Water Supply Bulk Metering: Pumping Station Meter Readings

Published on by in Technology

I have been monitoring the performance of pumped water supplies in Uganda and have noticed that bulk meter readings at the pump station are reading less than reservoir supply bulk meters. Any idea on the likely cause? Calculation of performances KPIs like NRW, billing efficiency cannot be made with this anomaly.

Taxonomy

13 Answers

  2. As already noted there are many factors that may be influencing the installed accuracy of the readings you are getting from the two meters. You do not state how big the discrepancy is - if they are both good quality electromagnetic meters with ideal installation conditions, then a difference of say 2% or less would not be a cause for concern. 10% or more could indicate a serious problem. Don't forget the telemetry system - incorrect scaling, sampling rates on plc's, etc etc can also give rise to differences.

  3. DImensions of lead & trail lengths of pipes.correct installations & calibration of measuring instrunents  need to be checked.

  4. Hi Julius,

    First question would be how big is the difference and is both meter has same pattern from time to time (same and consistent ups and downs between those two)? Based on my experience, the best practice of max difference allowed would be not more than +/- 10% (long explanation for this meter accuracy allowable error).

    Then the following questions might come up afterwards:

    1. Is the meter sizing correct? You need meter with correct flow measurement capability e.g. flow with 400-500 m3/h would need around 200-250mm meters.

    2. Are the both meters share same class (old versio is class B, C, etc. and newer ISO is R100. R160, R200, etc.) as well as class 1 or 2 in “error band” since you need same sensitivity to ensure you have same measured value.

    3. Have you installed the meter correctly? Like the 2D or 5D straight pipe installation of the inlet section? Most mechanical meters have short lifetime when installed near pump.

    Also the degree, spaces between bends, reducers, position of meters to ensure it is full pipe, etc.

    4. Have they been calibrated correctly at least within the last 2-5 years?

    5. Is it possible to have transmission leak?

    6. Is it possible to have backflow? Since most meters are designed to count forward not backward (it resulted uncertain errors when counting backwards).

    7. Is it possible to have air measured due to wrong or incomplete installation?

    I can help regarding metering theories and methods, so do not hesitate to contact me at muhammadalfalahfauzi@yahoo.com

     

    Regards,

    Muhammad Alfalah Fauzi

  6. 1) calibration issues. Design flow range vs avtyal beee ti check..

    2) air passing alingbwaloth water to shiw High  readings.

    3) not sure if both are on same line. Else reservoir level need to be taken into consideration .Difference in flow meter readings resulting in reservour level variations.

    4) type of flow meter. Turbine/magnetic/ vortex and design of piping and reducer fir laminate flow. 

     

  7. While I concur with the opinion of Mr Andre Vandeloo, the actual readings of any flow meter will depend on the location of the flow meter so as to maintain a straight reach upstream and downstream of the meter, accuracy (tolerance) as well as the calibration status of the flow meter. The actual reading also depends on the actual range of flow compared to the calibrated range of flow. The nearer (more or less in the middle portion) the actual range of flow is to the calibrated range of flow, the more will be the accuracy. You may have to consider the above issues and make necessary corrections to the readings of the flow meters to enable any comparison. Another cause for the variation may be NRW in between the pump station and the inlet to the reservoir. Hope I am clear my friend.

  8. I will assume you to mean that you have two flow sensors on the same line, one after the pump and one on discharge to the reservoir, magnetic type. Will mention the basics.

    Apart from the obvious regarding calibration (which depending on the quality of build, can be in-situ or has to be removed and calibrated), flow meters are most accurate if installed in as straight a pipe run as possible (15 x ID in length before and >5 x DI after) to dampen a turbulent flow with nil air and at a respectable velocity (>1 m/s minimum, 3 to 5 m/s preferred). Which of the two locations better fits the above is the one i would trust more.

    Pump stations are typically housed and therefore tend to be compact in installation, including the attainable straight run of pipe. The pump station piping tends also be larger to account for downstream pipe friction which then affects the measured velocity, and therefore the accuracy.

    In this situation the flow sensor is better installed in a reduced ID section of pipe which helps to stabilize the flow, shortens the foot print, reduces instrument cost whilst increasing reading accuracy.

  10. Are the water meters at the reservoir inlet or outlet? If at the inlet, then the cause maybe due to inaccuracies caused by lack of calibration. If at the outlet of the reservoir then check for any loss through overflows or reservoir leakages.  At the same time check for Rising mains to identify any leakages, check other loses eg at AV points. Check if you do frequent flushing of the line. 

    The other option you can install another meter in series with the reservoir one.

  11. large meters can have significant inaccuracies if not calibrated. if having the meters calibrated is not in the immediate future, maybe do a rough calibration by measuring the volume change in the reservoir over a day and compare to pump volumes. 

  13. Dear Julius,

    I am modelling Latvia capital Riga water supply system. We have several pumping stations with reservoirs before them. In our case it is common, that in same days it could be more inflow in reservoir than outflow from it. In our case total inflow in reservoir is same as outflow from it in cycle consisting from several days and not in one  day cycle. May this is also in your case?

    Br,

    Normunds

    1 Comment

    1. Along with the other comments above. I know it might sound to simple or obvious, but the location of your meter in-line from any offset, bends (elbows) especially 90 deg, or to close to the exit points from reservoir or pumps can effect accuracy due to turbulent flow conditions. 10++ x the pipe diameter clear needs to be considered/sometimes overlooked due to site constraints and security.