Interface level detection — what to consider

Compared to standard level measurements in liquids, the detection and measurement of interfaces has its own particular challenges, leaving many interface applications without reliable level measurement. Most popular today are the electronic or non-mechanical solutions with no moving mechanical parts which offer much higher long-term stability and low-maintenance operation.

Ideally, an interface would be a clear separation between two media with different densities. Practically, however, a transition or emulsion layer often exists between the two media (see Figure 1). Consequently, multiple interfaces exist if one considers the transition from the first medium to the emulsion layer, and then from the emulsion layer to the second medium. So with the addition of each different medium two new interfaces are presented. One example is an oil-water separator in the crude oil extraction process. At the bottom there will be a sedimentation layer, consisting of sand and other heavy substances, followed by a liquid-solid interface to the water layer. During the separation of oil and water an emulsion layer will be formed, and frequently a foam layer can be found on top of the oil — five interfaces!

Figure 1: Different situations for interface measurement.

Process conditions not only affect and change the properties of the media but also influence the performance of the selected interface measuring technology. For example, traditional displacer or differential pressure technologies operate based on the different densities of the media. Changes in process temperature and densities of the media will cause measurement errors. Similarly, build-up and scaling on the sensing elements is another factor that affects mechanical systems with moving parts.

When measuring interface levels, there are further considerations. Is the overall level constant or variable — and, if variable, what is the variation? Is only the interface level measurement required, or is the overall level needed as well? Answering these questions will help in the selection of the appropriate instrumentation.

Commissioning procedures also have to be taken into account as some measuring systems can only be commissioned using a ‘wet calibration’, which means the vessel must be filled with both process fluids, one at a time. Not only does this add to the complexity but also incurs costs during commissioning and throughout the life cycle of the instrumentation. These costs of ownership need to be considered.

Mechanical systems

Although these systems still offer a wide range of applications, their use has declined in favour of electronic systems which offer the following advantages over mechanical systems:

• Capital costs: Although there are cost advantages on the smaller and simpler measuring applications, on larger and more complicated requirements, mechanical systems can cost up to three times more than electronic solutions.
• Poor dynamics: When interface measurement is between two products with small differences in specific gravity, or where the span of interface change is small (small dynamic), mechanical systems cannot offer a high enough measurement resolution.
• Poor flexibility: If either the media changes or the density changes at any time, the complete mechanical design has to be resized and the construction modified.
• Poor reliability: Sedimentation or scaling and corrosion on the mechanical parts not only influences accuracy but can lead to the complete failure of the system.
• High maintenance: Today, plants run with lean engineering resources and maintenance-intensive systems only add to operational costs.

Electronic systems

With the increasing popularity of electronic over mechanical systems, following is a brief overview of the most recognised electronic methods.

Differential pressure (DP) DP transmitters can be used to measure the interface between two fluids that have different densities (ρ). Unlike hydrostatic level measurement, the difference between the densities (ρ1 - ρ2) of both liquids has to be considered for the calculation of the differential pressure. Since the calculation for interface level measurement is based on the difference of the product densities, the resulting DP is small. This can make the accuracy of the DP method unreliable. External factors such as the way temperature changes affect remote chemical seals and capillaries, or density changes in the products, can produce side-effects that prevent the desired accuracy.

Figure 2: Interface measurement using differential pressure.

While DP interface measurement is one of the most popular and cost-effective solutions, and offered by most reputable vendors, it is also most vulnerable to external influences such as fluctuating ambient temperatures and changing product densities, and users need to carefully take this into consideration.

Capacitive probes

The instrument probe in combination with a metallic tank forms the two plates of a capacitor, with the medium in the tank acting as the dielectric. When the dielectric changes, the capacitance changes. Consequently, capacitance offers a simple yet precise interface measurement of media with different dielectric constants (ε_r).

This technology offers certain advantages over other measurement technologies:

Overall level changes: The measurement of the interface is still reliable as long as the εr of the lower medium is sufficiently higher than that of the upper medium. Emulsion layers: Since capacitance measures the combined εr of both media, it is unique in that it produces a stable measurement even when faced with emulsion layers with no well-defined interface. Density and temperature variations: Capacitance is immune to these changes. Suitability software: The suitability of capacitance technology for an interface measurement is easily determined using application software tools. Dry calibration: Using application software, wet calibration can be completely avoided, resulting in significant savings in user effort and cost.

Figure 3: Interface measurement using capacitive probes.

Radiometric measurement

Radiometric (gamma) measurement is non-intrusive and thus ideally suited to level measurement applications in which other technologies will not be able to operate due to harsh process conditions such as high temperatures and pressures, or due to physical limitations. These systems are high cost and therefore are only considered after all other technologies have been exhausted.

Figure 4: Interface measurement in a separator using radiometric systems.

Although high end, radiometric measurement offers solutions for problems that cannot be solved by other methods. For example, the use of multiple sources and detectors makes it possible to implement the detection of multiple interfaces, providing a product ‘profile’ in the vessel, and is far less costly than other ‘profilers’ available in the marketplace.

While radiometric systems have a high initial investment, the long-term reliability of this instrumentation solution drastically reduces downtime. Modern technology minimises radiation concerns as very low nuclear source activities are required. Additionally, reputable vendors today offer a single device that can measure level, interface and density, thereby reducing the investment in spare parts.

Guided radar Level measurement with guided radar operates using high-frequency radar (electromagnetic) impulses which are guided along a wave guide, a suspended conductive rod or rope. When the impulse reaches the surface of a medium, the change of the electrical impedance causes part of the energy to be reflected. The measured time of flight between emitting and receiving of the impulse is directly proportional to the distance between the measuring instrument and the product surface. Only a part of the radar impulse is reflected by the medium surface. The remaining energy penetrates through the medium and part of it is reflected once it reaches the surface of another medium further down. This effect is used for interface measurement with guided radar. This technology does have some boundary conditions which need to be satisfied for reliable measurements. These are: The dielectric constant (εr) of the upper medium has to be known, needs to be <10 and needs to remain constant. It must also always be >60 mm thick. The difference in εr between the media has to be >10. Emulsions layers need to be <50 mm thick.

Figure 5: Interface measurement using guided radar.

Although the boundary conditions are strict, guided radar offers the following advantages:

• No wet calibration required.
• Independent of medium density.
• Direct replacement of mechanical displacers in the existing chamber.

Conclusion

There is no universal technology that covers all interface applications, highlighting the importance of the correct selection to provide best performance with optimised cost of ownership.

Endress+Hauser Australia Pty Ltd
www.au.endress.com