Critical Analysis of the Use of Standard Fire Curves for Determining Fire Resistance - Term Paper Example

Critically Analyse The Use of Standard Fire Curves For Determining Fire Resistance. How And Why Does The Approach Vary For Offshore Applications. Fire resistance as a phenomenon generally refers to the ability of structural members to withstand conditions of heat and combustion in circumstances or incidents of fire outbreak. Typically, the component parts of buildings or structures are the primary objects of assessment in such conditions, within the context of tests on them to determine their fire resistance rating, where design fires are used to simulate real fire. Standard fire curves are used to generate temperature-gradient properties of the members under analysis, with a view to predicting the limiting temperature of members in structures gutted by fire- as a means of determining levels of fire resistance. Because the fire curves are modelled on the basis of design fires, the analysis becomes vulnerable to the difference between design fire parameters and the real fire situations that may occur. Furthermore, varying conditions and the global proliferation of standards for fire tests make it imperative that an approximate approach to different situations/standards, for structures (at sea and on land) be adopted. This work aims to analyse fire curves in terms of their vital components/features, simultaneously considering the critical levels at which these vital variables of the standard curves experience a change, with the associated consequences of such, regarding the effectiveness of fire curves at carrying out their function of fire resistance determination. A further look will be taken at the applicability of these curves at sea, with analysis involving a look at the means and extent to which these applications vary, in the approach associated with them. The level of fire resistance in structures (onshore) is generally assessed or analysed by the use of engineering fire design curves. Different types of fire curves and standards are adopted in various parts of the world. Different assumptions are also associated with these different curves. Technically speaking, fire curves are applied to determine the fire resistance rating of passive fire protection systems. This rating is a measure of the time over which a passive fire protection system can resist or withstand a standard fire protection test. Some of the most commonly used standard fire curves include the ISO-834 (BS 476 or DIN 4102), which are relevant to cellulosic structures. There are also the ASTM E119 and UL 1709(or Eurocode 1), applicable to hydrocarbon situations. Engineering design practice generally uses these curves in the context of design fire temperatures, which attempt to simulate real fire scenarios. But this is an exercise which requires as widely applicable and accurate fire model as is practicable (Barnett and Clifton 2002). The difficulty of this arises from the fact that there are diverse standards globally and a wide variety of curves are in use around the world. Various assumptions underpin the use of the different fire resistance assessment curves. The Eurocode temperature-time curve, for instance, is treated as appropriate in fire situations where the assumption is that of fire occurring in an enclosure where the fire is ‘powered’ by fuel that is considered ‘celluloic’, and uniformly distributed over the area of enclosure (Barnett and Clifton 2002). A fire time curve is generated from this to determine temperature rise in the structure’s members, up to a limiting temperature value, which is compared with the maximum temperature reached by the structure’s members. However, as the authors note, in this parametric fire curve (i.e eurocode), there are three equations each for the heating and cooling sides of the time-temperature curve, making it difficult to represent such an equation on a spreadsheet. Another criticism of the eurocode is that its decay curve does not properly represent the exponential time-temperature cooling characteristics of experimental fire tests(Barnett and Clifton 2002, pg 381). Generally, the eurocode, the ISO-834 and indeed other standard fire curves make use of the simulation of design fire scenarios, as earlier mentioned. However, as noted by Kruppa et al (2001), calculation methods for essential design fires have an accuracy level that depends on the design objectives and the design scenario. A design fire typically gives a quantitative description of the development or course of a fire with respect to time and space. It generally incorporates a consideration of features such as the ignition mechanism or source of the fire, the growth and spread of the fire, its interaction with the environment, occupants and fire safety equipment in the building/structure under consideration, as well as the fire’s decay and extinction. Practically, there are an infinite number of fire scenario possibilities in a combustible structure or building; hence there is usually a resort to a finite set of design fire scenarios, which are actually based on worst-case fire scenarios (Kruppa et al 2001). This, however, has a shortfall in that there are certain extreme fire scenarios with a low possibility of occurrence which are neglected, in this approach to formulating the design fire assessment curves. A negative result of this is that the risk of such omission is borne by the rest of the society (Kruppa et al 2001). The design fire is generally simulated by describing or defining a rate of heat release as a function of density of fire growth, the fire growth rate, and area covered by the fire (which is usually taken as the fire compartment area. However, a design fire- as Kruppa et al note, contains information/data which cannot actually be obtained by calculation. Furthermore, there are other factors affecting development of the fire, which are not really woven into the fabric of the fire assessment analysis through the standard fire curves. The authors note that the type of ignition source, the density of fire load, the distribution and type of fuel, and indeed the size of the ignition source, for instance, are not accounted for, in analyses using the fire assessment curves to determine fire resistance. Other factors also play a prominent part, e.g the internal ventilation condition, external environmental conditions, the air handling system of the building/structure involved. There are also structural engineering considerations which need to be taken into account, such as the position of the fire relative to columns, bracing systems and other loaded elements/members in combustible structures. The first stage of fire, generally taken to be temperatures below 100 degrees Celsius, is normally neglected when design fires are used in determining fire resistance. The assumption here is that, such temperatures are generally too low to affect structural elements. This can however be criticised on the grounds that it may become vulnerable to the condition where there is a developing fire that is not consistent with the assumption or ‘single layer treatment’ establishing the input of heat to the structural members(The Royal Society of Edinburgh, 2004). The next stage in design fire occurs at about 400 degrees Celsius, where temperature of localised gas affects the members or load-bearing structures. It is generally the point at which flashover occurs, where the fire is regarded as fully developed. At this point, heat radiates from the fire at about 20 – 25 kW/m2; pyrolysis gases cannot burn at this level (due to oxygen deficiency), and flames appear in boundaries through cracks or openings. The treatment involves, where calculation methods are unable to predict fire spread beyond the room or point of origin, assumptions on the reliability of the separating elements, to obtain the time when the fire reaches another room. This can however be criticised, considering that design fires are not necessarily or exactly fitted to prescribed regulatory construction standards for the structural elements. Parkinson ( 2002) notes that it became problematic when trying to relate the time to failure (reliability) to the requirements of the building code, when the time-temperature curve established in 1908- which was not based on the response of the structural elements to a real fire, but a worst-case scenario that could be expected in a real fire- was applied, for example. The standard temperature curve (ISO-834) and the parametric curves are based on the evolution of temperature with time. A report by the Royal Society of Edinburgh on fire structures in 2004 notes however that only the region of maximum temperature and decay stage are consistent with the assumption of this single layer treatment. The report further asserts that the thermal inertia of structural elements is larger than that of the gas phase, so that typical times for temperature changes within solids are longer than those for gas phase changes. Thermal insulation further results in very minor temperature changes through the fire growth period. Spatial temperature distribution within the fire compartment may not be adequately represented and this, again, goes against the assumption or treatment of the fire as one having homogeneity of temperature throughout the compartment. Indeed, the authors of the report note that drastic temperature variations are believed to exist within the fire compartment, and that observations establish that the assumption of a single compartment temperature may actually be over-simplified. Another criticism of the approach is that of the assumption that the structural element’s surface temperature is that of the gas. Although this is simpler, introducing less parameters into the equation, it is deemed as not describing adequately the technicalities of the heat transfer process. The current practice in application of fire curves ignores the fire growth period. While this may be ok for first order magnitude approach, it is not adequate for compartmental fire modelling, as the Royal Society of Edinburgh report notes. Quite significantly, the authors expressed that within a fire scenario it is possible that flashover is attained within the compartment of origin before any of the structural elements have undergone significant heating(!). Therefore, if compartmental fire models are to be incorporated in fire resistance analysis of structural members, realistic time scales must be established for fire growth beyond the compartment of origin. Characteristic conditions also need to be more realistic, in the context of the above, as the authors note. Crucial attention is (and must be paid), in analysis involving the use of standard fire curves, to the limit state conditions of members. Indeed, Kruppa et al (2001) note that fire resistance performance criteria using the fire curve analysis must include recognition of factors such as the limit state condition of thermal insulation, where temperatures at the unexposed end of the fire barrier attains a level of intensity as to ignite combustible materials either by contact or radiation. The limit state condition arising from smoke is also vital, where combustion products develop the ability- through propagation of toxic gases and smoke- to pass through separating elements. Other important limit state conditions include load bearing capacity where, as a result of reduced mechanical strength at high temperatures occasioned by fire, this capacity becomes less than the actual load. The limit state condition of integrity is crucial in the considerations as it relates to the spread of fire- sequel to the emergence of cracks- as a result of combustible gases passing through barriers. In order to properly analyse the thermal effects of design fires as used in standard fire curves, the behaviour of the boundary elements need to be understood. Kruppa et al (2001) note that the accuracy of the assumed time-temperature curve must be verified. A general drawback, however, is that integrity failure cannot be accurately predicted through simulation and calculation. Assumptions therefore have to be made on the basis of test results- which are in danger of being rather ad-hoc and therefore not very reliable. In defining design fires- in an enclosure or compartment- assumptions are made. Kruppa et al (2001) identify the following: (1) Doors assumed closed if the compartment has other openings, (2) doors are assumed open if the enclosure has no other opening, (3) Assumption of broken glazing if such has no fire rating. An ‘opening factor’ is normally applied in simple models, to model the openings in an enclosure, whereas in more complex analyses flow calculations are applied using information on the actual flow through the opening. An obvious criticism of the assumptions is that they are not exactly accurate- doors being open, for instance, simply because there are no other openings in the enclosure is not necessarily true in a real situation (all openings may, in fact, be closed- in a fully air-conditioned interior, for instance); so that a wrong assumption becomes the starting premise for the fire analysis (!). And in the true manner of all mathematically-based analyses, everything based on such wrong assumption may be in real jeopardy. In fire resistance analysis using standard fire curves, grounds for further criticism exist when we consider ‘accidental actions’. Earthquakes, for instance, are normally not considered in such assessments/analyses except in countries particularly vulnerable to such ‘acts of God’, in which case account is taken of possible damage to buildings. Obviously, neglect of such accidental actions, in a general sense, would impact upon fire resistance analysis results- and not in a particularly positive way- as there has been omission of an element /variable contributing negatively to the structural and fire resistance status of the members/elements being assessed, using standard fire curves. Temperature may, for example, break down a weakened wood/steel member much sooner than it normally would, and this certainly jeopardises the time-temperature curve applied in the analysis. It is also to be noted that there is no real correction for such omission, in the curves. There are other considerations in analysis of fire resistance. Impact, for instance, is taken into account because of collapse of elements on the side exposed to fire; hose-stream impact from fire-fighting activities affecting unexposed side of separating elements; forces and moments induced as a result of thermal elongation of smoke, or even the shrinkage of surrounding elements; the deformation of single elements, resulting in the use of load bearing separating elements; also, the pressure of the fire in its developmental stages is taken into account in the assessment curves, because a vertical pressure gradient develops in this particular phase of fire. Fire assessment calculations are made from the geometry of the structure/building, the rate of heat release of final items, the thermal behaviour of boundary elements, the type and amount of fire load. The design value of these factors impacts on safety levels. In offshore applications, jet fires generally result from gas leakage occurring under very high pressure. This typically occurs in offshore engineering work such as oil-drilling or related petrochemical activities. Traditional fire resistance tests carried out onshore do not effectively recreate or replicate the processes of heat transfer, gas velocities and thermal shock- major factors with regard to passive fire protection in actual fires- and in particular, jet fires resulting from high-pressure gas leaks (Chamberlain 2002). There are generally two broad categories of jet fire test procedures: (a) Interim Jet Fire test, which assesses the effectiveness level for passive fire protection materials, and (b) Jet fire resistance test- this makes use of a 0.3kg/s propane gas jet flame to simulate conditions of large scale natural gas jet fire. Offshore applications also include compartmental fire modelling (CFD), applied for the purpose of determining heat flux to offshore vessels, or indeed the elements or specimens being assessed. In this technique, there is a comparison of properties such as convective and radiative transfer of heat- which are normally not amenable to easy measurement; temperature anywhere in the floor field, and gas velocities are also measurable, in this procedure. In some offshore applications, such as those involving GRP piping, the approach may involve minimising the number of fire test variables. Analysis may not involve a cellulosic fire scenario in fire resistance testing, while hydrocarbon and pool fire resistance assessment is carried out. The logic of this is simply that hydrocarbon fire is more severe- hence the cellulosic scenario is automatically taken care of. Offshore applications may further involve hydrocarbon tests simulating the effect of the rupture of a vessel. A variety of codes/standards are available which use different values for the effect of such fires. In certain (offshore) applications, the fire severity is mitigated by the cooling effects of firewater. Hydrocarbon jet fire- unlike onshore fire generally- comprises a turbulent diffusion flame which results from the combustion of gaseous or liquid fuel. This, continually released in a particular direction, is characterised by instantaneous high heat flux. It is highly erosive because of the high velocities of flow that result. The heat flux may vary widely, and release conditions will be affected by the size of the opening through which the hydrocarbon gas is escaping (Wright et al, 1994). Fuel type will also have major effects on the characteristics of a jet fire. Typical conditions may involve the sonic release of pressurised methane gas through a small opening- at a rate of 2-3 kg/s- producing a 20 metre-long flame having a heat flux of about 250kW/m2(this contrasts sharply with a typical value of 20 – 25kW/m2 in flashover cases onshore), and a velocity of about 60m/s. In offshore application, the BS 476 curve may involve consideration of the combined effect of ignition characteristics and the extent to which the surface of the product spreads flame. Also, the influence of underlying materials in the context of their ability to influence the rate of fire growth is taken into account. The test result is a function of the distance travelled and the rate of lateral spread of flame. Offshore applications also include the ISO 5660-1, 1993; also known as the cone calorimetric. It simulates heat release (by the oxygen consumption principle) of the residual oxygen concentrated in the stream of the exit gas. This technique allows fire reaction properties such as ignition time and heat release to be studied. Smoke and toxic product generation are also measurable. Pool fire, burner and furnace fire tests are generally also applied offshore. In the furnace test, for instance, structural samples are assessed in the form of panels, with the side in the furnace subjected to a temperature profile following one of the fire curves. The time taken by the cool face of the panel to attain a temperature of 140 degrees Celsius represents the fire resistance. Composite materials are in use offshore, where the behaviour of thick composite laminates enables the retention of integrity and the fire-protecting effects of the materials to be predicted- at given values or levels of hot face temperature or heat flux. Perhaps the most important factor is the endothermic nature of the resin decomposition process, because it delays heat transmission through the laminate. . Bibliography Wright, P et al, “Specifications And Recommended Practice For The Use of GRP Piping Offshore”, 1994 05 May 2010 Gibson, A. G, “The Cost Effective Use of Fibre reinforced Composites Offshore”, 2003 05 May 2010 The Royal Society of Edinburgh, “Fire and Structures: The Implications of The World Trade Center Disaster”, 2004 04 May 2010 Parkinson, David “Performance Based Design of Structural Steel For Fire Conditions”, 2002 04 May 2010 Kruppa et al, “Rational Fire Safety Engineering Approach To Fire Resistance In Buildings”, 2001 04 May 2010 Barnett, C. R and Clifton, G.C, “Examples of Fire Engineering Design”, 2002 04 May 2010 Chamberlain, G.A “Controlling hydrocarbon Fires In Offshore Structures”, 2002 04 May 2010 7 Read More