Kerb and parapet load

[Shz713]

Hi everyone,

what is the best practice to represent parapet and kerb load of bridges as superimposed dead load? Should it be line load (N/M)?

If parapet is made up of steel and kerb made up of plain concrete, then we simply get cross section and multiply by density of material to get load per meter length. I have adapted such approach but I got 12.4 KN/m line load for parapet and kerb. Is this approach correct?

Besides, for deck wearing surface, I multiplied thickness of deck wearing to concrete density and applied it as pressure load (N/m^2) over entire deck surface. Is this correct practice?

Thanks Engineers

 

[BridgeEI]

The application depends on how you or your software is doing the analysis. But yes your approach is reasonable, line load for parapets/curbs and an area load for the wearing surface.

 

[Shz713]

Dear bridgeEI,

Thanks for your response. Well, I'm using CSiBridge for analysis. Attached are examples of how I applied parapet and wearing surface as superimposed dead loads.
When I check the dead loads with that of grillage model (prepared by someone else) I have, there are some discrepancies. I tried to match my results with dead loads obtained by grillage model by no luck so far

http://files.engineering.com/getfil...-183a-43a3-af83-f260c13bee69&file=parapet.PNG

http://files.engineering.com/getfil...fb-b9b7-725ebb2051e9&file=wearing_surface.PNG

 

[BridgeEI]

It looks like your loads are applied correctly. Your discrepancy is probably due to how you distribute the parapet in your calls. I'm not familiar with the program you are using, but I assume your exterior girder is carrying more of the parapet than the interiors.

 

[Shz713]

Dear bridgeEI,

Thanks for your prompt reply. In fact, parapet is located at outer side of the cantilever portion in both sides of box girder; thus I applied them directly to the portion in which they been designed. Thi is same as picture shown below. For the case of deck wearing, it is been applied only to the portion of deck which is covered by wearing, not the footway and kerbs. Hence, parapet load is imposed as line load and deck wearing as pressure load to account for entire deck portion.
My concern is about how loadings are calculated in grillage model. Because in grillage model the superimposed DL (parapet and wearing) are calculated as well bu values are not matching with what I have found based on drawings.

Cheers

 

[BridgeEI]

If the values are on a drawing or calculation they may be from another party. If that's the case, the difference is probably due to how they assumed the load is distributed.

Normally the reaction would be calculated as a total parapet load of all parapets divided by the number of girders and that is applied as a line load to the girder. So for example, say you have 2 parapets at 10 k/ft and 5 girders. The load I apply to each girder is 2*10/5 = 4 k/ft. Then I determine the reaction from that. I do the same for the wearing surface. Just note that some jurisdictions may not divide by all girders in the bridge and only a few of them.

 

[Shz713]

You are right that different jurisdictions have different approach, particularly in Australia every approach for load assessment is different!

What I didn't understand is that why you divided the parapet loads to all girders? I mean if we assume that parapet contributes to overall stiffness of bridge, then we probably do that. However, what if it is found that parapet has marginal effect on stiffness of bridge. Do we still share the load effect of parapet to all girders? because what I did for parapet load the most portion of it is merely carried by outermost girder.

Cheers

 

[BridgeEI]

The parapet will be carried by all of the girders. In actuality Some more than others. It's just a simplification in the code to make our lives easier.

The parapet isn't going to contribute to stiffness of the bridge until it's in place and semi cured. So all load supporting elements pick up this load to some extent.

For the most part in the states, anything above the deck slab is ignored fit contributing stiffness. That's just the practice here. You can make the argument to do it, but are you getting all of these refined calcs covered in your fee? I don't, so we make assumptions to simplify things that aren't too significant.

 

[Shz713]

That's correct. I'm involved in dynamic analysis of bridge not design so I need to take into account effect of even non-structural members on modal parameters of bridges.
Thanks again