Back to blog

Magnetic Flowmeters: Commissioning and Diagnostics

A field-focused checklist for selecting, installing, commissioning, and diagnosing magnetic flowmeters, with emphasis on conductivity, full-pipe conditions, grounding, wiring, zero checks, and PLC ...

Magnetic flowmeters are attractive because the measuring tube has no rotating element in the flow path. That does not make them installation-proof. Most commissioning problems come from application fit, pipe conditions, electrical reference, or signal handling—not from the transmitter’s arithmetic.

This checklist separates the checks that should happen before purchase, during installation, and at loop acceptance. It applies to full-bore and insertion designs, but the manufacturer’s model-specific limits always take priority.

1. Confirm the liquid can be measured

A magnetic flowmeter induces a magnetic field across the pipe and senses a voltage produced as a conductive liquid moves through that field. The method therefore requires an electrically conductive liquid. It is not the right starting point for gases, steam, hydrocarbons with very low conductivity, or a line that regularly runs empty.

Do not copy a conductivity threshold from another model. For example, Bürkert specifies conductivity greater than 20 µS/cm for its Type 8045 insertion meter; another meter may have a different limit. Check the selected meter’s current data sheet and the lowest expected process conductivity, including startup, cleaning, and batch-transition conditions.

2. Verify materials and process limits

Confirm the liner, electrodes, seals, grounding components, pressure rating, temperature rating, and enclosure are compatible with the actual fluid and cleaning regime. A chemically compatible electrode does not automatically make the liner or gasket suitable. Abrasive slurries, coating fluids, vacuum conditions, and aggressive cleaning chemicals deserve explicit review.

Record the nominal pipe size and expected minimum, normal, and maximum flow. The meter should operate inside the manufacturer’s specified velocity and turndown ranges without creating an unacceptable pressure or maintenance problem.

3. Choose a location that stays full

The electrodes must remain wetted. Avoid the top of a horizontal pipe if gas can collect, and avoid a location where a downward run can drain. A vertical upward flow is often useful when entrained gas or partial filling is a concern. If the meter is near a pump, valve, elbow, reducer, or other disturbance, apply the straight-run guidance for that exact meter rather than a generic rule.

Also provide access for grounding connections, transmitter viewing, cable entry, and future removal. Mechanical stress from misaligned flanges can damage liners and seals, so support the pipe independently and follow the specified bolt sequence and torque.

4. Establish a reliable electrical reference

The sensed electrode voltage is small. Bonding and grounding arrangements help the transmitter distinguish the flow signal from electrical noise. Follow the manufacturer’s diagrams for conductive pipe, lined pipe, plastic pipe, and cathodically protected systems. Ground rings or electrodes may be required in some installations.

Keep electrode and coil cables within approved lengths and routes. Separate low-level measurement wiring from motor leads, contactor wiring, and variable-frequency-drive output cables. Terminate shields exactly as instructed; indiscriminate bonding at both ends can create an unwanted current path.

5. Configure the transmitter deliberately

Enter the correct sensor size, calibration data, engineering units, flow direction, damping, low-flow cutoff, empty-pipe behavior, and output range. For a 4–20 mA signal, document the relationship between current and flow and decide how underrange, overrange, and faults will be represented. For pulse or frequency outputs, document pulse weight, maximum output frequency, and counter rollover behavior.

If the meter communicates over an industrial network, treat its quality and diagnostic bits as part of the control design. The PLC should not use a numeric flow value as valid when the device reports an empty pipe, sensor fault, or invalid measurement.

6. Perform a controlled zero check

A zero check requires a completely full, stationary liquid column. Closing a valve does not prove zero flow if leakage, thermal circulation, trapped gas, or pipe vibration remains. Stabilize the process, confirm the pipe is full, and use the manufacturer’s zero procedure. Do not use zero adjustment to hide installation noise or a partially filled tube.

7. Test the complete signal path

Commission from the sensor to the PLC and operator display:

  • Verify displayed flow direction against a known process movement.
  • Compare local indication with the PLC value at zero and at one or more controlled operating points.
  • Simulate or observe empty-pipe and device-fault conditions.
  • Confirm analog scaling, byte order, pulse accumulation, totalizer units, and alarm delays.
  • Check that loss of power or communications produces the intended control response.

Store the final parameters, firmware version, calibration identifiers, wiring drawings, and baseline diagnostics with the loop record. These references make later troubleshooting far faster than changing damping or scaling by trial and error.

Fast diagnostic sequence

If the reading is unstable, start with process reality: full pipe, entrained gas, conductivity, coating, and flow profile. Then check grounding, cable routing, shield termination, and supply quality. If the local value is stable but the PLC value is wrong, focus on output configuration and scaling. If both values are biased, review zero conditions, installation geometry, and calibration data before replacing the meter.

For related hardware context, see PLC ProTech’s communication and networking components and the broader automation Knowledge library. A representative device-specific conductivity requirement is available on Bürkert’s official Type 8045 page.

Leave a comment

Please note, comments need to be approved before they are published.