Level Interface Measurement
Measuring a Level Interface is a very common requirement in many industries, for example in the Oil & Gas industry, the oil/water interface in the main separator needs to be controlled. There are many techniques employed to measure level interface and depending on the fluids being measured, some work better than others.
This guide explores the pros and cons of some of the more commonly encountered.
What is a Level Interface?
The disimiliar density, or specific gravity, of two liquids contained in the same vessel or tank, means the higher density liquid will sink to the bottom allowing the lower density liquid to float above above. Where these two liquids meet is called the Interface.
Sometimes the liquids will separate readily and the interface will be distinct, whereas other times an emulsion layer will form. In the US and Canada an emulsion layer is often refered to as a Rag Layer.
What is an Emulsion?
An emulsion is a mixture of two or more liquids that are normally immiscible, e.g. oil and water. Continuing with this example we can have an oil-in-water emulsion, where the oil is the dispersed phase and the water is the dispersion medium, or we can have a water-in-oil emulsion where the water is the dispersed phase and oil is the dispersion medium. It is also possible to have multiple emulsions, for example a water-in-oil-in-water emulsion.
Why is it Important to be Aware of Emulsion Layers?
An emulsion layer can lead to inaccuracies in the measurement of level interface. The density of an emulsion layer will vary depending on the how much of the dispersed phase is dispersed in the dispersion medium, and this variation in density can fool the measuring sensor into thinking it is measuring either the upper or lower liquid layer when it is actualy measuring the emulsion layer. For example, with an oil and water interface if there is a high percentage of water in the emulsion (approx 80% or more) then many measuring devices will consider the full emulsion layer to be water.
Guide to Choosing Level Measurement Sensors for Interface Duties
There isn't a single sensor that will be suitable for all level interface duties, and quite often there will be more than one measurement technique that will be suitable. So each application should be considered on its own merit with regard to the physical properties of the liquids encountered, how these properties may change with changes in process pressure and temperature, number and size of nozzles availabe for use on the vessel or tank, cost of installation and maintenance of transmitter, and familiarity of the plant operators with the chosen technology.
Generally speaking, as long as there is a good difference in a physical property between the two liquids then we will be able to choose a sensor, e.g. a difference of densities (0.8:1.1 would be considered a typical ratio), electrical conductivity (1:1000 is common), thermal conductivity, opacity, or sonic transmittance.
The following chart will give a good starting point for which type of sensor may be suitable for some common applications, and should be used in conjunction with the notes and discussion below:
Differential Pressure transmitters can continuously detect the interface between two liquids, but if their difference in density is small, a DP cell will only detect a small pressure differential potentially leading to accuracy errors. Changes in liquid density (perhaps due to temperature variations) typically produce 5 to 10 times the error on an interface calibration than they do on a single-liquid calibration. A major limitation is that the range of interface movement must cause a change that is as great as the minimum DP span. DP transmitters will not be able to accurately measure the thickness of an emulsion layer due to its variation in density, though approximations can be made by using the average density between the upper and lower density fluid to represent the emulsion density.
Float and Displacer type instruments can be used on clean fluid services. For float-type units, the trick is to select a float density that is heavier than the light layer but lighter than the heavy layer. With displacer type sensors, it is necessary to keep the displacer chamber flooded with the upper connection of the chamber in the light liquid phase and the lower connection in the heavy liquid phase. By so doing, the displacer becomes a differential density sensor and, therefore the smaller the difference between the densities of the fluids, and the shorter the interface range, the smaller the force differential produced. To produce more force, it is necessary to increase the displacer diameter. The density of the displacer must be lighter than the density of the heavy phase. Float and displacer instruments suffer the same issues as DP transmitters when emulsions are present.
Guided Wave Radar (GWR) transmitters have a rod or cable that extends into the vessel or tank to guide the radar signal. Part of the radar energy is reflected at the surface of the upper liquid and travels back to the transmitter. The time it takes the signal to go down and come back is proportional to the distance to the liquid. The remaining radar energy travels through the upper media, and it is reflected at the interface where there is a change in dielectric. Therefore GWR transmitters can give an output for the overall top level and the interface level.
GWR measurements are unaffected by density and can detect the interface between two liquids as long as the dielectric constant (dk) between the two media is greater than 10 dk and any emulsion or rag layer is less than about two inches thick. GWR is not suitable for thick rag layers. The upper media has to have a dielectric of less than 10 dk, or not enough of the radar signal can penetrate through it to read the interface.
In the unusal case of having to measure the interface between two insulating liquids GWR can be a good choice.
Capacitance transmitters have a coated rod or cable, often referred to as a probe, that extends to the tank bottom and measures capacitance of liquid that is in contact with the probe. The rod or cable of a capacitance transmitter acts like one plate of a capacitor, and the metallic tank wall acts as the other plate. Any liquid between the two plates changes the reading from the probe.
For a capacitance transmitter to measure the liquid interface, one of the liquids must be conductive and the other non-conductive - depending on your liquids, this may prohibit the use of a capacitance transmitter. The influence of the conductive material on the sensor is what drives the reading, as the non-conductive material will have a very small effect on the output. An advantage of using capacitance in interface applications is it is not affected by emulsions or rag layers.
A capacitance transmitter can have inaccuracies if there is a change in the conductivity of the liquids. Note that, unlike GWR, capacitance will only give one output which will be the interface (or the thickness of the conductive media according to how the unit is configured).
Hybrid Capacitance/GWR transmitters are increasingly being used for interface measurement. Typically these combine guided wave radar and a capacitance sensor on one coated rod or cable. The transmitter will normally use guided radar to make the interface measurement. If the emulsion layer gets too thick and the transmitter loses the guided wave reflection from the interface, it will automatically switch over to capacitance to make the measurement. The use of these two technologies has the potential to eliminate many of the problems associated with interface level applications when an emulsion layer is present.
Nucleonic uses nuclear radiation to detect levels and interfaces. A radioactive source is mounted on one side of the tank and emits energy through the wall of the vessel and process liquid. The radiation is measured by a detector or multiple detectors on the other side of the vessel. When used for interface measurement, the attenuation of the radiation energy is different as it passes through products with dissimilar densities. Multiple detectors are commonly used to locate an exact interface location and to identify multiple interfaces or layers.
For large diameter, thick walled vessels it is common for the radioactive source to be inserted in a stilling well inside the vessel to reduce the distance the radiation must travel, thereby allowing a less strong source to be used.
Nucleonics can cope with most interface measurements, but they are comparitively expensive compared to many of the other available technologies and must comply with sometimes onerous nuclear radiation rules and regulations.