One of the most challenging issues in Fire and Gas (F&G) detection is deciding: ‘How many detectors do I need, and where do I put them?’ A fair question where traditional detection codes are simply not applicable. Placing heat detectors on a grid is intuitively not adequate for external processing facilities exposed to the harsh external environments like the North Sea. How, therefore, can we adequately position flame and flammable gas detectors, such that target fire sizes and gas clouds of concern can be detected reliably?
As a result of this challenge, the Oil and Gas Industry led the way with what is widely referred today as ‘F&G Mapping’. Companies including Shell and BP formalized their practices into internal guidance documents, detailing methodologies against which F&G detection systems should be designed where national / international codes are not applicable. In 2005, Micropack (Engineering) Ltd. introduced the concept of F&G Mapping to the ISA84 committee, which later produced ‘ISA TR84.00.07 Guidance on the Effectiveness of Fire and Gas Detection’ 1. This document, although well intended, has created confusion in how to apply it, the result being that greater inconsistency of design has emerged.
A major driver behind the development of the BS60080 standard was to address design inconsistency in relation to F&G Detection devices and the suitable application of performance-based design in mapping.
The BSI standard attempts to improve upon the guidance which has gone before and to provide a platform for the performance-based approach. The standard provides guidance on how and where to apply the various methods, and places emphasis on competence when conducting such an analysis. The dangers of applying an inappropriate methodology to a certain facility (i.e. applying scenario-based modelling to a standard offshore oil and gas facility) can result in either an over-engineered detection system or, even worse, an unsafe design2.
How many detectors do we need and where do we put them?
To simplify the intention and content of the standard, BS 600803 aims to answer the following questions:
1) How many detectors do I need?
2) Where do I place them to maximize effectiveness?
Throughout the development process these two questions were at the forefront of the group decision making. These questions helped to decide what content was required, and the subsequent sections that would be included. If additional content were proposed at any point, it would be included only if it helped to answer one or both of those pivotal questions.
Such factors include the detection design methodology selection, hazard analysis, detection technology selection, management of change for the facility, practical considerations (installation / maintenance), how to determine the adequacy of the detection coverage etc. The potential of spurious trips also informs technology selection and requires consideration in the positioning of detectors.
Life cycle guidance on a flame and gas detection system is also included as it is important to emphasize that mapping and/or modelling is an ongoing activity and not simply a starting position for a new facility. Routine surveillance of detector coverage ensures that facility modification, change in hazards, etc. are all considered, and the management of change highlights any deficiencies.
F&G mapping techniques
In order to provide guidance on the specific techniques that can be applied to flame detection coverage mapping, the standard provides guidance on the considerations that should be made regarding the fire risk. These factors include the properties of the process stream and consideration of the spectral characteristic of the anticipated flame. This will influence the detection technology that can be applied and the positioning of devices to optimize detection capability to the specific portion of the flame that the devices will be most sensitive to.
Guidance is also presented on voting of devices to reduce false executive action and provision of system redundancy, along with guidance on the notion of target fire size in assessing coverage. As flame detector response is governed by the inverse square law, the typical practice of assessing risk grades based on the allowable maximum target fire is expanded in the standard. This aims to ensure an appropriate detector sensitivity is applied based on the environment to reduce false alarms, but also to ensure this is accurately accounted for within the design. The following table and figures present an example of how flame detection coverage can be modelled.
Flammable gas detection
With respect to flammable gas detection mapping, guidance begins with the overview of considerations to be made prior to the mapping process. The guidance discusses the impact of whether the risk involves a gaseous release, gases liquified by the application of pressure, gases liquified by refrigeration, liquid releases with gas accumulation potential etc.
Unlike flame detection, gas detection set points require consideration. The set points of flammable gas detectors can impact the resulting design and placement of the device as this has an impact on the sensitivity of the device.
Flammable gas detection also differs to flame detection in the sense that the devices are passive, requiring gas to come into direct contact with the device. The guidance therefore focuses on the target accumulations or releases that would aim to be detected. As it is virtually impossible to detect all gas leaks4, the risk assessment focuses on the type of accumulations and releases that could develop to an explosion hazard, for example an accumulation of flammable gas within a heavily congested and confined processing area of an offshore facility.
Despite this fundamental difference in assessment, the guidance within the standard follows a similar route in providing guidance on change management, such as the consideration that applying a dispersion based scenario gas mapping design may cause significant problems when changes are implemented during the life of the site. This guidance could therefore mitigate a potentially expensive and time-consuming process of re-design and detection layout alteration when even the most basic of changes are implemented in the process area. The following tables and figures show an example gas detection analysis.
Toxic gas detection
Toxic gas detection design presents a significant challenge as there is limited design guidance, literature and empirical data that can be used as a basis against which to design the placement and quantity of fixed toxic gas detection. While guidance and data exist which refers to the design of devices and their performance requirements, these do not answer the fundamental questions posed by BS60080 – how many do we need and where do we place them?
Considering this, the group focused on the philosophy behind why a site would require fixed toxic gas detection. From this mindset, the guidance for toxic gas detection differs slightly from flame and flammable gas detection. Much of the guidance takes the form of considerations that should, as a minimum, be taken into account to ensure personnel protection.
This guidance includes considerations ranging from the standard considerations of set points and voting, but focuses more on the relationship between response of the system, and action which can be implemented (whether this is process isolation, evacuation, stay in place etc.), as demonstrated in the Figure 7.
As part of this analysis the designer should consider the occupancy, means of escape, purpose of the area, nature of the gas (i.e. toxic concentration), inventory size, facility geometry, mitigation factors etc. Adequacy can then be analysed based upon the consideration of the specifics and the risk grades and performance targets which are set to meet the facility requirements after the risk assessment.
Assessment of adequacy
Each of these factors included in the standard are present with the aim to ensure that adequate coverage is maintained. Which introduces the complex issue of what can be deemed as ‘adequate’ coverage. Guidance is presented on this process of analysis and promotes the application of risk assessment of the escalation potential, to assign a performance target for the system, and to subsequently map the area to verify if the coverage meets this target. This can be through the application of a target percentage coverage; however, this can be a problematic form of adequacy analysis as the following figure shows.
Figure 8, as presented in the standard, shows that 50% coverage for example can result in drastically different configurations. While a flame is shown here, this guidance is applicable to flame, flammable and toxic gas detection analysis. The example on the right could not be classed as acceptable as such a large fire could exist undetected and cause a significant escalation event. The left-hand-side chess-board-type coverage, however, could be classed as acceptable dependent upon various factors including but not limited to: the specific fire which can be present; the specific detector which is applied; the specific portion of the flame exposed to the device; the elevation in which the small blockages occur in relation to where the flame can exist; the location where the blind spots are located etc. Such information then allows the designer to make an informed decision regarding whether the design is adequate or not. This helps to demonstrate that a target % coverage is rarely an adequate form of design strategy and that each individual area should be analysed within its own merit by a competent designer who understands the environment, detection and risk in place. It also aids to show why automated coverage tools that make sweeping assumptions on detection capability may be dangerous.
To conclude, the new standard, BS60080, is intended to provide guidance on the design of fixed F&G detection devices. This aims to bring the hazardous industries in alignment with respect to flame, flammable and toxic gas detection systems when considering how many devices are required and where they should be located. The standard will be available through the British Standards Institution (BSI Group) and is intended for publication Summer 2020.
For more information, go to www.micropackfireandgas.com
- ISA, 2018, TR84.00.07 Guidance on the Evaluation of Fire, Combustible Gas, and Toxic Gas System Effectiveness
- Sizeland, E., and McNay, J., 2019, Using CFD to optimise gas detection layouts: Are we barking up the wrong tree?’, FABIG June 2019
- BSI, 2020, BS 60080, Explosive and toxic atmospheres: Hazard detection mapping – Guidance on the placement of permanently installed flame and gas detection devices using software tools and other techniques
- Hilditch, R. and McNay, J., 2019, Addressing the problem of poor leak detection rates on UK offshore platforms, Proceedings of Ninth International Seminar on Fire and Explosion Hazards (ISFEH9), pp. 1198-1209