1. How do paddle wheel flow meters work?

Paddle wheel flow meters (also known as paddle wheel flow sensors) are devices which measure the volume flow rate of liquids. They are based on a similar concept to that of the historical water wheel. The sensor consists of a paddle wheel inside a stainless steel or plastic housing. The paddle wheel is a rotor with a number of flat blades extending perpendicularly from the rotor hub. The paddle wheel axis is positioned above the liquid flow so that the flow strikes only the blades below the axis, causing the paddle wheel to rotate. The rotational velocity of the paddle wheel is proportional to the flow velocity and therefore also to the volume flow rate. The flow rate can therefore be determined by measuring the rotational velocity of the paddle wheel, which is performed by counting the blades using either a hall effect proximity sensor or an infrared through beam sensor. The preference between the two methods depends on the material from which the paddle wheel is made. Hall effect sensors, which can only detect ferrous materials, are used to count the blades of stainless steel paddle wheels. Infrared through beam sensors are used to count the blades of plastic (normally PVDF) paddle wheels.

Illustration of paddle wheel flow meter cross section

The output of the Hall effect (or infrared through beam) sensor is converted into a square wave of frequency proportional to the flow rate. A notable characteristic of paddle wheel flow meters is that they cause much smaller pressure losses than turbine and oval gear flow meters. Furthermore, they are comparatively insensitive to liquid viscosity, as long the Reynolds number is sufficiently high that the flow regime remains turbulent. A non-turbulent flow regime leads to a non-uniform velocity profile across the pipe and therefore to an underestimation of flow rate (because the paddle wheel is positioned off center). Paddle wheel flow meters are bi-directional meaning that they can measure flow rate in both directions. However, most cannot detect the flow direction. Detecting flow direction requires the use of two rotational velocity sensors, fixed at a 90° phase difference. The two sensors output two square wave signals at a 90° phase difference, known as a quadrature output. By determining which square wave is leading, the flow direction can be determined. This is precisely the same method as that used by angular encoders to detect the direction of angular motion.

Illustration of paddle wheel and hall effect proximity sensor

2. General characteristics of paddle wheel flow meters

Paddle wheel flow meters measure the volume flow rate of liquids. They operate on the same principal as the historic water wheel, in which the velocity of water flowing past the lower rotor blades causes the wheel to rotate. They are commonly made from either stainless steel or plastic (PVDF, Polypropylene, PVC). Paddle wheel flow meters are less accurate and less robust than turbine and oval gear flow meters. Their main advantages are their low cost, small pressure drops and high turn down ratios (as large as 1:100). Furthermore, paddle wheel flow meters are less severely affected by changes in liquid viscosity than turbine flow meters, as long as the flow regime remains turbulent. Paddle flow meters are unsuitable for use with liquids containing solid particulates. It is therefore recommended that a filter is placed prior to the flow meter inlet to trap solid particulates. Paddle wheel flow meters require straight sections of pipe before and after the sensor to straighten the flow and reduce swirling.

3. Input and output signals

Paddle wheel flow meters require a 10-30 Vdc supply voltage. The supply voltage is used to power the Hall effect or infrared through beam sensor and the built in electronics. The sensor outputs a square wave signal at a frequency proportional to the volume flow rate. Paddle wheel flow meters are also available with 4-20 mA analog output signals. Furthermore, models are available that generate two square wave outputs, 90° out of phase with one another. The dual output increases resolution and enables flow direction to be determined. The square wave output signal often necessitates that an external pull up or pull down resistor be used, to prevent the occurrence of a floating output (see the application tips).

4. Applications of paddle wheel flow meters

Paddle wheel flow meters are used in many of the same clean liquid metering applications as turbine flow meters. They are particularly attractive for applications that require low cost, high chemical compatibility and low pressure loss but for which high accuracy is less important. Typical applications for paddle wheel flow meters include metering of flow volume within the chemicals manufacturing, process manufacturing and oil industries. Paddle wheel flow meters are also commonly used to provide closed feedback for pump control systems.

5. Typical specification

CostLow cost
Measurement range0.1-1 L/min to 1-500 M3/min
Liquid temperature-18 to 85 °C (up to 350 °C available)
Max pressure15 bar typical (150 bar available)
Turndown ratio1:100
ViscosityRe > 4,000
Accuracy1-10%
LifetimeLow to medium
Supply voltage6-28 Vdc
Output signalSquare wave or 4-20 mA
Passive / activeActive
Contact / non-contactContact

6. Purchasing tips

  • Material compatibility: The housing and paddle wheel of the flow meter are often made of stainless steel, with a ceramic axle and elastomer seal. However, the housing may also be made of aluminum or PEEK and the paddle wheel is often made of PVDF. Stainless steel provides significantly better abrasion resistant than PVDF but PVDF has superior chemical compatibility.
  • Uni/bi-directional: All paddle wheel flow meters are able to measure flow in both directions. However, most cannot detect the flow direction. Detecting flow direction requires the use of two rotational velocity sensors, 90° out of phase with one another. The two sensors generate two square wave output signals 90° out of phase with one another. By detecting which square wave leads the other, flow direction can be determined.
  • Pressure drop: The pressure drop across the sensor is normally small because the paddle wheel only blocks the upper portion of the pipe cross section. However, some paddle wheel flow meters, particularly those designed for low flow rates, do generate large pressure drops, often in excess of 1 bar.
  • Viscosity: Few manufacturers quote a viscosity range for the paddle wheel flow meter. However, for accuracy to be maintained, the flow regime must remain turbulent so that flow velocity is uniform across the pipe. To ensure a turbulent flow, the Reynolds number of the flow must be greater than 4,000.
  • Output signal: Most paddle wheel flow meters output a single square wave signal. Others output two square waves (for detecting flow direction) or a 4-20 mA analog output.
  • Turndown ratio: The turndown ratio is the ratio of smallest to the largest flow rate that can be measured. Like turbine and oval gear flow meters, paddle wheel flow meters cannot measure very low flow rates. However, many paddle wheel flow meters have very high turn down ratios, often 1:100 or more.
  • K factor: The k factor is the number of pulses (i.e. peaks and troughs of the square wave) per unit volume of liquid and often has the units of pulses per liter. The k factor is a measure of the resolution of the sensor.

7. Advantages of paddle wheel flow meters

Paddle wheel flow meters:

  • Are probably the lowest cost flow meter available.
  • Very large flow rates and high turndown ratios are available.
  • Do not create large pressure drops.
  • Non-metallic, highly chemical resistant models are available.
  • Can measure flow in both directions (though most cannot determine flow direction).
  • Can be installed in any orientation.

8. Disadvantages of paddle wheel flow meters

Paddle wheel flow meters:

  • Are of relatively low accuracy, particularly when compared to other rotary flow sensors (e.g. oval gear and turbine flow meters).
  • Are susceptible to errors due to non-uniform flow velocity profiles.
  • Are not suitable for gases or high viscosity (laminar flow profile) liquids.
  • Are not designed for high pressure operation.
  • Are not suitable for use with liquids containing particulates.
  • Require straight pipe sections to be installed at their inlet and outlet for straightening the flow.

9. Application tips

  • Calibration: Paddle wheel flow meters can display some sensitivity to liquid density and viscosity. They therefore need to be calibrated with the specific process liquid if maximum accuracy is required. Unfortunately, because viscosity is highly sensitive to temperature, the output of the paddle wheel flow sensor may also be sensitive to temperature. We are not aware of any paddle wheel flow sensors with built in temperature compensation. However, turbine flow meters are available with built in temperature compensation.
  • Mounting: Paddle wheel flow meters may be mounted in any orientation. If there is a significant pressure loss across the meter (as is common with low flow rate meters) and you intend to mount vertically, it is recommended for the liquid to flow upwards.
  • Straight pipe: Paddle wheel flow meters require straight pipe sections to be connected before and after the sensor to straighten the flow. Typical minimum recommended straight pipe lengths are 20 pipe diameters at the inlet and 10 pipe diameters on the outlet. However, longer straight sections may be required for small diameter pipes.
  • Filters: Particulates will wear the paddle wheel over time. It is therefore recommended to place a filter at the sensor inlet to trap any solid particles in the liquid.
  • PNP/NPN transistor: The flow meter output is controlled by a transistor of PNP or NPN type. A transistor is a switch which either conducts or does not conduct electricity. A PNP type transistor conducts during the logic 0 portion of the square wave and an NPN transistor conducts during the logic 1 portion. When the transistor is not conducting, the sensors output is floating and will pick up background noise. The output must be tied to either the supply voltage (for PNP) or to ground (for NPN). This is performed using a pullup resistor (for PNP) or a pulldown resistor (for NPN). Process controllers normally have built in pullup or pulldown resistor. However, care must be taken to make sure that the sensor is connected to the corresponding port type on the controller. Additional information is available on the Azom and se websites.