This handbook has been written as a part of Christian Michelsen Research's (CMR) research programme "Gas Safety Programme 1990-1992" (GSP90-92). The participants of the programme are: BP Norway Limited U.A., Bundesministerium für Forschung und Technologie, Conoco Norway Inc., Elf Petroleum Norge A/S, Esso Norge A/S, Gaz de France, Health and Safety Executive, Mobil Exploration Norway Inc., Norsk Hydro, Norwegian Petroleum Directorate, N.V. Nederlandse Gasunie, Phillips Petroleum Company Norway and Statoil.

The purpose of this handbook is to give a brief introduction to gas explosion safety, based on our current knowledge of the subject and on our experience in applying this knowledge to practical problems in the industry. Because of the intended brevity and simplicity of the handbook the information provided may in some cases be strongly simplified and/or incomplete. For in-depth information on the various subjects the reader is referred to the literature described in the References .

The user of this handbook is intended to be a process- , design- or structural engineer, but the handbook should also be useful for safety engineers.

How to Read this Handbook

The handbook is divided into 15 chapters. The individual chapters can be grouped in four categories: i) Introduction, ii) Background and Basics, iii) Practical Aspects and iv) Tools and Analysis. The chapters of each category are shown in Figure 1.

The chapters in the first category "Introduction" contain the description and physics of gas explosion phenomena, definitions and loss experience. Under "Background and Basics", cloud formation, gas explosions, blast waves and structural response are described. The part "Practical Aspects" relates the information presented in "Background and Basics" to different industrial situations The final part on "Tools and Analysis" focuses on what one can do to improve gas explosion safety. In particular a description is given of the FLACS and µFlacs codes, and how these tools can be applied for predicting the consequences of gas explosions in an industrial environment.

When writing this handbook, it was our intention that the reader could start directly in one chapter listed under "Practical Aspects" or "Tools and Analysis" and use the other chapters for supplementary information. However, if the field of gas explosions is new to the reader, we would recommend going through Chapters 1 and 2 as a first introduction to the subject. Chapter 15 contains a list of terms and expressions.

To avoid too much cross-referencing and thereby making the use of the handbook more cumbersome, there is a certain amount of overlap between what is contained in the two first parts and the two last parts. This makes each of the two halves more self-contained and the handbook may therefore be used as a reference work without having to read it all.

Objectives of the Handbook

Today there is a lot of information available in scientific papers and reports on gas explosions. However, in most cases the practical implications of this information are very hard to extract. A need for a handbook with simpler presentation of the available information that can be used in the industry, has therefore been identified.

This handbook summarises the main results and experience from our previous research programmes and consultancy activity on gas explosion safety (Bakke et al., 1991). We are focusing on pressure build-up during gas explosions. Important areas of gas explosion safety, such as how to prevent leaks and what is the ignition probability, are not covered. In this handbook we assume that the premixed combustible gas has been generated and ignited. Phenomena of flame propagation and pressure build-up are discussed. The important factors influencing pressure build-up are pointed out and some simple guidelines are presented. The use of numerical codes (FLACS and µFlacs) for simulation of gas explosions in industrial environments is also covered.

The handbook is one of three tools for analysis of gas explosions, which have been provided by CMR. The two other tools are FLACS and µFlacs. FLACS is the most advanced code of the two. FLACS is used for detailed analysis, while µFlacs is a PC screening tool and does not require the same amount of detailed input and resources as FLACS. Our goal is that the handbook, FLACS and µFlacs are used together in gas explosion analyses as supplementary tools. Which tool to use will depend on the stage of the analysis and the detail of information required.

Our intention is that the CMR gas explosion handbook is a "live" document that will be updated when new information is available. Further comments and suggestions for improvement of the handbook will be gratefully received. A comment form is included in the handbook.

Gas Explosion Activities at CMR

In the late 60's and in the 70's, large oil and gas fields were discovered in the North Sea. It was recognised that gas explosions might constitute a hazard for the drilling and production installations in the North Sea and that the knowledge about gas explosions in industrial environments was limited. At Chr. Michelsen Institute's Dept. of Science and Technology (CMI-DST, from June 1992 Christian Michelsen Research - CMR), gas explosion research was started in the late 1970's as part of the programme "Sikkerhet På Sokkelen". Since then gas explosion research has been an important activity at CMI/CMR, as shown in Figure 3. The gas explosion research was a continuation of the research work on dust explosions. The dust explosion research work is described by Eckhoff (1991).

In the period 1980-1990 CMI carried out two major research programmes on gas explosions. A total of 80 man-years of research have provided new insight into gas dispersion and gas explosions in industrial environments. The primary objectives of the work have been to generate know-how and tools for minimising the effect of accidental explosions. CMI's strategy has been to combine large- and small-scale experimental work with the development of advanced fluid-dynamic codes.

In GEP 80-86 pressure development due to flame acceleration by obstacle-generated turbulence was studied. This mechanism was identified as being mainly responsible for explosions occurring in complex geometries typically found on offshore platforms. The phenomenon was studied both in small and large scale (0.2 m³ and 50 m³) for geometrical layouts of increasing complexity. The most complex layout was a scale 1:5 offshore module. Different gases and ignition strengths were studied, as were ignition positions and vent arrangements.

Results from these experiments served two purposes: they provided new insight about flame acceleration in complex geometries and also data for validation of computer codes developed for explosion overpressure prediction. These codes, which were named FLACS (Flame Acceleration Simulator), were developed in the modelling activity which concentrated on modelling of compressible, turbulent reactive flows and on numerical solution of the resulting set of partial differential equations. Some results from the research have been published in the open literature (Bakke et.al., 1986, 1987; Hjertager et.al., 1988a, 1988b).

Following the seven-year programme which ended in 1986, a three-year programme focusing on gas dispersion in complex geometries and on explosions in onshore plants was initiated in 1987. The objectives of the programme were to provide more knowledge by performing experiments and to apply this knowledge to evaluate an enhanced FLACS code, taking into account gas dispersion and explosions for different fuels mixing with air.

GexCon, CMR's gas explosion consultancy, was established in 1987. GexCon is an independent unit that operates in close Cupertino with CMR's Gas Explosion and Process Safety section. Through GexCon CMR's research has already had an impact on engineering practice. CMR personnel are used actively by the industry as consultants on gas explosion safety. Our experimental facilities are used to test the integrity and functionality of equipment and structures exposed to gas explosions of predetermined strength. The FLACS code is used extensively to provide quantitative information about pressure loads from accidental explosions, both offshore and onshore. Since 1989 more than 40 projects have been performed for specific plants and installations. This work includes safety analyses, explosion simulations using FLACS, experimental studies and gas explosion courses including demonstrations of actual explosions.

In 1990 a new large multi-sponsor programme was started. The objective of this programme is to improve gas safety. This may be achieved by providing knowledge, predictive techniques and testing procedures / facilities, and by transferring results to the industry in such a way that everyday working procedures, rules and regulations as regards both design and operations may take proper account of state-of-the-art knowledge. Many of the current R&D subjects are general in the sense that they are of importance both off-shore (exploration / production / storage / transport) and on-shore (transport / storage / processing / utilisation). These are described briefly below.

Experimental Test Programme

The objective of this part is to provide knowledge which is directly applicable to engineering work, to provide data for validation of simulation codes, and also to provide tests of equipment and explosion safety concepts. For instance work is done on developing:

• knowledge on how real process streams (fuel mixtures) explode
• guide-lines for use of water deluge systems for explosion protection
• knowledge on how to ensure integrity of vital equipment by assessing loading on structural parts and equipment
• knowledge about the effect of small, complex process equipment on explosions
• data on how mist explosions compare with gas explosions in large-scale geometries
  

Enhancement of FLACS

Emphasis is put on improving FLACS interfaces (both to users and to other software) to facilitate its use, and on improving the predictive capability of FLACS. The objective of this part of the programme is hence:

• to provide a comprehensive simulation package for integrated safety analysis and design, using a common framework and user interface. The simulation package will be an extension and improvement of CASD/FLACS
• to increase the accuracy and reliability of the code by improving physical submodels and numerical solution schemes
  

Applied Safety Technology

Many smaller industrial companies do not have access to the often large amounts of knowledge and expertise in specialised areas which exist in safety research groups today. Furthermore, technology transfer is frequently considered as the final activity of research projects. In this programme, technology transfer to the industry is a major continuous effort from the very beginning and constitutes one of the three main parts of the programme.

During the last decade research has focused on developing knowledge which is now at a stage where guidelines and practical results can be formulated. Large amounts of data that can be analysed and systemised now exist. Areas where CMR at present is making an effort along these lines, include:

• gas explosion handbook
• PC tool for gas explosion analysis (µFlacs)
• safety walls designed with regard to working environment, gas dispersion (natural ventilation) and gas explosion safety (these three aspects may not be compatible!)
• water deluge systems for explosion mitigation.