Catch Basins in HydroCAD

[ClarkBart]

There is a lot of discussion concerning modeling Catch Basins and their grates on this discussion forum and in the HydroCAD Manual. But, I still have not found a solution/approach to this problem.

I am currently trying to model a catch basin/culvert drainage system. I am getting the error message that says: [58]Pond X peak YY.YY ft above flood elevation. The amount that distance is above the flood Elevation is unrealistic. (I've set the flood elevation at the grate elevation.) So, the manual/help screen says "you must include an appropriate overflow device."

There is no definition of "appropriate device." So, I have experimented with a few outflow devices: custom, broad crested weir, etc. I set this flow just above (.5' +/-) the flood elevation. I have tried to route this flow as secondary and/or discarded. In either case, I get an interesting result.

The result is some wide variations in hydrograph. My expectations are that this would shave the peak off of the hydrograph. What I get is the primary outlet flow drops to 0. The secondary or discarded flow spikes to the full discharge. Then, they swing back with the Primary device matching the inflow and the secondary/discarded device back to zero. the dt is at the finest setting and there is no error message concerning oscillations for the pond/catch basin in question.

Obviously, I don't have an appropriate outflow device. Any guesses on just what steps are needed to model an appropriate device?

Thanks in advance for any ideas.

 

[psmart]

This is probably an issue with the device routing, rather than the specific overflow device you're using.

Make sure each outlet (the pipe AND the overflow) are routed to Primary. Or set the overflow to secondary if you want separate reporting and routing for the overflow. But do NOT route one device into the other or the flow will be limited by the two devices in series.

A weir or horizontal orifice will work fine for the overflow device. Set the invert at the elevation where the overflow begins.

Without an overflow device, the model doesn't "know" what to do with the excess flow and volume, so it will generate additional head on the pipe/culvert as required to handle the flow. In other words, the HGL will exceed the top of the CB.

When you include the overflow device you will still get discharge through the normal culvert/pipe outlet. There is no reason for the primary discharge to drop to zero as you are reporting. This would be a separate issue, such as the device routing discusses above.

If you still have problems please post your HydroCAD project file or send it to HydroCAD support. It saves a lot of time to look at the actual file.


Peter Smart
HydroCAD Software

http://www.hydrocad.net

 

[ClarkBart]

Thanks Peter:

I understand what you are saying about how the excess flow are modeled without an overflow device. Some of the particulars that raise a few questions in my mind are:

1. If a horizontal grate is used (device 2), wouldn't that be normally routed through the culvert (device 1) since they are in series?
2. The grate is treated as weir flow at low heads. So, water must back up to an elevation the discharge over the grate and into the culvert.

So, we use the grate as an overflow device. Wouldn't using a horizontal grate as an over flow device cause the water to short circuit the culvert when it doesn't? Even if you route it as a primary outlet?

3. WE are back to the grate being in series with the culvert. There is also and overflow device at the elevation of the grate. Water is backing up over the grate as an inlet (or as an outlet as the model sees it) and it exceeds the height of the overflow device invert, could could this potentially be showing flow out of the overflow device when it shouldn't?

4. For all of these reasons, modeling the inlet grate in HydroCAD seems to be problematic in a typical catch basin culvert arrangement. So, simplify and leave it out. The only overflow the model seems to be truly set up to look at is the condition when the pipe is undersized and the water level builds up in the CB to the point where it overflows the top of the basin. So, the basin top is an outlet. My preference would be to route it as secondary or discarded if it flows out of the watershed being studied. So, would it be best to identify this outlet as a custom weir using the shape of the road cross-section as the information to input? Again, trying to simplify, maybe it doesn't matter?

I've attached the model that I have been working with. Look at Pond 1A in the 50 and 100 yr storm.

Sorry for the ramblings, but, this issue might warrant further treatment in the manual and help screens so, that we can see what the best practice is here.

Thanks again, Bart

 

[psmart]

Since you're modeling overflow, you MUST use parallel routing for the culvert and overflow device. This can be done by routing both to Primary, or one to Primary and the other to Secondary.

As you point out, the grate is often ignored in a HydroCAD model. You could use the grate to control the overflow, but if the grate is in any kind of sump situation ponding will occur, and the control point would be the roadway crown or other passage that ultimately controls the overflow.

For further details please see

http://www.hydrocad.net/sewers.htm

Peter Smart
HydroCAD Software

http://www.hydrocad.net

 

[TerryScan]

ClarkBart,

Here is something else you may try that I have had success with. Rather than using "Catch Basin", use Detention Pond for your node. Enter the actual storge area within the inlet, then the actual storage above the grate (in my cases it may be something like a half acre 0.5ft above the grate, 1.5ac 1.0ft above the grate etc.). This usually produces sensible WSE above the grate along with appropriate pipe hydraulics.

 

[ClarkBart]

So, in HydroCAD water can leave the watershed only through the final outlet or through infiltration.

There is still something that doesn't make sense in the way flows are treated with the over flow device. Possibly, it doesn't matter in the long run, but, take a look at the attached file for a few mark-ups on the hydrograph table.

Thanks again, Bart

 

[psmart]

Water will leave the system through a discarded outflow or any unrouted outflow. For this reason it's important that you route all the outflows as appropriate. For example, your CB overflows should probably be routed to secondary (not discard) so their flow can rejoin the system at an appropriate downstream point. Discarded is usually used just for exfiltration or other true "losses". But a CB overflow if rarely "lost", it just follows a different path.

I believe the apparent differences you've noted on the PDF report are due to varying tailwater. Since the culvert is exiting to another CB with a rapidly rising WSE, this will reduce the flow through the culvert, despite the increasing inflow. That's why the overflow increases as the pipe flow decreases. Makes perfect sense.

If you remove the tailwater (just disconnect the outflow) the results will be as you expect.

Routing time-varying flows through a pipe network is a non-trivial matter, and some time may be required to fully understand the results. But you don't need to get buried in the details to create a working model.

Peter Smart
HydroCAD Software

http://www.hydrocad.net

 

[psmart]

If the above-grade storage is significant, it should definitely be included in the model as TerryScan suggests. Depending on the available storage, this may cause significant detention and peak reduction.

Peter Smart
HydroCAD Software

http://www.hydrocad.net

 

[ClarkBart]

Thanks for all of your thoughts and patience answering my questions. I still have some oscillations in the model that I can't seem to find a source for, but, I'll keep working.

I can appreciate the behind the scenes complexities of software that makes the whole thing seem easy. Thanks for your work.

Bart

 

[psmart]

Given the small Tc values in your model, you may need an even smaller time step than the normal minimum of 0.01 hours. This can be achieved by increasing the number of decimal places for the "TimeInc" parameter.

Open the Settings|Units screen, click Custom, scroll down and select "TimeInc", and increase the decimals.

Peter Smart
HydroCAD Software

http://www.hydrocad.net