AREMA Rating Steel Bridge Load Combinations 1

[PatrickM111]

I have to complete the rating of a steel truss bridge and since AREMA Manual for Railway Engineering does provide load combinations for steel structures, it's a bit of a challenge.

There are no centrifugal forces or CWR, therefore the only loads I can see including are: DL, LL, Imp, Wind Forces and Longitudinal Forces. Are the longitudinal forces usually included?

I was thinking the following:

1. DL + LL + Imp
2. DL + LL + Imp + Wind on Train + Wind on Bridge
3. DL + LL + Imp + Wind on Train + Wind on Bridge + (Braking or Traction)
4. DL + LL + Imp + (Braking or Traction)

Does anyone have any suggestions?
Thanks

 

[bridgebuster]

For a highway bridge, your Rating Factor would be:

(Member Capacity - Dead Load)divided by (Live Load + Impact).

The other forces you cite are normally used for substructure design.

 

[LobstaEata]

bridgebuster

Having rated many railroad truss bridges over the years, I must respectfully diagree. Wind on LL, and longitudinal braking loads will affect the rating of diagonal truss elements. Rocking of LL must also be considered when calculating impacts loads. Speed ratings will also a factor that must be considered.

Patrick: You may already know this, but keep in mind that when rating a "truss" bridge that ALL truss elements must be rated considering every position of live loads along the span. Using influence lines or a stiffness matrix program with a transparent moving load generator will be necessary. The overall rating of bridge will depend on the weakest element of the truss. I'm not sure if AREMA has caught up with evaluation of connections yet for ratings, but considering the collapse of the I35W bridge in the US, it probably should be part of your rating now.

If this is the first truss bridge that you have rated, it will be helpful to check with someone in your firm that has done it before. There is a lot of work involved to come up with the correct rating values. Be careful to set things up right from the start or you will be chasing your tail and burning right through your budget.

 

[bridgebuster]

LobstaEata,

Does AREMA permit an overstress when taking into account wind on liveload, etc? Do you also consider wind on superstructure?

ON a highway truss we would never take these into account since the DL+LL+I case would govern.

 

[LobstaEata]

Highway and rail bridges are two very different structures. Highway truss bridges are much wider between trusses than rail bridges for one.

Wind load cases do permit a 25% increase in steel stress when using ASD. Wind on superstructure when combined with wind on LL should be considered, but will be highly dependent upon structure type as a controlling case.

AREMA PASSAGE

h. Wind Load on Structure. The wind load acting on the structure shall be assumed as 45 lb per square foot
on the vertical projection of the structure, applied at the center of gravity of the vertical projection. The
wind load shall be assumed to act horizontally, in a direction perpendicular to the centerline of the track.

i. Wind Load on Live Load. A wind load of 300 lb per linear foot on the train shall be applied 8 feet above
the top of rail horizontally in a direction perpendicular to the centerline of the track.

A long span deck truss is a structure type where wind may come into play. Wind on LL for this type of bridge has the potential of producing a long moment arm between the centroid of applied wind load and bearings than the shorter couple distance between trusses.

 

[PatrickM111]

Thanks for the tips guys!

I am using a structural analysis program with a moving load generator, therefore I will create force envelopes in order to identify the "worst case" members.

From what it sounds like, I probably will have to the check the wind load cases with the braking (or traction force).

Although, there were no signs of cracking or significant section loss at the connections, I was going to identify a few typical connections and check the capacity.

If anybody has further guidance, please provide.

 

[yeseri]

On top of the excellent comments of LostaEata, I would like to add my two cents on this topic.
Please forgive me in advance if my comments are too lengthy and boring.

According to AREMA 2006 Part 7 Existing Bridges and Part 9 Commentary, LF should be applied 8 ft above top of rail for Braking, and 3 ft for Traction. Section 15-9.1.3.12 says, “This force is transferred from vehicle to rail as a horizontal force at the top of rail and a vertical force couple transmitted through the wheels.”

It is fairly straight forward for the horizontal force applied at top of rail. However, it is not so clear for the vertical force couple. To understand the vertical force transfer into the bridge members, I would like to introduce two modeling concepts: (1) External Mass and (2) Internal Mass. The first model can be used for calculating forces at bearings and the second one will be useful in understanding how each structural member reacts to LF.

1. External Mass Model for Forces at Bearing

The external mass consists of a solid rectangular box rests on two simple supports. The rectangular solid box represents the entire superstructure as a whole. When LF is applied at 8 ft or 3 ft above the rail, this solid box directly transfers the load to the bearings creating a vertical force couple at the bearings. Some people may raise a question, “How can a force go up? Well, there are plenty of other downward loads such as Live Load and Dead Load to prevent one end from being lifted up. So, moment (M = LF x Arm) will be resisted by the vertical force couple: one goes up and the other goes down at the bearings.

The horizontal force will be transferred to the bearings also based on their fixed conditions. These horizontal force and vertical forces will be used to check the bearing materials as well as anchor bolts capacity and substructure members.


2. Internal Mass Model for Individual Structural Members

Unlike the external mass model, the superstructure is no longer a solid box in the internal mass model. It rather consists of many different structural members such as stringers, floorbeams, plate girders and truss members. In bridges with stringers and floorbeams, LF is first applied to the stringers and then must be transferred to the members to which stringers are connected, usually the floorbeams.

When modeling this internal mass in computer analysis software, it is recommended to apply LF at as many locations as possible rather than at only one location. If LF is applied at one location, the structural elements right underneath the application point will experience excessive stress. It is recommended to apply LF at the location of each floorbeam. For example, if the total LF is 200 kips and total number of floorbeams is 20, then 10 kips of horizontal force will be applied at each floorbeam location at the elevation of 8 ft above the rail top. For application arm linked to 8 ft for braking and 3 ft for traction, it is recommended to create weightless rigid member and apply LF on top of those dummy members.

When calculating the forces by hand, the floorbeams will get even horizontal force. When it comes to vertical force at floorbeams, it is recommended to use a group of pile analogy. Imagine there are 20 piles lined up in one row and then the moment created by LF x Arm will be resisted by this 20 piles. The outer piles will get more loads than the inner piles. The vertical pile reaction, p = M* xi / sum (xi ^2) from pile design.

I think that’s enough for today.
Thanks for your attention.