Storm and Sanitary Analysis 2024 Sag Inlet Capacity Limit

[qchalmersKACEH]

The User Guide for SSA 2024 states the following definition for the ponded area above an inlet located in a sag:

My understanding is that SSA calculates the weir (or orifice) flow which is entering the sag inlet, and then determines a depth of flow at the inlet. In theory as the flow along the gutter to the inlet increases, the depth of water around the perimeter of the grate (weir flow) or above the grate (orifice flow) continues to increase as long as the curb is high enough and the road wide enough to contain the flow depth and width. What I am struggling to understand is how the ponded area is used  in SSA. The definition above states that any flow which exceeds the inflow to the sag inlet is routed to the ponded area. How does SSA determine that the flow to a sag inlet has exceeded its capacity?

[qchalmersKACEH]

I have been testing the ponding feature in SSA and have come to the conclusion that the definition of ponded area in the 2024 User Guide is incorrectly explaining it. Ponding at sag inlets does not occur because of an inlet capacity issue. It occurs when the downstream system from that inlet is under capacity, which results in a backwater condition where the water surface elevation of the catch basin at the sag inlet increases above the rim elevation of that inlet. In this scenario SSA starts to store the volume of flow above the inlet in the ponded area. It is not clear if SSA uses the depth of head in this ponded area to determine the flow rate into the inlet or if the depth at the inlet is still determined based on the upstream flow entering the inlet. Regardless, from a design perspective any ponding, shown as "Total Flooded Volume" and Total Time Flooded" in the SSA results is an indication that the stormwater system is over capacity.

[matt_anderson_pe]

Eh, no.

SSA does not have the capacity of back water up and into the surface via an inlet.  It never calculated that.

SSA uses ponding area to determine the depth above the inlet, and calculate the capture curve for the inlet.

During a simulation, the curve is used, not the native inlet capture calculations.

[qchalmersKACEH]

Hi Matt, thank you for the reply.

We can help determine how the model is calculating ponding volume from testing multiple different scenarios with a single sag inlet connected to a single downstream pipe outlet. For example, in scenario #1 I have reduced the inlet capacity at the sag inlet by drastically reducing the grate area to 10" length by 3.5" width.

As you can see the restricted inlet has greatly increased the gutter spread and flow depth at peak flow. At this depth the inlet is operating as an orifice (rather than a weir). The SSA results show a total flooded volume and total flooded time of zero. This indicates that even at a flow depth of 20.44' above the sag inlet (which is completely unrealistic considering a standard curb height of 6") SSA does not calculate a ponding volume.

In scenario #2 I have kept the sag inlet grate at high capacity, with dimensions of 24" length by 20" width. In order to simulate a backwater condition I have restricted the outlet pipe from the sag inlet to a small diameter of 2".

The grate inlet has adequate capacity for the flow entering it, resulting in a flow depth and gutter spread of 0.06' and 3.65'. However the water in the downstream pipe backs up into the inlet and increases above it's rim elevation. A time series plot of the water surface elevation at the sag inlet shows the water level rising from the outlet invert of 2116.23' all the way to the rim elevation of 2119.76'.

Examining the SSA output shows a total flooded volume and total flooded time of more than zero, indicating that although the inlet itself has capacity for the flow entering it the backwater effect has created a ponding volume above the inlet rim.

Would you have time to test the same scenarios for a sag inlet and tell me if you get the same results?

 

[matt_anderson_pe]

Let's try this.
In your example - click the "more" to expose the Hydraulic Efficient and Gutter Depth/Spread per inlet graphs.

This is what the application uses to determine the performance of the inlet. 
Your first set of values is not valid.  If you change Ponded Area - nothing on those charts will be change.

However, if you review your models continuity - its going to be messed up.  The depth on that structure will not be in agreement with your 20 feet - primarily because is assuming the flooded area is only 10 feet.

Your second inlet condition fails because the top of the hydrograph gets lost and is not counted in the surface flooding.

Best case scenario - abandon SSA, and use EPA-SWMM5 for this analysis.  Your answers will be different - by alot.

[qchalmersKACEH]

Hi Matt,

I agree that changing the ponded area at the sag inlet does not change the Inlet Hydraulic Efficiency or Gutter Depth & Spread per Inlet charts as well as the inlet peak flow, gutter spread during peak flow, and gutter depth during peak flow (assuming the model is not fluctuating and providing results that make sense). What I am trying to understand is how SSA is using this ponded surface area input. The 2024 SSA User Guide does not explain how it is being used even though the model will automatically assume a value of 10 square feet if you do not enter a ponding area for a sag inlet. Realistically the ponding area of a sag inlet would change with depth, as the pool of water expands both up and down the curve of the roadway longitudinally and into the roadway perpendicular to the inlet (spread). From what I understand it should not be a constant value.

In scenario #2, could you please elaborate on the top of the hydrograph being lost and not being counted for surface flooding? If SSA was calculating ponding correctly I would assume it should be using the ponded area to determine the amount of head above the inlet rim as the flooded volume changes over time. It appears my assumption here may be incorrect and that SSA is not calculating this correctly?

As I explained to you on a previous thread, I do not have the option of using EPA-SWMM5 at my consulting firm. I am trying to make the best of the tool that my firm is willing to try. For our design purposes we do not want to see any total flooded volume or total flooded time more than zero. It appears that the ponding area input should not matter for my design purposes as long as the rest of my model inputs are correct and the model results not fluctuating.

[matt_anderson_pe]

SSA inlets are "capture" only.   They accept flow using those "curves" and do not process reverse flow.  They have no negative flow.

If you change your option 2 to a manhole and redo - you will see water stored over the rim.  Its been a fatal flaw in the SSA application since Autodesk bought the product.  Great for design when its unlikely you are designing for inlets failing, but horrible to determine where the water goes.

Post your model solution No 2's continuity model or zip it up and post.

[qchalmersKACEH]

"Great for design when its unlikely you are designing for inlets failing, but horrible to determine where the water goes" - This is exactly the scenario I am in. My firm intends to use SSA to design new stormwater systems without any of the inlets overtopping during the design storm.

I have attached my model if you would like to take a look. This is something I have been playing with to try to determine how SSA is working and by no means represents an actual system. Inlet #76 is the one I have been messing with regarding ponding.

[matt_anderson_pe]

Well - then test Inlet 76.

Select Inlet 76, right-click and Convert to a Junction Node.   It should produce about 4.2 feet of head with your 1000 sf of flooding due to the 8 inch diameter pipes.

Modified the Ponded Area to an acre - it produces about 4.1 feet.

Flip it back to a sag inlet. Give it a Sweeper Curb Opening inlet—say 10 feet, with a 6-inch opening height—and keep the same acre ponding area.

Amazing!  The ponding head is gone - and the spread/depth is tiny.  Where did the water go?  It's not accounted for.

BTW - I would not use circular pipes for gutter links.  [My numbers will likely differ as I swapped those gutter links for a simplistic gutter cross-section.]

Ponded area only works with Junctions that allow reverse flow.  Inlets only "accept" surface water and do not surcharge. 

[qchalmersKACEH]

Hi Matt,

Thank you for taking the time to examine my test model. The smallest pipe diameter downstream is 2" (not 8") as this is how I left the model right before attaching.  If I keep the 2" pipe in place, convert #76 to a junction, and run the model without any other changes I get the following results:

This doesn't make sense to me as it is showing no ponding, even though the WSEL is clearly backed up above the rim elevation of the junction. How did you get the results to show a flooded volume for this junction?

I do see the comparison you are trying to complete. If the downstream pipes are backing up the WSEL above the rim of the sag inlet it should cause ponding. This ponding should not disappear if you change the inlet capacity of the inlet. I can test this by causing a backwater condition in the downstream system (with a 2" pipe) and then running different types of grates/sweeper inlets.

In scenario #1 I purposely restrict the sag inlet with a very small grate and get the following results:

The calculated gutter spread and gutter flow depth appear to be based on the inlet capacity only and are incorrect as ponding should be occurring due to the backwater condition. If I look at the excel inlet report however I do see that ponding is occurring:

In scenario #2 I greatly increase the inlet capacity with a sweeper inlet and get the following results:

Again, the calculated gutter spread and gutter flow depth appear to be based on the inlet capacity only and are incorrect as ponding should be occurring due to the backwater condition. Also again if I look at the excel inlet report ponding is occurring, and its the exact same volume and time as the much smaller inlet:

This appears to show that SSA is correctly determining when ponding is occurring but is displaying it in the results incorrectly. A gutter spread and depth of flow should NOT be shown if ponding is occurring unless they are correctly calculated based on the ponding volume and cross section of the pond which SSA does not appear to be doing.

Regarding a gutter cross section vs pipe cross section for bypass links, I have been trying to determine how much this effects the model and also have referenced some other modelers to see what they do. FDOT uses a direct link, which is odd as I have been unable to get this to work in any SSA model. This modeler recommends using a pipe with a length of 1' for all bypass links. I have tested this myself by making a copy of a model where one version is 1' long bypass pipe links and the other is exactly the same except all the bypass links are a gutter cross section. I overlaid the results in SSA and was unable to see any real difference:

Would you have time to run a similar modeling comparison and see if you can find differences in the results when bypass links are modeled as pipes versus gutters?

[matt_anderson_pe]

Oh, bloody hell.

This is going to take some time.   I guess I cant say its a bug, but something that StormNET never had capabilities for, and Autodesk never spent the money to extend.  XPSWMM does it better but it has bugs with curb opening inlets in sag.  InfoWorks ICM does 1D best, but gutters need to be gutters as velocity of those gutters is what determined capture.  InfoDrainage doesn't consider inlet surcharge/capture during the simulation.  TUFLOW-SWMM handles this best in the 2D surface and 1D underground with HEC-22 inlets.

[qchalmersKACEH]

No need to rush and I appreciate your help.

[qchalmersKACEH]

I decided to test turning the ponding feature of SSA off entirely to see how it changes the model results. To my shock it makes absolutely no difference, even when modeling situations where ponding would be occurring at sag inlets. Interestingly SSA also still calculates a total flooded volume and time when the ponding option is off. Here are three separate comparison time series with the exact same model, ponding on vs off.

 

 

 

[fcernst]

Here’s a good one, but SSA is dead now anyway ..not supported anymore..

https://forums.autodesk.com/t5/civil-3d-forum/ssa-inlet-upwelling-flow/m-p/5664854#M317112

[qchalmersKACEH]

Hi Fred,

I agree with you in that I am disappointed in Autodesk's support for SSA. What I don't understand is why the program was developed at all if Autodesk didn't intend on supporting it. The 2024 SSA User Guide is 544 pages, which clearly took quite a bit of time to write and yet the support from Autodesk itself is nonexistent for anything which isn't answered in the user guide.

[fcernst]

It’s long, long, been over…Dead

[drehner89]

So which is the program to use now?  Infraworks?

[qchalmersKACEH]

My company is looking at Autodesk Infodrainage. Unfortunately it is quite expensive while SSA was free (already included with our Autodesk software package). C3D 2026 has a new analysis tool built in called Drainage Analysis but unfortunately it has nowhere near the capabilities of SSA (for example it doesn't support bypass inlets). Autodesk claims it will be added but who knows when.