Upgrading from Pellistor Gas Sensors to Infrared Technology

In this article Andy Avenell, Crowcon’s Fixed Systems Product Manager, considers the shift towards infrared (IR) gas sensor technology in the oil and gas industry.

Abstract

Flammable gases and vapours can present considerable dangers in many industrial applications, none more so than the processes involved in extracting, transporting and processing oil and gas. Fast and reliable detection of gas accumulations in sub-LEL (Lower Explosive Limit) levels is essential in order to prevent potential explosions. The longest established and most prevalent sensor technology employed for detecting flammable gas is the Pellistor (or Catalytic Bead as it is otherwise known).

Pellistor-based detectors are widely available at low cost, but are vulnerable to permanent poisoning by contaminants, do not fail safe, require very regular maintenance and calibration and have a limited life-span. Infrared (IR) gas detectors overcome all of the limitations associated with pellistors to provide fast and reliable detection of hydrocarbon gases. IR detectors provide rapid gas detection, fail-safe operation, and in some cases compliance with the IEC61508 safety standard. IR detector prices have fallen significantly in recent years, and although the price per point is still higher than pellistor based detectors, industry experience confirms that IR detectors quickly pay for themselves by reducing operational/maintenance costs.

Replacing pellistor-based detectors usually means that the control equipment will also need upgrading. This is due to the mV bridge type signal interface between detector and controller. IR detectors typically require a power supply of between 12 and 30 Volts dc, and provide a 4-20mA signal to the controller. Thus a pellistor control card is inherently incompatible with conventional IR gas detectors. ‘Pellistor Replacement’ IR gas detectors simulate a pellistor mV type signal and are specifically designed to replace pellistor sensors with IR technology, whilst retaining the original control equipment, cabling and detector junction box. Thus the very significant cost associated with upgrading controllers and re-installation is avoided.

One of the more difficult problems to overcome with pellistor IR replacement technology is signal drift. Systems operating using a mV (Wheatstone) bridge, such as pellistor based systems, are very vulnerable to this problem. In order for a control card to show a zero gas reading the four resistances in the bridge must be balanced. Resistance imbalances can however be introduced by poor cable connections either due to loose terminals, temperature effects or through oxidation of conductors. Any cable/terminal resistance changes are manifested as zero drift on the control card. The challenge is that to be a genuine pellistor replacement, an IR detector must have very low power consumption (ie less than 1 Watt), and a well designed power supply so that potential cable influences on the signal are negated.

Pellistor replacement IR detectors should be rigorously designed and tested to ensure reliable operation at all times. High integrity products are designed to comply with the demanding requirements of IEC61508 (ideally to SIL 2) in terms of both hardware and software. Gas detection performance (response time, short and long-term stability, linearity, accuracy, temperature stability etc) should be verified to a performance standard such as EN61779:2000.

Finally, independent verification of the product by a recognized testing body should have been conducted. The product should be performance tested in simulated offshore conditions to ensure reliable performance in even the most extreme conditions.

Introduction

Flammable gases and vapours can present considerable dangers in many industrial applications, none more so than the processes involved in extracting, transporting and processing oil and gas. Fast and reliable detection of gas accumulations in sub-LEL (Lower Explosive Limit) levels is essential in order to prevent potential explosions. The longest established and most prevalent sensor technology employed for detecting flammable gas is the Pellistor (or Catalytic Bead as it is otherwise known).

Pellistor technology

Pellistors were originally developed for the mining industry during the early 1960’s; earlier simple platinum coil sensors were unsuitable for continuous operation due to their high power consumption and excessive zero drift.

Pellistor detectors consist of two matched platinum coils, each embedded in a bead of alumina. The detecting element is coated with a catalyst which promotes oxidation when in contact flammable gases; the compensating element is treated so that catalytic oxidation does not occur. The compensating element is fitted to ensure that signals are not generated due to environmental effects (eg changes in ambient temperature or gas flow rate).

Pellistor-based systems operate in a Wheatstone Bridge circuit whereby the pellistor and compensator in the detector represent one half of the bridge, the other half being fitted to the control card usually located in the control room. The control card supplies a voltage to the bridge (typically 2V to 3.5Vdc), which generates a current flow and raises the temperature of the beads to a level where oxidation of gases will occur (>300°C). The control card measures a small voltage offset in the bridge due to the increased resistance in the pellistor element when gas is present. This voltage is then amplified and used to display the gas level and activate alarms.

Because pellistors are relatively high power devices, and as they operate at a temperature that will ignite flammable gases they need to be sealed behind a flame arrestor (sinter).

Pellistors are typically fitted within a stainless steel housing, mounted on an Exd (Flameproof) or Exe (Increased Safety) certified junction box. The detector is connected to the control equipment via a 3-core (or sometimes 4-core) cable.

Advantages and Disadvantages of Pellistor Technology

Advantages:

•    Low cost technology; pellistor-based detectors are widely available at low cost.
•    Pellistors will detect a wide range of gases and vapours. Correction factors can be applied so that the sensor can be scaled for a particular substance.
•    Pellistors are very simple devices; apart from calibration gas, no special equipment is required for commissioning or maintenance.

Disadvantages:

•    Pellistors are vulnerable to permanent poisoning by silicones, lead, sulphurs or chlorinated compounds. If exposed to these compounds, a pellistor may fail to respond to flammable gas.
•    Pellistors must be operated behind a sinter (flame arrestor) which may become blocked, thus preventing gas from reaching the sensor.
•    Pellistors do not fail-safe; poisoned pellistors remain electrically operational; thus the control system will continue to display zero gas when flammable gas may be present.
•    Sensitivity to flammable gas is reduced in the presence of some compounds (notably hydrogen sulphide and halogens).
•    Pellistors need at minimum of 12% volume oxygen present to operate. Their efficiency reduces in oxygen deficient atmospheres.
•    Pellistors may burn-out and require replacement if exposed to gas concentrations greater than 110% LEL.
•    Pellistor sensitivity degrades over time.
•    Pellistors have a limited life-span, sensors typically last 3-5 years.
•    Pellistors require regular gas testing to ensure they are operational, and regular calibration of offset signal loss due to poisoning or bead contamination.

Typical Pellistor Gas Detector

Infrared technology

Gases which contain more than one type of atom absorb infrared (IR) radiation. Therefore hydrocarbons and other gases such as carbon dioxide and carbon monoxide can be detected by this means, but gases such as oxygen, hydrogen, helium and chlorine cannot.

Specific gases are detected by measuring their absorption at particular frequencies of infrared light which correspond to the resonance of the molecular bonding between dissimilar atoms. For example the wavelength at which the carbon atom and each of the four hydrogen atoms resonate in a methane molecule is 3.3µ (microns). Most commonly encountered hydrocarbons absorb IR energy in the range 3.3µ to 3.4µ. IR gas detectors are therefore filtered to respond to IR absorption in this range. Carbon dioxide absorbs IR energy in the 4.2µ range and thus different filters are required for a CO2 detector.

The output from the IR sensor is non-linear, and will vary with ambient temperature (due to thermal expansion effects on optical components). Therefore IR detectors use sophisticated software algorithms to ‘linearise’ the output signal to correspond to 0-100% LEL for the target gas, and also compensate for temperature shifts. The sensor will respond differently to each gas or vapour, and therefore a unique ‘linearisation’ algorithm must be developed for each target gas. Depending on sensor quality and/or production repeatability, individual sensors may need linearising.

In order to differentiate between IR absorption due to gas and other substances such as dust, dirt or water, an additional sensor with a bandwidth of (typically) 2.7-3.0µ is employed: hydrocarbon gases do not absorb IR energy at this wavelength. This prevents false alarms occurring and compensates for a reduction in signal from the interfering substance. This ‘Dual Beam’ design is also used to provide a fault alarm to notify operators of contamination of optical components.

In a typical fixed point detector, the IR source(s) and receiver(s) are mounted in the main body of the housing, with the light beam being reflected by a mirror at the far end of the housing. Parts of the light beam are exposed to atmosphere so that, using natural or forced diffusion, gas can intersect the beam. As the gas concentration increases more infrared energy is absorbed by the gas and less reaches the sensors. Using this method, received energy is inversely proportional to gas concentration.

IR Gas Detector

Advantages and Disadvantages of IR Technology

Advantages:

•    Very fast response: T90 response typically less than 7 seconds.
•    Fail-safe operation: no un-revealed failures (power faults, signal faults, software errors are always reported to the control system).
•    Immune to signal inhibition by contaminant gases.
•    No consumable parts; life-span typically > 10 years.
•    Reduced maintenance costs.
•    Does not require oxygen to be present.
•    Will not burn-out in high gas concentrations.
•    Premium models do not utilise a sinter (flame arrestor), and thus associated blockages cannot occur.

Disadvantages:

•    Purchase price is higher than pellistor based detectors.
•    IR detectors cannot detect hydrogen.
•    IR detectors cannot provide a linear response to a group of different gases: the detector is ‘linearised’ for a particular gas, and will respond to others but in a non-linear fashion.

The Move Towards IR Technology

IR gas detectors (in combination with other detector technologies such as Open-Path Infrared and Acoustic sensors) are now the accepted technology for protecting oil and gas installations against flammable gas hazards.

Independent reference to the market shift from pellistors to IR gas detectors is made in the 2006 Frost and Sullivan report F868-321.

It is now common practise to validate safety systems in accordance with IEC615082 (“Functional Safety of electrical/electronic/programmable electronic safety-related systems”); it is however difficult to achieve a satisfactory SIL rating (Safety Integrity Level) using pellistor based gas detectors. This is due to the significant possibility of un-revealed failures of sensors due to poisoning or sinters becoming blocked (the sensor is electrically operational, but will fail to respond to gas).

IR detectors also provide significant maintenance cost reductions: pellistors require very regular testing (by application of test gas). Some offshore platforms test sensors as often as every six weeks. Many platforms have 400+ gas detectors fitted, and thus such a regular test regime, combined with the need to replace sensors every 3-5 years represents a huge cost. Sinter-free IR detectors are self-checking (ie lamps, sensors, windows, mirrors, software) and thus the risk of an un-revealed failure is minimal. This combined with very low levels of zero and sensitivity drift means that calibration/testing routines can be extended to six or even twelve months on IR detectors. Routine maintenance is usually restricted to cleaning optical components, and a test with calibration gas. IR sensors typically last in excess of 10 years and thus parts replacement is usually restricted to consumables such as filters that may be needed for very dusty environments.

IR detector prices have fallen significantly in recent years, and although the price per point is still higher than pellistor based detectors, industry experience confirms that IR detectors quickly pay for themselves by reducing operational/maintenance costs.

Upgrading From Pellistors to IR

Replacing pellistor-based detectors usually means that the control equipment will also need upgrading. This is due to the mV bridge type signal interface between detector and controller (refer to the Pellistor Technology section for details): IR detectors typically require a power supply of between 12 and 30 Volts dc, and provide a 4-20mA signal to the controller. Thus a pellistor control card is inherently incompatible with conventional IR gas detectors.

‘Pellistor Replacement’ IR gas detectors simulate a pellistor mV type signal and are specifically designed to replace pellistor sensors with IR technology, whilst retaining the original control equipment, cabling and detector junction box. Thus the very significant cost associated with upgrading controller and re-installation is avoided.

Pellistor Replacement IR detectors are typically fitted with a mounting spigot compatible with the most commonly used junction boxes (eg M20 thread). This enables the unit to be directly screwed into the original detector junction box. Pellistor Replacement IR detectors are certified for use in hazardous areas (usually to ATEX3, IECEx4 and/or UL5, standards), and thus can be installed in any area for which the original detector would have been certified. The detector wires are simply connected to the original terminals (and thus the original cable). Sophisticated Pellistor Replacement IR detectors operate from the voltage source from the original control card, and are zeroed and calibrated in exactly the same way as the original pellistor: no adjustments are necessary at the detector.

Operation and Maintenance

One of the more difficult problems to over-come with pellistor IR replacement technology is signal drift. Systems operating using a mV (Wheatstone) bridge, such as pellistor based systems, are very vulnerable to this problem. In order for a control card to show a zero gas reading the four resistances in the bridge must be balanced. Resistance imbalances can however be introduced by poor cable connections either due to loose terminals, temperature effects or through oxidation of conductors. Any cable/terminal resistance changes are manifested as zero drift on the control card. The challenge is that to be a genuine pellistor replacement, an IR detector must have very low power consumption (ie less than 1 Watt), and a well designed power supply so that potential cable influences on the signal are negated.

Flammable gas detectors are often installed in areas that may be difficult to access. To enable testing and calibration detectors may be fitted with a pipe connector. A flexible pipe can then be fixed to the connector and run to a more accessible point. Calibration gas can then be applied to the pipe, and the performance of the detector can be verified without needing to access the detector directly. More advanced IR detectors may actually be calibrated via this means, as oppose to the traditional method of temporarily replacing the detectors’ weather-cap with a calibration cap.

Routine maintenance should be restricted to gas testing (with re-calibration only as required: typically annually at most) and cleaning of optical components (algorithms are utilised to provide a fault signal if the window or mirror are more then 75% obscured by contaminants).

Some pellistor replacement IR detectors utilise sinters to achieve Exd Flameproof compliance. Sinters slow the response time of the IR detector significantly (T90 response time may actually be longer than achieved by pellistors), and are vulnerable to blocking. Sinters blocked by contaminants will prevent gas reaching the sensor; this represents a potentially dangerous ‘un-revealed failure’ which necessitates regular testing to avoid.

Pellistor Replacement IR detectors need to operate continuously in very harsh environments. 316 stainless steel construction and an effective weather-cap are essential to protect the optical components.

Pellistor Replacement IR Gas Detector

Performance and Testing

Pellistor replacement IR detectors should be rigorously designed and tested to ensure reliable operation at all times. High integrity products are designed to comply with the demanding requirements of IEC615082 (ideally to SIL 2) in terms of both hardware and software. Gas detection performance (response time, short and long-term stability, linearity, accuracy, temperature stability etc) should be verified to a performance standard such as EN61779:20006.

Finally, independent verification of the product by a recognized testing body should have been conducted. The product should be performance tested in simulated offshore conditions to ensure reliable performance in even the most extreme conditions.

References

1. Frost and Sullivan Report F868-32, 2006: World Industrial Gas Sensors Detectors and Analyzers markets. Relevant references are made on pages 3-2 and 3-42.

2. International Electrotechnical Commission (IEC), IEC61508 Functional Safety of electrical/electronic/programmable electronic safety-related systems

3. ATEX: European Directive defining standards for equipment for use in potentially explosive atmospheres.

4. IECEx: is an international certification scheme created by the IEC to facilitate international trade in electrical equipment intended for use in explosive atmospheres.

5. UL: Underwriters Laboratories Inc is a privately owned company that tests to make sure that products meet safety standards.

6. EN61779:2000: Electrical apparatus for the detection and measurement of flammable gases. General requirements and test methods.

Filed under Uncategorized. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.

Name (required)

Email (required - will not be published)

Website