Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
17 The findings of this synthesis study on the erection of steel bridges are based on survey questionnaires and interviews. They are summarized in these conclusions. The overwhelming majority of respondents agree that most of the common problems that occur during the erection of steel bridges can be prevented by the following: ⢠Verifying horizontal and vertical alignment before and during erection; ⢠Installing enough crossframes to maintain geometry and girder stability during erection; ⢠Properly using temporary falsework or additional cranes; and ⢠Rigorously following pinning, bolting, and tightening procedures. FINDINGS FOR OWNERS In regard to the procedures used by the designer, states reported that consideration is merited as to whether the designer should include an erection procedure in the design. Many states have reported problems with the deck profile resulting from the deflection, rotation, and translation of deck cantilever brackets. These deformations can be controlled through designer input on support locations to the contractor or contractor-developed forming plans. Also, states reported that problems develop in stage con- struction as the result of differences in elevation between the Stage 1 deflected position and the undeflected position of the Stage 2 members before pouring the Stage 2 concrete. Deck alignment between Stage 1 and Stage 2 and crossframe con- nections between Stage 1 and Stage 2 girders require special considerations. Successfully implemented strategies include the use of ⢠At least three girders in either or both stages to reduce transverse movement during deck pour, ⢠A closure or construction pour between the two stages, or ⢠Only a top and bottom strut between girders between stages. If deemed necessary, a strategy would be to add cross bracing after the deck pour. Respondents reported problems relating to girder stability owing to wind, crossframe erection sequences, temporary supports, and deck pouring sequence. Considerations toward solving the problems could include: ⢠Verification of stability in using the pouring sequence in positive moment areas, ⢠Checking for stability of the cantilever end of girder field section from pier to field splice, ⢠Checking the member length-to-flange-width ratio (several states provide guidance with preferable val- ues between 80 and 90), and ⢠Evaluating the need for lateral bracing. Where there are differential deflections between girders at the ends of crossframe connections, the girders will rotate transversely as (1) the dead load of the steel is applied, and (2) the concrete dead load is applied. Curved bridges and skewed bridges represent the most common examples of where that condition will occur. The designer should address the condition and should show on the design drawings whether the structure should be detailed so that the webs are vertical in no-load, steel dead- load, or full dead-load condition. Article 1.6.1 in the Guide- lines for Design for Constructability (AASHTO and NSBA) discusses this issue in detail. Also, it is not uncommon for girders to be out of plumb, and designers should evaluate the condition rather than spec- ulating how much the out of plumb is problematic. The twisting of box girders is another situation that needs to be considered if there is more than one bearing on either end of the box. Because of the rigidity of the boxes, provi- sion must be made to allow for field adjustments in the bear- ing height to account for any twisting that will occur. External crossframe connections can also be difficult because of the rigidity of the boxes to both transverse and twist movement. Article 3.9 of the Guidelines for Design for Constructability recommends that âIf multiple straight boxes or tub girders are adequately braced internally, external inter- mediate crossframes are not required. For curved multiple box or tub girders that require crossframes between mem- bers, use permanent crossframes.â CHAPTER SEVEN CONCLUSIONS
Bearing rotation was also mentioned by respondents. For tangent bridges on skewed supports, there is the potential for transverse rotation of the girder at the bearings, owing to dif- ferential deflections as well as to skewed pier diaphragms. Bearings for these types of structures should be designed to allow for this transverse rotation or, as a minimum, distortion- forgiving bearings such as elastomeric pads should be used. Other survey responses pertained to certification of fabri- cators and erectors. The owner should mandate that the fab- ricator and erector be certified by the American Institute of Steel Construction or another suitable program. Furthermore, the owner should enforce submittal and review, accept, or approve, according to agency practice, the erection proce- dure prepared by the erector. FINDINGS FOR FABRICATORS The fabricator should strive to understand the geometric fea- tures and how they affect erectionâparticularly curvature, differential deflections, skew effects, tolerances, and mem- ber rigidity. On complex structures, the fabricator should consult with the designer and the erector to determine the load condition for which the webs should be vertical. In this manner, all par- ticipants will understand the geometric assumptions. Shop-assembly methods were also discussed. The com- plexity of the structure must be considered when determining the shop-assembly method. The most common method is the progressive girder assembly, with at least three members in an assembly, often including one span or bearing to bearing. Records of the actual shop-assembly blocking dimensions should be maintained and made available to the erector. According to respondents, the use of standard size holes is encouraged. Oversized holes should generally be avoided, because the geometry of the structure can easily become imperiled. In regard to fabrication details, particular emphasis should be considered on the following: ⢠Holes should be drilled accurately, ⢠Splice material and main members should be accurately match-marked, and ⢠Members should be fabricated to appropriate sweep and camber tolerances. Attention should also be given to shipping stability. The fabricator should check the member length-to-flange-width ratio to ensure shipping stability. Where values exceed 60, computations should be made to determine if temporary brac- ing for shipping is required. 18 FINDINGS FOR ERECTORS The erector should submit for review, acceptance, or approval (based on the agencyâs practice) an erection procedure that addresses all of the pertinent issues. This procedure should lead to a properly erected structure. The issues that follow should be included in the erection procedure, although they are listed separately here for emphasis. To ensure erection stability, the erector might take the fol- lowing several measures. ⢠Check the ratio of member length-to-flange width for erection stability, ⢠Install enough crossframes to avoid flange buckling owing to the dead load of steel and concrete, ⢠Verify the stability of the partially erected structure for wind loading, and ⢠Use falsework as appropriate. Geometry control should be maintained at all stages of erection. This can be successfully accomplished by ⢠Determining, in conjunction with the fabricator and the designer, the condition at which the webs are detailed to be vertical and erecting them accordingly; ⢠Checking the vertical and horizontal alignment of bear- ings, falsework, and anchor bolt locations before erect- ing steel; and ⢠Using appropriate pinning and bolting procedures as detailed here. The geometry of the erected structure may be significantly affected by the procedures and sequences used for pinning and bolting the members during the erection process. The procedures detailed in this report represent a reasonable bal- ance of the various state requirements. They apply impor- tantly to splices in continuous members and other connec- tions where small movements or placement errors can have a substantial effect on geometry. The erector should review the shop-assembly blocking records to determine the effect of the camber fabrication tolerances on the final shape of the structure. A recommended summary procedure might con- tain the following: ⢠Initial pinning and bolting should consist of filling the holes in the connections with 25% pins and 25% bolts and the bolts at least snug tightened before releasing the crane and having the adjacent girders erected. ⢠The balance of the holes in the connections should be filled with snug-tight bolts. ⢠Final tightening of the bolts to installation tension should not start until a continuous line or at least adjacent spans have been erected and the vertical and horizontal align- ment has been verified. ⢠Pins should not be removed from the connection until after the previous step has been accomplished.
19 Erectors should also be aware of potential thermal effects, such as heating from the sun. Also, only experienced road crews should be used. An analysis of the questionnaire responses raised two general questions for the bridge community: 1. Do the reported problems caused by deviations from the vertical and horizontal alignment of the superstructure have a detrimental effect on the performance of the constructed bridges? 2. Are the problems described by the respondents endemic or more isolated? There were also comments on the effects of deviation from planned alignment. During construction, when dealing with any of the erection problems as discussed herein, the impor- tant question to ask is this: Is corrective action needed when something does not go as expected? Many owners reported problems encountered during erection, such as out-of-plumb girders, which in the end were allowed to remain unaltered or that required additional manipulation to complete cross- frame connections. No detrimental effect of such misalign- ments was subsequently reported. Is this truly a problem, or must the ramifications of the misalignment, or lack thereof, merely be better understood? One astute fabricator noted that the term âplumbâ has little meaning and that acceptable tol- erances based on subsequent adequate performance need to be developed. The erection of steel bridges, although based on science, is an art or craft. The practices and specification requirements are based on rules of thumb, experience, and intuition more so than on rigorous analysis. Rigorous erection analyses, including the prediction and reporting of intermediate deflec- tions (deflection before the final erected condition) are not made, according to the survey responses. Without such analy- ses for certain types of structures, problems of fit can be expected and, as the responses suggested, they do occur. The problems are exacerbated by stage construction. Determining acceptable tolerances of deviation from the planned vertical or horizontal alignment of the superstructure based on subsequent performance of the bridge could aid owners in determining whether a true problem exists and lim- iting or preventing much frustration on the part of fabricators and erectors. The current frequently cited use of the term plumb without associated tolerances given in specifications can be considered too restrictive and unenforceable. Finally, many erection problems were reported by owners, fabricators, and erectors. For the most part, owners related specific bridges where problems were encountered during erection, whereas fabricators and erectors provided more general discussions of problems. Although most of the own- ers could cite a problem bridge, the problems seemed iso- lated. When asked to discuss the solution to the problem, the owners provided much information on how the problem was solved on the problem bridge. However, few offered a global solution to the problem, such as a change in their specifica- tions or practices. This observation suggests that while the problems appear real to the owners, they are not endemic. In many cases, when problems with alignment arose, the owners chose the do-nothing option with apparently no adverse impact on the performance of the bridge. It could be asked that when doing nothing is acceptable, does a problem exist? For some specific field problems cited, other than those resulting from failure to follow appropriate specifications or acceptable practices, a rigorous incremental analysis of the erection process could solve problems. Today, such analysis is routine for more complex forms of bridge construction, such as segmental concrete bridges and cable-supported bridges. However, with such analysis, additional costs are incurred. In the majority of the reported cases, additional effort in the field solved the problem. Before more rigorous incremental analy- ses are instituted, the question to be answered is whether the potential field costs to solve problems exceed any proposed rigorous analysis costs before erection.
