Heat Energy is a form of energy characterized by vibration of molecules and capable of initiating and supporting chemical changes and changes of state NFPA In other words, it is the energy needed to change the temperature of an object - add heat, temperature increases; remove heat, temperature decreases. Temperature is a measure of the degree of molecular activity of a material compared to a reference point. Human body core temperature that may cause death 3. Human skin temperature causing a first degree burn injury 4. Hot water causes a scald burn injury with 30 s exposure 5.

Human skin temperature with blistering and second degree burn injury 4. Temperature when burned human tissue becomes numb 4. Human skin temperature at which tissue is instantly destroyed 4. Temperature when water boils and produces steam 6. Modern synthetic protective clothing fabrics begin to char 7.

Temperature of gases at the beginning of room flashover 8. Temperature inside a room undergoing flashover 8. References: 1 Klinghoffer, Max, M. NFPA, pp. Heat transfer is a major factor in the ignition, growth, spread, decay and extinction of a fire. It is important to note that heat is always transferred from the hotter object to the cooler object - heat energy transferred to and object increases the object's temperature, and heat energy transferred from and object decreases the object's temperature.

These values are found empirically , or, by experiment. For free convection, values usually range between 5 and But for forced convection, values can range anywhere from 10 to Fire Development is a function of many factors including: fuel properties, fuel quantity, ventilation natural or mechanical , compartment geometry volume and ceiling height , location of fire, and ambient conditions temperature, wind, etc. In other words, the fire growth is not limited by a lack of oxygen.

As more fuel becomes involved in the fire, the energy level continues to increase until all of the fuel available is burning fully developed. Then as the fuel is burned away, the energy level begins to decay. The key is that oxygen is available to mix with the heated gases fuel to enable the completion of the fire triangle and the generation of energy. The element matrix equation is obtained from the governing Eq. Using Eq. Note that the element nodal temperatures cannot be solved from the element matrix Eq. Note that, Rq is other than zero only when it is in the position corresponding to the node that is on a boundary.

To solve Eq. Integration techniques for transient non-linear solutions are typically a combination of the methods for linear transient solutions and steady-state non-linear solutions Huebner, Thornton and Byrom, ; Zienkiewicz and Taylor, The transient solution of the non-linear ordinary differential equations is computed by a numerical integration method with iterations at each time step to correct for non-linearities.

Equation 6. Calculation approach Equation 6. The Newton—Raphson iteration method is often used to solve the non-linear equations at each time step. The explicit and implicit algorithms have the same trade-offs as occur for linear transient solutions. The explicit algorithm requires less computational effort, but it is conditionally stable; the implicit algo- rithm is computationally expensive, but it is unconditionally stable. The non-linear implicit algorithm requires even greater computational effort than in linear implicit solutions because of the need for iterations at each time step.

The analysis, however, needs to consider the continuing changes in mate- rial properties due to rising temperatures. But then it is important to continue the calculations in order to deter- mine whether these local failures could lead to progressive collapse of the whole structure. Therefore, the structural analysis should be time dependent although the inertia and damping forces may not necessarily be involved in the analysis.

The analysis with history dependent based on time steps but without considering the inertia and damping forces is called the quasi-static analysis. Calculation approach 6. The main assumption of the Bernoulli beam is the linear distribution of the axial strain in the cross section. It is known that, when the strain involves the plastic strain, the stress—strain relation is usually expressed in the increment form.

According to Eq. Similarly, the increment form of Eq. In the case where the temperature and internal forces are independent of the longitudinal coordinate, Eq. Examples of this include the column subjected to pure compression and the beam subjected to pure bending. Otherwise, the membrane strain and curvatures must be solved by considering the com- patibility along the longitudinal direction of the member with imposed or calculated end conditions Purkiss and Weeks, ; Purkiss, a. Thus, iterations are required in solving the equations at each time step to cor- rect for non-linearities.

The method can be applied to steel, concrete and composite steel—concrete members. For timber, which chars substantially when sub- jected to heat leaving a relatively unaffected core, calculations can be undertaken using normal ambient methods with the use of temperature- reduced strengths if considered appropriate on the core after the parent section has been reduced by the appropriate depth of charring. The analysis considers the whole frame action and includes geometrical and material non-linearities within its beam—column and slab elements.

For steel this includes the clas- sical creep strain and the strain induced by the mechanical stress. In the case where the classical creep strain is negligible, this reduces to just the strain induced by the mechanical stress and therefore the stress—strain temperature behaviour can be simply represented by the temperature- dependent stress—strain equations as described in Chapter 5. Although there are many computer simulation results published in literature, there have been limited comparative, or benchmark, tests com- missioned for both thermal response and, more importantly, structural response.

The latter is very much affected by materials models and the exact formulation of the analysis techniques used. For the thermal analysis programs, particularly for steel structures, it seems that most packages gave comparable answers and that these answers were also in reasonable agreement with experimental data. For the structural analysis packages investigated, a greater spread of acceptability was found. It appeared that many of the programs predicted correct trends but that the absolute results did not agree with experimental data.

It was also noted that the effect of classical creep on the behaviour of steel was neglected. The problem was solved using the simple approach described in section 6. Because of the axial symmetry of the problem, the temperature and axial compressive stress are axial sym- metric. As is seen in Fig. This is because steel has a much greater thermal diffusivity than concrete. Two main steps were followed in the analysis procedure. Both geometric and material non-linearities were included in the simulations to account for the expected large displacements, plastic deformations and creep.

The highly non-linear problem was solved using iterative procedures with automatic time stepping. The time to collapse is about 45 min.

Ceiling jet fire

The minimum clearance required between the frame and the wall increases with the length of the fuel burning. Copy with permission from Ali, Senseny and Apert, Calculation approach Figure 6. The problem shown here is a simple concrete wall of mm thick. The wall has an initial porosity of 0. The temperature results for the case where the moisture transfer is considered are taken from Tenchev, Li and Purkiss b. The experimental data are taken from Ahmed and Hurst Experimental data are taken from Ahmed and Hurst Coupled model considers the transfer of both heat and moisture and the results are taken from Tenchev, Li and Purkiss b.

The problem is the same as that shown in Fig. The two sets of results shown in Fig. This is simply due to the high tem- perature that weakens the stiffness of the structural member. Therefore, the geometric non-linearity or the second-order effect, i. The prescriptive approach, in which tables of minimum dimensions, minimum axis distance, etc. Because of the complexity of the stress—strain relationships involved for concrete in compression where the elastic strains, unrestrained ther- mal expansion and transient strains need to be taken into account, it is not generally possible in simple design methods to consider the deforma- tion history of the structural element.

However, using numerical stress—strain curves such as those proposed by Li and Purkiss , it may be possible to determine deformation history using spreadsheets. The time to failure for a structural element may be determined by calculating the moment capacity or axial capacity at a series of discrete time steps. The previous chapter has dealt with matrix techniques for both temperature distribution and stresses within structural elements, this chapter will concentrate on hand methods, although the use of spread- sheets may be found to be advantageous.

The methods presented will assume that the thermal analysis and the structural analysis may be decoupled. Design of concrete elements 7. It is generally accurate enough to use such data for the end-point design of concrete members. There are two such available methods. Both methods are applicable to different concretes as the ther- mal diffusivity is entered as data.

The presentation of both methods, given below, is limited to exposure to the standard furnace curve. Note that for exposure to the standard furnace, temperature—time curve L does not need calculation, since f3 is always zero. For exposure to the standard furnace curve, values of C, D and E are given in Table 7. It should be noted that C is equal to twice the required time period. For the values of these parameters when exposure is to a parametric compartment, temperature—time response reference should be made to Hertz. Design of concrete elements Table 7. Source: Hertz b by permission Table 7.

This comparison is carried out in Table 7. Thus it would appear acceptable to use any of the methods to predict temperature rise. EN does not give any guidance on this. A closed form solution cannot be obtained from Hertz, thus a spread- sheet was used to obtain values of x for the same values of ac. Additionally, values of x obtained from Fig. A2 of EN are tabulated in Table 7. It would appear from Table 7. Hertz consistently overpredicts the value of x and is thus conservative. From Table A1.

EN provides graphical data for reduction of concrete strength and section width in Annex B Fig.

EN imposes a restriction on the zone method, that it may only be used for exposure to the standard furnace curve. Example 7. Continuity steel in the top face has been provided purely to resist the effect of cracking over the support. The cover provided is that to satisfy durability only. Thus the slab is satisfactory.

Divide the slab into 5 slices Table 7. Thus the slab is satisfactory, but only if the top steel is mobilized. The data for the example are given in Fig. These values are obtained from Table 7. Design of concrete elements Determination of reinforcement temperatures Tables 7. As all the reinforcement is the same size, Eq.

Fire Safety Engineering Design of Structures

The results are given in Table 7. The temperatures corresponding to depths of 68, and mm are given in Table 7. The determination of reinforcement temperatures and strength reduction factors are tabulated in Tables 7. However, in both cases the beam will last for more than 4 h. EN — Annex B3 gives a method of handling this situation, but it is iterative as the column curvature s need to be taken into account.

The values of the loss in concrete area due isotherm rounding are given in Table 7. Ignoring rounding: The values of x are taken from Table 7. The results of the calculation are given in Table 7. Table 7. The results, considering rounding of the isotherms, are given in Table 7. Note, as the number of strips is even the mean concrete temperature at the centre will need calculation separately.

The heat transfer is considered through the narrower dimension of mm and the results are given in Table 7. However, this will not be the case for pre-stressed concrete due to the moments induced in the sec- tion by the pre-stress. Bobrowski and Bardhan-Roy indicated that the critical section for shear was between 0,15 and 0,2L from the support, where L is the span.

The problem is more likely to be worse in pre-stressed concrete construction where bond in the anchorage length is needed to transfer the pre-stress force into the concrete. However, there appears to have been few, if any, failures in pre-stressed concrete directly attributable to loss in bond.

The second form is known as sloughing, whereby the concrete gradually comes away due to loss of effective bond and strength loss. The magnitude of the effects of spalling is demonstrated both by actual test results and computer simulation. Results are given by Aldea, Franssen and Dotreppe quoted in Table 3. The exact mechanism of explosive spalling is still not understood, but it is affected by the following factors Malhotra, ; Connolly, , : 7. The original proposal by Meyer-Ottens also suggested stress limits. However, it must be pointed out that a porous concrete will give a poor performance with respect to durability.

It has also become clear that it is a combination of moisture content and permeability is critical Tenchev and Purnell, This is indicated in Fig. The values of water content W in Fig. Tenchev and Purnell, , by permission. It is possible for a homogenous concrete to determine pore pressures using a coupled heat and mass transfer model Tenchev, Li and Purkiss, a, b; Tenchev et al. The following results are given in Tenchev, Purkiss and Li c.

In deducing Eqs 7. This can partly be explained by the fact that in areas of compressive stresses, cracks cannot open up to relieve internal pres- sures. This does not mean that spalling cannot occur in areas of sagging moments where tensile cracks exist, since it is possible that pressure build up will still occur as in general tension cracks are discrete and not part of a continuum.

This is likely to be linked to the basic porosity of the aggregate, in that siliceous aggregate is impermeable com- pared to the others and that moisture transport has to occur through the mortar matrix. However, there is now some evidence that limestone and lightweight aggregates may give problems, especially in younger concretes as the pore structure of the aggregate may provide convenient reservoir storage for free water Connolly, Spalling is also exacerbated in thin sections, partly since the depth of spalling is a greater proportion of the section dimension and hence proportionally worse, and partly due to the fact that there is less of a cool reservoir for any moisture to migrate toward Khalafallah, High covers are also likely to produce greater amounts of spalling.

These restrictions often concern the placement of a light mesh with 4 mm wires at a spacing of mm at the surface of the concrete cover when the axis distance exceeds 70 mm, in order to retain the cover EN The rate of heating is therefore critical to an assessment of the like- lihood of spalling.

However, if a concrete although designed as normal strength has a much higher strength than that designed for, problems may ensue. The actual strength at the time of test was 61 MPa cube or approximately 50 MPa cylinder. The problems are exacerbated for high strength concretes. The higher tensile strengths of such concretes do not remove the problem as tiny pores will act as stress raisers and hence reduce the effective tensile strength. EN allows a number of methods to reduce the effect of spalling on high strength concrete. As the thermal properties of steel are sensibly indepen- dent of the steel strength it is also possible to utilize empirical equations to determine the temperature rise within an element.

This chapter covers both the calculation of temperatures within an element and the methods by which the load-carrying capacity at elevated temperatures may be determined. A full discussion of the evaluation of the convection and radiation boundary conditions is given in Chapter 6. However, for steelwork there are three components including that due to the insulation. EN cl 4. However, this is likely to be conservative and Eq. Example 8.

The use of Eq. The reason for this sub- division was that the equation used for calculations where the insulation had substantial heat capacity was unstable when the other situation was operative. However, for historical completeness the ECCS method is included. The external face of the steelwork may then be considered to have the same temperature as the furnace gases. The second case is where the insulation has substantial heat capacity or there may also be a substantial moisture content.

Join Kobo & start eReading today

The insulation temperature can then be taken as the mean of the steel temperature and the gas temperature. It should be noted that the calculations for both uninsulated and insu- lated sections are best performed on a spreadsheet, as is done for the examples in this chapter and the next. There are two approaches that may be used to deal with this effect. The latter approach is indicated in EN , although no explicit method is included but reference is made to EN Note that Eq. Thus when the insulation has substan- tial heat capacity, Eq.

The heated perimeter Am will depend on the type of insulation, e. Lawson and Newman give similar data. Three examples on calculating temperature rise will be carried out, two on uninsulated steelwork and the other on insulated steelwork. The value of ksh will initially be set equal to 1. The governing equation for this case is Eq.

The calculations are presented in Table 8. It will be noted that there is little difference between both sets of results. Also, opportunity was taken to evaluate the effects of using the values of ca which vary with temperature, Eqs 5. For I sections, EN — cl 5. The results are plotted in Fig. The results from these calculations are also plotted in Fig.

It should be noted that after around 20 min, the predicted temperatures are sensibly independent of assumptions. This is not unreasonable if Fig. The governing equation is identical to that of Example 8. As in Example 8. These negative values have been ignored and set equal to zero when calculating the steel tempera- tures.

Typical calculations for the start and end of the heating period are to be found in Table 8. If the time shift calculations are carried out according to Eqs 8. Design of steel elements Table 8. It is convenient because of the different approaches used to con- sider non-composite and composite construction separately. Composite steelwork is covered in the next chapter. The complete test data are given in Wainman and Kirby , Results from unprotected beams and columns with partial or total shielding and varying loadings are given in Figs 8. For tension members with a non-uniform temperature distri- bution, the axial capacity may either be obtained by summing the contributions of incremental areas or conservatively using the maximum steel temperature reached and assuming constant temperature.

Figure 4. Web is Class 1. Design of steel elements From Eq. The use of the quadratic equation for determination of protection thicknesses and the heat transfer calculations give almost identical results. As the beam can suffer lateral torsional buckling, the critical tem- perature approach cannot be used.

As the solution to Eq. Thus from Eq. The effective length of the column is 3,5 m. Methods are therefore needed that can be used to calculate the temperatures in external steelwork, since it is possible if the temperatures attained are low enough that no protection need be applied to the steelwork. The problem therefore is one of temperature calculation rather than strength response. The weld on the underside of the angle is to be neglected. Table 8. The example chosen is that reported in Data sheet No 35 in Wainman and Kirby Thus the beam would still have had the ability to resist the applied moment for slightly longer than the time period quoted in the test report.

From Table C. The relevant zones and dimensions for calculating temperatures are given in Fig. The calculations are carried out for one-half the beam in Table 8. Thus the neutral axis lies within Zone 4. Part of the dis- crepancy is due to the temperatures measured in the test and the assumed values in the calculations. These are summarized in Table 8. The latter will have a lesser effect on the moment capacity as they are closer to the centroidal axis.

Wainman and Kirby also reported that the actual strengths were for the Grade beam MPa and for the Grade shelf angle MPa. These actual strengths would also increase the capacity of the section, thus overall the calculated moment capacity is acceptable. Generally spray systems are limited to beams where there are false ceilings. The non-loadbearing masonry effectively acts as a heat sink and thus reduces the average temperatures within the steel- work.

The design of such columns is covered in Chapter 9. Since formwork is needed for the concrete together with some reinforcement and the need for curing time, it is probably more economic to use full reinforced concrete construction. Some of the problems can be mitigated by the use of pre-cast concrete pro- tection, but concrete will still need to be cast around the beam-to-column or beam-to-beam connections. Design of steel elements 8.

It is known that this approach is conservative and that steel members can achieve higher temperatures than these at failure. Equation 8. Table 1 of AD then gives multiplication factors for increase of protection thicknesses. The appropriate strength reduction factors are given in Annex D of EN The effect on cooling will be to maintain the increased deformations even though there will be little resultant effect on the residual strength of the steelwork section Having thus considered non-composite steelwork elements, it is necessary to turn to composite steel—concrete construction.

As with reinforced concrete construction the slab is required to satisfy both the load-bearing capacity and insulation limit states. Composite construction h1 h2 yp l2 l3 l1 a h1 h2 yp l2 l3 l1 b Figure 9. The required values of heff for normal-weight concrete are given in Table 9. Source: Table D. This is not unreasonable as the depth of the concrete stress block will be small since the tensile force in the reinforcement that the concrete in compression is required to balance is low.

This means that it is very unlikely that the concrete strength will be affected by excessive temperature rise. The moment capacity is then determined using conventional rein- forced concrete theory. Only the concrete cross section with temperatures less than a limiting u1 u2 u3 a u1 u3 u2 b Figure 9. A conservative approach to determine the position of isotherms in a mm thick normal-weight concrete is given in Table 9.

The strength reduction factors for normal-weight concrete are given in Table 6. Where appropriate, the reinforcement strength should be reduced using the concrete temperature at the level of the reinforcement. The moment capacity must then be calculated using basic theory, but with an iterative procedure as the depth of the neutral axis is not known a priori.

Example 9. In this example both will be used, although clearly in practice only one is necessary. The dimensions of the deck and sheeting are given in Fig. It would possibly be expected that the effective thickness approach would have been more conservative! Composite construction The temperature at heff minus centroidal distance to top steel, i.

From Table 5. Two options can be considered: design and detail the mesh such that no bottom reinforcement is required to carry any sag- ging moments, or design and detail the mesh such that bottom reinforcement is only required in the end span. Although the former is more economic and more practical on site, both solutions will be investigated here in order to demonstrate the principles involved. This must be balanced by the force in the strength reduced concrete. Take a series of 5 mm deep strips in the concrete, determine the concrete temperatures at 0, 10, 15 mm etc. This calculation has been carried out in Fig.

Place the required reinforcement level with the top of the dovetail, i. Supply H6 bars in alternate ribs, i. The two types of deck can generically be described as trapezoidal and dovetail. It is also important to distinguish the rela- tive direction of the beam span and deck span as this may affect the temperatures within the steelwork Newman and Lawson, For the determi- nation of structural behaviour EN adopts one of two approaches concerned with the calculation of either a critical temperature or the full moment capacity.

In both cases, the calculation of temperatures uses the equations for non-composite steelwork which assume no thermal gradients. Composite construction 9. This may only be used to determine the sagging moment capacity. The notation used is given in Fig. For unprotected composite steel beams, Eq. Annex E details the model that may be used to determine both the sagging and hogging moment resistances of composite beams.

The tensile capacity of the steel beam T calculated using temperature reduced steel strengths Table 5. The position of hcr and the elemental temperatures should be taken from Table D. The design data for the composite beam are given in Fig. The beam is at an internal support of a composite deck system. From the ambient structural analysis of the decking, the application of a udl of q to the deck, gives a beam reaction per unit run of beam is 2,98q.

1. Hormones, Signals and Target Cells in Plant Development: 41 (Developmental and Cell Biology Series).
2. Fire Dynamics.
3. Connect with us.

Composite construction From Eq. However, rather than to use another example, the calculations that follow are illustrative only. Composite construction The design moment of resistance of the composite beam is kNm, so from Eq. It will be necessary to sum the contributions of the concrete and reinforcement incrementally. If the loading is eccentric, the eccentricity should not exceed 0,5b or 0,5d.

For column in continuous construction where local buckling did not occur, then the buckling length may be taken as 0,55 times the actual length Bailey, Where local buckling is included in the analysis, then the factor should be increased to 0, For column continuous only at one end the buckling factor is increased further to 0,8. For unprotected columns, the contribution of the concrete will need to be determined incrementally. Newman also provides design tables for UK sections. There is also generally no need to consider any strength reduction in the residual section as any temperature rise can be consid- ered small and therefore ignored.

Beech should be treated as a softwood. Design of timber elements The background to the determination of cross-sectional resistance determination is given in Kersken-Bradley The beam is therefore satisfactory. Example For both the beam and column example, the reduced properties method is less conservative than the reduced cross section method.

It is convenient to consider these in historical order. So from Eq. Equation Values of the parameter f are given in Table Table However, most columns are neither short nor extremely slender but have a slenderness ratio such that failure is by a combination of squash- ing and buckling. Design of timber elements For a square column, Eq. Calculate the loss of section for the side faces from Eq. Although this is low it would be acceptable in an accidental limit state.

Use Eq. For a void cavity with depths from 45 to mm, tins,0 is taken as 5,0 min. The term plastics also cover the use of plastic-based composites. Notes are also included on the behaviour of glass.

Fire Safety Journal

Often where masonry is used as external cladding, again the wall is carrying no applied vertical load, but only has to resist the effect of horizontal wind loading. In the UK, at least, load-bearing masonry, i. Data on these requirements may be found in EN It is possible under certain circumstances to pro- pose either interaction formulae between individual layers or to allow for the effects of coatings de Vekey, Also it should be noted that whereas a wall panel, typically some 3 m square, when tested is likely to be stable, extrapolation of stability and any resultant reduction in load- carrying capacity cannot be made to larger panel sizes found in normal building construction.

A wall with no vertical or top restraint will show some bowing, although this will be reduced in the initial stages by the effects of any ver- tical loading applied to the wall. The vertical load is only likely to cause problems after a substantial period when the wall may tend to become lat- erally unstable with the vertical load inducing high tensile stresses owing to the moments induced by the lateral deformations.

A comprehensive series of tests on mm and mm thick with var- ious load levels were carried out in Australia Gnanakrishnan, Lawrence and Lawther, The results are given in Table Temperature measurements during the tests indicated that the temperature rises in the external leaf were relatively low with a substantial gradient across the mm cavity. Some attempt was also made to model the behaviour of the cavity walls but, due to the assumptions made and the paucity of high strength data on masonry, the results appear only to show the correct trends.

Thus the temperature within an aluminium section can be calculated using the equations given in Chapter 8 using the relevant properties given in Chapter 5. Owing to the extensive use of aluminium in the off-shore industry, it may also be necessary to consider the use of the hydrocarbon curve rather than the standard cellulosic curve to give the gas temperature. The limiting temper- ature is taken as a function of the exact aluminium alloy in use as the temperature-related strength loss is very dependant on the amounts and type of alloying constituents.

This therefore means that such materials need extensive levels of protection in order to retain load-carrying capacity at elevated temperatures. This therefore means that the insulation thicknesses need to ensure that the temperatures within the plastic element need to be kept close to ambient. Unfortunately the temperature levels for which the insulation was designed when applied to the extruded sections is not stated.

Rather than to repeat the material produced recently in a considerable number of papers, tests on the various frames at Cardington are posted here so that the emphasis is placed on discussing the implica- tions for design. Additionally, the performance of connections and portal frames will be highlighted. However, before continuing to discuss Cardington, it is perhaps worthwhile to examine the tests on beam—column connection behaviour which was also highlighted by the Cooke and Latham test.

The results from the tests are given in Table It should be noted that the tests were on assemblies which modelled internal columns where the net moment on the column would sensibly be zero owing to the balanced nature of the loading. The tests also provided data on the temperatures reached in the bolted connections. These data enable capacity checks to be carried out on the connection. Experimental work Leston-Jones et al. P indicates that both the beam and column were protected to 60 min, U both the beam and column were unprotected, and B the web of the column was blocked in with lightweight concrete blocks.

Source: Lawson a equation similar to the Ramberg-Osgood equation for the ductile stress—strain curve, i. The results from this test allowed a relaxation in the England and Wales Building Regulations on timber frame construction, and allowed the Scotland and Northern Ireland Regulations to be brought into line.

Also, the columns were kept at the same size throughout the structure with the concrete strength being increased in the columns in the lower storeys. The columns in the lower storeys were constructed using Grade 85 high strength concrete with limestone aggregate and microsilica. The test was carried out in September shortly before the Cardington Test Facility was closed. The technical data are taken from Bailey Since the author was present at the test and was able to inspect the structure shortly after the test, the interpretations of the results are those of the author.

This would put the slab concrete into the category of high strength concrete. From Fig. In the actual test, spalling commenced at around 6—7 min after ignition. The actual depth of spalling was around 20—25 mm as much of the bot- tom reinforcement was exposed. The test also indicates the possible need to consider maximum as well as minimum strength requirements for concrete. It should be noted that the central column Grade C85 with a cube strength of MPa cylinder strength 80 MPa at 28 days did not spall. After the test, cracking was observed at the corners of the column — it is not known when all this had occurred.

It needs to be noted that the moisture contents were high as the building was unable to dry out as would occur in a structure built in an open air. This gave a load of kN in the central column and kN in the edge column. A much stiffer structure with lift shafts and stairwells may have induced higher compressive stresses into the slab and hence increased the spalling levels. The composite deck was lightweight concrete cube strength 47 MPa, cylinder strength 38 MPa.

There were six main tests carried out with applied loading in each test of kPa Moore and Lennon, The columns were protected within mm below the connections. This test also produced distortional buckling of the column heads with the columns shortening by around mm. The secondary beams were heated over a length of around 1 m. Again, this was due to high tensile forces induced during cooling.

Test 3: Corner compartment no. The opening factor was initially 0, m0,5 and subsequently increased slightly to 0, m0,5. It was estimated that the heating regime had a time equivalent of around 85 min. The internal compartment wall was placed slightly eccentrical to the 9 m internal beam, but under the 6 m internal beam. Little damage to the blockwork compartment walls was observed. Test 4: Corner compartment no. One external face length 9 m was formed using double-glazed aluminium screen. Additionally, local buckling of the beams at the connections and end-plate fracture in the connections caused during cooling were observed.

A substantial amount of cracking in upper surface of the composite slab around the column was also observed. The lift shaft was also provided with additional protection. Double glazing was installed on two sides of the building with the middle-third of the open area on each side left open for ventilation.

All the beams were left unprotected whilst internal and external columns were protected up to and including the connections. After the test, it was observed that extensive local buckling occurred at the beam to beam connections with a number of failures in the end plates. One case of a complete fracture between the beam web and end plate was noted. Ventilation was provided by a combination of blank open- ings and widows. All the beams were left unprotected whilst internal and external columns were protected up to and including the connec- tions.

This was in part due the mesh in the top of the slab being inadequately lapped. Frames Observations from the tests: 1 It is very obvious that the overall behaviour of members in a frame is far superior to that predicted by isolated furnace tests. This is in part due to continuity and in part due to alternative load paths or load-carrying mechanisms. These tensile forces may cause failure of the end plates, welds or bolt shear. The performance of the frame under test conditions clearly has impli- cations for design, but these implications can only be realized if it is understood how the frame is actually behaving.

This can only be achieved through computer simulation. Much effort has been expended on this to the extent that the results from the Cardington Tests can be reproduced e. The membrane action is generated by a compression ring around the edges of the slab and a central tension zone. The approach can be outlined as follows. The total load capacity is due to any of the unprotected beam within the area being considered plus that due to the slab determined by yield-line response. However, the yield-line load capacity may be enhanced due to membrane action.

Frames From Eq. This means that they are both unable to carry any of the remaining dead load from the roof and to supply any propping restraint for the stanchions or external cladding. The principle is to ensure that when movement occurs the stanchion will rotate about its foot into the space occupied by the structure. The full background to the calculations is given in Simms and Newman At collapse hinges will tend to occur at either side of the apex con- nection and in the rafter at the end of the haunch.

With these assumed positions of plastic hinges, it is then possible to determine the moment and other forces required to provide stability at the foot of the stanchion Fig. Further guidance on this is also given in Simms and Newman Table 2. The vertical reaction is simply half the load on the frame. This situation is exacerbated by the amount of non-structural debris lying around together with the acrid smell of many combustion products. In most cases, the damage is not as severe as is initially thought, even though immediate decisions must be taken on the short-term safety of the structure and whether any temporary propping is necessary or, indeed, whether some demolition work is necessary.

This question can often be answered after a thorough visual inspection has been carried out. It is also necessary to consider the stability if excessive bowing has occurred in any masonry cladding or internal compartment walls. In the case of concrete construction, attention should be given to damage due to spalling on beams and columns as this may reduce the load-carrying capacity of the member due to excessive temperature rise in any reinforcement. It is thus important that no debris is removed until such a study is carried out; otherwise, vital evidence may be lost.

Also, this method only gives an indication that particular temperatures were reached and not the duration of exposure to that temperature. The charring depth can be related back to the standard furnace exposure since timber of known, or established, density can be assumed to char at a constant rate between 30 and 90 min standard exposure section 5. The position of the timber specimen in the compartment should also be noted. In practice, no one of the above methods is completely reliable and therefore a combination of methods must be used to give a reasonable answer.

This last category merits no further discussion. If repair is considered feasible, then a much more thorough investigation is required to ascertain the exact extent and severity of any damage and the residual strength of the structure. The second stage ascer- tains the residual strength of both the individual members and of the complete structure. Other observations needed in the survey depend on the main structural material: steel, concrete or masonry. In concrete structures, it is necessary to note the existence of spalling and therefore exposed reinforcement.

It is useful to note the colour of the exposed concrete face as this can give an indication of the temperature to which the ele- ment was exposed; care is needed, though, because spalling may nullify the observation, and some aggregates do not exhibit colour changes. Note should also be made of the formation of cracks. Cracking is unlikely to be deleterious in the tension zones of reinforced concrete beams, but will indicate the existence of severe problems should it occur in the compres- sion zones of beams or slabs or in columns.

In steel structures, since most structural steels regain more strength on cooling see section However, the resultant deformations are likely to indicate the state of the structure. In this case, it is important to assess the integrity of the connections; it is possible that bolts could have failed within the connection or could have become unduly deformed.

This separation can still occur even if thorough deck stud welding was used. Another potential point of failure is the shear bond between the decking and the in situ concrete. Masonry is either used in low-rise load-bearing structures or as cladding to framed structures. This is less likely in low-rise construction where substantial amounts of timber are likely to be used. If the damage is restricted to the inner leaf it may be possible to retain the external leaf and rebuild only the inner leaf, provided the wall ties are still reusable.

Whilst carrying out the visual survey, attention should be given to the need for carrying tests on the structural materials to ascertain their resid- ual strengths. The testing methods used may either be non-destructive or involve the taking of samples from damaged portions on the structure, together with control specimens from undamaged areas.

BS Part , and then relate the measured strength to an equivalent cube strength using appro- priate empirical formulae. Great care is needed with the use of cores to assess residual strengths as it is necessary to attempt to extract cores free from any reinforcement, although the presence of reinforcement can be allowed for in assessing equivalent strengths. To aid the assessment of loss of strength, it is useful if at all possible to obtain the original cube or cylinder control test records when the structure was built. It is also useful if any colour changes in the concrete along the length of the core are noted, as this can help assess the residual strength of parts of the structure where it may not be possible to extract cores.

There are a series of non-destructive test methods available discussed below , although they all have problems. In the former case, it is necessary to be able to gain access to both sides of a member, together with the further limitation that the thickness cannot exceed about mm. In the latter case, the surface must be good enough to allow a series of readings to be taken and that a sim- ilar procedure is used for the reference value. Provided reference values of both the pulse velocity and strength are known, then it is possible to estimate the loss in strength if the loss in UPV is known.

Benedetti suggests that a linear degradation model of elastic modulus with temperature is adequate. The method does rely on a relatively undamaged surface. It also, however, requires very specialist equipment that may not be readily available. Cements with pulverized fuel ash PFA or ground granulated blast-furnace slag GGBS do not show any or very little change in response when heated.

More recently it has been demon- strated Short, Purkiss and Guise, ; Short and Purkiss, that it is possible to quantify the relationship between crack density and temper- ature reached due to heating Fig.

• Shop by category?
• Environmental Contamination and Remediation Practices at Former and Present Military Bases.
• SteelConstruction.info;
• Description:!
• From Short, Purkiss and Guise, It is not clear what the effect of a temperature gradient along the specimen would have on the results. For reinforcement, similar techniques are available to structural steel. It should, however, be noted that where specimens are taken from either tensile steel in beams or compressive steel in columns, the elements or structure must be propped since removal of the specimen will reduce the strength of the member. It may be possible to remove samples from shear links at the mid-point of a beam or a column without propping.

Care should taken in removing test specimens in that the damaged structure is not further weakened, and that again any necessary propping should be used. The second is to use non-destructive tests of which the most suitable is a hardness indentation test usually measuring the Brinell hardness. There is a direct, sensibly linear, relationship between the Brinell hardness number BHN and tensile strength Fig. It is important that care is taken in using this test since a number of results are needed before the strength estimates are statistically reliable.

Often, it should be noted that combination of these two approaches will be needed. Effectively, any strength assessment of an element of a struc- ture can be undertaken using the same basic approaches as outlined in previous chapters for the assessment of structural performance at elevated temperatures. Typical strength data for normal strength concrete from Malhotra , Abrams and Purkiss , are plotted in Fig. From the plotted data, it may be observed that older, historical concretes appear to give a worse performance than more modern concretes.

Chan, Peng and Anson and Poon et al. The residual strength of concrete is lower than that strength measured at elevated temperatures as there is further degra- dation on cooling caused by differing thermal properties between the aggregate and the cement matrix.

However, it is not usual to take account of any pre-load applied to spec- imens as it is conservative not to. The concretes with silica fume behaved better than those without. Poon et al. The effect of water immersion on specimens heated for 24 h was negligible com- pared to specimens tested dry.

For concretes of strength around 50 MPa, the effect on the compressive residual strength is that the addition of silica fume increases the resid- ual strength at a given temperature but that the pattern does not appear consistent Saad et al. It is thought that the peak is due to the presence of PFA in the mix. The effect is also noted by Short, Purkiss and Guise The variation in residual strength between these temperatures is sensibly linear. Bessey and Ahmed, Al-Shaikh and Arafat report visually observed colour changes in heated concrete.

The application of colour image analysis techniques can overcome this problem. By determining the change in hue when con- crete is heated, there is an obvious change in the frequency of occurrence of red Short, Purkiss and Guise, The other primary colours yellow and green appear to have little impact Fig. Some sources of siliceous aggregates may not produce any colour change. The results from the work of Bessey on siliceous aggregates and Ahmed, Al-Shaikh and Arafat on limestone aggregates are given in Table Absence of or a different colour change to those noted above should be treated with care.

Sources: Bessey Building Research Establishment: Crown Copyright and Ahmed, Al-Shaikh and Arafat by permission Thomas Telford Publications Following the structural assessment when it determined that the resid- ual structure is either strong enough to carry the imposed loads or that only minor strengthening is required, attention must be given to methods of repair.

However, before these are chosen, the economics of the situation must be considered. Thus, estimates should be prepared of the cost and duration of both repair and demolition and rebuild must be made. It is likely that if only minor repairs are required, albeit with some minor demolition and replacement, repairing the structure will be much more economic. Any intumescent paint systems will certainly need renewing. For further information on the reinstatement of steel structures reference should be made to Smith et al. In any other case replacement of either or both leaves is likely to be necessary.

Timber structures will generally need total replacement. Concrete structures generally provide the greatest scope for repair and strengthening. There are a large number of choices available to the engineer in such cases. It is only intended here to give an overview of the situation.

More detailed guidance is given in a Concrete Society Report Where the structure needs strengthening, it is essential that the new sections of the structure are not only capable of carrying the forces within the new sections, but must also be capable of transmitting the forces from the existing sections of the structure. Repair can be effected by concrete spraying gunite , resin repairs or overcladding. For gunite repairs, it is essential that all exposed concrete faces are thoroughly cleaned to ensure that the gunite bonds fully to the existing concrete.

It will often be necessary to place very light mesh within the depth of the repair to aid integrity unless the repaired area is very small Fig. Resin repairs are usually only applicable to lightly damaged areas where spalling is shallow. There are however additional aspects that warrant consideration Purkiss, b. In both cases, the asbestos will need specialist handling before demolition can start. This is due to the possibility of alternative failure mechanisms or second order effects becoming far more critical.

3rd Edition

In each case lateral restraint should be provided by stairwells and lift shafts. There is also a set of design guides from BRE due in the near future. Fire safety engineering still has a very promising and potentially rewarding future provided it is not fettered by prescriptive or legislative rules which prevent the engineer from taking properly argued engineer- ing judgements. Fire safety engineering design of structures must be considered of equal importance within the overall process of design as either the conven- tional ambient structural design, whether at ultimate or serviceability limit states, or the effects of other accidental actions such as earthquake or explosion.

References Abrams, M. Ahmed, G. Ahmed, A. Magazine of Concrete Research, 44, — Aldea, C. Phan, N. Carino, D. Duthinth and E. Garboczi , Gaithersburg Feb. Ali, F. Magazine of Concrete Research, 53 3 , — Ali, H. Engineering Structures, 26, — Allen, B. New Steel Construction, 14 1 , 24—8. Anderberg, Y. Fire Safety Journal, 13, 17— What Next? Gamberova, R. Felicetti, A. Meda and P. Fire Prevention, 1— Fire Prevention, , 13— Fire Prevention, , 12— Anon Fire tragedy at football stadium.

Fire Prevention, 81, 5—6. Anon News scene. Fire Prevention, , 5. Fire Prevention, , 5—6. Fire Prevention, , New Civil Engineer, Sept. Anon Goodnight Vienna. Fire Prevention, , 40—1.