[Hydrojock]

Doug,

Is this due to the length being defined by the river centreline or something? Any ideas how to fix this?

Sort of. It has to do with the difference in downstream reach lengths in the overbanks versus the true length of the lateral weir. The User’s Manual actually has a very good description of this and how to ‘fix’ it. See page 6-80 or so. Basically, you can manually enter the correlation between river stationing and lateral weir stationing. It’s done in the Weir/Embankment Editor.

[Hydrojock]

Sarfraz,

No, I don’t believe that has been done. CAN it be done? Perhaps. How much programming skill and time do you have? =)

[Hydrojock]

Netra,

KTWW is correct, you can make an interpolated cross section into a normal cross section by deleting the asterisk (*) from the end of the river station number. You can do this in the Cross Section Data Editor under Options>Rename River Station…

Note that you can only delete an asterisk within RAS, you can’t add it back. In other words, you can convert an interpolated cross section to a normal one but you can’t reverse the process. That can only be done by editing the river station name in the geometry file with a text editor.

[Hydrojock]

What are you using for your downstream boundary condition? Is it normal depth? If so, is your friction slope reasonable (i.e. does is it roughly equal to the bed slope)?

[Hydrojock]

John,

Can you give a screen shot or something of your geometry? It sounds as though you’re using a junction to connect two reaches of the same stream. This is not recommended as RAS computes losses across junctions differently than it does from one cross section to the next within a single reach.

Junctions should only be employed when modeling a tributary or a significant split in channel geometry (as in a looped or dendritic system).

[Hydrojock]

Ross,

A couple of quick thoughts:
1) do you have an initial water surface elevation defined in the storage areas?
2) the profile plot you provided shows the culverts are quite submerged. at the start of the simulation, is the water surface below the invert of the culverts? if it is already above the invert of the culverts for the first time step, the model might be having difficulty coming up with a valid starting solution. best to start below the culvert invert and gradually bring it up.
3) presumably you’ve checked the obvious things like the Cd for the culverts, geometry integrity (elevations for all connections make sense), etc…

[Hydrojock]

Godfrey – can you clarify a bit more what you’re trying to do? It sounds like you might have observed elevations and want to determine the flows in the model necessary to produce those elevations?

[Hydrojock]

drainage321,

1) It would be fine to end your bridge data mid-cross section. You might consider defining the crossing as a multiple opening in that case. The bridge section and a conveyance section out in the overbank. Cross sections bunching close together in the overbank is usually not a problem, so long as your downstream reach lengths are entered correctly.
2) I wouldn’t worry too much about how the x-y-z perspective plot draws. It’s a relatively crude rendering. Double check to make sure all your data is entered properly. The misalignment of the u/s and d/s bridge opening likely has to do with the zero station of the cross sections. Is you model geo-referenced? There could be a simple stationing error. Either way, the program is only 1-D so it doesn’t know the difference.
3) Breaking the model into two sections is fine. A little extra work, but if it makes it more conceptually ‘simple’ there’s certainly no harm in it.

regards,
James

[Hydrojock]

If I calculate the expansion length, I come up with a downstream expansion length of 1,126′. Does that truly mean my downstream x-section should be located that far away – that is almost to the next bridge?

That’s a long way between cross sections, as you note, which probably violates the gradually varied flow assumption built into RAS. This is a common issue, especially with wide floodplains. The best way to handle it is to use additional cross sections within the contraction/expansion zones to keep the d/s reach lengths reasonable. Just be careful how you define your ineffective flow areas (gradually contracting/expanding as you move towards/away from the bridge) and set your expansion/contraction coefficients properly.

[Hydrojock]

Okay. Thoughts:
– If you’re going to have flow in that right overbank (beyond the roadway) you’ll need to be careful how you define your ineffective flow areas. You want to make sure that you maintain flow consistency for a given event along the entire study reach. For example, you don’t want the 50-year flood overtopping the roadway and flowing in the overbank at one section then not overtopping and flowing at the next; that’s not consistent (where did all that flow go? etc…)
– No idea why RAS would be crashing like that. It’s usually a very stable program. The installation process is pretty short, so I’d try re-installing it. Might save a bunch of time in the long run.

Good luck!
James

[Hydrojock]

You ask some good questions. I’d first like to ask a couple of my own:
1) What’s the objective of the modeling? Are you trying to determine flooding extents, design the bridge, etc… The ‘best’ course of action might well depend on your objective.
2) Are you running a steady-state or unsteady-state model? (constant flow vs. a hydrograph)
3) What range of flows are you planning to model? Specifically, do you know if the flows will remain contained within the main channel? When high flows occur, do you know if the right overbank/floodplain is wet? Is it able to flow?

You have the ability to ‘make’ the model behave in quite a few different ways. The more information you have about how the system is supposed to behave (have you or anyone else seen it during flood stage?) the more accurate your modeling will be.

To you specific questions:
– Length of cross section is widely dependent on what you’re trying to model. In general, make the cross section wide enough to fully contain whatever flow event you’re modeling. If the flow is contained within your roadway/levee, then the right overbank beyond the roadway is really not necessary, etc…
– Approaching the roadway as a levee is probably fine, depending on what your objective is (see above)
– You will have to use the tools available to you within the model to ‘train’ the flow as it approaches and flows through/around/over the bridges. Ineffective flow definitions are key to accomplishing this properly. See the User’s Manual for a good description of how they should be properly employed.
– The additional drainages that flow into the main channel through your study reach can probably be ignored, depending on what you’re attempting to model, of course. Is it reasonable to assume co-incident peaks for the drainages? If so, you might need to look at adding their flow to the model, etc. As to the geometry of the drainage channels, you can probably ignore them. (again, it depends on what you’re modeling…)
– Finally, I’m not sure what to make of RAS crashing when you try to use multiple blocks of ineffective flow. That’s pretty odd. Is it the program (RAS) that crashes or the entire OS? Perhaps try re-installing RAS.

Hope this is helpful. Let me/us know how it goes!
James

[Hydrojock]

Pallo,

That would be cool, but alas, it’s not a feature in RAS. That said, I’ve ‘stumbled’ upon several features like this in the past that were not described in the User’s Manual. If you figure out a way to do it, make sure you re-post!

James Heyen, P.E., CFM
Senior Hydraulic Engineer
WEST Consultants, Inc.

[Hydrojock]

Hello RPL

It sounds like you have a simple issue. There is an option in steady flow analysis where you can specify if you want RAS to calculate critical depth at all cross sections. By default, it does not – only calculating it where it is necessary. The User’s Manual goes into this in detail.

Go to the Steady Flow Analysis and select Options. Then select Critical Depth Output Option and choose Critical Always Calculated. When you re-run your model, you should have critical depth solved at all locations.

Regards,
James Heyen, P.E., CFM
Senior Hydraulic Engineer
WEST Consultants, Inc.

[Author]

Posts