Flow Diversions

This tutorial illustrates a procedure for diverting flow in two directions and then continuing the analysis. A possible scenario that this would be used is for the design of a Water Quality Swale or just a splitter of some type. Before you look at this tutorial, you will need to have gone through the Simple Detention, Simple Layout, and InLine tutorials.

Lets setup the example:

We are punting on this one and just modifying the Inline tutorial. Basically, assume that you have flows entering a detention node of some type. The purpose of the node is to split the flow in two directions. From the Inline tutorial, we will use node Pond and just change the control structure (which was just an orifice to a combination type control structure consisting of an orifice and a overflow riser).

To refresh your memory, this is what the Inline layout looked like.

In the Inline tutorial, the discharge structure was an morif structure. Here, we need to make it a combo consisting of an orif and a riser. We are also going to assume that somehow the water that goes into the orifice is diverted in a different direction from the water that goes over the riser. (Construction detail that I will leave to you to work out.)   Based on the InLine tutorial, we know that the peak stage of the 10 year design event is 106.9154 ft.  We will use that as our overflow weir.  Create new discharge structure called oweir.  Make it a Riser type and set the starting elevation at 106.92 ft. The next step is to define a combination discharge structure: that consists of the orif and oweir controls.  Note that we have checked the "split total outflow . . ." checkbox.   The next step is to tell the program to use the combo control structure instead of the orif.  We do this by modifying the Pond node. We are essentially done, but we have forgotten an important step.  The orif structure was originally designed by the program to discharge the 2, 10 and 100 year events.  If you look at the control, it has thre orifici defined.  Now that we have replace the 100 year outflow with a weir, we need to edit the orif control to reflect only two orifices.  Open the Orif control and press the zero button next to the "Distance 2nd to 3rd (ft):" and "3rd Diam (in)" fields. So far, everythng is pretty straightforward, we have replace the control structure in the Pond with a combination structure that consists of an orifice and riser.  Our intent is to route the runoff that goes through the orifice in one direction and the runoff that goes through the riser in another.  The question you should be asking is how do you tell the program which control structure goes in which direction? The answer is by assigning the contrib discharge in the downstream reach. To tell the program that reach P-004 should get the discharge from the Orif discharge structure, just assign it in the reach!  Double click on reach P-004 and assign Orif to the Contrib Discharge field.   At this point, the obvious guess is to simple connect another reach to node N-004 along with its connecting node. It would represent the flow for the riser discharge structure. WRONG! The program will not let a node have multiple outlets from a node! The strategy would be to either create a new layout and redefine the node or put another segment in the same layout. In this example, we will just put it in the same layout. Add the new nodes to the layout as shown below. The procedure is to define a duplicate node representing the pond node.  The layout calls it a ShadowPond.   We also need to define a new reach P-008 to carry the riser flow and of course a new dummy node, called OtherDirection. Notice that the ShadowPond only needs to be a dummy node, not another rlpool node. Now set the Contrib Discharge in P-008 to the OWeir discharge structure.    Almost done. When you compute the layout the FIRST time, there is a good chance that there will be no flow in the P-008 reach. The reason is in the sequence of computation. The program could have computed the smaller network first. If it did, nothing would have been assigned to the riser control structure yet.  Compute it a second time.  OK, what hasn't been explained. Look at the HGL elevation for nodes Pond and ShadowPond. They are different. The ShadowPond is not correct. The primary reason is becaues we never set the proper elevations of the P-008 pipe and ShadowPond node in this example.  Even if we did, the HGL for the Pond and ShadowPonds would not agree.  The reason is because the HGL for the ShadowPond node is based on the HGL computation for the flow going through the riser while the HGL for the Pond node is really the maximum water surface elevation from the detention pond computation. It should always be higher than the HGL computation from the ponds outlet pipe and should always govern.  The program will always choose the higher of the two to present for the detention node.  It would not hurt to make it a habit to always check to make sure that the HGL elevation reported for detention ponds is the max water surface from the pond routing and not the HGL elevation.  The program can only report the higher of the two.