Table of Contents

Single-Sensor Measurement of Flow
in Filled or Partially Filled Process Pipes
A new sensor designed for municipal and industrial wastewater treatment facilities combines electromagnetic flow rate measurement with capacitive level detection in a single unit that can operate on pipes filled to only 10% of capacity.

John Flood, Krohne America, Inc.

Photo 1. The Tidalflux looks like and functions as a conventional magmeter, but can also accurately measure flow in partially filled pipes.
Current concerns about the environment have called for new solutions, particularly in wastewater applications where there is a demand for accurate measurement of the flow rate in piping systems that may at times be only partially full. To avoid inaccuracies caused by partially filled pipes, conventional magnetic flowmeters (magmeters) must be installed in sluice underpasses. This approach, however, can cause pressure loss within the fluid and a chance of clogging that cannot always be tolerated. Furthermore, installing a sluice underpass creates additional costs that can be higher than the price of the flowmeter itself, especially when large pipe sizes are involved.

The Tidalflux
   The Tidalflux (see Photo 1), a maintenance-free magmeter combined with an independent capacitive level measuring system, is capable of measuring flow to an accuracy of 1% in both filled and partially filled pipes.

   Flow Velocity Measurement. Flow velocity is calculated based on Faraday's law of induction, which states that if a conductive liquid flows through a magnetic field a voltage will be induced. This voltage is directly proportional to the average flow velocity of the medium. With this, and the tube diameter of the primary head, the signal converter calculates the volumetric flow rate:

Figure 1. The electromagnetic flowmeter measurement principle is based on Faraday's law that states that a conductor passed through a magnetic field will generate a voltage.

U  = K1 · B · V · D (1)


   U  = voltage

   K1  = instrument constant

   B  = magnetic field strength

   V  = mean flow velocity

   D  = electrode spacing/tube diameter

   In the Tidalflux, two coils generate a magnetic field that is perpendicular to the nonmagnetic measuring tube (see Figure 1). The electrodes are not placed opposite each other in the middle of the pipe as is the case in conventional electromagnetic flowmeters, but rather at a point near the bottom representing ~10% of the tube diameter. The flow velocity can therefore still be determined with certainty in a measuring tube filled to only 10% of its diameter:

   Q(t)  = V · A (2)



   Q(t)  = actual flow

   V  = flow velocity

   A  = wetted cross section of the pipe, computed by means of capacitive level measurement

Figure 2. Because the level measurement is done in the background, the output signals of the Tidalflux are the same as those of a conventional electromagnetic flowmeter.

   Capacitive Level Measurement. The integrated capacitive level measurement system consists of a detection plate on one side of the measuring tube and four transmission plates on the other side, all embedded in the Irathane (polyurethane) flowtube liner (see Figure 2). After the liner is cast in place, the level sensor is completely covered and fully insulated from the liquid to be measured. A high-frequency voltage induced into these transmission plates influences the current in the receiver plate, which is then measured. The bottom transmission plate is always completely covered by the liquid of interest and is used as a reference by which the offsets caused by varying fluid conductivity can be determined and compensated for. With these parameters, the filling level in the tube can accurately be determined:

   C  = K2 · A (3)

   C  = capacitance

Figure 3. Incorporation of the level measurement plates within the flowtube eliminates the need for open-channel flow measurement, a feature than can help in odor control.
   K2  = constant

   A  = wetted cross section.

   The compact electronics, which are mounted on the measuring tube, calculate the filling factor, defined as the relation between the filled area of the measuring tube at partial filling and at complete filling. This filling factor b is between 0 and 1 and is a function of the flow. Combining the actual flow velocity at partial filling with the filling factor b yields:

   qpart  = Vpart · K3 ·b (4)


     Vpart  = velocity of partially filled flowmeter

   K3  = flowmeter constant

   The level data are linked via serial digital communication to a remote flow converter that provides the necessary flow outputs. In full pipe conditions, the Tidalflux functions as a conventional electromagnetic flowmeter.

   The Tidalflux is calibrated in two phases. First, the calibration constant is determined with a completely filled measuring tube in a conventional calibration facility.

   Calibrating the level measurement feature addresses both static (no-flow) and dynamic calibration. For the former, the correction factors for the level meter are determined by a number of filling levels and fluids compared with a reference level measuring system. By using a 3rd degree polynomial, the so-called a-function is determined. This function is stored in the EPROM of the level electronics and used by the microprocessor to calculate an output signal, which is in proportion to the filling fraction and independent of the properties of the liquid.

   For dynamic calibration, a specially designed calibration facility has been designed for partially filled flowmeters. The partially filled flowmeter is calibrated against a reference magmeter, a process that allows corrections to be found at a number of filling levels and flow velocities. This is the b function, and is also stored in the EPROM of the level measurement electronics.
Figure 4. Placing the electrodes near the bottom of the pipe allows flow measurement in partially full pipe conditions. The electrodes are high enough to prevent measurement problems due to sediment buildup in the flowtube.

   The Tidalflux has an inaccuracy of max. 1% of F.S. for partially filled pipes and max. 1% of measured value (V $ 1 m/s) or max. 0.5% of the measured value + 5 mm/s (V < 1 m/s) for completely filled pipes.

   The magmeter has no moving parts to wear out and no obstructions in the flow to collect dirt or debris, and requires no sluice underpass. Its single set of electrodes minimizes potential downtime. The Irathane flowtube liner has been found to be resistant to abrasion and corrosion in this application, based on thousands of successful installations of conventional magmeters.

   The use of a continuous pipe brings its own set of benefits: because the pipe is closed, rather than opening out into a flume, hazardous materials in the flow stream do not endanger personnel working nearby. There is only a minimal chance of plugging from surface contamination, and no restriction in the flow stream. The continuous pipe diameter eliminates the risk of flooding that can occur when a fast-moving stream coming from a large inlet is forced into a narrower inlet.

   In partially filled measuring tubes there is usually a formation of waves on the liquid surface. These ripples, especially when they are on the level of the electrodes, could cause a fluctuation in the sensor's output signal (see Figure 4). With the help of fuzzy logic and special filtering circuits included in the level measurement system, this problem is eliminated.

John Flood is Magmeter Product Specialist, Krohne America, Inc., 7 Dearborn Rd., Peabody, MA 01960; 508-535-6060, fax 508-535-1720.

Questex Media
Home | Contact Us | Advertise
© 2009 Questex Media Group, Inc.. All rights reserved.
Reproduction in whole or in part is prohibited.
Please send any technical comments or questions to our webmaster.