Culvert Modelling

[dmad1]

Does anyone know the proper procedure and software to model a series of 3 culverts. I have developed the hydrograph at the first (upstream) culvert as input data. However I think that I will need to do an unsteady flow analysis to determine whether there is detention at the culvert inlets, or whether the headwater at the downstream culvert inlets creates the tailwater for the upstream culverts, etc.. Perhaps there is detention at each culvert entrance?

Also, these are very old culverts under an interstate highway that have been extended over the years, and one in particular was originally a brick arch section that has been extended on one end with a conc. box section, and on the other end with a pipe section. To further complicate things, there is also a 40 degree bend at one of the transition sections.

The programs that I have used, HY-8, culvertmaster, etc. do not allow for multiple barrel sections and significant bends. If someone has dealt with these problems before I would really like to hear from them.

 

[DPB]

For a series of culverts AMI HydroCAD or Haestad Methods PondPack will work. HydroCAD gives the "most bang for the buck", but has a few limitations.

The "old culverts" may be modeled by increasing the default entrance loss coefficient (Ke) and the contraction coefficient (Ce). The bend MAY even be modelled as a pond with essentially zero volume with a culvert outlet. This will allow you to apply an additional (Ke) and (Ce) if you believe it is warranted.

Both programs account for tailwater elevation in their analysis. The headwater for the most downstream culvert is the tailwater for the next upstream culvert, and so on. Storage effects are calculated by the storage indication method.

The downside of HydroCAD is that it will not account for reverse flow in the event of tailwater exceeding headwater. You merely get a warning. PonPack will account for reverse flow, but costs about five times as much to purchase.

Good luck.