HEC RAS "bent" bridge modeling

[Crystalct1]

I have a bridge to model in HEC RAS that has a bend in it. Apparently, the road had been widened years ago and the added portion of the bridge had to be put on a slant to avoid conflicts. So the result is a bridge that is perpendicular to the roadway up to the median, then the rest has a 20-degree-from-the-vertical slant. I've seen culverts like this before, but I never modeled one in HEC RAS - I've only done HGL calcs for drainage design for this type. I know there are some bend losses to consider, but when using a bridge, HEC RAS doesn't do losses, does it? I did find a way to enter my own orifice coefficient and calculate this including the bend losses.

Or I could model it as a buried culvert. Two buried culverts?
- one for each section and then put no distance between them.

Any ideas?

 

[JEmH]

While I've not had this specific situation, I have seen similar HEC-RAS problems. If I have a weird bend from a wing or something, I generally try to imagine how the flow will expand or contract in "real life." It helps to have a big storm so you can go watch the flow. If I can see that there is going to be a dead spot(ineffective flow area)under the bridge I'll model the abutment as just behind the eddy line.
In your case I'd be tempted to imagine that the downstream wall on the inside of the bend is straighter than it actually is and pinch in the downstream opening.

Good luck

 

[Crystalct1]

For all who are interested:

You have use the orifice coefficient to model losses: C = 1/SQRT(Kent + Kex + Kfr + Kbend)

You can find these values just like you do for pipes.

Then use the computed value for C instead of the default which is 0.8

For my bridge the coefficient only differed to 0.72 with the bend and 0.79 without the bend, so it actually doesn't make a whole lot of difference for a 20 degree bend.