pipe bridge support

[StructDave]

A client (large Eng. company) asked me to review a single-span pipe bridge design, span >100ft. In the initial review (3 weeks ago) we introduced the subject of bridge camber, and had extensive discussions about weight, erection, galvanizing etc.

In the second round it became clear the design team is proposing the bridge be fixed at the towers, i.e. extended over both supports with bottom chords anchored front and rear. The bridge will project some 15ft over each tower, so the front-rear support spacing at each end of the span would be about 15ft. So we discussed negative bending in the bridge, the reaction-couple on each truss, the fact the front-reaction will be increased by the magnitude of the the rear/negative reaction, and how the towers will transfer the loads to foundations.

The client's design is taking advantage of reduced overall truss moments. They propose to include sliding connections at one end, in which slotted bolt holes allow sliding but can also provide the negative reaction force to tower.

To me this seems more complex and risky, than the simple-span approach. I've counseled thorough attention to detailing at the towers, to ensure all loads are addressed. I am accustomed to industrial bridges with simply-supported spans, and would naturally have used pin-connections, with expansion accommodated at one support by either sliding bearings or a doubly-pinned bent (or hanger).

The client has modeled the bridge and towers as one structure. My approach would require separate models, and there seems to be some resistance to the idea of back-pedaling.

We are also discussing lateral wind transfer from the top horizontal frame, as this force has to find its way thru the same set of connections, and we are discussing the possibility of a simple-span condition during erection. I should add, the foundations are very substantial and likely can take the negative moment.

Has anyone here used rigid/fixed connections for a single-span bridge (for pipes, conveyor, etc.)? Successfully? I really appreciate your thoughts.

 

[JAE]

The main thing that pops out at me is the idea of using longitudinal slotted holes where there is potential uplift, or perhaps sustained uplift, on the bolt nuts.
This requires the nuts and washers on top of the slotted connection to allow sliding with thermal movements.
Most bridge slide details I've ever seen deal with full-time downward reactions so a continuous uplift force on a nut/washer while thinking that will always allow "free" movement is a pipe dream (no pun intended).



Check out Eng-Tips Forum's Policies here:
faq731-376

 

[StructDave]

Thanks JAE, pun appreciated. Also the observation; my client would have to devise a sliding guide with hold-down capability for the negative reaction. Forces are such that a Teflon-SS type bearing is warranted, but they'll have to invert half of them. Definitely something I will point out.

 

[racookpe1978]

From your description, it is not clear to me whether you are proposing to suspend the pipe from the two towers (with the pipe acting as its own structural member), or whether the pipe is supported on top of an independently supported bridge.

The first obviously requires less structural members/weight, but the second separates the pipe movement from the structure movement at both towers, which simplifies the both structural analysis and pipe stress/movement analysis and makes the system more robust.

 

[StructDave]

Thanks racookpe1978; the intent is pipes will be supported on a bridge structure. The "Towers" will be braced frame structures no higher than the bridge. Pipes will have guided supports that allow their thermal movement to take place independently of the bridge. My dilemma is only about whether the bridge (a dual-truss box arrangement) should be simply supported (zero-moment at supports) or should have fixity at supports (with negative end-moments).

 

[racookpe1978]

Hmmmn.

Your (relative) movement then is between the pipe and pipe supports - which are fixed to the bridge structure.
But!
The bridge structure then must be expected to move (must move!) with respect to the land on both sides of the bridge endpoints.

The only force pushing the bridge sideways or end-to-end will be wind loads and the friction remaining between pipe and pipe supports. Sounds like you're planning to accommodate both. Friction loads from a good pipe support system, even after rust and corrosion and bad maintenance) will be minor compared to gravity loads on a full pipe and sideways wind loads.

I'd go with the solution giving you the least motion of each end of the bridge structure compared to the land-locked ends: heat changes from summer to winter could be as much as 3 inches axially at each end. The sideways land-support loads need to be anticipated.

 

[BridgeSmith]

Sounds similar to our overhead sign support structures, except your 'posts' at the ends may not be nearly as tall. We typically pin ours at the top of the posts to eliminate most of the potential fatigue-prone locations within the structure. Generally, that leaves only the post to base plate connection to be concerned about, and there is much experience and research into making those connections fatigue resistant. All that to say, I recommend the simply supported approach.

 

[StructDave]

Thanks everyone. In a conference with the client I discovered the foundations are soil-based (not anchored to bedrock) and we are in a marine environment, which tells me the foundations will eventually creep (rotate) out of position. A permanent negative moment from the pipe bridge will encourage foundation creep, to a point where the foundations can no longer provide bridge end fixity. At that point it was an easy sell to have them change the bridge design to simply-supported.