2D Culvert Instabilites

[cameron]

Anyone have tips on how to get 2D stable? I converted 25 bridges to culverts with equivlent opening area. Does not matter what time step I try or parameters I change, the discharge through through the structures oscillate even though I have a constant flow. I have tested low, medium, and high constant flows and all show the same thing.

I am thinking of maybe increasing the inverts to see if that helps. I am also going to try to use a rating curve from the 1D HEC-RAS model instead for the opening.

[Anonymous]

I’ve had similar stability issues. Shorter time steps are better.

If you have any bridges where the water does not hit the low chord, just remove them and the model the bottom of the bridge opening with the terrain. If need be, you can add any piers to the terrain.

[cameron]

We are modeling up to the 3,000 year event for some probabilistic modeling so all bridges are impacted.

[Anonymous]

Are you using the courant advanced time step manager? If so turn it off. Whats our Froude mixing coefficient? Have you tried to isolate just one of the bridges/culverts? Im assuming this is full 2D? Whats the DS boundary condition? and how far is it from your nearest bridge/culvert

[cameron]

I have tried using the variable time step with Courant and with it off.

For the Froude mixing coefficient are you talking about the Eddy Viscosity or the something else? I usually don’t use the eddy viscosity, but can add it in to test it.

I have not yet isolated the 25 bridges, but that is my next step. With a time step of 0.1, some of the bridges give ok results now so I am skipping those. I am placing a flow upstream each bridge one by one and testing different things to try and fix each problem structure.

It is full 2D model and the DS boundary condition is a rating curve from a 1D FEMA model which is 1 mile downstream the closest bridge and 2 miles downstream the project area.

I plan on running about 2,500 runs so I need to optimize the model to run fast.

[Anonymous]

Then adjust your htab parameters for all these structures to either much higher or much lower. add a max flow and also only run the model for a couple hours at a time until you stabilize

[cameron]

HTAB parameters for the culverts in 2D? In the Connection Data Editor there is no option for HTAB.

There is no 1D in these models so no HTAB.

[Anonymous]

Is it possible the terrain upstream of the culverts is uneven and causing unstable flow into the culverts. Maybe try flattening the few rows of cells leading into the culvert entries to smooth out the flow.

[Anonymous]

In version 5.0.3 there is so im assuming you are using 5.0.5 then

[Anonymous]

NO i meant variable for mixed flow options. The viscocity coefficient shouldnt play a factor. But now that i realized you are using 5.0.5 that suggestion is out due to mixed flow obviously always being on for unsteady models. Try reverting back to a previous version though. I dont use 5.0.5 yet at all…only for GIS features that RAS added to help out efficiency

[Author]

Posts