The art of level instrument selection

By Hunter Vegas

May-Jun automation basicsIn the January/February 2013 issue of InTech, Donald Gillum wrote an excellent article called “Level measurement.” In that article, Gillum discussed the various types of level instrumentation used by industry today and described how they work and some of the advantages and disadvantages of each.

This article is a follow-up to Gillum’s article and will focus on how an engineer might select a level instrument and the specific reasons why one type might be chosen over another. Before beginning this topic, it is important to stress that no technology is perfect for every situation. Each type of level instrument has advantages and disadvantages and may work great in one situation and be inoperable in another. Also, instrumentation vendors are continuously improving their equipment to overcome weaknesses in their particular class of level device. Therefore it may be possible for one brand of instrument to overcome or compensate for an issue that normally vexes competing products. However, before trying a particular device that might have problems, consider obtaining an agreement from the vendor guaranteeing the device and allowing a full refund should it fail to function as promised. The refund will not cover the cost of the installation and replacement, but the vendor will have a strong incentive to make sure the device is successful.

Where to start?

To determine the right technology for a particular application, you must know the following:

1. How does each type of level technology actually work? You cannot determine if a particular level instrument will function in a given application if you do not understand how that device works.

2. In what conditions must the instrument perform? There are obvious conditions, such as temperature, pressure, and the general contents of the vessel, but there are also less obvious issues such as:

  1. What is minimum/maximum temperature and/or pressure the unit might see, and do those temperatures/pressures vary over time?
  2. Are there abnormal conditions, sterilization, or special chemical cleaning procedures that might affect the instrument?
  3. Are dust, fumes, or heavy vapors present?
  4. Is there foaming or excessive vessel agitation?
  5. What is the material in the vessel and does it change over time?
  6. Are there changes in specific gravity, electrical properties, or material consistency?
  7. Does the material tend to cling, plug, coat, or build up?
  8. Could the instrument become a source of contamination to the product (from diaphragm leakage or could bacteria hide in cracks and crevices)?

3. What is the required accuracy and repeatability of the level reading?

4. Can the vessel be taken out of service for maintenance, and is there any way to independently verify the level for calibration?

5. What are the details of the vessel itself? The vessel geometry and layout of the associated piping will often affect the selection of an instrument. You need to answer such questions as:

  1. Are there existing nozzles, or is this a new vessel and the nozzle can be located and sized to suit a particular instrument?
  2. If the nozzle is existing, you need to know its size, length, distance from the wall of the vessel, and the interferences that might be below it.
  3. What do the vessel internals look like? Are there internal baffles, coils, thermocouples, agitator, etc., which can create interference or make installation of a vertical probe practically impossible?
  4. Where are the inlet feed nozzles and outlet discharges? Can they affect the measurement?
  5. Is the vessel small enough to fit with weigh cells?
  6. Can an internal or external stilling well be installed? What size? Is the level within the external stilling well indicative of the vessel level?

Without complete answers to these questions, the engineer will find it very hard to choose an instrument, because the answers will usually determine which instruments will work and which will not. Let us reconsider these same questions and see how they can quickly eliminate many level instrument options:

Questions 1 and 2: There are many “show stoppers” that can eliminate a particular level technology. Here is a short list of some of them:

  • Capacitance – if the electrical conductivity is low and variable, a capacitance-type level device can drift significantly.
  • Ultrasonic – if the material has foam or heavy fumes or heavy dust, or the temperature, pressure, or material in the vapor space changes, an ultrasonic transmitter can either drift or not function at all.
  • Radar – if the dielectric constant is too low, radar may not function. Also, some vessel geometries can eliminate different types of radar devices.
  • Differential pressure (DP)/head – if the specific gravity of the material is changing, most DP measurements will drift significantly. (Compensation may be possible at added expense.) If the material tends to plug, it can render a DP measurement meaningless over time.

Question 3: The required accuracy and repeatability of the measurement can significantly impact the selection of the level instrument. If the level instrument is used for custody transfer, then maximum accuracy and repeatability is paramount. However, in other cases, accuracy within a percent or two is perfectly acceptable. Some level technologies are capable of extremely high accuracies, while others can offer improved accuracy by employing additional compensating measurements. Before beginning the selection process, it is important to understand how precise the measurement must be.

Question 4: The ability to take an instrument out of service can radically affect the choice of instrument. If the instrument is adversely affected by product build over time, then the instrument must be removed for routine cleaning to ensure reliable operation. If the instrument cannot be removed for service, then a different technology should be considered.

It is also important to know if the device’s measurement can be independently verified. Safety integrity level devices usually require some type of routine calibration to document that the device is functioning properly. Some technologies, such as DP transmitters and weigh cells, are inherently easy to calibrate. Others must be calibrated on an actual level, such as capacitance or nuclear technologies. Still other technologies, such as radar or ultrasonic, usually require no calibration. These types of instruments are easier to set up initially. However, once they are in service, it is difficult to verify their operation unless there is another way to determine the true level in the vessel.

Question 5: The geometry of the vessel itself may eliminate several technologies. Large agitator blades or internal coils may render several technologies unusable unless some type of stilling well can be installed. There may be no available nozzles to install a DP transmitter, or the nozzles on the top of the tank may not be large enough for a radar device.

Clearly these five questions can significantly affect your options in choosing a level device—so take the time to fully understand the application before considering the measurement options.

Level technology review

The following section broadly reviews the issues and weaknesses associated with the more common level measurement technologies. This information, combined with the answers to the previous questions, helps an engineer select the best level instrument for a particular application. Relative cost information is also provided.

Displacer/float:
Float-type devices usually require a reasonably clean service or the device cannot move up and down easily. Displacers can be installed in dirtier services if polymer build up in the vapor space does not restrict the displacer movement. In either case, the device is dependent on material density. A float may rise higher or lower if the density is changing and may not float at all if the specific gravity is too low. A displacer is calibrated for a particular density and will read incorrectly if that density changes.

Cost: $$–$$$. Floats are usually less expensive than displacers.

DP – bubblers:
The bubbler is a simple technology that can work well in a large variety of applications. However, the environmental impact of a continuous vapor stream may eliminate the bubbler as an option. A bubbler’s performance can be affected by plugging of the bubbler tube, pressure drops in the tubing, tubing leaks, and the consistency of the specific gravity of the liquid.

Cost: $$. Installation and rotameters make this option cost more than a DP-tube.

DP – pad type:
The pad-type DP transmitter is a very common level device in industry today. However, the transmitter assumes that the gravity of the material does not vary (or if it does vary, then the installation has a means of measuring that variation and compensating for it).

Many issues can affect a pad-type DP’s performance. Vacuum, low temperature, and high temperatures can render many sealed diaphragms inoperable or very slow to respond. Also, variable process or vapor space temperatures can create drift in the level reading. Remote level seals can also be affected by changing ambient temperatures.

Cost: $$–$$$. Large diaphragms are expensive, and if a low side seal is needed the price will be higher still.

DP – tubing:
Installation of a DP transmitter that uses static tubing to sense the process is another very common level device. The chief issues here are plugging of the static lines and the dependence of the device on the gravity of the material in the vessel and the gravity of the material in the static lines. (Turning off the heat trace on the static lines of a level transmitter on a boiler drum can cause the reading to drift significantly!) Also watch out for issues on the low side impulse line. Trapped condensate or even continuous gas condensation can shift the zero.

Cost: $. This is one of the cheapest level options.

Ultrasonic transmitters:
Ultrasonic transmitters depend on the vapor within the vessel to transmit the ultrasonic waves. The level reading is affected when those vapor properties change due to varying pressure, temperature, or composition. The signal is also attenuated if there is agitation, foam, fumes, or dust present. If any of these conditions are present or could be present, an ultrasonic device may not work. Also, know and understand the blanking distance of the device. If the level should rise within this distance of the sensor, the transmitter will suddenly switch from reading a high level to reading no level at all.

Cost: $$.

Radar transmitters:
Unlike ultrasonic transmitters, radar units are generally unaffected by the composition of the vapor space, unless the temperature or pressure swings significantly. Radar transmitters do require a minimum dielectric constant of the material they are measuring, and they can be sensitive to vessel geometry. Also a noncontact radar can struggle with a slanted surface or excessive surface turbulence. Perhaps the most troublesome problem is inconsistent readings at very low levels. Obtaining accurate readings at empty/near-empty conditions can be difficult. There are a lot of options when selecting a radar—frequency, antenna type, guided versus noncontact—and each has advantages and disadvantages. It is usually wise to consult the vendor’s experts, and let them help you select the best option for your application. Be sure they have accurate vessel drawings and a good understanding of the process conditions. Also look out for the location of feed piping, as the incoming materials can affect the reading.

Cost: $$–$$$. Radar costs have dropped significantly in recent years.

Nuclear level:
Nuclear level transmitters are generally more expensive and require special licensing and hygienic testing, but they can work where no other level transmitter will function. Their ability to “see” through vessel walls makes them impervious to extreme process conditions (e.g., temperature, pressure, plugging). Radiation exposure can be a concern if vessel entry is possible.

Cost: $$$–$$$$. Price can vary depending upon the configuration of the unit.

Weigh cells:
Weigh cells are excellent level devices, because they are generally unaffected by the chemical composition of the vessel contents. However the device is measuring mass, not level, so if the gravity of the material changes, then the level will change as well. Not all vessels can accommodate weigh cells, and proper installation of the cells is critical for success. Any attached piping must have flex joints, and there can be no interferences for proper vessel movement. Watch out for thermal growth (of the piping or the vessel itself) and variable weight caused by vessel jacket contents (steam/water/brine). Also watch out for operators standing or leaning on the vessel and changing the weight readout.

Cost: $$–$$$$. The price can vary depending upon the tank configuration and the amount of piping involved.

Conclusion

A successful level instrument design starts with a thorough understanding of the application and the strengths and weaknesses of the available level measurement technologies. Taking the time to fully research the application and the available measurement options can take you a long way toward avoiding angry phone calls from your operations department!

Where to Start?

To determine the right technology for a particular application, you must know the following:

  1. How does each type of level technology actually work?
  2. In what conditions must the instrument perform?
  3. What is the required accuracy and repeatability of the level reading?
  4. Can the vessel be taken out of service for maintenance, and is there any way to independently verify the level for calibration?
  5. What are the details of the vessel itself?
 

About the Author

Hunter Vegas (phvegas50@gmail.com) has worked in the automation industry for nearly 29 years and has executed more than 2000 automation projects in the nuclear, pulp and paper, and specialty chemistry industries. He is a frequent contributor to several controls magazines and recently co-published his first book, 101 Tips for a Successful Automation Career with Greg McMillan. Vegas currently works for Wunderlich-Malec as a project engineering manager and lives in North Carolina.

Reader Feedback

We want to hear from you! Please send us your comments and questions about this topic to InTechmagazine@isa.org.

>