multiple lift stations 1

[gjeppesen]

I have a project that involves manifolding a new force main into an existing force main system with multiple lift station on it. What are everyone's thoughts on accounting for the existing stations. For example, what assumption do you make on what stations are pumping at the same time, etc.
What if you can't get information on the existing stations?

Thanks.

 

[redbridge]

You really need to get the information on the existing lift stations and simulate the system. You will need to use some sort of software for the simulation. The utility should be able give how many hours the station runs per day and the conditions of the pumps. Check the system with all of the pump stations running at the same time. This will be your worst case scenerio. If all the pumps will run and pump the design flow rate than you are fine. Otherwise you will have to use your judgement in coming up with a way to simulate the system. I would also check with the state regulatory agency to see what is required. With multiple lift stations on a system you may not be able to get the flushing velocities at all times so you may want to add an entry point for some sort of pigging device.

 

[francesca]

The other scenario you have to test is your pump running only to make sure it doesn't run so fast that you get cavitation.

I guess you could do some kind of iteration in a spreadsheet or maybe even a goalseek, but setting up that kind of spreadsheet, depending on how many pumps you have, will probably end up costing as much in billable hours as a piece of software will.

I do NOT recommend SewerCAD for this if your controling point on the trunk forcemain changes depending on the flow rate. SewerCAD cannot handle an ARV and you end up having to fudge the downstream part of the forcemain as bolted manholes on a gravity line and set your end point on the upstream side's elevation to the HGL elevation. If there's only one controling point then the HGL will be top of pipe elevation, but if your controling point shifts downstream, then you get to manually iterate starting out with a guess of the HGL in order to model your forcemain correctly. (And you also blow your project budget figuring this quirk out unless you have the Rolls Royce support, at least prior to Bentley buying them out.)