Modeling Barrier Separated Grates

Hello Robert,

For part one, hydraulically you can use the approach mentioned in this article (which I just edited a bit to include pictures):

Modeling one structure with two inlets (grates)

Basically you would model it as two catchbasins with a short, "fake" pipe between them.

Now, this doesn't work out quite so well for a physical model in Subsurface Utilities, since there physically would not be two separate catchbasin structures. Perhaps you could model it as one catchbasin structure with appropriate feature definition for the appropriate physical representation, and set up the inlet to be equivalent to the two grates. Of course, each grate would have a separate contributing area, but you may be able to do a hydraulic analysis of the capacity for a range of storm events using approach #1, then set up a catalog inlet to use for the physical model that provides hydraulically equivalent results. (perhaps using the inflow-capture curve method).

If the contributing areas of each side of the inlet are significantly different, it could pose a bigger challenge, though that may depend on which results are important to represent in Subsurface Utilities (spread and depth?). I'll consult with some colleagues for further advice.

Thanks Jesse: I suspected that would be the answer hydraulically but struggling to reconcile the physical model. I'll follow up with some information when I've got a workflow in place.

Hi Robert.

I've been thinking about this. I haven't tried it, but if the hydraulic solution is to use two inlets, then the SUDA solution could use the same technique.

I'm thinking that you could use three-sided cells for the top and bottom of each inlet. I'm picturing this as cutting the chamber in half vertically, along the line of the centre of the barrier. That way, the extrusion between the top and bottom cells should work. As Jesse notes, you would need to connect the two inlets with a pipe that doesn't really exist, but, since it's inside the chamber, it wouldn't normally be visible.

The connection regions for the two bottom cells would need a little thought, because you need to place the connecting pipe between them, and also allow the correct location of the trunk pipes. Because the trunk pipes would actually be connected to one of the two bottom cells, then you may need to manually adjust the invert elevation of the other inlet to suit.

This technique should also be able to cater for the situation where the grates on either side of the barrier can have different cover levels, which will sometimes occur.

Regards,

Jon

Thanks Jon:

Your suggestion is one that I have been considering as a solution. It is the most elegant solution for the physical model, especially when the two grates are different elevations. The down-side to this approach is that the two inlets would get counted as two catch basins.

The other option is to create "dummy" nodes to account for the hydraulic "trick". Such dummy nodes could be easily ignored in reports. The down-side to this approach that the physical model would be incorrect when the two grates are at different rim elevations.

In the short term, barring an epiphany for something better, I'll need to develop examples of both and let my client decide.

Robert

The final solution is much more elaborate:

You need the first inlet to be complete in itself because sometimes there will be a total structure but only one inlet, such as high side super.

Then the other side needs to be a skeleton. This skeleton is not intended to be used alone because there is no vault. Hydraulically, it is complete, but for the 3D model, there is only a skeleton of a vault. 

And you need a false pipe between the two.

More details coming soon on our web site (EnvisionCAD.com).

Robert