Mike Powers, Product Marketing Director for Gems Sensors and Controls, looks at developments in pressure transducer technology

Phil Black - PII Editor 22/07/2011

138 6 minutes read

Listen to this article

Pressure transducers are widely used in process and industrial applications, where they often have to withstand aggressive operating conditions, ranging from high pressure hydraulic spikes to extremes of temperature, mechanical shock and vibration. In each case transducers have to function reliably, consistently and accurately without maintenance or recalibration; indeed, modern transducers are now expected to be genuinely ‘fit and forget’ instruments.

Yet within each transducer there is typically an extremely sensitive pressure sensing mechanism, combined with a sophisticated electronics package. Together, these can provide accuracy better than 0.25% of full scale output, with almost zero drift over time, yet with a response to changes in pressure of 1msec or less, and an operating life in excess of 100 million cycles.

These outstanding levels of performance are only possible thanks to a number of highly innovative and carefully controlled methods of construction. Three of these make use of advanced strain gauge technology: sputtered thin film, chemical vapour deposition (CVD) and micro-machined silicon (MMS); while a forth uses capacitance as a method of detecting changes in pressure.

Atomically bonded
Sputtered thin film technology was developed some thirty years ago and evolved from the manufacturing processes used by the electronics sector for the production of integrated circuits. The production technique, although relatively straightforward, requires advanced engineering systems and carefully controlled conditions to create an atomically bonded strain gauge sensor on a stainless steel diaphragm.

This is produced by placing a material such as silicon dioxide in a vacuum and then bombarding it with argon ions. The atoms that are subsequently released are deposited in a layer on a stainless steel beam to form the base insulating layer for the strain gauge; using the same process this is then coated with further layers of a suitable gauge material, before being patterned using photoresist techniques. The unwanted areas are removed by sputter etching, to create a dielectrically isolated strain gauge in a conventional Wheatstone bridge arrangement, which is mounted on the reverse of a stainless steel diaphragm.

Displacement of the diaphragm thus causes the strain gauge to flex, either in compression or under tension, with the electrical output being directly proportional to the pressure or vacuum applied. Output from the sensor is connected to onboard electronics, with the entire unit being contained in a compact and sealed stainless steel housing.

This construction is extremely robust, with the diaphragm being suitable for direct contact with almost all liquids, oils and gasses. In addition, the materials of construction for both the sensor mechanism and the transducer as a whole are thermally compatible to minimise non-repeatable errors for hysteresis and thermal stability, and to ensure that coefficients for thermal zero and shifts in sensitivity remain constant across a wide temperature band.

Volume production
Although thin film pressure transducers provide exceptional levels of performance and long term stability, the mechanical complexities of the ‘first generation’ sensors, mean they could be relatively expensive to manufacture in large volumes. It was to address this need that the chemical vapour deposition process was originally pioneered, using semiconductor manufacturing processes to produce multiple sensors at lower cost, while retaining many of the benefits and performance characteristics of thin film devices.

CVD sensors are produced on wafers in large batches, using polysilicon deposited on a stainless steel substrate, with the strain gauge patterns being chemically milled. The wafer is then divided to produce individual sensor beams, which are laser-welded to a stainless steel summing diaphragm and pressure port, before being connected to internal electronics for signal conditioning and amplification.

This process enables sensor assemblies to be produced in volume and at low unit cost. Each sensor generates a high electrical output from a small mechanical deflection, thereby simplifying signal processing, and is inherently stable with an accuracy to within 0.5%. It also offers a long operating life with excellent resistance to pressure shocks and mechanical vibration. Additionally, the use of high temperature vacuum brazing of the stainless steel during sensor production creates a structure with low hysteresis and creep, together with high strength and corrosion resistance.

Smaller solutions
Traditionally, most pressure transducers have been manufactured to an outer diameter of around 25mm. This is satisfactory for many process and industrial applications; in the water, waste and drilling sector, however, where the trend is towards smaller diameter boreholes, a narrower diameter unit of around 19mm is required. This is difficult to achieve with thin film and CVD production methods, where the size of each sensor is limited by the mechanical requirements of the diaphragm.

By comparison, micro-machined silicon sensors are produced with similar technology to that used in the manufacture of integrated circuits on silicon wafers, with ion implantation allowing a strain gauge structure to be diffused into the internal lattice of the silicon. This optimises the unique mechanical and electronic properties of silicon, enabling sensor and diaphragm size to be reduced proportionately, and without adversely affecting factors such as hysteresis, linearity or performance under tough operating conditions. In common with CVD technology, the production methods used for MMS transducers allow larger volumes to be manufactured at lower unit cost, although unlike CVD devices where the diaphragm is in direct contact with the media MMS sensors are generally protected with oil filled isolation diaphragms.

Lower demands
In a number of applications, the requirements for pressure monitoring and control can be relatively modest, calling for a low cost, reliable but not necessarily highly specified device; or conversely a device that is capable of providing measurements at low pressures and at reasonable levels of accuracy, typically ±1% of full scale deflection.

This need can be met by the use of capacitance sensors, where a flexible ceramic diaphragm and a fixed plate form the two capacitance surfaces. Pressure or vacuum applied to the diaphragm will therefore cause a proportional change in capacitance, with the output signal again being fed directly to integrated electronics for subsequent conditioning and amplification.

It should be noted that although these devices are low in cost, with good temperature performance, they are also limited in application; in particular, the sensor construction often employs the use of O-ring seals which can be prone to leaks. Great care has also to be paid to potential media compatibility issues with the seals employed.

Electronics hold the key
The transducer technologies described above have been available for some time and are proven in many different applications. In each case, there has in recent years been developments in the respective manufacturing processes which, combined with detail changes in design and materials of construction, has led to a steady improvement in areas such as performance, stability and reliability.

For example, the latest CVD devices now use smaller sensor assemblies and have been engineered to reduce the overall dimensions, to be used for low (0-16bar) pressure ranges. Conversely, the capability of thin film devices has been extended to 2,200bar, while size, weight and manufacturing costs have been reduced, to meet the needs of high pressure hydraulics and diesel rail injection systems.

Perhaps the most important developments, however, have been in the electronic packages that are increasingly being supplied with pressure transducers. In particular, many of the latest devices and thin film transducers incorporate advanced ASIC (application specific integrated circuits) technology, which enables the performance and functionality of each transducer to be tuned to meet the specific requirements of a market, application or customer.

These improvements can range from total accuracy enhancements, on board diagnostics, and the ability to carry out data analysis, with digital output direct to higher level process control systems. Digital processing also enables devices to be easily networked and linked to a PLC, to form a discrete and cost effective control system for areas where the expense and complexity of higher level technology would be inappropriate.

Similarly, for devices such as Gems MMS transducers, where they are used for hydrostatic level monitoring with liquids with changing densities, the use of micro processor technology allows both temperature and pressure to be monitored and the results to be processed with a known specific gravity, to produce an extremely accurate level reading.

The use of ASIC technology, combined with improvements in volume manufacturing techniques, has in many instances reduced the unit cost of transducers by a factor of 10; for example, the performance previously associated with units that sold for £300 can now be delivered from devices costing £30 or less. As a result, the potential exists for the use of pressure transducers to become even more widespread and for the technology to play an even greater role in data acquisition and system control.

For more information on Gems Sensors please contact:

Gems Sensors and Controls
Basingstoke
Hampshire.

Can be contacted on

Tel: +44 (0) 1256 320244.
Email: sales@gems-sensors.co.uk
Web: www.gemssensors.com