At process plants, a significant percentage of measurement devices aren’t correctly matched to their application, leading to decreased quality and consistency of the operation. Often, the source of this problem is the assumption that one type of level measurement sensor suits multiple applications. For example, a float sensor may serve its purpose adequately in a completely liquid environment but, when the liquid contains particles or suspended solids, the circumstances change, compromising the accuracy of the sensor. Maintaining a high level of output quality in any plant requires selecting the proper level measurement device for each application. There’s no “one size fits all” solution.
Unfortunately, companies and individuals often are tempted to skip a lengthy and involved sensor selection process for a cheaper or faster outcome. This is ill advised, though. To ensure picking the right sensor for long-term durability and safety, engineers must consider all possible factors in a design scenario. After all, in a dangerous application, an incorrectly specified level sensor easily could become the source of a major incident.
A common question is, “Are all these detailed questions really worth it for deciding on a simple level sensor?” The answer is “yes” — because there’s no simple sensor.
THE SELECTION PROCESS
You must consider several factors to ensure the accuracy and effectiveness of a level measurement device in a particular application. These factors include the design conditions, the specific media the sensor will contact, how the information gathered from the device must be transmitted, and what additional accessories are needed to complete the operation. Properly considering each of these criteria will lead to choosing a sensor that can provide more accurate inventory, increased product quality and maximized output due to a lack of disruptions.
The first order of business in the selection process is to narrow down the application for which the sensor will be used. In other words, what will its job be? This could be as simple as giving a visual readout or as complex as serving in a multistage automated response system. Answering this question requires knowledge of how the gathered information will be used. Must the sensor continuously monitor the level of a volatile substance? If so, a continuous level sensor is necessary. On the other hand, if the aim is to stop a tank from overflowing or spilling, a point-level sensor might suffice. You easily can determine the complexity and versatility required of a sensor by first assessing the application.
The next step is to determine design conditions; these are crucial in choosing the correct option. Questions to ask at this stage include:
What material(s) will the sensor contact? A sensor must be compatible with the fluids or materials to which it will be exposed. If the sensor will contact any caustic, corrosive or aggressive materials, it must be able to withstand them while maintaining proper functionality. For example, a metal sensor used to measure critical process fluids could release metal ions or particles and contaminate the fluid. In this situation, selecting a sensor made out of a fully compatible material is best.
Are solids or liquids being measured? This is an extremely important question that must be considered upfront. Float-type sensors normally are a good solution for measuring liquids. However, when measuring solids, float-type sensors are impractical.
The following is a list of popular sensor types and the materials with which they are compatible:
• mechanical float sensors —fluids only;
• electromechanical (tuning fork or staff-based) —solid substances only;
• ultrasonic — both solids and fluids; and
• radar — both solids and fluids.
Remember, other design conditions besides the type of substance being measured could cause one of these sensors to perform better than another. The material being measured is just one factor to consider in device selection.
Where and how will the level sensor be placed — internally or externally? This question is important from a logistical standpoint. External sensors could require additional plumbing and installation costs while internal sensors may throw off production amounts or limit tank capacity. Some systems incorporate an internal sensor and an external display to convey information. In this case, always consider the requirements of both the internal and external components.
Is the material at rest or in motion? In a silo or storage vessel, a material may be stagnant at all times except during filling or extraction. In contrast, in a mixing tank, for instance, a substance may experience constant motion or agitation. In the latter case, ensure the sensor can withstand the movement of the material.
What temperatures and pressures will the sensor face? When considering temperature, sensors generally fall into two categories: those built to withstand extreme heat or cold, and those limited to common room temperatures. For example, a metal sensor will withstand an extremely hot environment much better than a polytetrafluoroethylene one. As far as pressure, most sensors can handle 0 to 100 psi — but an extremely high pressure environment or vacuum could cause rupture or malfunction.
What is the density of the material? This question applies almost exclusively to fluids. Those that are fairly dense or contain floating solid particles often require more-complicated level measurement sensors. So, for example, a simple float sensor will suffice in a water cooler tank but may experience difficulties measuring rough crude oil because of the viscous nature of that fluid. A material also may include a mix of fluids such as oil and water. In such a case, a capacitance sensor can accurately measure the mixture. A magnetic float sensor with the buoyancy factor adjusted to the material would work, too.
Are multiple sensors required? Operations demanding more-detailed information than simply monitoring an overflow level often call for multiple sensors. A common application is when tanks have a recommended capacity as well as further measurements above or below that capacity. For example, a processor may need to know if a tank reaches the level considered “low” and continues to drain. The more advanced a sensor is, the more flexible it will be in these types of situations.
Must the sensor comply with any specific design codes? This is a very important question when dealing with hazardous or flammable materials. The sensors in these services typically must be either explosion-proof or intrinsically safe. Basically, can the sensor in any way be a source of ignition? Can a spark come from it under any conditions? If the answer to either of these questions is “yes,” the sensor likely won’t meet the required design codes for applications in the chemical, petrochemical, biotech, natural gas and semiconductor industries.
Does the liquid have a boiling or flash point? A sensor that can’t withstand the boiling or flash point of a material could become a safety concern. For example, alcohol has a low boiling point and its vapors may degrade the seals of some sensors over time. This is a particular concern in some biotech and pharmaceutical applications and in semiconductor processes.
What level of measurement precision is desired? This impacts a sensor’s price as well as the delicacy and intricacy of the device. A radar or infrared sensor accurate to ±0.5% costs substantially more than a simple float-level sensor accurate to ±5%. However, for certain critical applications, this investment quickly will pay for itself with more-accurate level measurement.
Is steam present in the process? Steam requires use of a more-robust, durable sensor. In biotech processes, steam-in-place cleaning often is used to sterilize tanks and components. A sensor present in this environment must be able to stand up to the harsh conditions.
What is the size and shape of the tank? A vessel’s dimensions and form directly determine the placement of the sensor in relation to the material. This is because different materials fill containers in different ways. Liquids occupy the lowest areas first and rise with a level surface while powders and other solids often stack to a point and leave lower areas empty — perhaps requiring placing the sensor directly in the middle so it can accurately measure the highest point of the material.
Do you need a point value or a continuous measurement? In a point-value measurement application, the sensor only will determine whether the material is above or below a particular point. In contrast, a continuous level sensor will measure the current level of material. Depending upon the application, this can be a very important distinction.
What other factors should be considered? You also may need to account for less-common design conditions such as mechanical shock (vibration or agitation caused by external forces), electrical interference or noise. For example, electrical interference from a radio transmitter like a walkie-talkie can affect the readout of a radio frequency sensor.
Be sure to thoroughly review each design condition of a given application before selecting a level measurement sensor. The specific conditions can make one measurement device suitable but another completely useless. For example, the distance between the sensor and the top of a “full” tank can determine whether or not an ultrasonic device is viable. Furthermore, a film or layer of floating solids at the surface of a liquid may disrupt the measurements of certain sensors. In short, a plethora of variables can and should contribute to the proper selection of a level-measurement sensor.
A specific concern at this point is assessing the potential impact of a material on the level sensor over time. For example, a material that’s corrosive, viscous or has suspended solids, that can transition from one state to another with changes in pressure or temperature, or that’s reactive or hazardous markedly will impact the selection of sensor type, materials of construction and other safety or durability features needed.
The next step in the selection process is to establish what will be done with the information collected by the device. If an alert must be sent when a tank reaches a certain level, the measurement device must be paired with the correct switch or output for transmitting that information. If multiple points along the path of a float sensor are to trigger different operations, each switch must be routed properly to its corresponding destination.
The last step in fully integrating a sensor into an operation is to choose accessories with appropriate functionality with which the sensor must interact to complete its desired task.
If each of these steps is considered with a desired end goal for the level measurement device in mind, the dependability, accuracy and consistency of level measurement should increase in any given operation.
BRIAN SULLIVAN is director of sales – technology for Valin Corp., San Jose, Calif. E-mail him at BSullivan@valin.com.