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

Modeling Flow Rate and Head from a Pumped Artesian Well to Ground Storage Tank

Good Evening! In looking in Advanced Water Distribution Modeling and Management (p 319-322) and the help section "Modeling a Pumped Groundwater Well", I think I have it understood but looking for reassurances. The well installer recently installed a new well pump at X depth. The installer provided the pump information and the elevation of the pump set point. Static water level was reported at 0 since the well is artesian but pump level steadied out at Y elevation. What I don't have is time to get to pump level but quite frankly I don't think it is necessary per se (exception to determine draw down in storage to cycle the well pump enough to keep the well from pouring out, but that is another study). Reason why I say it is not necessary is that at this point in time, I know the flow rate out of the well head at PL which then will correspond to particular point on the curve to give me the max head of that pump at any time (after so many seconds). So it can only go down from there and with only two data points I will have a linear line which will not necessarily be representative of the true curve. What I am trying to model at this point in the software is I have my storage tank and I am going to set it up with a separate inlet and outlet to ensure that in the future disinfection storage is satisfied up front but based on what I need in the tank, I need to ensure mixing in the tank and I am looking to internal nozzles/diffusers from the inlet pipe. I need to determine the flow rate and head being delivered to nozzles to ensure that I specify the right mixers to help keep age in the tank minimal and to prevent dead zones. So really I am looking for opinions on the modeling. Do I just assign a junction instead of modeling a pumped groundwater well? What other information could I learn from modeling the groundwater well with the information I have now without having to obtain additional data points? Thanks in advance!
Parents
  • Hello,

    In general, it would be best to model what is there. That will allow you to work on other aspects of your model as you gather data. If the tank and/or the pump are not important to what you are modeling, you can use an inflow node with a negative flow. But if possible it is better to model what is in the actual system.

    For the nozzle, you might be able to apply a minor loss coefficient to the adjacent pipe. How you model this will be based on the hydraulic impact of the device.

    If you have further questions, please include some clarification of what you are currently modeling and the data that you have.

    Regards,
    Scott
  • I have a few observations and a few questions.

    Is this storage tank before the first customer to provide disinfectant contact time or is it distribution storage after the first customer? if it is a contact tank, you don't want complete mixing; you want plug flow with separate inlet and outlet to maximize the average contact time. You can install baffles to improve this. If it is for distribution storage, you want complete mixing but you don't need two pipes to achieve that.

    The actual elevation of a submersible pump isn't terribly important as long as it is deep enough to prevent air intrusion. But it is good practice to use the actual pump elevation and the correct length and diameter of the riser pipe.

    Is elevation Y the drawdown elevation when the pump has reached steady state. You can use that as the reservoir elevation.

    Be careful to use the same datum for all of your elements. My preference is mean sea level so that the junction at the top of the well would be at that elevation, not 0.

    I'm assuming you are using the actual pump curve for the well pump. The model will find the actual operating point.

    What is PL?

    What do you mean by "max head" and "going down from there"?
  • I will work from the bottom up on answers to your questions:

    Max Head and Going Down from There...HGL out of the pump and the decline to the tank.

    PL = Pump Level

    Yes I/will be using the well pump curve based on where it is set and the current output as determined by the well installer.

    Yes, my other post "Negative Pressures - WaterCAD" caught me regarding the datum so I will double check.

    Good idea on the reservoir and yes, Y is the drawdown elevation at steady state.

    I figured that modeling the hydraulic loss in the drop pipe would be good practice for the overall model; but if I know what my output is at the well head those loses are already taken into account. If my model seems out of sync this is at least a place to put in a "known" to figure out where potential problems are located.

    This is storage/disinfection before the first customer. Basically it is a straight forward setup well -> storage -> boost pump -> bladder tanks -> customer. Where this gets challenging is that we do have an artesian well. The existing setup uses a 2,000 gallon storage tank in an existing building and the operator short cycles his system so that his well pump comes on frequently enough so that water does not flow freely out of the top of the casing. The new tank will be much larger and located in the elements (heat/cold). My original intent, and still is, for separate inlet and outlets because should a problem with the system occur emergency chlorination is onsite. However, with a larger tank I have a possible issue with age even if I "short cycle" the well pump. The idea of mixing came from the idea that from a couple AWWA magazines, OpenFOAM models demonstrated dead spots with tanks. Chlorination may have to be introduced permanently but I want help try to mitigate it. Short cycling of a well pump is also another wrinkle. Franklin motors do not like more than 50 starts a day and there needs to be cool down period for the windings as well. We have discussed that it is probably going to be inevitable that the wellhead will have to have an over flow to a nearby ditch to waste water which I have a problem with. Those are some of my thought processes in this modeling.
  • In the meantime, I have been looking and thinking about your discussion about setting the elevation to MSL or something similar. For this project I am tying the floor elevation as the benchmark at 100'. Add 18" for top of casing to get 101.5' the subtract 220' from the 101.5' to get me my reservoir elevation -118.5 and my pump is 270' which puts my pump elevation at -168.5'. Reason why I am asking is that I am getting errors in my model and I am hoping that is not a reason.
Reply
  • In the meantime, I have been looking and thinking about your discussion about setting the elevation to MSL or something similar. For this project I am tying the floor elevation as the benchmark at 100'. Add 18" for top of casing to get 101.5' the subtract 220' from the 101.5' to get me my reservoir elevation -118.5 and my pump is 270' which puts my pump elevation at -168.5'. Reason why I am asking is that I am getting errors in my model and I am hoping that is not a reason.
Children
  • nkg2,

    You mentioned that you're getting errors in your model. I'm a bit unclear on the numbers that you mention above that you're adding and subtracting - if you can provide a copy of the model, we could at least explain the reason for any error or particular result that you're getting in the model.


    Regards,

    Jesse Dringoli
    Technical Support Manager, OpenFlows
    Bentley Communities Site Administrator
    Bentley Systems, Inc.