How Do Cross Section Nodes Control the Shape of Channel Cross-Sections?
When you connect a channel between two cross section nodes, the cross-section of the channel is interpolated between the two cross section nodes. When there is only one cross-section node at either the upstream or downstream end of a channel, then the channel will have a constant cross-section, as defined by that one cross-section node.
If there is a cross section node at each end of a channel, then the channel will start with a cross section as defined in the upstream cross-section node, and will make a transition to the cross-section defined in the downstream cross section node.
When you connect a channel to a conduit at a cross-section node, a transition is added between the channel and the conduit. You can specify the type of transition in the Property Editor for the cross-section node as either Gradual or Abrupt. If you specify Abrupt, the top width of the channel cross-section is used as the length of the transition part. If you select Gradual, you enter a value for the Transition Length. If the Transition Length is larger than the top width of the cross-section node, the Transition Length value is used as the length of the transition part.