Calibrating a Parshall flume

Share Print Related RSS
 PROCESS PUZZLER From February 2004's Chemical Processing Magazine

We must measure the flow of a waste water stream through a Parshall flume. An accurate rate is important to ensure environmental compliance. The flume is the last device in line right before the pumping station. We suspect that the flume is way off because the readings don't make sense and are consistently low (they don't match our water consumption figures). How do we end our calibration nightmare, determine what's going on and verify what the actual flow rate is?



Decrease water velocity
If the water velocity is too fast, the relationship between depth and velocity will be way off A small increase in depth will give you a disproportionately large change in discharge rate. You need to find some way to dissipate the velocity energy, and that will result in an increase in depth of flow. You can see if there is room in the pipe for artificial barriers, which will slow down the flow, or you can rebuild the inlet box on the flow meter.

If you are going to rebuild the flow meter entrance box, then read either the ISCO Open Channel Flow Measurement Handbook or, on the Web, chapter eight of the U.S. Bureau of Reclamation's "Water Measurement Manual."

When rebuilding the entrance box, bring the flow in at a right angle to the flow meter (even from the bottom is fine), and give it a hard surface on which to dissipate the energy of the stream. Then your flow meter will work just fine.
David Russell, P.E., president
Global Environmental Operations Inc., Lilburn, Ga.

Ultrasonic level transmitter
Install an ultrasonic level transmitter to measure the height of water in the throat of the flume. Flume calibration charts are often built right into the transmitter. The transmitters have no moving parts and nothing comes in contact with the measured media. We have been using an ultrasonic meter in this service for about eight years and have had good results.
Chuck Rogge, process development specialist
BP Chemicals, Lima, Ohio

Determine flow type
Two conditions of flow can occur through a Parshall flume: free flow and submerged flow. In general, submerged flow should be avoided whenever possible since it greatly complicates the calculation of the discharge rate. During free flow conditions, the discharge rate can be determined by a single depth measurement upstream from the point of critical depth (i.e., the throat of the flume). Submerged flow requires the measurement of both an upstream and downstream depth to obtain a correction factor. You have two options: keep this flume and measure both the upstream and downstream depths as required for submerged flow conditions to obtain the corrected discharge rate, or install a smaller flume that will constrict the flow through the flume such that free flow conditions are created.
Donald A. Smith, P.G., principal hydrogeologist
Converse Consultants, Parsippany, N.J.

Measure velocity with a pitot tube
Use a pitot tube to determine the liquid velocity (based on the differential pressure created by the flow), then apply this to the cross-sectional area of the flume to determine the flow rate.
George Basile, environmental specialist
Consolidated Edison Co. of N.Y. Inc., New York

Determine flow calibration
For Parshall flumes, the relationship of the flow rate to measurable liquid height is linear -- meaning, if you plot the manufacturer's data for liquid height versus flow rate you should get a straight line. If you have a flow rate and a corresponding height that does not match what you expect, use these two data points along with two other known points to generate your own linearly extrapolated calibration data.

If you have to calibrate the flow in this manner, then the Parshall flume you have is too small. Any measurement an eighth of an inch off could yield hundreds of gallons per minute more or less than actual. The larger the flume, or, perhaps, better designed for your application, the better the accuracy in measurement.
Ron Johnson, process engineer
Alpharma Pharmaceutical, Baltimore

Use LiNO3 injection
The measurement of large flows is a problem we deal with all the time. We have developed a method that works quite well:

  • Prepare a concentrated solution of LiNO
  • Assemble a reliable metering pump with a good flow meter.
  • Pump a constant flow of the concentrated LiNO
  • Allow the stream to reach steady state, and obtain 10 samples of the water flow downstream of the injection point.
  • Analyze the water solution for concentration of Li (call this value C1). Also, analyze the concentrated LiNO3 solution for concentration of Li (call this value C2).
    Compute your flow by:
    Unknown flow (gpm) = F1 x C2 / C1

    LiNO3 works quite well for this application as it has a low background value (usually zero) in typical waters, it can be analyzed accurately with an AA, and most important, it is not adsorbed (removed from the water phase) by any contaminating solids that may be present in the water stream. We have used this method on flows as large as 50,000 gpm.
    Glen Oswald, process engineering superintendent
    IMC Phosphates, Mulberry, Fla.

    Make sure depth sensor is clean
    The flow meter may be reading low due to a problem with the sensor. Typically, a bubbler or ultrasonic sensor is used to determine the height of water at the entrance to the flume. Bubblers can get dirty while other sensors can give an inaccurate reading in the presence of foam.

    Incidentally, in relating sewage effluent to water consumption, it's important to remember to allow for evaporation by any cooling towers within the plant. High evaporation rates can give suspiciously low discharge numbers.
    Mark Smith, plant manager
    Nippon Carbide Industries Inc., Greenville, S.C.

  • 3solution into the water flow you wish to measure.
  • 3(you will have to use hot deionized water to dissolve it). This compound has about a 0.5 g/mL maximum solubility, so you have to make a solution of slightly lower concentration than that and keep it warm to prevent crystallization.
Share Print Reprints Permissions

What are your comments?

Join the discussion today. Login Here.

Comments

No one has commented on this page yet.

RSS feed for comments on this page | RSS feed for all comments