Hec-Ras Culvert modeling 1

[TheEnginerd]

Hello,

I am currently using hec ras 4.1.0 and have a situation where i have the stormwater from a 300 acre drainage area traveling down a stream and the stream becomes a 30" pipe that is 200 feet long, and then the stream picks up again on the other side. The pipe is way under capacity obviously, and what i think happens is that the pipe passes X amount of water and the rest is overland flow (above the pipe) for the 200 feet until you pick up the stream on the other side.
Is there a way to model the culvert and overland flow all in RAS?
This isn't exactly a bridge/culvert scenario because the grading for the 200' of overland flow is not consistent. I couldn't for example just treat this as a 200' wide bridge deck with a culvert because i need to describe how the grade changes at every 25' which is what our state environmental dept. reviewer is requesting.

If i were to model the pipe OUTSIDE of RAS, what would be the best program? I tried to just model the pipe in STORMCAD, and by trial and error kept increasing flow until the hydraulic grade was above the upstream pipe headwall (overland flow). but i don't know if there is a better way???

thanks in advance

andy

 

[froude]

Andy,

You can use lidded sections in HEC-RAS which will allow you to do this. I've seen varying success with this method...it is very finicky based on whether you are using steady/unsteady flow modeling and the ratio of overland to pipe flow. In instances, I've seen very crazy results where several profiles (from 2- to 500-year) cross and zigzag in and out of the lidded sections. But it may be worth a shot if you want everything in one model.

Unless have a more robust model (such as SWMM), your HEC-RAS/StormCAD combo is a good bet! Just reduce your flow in HEC-RAS by your pipe capacities.

Good luck!

 

[gbam]

HY-8 or even just hand calcs for a pipe could be used to develop a rating curve that splits your pipe and overland flows. Then you could model the overland flow using RAS. You may have to play with the model to try to get it all in one run using different reaches. Maybe you could use a junction at the pipe inlet with the rating curve and one at the outlet to bring the overland flow back into the creek.

 

[TheEnginerd]

thanks froude and gbam. i do not want to try to put the culvert into hec ras with the lid either, i may try the rating curve idea and see how the results compare.
thanks again.
andy

 

[TheEnginerd]

ok, now for a follow up question.

i was able to generate a ratings table from Hydrocad. Basically i modeled a small basin of insignificant storage, and can get the pipe flow at each elevation of headwater. HOWEVER! i need to put something as my downstream TAILWATER elevation which would significantly reduce the flow. I'd obviously like to use FREE DISCHARGE, but i know that there will be a flood elevation....

This is a chicken and egg situation. I want Hec Ras to tell me the Flood elevation for the entire stream, but in order to get the proper ratings curve for the pipe, i NEED the flood elevation. i guess i could use the top of bank elevation downstream as the tailwater elevation, assuming that the stream is totally flowing full - that would be super conservative.... i just hope there is a better way

 

[froude]

If you are running subcritical calculations, you could run the "lower reach" (reach downstream of pipe outfall) and use those results to determine the tailwater condition on your pipe.

 

[gbam]

Okay if outside of HECRAS you will need to iterate based on pipe flow and tailwater because this is something you do noit know in the beginning. You can initially estimate the flow split using HY-8 and coding the roadway weir section. HY-8 incorporates the entire overflow discharge to the tailwater condition. I think that HY-8 with weir overflow is a good approximation and may even be enough for your model. You will need to evaluate it. Once you have the rating of pipeflow vs overflow you can code your HECRAS model to rout the overflow and then check you resulting headwater to see how close they are between the two methods. Once can adjust weir coefficient to adjust the weirflow to obtain equal headwater elevations.

 

[TheEnginerd]

thank you for the suggestions.
i actually ended up adding the culverts to the overland cross sections, and added lids above the culverts. The cross section looks complicated, but at least it's all in ONE program. It gives me a result that is similar to what i would have expected. i now need to see how it compares to the method of doing the pipe calculation in another program. i want to see how close they are.