Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations JAE on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!
Minnesota Bridge Findings 4
- Thread starter jike
- Start date
Hgtx
Thats a shame some one would want to know why you would want to look at one of your projects. I used to work for a consultant before going over to the dark side. However one of the advatages of contracting is nobdy asks how many hours did you bill this week.
I gree the system does work, howeverthe contractor focus on putting the structure together, somehow. In staged construction, which is the most common, there is a lot of locked in errection stress that I do not think are accounted for in design.
Further, bad details construction wise are repeated for years because no one sees the work in the field. I have encouraged our DOT, to no avail, to have the designers visit the jobsites periodically so they can get some feed back. They have not done it -0 again can't justify the cost to the project.
Hope you got your approval for you visit.
Thats a shame some one would want to know why you would want to look at one of your projects. I used to work for a consultant before going over to the dark side. However one of the advatages of contracting is nobdy asks how many hours did you bill this week.
I gree the system does work, howeverthe contractor focus on putting the structure together, somehow. In staged construction, which is the most common, there is a lot of locked in errection stress that I do not think are accounted for in design.
Further, bad details construction wise are repeated for years because no one sees the work in the field. I have encouraged our DOT, to no avail, to have the designers visit the jobsites periodically so they can get some feed back. They have not done it -0 again can't justify the cost to the project.
Hope you got your approval for you visit.
bridgebuster
Active member
When I first read about the plates being undersized it seemed credible as someone pointed out that it was drafting error 1 1/2 vs 1/2. However, going through the calculations 1 1/2 would be too much; 3/4" would've been OK. It could still be a drafting error: The designer's intent might have been A514 ( a bit of overkill) in lieu of the A441 but without the calculations we'll never know.
Maybe the design wasn't checked; maybe the design was done by an inexperienced engineer; there are so many what ifs. I worked for a firm (now defunct) that designed a number of long-span trusses; we would design and detail connections simultaneously. The reason being if something didn't fit it was easier to revise a design in progress. I looked at some plans prepared by my former firm, including one bridge designed a few years after 35W, and compared plate sizes for similar loads, we never used a 1/2" plate for a main truss connection.
As far as finding the error in the shop drawing review, let's face it, shop drawings are usually checked by someone at the bottom of the food chain. Would we expect someone at that level to know 1/2" was undersized?
I've also performed about two dozen inspections of long-span trusses and to be honest, we're not looking to see if a plate is the right size. Granted, during bridge inspection, we look to see if something jumps out at us but whether a plate should be 1/2" or 3/4" is not going to be noticed. It's not as if someone put a 1/2" plate where a 2" plate is needed.
Anyway, I still think there's more to the collapse than an undersized plate. The stress was above the allowable but below the yield point. Let's wait for the report.
Maybe the design wasn't checked; maybe the design was done by an inexperienced engineer; there are so many what ifs. I worked for a firm (now defunct) that designed a number of long-span trusses; we would design and detail connections simultaneously. The reason being if something didn't fit it was easier to revise a design in progress. I looked at some plans prepared by my former firm, including one bridge designed a few years after 35W, and compared plate sizes for similar loads, we never used a 1/2" plate for a main truss connection.
As far as finding the error in the shop drawing review, let's face it, shop drawings are usually checked by someone at the bottom of the food chain. Would we expect someone at that level to know 1/2" was undersized?
I've also performed about two dozen inspections of long-span trusses and to be honest, we're not looking to see if a plate is the right size. Granted, during bridge inspection, we look to see if something jumps out at us but whether a plate should be 1/2" or 3/4" is not going to be noticed. It's not as if someone put a 1/2" plate where a 2" plate is needed.
Anyway, I still think there's more to the collapse than an undersized plate. The stress was above the allowable but below the yield point. Let's wait for the report.
bridgebuster
Active member
I'm just wondering if the bolt holes in the gusset plates were punched full-size. It could help explain why they cracked.
I suspect that a lot of people haven't realised that the gussets are also acting as splices in the chords.
Looking at the published calcs, we see that the compression diagonal at U10 is designed to carry 2288kips at less than 22ksi giving a sectional area of at least 104 sq.inch. Scaled flange width is about 30in, so probably about inch & half thick. It seems hard to imagine why nobody thought it odd that the gusset thickness was only a third of the connected flange thickness. Maybe they did but decided that other people would know better.
If we play around with the calculation results for gusset U8 (above pier) we could conclude that a 5/8in gusset plate will just about do on a joint where the tension diagonals are carrying at least 2000kips. Follows that the calcs given are insufficent on their own and that additional requirements need to be met.
Personally believe there should be a code rule giving a minimum gusset thickness relative to connected plates or flanges.
Looking at the published calcs, we see that the compression diagonal at U10 is designed to carry 2288kips at less than 22ksi giving a sectional area of at least 104 sq.inch. Scaled flange width is about 30in, so probably about inch & half thick. It seems hard to imagine why nobody thought it odd that the gusset thickness was only a third of the connected flange thickness. Maybe they did but decided that other people would know better.
If we play around with the calculation results for gusset U8 (above pier) we could conclude that a 5/8in gusset plate will just about do on a joint where the tension diagonals are carrying at least 2000kips. Follows that the calcs given are insufficent on their own and that additional requirements need to be met.
Personally believe there should be a code rule giving a minimum gusset thickness relative to connected plates or flanges.
Personally believe there should be a code rule giving a minimum gusset thickness relative to connected plates or flanges.
That knee jerk response that would hurt more than it would help. I do these sorts of gusseted connections all the time for buildings. Sometimes you have a member that is much heavier than the other members at a joint. It may be at a low stress at this location. Why should the gusset plates be penalized because one of the members has thicker flanges than the others?
This bridge failure was preceded by a design error. There's no need to dumb down all design because of this mistake.
That knee jerk response that would hurt more than it would help. I do these sorts of gusseted connections all the time for buildings. Sometimes you have a member that is much heavier than the other members at a joint. It may be at a low stress at this location. Why should the gusset plates be penalized because one of the members has thicker flanges than the others?
This bridge failure was preceded by a design error. There's no need to dumb down all design because of this mistake.
- Status
Similar threads
- Question