Storm Sewers Version 11 - HGL Not Right

[Anonymous]

I've noticed that the HGL isn't calculating or displaying correctly in Storm Sewers Version 11 in some instances. I've seen several examples where the HGL is sloping the opposite direction of the flow, which is not possible. Here are some examples.

[Anonymous]

try to change the direction of the flow of the pipe.

in my case, I used storm cad and sewer gems for hydraulic model and calculations

[Anonymous]

Thanks for the tip. I gave this a shot and it did not fix the problem. I revised the pipe network to make sure the Pipe Flow Direction Method for all pipes was set to "By Slope", then exported the network to Storm Sewers and re-ran the calcs. Unfortunately, I got the same results with segments where the HGL and EGL were sloping opposite the flow direction, which isn't correct.

[Hidden_Brain]

usually the HGL adverse slope is caused by a hydraulic jump occurring inside the pipe/culvert, both under inlet or outlet control conditions.

[fcernst]

"..where the HGL is sloping the opposite direction of the flow, which is not possible.""

Why do you keep saying that? Remember your water surface profiles from Hydraulics?

Do you have any laterals coming in to this system?

Can you post the file?

[Anonymous]

Yes, I understand when going from supercritical to subcritical flow there will be a hydraulic jump, which is a local jump in the HGL. The supercritical flow into the jump will still have an HGL that slopes in the direction of the flow until it hits the backwatered condition (which will also have a HGL that slopes in the direction of flow) from the subcritical flow, then the jump will occur over a very short distance. This may be what's occurring here, but that's not what the HGL is showing. I would expect the HGL from the downstream junction to propagate upstream to the point where it intersects with the supercritical flow.

What may not be clear in the exhibits is the EGL is also sloping opposite the flow direction. Even with a hydraulic jump, the EGL should always be sloped in the direction of flow (water flows down hill).

I think the program is not recognizing the boundary condition at the downstream junction when calculating the upstream HGL & EGL.

[Anonymous]

Thanks for your input. I do remember my hydraulics classes and that's why I'm questioning the results. I think what you're showing below is the how the HGL changes at transitions from supercritical to subcritical flow. As I mentioned in a previous reply, it could be that a hydraulic jump is occurring, but that's not what the HGL is showing.

Also, the EGLs are also sloping opposite the flow direction. Even with a hydraulic jump, the EGL should always be sloped in the direction of flow (water flows down hill).

There are laterals coming in and changes in flow at these locations. I think the program is not accurately accounting for how the water surface from those flows would propagate up the system.

I've attached the file.

[fcernst]

Hi There-

I have to run to a meeting, I will check later.

My first thoughts were that you had laterals coming in and this pipe has a small flow that creates an M1 profile from the backwater. It's in the mild slope category due to its parameters, no hydraulic jump.

[fcernst]

I don't see the attachment..

[Anonymous]

Sorry, hopefully attached now.

[fcernst]

Yes, with that flow of 1.3 cfs for the 24" you are just barely supercritical at Fr = 1.04, and critical slope of 0.46%. The 0.39' flow depth is essentially critical depth.

[Anonymous]

Sure, but this is all beside the point since the software is showing the EGL sloping in the opposite direction of the flow, which is really what I'm concerned about. This simply isn't correct - water flows from higher energy to lower energy. Do you agree?

[Hidden_Brain]

EGL = HGL + velocity head.

I might be missing something, but with a rising HGL and an increase in velocity head in d/s direction , wouldn't EGL rise d/s?

[fcernst]

Yes, you're right about the EGL's.

Take a look at Line 42 in your model.

It seems to be having trouble with with these pipes that are so close to critical slope.

[Hidden_Brain]

I found these discussions online that might be helpful:

https://communities.bentley.com/products/hydraulics___hydrology/f/haestad-hydraulics-and-hydrology-f...

https://communities.bentley.com/products/hydraulics___hydrology/w/hydraulics_and_hydrology__wiki/101...

different software but same topic.

[fcernst]

EGL can't rise downstream without pump energy added.

Here's the classic jump diagram showing vhead being converted to phead + losses.

[fcernst]

Brought this into Analyze Gravity for comparison:

[Anonymous]

Thanks Fred, the results from Analyze Gravity look like what I would expect. They show the pipe segment backwatered from the downstream junction, creating very slow subcritical flow with very small losses due to friction, so essentially flat HGL and EGL, but definitely not reverse sloping.

I think what Storm Sewers should be doing is calculating the HGL (and EGL) using a backwater analysis starting at the downstream junction for these segments to account for the backwatered condition, and also a frontwater analysis to identify any hydraulic jumps (as discussed in the StormCAD discussion linked above).

I think it's clear that we've identified an error in the way Storm Sewers is performing these calculations. I'm not sure how best to notify Autodesk, but they probably monitor these discussions. Hopefully they can provide a fix in the next version of the software.

[fcernst]

Not so fast.. 🙂

I ran it in EMS mode with fix EGL Discrepancies toggled on, and get similar results to Analyze Gravity, and another software product.

It probably needed more iterations to converge on these critical slope pipes.

[Anonymous]

That looks good! For some reason I can't get the same results in my model. I ran it in EMS with the calc settings below. Do you have anything set differently?