This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

WaterCAD Multiple Pumps

Hello,

I have a network of 110 grinder pumps that connect to a force main. The grinder pumps are a combination of a 32.5 and 70 gallon storage tanks and the grinder pumps only turn on when they have reached tank capacity. I also have a master pump stations that serves as a re-pump to the force main. I am trying to determine how many grinder pumps can run at the same at peak hour (maybe from 7am to 8am for example). How do i model this in watercad?

I am using 2.5 persons or 250 GPD usage per grinder pump. I understand i will have to assume how much of the 250 gallons is used at peak hour (7am to 8am example).

Thank you for your help!!

Senhad

  • how many grinder pumps can run at the same at peak hour

    The short answer is "All of them could be running".

    They may not actually be delivering much individually, but if the wetwells are above the "turn on" level then the controls will tell the pumps to energize and won't stop until the wetwell is pumped down.  As more and more wetwells reach the "turn on" level, their corresponding pumps will energize and the output of every other operating pump in the system will be reduced by some amount.  They will then run at some reduced capacity, with the ones closer to the discharge end of the force main pumping a little more than the ones at the far end of the system.

    If you set up your pumps, wetwells and controls and run it as an EPS, WaterCad will tell you how your system will react.

    What I would be interested in is what happens to wet well levels as the pumps cycle on and off.  As I stated above, if a lot of pumps are running then none of them will be outputting much flow, and consequently your wetwell levels may continue to rise.  What I would want to find out is if any wetwell levels rise to above the overflow elevation.  If they do then you have a problem.

    It sounds like this should be a fairly straightforward system to model, although it may take some time to set up depending upon the topography of your system, how many different pump curves you have, and how many different control settings you have.  In my experience getting the controls set up properly will be the most time consuming (and most frustrating) part of the job .

    Good luck.

  • Adam, thank you for the reply. 

    I just want to add that I have modeled all the grinder pumps but i am new to this so i want to assure myself that i did it the right way. My grinder pump modeling is as follow; resevoir-pump-juntion-forcemain. Yo your question regarding what would happen to the grinder pump wetwell, the grinder pumps have a control switch that turns off the pumps if they are forced to pump against high head but also only turn on when the tank is full. Ultimately the pumps that are the furthest away from the discharge end will be prone to high head and will ultimately shut down and flood from my assumption. I do not know what the shutoff parameters (high head) are for the grinder pumps.

    My model is all set up. I technically have 2 zones i am working with. Zone 1 has a 110 grinder pumps connected to a forcemain ranging from 1.5"-4" diameter and one master pump station that is a re-pump station to zone 1. Zone 2 goes from repump station for zone 1 to a second pump station that discharges to an outlet (county line), forcemain in zone 2 range from 4"-6"in diameter (from pump station in zone 1 to pump station in zone 2) and then 8" outlet to county line from pump station in zone 2 . Zone 2 has an additional 29 grinder pumps. My goal is to determine the efficiency of this system for full development (110 + 29 grinder pumps). 

    I have a total of 4 curves, 2 curves for the 2 pump stations and 2 curves for two different grinder pump stations (32.5 gallon and 70 gallon tank). Although my goal is to analyze the entire system i am more concern with zone 1 due to higher number of grinder pumps connected to a smaller forcemain diameter and also the longest forcemain of the two. 

    I have run a simulation on the model and analyzed the results and what you mentioned above agrees with the results i am getting. As i mentioned i am new to this but i would essentially like to know how many grinder pumps will turn on and run during a peak hour, say morning at 7 am to 8 am. Considering that grinder pumps only turn on when they reach capacity (32.5 gallons or 70 gallons). And they will not all turn on at the same time due to to the fact that they only turn on when the tanks are full. I will also have make some assumptions and say each grinder pump will receive for example 50 gallons from 7am to 8 am. So i guess my question is to find out if i can set up parameters for the grinder pumps to turn on only when tanks are full and if can assign assumption to say between 7am to 8am each grinder pump will receive 50 gallons within this hour?

    Right now when i run my analysis, it is essentially running all the grinder pumps at once and like you said some are pumping and some are not. But this does not align with the assumption that not all of the grinder pumps will turn on at the same time, at least i don't think it is .

    My apologies for not providing more detail and possibly making it more confusing.

    Thank you again.

  • You would like to simulate a reasonable number of combinations of pumps' running. If you have 110 pumps, there are 10^33 combinations so you can't run an exhaustive number of scenarios.

    The eOne company has some tables on their web site suggesting the average and high number of pumps that are likely to run at any time. We have a section on this in our Wastewater Collection System Modeling and Design book. With 110 pump, a likely high number of pump running is 8. You don't need to include 110 pumps in your model, just a sampling of 8. You also want to check what happens when say 2 pumps are running to make sure you have a reasonable velocity (about 2 ft/s)  to prevent setting in the pipes.

    There is one exception to this guidance and that is the case when there has been a long power outage and all the tanks have filled. When the power comes back on, all of the pumps will start at once and, depending on the pumps, the pressure can be very high for a short period. You don't need to size the pipes for this case but you need to check the pressure.

  • 1. If you have the pump curves then you have (or should have) the shutoff condition (the head the pump produces at 0 flow). This information should be input into your WaterCad pump curve information if it is not already.

    2. I don't understand why you would have a separate pressure switch to disable the pump on high pressure.  Without the pressure switch the worst that could happen is that the pump overheats and you have a seal failure.  In my opinion the addition of this pressure switch is just another potential failure point, but that's just me.

    3. I understand what you are trying to do, but unless you have real-world data about the actual inflow rates & times into the individual wetwells, you will have to artificially stagger or alter the rates/times of inflow into the wetwells. Otherwise they will all fill at the same rate and thus start pumping at the same time.

    WaterCad has no idea of either the fill rate of the wetwells or when they start filling.  You must give it this information.  If you tell WaterCad that the wet wells start filling at a certain rate at 8 am every day, all of the 32.5 gallon wetwells will fill up at the same time and all of the pumps will start at the same time. With the high pressure shutoff switches in there (assuming they are all set to the same pressure) it's probably somewhat random as to which pumps would be disabled.  I can easily see how you could get into a very fast "on-off-on" cycling situation with this kind of setup. The same holds true for the 70 gallon wetwells.  Perhaps either Jesse Dringoli or Dr. Walski can address this.

    It sounds to me that you don't really have enough information about exactly how your system will operate to provide you with what you want to know.  I would set up the system without the high-pressure shutoff switches, run an EPS, see what the results are and go from there.  You may want to look into setting your pump "on" level settings slightly differently for groups of stations (i.e. - set 10 of the stations to start at elevation X, then 10 more to start at elevation X+0.5, etc).

    Again, good luck!

  • Mr Walski,

    Thank you, this was helpful.