Results 1 to 3 of 3

Thread: Pumps operating off curve

  1. #1

    Join Date
    May 2014

    Pumps operating off curve

    Hello, I'm still learning InfoWater, so my apologies if I'm not understanding something simple.

    I have a model from another engineer that has several pumps. The pumps are variable speed but are not operated as such, they run at a set speed and turn on and off based on a tank level. They are modeled with a multipoint curve from the manufacturer of head vs. flow, but without a datapoint for zero head (see below minus the last row). When the model is run, the pump curve shows the operating point moving past the specified flowrate onto an extrapolated part of the curve. Since my client has historical data that shows a flow rate much lower than what the model is doing, I need a quick fix to keep the pump from operating beyond a certain flow rate. To do this, I tried adding a data point with zero head (see image).

    The model then sticks to this flowrate for some time but then exceeds the flow rate again while showing -25,000 psi.
    pump graph.jpg

    Is this method workable? Is there a better way to limit the pump flow rate? I have limited information beyond what the client believes the actual flowrate is and the manufacturer curves.

    Thanks for any discussion or input!

  2. #2
    Forum Moderator

    Innovyze Employee

    Innovyze Employee

    Join Date
    May 2015

    If you have a pump flowing off the curve that indicates that the system curve is intersecting the pump curve to the right of the last operating point on the pump curve. This indicates that the head difference between the suction and discharge side of the pump is lower that expected by the pump. When this occurs it is a good idea to check the following:

    1) Is the pump curve correct? (Always a good idea to do a quick check as the error could be there). Also does the pump curve accurately represent the high end range shown on the curve?
    2) Is something boosting the suction HGL higher than normal,or is there an error in the element settings that are controlling the suction HGL (I.e a tank elevation or Reservoir HGL). This could also be caused by an open interface between pressure zones where the suction zone is the zone receiving water and that is boosting the HGL in the suction zone.
    3) Is something artificially depressing the discharge HGL so that it is lower than expected. This can be caused by an error in the elements controlling the HGL (like tank elevation and level) or potentially by an open connection between pressure zones which could cause large flows and depress the discharge HGL.

    You may also want to review the presentation indicated in this forum post on system curves to better understand why a pump operates where it does:

    We would generally NOT recommend the procedure you are using as you need to find out why the pump is operating where it is and fix that as you likely have a few settings just barely off. By adding a zero head point as shown the model will use the last two points to extrapolate the curve when the flows exceed the last point. Using the points you have it should have a huge negative head gain which would result in meaningless results.

    If your curve is good then check the suction and discharge HGLs seen in the field (HGL(ft) = Elevation(ft) plus pressure in psi * 2.31 ft/psi and compare them to what the model is seeing. Make sure to use junction elevations that match the field elevation of the gauge.

    You likely just have things slightly off on the suction or discharge HGL's and this causes the pumps to operate past the last point of the curve.

    Manufacturers generally don't recommend operation past the last point they show on their curve as this is where the pump can experience damage due to cavitation.

    You may also want to ask the client to verify the quality of the flow measurement device and check its calibration as well. It is not unheard of to find field flow devices not accurately representing the true flows. If the flow device is well maintained and calibrated you are probably fine, but if the device is poorly maintained or poorly calibrated its measurement values could be inaccurate.

    Hope this helps.

    Patrick Moore

  3. #3

    Join Date
    May 2014
    This is a lot of great info for me to look into. Thank you Patrick!

Tags for this Thread

Posting Permissions

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