How to select a flow meter

Flow meters are frequently used and installed in nuclear plants. But there are numerous factors to consider when selecting the correct one for the job. This guide does a pretty good of summarizing them…

Omega Engineering manufacturers measurement and control products. This includes a line of flow meters. They also produce a technical reference series, and Volume 4 is focused on flow and level measurement. It offers an introduction to different kinds of flow meters, as well a brief history of the field. The document also provides a comprehensive set of considerations that can be helpful for selecting the best kind of instrument for the job.

Their technical guide provides a good overall introduction to the topic. So, I have tried to summarize its content in this article.


The first criteria listed are not engineering in nature, but personal: the preferences of plant operations and maintenance personnel; and operating experience. This includes a consideration of whether spare or replacement parts are available, and for how long, as well as the failure rate of particular kinds of meters at the site.

Next should come a consideration of cost. Omega’s guide stresses the risk of elevating cost in priority over personnel preference and site OE. You can be tempted to buy a cheap meter that really doesn’t perform as desired, and then have to (reluctantly) justify its acceptability later.

The nuclear power industry’s design process mitigates the risk of falling into this particular pitfall. Rigorous vetting and evaluation are performed up front before materials are procured. This includes OE searches not just within the site’s database, but within the INPO industry OE database, too. Particular brands or models that experience higher failure rates than others can usually be identified by an OE search.

Furthermore, the site stakeholders are involved from the beginning of the design process. This group of stakeholders will usually include representatives from maintenance and operations. They will have their personal opinions on what works and what doesn’t, or on which model meter is easier to calibrate and provides less of a maintenance headache over time than others.


Moving on, the selection guide comes to the actual application requirements. Part of this is gaining a full understanding of the “process fluid,” as well as how the system functions.

There are several requirements that should be decided first:

  • Does the flow rate data need to be reported continuously or totalized (summed over time)?
  • Does it need to be displayed locally, remotely, both, or neither?
  • If remote, will the data transmission be analog, digital, or a combination of both?
  • What is the minimum data-update frequency needed?

After answering these questions, the characteristics of the process fluid and the system piping should be evaluated. The guide provides a detailed matrix to assist this process, which, frankly, can be overwhelming if it’s your first time looking at it. But generally, the analysis should consider:

  • Fluid and flow characteristics, like operating pressure, temperature, allowable pressure drop, density, conductivity, viscosity, and vapor pressure at operating temperature. This includes an understanding of whether there will be bubbles in the fluid, whether there will be solids present, and how transparent the fluid is (for letting light through).
  • Minimum and maximum temperature and pressure values, along with those of normal operating conditions. Other items to consider are possible flow directions (is it reversible?), can slug flow develop, and what special precautions may be needed during cleaning and maintenance.
  • Piping direction, size, piping material, schedule (wall thickness), and where valves, regulators, straight-runs, and bends are located.
  • Special area requirements (i.e. are there explosive hazards or, particular to this industry, radiation exposure over time).

These items are especially worth considering for nuclear applications. In a plant, for example, you will have both carbon and stainless steel piping.

If the plant is located on the coast, the meter should probably be available in a stainless steel option to resist corrosion due to salt in the air. An ABB article on flowmeter selection elaborates on the importance of knowing where bends and valves are:

Obstructions in the pipeline such as joints, bends or valves in close proximity to the meter can all cause distortions in flow, affecting flowmeter accuracy and repeatability. To ensure best results, flowmeters should be installed in locations where there are several straight-lengths of unobstructed pipeline both upstream and downstream of the meter.

It is therefore important to find out the manufacturer’s installation recommendations before buying a flowmeter, particularly where installation space is limited.

In a separate article published by Titan Enterprises, Trevor Foster (founder) recommends also evaluating a fluid’s chemical properties. The fluid’s chemistry should be compatible with the flowmeter’s construction materials. “Remember to consider all of the flowmeter materials, not simply the material of the body. Typically a flowmeter consists of an O-ring, a rotor, turbine or Pelton wheel, embedded ceramic magnets, gears, bearings, etc. All of these could affect, or be affected by, the fluids they come into contact with. Check each material separately against a reputable chemical compatibility table and double check your selection with the manufacturer of the fluid you wish to measure to ensure long term durability.”


You must also determine the range of rates that an acceptable meter needs to measure. This, too, is critical for meter selection in nuclear applications. This is done be examining:

  • Flow rate (maximum, minimum, and normal conditions)
  • Whether you are more concerned with mass flow or volumetric flow. Volumetric flow can be used to calculate mass flow, but only if actual process conditions (pressure and temperature) are known.
  • The requirement measurement accuracy, which is typically specified in percentage of “actual reading,” (AR), percentage of calibrated span (CS), or in percentage of full scale (FS) units.
  • What the accuracy needs to be across minimum, maximum, and normal flow rates so that the meter’s performance will be acceptable over the full range of flow rates.

Another important characteristic to consider is turndown. As explained in the ABB article on flowmeter selection, “turndown is the ratio of the maximum and minimum flow rates a flowmeter can measure within its specified accuracy range.” For example, assume you know you need to measure a flow between 200 and 1900 SCFM. You would need a flowmeter with a minimum turndown of 9.5. ABB recommends going with a meter that gives you the widest turndown possible since there is some uncertainty in knowing exactly what your range of flow rates will be.

The Omega selection guide then goes into a discussion about accuracy versus repeatability. Different applications will value one over the other, so they recommend establishing separate accuracy and repeatability requirements and state them both.

Accuracy can be stated in multiple ways: percent calibrated span; percent full scale units; or percent actual reading. The guide recommends normalizing the accuracy requirements to percent AR so that different meters can be compared fairly.

ABB issues a caution to maintain a questioning attitude when it comes to manufacturers’ calibration accuracy claims. “Even under stable reference conditions, the best accuracy that manufacturers can hope to achieve is 0.1%.”

Another item to consider is whether you will configure a pressure transmitter to report flow by enabling square root extraction. Manufacturers may recommend disabling square root extraction within the meter if the turndown exceeds a certain ratio. For example, the Rosemount 3051 differential pressure transmitter’s manual recommends disabling square root extraction if the turndown is greater than 10:1. In that case, a separate software routine will have to perform the square root extraction.

The next step listed is to determine the maximum and minimum Reynolds numbers for your application. The Reynolds number is a value that predicts fluid behavior in different situations. Fluids at low Reynolds numbers tend to exhibit laminar flow. At high Reynolds numbers, the flows become more turbulent. The flow meter needs to be able to perform adequately over the range of flow conditions described by the Reynolds numbers.


Wrapping up, the Omega guide offers some parting wisdom.

If you get this far, and two different flow meters perform acceptably, then the greater weight should go towards the meter without any moving parts. Moving parts mean fewer problems over time.

Similarly, if it comes down to equal performance between a full flow meter and a point sensor, go with the full flow meter. It will tend to be more accurate over time.

Next, break the tie by going with the meter that has less pressure loss.

Finally, if you are more interested in measuring mass flow (particularly if the material is compressible), then go with a mass flow meter. Meters that only measure velocity can produce erroneous readings due to the physical effects of mass flow.


The guide probably offers more detail than you might need for a typical nuclear application. However, it is still beneficial to at least be familiar with matching the right flow meter to the application. You can build a checklist from the Omega technical reference document that can help guide the flow meter selection process.

One thought on “How to select a flow meter

What do you think?