Drop structures

Todd - you will probably not have any trouble at low flows. At higher flows you may generate transients, which are most easily detected with a graphical representation of each pipe's hydrograph. If you do not get any consider yourself lucky and move on. If you do, then for this problem a dummy slot conduit equal to the full depth of the MH, a narrow width and a high n will force compliance with this criteria. Even though Wayne has indicated this should not be a problem any more, I still run into it now and then.

Good luck, Ron Kilmartin

Hello All

First, Thanks to all who responded to my question on the outfall elevation. Things are running smoothly now.

The SWMM manual states the no conduit should have an invert above the crowns of all other pipes.

I am modeling a system with several drop structures, and bypasses where this is the case.

The manual states the these are not fatal, but depending on the critical/subcritical decisions made by Subroutine HEAD in the assignment of the junction areas, they may cause instabilities and continuity errors.

What is the lists' experience with this situation? Is it better to break the model into sections at nodes where this occurs?

It does not appear to be a problem, but I am only running dry weather flows at this time.

Thanks
Todd Williams
Parsons Engineering Science

Todd,

> The SWMM manual states the no conduit should have an invert above the crowns of all other pipes.

This was only a problem in Extran version 3 because there was not a default surface area for a manhole. In the numerical solution of the head in a node the area would be zero when the head was between the crown of one conduit and the invert of the conduit above. Since 1988, there is a default surface area for each manhole eliminating the division by zero.

Regarding the machinations of subroutine head - this is usually well behaved unless you have a large conduit connected to a large storage node. Often in this case the model might oscillate between critical upstream and critical downstream, thus changing the surface areas of the upstream and downstream nodes. The changing surface area would cause oscillations in the head and flow of the connecting conduits.

In XP-SWMM32 and Visual Hydro we added an option to make the model allocate half of the conduit area to both the upstream and downstream nodes at all times to prevent this oscillation. You can do somthing similar in EPA SWMM by adding a vertical pipe thus bypassing this section of subroutine head.

Regards,
Robert Dickinson
CAiCE Software

Bob,

What do you mean by a vertical pipe? A riser? Or a pipe with a slot like Ron Kilmartin was talking about?