[Chris G.]

First make sure your model is running to completion without going unstable (make sure there are no “red” bars). Also, make sure your detailed output interval is small enough so that you have output profiles to look at.

[Chris G.]

I believe the floodway will contract and expand at the bridge at the same rate as the ineffective flow area limits. I.e, the floodway should follow the ineffective flow boundaries on both sides of the bridge.

[Chris G.]

In the initial conditions tab of the unsteady flow editor. I don’t believe it changed locations since the last version.

[Chris G.]

It’s no problem. However, you must have at least one cross section between the bridge and any part of the lateral structure.

[Chris G.]

Try running a steady flow version of this project. Then look for errors (summary of errors warnings notes) and places where the computed water surface elevation defaults to critical depth. Get it all fixed up nice in steady flow, then retry in unsteady. In unsteady, make sure your initial conditions are good and consistent. For example, if your first time step flow is 100, then your initial conditions flow should also be 100. Make sure your computation time step is appropriate (not too big). And make sure you have enough water in your reach (RAS will crash if it goes dry, or even too close to dry). The steady flow version you run should reveal if this is a problem.

[Chris G.]

longitudinal slope is the legnth along a reach divided by the change in water surface elevation (more precisely the change in energy elevation) over that length. x is the length, y is the elevation change.

[Chris G.]

Here you go. The main channel reach length follows the thalweg of the channel, the overbank reach lenghts follow the assumed center of mass of flow in each respective overbank.

proceedings esri.com

[Chris G.]

The best thing for you to do at that stage would be to read through the Users Manual and Hydraulic Reference manual where they address lateral structures. Then open up one of the lateral structure example projects and see how its done.

[Chris G.]

Keep in mind that culverts in RAS are not meant to model tunnels or pipe systems. They are meant for your typical road or highway crossing culverts. That being said, there is no limitation to the number of culverts or length of culverts you can have in your model. BUT, a single culvert cannot have a change in it’s shape, slope, or other basic geometric/hydraulic characteristics. 90 degree turns are no problem in HEC-RAS, just keep in mind that turns are purely geometric-they do not factor directly into the computations. If you want to account for energy losses due to bends (or turns), you can only do this through adjustments to Manning’s n values or contraction/expansion coefficients.

[Chris G.]

There are too many steps to include in this reply. A thorough read of the User’s Manual will help you get your sediment model set up.

[Chris G.]

A run-time error is a bug in the source code. You should send a bug report to HEC. Until you hear back from them, you can go through all of your input data and make sure it is all correct and logical. If input data is all good, you may want to try a different version of HEC-RAS and see if that helps.

[Chris G.]

Great question! And I hate when this happens, but unless there is a gage in the general vicinity of your desired downstream boundary location, you’ll either have to extend your model downstream to an established gage, or all the way to the estuary.

[Chris G.]

Are you sure the junctions are “snapped” in ArcGIS?

[Chris G.]

RAS cannot handle negative (upstream) flow in steady flow. Good news is you can run unsteady flow and have flow in both directions for a given reach.

[Chris G.]

Your model is not robust enough to take a wide range of n values. I’d say an n value of 0.0135 is way too low, anyway. Even if on concrete, a dam break flood wave will be very turbulent with a lot of debris-higher n values are warrented. Try very high n values. Get the model stable. Then slowly lower the n values, running the model between each change. When the model crashes, find and correct the source of the error, then continue on lowering your n value.

[Author]

Posts