Slab Weight Calculation on Flared Girder

[Son of a Bridge]

Hello everyone,

I am trying to create a spreadsheet that calculates DC loads for bridges. However, I do not know how software programs calculate the slab weight on flared girders. I have attached an image to see if anyone has a general idea of how to calculate the slab weight on the flared girder (girder 2). I know there is missing information, but I just need the general concept. By the way, these numbers are theoretical.

The only reason I am creating a spreadsheet is to compare my own calculations against software programs such as PGSuper or Conspan.
I do not want to be overly conservative. The objective is to obtain results similar to other software programs.

Has anyone ever validated DC loads for flared girders using hand calculations?

Any references or examples would be helpful!

Thanks,
Danny

 

[BridgeSmith]

Not sure how the programs would handle the loading, whether by integration or as an incremental series of discrete nodes approximating the linearly varying load. Using a spreadsheet, the only way I would know how to do it would be the incremental approach. Using small enough increments, that should compare very closely with the software, regardless of the method it uses.

The dead load is the fairly straightforward. The difficult part is approximating the live load. I've been a DOT bridge designer for 17 years, and I'd have to think long and hard about how to approach the truck loading for something like what you've depicted. We have LUSAS, so I'd probably have to model it using that, and compare it to some rough hand calcs and hope they were in the same ballpark.

 

[Settingsun]

7' uniform load plus 3' triangular load? Uniform and triangular loads on simply-supported beams have practically the same maximum bending moment and deflection, when calculated using total load on span. I know many (even most) designers will ignore creep redistribution for basic bridges.

 

[Son of a Bridge]

@HotRod10

Where would the boundaries be located to use the incremental approach?

Would it be similar to the new attached image?
Originally I was thinking the load would be determined half way along the girder (half the area goes to one girder and the other half goes to the other girder) to determine each girder reaction. However, the abutment skews are what's tripping me up. Not sure if I could use this approach or maybe I'm completely wrong.

I'm fairly new to bridges. I also want to learn how to determine the distribution factors for flared girders. Do you happen to know of any good references so I can learn about it?

TxDOT has a live load distribution spreadsheet for I girders if I'm not mistaken. The values obtained from the spreadsheet are the values that the software program PGSuper uses. That's one quick way I would go about validating the distribution factors for flared girders.

Danny

 

[Son of a Bridge]

@steveh49

Could you maybe create a sketch so I can visualize your explanation?

Would it consider the different skews?

Danny

 

[Settingsun]

I assumed that the slab is cast-in-place so the slab weight is distributed according to the (sacrificial) formwork system which is usually simply-supported between the girders IME.

The skew doesn't really come into it unless you try to do the creep redistribution calc which is built on assumption after assumption after simplification.

PS you can insert images directly in the post rather than attaching them. I can't see the images now because my phone hates .PNGs. That's my problem but in-line images are still easier to read.

 

[Settingsun]

Whoops. The second picture is a jpg so I can see it. The shading is what i had in mind and is equivalent to a uniform load plus triangular load. But 2/3 of the triangular load will go to the 10' abutment end, rather than the 50:50 split you've shown.

 

[BridgeSmith]

I don't think the distribution factor methodology in the AASHTO spec, which is the only one I'm familiar with, is valid for flared or splayed girders. As I alluded to earlier, analyzing the configuration you've shown accurately would take considerable thought and time, even for an experienced bridge designer like me. No disrespect, but analyzing the live loading for a bridge of this type is not project for someone who is "fairly new to bridges", unless you have considerable time to become familiar with how the truck configurations and magnitudes were derived.

Faced with this design, I would most likely take the simple and conservative route for the girders of calculating the loads as if the girder spacing was a constant 10' and using the 7' wide cross section to calculate the capacity. This is actually the approach suggested in the earlier versions of the AASHTO LRFD Bridge Design spec (haven't looked to see whether similar wording is in the latest versions).

It's not a common enough situation for us to warrant the labor expense to do more in-depth analysis, when the cost difference in construction will be minimal (a nominal increase in the volume of steel of one kind or another, whether that be mild reinforcing, prestressing strands, or thicker flange plates).

If all you're looking to get is the dead load, again, it's fairly straightforward, and steveh49 has given you a good approach to that.