Considerations when selecting temperature sensors

Driven by the opposing needs of meeting the profit expectations of the shareholders and the seemingly continuing demands of the various regulatory bodies.

Recent years have seen many of the smaller organisations swallowed up by their larger competitors followed by many of the major multinational organisations combining in an attempt to both reduce overhead costs and increase their ability to compete in the market place.

Whether these efforts have improved their market positioning or led to increased cost benefits elsewhere appears to be open to question, judging from the various company reports of recent years.

Where then, can costs be controlled or improved without further significant expenditure? Within the industry, the elimination of micro-organisms to the point at which they are no longer detectable in standard media culture is the basic and essential process of sterilisation.

This process plays a fundamentally essential part during research and manufacture, with the autoclaves employed, currently costing anything from £30,000 to £500,000 depending on their duty.

The items requiring sterilisation, particularly parenteral and other medical products in a large autoclave, can also be valued at many thousands of pounds, all amounting to a considerable investment by the pharmaceutical manufacturer.

With such large investments made in the main process equipment and in production materials it is of the utmost importance that the autoclave system components offer a high reliability.

There are various forms of sterilisation, but perhaps the most prominent of these is steam or moist heat sterilisation and for those products in liquid form and held in glass/plastic containers, the water cycle system.

In both of these processes, time, pressure and temperature are the three main parameters that contribute to a successful sterilisation cycle.

Of these three, reliable temperature measurement is the parameter most generally regarded as the one that causes most problems and frustrations to all those concerned with the efficient operation of an autoclave.

So why should this be? In many cases it is down to the fact that many manufacturers of temperature sensors are not fully conversant with autoclave practise or the users requirements.

It is also a fact that a large number of autoclave manufacturers are prepared to employ assemblies that fall below the minimum requirement.

Consider the types of temperature sensors that are found in most modern autoclaves.

In general they will fall into two major types, the thermocouple and the resistance thermometer.

Both have been around in various forms since the early and late 1800s, but the advances made during the 20th century have enabled them to be considered for almost all applications involving the measurement of temperature.

The thermocouple type now recognised internationally as the premier choice for autoclave use is the type 'T' thermocouple which consists of two conductors of dissimilar material, the positive leg in copper and the negative leg in a proprietary metal known as Constantan.

This proprietary material being an alloy of 55% copper and 45% nickel.

Type 'T' material is available in three grades, but only Class 1 material should be considered as acceptable for autoclave use due to its higher tolerance.

Thermocouples offer a simple, but effective method of temperature measurement in an autoclave with the sensing element or hot junction generally providing a more robust sensor than the resistance thermometer.

However it has a shorter, long term stability characteristic and is less accurate than the resistance thermometer.

The principle of operation of a thermocouple is simply that when their opposing junctions (hot junction as the measuring junction and the cold junction as the reference junction at the instrument terminals) are at different temperatures, then an EMF or electromotive force will be generated.

Although this generated voltage is only in microvolts it follows a known, non-linear curve which meets internationally accepted calibration standards.

This low signal is then amplified by the receiving instrumentation to enable adequate control, recording and data logging to be carried out.

However it is important that the thermocouple wires running between the point of measurement and the instrumentation are not located close to power lines or equipment operating high voltages, otherwise the signal may give rise to errors induced by these higher voltages.

Although there are still many machines using thermocouples, the majority of modern autoclaves are fitted with resistance thermometers.

The modern resistance thermometer consists of a length of platinum wire, wound into a coil and housed within a ceramic former, such that at 0C it will measure 100ohms across the coil and is designated as a Pt100 resistance thermometer.

The Pt100 detector has greater accuracy over a similar temperature range to the type 'T' thermocouple, with excellent long term stability, but it is more susceptible to physical shock.

When continually subjected to this kind of treatment, the device will both decline in accuracy and increase in drift.

Recent years however have seen the development of the Pt100 thin film detector, which has far greater resilience to physical ashock; although without quite matching the long term stability of the conventional wound type.

Thin film Pt100.

Several tolerance grades are available with Pt100 detectors, the two most frequently used in industrial applications are Class 'B' and 'A', followed by the 'fractional DIN' classes.

Class 'A' should always be the minimum grade of tolerance for autoclave applications, the latter range of 'fractional DIN' classes only being available in the Pt100, conventional wound format.

The term 'fractional DIN' arose as detector manufacturing techniques improved, allowing detector manufacturers to offer commercially available detectors with higher tolerances.

Since these could be of a tolerance several factors better than current international standards, they became known as 'fractional DIN'.

These higher tolerance detectors are not specifically produced to meet this higher tolerance, but are the result of selection within a standard manufactured batch.

Typical drain and air detector probes.

Having decided upon the type and grade of sensor to be used, we need to give consideration to the intended application.

The autoclave chamber or load probe is the sensor that invariably causes the most problems, mainly because of the need for the operator to continually handle the unit.

The other fixed sensors such as the air detector, if fitted and the drain probe will usually provide excellent service with minimum drift over long periods and for many cycles.

Autoclave chamber sensors usually fail due to signal errors caused by the chamber atmosphere penetrating the assembly, calibration errors due to physical shock or long term use, or detector breakage from mishandling.

Because it is necessary for the operator to handle the chamber/load probe at the start and conclusion of each sterilisation cycle, it is essential that at least a section of this probe is of a flexible construction.

This in itself can present a major obstacle, because to provide this flexibility, a seal of high integrity and resilience is necessary at the transition point of the rigid and flexible sections of the probe.

Unfortunately though, the environmental conditions within an autoclave chamber are harsh, extremely demanding and conspire against most conventional sealing arrangements.

Temperature cycling within an autoclave chamber will cause the various materials of construction in the probe assembly to expand and contract at differing rates and amounts.

The stresses and strains that this imposes on the chosen type of sealing arrangement will soon be evident if an inadequate seal has been used.

The slightest weakness in a seal, will soon be exploited by the chamber atmosphere and it's penetration by the atmosphere will follow.

The ensuing ingress of moisture or steam will contaminate a Pt100 detector and unsatisfactory signals will then be derived from the detector or the thermocouple.

This results in erroneous data being received by the instrumentation and the possibility, ultimately, of a rejected product load.

The design of the chamber/load probe assembly must also take into account the opposing ideals of ruggedness, to resist the attentions of the operator, whilst being sensitive enough to respond to temperature variations in the chamber within an acceptable time period.

By careful selection of the appropriate materials and then correctly engineered and applied a chamber/load probe should provide an economical and acceptable service life.

However, the autoclave chamber/load temperature sensor has long been treated as the 'Cinderella' of autoclave components.

Partly due unfortunately, to their historically low reliability record.

The attitude of many purchasing managers, influenced by this poor record perhaps, is to drive down the cost of these sensors, yet still with the expectation that they will provide a long and reliable service life.

The autoclave sensor, once fitted, is then generally subjected to either mistreatment or at best, casual indifference by the various autoclave operators.

The consequence of this imbalance of requirements is that most suppliers of these devices endeavour to provide, out of necessity, a product that satisfies the purchasing manager only.

When finally put to use, the product is then often found to display questionable reliability and generally falls well below an acceptable confidence level.

The eventual cost of ownership in this kind of situation can then become very high.

The cost of lost production through down time, caused by unreliable temperature measurement on a production autoclave, frequently contributes to overhead costs of no less than £1000 per hour.

In addition to this is the costly probability of having to reject a spoilt product load.

Frustration is caused to the production department, often followed by a degree of animosity between them and the engineering department for providing unsuitable equipment.

This general attitude regarding autoclave temperature sensors, in many cases stems from the autoclave manufacturer's sensor choice being based mainly on cost, thus often resulting in the inclusion of inappropriate sensor assemblies.

This decision is frequently driven by the need to maintain a competitive edge, coupled with the client's main interest being centred primarily on the control specification and the other more obvious aspects of the autoclave.

It is not unusual for the client then to continue purchasing identical replacements, due to a lack of market knowledge or an inability to convince the sensor manufacturer of the necessary application requirements.

It is in the interest of both the autoclave manufacturer and the end user that appropriate temperature sensors designed for the application, should be specified in order to ensure the confidence of reliability, so necessary in today's pharmaceutical industry.