Under pressure: an engineer's guide to pressure switches

When systems fail, engineers tend to focus on the big, the expensive or the exotic.  However, it is often a problem with a simple or inexpensive part that shuts an application down or affects performance. The cost of component failure can be significant and is measured in unscheduled outages, costly downtime and lost production.

Pressure switches are one of the most common types of component in process plants. Yet they are also one that design engineers know very little about. Despite this, a clear understanding of the basics and a reasoned, methodical approach will make the selection process much easier. In its simplest form, a pressure switch is a device capable of detecting a pressure change, and, at a predetermined level, opening or closing an electrical contact. 

Traditionally, pressure switches have been electromechanical devices. Their basic design has been employed for more than two hundred years. But today competition for this traditional technology is being provided by solid-state devices. The most common electromechanical pressure switches consist of a sensing element and an electrical snap-action switch. A number of different types of sensing elements can be used but they have one thing in common: they move in response to changes in the system pressure. Through their movement they directly act on the opening and closing of the snap-action switch's contacts. 

The current market provides a wide variety of solid-state pressure switches, with one to four or more switch points, digital displays, analogue and digital outputs, and full programmability. In many cases they cross the line from simply being a switch to becoming an open-loop controller. In addition to opening or closing the pressure switch circuit or circuits, they provide a proportional analogue 4-20 mA signal or digital output. The analogue signal can interface with PLCs, DCSs or stand alone industrial computers.   

Solid-state pressure switches provide a number of advantages over electromechanical switches, including a much longer cycle life, improved accuracy to ±0.25%, high resistance to shock and vibration, the ability to handle a wide range of system pressures, broad frequency response and excellent long-term stability. But the biggest advantage lies in cycle life. Solid-state switches routinely have an operational span of 100 million cycles. 

Despite this, one problem with using solid-state switches in industrial and process settings is the fact that electromagnetic interference (EMI) can corrupt signal data. In addition, a solid state switch requires an input power source to function. EMI and radio frequency interference does not affect electromechanical switches because the circuit is a mechanical switch that is either open or closed.

The first and most important step in selecting a pressure switch is to fully establish your requirements before beginning the process. Armed with that knowledge, you must consider a number of parameters in making a final selection: what kind of pressure sensor you need, cycle speed and life, pressure range, accuracy, number of switch points and deadband. You must also ask yourself if you need an adjustable or non-adjustable switch and decide between electromechanical and solid state technology. 

The frequency with which the switch is activated will have direct impact on switch life, system downtime and the maintenance schedule. Due to their design, electromechanical switches are subject to metal fatigue although solid state switches aren't. Cycle speed will also affect switch life and preventative maintenance programmes once the design is used in anger. A solid-state switch should be selected whenever the cycle rate exceeds 50 cycles per minute so that metal fatigue is not a problem. 

Establishing the right relationship between the switch point and the operating pressure range of a switch is also important. When a solid-state pressure switch is selected, the switch point should normally be in the upper 25% of the operating range. For an electromechanical switch, the switch point should be in the middle of the operating range. Thus, a system that requires a switch to activate at 140 psi should use a solid-state pressure switch with an operating range of 150 psi, or an electromechanical switch with an operating range of 300 psi. 

The location of the switch point versus the operating range is critical to both accuracy and life.

Pressure switch accuracy is defined as the ability of the switch to operate repetitively at its set-point. If the switch is used to trigger an alarm, ±2% accuracy is sufficient. If one is controlling a process where the error of various devices is cumulative, then ±0.25% accuracy may be absolutely necessary. Accuracy is referenced at the high end of the operating pressure range and decreases at lower pressure. 

Once the required accuracy is established, we should decide on the number of switch points need. When sensing pressure at one point, it is normal that only one switch point is required. Nevertheless, it's not unusual for a system to require two or even four switch points to be monitored, controlled or alarmed. In designing a system, one could select a single switch for each switch point, or a single pressure switch capable of handling as many as three separate switch points. 

A related issue is deadband, or the difference between the actuation point and the reactuation point in a pressure actuation switch. For example, if the device is set to operate at 100 psi on increasing pressure, the switch will close when pressure rises to that point. As pressure drops to 95 psi the switch opens - this is the re-actuation point. In this case, the deadband of this switch is five psi, the difference between the set point of 100 psi and the re-actuation point of 95 psi. 

If one considers all of the factors that have to be taken into account when designing a pressure switch into an application, the hidden complexity in one of the manufacturing and process industry's simplest components is revealed. However, despite this hidden complexity, this kind of switch remains one of the simplest to integrate and maintain. The job of the design engineer is to make that process even easier.