CivilStorm - Pond HGL not matching outlet pipe HGL.

I have a pond with an outfall in and outlet out.   The HGL in the pond and outlet pipe are not matching up.   What are some potential causes for this?  or is this typical?

Profile:

Layout through pond:

Parents
  • Here are a few other considerations in addition to Scott's answer:

    1. Check to make sure the outlet structure (POS-1) is configured to support forward and reverse flow.
    2. Models with ponds are often more stable when using the Explicit (SWMM) solver. (see "Ponds and pond outlet structures" in the implicit troubleshooting article) So, if you are using the Implicit solver currently and item #1 above does not help, consider switching to the Explicit solver. I recommend changing the Explicit solver's "Routing Step" to 1 second and the "SWMM Output Increment" to 0.025 hours.


    Regards,

    Jesse Dringoli
    Technical Support Manager, OpenFlows
    Bentley Communities Site Administrator
    Bentley Systems, Inc.

  • I've have gone through and made some changes.   However, now I am running into a  HGL in the outlet pipe that matches the exit of the pond, but it slopes up after leaving the pond.  The EGL matches the HGL throughout the profile, so I don't believe I can attribute this to hydraulic jump.   Is there something I am missing here.

    Profile:

    Solver:

    Outlet Structure:

  • Thanks for sending the files.

    In the properties of the composite outlet structure (click the name "Composite Outlet Structure - 1" in the Composite Outlet editor), you will see that the Tailwater is set to "Free outfall". This means that the pond outflow will not be influenced by the downstream tailwater, and that is why you see the downstream HGL rise above the pond water surface even though the pond experiences outflow.

    If you choose "interconnected ponds" for the tailwater type and select the downstream pipe to get the tailwater range, the program will be able to account for the downstream tailwater. However, you will notice that the Implicit solver will struggle with this and results will still be problematic. The Implicit solver tends to not do as well with tailwater (from the time-elevation curve on outfall "Ex.POND") interactions with ponds and outlets, and the SWMM solver is a better choice for this situation. The following article mentions this: Troubleshooting unstable SewerGEMS and CivilStorm model results using the Implicit (SewerGEMS Dynamic Wave) solver

    Your earlier screenshot suggested you were using the Explicit (SWMM) solver but the scenario you're running (25 year) is using the other calculation option set configured to use the Implicit solver. To use the calculation options with the SWMM solver, you will need to select it from the calculation options dropdown in the properties of your scenario. See more here: Scenario and Alternative Management

    You will also need to enter a valid "Maximum Gutter Depth" in the properties of your catchbasins as they are set to zero and a valid value is a requirement for the SWMM solver.

    Once this is done, the results with the SWMM solver look reasonable and the pond water surface rises with the tailwater appropriately.


    Regards,

    Jesse Dringoli
    Technical Support Manager, OpenFlows
    Bentley Communities Site Administrator
    Bentley Systems, Inc.

  • Jesse,

    Thanks a ton.  I appreciate it buddy.

    Thanks,

    Tony

  • Jesse,

    I had two more questions for you.

    1) The initial state of the profile shows a crazy high EGL and doesn't show the anticipated initial condition HGL that I would expect (as shown in the image below).

    2) The max HGL at some inlets in the model are really high and I can find any point in the time-series where the HGL reaches these limits.

    Are these issues typical with models?

    Thanks,

    Tony

     5861.BW Middle.stswpkg.zip

  • Hi Tony,

    1) The initial state of the profile shows a crazy high EGL and doesn't show the anticipated initial condition HGL that I would expect (as shown in the image below).

    This appears to be from initial instability related to the high downstream tailwater filling up the pipes at the start of the simulation. 

    Generally speaking most dynamic model runs tend to have some initial instability which you can often ignore. You can use the guidance from the article previously provided to fine tune the calculation options and look for data entry issues, to achieve more stable results.

    2) The max HGL at some inlets in the model are really high and I can find any point in the time-series where the HGL reaches these limits.

    If you look at the "Time to maximum hydraulic grade" result field, you will see that it is occurring very close to the beginning of the simulation for the cases i question such as Ci7. With a Routing step of 1 second and the reporting timestep of 0.025 hours, graphs and profile animations skip over that time.

    Similar to item #1 we can attribute this to initial instability from the downstream tailwater. If you change the Surcharge Method to Slot (calculation options), the results seem to be much better, though some slight initial instability can still be seen.

    One other thing I noticed is that you have several on-grade inlets with no downstream gutter link element. You will see some user notifications about this too. If you intended for some of the catchment surface flow to potentially bypass the inlet and approach the next downstream inlet, you will need to draw a gutter between the catchbasin nodes. You can hold down the CTRL key and click to place a vertex away from the conduit, so you can more easily select the gutter vs. the conduit.


    Regards,

    Jesse Dringoli
    Technical Support Manager, OpenFlows
    Bentley Communities Site Administrator
    Bentley Systems, Inc.

Reply
  • Hi Tony,

    1) The initial state of the profile shows a crazy high EGL and doesn't show the anticipated initial condition HGL that I would expect (as shown in the image below).

    This appears to be from initial instability related to the high downstream tailwater filling up the pipes at the start of the simulation. 

    Generally speaking most dynamic model runs tend to have some initial instability which you can often ignore. You can use the guidance from the article previously provided to fine tune the calculation options and look for data entry issues, to achieve more stable results.

    2) The max HGL at some inlets in the model are really high and I can find any point in the time-series where the HGL reaches these limits.

    If you look at the "Time to maximum hydraulic grade" result field, you will see that it is occurring very close to the beginning of the simulation for the cases i question such as Ci7. With a Routing step of 1 second and the reporting timestep of 0.025 hours, graphs and profile animations skip over that time.

    Similar to item #1 we can attribute this to initial instability from the downstream tailwater. If you change the Surcharge Method to Slot (calculation options), the results seem to be much better, though some slight initial instability can still be seen.

    One other thing I noticed is that you have several on-grade inlets with no downstream gutter link element. You will see some user notifications about this too. If you intended for some of the catchment surface flow to potentially bypass the inlet and approach the next downstream inlet, you will need to draw a gutter between the catchbasin nodes. You can hold down the CTRL key and click to place a vertex away from the conduit, so you can more easily select the gutter vs. the conduit.


    Regards,

    Jesse Dringoli
    Technical Support Manager, OpenFlows
    Bentley Communities Site Administrator
    Bentley Systems, Inc.

Children
No Data