Results 1 to 6 of 6

Thread: Modeling Multiple Tanks on level elevation

Hybrid View

Previous Post Previous Post   Next Post Next Post
  1. #1

    Join Date
    Nov 2016
    Posts
    23

    Modeling Multiple Tanks on level elevation

    Hi,
    I have a scenario where there are multiple tanks that discharges to a common pipe and water is transported by gravity and head to a receiving point. I've been trying to determine if the pipe sizes specified is adequate for the network but coming up with negative flow and other issues.

    Any help?

    Thanks

  2. #2
    Forum Moderator

    Innovyze Employee



    Innovyze Employee



    Join Date
    May 2015
    Posts
    420
    David,

    It is difficult to say with the limited information provided and no insight into the system layout, but here are a few initial thoughts:
    1) What software and software version are you currently using?
    2) Water flows from High Head to lower head, so if you are getting unexpected reverse flow it may be due to the receiving head being too high. Is your "receiving point" modeled as a point demand or as a model reservoir?
    3) If you are using reservoir to reservoir flow without demands in the model, are your head values for all locations correct? (supply and receiving point)
    4) If you have hills and a receiving tank you may get negative pressures at certain points because the head needed to drive the flow may not be adequate to keep positive pressure at the high points. You may need to have a pressure sustaining valve before the receiving tank if you need "backpressure" to maintain positive pressure at the high points.

    If you can provide more insight into your model and results we may be able to assist more.

    Patrick Moore


    Quote Originally Posted by D.Assan View Post
    Hi,
    I have a scenario where there are multiple tanks that discharges to a common pipe and water is transported by gravity and head to a receiving point. I've been trying to determine if the pipe sizes specified is adequate for the network but coming up with negative flow and other issues.

    Any help?

    Thanks

  3. #3

    Join Date
    Nov 2016
    Posts
    23
    The version installed is 12.4 Update #2 and the tanks are all level but the levels vary. One tank discharges into four tanks, and those four tanks discharges to a common pipe, which in turn picks up flows from four other tanks before ending at a point to be pumped out. I have attached a sketch to illustrate this.

    TankModel.PNG

  4. #4
    Forum Moderator

    Innovyze Employee



    Innovyze Employee



    Join Date
    May 2015
    Posts
    420
    David,

    Thanks for the further information.

    1) Latest update is 12.4 update 5. We highly recommend updating as soon as you can. There were a few known issues that have been resolved since 12.4 update 2.
    2) Having that many tanks that close to one another can definitely create unusual behavior within the model with large unexpected flows between the tanks. The reason is tiny changes in tank level between the tanks will mathematically cause large flows to occur to induce headloss to account for the head differences. When tanks are very close to one another in distance and strongly hydraulically connected (usually with large connected pipes) that to get the headloss caused by the head difference this generally causes large flows to get induced. If using an EPS run, in the next analysis timestep, the flows generally are reversed as the tank receiving water now has higher head that the other tank. This causes water to flow back and forth between the tanks unexpectedly. This obviously would be even more complicated when you have 8 tanks fairly near one another.

    Out of curiosity, what exactly are you trying to model in this situation? This is definitely a somewhat "unusual" model. Are you trying to check pipeline capacities or something else here? Perhaps there is a way to better model this to avoid the confusion caused by the other tanks? Liek if 100% of the supply went to your Exit and none to the tanks perhaps this could best size this line w/o having the instability caused by the tanks? KNowing what you are primarily after would help with providing the best suggestions of how you might best resolve this.

    I suspect the multitude of tanks is likely the cause for the "unusual results" you had noted.

    Please let us know if you have more information regarding the above and reply back if you need further information.

    Patrick Moore

  5. #5

    Join Date
    Nov 2016
    Posts
    23
    Thanks Patrick, the tanks are all installed on a flat grade and are connected with 8-14" pipes as part of a treatment process and sends the treated water to the "Exit" for pumping. There is little gradient here and the only positive is that the entry pipes into the first four cluster of tanks is 3-ft above the exit pipes which connects with the later four tanks.

  6. #6
    Forum Moderator

    Innovyze Employee



    Innovyze Employee



    Join Date
    May 2015
    Posts
    420
    David,

    This definitely is a challenging thing to model hydraulically due to the potential mathematical issues due to multiple tanks closer to one another as described above.

    If you can live with the potential oddities, you could certainly model this, but this due to the match is a very difficult thing to model because of this.

    What exactly is the "question" you are trying to answer with this model?

    As noted, you may be able to simplify things under certain circumstances and still adequately size elements, but this layout as shown has a great potential for back and forth flows between tanks to occur just by the layout that could lead to unclear results.

    Don't forget you always can look at things form the perspective of a) What is all the water supplied (at maximum) went to the "Exit".. Is my pipe large enough to do that?

    You could also look at what flows might be expected to go through each tank to make sure you have sufficient pipe capacity for your issue.

    As noted, you may run into unexpected flows if you model it as shown simply due to small head differences in each of the tanks, but just taking a step back and thinking how you might still be able to answer this form an engineering perspective this should potentially help you.

    As we discussed off line, please give me a call if it would assist in talking further.

    Patrick Moore



    Quote Originally Posted by D.Assan View Post
    Thanks Patrick, the tanks are all installed on a flat grade and are connected with 8-14" pipes as part of a treatment process and sends the treated water to the "Exit" for pumping. There is little gradient here and the only positive is that the entry pipes into the first four cluster of tanks is 3-ft above the exit pipes which connects with the later four tanks.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •