Electromagnetic flowmeters with wetted electrodes have the conductivity limit for water at 20 µS/cm, while the conductivity limit for other fluids is set at 1 µS/cm. More and more people are asking why this limit is set and in addition, why other flowmeter suppliers advertise with other conductivity levels.
This report provides background information regarding the conductivity limits with the Krohne Optiflux range.
Flowing water (with a low electrical conductivity) generates electrical noise signals. These signals from the measured liquid approach the level of the flow signal from the electrode. So much so, in fact, that the amplitude of these noise signals can be much higher than the electrode signal at a flow velocity v = 3 m/s. The result can be an unstable, faded and violently fluctuating indication of the actual flow.
Research in the 1970s, together with many field experiences showed that this noise effectively (or only) appeared with water with a conductivity of < and equal to 20 µS/cm. With other fluids, this effect only appeared at a conductivity of < and equal to 5 µV/cm. That is why Krohne printed these conductivity limits in its technical data sheets. With Optiflux (after various tests) Krohne have reduced this limit to 1 µS/cm.
Most competitors followed Krohne's first specifications in the 1980s and early 1990s. These days, when a manufacturer claims they are able to measure water with an electrical conductivity of 5 µS/cm - while using a standard electromagnetic flowmeter with pulsed DC field - they have either no experience with this medium or with electromagnetic flowmeters. Or they were simply lucky once...
One particular brand still specifies its flowmeter with an electrical conductivity of 5 µS/cm (no reference to water is made). But it is known that this brand has exactly the same limitations; or problems, when the previously discussed limitations are exceeded.
Background
The effect strongly depends on various properties of water. The electrical noise level of water with low electrical conductivity mostly appears only with turbulent flow. Turbulent flows have a transition phase, in which the flow velocity goes from zero (at the pipe wall) to the higher value just a little bit away from the pipe wall. The friction of the flow paths with each other is the largest in this transition phase.
Electrical effects in this transition phase
Most people know about static electricity: if you rub isolated or poorly conducting materials against each other, electrical voltages build up (sufficient to cause sparks in many instances).
Rubbing non-conductive fluids on non-conductive pipes also causes static build up. Benzene in plastic pipelines can develop static voltages as high as 10 kV. This voltage depends on the conductivity of the fluid and results in an electrical charge. The charge is higher for mediums having higher dielectric constants.
Water's dielectric constant is 40 times higher than most other fluids, so water, for a given voltage, charges as much as 40 times more than other fluids. This charge remains present, and at a higher voltage for up to 40 times longer. The charge voltage is transferred to the electrodes, and with this action, together with the swirling motion of the turbulent flow present, the voltage fluctuates - this is the noise that begins to swamp the signal that needs to be measured.
What to do with water < 20 µS/cm
Krohne recommends the Optiflux 7300 C (Capaflux), which has larger electrodes for measuring the average value of the voltage from many small whirls, thus reducing the noise in the measured signal. A VA-meter measures non-conductive media, and there are also the UFM and the Optimass, which measure flow of non-conductive media. Krohne believes that it has a flow solution for almost all media.
| Tel: | +27 11 314 1391 |
| Email: | [email protected] |
| www: | www.za.krohne.com |
| Articles: | More information and articles about KROHNE |
© Technews Publishing (Pty) Ltd | All Rights Reserved
printer friendly version