangle steel frame that carry the rotating counterweight (CW), operating motors and the relevant members. The members are primarily riveted lattice truss work, double angle sections, heavy H-shape built-up sections, and plate girders with stringers. The substructure comprises: the rest pier (P1), main trunnion pier (P2), the CW pier (P3), and two massive concrete abutments. They rested either on rock or on timber piles driven through soft soils to reach the rock. The bridge was inspected , and the client was concerned to get sound engineering judgment regarding the bridge’s capacity against earthquakes.
My Role in the Project (in Detail):
1) I reviewed the inspection reports, photos, and as-built drawings for the whole bridge.
2) I checked and modified the finite element model, and add the boundary conditions, see fig (2).
3) I communicated with UBC, Geological Survey of Canada (GSC) and National Resources Canada for collecting information on the site specific response spectrum, Peak Ground Acceleration (PGA), and earthquake history in the bridge area.
4) I performed the seismic vulnerability assessment of the existing bridge by applying three kinds of analyses: static analysis (DL & LL), modal (free vibration eigenvalue analysis), and response spectra analysis. Thorough observation of the modal analysis results was made to figure out the weak planes of the bridge, the structural behavior during earthquake events, and the shape of displacement under consideration, see fig. (3). I realized from the results that longitudinal direction of ground motion has fewer effects than the transverse direction. Because of the weak bracing used in the existing structure, much of the bridge’s strength was not engaged. Consequently the bascule span and the CW tower act as discrete units rather than as a continuum, see fig (4). Also from the analysis I found that the stresses were concentrated in the some part of the bridge. Also, during earthquake events there will be a great probability for collision between highway and railway bridges (i.e. two different periods for each of them).
5) The bridge had survived some earthquakes, the 2001 Nisqually WA earthquake, and the 1946 mid Vancouver Island earthquake. For that reason the City of Victoria inquired (in a meeting) about “the Do-Nothing option”, which represents the bridge surviving earthquake in its existing condition. I employed failure criteria based on maximum deflected displacements (stability criteria) and maximum stresses (strength criteria). Those criteria were for the steel superstructure, concrete substructure, and timber piles. After careful interpretation of the results, the lower allowable earthquake level was defined as an event with an intensity having a 35% probability of exceedance (PoE) in 50 years (119 years return period). The PGA associated with this event was bigger than the 2001 and 1946 earthquakes. The failure would start in the timber piles causing a stability issue (rigid body motion) of the whole bridge. The “Do Nothing option” has considerable risk particularly when considering the volume of daily traffic, historical factor of the bridge, post-earthquake repair work expenses and potential for negative economic impacts if the bridge ever was closed after a seismic event and as such is not recommended. This risk is accentuated by the fact that to-date, the bridge has experienced relatively low earthquakes as compared to the design earthquake recommended by the CHBDC.
6) I proposed retrofit options to improve the structural behavior of the bridge, and to reduce the seismic demands. Those options were adopted after careful consideration of weight changes, existing strength of bridge’s electrical/mechanical equipment and the required services of the bridge during opening for traffic (vessel, trains and vehicles). These options were:
· Option 1 (CW reduction): It was the reduction of the Counter Weight (CW) mass, and replacement of the existing electrical/mechanical system to provide a more powerful motor. The stresses dropped for two reasons, however, the ultimate capacity of the bridge that meet the requirements of the Capacity on Demand C/D ratio = 1 (i.e. the critical case) was found to be 12% PoE in 50 years, and the PGA = 0.3 g (less than required in the code).
· Option 2 (CW mass relocation): It was replacement of the existing CW with new one located in a cavity under the deck. This would also require replacement of the existing mechanical system.
· Option 3 (structural strengthening): It was to improve the seismic performance of the bridge and get better internal load paths by changing the structural behavior, fundamental period and mode shapes. This would involve adding/replacing steel members. I proposed the energy dissipation eccentric frames, see fig (5), which used as new bracing. The high ductility capacity is achieved by hysteresis of short, ductile, replaceable link elements. This redistribution resulted in an overall reduction in the stresses, see fig. (6). The analysis indicates that this option presented an acceptable solution if the bridge is to function as an emergency route structure but not an adequate solution to make the bridge into a lifeline structure.
· Option 4 (seismic isolation): it was the minimization of the seismic forces applied on the bridge during earthquakes by shifting the fundamental periods and increasing the damping effect. I entered different effective stiffness of the isolators for each support in order to gain the best displacement demands that would not affect the stability criteria (i.e. the maximum drift). The C/D values for this option indicated improvement of the bridge’s response for the three earthquake levels (i.e. 2%, 5%, 10 % PoE in 50 years). Fig (7).
· Option 5 (substructure upgrading), see fig (8): it was the assignment of a number of members in the system (bridge) which have ductility, and let the whole elements to form multi-load paths and to contribute during the earthquakes. It implied to build new substructure under the fixed part of the bridge. In this option the failure mechanism would be clearly defined and at the end it would prevent undesirable failure modes. I suggested this option because it has the benefit of adjusting more variables in the system to gain the required seismic performance, i.e. number, properties, material, dimensions, depth, tapering and detailing of the shafts, as well as the depth of the detached hollow steel pipes that surround the shafts (the sleeves). I determined the number of the shafts to increase the redundancy of the substructure, and that reflected on a useful response modification factor. Attention was paid by our team to the constructability of this new substructure, so I chose the properties and dimensions to achieve the required stiffness in the assigned limited space (available in the existing structure). The length of the proposed shafts is approximately 38 m and the diameter changed regarding the required location of the plastic hinges (maintenance requirements). There was a great stress reduction and the maximum displacement of the bridge elements was acceptable. This alternative would be more durable and reliable for major earthquakes (i.e. M = 6.5 Richter or more). Fig. (9) represents the FE model for option 5, and fig. (10) represents a proposed detailed drawing for option (5).
I supervised the drafters during the drawing process of the existing bridge and this option.
7) I made comparison for the options on the basis of their effectiveness in reducing seismic risk, traffic disruption, impact on the historic character, and cost.
8) I participated in establishing a rehabilitation program with the intention of keeping the existing bridge in service for an additional 40 years.
I depended on CAN/CSA S6-06 (CHBDC), BC MoT supplement to S6-06, BC MoT Seismic Retrofit Design Criteria, and other famous seismic textbooks (listed separately in other page). Further work for this project is still under negotiation with the city council.
2) I checked and modified the finite element model, and add the boundary conditions, see fig (2).
3) I communicated with UBC, Geological Survey of Canada (GSC) and National Resources Canada for collecting information on the site specific response spectrum, Peak Ground Acceleration (PGA), and earthquake history in the bridge area.
4) I performed the seismic vulnerability assessment of the existing bridge by applying three kinds of analyses: static analysis (DL & LL), modal (free vibration eigenvalue analysis), and response spectra analysis. Thorough observation of the modal analysis results was made to figure out the weak planes of the bridge, the structural behavior during earthquake events, and the shape of displacement under consideration, see fig. (3). I realized from the results that longitudinal direction of ground motion has fewer effects than the transverse direction. Because of the weak bracing used in the existing structure, much of the bridge’s strength was not engaged. Consequently the bascule span and the CW tower act as discrete units rather than as a continuum, see fig (4). Also from the analysis I found that the stresses were concentrated in the some part of the bridge. Also, during earthquake events there will be a great probability for collision between highway and railway bridges (i.e. two different periods for each of them).
5) The bridge had survived some earthquakes, the 2001 Nisqually WA earthquake, and the 1946 mid Vancouver Island earthquake. For that reason the City of Victoria inquired (in a meeting) about “the Do-Nothing option”, which represents the bridge surviving earthquake in its existing condition. I employed failure criteria based on maximum deflected displacements (stability criteria) and maximum stresses (strength criteria). Those criteria were for the steel superstructure, concrete substructure, and timber piles. After careful interpretation of the results, the lower allowable earthquake level was defined as an event with an intensity having a 35% probability of exceedance (PoE) in 50 years (119 years return period). The PGA associated with this event was bigger than the 2001 and 1946 earthquakes. The failure would start in the timber piles causing a stability issue (rigid body motion) of the whole bridge. The “Do Nothing option” has considerable risk particularly when considering the volume of daily traffic, historical factor of the bridge, post-earthquake repair work expenses and potential for negative economic impacts if the bridge ever was closed after a seismic event and as such is not recommended. This risk is accentuated by the fact that to-date, the bridge has experienced relatively low earthquakes as compared to the design earthquake recommended by the CHBDC.
6) I proposed retrofit options to improve the structural behavior of the bridge, and to reduce the seismic demands. Those options were adopted after careful consideration of weight changes, existing strength of bridge’s electrical/mechanical equipment and the required services of the bridge during opening for traffic (vessel, trains and vehicles). These options were:
· Option 1 (CW reduction): It was the reduction of the Counter Weight (CW) mass, and replacement of the existing electrical/mechanical system to provide a more powerful motor. The stresses dropped for two reasons, however, the ultimate capacity of the bridge that meet the requirements of the Capacity on Demand C/D ratio = 1 (i.e. the critical case) was found to be 12% PoE in 50 years, and the PGA = 0.3 g (less than required in the code).
· Option 2 (CW mass relocation): It was replacement of the existing CW with new one located in a cavity under the deck. This would also require replacement of the existing mechanical system.
· Option 3 (structural strengthening): It was to improve the seismic performance of the bridge and get better internal load paths by changing the structural behavior, fundamental period and mode shapes. This would involve adding/replacing steel members. I proposed the energy dissipation eccentric frames, see fig (5), which used as new bracing. The high ductility capacity is achieved by hysteresis of short, ductile, replaceable link elements. This redistribution resulted in an overall reduction in the stresses, see fig. (6). The analysis indicates that this option presented an acceptable solution if the bridge is to function as an emergency route structure but not an adequate solution to make the bridge into a lifeline structure.
· Option 4 (seismic isolation): it was the minimization of the seismic forces applied on the bridge during earthquakes by shifting the fundamental periods and increasing the damping effect. I entered different effective stiffness of the isolators for each support in order to gain the best displacement demands that would not affect the stability criteria (i.e. the maximum drift). The C/D values for this option indicated improvement of the bridge’s response for the three earthquake levels (i.e. 2%, 5%, 10 % PoE in 50 years). Fig (7).
· Option 5 (substructure upgrading), see fig (8): it was the assignment of a number of members in the system (bridge) which have ductility, and let the whole elements to form multi-load paths and to contribute during the earthquakes. It implied to build new substructure under the fixed part of the bridge. In this option the failure mechanism would be clearly defined and at the end it would prevent undesirable failure modes. I suggested this option because it has the benefit of adjusting more variables in the system to gain the required seismic performance, i.e. number, properties, material, dimensions, depth, tapering and detailing of the shafts, as well as the depth of the detached hollow steel pipes that surround the shafts (the sleeves). I determined the number of the shafts to increase the redundancy of the substructure, and that reflected on a useful response modification factor. Attention was paid by our team to the constructability of this new substructure, so I chose the properties and dimensions to achieve the required stiffness in the assigned limited space (available in the existing structure). The length of the proposed shafts is approximately 38 m and the diameter changed regarding the required location of the plastic hinges (maintenance requirements). There was a great stress reduction and the maximum displacement of the bridge elements was acceptable. This alternative would be more durable and reliable for major earthquakes (i.e. M = 6.5 Richter or more). Fig. (9) represents the FE model for option 5, and fig. (10) represents a proposed detailed drawing for option (5).
I supervised the drafters during the drawing process of the existing bridge and this option.
7) I made comparison for the options on the basis of their effectiveness in reducing seismic risk, traffic disruption, impact on the historic character, and cost.
8) I participated in establishing a rehabilitation program with the intention of keeping the existing bridge in service for an additional 40 years.
I depended on CAN/CSA S6-06 (CHBDC), BC MoT supplement to S6-06, BC MoT Seismic Retrofit Design Criteria, and other famous seismic textbooks (listed separately in other page). Further work for this project is still under negotiation with the city council.
Figure (1) Johnson Street Bridge
Figure (2) Isometric view of the Johnson Street Bridge superstructure Model, created using Midas Civil.
Figure (3). The first 9 mode shapes of the existing bridge
Figure (4). Deflected Shape of Existing Bridge due to Seismic Loads
Figure (5), A) The existing bracings, B) The proposed energy dissipation system
Figure (6). the stresses in option 3 after forming the plastic hinges in the horizontal members.
Figure (7). The maximum displacement (the drift) of the seismic isolated option.
Figure (9) stress redistribution in option 5
Figure (10). Option 5 general arrangement.
