Top tips - detecting gas leaks in industrial environments
Published: 01 July, 2008
There are many factors which should be taken into consideration when determining the most suitable location for sensors, and site location is one of them. Indoor and outdoor site monitoring present different problems due to mechanical ventilation, building structure or the wind direction/speed.
For an objective and precise sensor location, a fundamental role is played by the measurements of air movement through a building and the identification of components that are potentially vulnerable to leakage (eg a break in the pipework, valves, flanges, pressure regulators, sight glasses, pump and compressor).
Another factor to take into consideration when deciding where to position the sensors is the type of gas that has to be detected.
Usually, for gases with a density lower than air (hydrogen, methane, etc), the sensors are uniformly distributed at about 30cm from the highest point of the ceiling, because these gases are easily defused in the air.
Useful also is to avoid places where there are air currents, or where air is likely to be very still due to irregularities in the ceiling (eg beams etc), which could stop the gas moving freely around.
For gases with a higher density than air, sensors should be placed near the floor and in proximity of possible leak points. Certain points such airspaces, junction boxes, manholes, and weigh-bridges should always be controlled by a sensor.
A general rule of thumb regarding gases of a similar weight to air – or for toxic substances in low concentration – is to distribute the sensors at different levels, thereby increasing the chances of detecting an eventual leak.
Up to now, a catalytic gas detector (aka“pellistor”) has been used for flammable gas detection. It is a technology based on catalytic oxidation and although it has a good relationship price/performances, it does require regular maintenance and exposure of the sensors to standard test gas mixtures at regular intervals in order to confirm its correct efficiency.
Some substances present in the atmosphere that is to be analysed can considerably alter the response of a sensor, and these – in the main – are chemical poisons. The most important poisons are:
l halides (compounds containing fluorine, chlorine, bromine and iodine).
l glycol sulphur (compounds which polymerise on the bead)
l heavy metals (tetraethyl lead).
Modern catalytic sensors do show a far greater resistance to the effects of these gases, nevertheless in all cases if the presence of these substances is expected on site, it is advisable to frequently verify the detectors’ sensitivity using calibration gas.
An alternative to a pellistor is the infrared flammable gas detector, which measures the amount of radiation that is absorbed by vibration of the C-H bond in the 3.4 _m region, where the position of maximum absorption depends on the particular hydrocarbons.
While the initial investment demanded by this technology is greater, sensors based on infrared absorption offer a potentially longer life, a faster response time, smaller costs of maintenance, and they can be made to be “fail safe”.
In the detection of gas or vapour clouds the IR (infrared) Open Path Detector has an important role.
Here the IR Source is separated – but aligned – with the IR Detector, and a beam of infrared radiation crosses the entire hazardous area to be protected. This method offers an alternative to the usage of large numbers of point detectors.
Usually the detection range for a flammable gas is between 0 and the Lower Explosion Limit (LEL). 0 divided by 5 per cent by volume (in the case of methane but may change in function of the different substances).
The explosive atmosphere is identified within the LEL and the UEL, the Upper Explosion Limit. Outside this range the mixture air and gas is not a potentially explosive atmosphere.
For toxic gases such as CO, H2S, SO2, NO, NO2, Cl2, NH3, HCN, HCl, PH3, SiH4/GeH4, the electrochemical cell is the most commonly-used technology.
An electrochemical cell is a type of gas-battery developed from fuel cell research. These sensors are usable in temperate climates, but their main weakness is their failure at high temperature and with an upper temperature limit of about 40°C.
Semiconductor gas sensors (metal oxide semiconductor (MOS) -based) is a technology in continuous evolution.
The use of these sensors has always been held back due to the limits of the technology (precision, linearity, selectivity), but thanks to the parallel evolution of microprocessors and microelectronics in general, and to the simplicity of integration of these technologies, we can say today that these sensors present a valid alternative to the electrochemical cell. What is more, they are also an option in hot climates – or even extreme environmental conditions – where electrochemical cells are inappropriate or where the application requires long life and reliable sensors, and there is no need for precision.
Many gases may be detected with this technology, eg ammonia, hydrocarbons, carbon monoxide, alcohol and hydrogen sulphide. In general, however, MOS detectors are sensitive to a number of gases, which means that is important to know all the possible gases that could be present in the environment, and take their effect into consideration.
For toxic gases, an indication of the detection range is taken from the occupational exposure standard limits, the long-term exposure limit TWA (8h time-weighted average), and the short-term exposure limit STEL (10 or 15 minutes reference period according to local regulations).
Apart from the different technologies used in the construction of the sensors, a great difference between sensors for flammable gases and toxic gases is also their sensibility. In the first case the range is expressed in percentage terms and in the second case in ppm, and we have to remember that 1 per cent equals 10,000 ppm.
The choice of the most appropriate technology for the detection of gas leaks is not the only aspect to be considered. The project should also include an adequate and easy-to-use signal acquisition system that can promptly inform personnel of the type of event and the degree of danger.
Communication of the gas concentration is typically achieved via a 4-20 mA current loop between a control card and the sensors. The control card must show if the apparatus is energised; have a fault signal in the event of sensor’s failure; and show loss of continuity in one or more of the wires to the sensor, or open or short circuit in the connection to the sensor.
Alarm output contacts or alarm signals are of a “latching type” requiring a manual acknowledgement and a reset.
The choice of a good and flexible modular system allows different hardware configurations, simplex or redundant, with fail safe and fault-tolerant characteristics.
Another important feature of the control unit is the capability to communicate via serial line with other systems such as a PC or DCS with standard communication protocols. With modern information systems, planned maintenance can be easily performed to ensure that the gas detection system is maintained in full working order.
It is also possible to use management software to support the user during all operating and diagnostic phases, including programs for the test of each part of the plant including sensors, electrical wiring, actuators and alarm indicators.
This continuous monitoring of the system allows the automatic creation of a data log of events, which can be a powerful tool in understanding the behaviour of the plant and its workers during the different parts of the process.
All this information is essential to improve the safety of the plant, allowing both a rapid recognition of an abnormal event, as well as prompt remedial action to prevent the development of gas or vapour clouds - which provoke a catastrophe.
Moreover the analysis of this data allows the identification of both critical phases of the process as well as false alarms events.
It also provides excellent feedback regarding corrective actions applicable to the process for the training of workers and for the improvement of maintenance operations.
In conclusion
The choice of the best sensor technology for detection of toxic or flammable gases is the first aspect to consider but it is not the only one. Very important is also the choice of control panel because that will guarantee the possibility of future upgrades and the capability to exchange information with other systems.
Technology allows ample choice among various gas detection tools, but the identification of the best solution for every specific case must be found during the project phase. Here, the correct compromise must be found between cost and benefit, without forgetting that the goal for the system must is safety and high reliability.
