The hydraulic grade in a conduit immediately downstream of a pump has a spike to an unreasonable elevation such as 1500 ft. where the hydraulic grade before and after it are relatively normal. How can I fix this?
This can be encountered when a model has bad data entry and causes the results to be unstable in one or more locations. The wiki article found here for troubleshooting unstable SewerGEMS or CivilStorm models that use the implicit solver will help. The solution below will focus on what to do with pumps in this situation that appear to be the cause of the problem.
One possible cause is that the pump is operating too close to the shut off head, as opposed to a point closer to the center or best efficiency point. Here is the original pump curve:
The screen shot below displays a pump curve (blue line) operating too close to the shut off head, which is a flow of 46 gpm. The red line is the system head curve.
The system head curve was adjusted using the GVF-Convex (SewerCAD) solver to bring the design point flow much lower, so it intersects at a reasonable point with the system head curve. This was achieved using a trial and error process. The GVF-Convex solver was used because the pressure engine is more stable for pump calculations. The adjusted pump curve and pump curve intersecting the system head curve are shown in the screen shots below:
After the changes are made and the new pump curve is assigned to the pump the spike in the conduit hydraulic grade occurs, but the scale is only 12 ft rather than ~1500+ ft.
Troubleshooting unstable SewerGEMS and CivilStorm results using the Implicit solver