17th March 2008 Section 3.4.4 amended
Draft Technical Report ??
Assessment, design and repair of fire-damaged concrete...
127 downloads
2252 Views
876KB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
17th March 2008 Section 3.4.4 amended
Draft Technical Report ??
Assessment, design and repair of fire-damaged concrete structures
Final draft
March 2008
1
17th March 2008 Section 3.4.4 amended
Contents Members of the Working Party Acknowledgements List of Figures List of Tables Foreword 1
2
3
Introduction 1.1 Scope 1.2 Process 1.3 Health and safety Assessment of damage 2.1 Objectives and methodology of assessment 2.2 Materials 2.2.1 Effects of high temperature on concrete strength and elastic modulus 2.2.2 Mineralogical changes in concrete 2.2.3 Cracking of concrete in fires 2.2.4 Spalling of concrete in fires 2.2.5 Residual thermal movement cracks 2.2.6 High-alumina cement 2.2.7 Reinforcing and prestressing steel 2.2.8 Degradation of other materials 2.3 Testing of fire damaged reinforced concrete 2.3.1 On-site inspection 2.3.2 Non-destructive testing (NDT) 2.3.3 Petrographic examination 2.3.4 Thermoluminescence tests 2.3.5 Core test 2.3.6 Tests on samples of reinforcement 2.3.7 Other laboratory tests 2.4 Assessment of fire damaged structures 2.4.1 Introduction 2.4.2 Testing 2.4.3 Assessment of fire severity 2.4.4 Heat transfer 2.5 Presentation of data Design 3.1 Design philosophy 3.1.1 Objectives 3.1.2 Building regulations 3.1.3 Codes of practice 3.1.4 Design assumptions 3.2 Structural analysis and member design 3.2.1 Structural analysis 3.2.2 Element design 3.3 Repair criteria
2
17th March 2008 Section 3.4.4 amended
4
3.3.1 Reduced material strengths 3.3.2 Residual strength factor 3.3.3 Bond strength 3.3.4 Bar size and spacing 3.3.5 Shear reinforcement 3.4 Member design 3.4.1 General 3.4.2 Beams and slabs – bending 3.4.3 Beams – shear 3.4.4 Columns 3.4.5 Walls 3.5 Design output 3.5.1 Demolition and construction sequence drawings 3.5.2 Key plans 3.5.3 Design details 3.5.4 Specifications 3.5.5 Design calculations 3.5.6 Method statements 3.6 Load tests Repair methods 4.1 General 4.2 Health and safety 4.3 Quality control 4.4 Surface cleaning 4.5 Breaking out 4.6 Reinforcement 4.6.1 Bar size and spacing 4.6.2 Connecting reinforcement 4.7 Mortar 4.8 Flowable micro-concrete and concrete 4.9 Sprayed concrete 4.9.1 General 4.9.2 Health and Safety 4.9.3 Substrate preparation 4.9.4 Layer thickness 4.9.5 Surface finishing 4.9.6 Curing 4.9.7 Repair details 4.10 Resins 4.11 Strengthening with fibre composites
References Further reading Appendix A A1 A2
Case studies Fires in buildings Fires under bridges
Appendix B
Worked examples
3
17th March 2008 Section 3.4.4 amended
B1 B2 B3 B4 Appendix C C1 C2 C3
Introduction Example 1 – Continuous slab Example 2 – Simply supported tee beam Example 3 – Axially loaded column Historical information Design codes Specification and strength of historic concrete Reinforcement C3.1 Early reinforcement systems C3.2 Standards and strengths C3.3 Detailing symbols
4
17th March 2008 Section 3.4.4 amended
Members of the Working Party Full Members Florian Block John Clarke Brian Cole Susan Deeny Alexander Heise Jeremy Ingham Nigel Pierce
Buro Happold The Concrete Society (Secretary) Buro Happold (Chairman) Edinburgh University Arup Fire Halcrow Group Limited CRL Surveys
Corresponding Members Simon Bladon CRL Surveys Pal Chana British Cement Association Stuart Matthews Building Research Establishment Ganga Prakhya Sir Robert McAlpine
Acknowledgements The Concrete Society is grateful to the following for the provision of photographs: Buro Happold: [To be checked.] Concrete Repairs Limited: Figures 16, 21–24 Jeremy Ingham: Figures 1, 2, 4, 5, 11, 15, 17–19.
List of Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Figure 12 Figure 13 Figure 14 Figure 15
The interior (left) and exterior (right) of a concrete framed structure shortly after a major fire during construction. View of the same structure as Figure 1.1 after repair of fire damage. Typical effect of heat upon the compressive strength of dense aggregate concrete after cooling. Appearance of flint aggregate concrete cores which have been heated for ½ hour (upper row) and 2 hours (lower row), at the temperatures indicated. View of floor below the fire showing thermal expansion cracks on the slab soffit. Surface crazing. Explosive spalling. ‘Sloughing off’. Spalling of a slab soffit owing to fire-damage of embedded plastic reinforcement bar spacers. Yield strength of reinforcing steels at room temperature after heating to an elevated temperature. Buckled bars. Tensile tests on untreated 0.76% carbon steel wire at high temperatures. Temperature effects upon relaxation of untreated cold-drawn prestressing wire. Ultimate strength of prestressing steels at room temperature after heating to an elevated temperature. Melting of aluminium formwork supports indicating that the fire reached temperatures in excess of 650°C. 5
17th March 2008 Section 3.4.4 amended
Figure 16
Figure 20 Figure 21 Figure 22 Figure 23 Figure 24 Figure 25 Figure 26 Figure 27
A fire-damaged reinforced concrete slab soffit showing pink/red discolouration of flint aggregate particles. Technicians diamond drilling core samples through the full thickness of a firedamaged concrete floor slab. View of spalled and discoloured fire-damaged concrete slab soffit, showing the location of a core sample (centre) that was taken to aid determination of the depth of fire damage. A photomicrograph of fire-damaged concrete seen through the optical microscope. Typical section of key diagram classification. Breaking out small area of concrete using hand-held equipment. Breaking out using hydro-demolition. Hand-applied mortar repair. Sprayed concrete application. Sprayed concrete repairs to beams. Sprayed concrete repairs to columns. Sprayed concrete repairs to floor slabs.
Figure B1 Figure B2 Figure B3 Figure B4 Figure B5 Figure B6 Figure B7 Figure B8 Figure B9 Figure B10 Figure B11 Figure B12 Figure B13 Figure B14 Figure B15 Figure B16
Original slab profile. Damaged slab. Temperature profiles. Repaired section. Original beam profile. Damaged beam profile. Temperature profiles. Temperature profile at corner. Support of added main bars. Anchorage of shear links. Repaired section. Original column profile. Damaged column profile. Temperature profiles. Temperature profile at corner. Repaired section.
Figure 17 Figure 18 Figure 19
List of Tables Table 1 Table 2 Table 3 Table 4 Table 5 Table 6 Table 7 Table 8 Table 9 Table 10 Table 11
Stages in the assessment and repair process. Mineralogical and strength changes to concrete caused by heating. Assessment of temperature reached by selected materials and components in fires Notional rate of charring for the calculation of residual section. A guide to the selection of test methods for fire-damaged reinforced concrete. Cycle of effects upon reinforced concrete structures. An example of a visual damage classification scheme for reinforced concrete elements. Initial repair classification. Typical section of schedule for damage classification shown in Figure 20. Features of methods of breaking out concrete. Advantages and disadvantages of sprayed concrete 6
17th March 2008 Section 3.4.4 amended
Table C1 Table C2 Table C3
The development of design codes. Reinforcement standards and associated strengths. Detailing symbols.
7
17th March 2008 Section 3.4.4 amended
Foreword Concrete has good inherent fire-resistant and concrete structures are generally capable of being repaired after a fire, even a severe one. The initial guidance on assessment and repair was published by The Concrete Society in 1978 as Technical Report 15, Assessment of firedamaged concrete structures and repair by gunite(1). In the late 1980s The Society was concerned that the guidance should remain useful and a Working Party was set up to update TR 15 and to include methods of repair other than gunite (sprayed concrete), which by then had its own Code of Practice. This led to the publication in 1990 of Technical Report 33, Assessment and repair of fire-damaged concrete structures(2). In 2007 The Society again reviewed the guidance given in TR 33 and concluded that much of it was still sound and that the Technical Report was widely used. However, there was a need to bring the material in the document into line with current Standards and repair techniques. The emphasis in TR 33 was still on the use of sprayed concrete, with little mention of other repair methods. In addition, assessment techniques, such as petrography, and analytical methods had advanced significantly. A small working party was formed, which prepared the present Technical Report. The emphasis of this report is on methods for assessing a concrete structure following a fire and hence for determining the extent of the required repairs. The design approaches used to assess the strength of repaired elements, illustrated by design examples, are in accordance with the relevant Eurocodes(3, 4, 5). The chapter on repairs is somewhat more limited than in previous versions of the Technical Report as the working party considered that techniques are common to all concrete repairs, irrespective of the cause of the damage, and not simply to the repair of fire-damaged concrete structures. Finally the report includes summaries of a number of case studies of the assessment and repair of structures damaged by fire.
8
17th March 2008 Section 3.4.4 amended
1
Introduction
Concrete is inherently fire-resistant and concrete structures are generally capable of being repaired after a fire, even a severe one. In the 1980s, Tovey and Crook(6, 7) summarised the information gathered from over 100 fire-damaged structures. They concluded that, almost without exception, the structures performed well during and after the fire. Most of the structures were repaired and returned to service; when structures were demolished and replaced, it was generally for reasons other than the damage sustained during the fire. Some more recent case studies are given in Appendix A of this Technical Report, which outline the damage caused by the fire and the subsequent investigation. The examples include residential buildings, commercial buildings and bridges. In all but one case the structure was successfully repaired. 1.1
SCOPE
The emphasis of this report is on methods for assessing a concrete structure following a fire and hence for determining the extent of the required repairs. The design approaches used to assess the strength of repaired elements, illustrated by the design examples in Appendix B, are in accordance with the relevant Eurocodes(3, 4, 5). In addition to structural damage, there may be smoke damage to partitions, electrical and mechanical systems etc. Although the associated costs of cleaning or replacing such systems can be significant, they are not considered in this report. The focus of this report is on fires in reinforced concrete buildings, including multi-storey structures, warehouses and factories, but the principles are equally applicable to civil engineering structures, such as bridges. However, tunnels are specifically excluded as an assessment of their performance will require specialised geotechnical input, which is beyond the scope of this report. There is a major difference between designing a structure to withstand a fire, allowing for safe evacuation and fire fighting, and assessing the extent of damage caused by a fire so that repair methodologies can be proposed. While designing structures is predicting performance during a future event, assessing structures is determining its residual strength after such an event. Hence, the focus in the latter case and in this report is on methodologies to measure on site the residual strength and deformations and to obtain evidence of the temperatures reached during the fire. Calculation methodologies are presented that may assist during the evaluation process, but the working party felt that any assessment needs to be based mainly on an on-site evaluation of the fire damaged structure, which is supplemented as necessary by laboratory testing, examination or numerical assessment. In all cases, it is important that the assessment work is carried out by a competent person, who is aware of the limits of applicability for any methodology and whether special considerations for certain construction methods are required. The competent person needs to be aware that material properties and calculation methodologies presented in Eurocode 2 may not be applicable to the specific situation, since effects such as cooling of the structure or restraint and residual stresses need consideration after a fire event. This means that although the structure may have served its purpose according to Building Regulations and allowed for safe evacuation and fire fighting, considerable effort may be required to strengthen the structure for future occupation after a fire.
9
17th March 2008 Section 3.4.4 amended
A brief chapter on repair techniques is included, which makes reference to more detailed guidance. The working party considered that techniques are common to all repairs, irrespective of the cause of the damage, and not simply to the repair of fire-damaged concrete structures. Finally appendices to the report includes summaries of a number of case studies of the assessment and repair of structures damaged by fire, worked examples and historical information on design and material properties given in British Standards and other documents. 1.2
PROCESS
After a fire the focus is on immediate measures for securing public safety. In the UK, the fire brigade will usually secure the building; if they have any doubts about the stability of the structure they will call in the local Building Control Officer to make an assessment. After a serious fire the Building Control Officer may require parts of the structure to be demolished or stabilised before anyone else can enter. If part of a damaged building is to remain occupied while repairs are carried out elsewhere, it will be necessary to establish that the remaining escape routes, fire separation, fire protection systems etc are adequate throughout. The responsible person, as defined in the Regulatory Reform Order(8), is required to assess whether the building is deemed safe. The Fire and Rescue Authority can request that compliance with the requirements of the fire safety order is demonstrated. The fire authority has the powers to take enforcement action where requirements of the order are breached or where a serious risk to life exists. Often the authority will also be notified by the police, who may investigate arson. Finally, the insurers may commission an investigation of the damage. The insurer will often have a major interest in finding the most cost effective solution for repairing the structure. When a fire has occurred, the requirements are generally for an immediate and thorough appraisal to be carried out, with clear objectives. Such an appraisal should begin as soon as the building can be entered safely and generally before the removal of debris. The competent person needs to establish whether the building is safe or not and propose propping of the structure, if required. Propping might not be required if the structure is too damaged for repair and demolition is proposed. The spalled and discoloured or blackened concrete surfaces and exposed reinforcement generally apparent after a severe fire often present a picture which suggests widespread and significant damage; Figure 1 shows the aftermath of a fire in a concrete multi-storey building under construction (see Ingham(9)). However, in practice the damage may be much less severe. It is necessary to be strictly objective and to consider the effect of high temperature upon the properties of the materials concerned. This is considered in Chapter 2.
10
17th March 2008 Section 3.4.4 amended
Figure 1: The interior (left) and exterior (right) of a concrete framed structure shortly after a major fire during construction. Remedial measures are likely to be based on a comparison of the cost of removal and the need for replacement. Experience shows that, following detailed appraisal, reinforced concrete structures can nearly always be repaired by means of a selection of suitable techniques. In the case of severe damage, replacement of some elements may be necessary. However, the actual fire resistance of a concrete structure is frequently well above minimum requirements due to the structural continuity present in most buildings. These reserves of strength may enable the structure to survive severe fires and still be reinstated. Reinstatement by repair will usually be preferable to demolition and rebuilding as it may require less capital expenditure, and may also produce a direct saving as a result of earlier reoccupation. As an example, Figure 2 shows the completed 10-storey concrete framed residential building mentioned earlier that was extensively damaged by a fire during construction (see Figure 1) and subsequently repaired, allowing earlier occupation than if it had been demolished and reconstructed(9).
Figure 2: View of the same structure as Figure 1 after repair of fire damage.
11
17th March 2008 Section 3.4.4 amended
Table 1 is intended as a simple guide to the process of assessment and repair; reference is made to the main parts of the report dealing with the various activities. The main contractor and, if appropriate, a repair specialist should be appointed as early as possible so that they can participate in the preparation of the programme and the strategy for the site work. Table 1: Stages in the assessment and repair process. Stage 1
2
3
4
5
1.3
Activities Carry out preliminary inspection of the structure. Take immediate steps to secure public safety and the safety of the structure; it may be necessary to prop members that are in a critical condition. Carry out on-site assessment of the structure to determine the extent of damage (by visual inspection, breakouts and/or nondestructive testing). Decide which elements require cosmetic repairs only (e.g. cleaning and repainting) and which, if any, will require further assessment under Stage 3. (At this stage the decision may be taken to demolish and rebuild parts or all of the structure.) Break out spalled concrete to determine depth of fire damage. Carry out laboratory testing of concrete and reinforcement samples to determine their residual strengths and confirm depth of fire-damage, supplemented by thermal modelling where appropriate. Possibly carry out dimensional surveys to determine the extent of deflections of beams and slabs and lateral movements of columns. Determine structural capacity of members that are to be repaired, using reduced residual material properties, and hence determine additional concrete and reinforcement required to reinstate original capacity. The opportunity may be taken to upgrade parts or all the structure to modern standards. (As at Stage 2, the decision may be taken to demolish and rebuild parts of the structure if the quantity of additional material required makes repair uneconomic or impractical.) Select appropriate repair methods and carry out work to restore the structure to its original capacity.
Reference Section 1.3
Sections 2.3 and 2.5 and Table 7
Sections 2.3 and 2.4
Chapter 3 and Appendix B
Chapter 4
HEALTH AND SAFETY
All repair work will be subject to the requirements of the Construction (Design and Management) Regulations(10), generally known as the CDM Regulations. The Regulations require a planning co-coordinator to be appointed if the construction work is longer than 30 days duration or requires more than 500 person days. All operatives should wear the appropriate personal protective equipment (i.e. safety helmet, gloves, boots etc) when carrying out repairs. Further aspects of health and safety are covered in Sections 4.2 and 4.9.2; the latter is specifically concerned with the use of sprayed concrete.
12
17th March 2008 Section 3.4.4 amended
It is essential to consider the safety of the structure at all stages, from the initial assessment phase through to the final repair. Where necessary, beams and slabs should be propped, with temporary bracing. Phased breaking out may be required in some circumstances. Special consideration is needed if the structure is to be left for some time before repair work is carried out, during which time further deterioration may occur. The obvious risk of collapse of the structure should be considered, as well as the risk to third parties, for example from falling debris. During the course of the remedial works risks, such as falling concrete during breaking out, should be assessed and appropriate actions specified to mitigate any identified events that could arise. Safe access to the area being repaired should be provided for personnel. Repair materials and equipment should be stored in a suitable location, taking account of any additional loads they may apply to the weakened structure. Temporary falsework may be urgently required to secure the structure, not just for individual members, but for the stability of the structure as a whole. Remaining applied loads and all dead loads should be calculated for all doubtful members. Special care is required to avoid the transfer of excessive loads and stresses to other members. This applies particularly where falsework is being used to relieve a column at an intermediate floor level. Relieving falsework may have to be carried through to foundation level to avoid creating excessive stresses in adjoining parts of the structure. Installation of the falsework should precede the detailed appraisal described in the following chapter. Attention is drawn to BS 5975, Code of practice for falsework(11).
13
17th March 2008 Section 3.4.4 amended
2
Assessment of damage
The aim of an assessment of a fire damaged structure is to propose appropriate repair methods or to decide whether demolition of elements or the whole structure is more appropriate. 2.1
OBJECTIVES AND METHODOLOGY OF ASSESSMENT
The total feasibility of repair is dependent on parameters such as the extent of local and global damage to the building, but also to direct losses, such as damage to the façade or mechanical and electrical (M&E) installations, and indirect losses to business, for example by relocation of people, interruption of trade and the costs of cleaning smoke and combustion products, which may include cleaning to provide acceptable air quality in future. The focus of this report is on the assessment of the damage of the concrete structure only. The respective stakeholders need to decide from their point of view if the suggested repair methodology is appropriate. A systematic approach will result in a damage classification for the various parts of the affected structure, which may be used in the selection of appropriate repair techniques. At best members may need no structural repair as they have sufficient residual strength, and at worst demolition will be required. The assessment can follow two methodologies, which can be used separately or combined depending on the nature of the fire and of the structure, as follows: 1. Test the fire damaged concrete to directly assess the concrete quality. 2. Estimate the fire severity so as to deduce temperature profiles and hence to calculate the residual strength of the concrete and the reinforcement. Following the first methodology, there are several levels and methods to test fire damaged concrete: • Visual inspection and hammer soundings • Non-destructive testing (e.g. rebound hammer, ultrasonic pulse velocity (UPV)) • Coring, sampling and subsequent laboratory testing (e.g. petrographic examination, strength testing of concrete and reinforcement samples). The second methodology involves three steps, which should be confirmed by tests: 1. Evaluation of fire severity – This can be performed based on debris or applying numerical evaluation methods, such as computational fluid dynamics. 2. Determination of temperature-profiles – This maybe performed applying numerical methods or simpler calculation techniques as provided for instance in Part 1-2 of Eurocode 2(5). 3. Assessment of residual strength of the concrete – See Section 2.2.2. With both methodologies, the result will be a damage classification, which ideally should be provided on drawings showing the actual condition of the fire damaged structure. It is advisable to assess the strength of the unaffected concrete to confirm the design assumptions. The assessment needs to provide sufficient information to finally prepare detailed drawings with instructions on how to repair the structure. 2.2
MATERIALS
14
17th March 2008 Section 3.4.4 amended
The following sections outline the material properties of concrete, reinforcing steel and prestressing wires to facilitate an understanding of the residual strength after a fire. 2.2.1
Effects of high temperature on concrete strength and elastic modulus
After cooling to ambient temperatures it has been observed that the strength of concrete may be further reduced from its strength at high temperature. Effectively during the cooling period a further loss of strength takes place because of continuing disintegration of the microstructure, see for example Schneider(12). This is one reason that a more conservative strength reduction factor to assess the residual strength of the concrete than that given in Part 1-2 of Eurocode 2 is proposed, see Figure 3.
residual strength factor %
1.00
residual strength
0.80
1 hr exposure (Bazant et.al)
0.60
0.40 2 hr exposure (Bazant et.al)
0.20
0.00 0
200
400
600
800
1000
o
temperature C
[Note: Change Bazant et al to Bazant and Kaplan.] Figure 3: Typical effect of heat upon the compressive strength of dense aggregate concrete after cooling. The stress:strain curves in the Eurocode are based on steady state as well as transient state tests. For this reason the stress:strain relationships given are solely valid for heating rates between 2 and 50K/min. Creep effects are not explicitly considered. Therefore, strictly speaking the Eurocode curves are not valid for the cooling phase, see Franssen(13). It should be noted that there is a great variation in the residual strength of concrete after cooling depending on factors, including the following: • The maximum temperature attained • Duration of heating exposure • Mix proportions • Aggregates • Conditions of loading during heating and stress level.
15
17th March 2008 Section 3.4.4 amended
Concretes containing certain synthetic lightweight aggregates, such as sintered pulverised-fuel ash are though to offer good fire resistance, provided that the concrete is dry. However, poor performance has been observed in conditions were the concrete is saturated at the time of the fire. The cement type and cement blend also influence behaviour of concrete in fire. Modern concretes often include a pozzolanic mineral addition in the binder such as fly ash (pulverised-fuel ash or pfa) ground granulated blastfurnace slag (ggbs). Their use is generally thought to give a slight improvement in heat resistance owing to the fact that they reduce the amount of calcium hydroxide (portlandite) within the hydrated binder. However, in the case of microsilica, its use significantly increases the risk of spalling due to the fact that it leads to very low permeability to the hardened concrete. Figure 2.1 also shows data from Bazant and Kaplan(14) giving the residual strength of concrete samples exposed to the same temperatures, but for different exposure times. It can be seen that a longer exposure to higher temperatures results in lower residual strength. In simple terms, for temperatures up to 300°C, the residual compressive strength of structuralquality concrete is not significantly reduced, while for temperatures greater than 500°C the residual strength may be reduced to only a small fraction of its original value. Consequently, the design methodology in the Eurocode discounts the strength of concrete exposed to temperatures higher than 500°C. On the basis of the uncertainties regarding the assessment of the residual strength of concrete discussed above, this report recommends a more conservative approach, discounting the residual strength for concrete exposed to temperatures above 300°C.
2.2.2
Mineralogical changes in concrete caused by heating
Heating concrete causes a progressive series of mineralogical changes that can be investigated by petrographic examination to determine the maximum temperature attained and deduce the depth to which the concrete has been damaged. A compilation of the changes undergone by Portland cement concrete as it is heated is presented in Table 2, which is based on Ingham(9). Table 2: Mineralogical and strength changes to concrete caused by heating. Heating temperature 70–80°C 105°C 120–163°C 250–350°C
Changes caused by heating Mineralogical changes Strength changes Dissociation of ettringite, Ca6Al2(SO4)3(OH)12·26H2O causing its depletion in the cement matrix. Loss of physically bound water in aggregate and cement matrix Minor loss of commences causing an increase in the capillary porosity and minor strength possible microcracking. (<10%) Decomposition of gypsum, CaSO4·2H2O causing its depletion in the cement matrix. Pink/red discolouration of aggregate caused by oxidation of iron Significant loss of compounds commences at around 300ºC. strength commences Loss of bound water in cement matrix and associated degradation at 300ºC becomes more prominent.
16
17th March 2008 Section 3.4.4 amended 450–500°C
573°C 600–800°C
800–1200°C
Dehydroxylation of portlandite, Ca(OH)2 causing its depletion in the cement matrix. Red discolouration of aggregate may deepen in colour up to 600ºC. Flint aggregate calcines at 250-450ºC and will eventually (often at higher temperatures) change colour to white/grey. Normally isotropic cement matrix exhibits patchy yellow/beige colour in cross-polarised light, often completely birefringent by 500ºC. Transition of α-to β-quartz , accompanied by an instantaneous increase in volume of quartz of about 5% in a radial cracking pattern around the quartz grains in the aggregate. Decarbonation of carbonates; depending on the content of carbonates in the concrete, e.g. if the aggregate used is calcareous, this may cause a considerable contraction of the concrete due to release of carbon dioxide, CO2; the volume contraction will cause severe microcracking of the cement matrix.
Concrete not structurally useful after heating in temperatures in excess of 550– 600ºC
Complete disintegration of calcareous constituents of the aggregate and cement matrix due to both dissociation and extreme thermal stress, causing a whitish grey colouration of the concrete and severe microcracking. Limestone aggregate particles become white. Concrete starts to melt. Concrete melted.
1200°C 1300–1400°C
The colour of concrete can change as a result of heating, which is apparent upon visual inspection. In many cases a pink/red discolouration occurs above 300°C, which is important since it coincides approximately with the onset of significant loss of strength due to heating. Any pink/red discoloured concrete should be regarded as being suspect and potentially weakened. In addition to the maximum temperature reached, the actual heat-induced concrete colour changes depend on the mineralogy of aggregate present in the concrete. Colour changes are most pronounced for siliceous aggregates and less so for limestone, granite and sintered pulverised-fuel ash lightweight aggregate (which shows very little colour change). Striking colour changes are produced by flint (chert); Figure 4, taken from Ingham(9), illustrates the colour changes of flint aggregate concrete.
Figure 4: Appearance of flint aggregate concrete cores which have been heated for ½ hour (upper row) and 2 hours (lower row), at the temperatures indicated.
17
17th March 2008 Section 3.4.4 amended
The pink/red colour change is a function of (oxidizable) iron content and it should be noted that as iron content varies, not all aggregates undergo colour changes on heating. Concrete which has not turned pink/red is not necessarily undamaged by fire. Also, due consideration should always given to the possibility that the pink/red colour may be a natural feature of the aggregate rather than heat-induced. In concrete containing aggregate that does not undergo colour change or is naturally pink/red, other mineralogical and physical indicators should be used for determining the presence of fire-damage. It should also be noted that the cement paste can also be discoloured by carbonation and this should not be confused with heat-induced discolouration. This is discussed further in Section 2.3.1. 2.2.3
Cracking of concrete in fires
At high temperatures, the unrestrained thermal expansion of steel reinforcement is greater than that of most concretes. This can lead to bursting stresses and cracking around the steel in heavily reinforced members. Experience suggests that such cracks concentrate at positions where, incipient cracks due to drying shrinkage, flexural loading, etc. were present (see Figure 5). In addition, the thermal incompatibility of aggregates and cement paste causes stresses which frequently lead to cracks, particularly in the form of surface crazing, see Figure 6.
Figure 5: View of floor below the fire showing thermal expansion cracks on the slab soffit. [Photograph required.]
Figure 6: Surface crazing. 2.2.4
Spalling of concrete in fires
Spalling involves the breaking off of layers of concrete from the exposed surface at high and rapidly rising temperatures. Spalling is complex and there are many parameters influencing
18
17th March 2008 Section 3.4.4 amended
the process, see for example fib Bulletin 38(15). It is believed that the main process involved in spalling is vapour pressure, which is released from physically and chemically bound water at the beginning of the heating process through concrete pores. Due to the small capacity of the pores pressure builds up, which eventually may lead to spalling. Whether the tensile stresses from the vapour pressure within the pore spaces exceed the tensile strength of the cement matrix resulting in spalling is dependent on the amount of moisture, the rate of heating, permeability, porosity and pore distribution, as well as inherent tensile stresses of the structure. Three main types of spalling can be recognised. Explosive spalling (see Figure 7) occurs early in the fire (typically within the first 30 minutes) and proceeds with a series of disruptions, each locally removing layers of shallow depth. Aggregate spalling, also occurring in the early stages, involves the expansion and decomposition of the aggregate at the concrete surface causing pieces of the aggregate to be ejected from the surface.
Figure 7: Explosive spalling. Sloughing off or corner spalling (see Figure 8), occurs in the later stages of the fire when temperatures are lower. It occurs chiefly in beams and columns, as tensile cracks develop at planes of weakness such as the interface between the reinforcement and the concrete. As this type of spalling occurs in the advanced stages, the concrete is already significantly weakened and thus there are no implications for structural performance. Due to the lateness of the onset of this type of spalling, the interior concrete and the reinforcement are unlikely to have been subjected to high temperatures, even though the latter is often exposed.
Figure 8: ‘Sloughing off’.
19
17th March 2008 Section 3.4.4 amended
Spalling of concrete surfaces can be caused by the deterioration of materials embedded in concrete other than reinforcement bars. Ingham and Tarada(16) indicate that plastic reinforcement bar spacers are one of the more commonly encountered examples of this (see Figure 9). The deterioration temperatures of materials other than concrete and reinforcement are given in Section 2.2.8
Figure 9: Spalling of a slab soffit owing to fire-damage of embedded plastic reinforcement bar spacers. Further loss of concrete may also take place after the fire has been extinguished and as the concrete cools. In such cases this concrete has remained in place long enough for the rise in temperature of internal concrete and reinforcement to be restricted. 2.2.5
Residual thermal movement cracks
During a fire concrete structures undergo deformations due to a number of reasons. Generally heating causes expansion and this can push columns, particularly edge columns, out of plumb. Differential temperatures through concrete elements, particularly slabs, can lead to thermal bowing. As discussed in previous sections the increase in material temperatures leads to a reduction in stiffness which again leads to increased deflection under gravity loads. Experience shows that these deflections are not fully recovered once the fire is extinguished. For example, as described by Chana and Price(17), the concrete building at Cardington underwent a fire test involving only four structural bays; the perimeter columns were all pushed out and in the worst case the residual horizontal deflection at slab level was 67mm. Such residual deflections should be recorded. These may have an important impact on the future serviceability of the structure. For example if a deflected slab compromises floor to ceiling heights it may require replacement even if it is structurally adequate. Similarly a column with a large out of plumb due to fire effects may attract substantially greater bending
20
17th March 2008 Section 3.4.4 amended
moments than assumed in normal design and this will need to be considered in any structural assessment. 2.2.6
High-alumina cement concrete
Special care should be taken when fire has damaged a structure containing high-alumina cement (HAC), also known as calcium aluminate cement, as the concrete may already have been weakened by the process of HAC conversion. In addition to the specific fire damage investigations it is recommended that the condition of the HAC concrete is checked in accordance with BRE guidance(18). It should be remembered that fire fighting water could potentially initiate or worsen mechanisms of HAC concrete deterioration such as sulfate attack or alkaline hydrolysis. 2.2.7
Reinforcing and prestressing steel
The effect of elevated temperatures and subsequent cooling on the residual strength of steel has been researched in detail, for example by Stevens(19) and Holmes et al(20). Significant loss of strength may occur while the steel is at high temperature and this is usually responsible for any excessive residual deflections. However, recovery of yield strength after cooling is generally complete for temperatures up to 450°C for cold worked steel and 600°C for hot rolled steel. Above these temperatures, there will be a loss in yield strength after cooling. The actual loss in strength depends on the heating conditions and type of steel but the conservative values given in Figure 10, for temperatures up to 700°C, will be sufficient for most purposes. Values above 700°C are not given due to the additional variations in properties that can occur due to phase changes in the steel. Therefore where the temperature of the steel has exceeded 700°C or where determination of the strength is critical to assessment, the matter should be discussed with the reinforcement manufacturer if known or, alternatively, tests carried out on samples taken from the member. Loss in ductility may occur after exposure to particularly high temperatures.
Figure 10: Yield strength of reinforcing steels at room temperature after heating to an elevated temperature. The initial survey should identify what type of steel was used in the original structure, i.e. cold worked or hot rolled. For older structures, Appendix C gives some information on the types and strengths of bars that were available at various times. For other materials, such as 21
17th March 2008 Section 3.4.4 amended
stainless steel reinforcement, it will be necessary to seek specialist advice. If the production process cannot be identified or for the very early (pre 1920s) proprietary reinforcement systems mentioned in Appendix C, it will be necessary to carry out tests on samples removed from the structure. An on-site indentation (hardness) test might be considered as an indirect means of measuring strength. However the hardness of the surface of the reinforcement may be different from that at the centre of the bar due to the possible effects of quenching that result from fire fighting. Any such tests should therefore be used with caution. Buckling of reinforcing bars often occurs as a result of compressive stress induced at high temperatures by restraint against thermal expansion. Figure 11 shows bars that have buckled out from the soffit of a slab and are no longer bonded to the concrete in this region.
Figure 11: Buckled bars. The effect of high temperature is more critical on prestressing steel than on reinforcing steel. The strength of prestressing steel during heating is likely to be reduced to less than 50% of the normal strength when the steel temperature reaches about 400°C. In terms of re-use, a more important factor is the effect of heat upon the tension of the steel. Loss of elastic modulus in the concrete, increased relaxation due to creep and non-recoverable extension of tensions all contribute to this loss of tension. The loss in strength for untreated wire when hot is shown in Figure 12.
22
17th March 2008 Section 3.4.4 amended
Figure 12: Tensile tests on untreated 0.76% carbon steel wire at high temperatures. Also shown in the Figure is the reduction in the limit of proportionality while the steel is hot; if the stress at which plastic elongation occurs is reduced at a particular temperature, the prestressing force will be similarly and immediately reduced unless it is already below the new limit of proportionality. Loss of prestress due to relaxation then proceeds from this new value of initial stress, as indicated in the diagrams of Figure 13. These two figures will be useful when assessing the residual prestress and when determining performance of a member under serviceability conditions.
Figure 13: Temperature effects upon relaxation of untreated cold-drawn prestressing wire. The residual strength of prestressing steels on cooling from an elevated temperature is shown in Figure 14 which is derived from Holmes et al(20). This will be useful when assessing the residual strength of a member under ultimate conditions.
Figure 14: Ultimate strength of prestressing steels at room temperature after heating to an elevated temperature. The maximum temperature reached in the steel, together with the temperature distribution and duration, are therefore clearly far more critical than in the case of reinforced concrete. In addition, it is necessary to consider factors such as the effect of temperature when hot and after cooling upon the elastic modulus and creep of the concrete, together with the effects of expansion and any restraints against expansion which may be present.
23
17th March 2008 Section 3.4.4 amended
For these reasons, precise guidance on the assessment and repair of prestressed concrete is beyond the scope of this Technical Report, but it should be noted that some fire-damaged prestressed concrete structures have been satisfactorily assessed and repaired, see Appendix A.
2.2.8
Degradation of other materials
The condition of other debris may be useful in determining the history and characteristics of fire (see for example Figure 15, which shows aluminium formwork supports that have melted during a fire). Table 3 gives a guide to the approximate temperatures which cause various materials and components to degrade in building fires.
Figure 15: Melting of aluminium formwork supports indicating that the fire reached temperatures in excess of 650°C.
Table 3: Assessment of temperature reached by selected materials and components in fires Substance
Typical examples
Conditions
Paint Polystyrene Polyethylene Polymethyl methacrylate PVC
Cellulose
Thin-wall food containers, foam, light shades, handles, curtain hooks, radio casings Bags, films, bottles, buckets, pipes Handles, covers, skylights, glazing Cables, pipes, ducts, linings, profiles, handles, knobs, house ware, toys, bottles (Values depend on length of exposure to temperature.) Wood, paper, cotton
24
Deteriorates Destroyed Collapse Softens Melts and flows Shrivels Softens and melts Softens Bubbles Degrades Fumes Browns Charring Darkens
Approximate temperature (°C) 100 150 120 120–140 150–180 120 120–140 130–200 250 100 150 200 400–500 200–300
17th March 2008 Section 3.4.4 amended Wood Solder lead
Plumber joints, plumbing, sanitary installations, toys
Zinc
Sanitary installations, gutters, downpipes
Aluminium and alloys
Fixtures, casings, brackets, small mechanical parts
Glass
Glazing, bottles
Silver
Jewellery, spoons, cutlery
Brass
Locks, taps, door handles, clasps
Bronze
Windows, fittings, doorbells, ornamentation Wiring, cables, ornaments Radiators, pipes
Copper Cast iron
Ignites Melts Melts, sharp edges rounded Drop formation Drop formations Melts Softens Melts Drop formation Softens, sharp edges rounded Flowing easily, viscous Melts Drop formation Melt (particularly edges) Drop formation Edges rounded Drop formation Melts Melts Drop formation
240 250 300–350 350–400 400 420 400 600 650 500–600 800 900 950 900–1000 950–1050 900 900–1000 1000–1100 1100–1200 1150–1250
An indication of the duration and severity of the fire may be obtained by examining charred timber that has remained in place throughout the fire. As a rough guide the char increases at a rate of 40mm per hour in the standard fire test. Variations from this can be seen in Table 4, which is based on Section 4.1 of BS 5268(21). From this Table, it can be seen that, if a piece of oak is found with a char depth of 45mm, it has been exposed to a fire equivalent of 1½ hours under furnace conditions. The information is less relevant if the timber had not remained in place and has fallen to the ground with the other debris.
Table 4: Notional rate of charring for the calculation of residual section. Timber
Depth of charring in 30 minutes 60 minutes
All structural species listed in Appendix A of BS 5268-2: 1989 except those listed below Western red cedar Hardwoods having a nominal density not less that 650kg/m3 at 18% moisture content
2.3
20mm
40mm
25mm
50mm
15mm
30mm
TESTING OF FIRE DAMAGED REINFORCED CONCRETE
There are a number of on-site and laboratory-based techniques available to aid in the diagnosis of reinforced concrete condition. Techniques conducted on site include visual inspection, non-destructive testing and the removal of concrete and reinforcement samples,
25
17th March 2008 Section 3.4.4 amended
which may subsequently be examined and/or tested in the laboratory. A guide to selected test methods appropriate for investigating fire-damaged concrete is provided in Table 5. Table 5: A guide to the selection of test methods for fire-damaged reinforced concrete.
2.3.1
Nondestructive
Test type
Partially destructive
Laboratory
On-site
Test location
Test method Colour changes Visual inspection Hammer soundings Rebound Hammer Ultrasonic Pulse Velocity Breakout/ drilling Petrographic examination Thermoluminescence Core Test Reinforcement test
Lateral extent of damage
9
9 9 9 9 9
9
Information gained Depth of Compressive damage strength of undamaged concrete 9 9
Tensile strength of reinforcement bars (damaged and undamaged)
9 9 9 9 9
On-site inspection
It may be sufficient to take ‘soundings’ on the damaged concrete to determine the degree of deterioration. The ‘ring’ of sound concrete and the ‘dull thud’ of weak material are readily distinguished, and this test may be successfully done with a hammer and chisel. Removing concrete with a hammer and chisel can therefore be used to determine the depth of the pink/red layer. Figure 16 shows red/pink discolouration on the soffit of a fire-damaged slab.
Figure 16: A fire-damaged reinforced concrete slab soffit showing pink/red discolouration of flint aggregate particles. Where it is difficult to assess the depth of the pink/red layer, a small-diameter core or a drilled hole can be made to determine it accurately, see Figures 17 and 18. Core samples may be required anyway for laboratory examination and/or testing.
26
17th March 2008 Section 3.4.4 amended
Figure 17: Technicians diamond drilling core samples through the full thickness of a fire-damaged concrete floor slab.
Figure 18: View of spalled and discoloured fire-damaged concrete slab soffit, showing the location of a core sample (centre) that was taken to aid determination of the depth of fire damage. As mentioned in Section 2.2.2, discolouration can occur as a result of carbonation and therefore care needs to be taken when investigating older concrete. Carbonation depth may be found by spraying a freshly broken surface with phenolphthalein indicator. If the depth of visual discolouration is beyond the layer shown by the phenolphthalein then it is clearly due to the effects of the fire. If it coincides with the layer shown by the phenolphthalein then it may be due to carbonation and not the fire. The boundary for the pink/red zone may be taken as being on the 300°C temperature profile and hence the strength loss and equivalent duration of the fire may be determined. An alternative on-site technique is the drilling resistance test, which uses a hammer drill to determine the depth of weakened concrete, see Felicetti(22).
27
17th March 2008 Section 3.4.4 amended
It is considered that hammer and chisel testing should still form the basis of investigations. Other methods are available for the assessment of strength and these are outlined in the following sections together with the comments on their particular use. 2.3.2
Non-destructive testing (NDT)
Visual inspection, hammer tapping and breakouts are the principal on-site methods of fire damage assessment. However, in certain situations there may be benefit in supplementing the normal on-site regime with some non-destructive testing (NDT). Guidance on the use of NDT methods is given in BS 6089: 1981(23) and BS 1881: Part 201(24). (Note that these Standards are in the process of being replaced by European Standards.) The NDT methods most commonly used for on-site condition assessment of fire-damaged concrete structures are the rebound test (Schmidt hammer) and the ultrasonic pulse velocity (UPV) test. The rebound test gives a measure of the surface hardness of the concrete surface. Although there is no direct relationship between this measurement of surface hardness and strength, an empirical relationship exists. Due to the need for a flat surface to test and as a large number of tests is desirable to reduce the effects of variability, the rebound hammer is not generally suitable for use on spalled surfaces, which is often the case with fire damaged concrete. The results of this test on fire-damaged concrete, even on flat surfaces, are somewhat variable and this is perhaps due to skin hardening effects that appear to occur. The apparatus is, however, commonly available and the method of test is given in Part 2 of BS EN 12504(25). The UPV test for the estimation of concrete strength is well established and, although there is no fundamental relationship between pulse velocity and strength, an accepted equation exists and the method is covered by Part 4 of BS EN 12504(26). Although an estimation of strength can be obtained by correlation, the method has perhaps a greater potential for comparing known sound concrete with affected concrete. This test also requires a flat surface and is, therefore, generally only appropriate for unspalled surfaces. The method has been found to be particularly suitable for use on the ribs of waffle and trough floors and for assessing the extent of damage of a localized fire. The UPV test can also be used to give an indication of depth of seriously weakened concrete. Most of the major testing companies have UPV apparatus.
2.3.3
Petrographic examination
Petrographic examination is the definitive technique for determining the depth of fire damage in concrete. It is performed in the laboratory by experienced concrete petrographers, using optical microscopes in accordance with ASTM C856(26). Concrete core or lump samples are subjected to visual and low-power microscopical examination. Followed this, samples are selected for thin-section preparation and more detailed examination with a high-power microscope. Petrographic examination provides a great deal of information and is offered by all of the major testing companies. A typical commercial examination would be expected to determine the following: • Type, mineralogy and approximate grading of the coarse and fine aggregates • Cement type • Presence of mineral additions and fillers • Cement content • Water:cement ratio • Air void content 28
17th March 2008 Section 3.4.4 amended
• • •
Depth of carbonation Presence of defects or deterioration Identification of deleterious reactions such as alkali-silica reaction.
Petrographic examination is invaluable in determining the heating history of concrete as it can determine whether features observed visually are actually caused by heat rather than some other factor. In addition to colour changes of the aggregate, the heating temperature can be cross-checked with changes in the cement matrix and evidence of physical distress such as cracking and microcracking. Careful identification of microscopically observed features allows thermal contours (isograds) to be plotted through the depth of individual concrete members. In the most favourable situations contours can be plotted for 105ºC (increased porosity of cement matrix), 300ºC (red discoloration of aggregate), 500ºC (cement matrix becomes wholly isotropic), 600ºC (α- to β-quartz transition), 800ºC (calcination of limestone) and 1200ºC (first signs of melting). Figure 19 shows an example of some of the microscopical features that may be observed in fire-damaged concrete. Some aggregate particles have been reddened indicating that the concrete has reached at least 300ºC at that point. Particles of flint have been calcined (brown mottled) and so have been heated to 250–450ºC. The cement matrix is bisected by numerous fine cracks (white) within the cement matrix (dark), some of which radiate from quartz grains (white) in the fine aggregate fraction. This deep cracking and cracking associated with quartz suggest that the concrete has reached 550–575ºC. Overall we can deduce that the concrete has been heated to approximately 600ºC in the area represented by the sample.
Figure 19: A photomicrograph of fire-damaged concrete seen through the optical microscope. In recent years, research has been conducted into the application of image analysis techniques to assessment of colour changes caused to concrete by heating, see for example Lin et al(27) and Short et al(28). These methods involve using computer software to analyse the colours of digital images captured from finely ground slices of concrete. Reliance on these methods to determine the depth of fire-damage has a number of drawbacks and should always be crosschecked with microscopical examinations, see Ingham(9). The limitations of petrographic examination are that it is relatively expensive and usually takes at least two weeks to complete. Also, although petrography will determine concrete condition in qualitative terms, it does not provide numerical values of concrete strength. When applied to fire-damaged concrete, petrography will usually: 29
17th March 2008 Section 3.4.4 amended
• • • •
2.3.4
Provide details of the concrete ingredients and mix proportions Identify potentially deleterious ingredients (e.g. high-alumina cement) Provide an assessment of general concrete workmanship and general concrete condition, including identification of any underlying problems Describe the effects of fire-damage and determine the depth of damage.
Thermoluminescence tests
The basis of this technique for investigation of fire damaged concrete is the measurement of the residual thermoluminescence in small samples of sand drilled from the concrete, see Placido(29) and Chew(30). A major loss of thermoluminescence occurs at around the same temperature that concrete begins to lose significant strength. This test has the advantage that only small holes are required for sampling of drilled dust on site and temperature profiles may be determined. However, it requires specialist laboratory equipment and experienced operators; the usefulness of this technique is somewhat reduced by its limited availability and relatively high cost, see Smart(31). 2.3.5
Core test
The most direct method of estimating the compressive strength of in situ concrete is by testing cores cut from the structure. The test procedure is given in Part 3 of BS EN 12390(32). Large cores should not be taken from positions where they would cause a significant loss in structural strength. 2.3.6
Tests on samples of reinforcement
The effect of elevated temperatures to the reinforcement has been considered in Section 2.2.7. To confirm the limit of deterioration, samples should be taken in the first instance from members damaged beyond repair (if any). If the tests are not satisfactorily further samples should be taken from representative elements. Obviously, the damage to the reinforcement should be evaluated when taking the samples, which need to meet requirements of test facilities with respect to their length. They should be tested for yield, elongation and tensile strength. The results should be compared with the relevant British Standard (see Appendix C) for the grade of steel concerned and if a reduced strength compared to code requirements is observed a re-assessment with modified properties should be performed. 2.3.7
Other laboratory tests
A number of other microscopical and chemical analysis methods have been used to investigate fire-damaged concrete. These include scanning electron microscopy (SEM) and mineralogical analysis by X-ray diffraction (XRD), see Handoo et al(33) ]. Thermal analytical methods used include differential thermal analysis (DTA), thermal gravimetric analysis (TGA) and derivative thermogravimetric analysis, see Alarcon-Ruiz et al(34) and Handoo et al. To date, these methods have been used mainly for academic research and are not routinely used to investigate fire-damaged structures commercially. 2.4
ASSESSMENT OF FIRE DAMAGED STRUCTURES
2.4.1
Introduction 30
17th March 2008 Section 3.4.4 amended
First, it is useful to try to build up a picture of the nature of fire and thus the likely nature and extent of any damage. An assessment of the materials burnt and the disposition of the fire can give information about likely temperatures developed and the duration at any location. Fire debris can also give useful guidance as to the temperatures experienced by evaluating which types of materials, e.g. plastics, glass, aluminium, or timber that have deformed, melted or burnt, see Section 2.2.8 and Tables 3 and 4. These observations are rarely enough to evaluate the extent of damage directly but are a useful guide in planning more specific examination and testing. Consideration of the fire characteristics may also prompt other specific issues, such as whether toxic or deleterious combustion products have been given off. The burning of extensive quantities of polyvinyl chloride (PVC), for example, may give off enough hydrogen chloride to initiate corrosion of steel elements or reinforcement. The three principal concerns in evaluating the effect of the fire on concrete structure are: • Depth of damage (spalling) or loss in strength of the concrete • Loss in strength of steel reinforcement or embedded structural steel elements • Damage or distress to the structure from movement, settlement or imposed loads. Reinforced concrete members exposed to a fire of a severity insufficient to cause collapse are likely to have undergone a cycle which is summarised in Table 6. Table 6: Cycle of effects upon reinforced concrete structures. Stage On heating
On cooling
1. Rise in surface temperature 2. Heat transfer to interior concrete 3. Heat transfer to reinforcement (accelerated if spalling occurs) 4. Reinforcement cools 5. Concrete cools
Probable effects Surface crazing Loss of concrete strength, cracking and spalling Reduction of yield strength Possible buckling and/or deflection increase Recovery of yield strength appropriate to maximum temperature attained (Figure 10) Any buckled bars remain buckled. Cracks close up Reduction in strength until normal temperature is reached Deflection recovery incomplete for severe fire Further deformations and cracking may result as concrete absorbs moisture from the atmosphere.
For prestressed concrete members the sequence of events is broadly similar to that for reinforced concrete, but the position is made more complex by the lower temperature at which prestressing steel looses strength and by the reduced modulus of elasticity, which results in a reduction in tension (see Section 2.2.7). Both of these effects may be offset by the greater amount of cover generally provided to the tendons in prestressed concrete. Outwardly, damage to the concrete will be seen most obviously as spalling, see Section 2.2.4. This will vary depending upon the location in the structure and the severity of the fire. Typically, soffits and thin ribbed slabs show more damage than the tops of slabs and the
31
17th March 2008 Section 3.4.4 amended
lower portions of columns. The absence of spalling, however, does not necessarily signify that no damage or loss in strength has occurred. As mentioned earlier, the assessment of fire damaged structures can follow two methodologies, which can be combined: 1. Testing of fire damaged concrete to directly assess the concrete quality 2. Estimation of fire severity to deduce temperature-profiles and hence calculate the residual strength of the concrete or the reinforcement. 2.4.2 Testing It is sensible to define a procedure to test the structure. The aim is to establish zones in ceiling areas or damage classes of individual structural members. The zones are defined from the seat of the fire to parts where no heat impingement has been observed. If possible, samples that have been taken from concrete that has not been affected by elevated temperatures should be tested for comparative purposes. The amount of testing, as well as the methodology, will be governed by the availability of equipment, the budget, the time available and other restraints. Health and safety must be considered at all times, particularly when extracting material samples from critical locations. Care should be taken not to cause more damage than the fire. 2.4.3
Assessment of the fire severity
Deduction of the temperatures reached in the concrete should be based as much as possible on observations. The history of a fire may help to determine the pattern of the fire intensity. Video footage and observations from the fire brigade are important sources of information. Often the duration, intensity and extent can be determined from eye-witness accounts. The aim is to determine the condition of the structure as accurately as possible. An examination of the debris (see Section 2.2.8) may not give as accurate a picture of the temperature of the fire and its effect upon the structure as would a detailed petrographic investigation or a visual inspection of the discoloration in the concrete. The condition of the concrete is a direct indication of the effect the fire had upon the materials, whereas the condition of the debris is subject to local conditions and flame temperature. In addition, predictive fire engineering tools, such as empirical equations or computer modelling used in design, can be used to assess the fire severity in the building, based on the fire load in the building, ventilation conditions, compartment size and shape and properties of wall linings. An estimate of the fire time:temperature curve can be based on the heat-release, the characteristic temperatures at flashover, the expected gas temperatures during a fully developed phase of the fire and the area of window openings providing ventilation to the fire. From eyewitness accounts it may, for example, be possible to determine when the fire started, how many windows were open and when flashover occurred. This allows estimates of the pre-flashover heat release rate. Further estimates of the heat release rate for the post-flashover phase may be obtained from temperatures within the compartment, deduced from debris. An assessment with a finite element Computer Fluid Dynamics program might then allow hot spots to be determined and may provide confidence in data obtained from testing and debris. The results can then be used to determine the amount and locations for testing and may confirm evidence provided by tests.
32
17th March 2008 Section 3.4.4 amended
2.4.4
Heat transfer
Once a credible time:temperature distribution within the compartment is determined, an assessment of the temperatures within the concrete is possible without relying solely on testing. As a result of heat transfer analysis it may be possible to reduce the amount of testing and the determination of required depths of samples. Temperatures greater than 900°C are frequent in fires in buildings. But, in a concrete member, only the temperature of the outside layers is drastically increased and the temperature of the internal concrete may be comparatively low. Temperature distributions in dense concrete elements in standard fire tests are given in Annex A of Part 1-2 of BS EN 1992(5), as follows: • Figure A.2: slabs (and walls exposed to fire on one side) • Figures A.3–A.10: beams of various sizes • Figures A.11–A.15: 300 × 300mm columns • Figures A.16 – A.20: 300mm diameter circular columns. It is important to note that the development of temperature in a real fire differs from the standard temperature time curve. Hence it is difficult to deduce the effects of a real fire from standard tests. A more straight forward way is to estimate realistic time:temperature distributions within the compartment and conduct heat transfer analysis. Such an assessment is more realistic than comparing results of standard tests with the situation after a real fire. Heat transfer calculations of concrete structures can be performed according to methods outlined in Part 1-2 of BS EN 1992. The numerical calculations are based on finite elements and can be performed with standard software. The required thermal properties of normal and lightweight concrete are provided in Part 1-2 of BS EN 1992. 2.5
Presentation of data
Before determining the necessary repairs to the structure, the degree of damage to the various elements must be determined. This section presents one approach, though other approaches may be more appropriate depending on the type of structure being considered and the nature of the fire. The degree of damage may be determined simply on the basis of a visual assessment of the structure. An example of a visual classification scheme is given in Table 7, which leads to Classes of Damage ranging from 0 to 4. However, visual inspection will generally be supplemented by the following evidence: • The condition of the concrete o Temperature reached o Depth of penetration of fire damage o Proportion of section needing renewal o Evidence from temperature of fire, duration of fire, sounding, colour, spalling, physical tests • The condition of the steel o Temperature reached o Reduction to yield point, ultimate strength, Young’s modulus, ductility o Loss of tension (prestressing tendons)
33
17th March 2008 Section 3.4.4 amended
•
o Evidence from loss of concrete cover, colour of surrounding concrete, samples of steel The quality of the original construction o Concrete: low strength, poor compaction, weak or contaminated aggregate, type of cement o Reinforcement: poor detailing, inadequate cover, closeness of bars o General: inaccuracies of line or level, non-axiality of column storey heights, poorly executed construction joints.
34
17th March 2008 Section 3.4.4 amended
Table 7: An example of a visual damage classification scheme for reinforced concrete elements. Class of damage 0 1
2
3
Element Any Column Wall Floor Beam Column Wall Floor Beam Column
Surface appearance of concrete Condition of Colour Crazing plaster/finish
Spalling
Some peeling
Normal
Slight
Minor
Substantial loss
Pink/red**
Moderate
Localised to corners Localised to patches
Total loss
Pink/red** Whitish grey***
Extensive
Localised to corners, minor to soffit Considerable to corners
Surface lost
Considerable to surface Considerable to soffit Considerable to corners, sides, soffit Almost all surface spalled
Wall Floor Beam 4
Column Wall Floor Beam
Destroyed
Whitish grey***
Structural condition Exposure and condition of main reinforcement* Unaffected or beyond extent of fire None exposed
Very minor exposure Up to 25% exposed, none buckled Up to 10% exposed, all adhering Up to 25% exposed, none buckled Up to 50% exposed, not more than one bar buckled Up to 20% exposed, generally adhering Up to 50% exposed, not more than one bar buckled Over 50% exposed, more than one bar buckled Over 20% exposed, much separated from concrete Over 50% exposed, more than one bar buckled
Cracks
Deflection/ distortion
None
None
None
None
Minor
None
Small
Not significant
Major Severe and significant
Any distortion Severe and significant
Notes: * In the case of beams and columns the main reinforcement should be presumed to be in the corners unless other information exists ** Pink/red discolouration is due to oxidation of ferric salts in aggregates and is not always present and seldom in calcareous aggregate, see Section 2.2.2 *** White-grey discolouration due to calcination of calcareous components of cement matrix and (where present) calcareous or flint aggregate.
35
17th March 2008 Section 3.4.4 amended
As a general guide, it is suggested that the repair that is required for the various Classes of Damage should be as indicated in Table 8. Table 8: Initial repair classification. Class of damage 0 1
Repair classification Decoration Superficial
2
General repair
3
Principal repair
4
Major repair
Repair requirements Redecoration if required Superficial repair of slight damage not needing fabric reinforcement Non-structural or minor structural repair restoring cover to reinforcement where this has been partly lost. Strengthening repair reinforced in accordance with the loadcarrying requirement of the member. Concrete and reinforcement strength may be significantly reduced requiring check by design procedure. Major strengthening repair with original concrete and reinforcement written down to zero strength, or demolition and recasting.
An example of how this information might be summarised for part of a building is shown in Figure 20, and the accompanying Table 9.
[Dimensions for columns sizes (380 × 380) and spacing (4.25m in N–S direction and 5m in E–W direction) to be added to Figure so that it ties up with the design examples.] Note: Circled numbers are damage class for each member.
Figure 20: Typical section of key diagram classification.
36
17th March 2008 Section 3.4.4 amended
Table 9: Typical section of schedule for damage classification shown in Figure 20.
Element Columns
Beams
Slabs
Location: Ground floor columns and first floor beams and slabs Class of damage Member reference number 1 1, 2, 11, 21, 31 2 5, 12, 15, 22, 25, 32, 34, 35 3 3, 4, 13, 23, 24, 33 4 14 1 11, 111, 211, 311, 112, 212, 312 2 21, 121, 331, 152 3 31, 41, 141, 221, 241, 341, 321, 132, 142, 232, 242, 252, 352 4 131, 231 1 101, 201, 301 2 102, 202, 203, 302, 303, 304 3 104, 204 4 -
The above approach is somewhat simplistic as it only considers the requirements for individual elements. Some adjustments would be required in the light of the overall performance of the structure, considering aspects such as: • The continuity of members • Stability including the need for robustness • Excessive residual distortion; the structure will probably have incorporated inaccuracies in the original construction, hence check by reference to items (e.g. lift guides) of known alignment prior to the fire. The method of carrying out the repairs will depend on a number of factors including the following: • Accessibility for the proper application of sprayed concrete • The performance of other repair materials in the event of a subsequent fire • The practicability of adding the number and size of reinforcing bars or links needed to restore the strength of the member. If the repair is Class 3 or Class 4 it may be necessary to consider whether it will be quicker and/or cheaper to carry out the repair or to demolish and reconstruct. It is also important to consider the general condition of the structure with respect to durability. This may influence the choice of repair method, and other aspects such as the provision of protective coatings.
37
17th March 2008 Section 3.4.4 amended
3
Design
This section of the report deals with aspects of the structural design of the reinforced concrete elements within the fire damaged building where the damage assessment procedure outlined in the previous sections has shown that repair is a viable option. The recommendations apply to conventionally reinforced structures. Similar principles apply to prestressed concrete; however, specialist advice may be necessary on how to deal with loss of prestress. Worked examples illustrating the design approach outlined in this chapter, applied to a slab, tee-beam and column, are given in Appendix B. 3.1
DESIGN PHILOSOPHY
3.1.1
Objectives
Repairs to a fire damaged concrete structure should provide the strength, fire resistance, durability and appearance appropriate to the proposed use and projected design life of the building. The intended use for the structure and the objectives for the repair should be agreed with the building owner before commencing the design of the repair work. 3.1.2 Building regulations The designer should consult with the local authority regarding the need for approval under the Building Regulations for the proposed reinstatement and repair works. It is possible that upgrading of both structural and non-structural aspects of the building may be required as part of the repair works. 3.1.3 Codes of practice In general the design of the repaired sections of the building should comply with current codes of practice. However, the damaged structure may have been designed to codes of practice which are out of date; for example, there have been significant changes to the requirements for shear reinforcement and provision for robustness in more recent structural codes. Where this is the case it may be necessary to formulate a strategy for the structural design of the repaired section of the building which is compatible with the original design. For example, limitations may be imposed on the restoration of listed buildings. Alternatively, if the assessment of the fire damaged elements of a building indicates significant deficiencies in the original design, it may be necessary to enhance the complete structure. However, before taking such a step it will be necessary to determine if there is any evidence to suggest that the structure was in any way inadequate or that there had been problems with similar structures. Wherever possible the original structural design drawings for the building should be obtained as these will be of considerable help in assessing the original properties of the structure; the steel and concrete strengths in older buildings were lower than those commonly used at present. Appendix C lists historical information, including the dates at which various structural codes were introduced, the steel strengths that were current at the time and the symbols used for reinforcement detailing.
38
17th March 2008 Section 3.4.4 amended
3.1.4 Design assumptions Any assumptions used in the design of the repaired sections of the building should take due account of the materials and methods used in carrying out the repair 3.2
STRUCTURAL ANALYSIS AND MEMBER DESIGN
3.2.1
Structural analysis
Structural analysis of the repaired structural frame should be based on methods of analysis and load arrangements as set out in current codes of practice. In addition to normal assumptions, the analysis should take due account of any dimensional changes, lack of verticality and residual forces which could have resulted from the elevated temperatures during the fire. As discussed in Section 2.2.5, a column with a large out of plumb will be required to carry substantially greater bending moments than would be assumed in normal design. 3.2.2
Element design
The design of the reinstated and repaired concrete elements should be based on design methods as set out in current codes of practice. Repaired concrete elements will comprise a combination of the remaining section of the existing member and the repair materials. Modified material parameters may have been established in the fire damage assessment process and must be taken into account in the redesign of the repaired members. The strength properties of the repair material should be used where appropriate. 3.3
REPAIR CRITERIA
3.3.1
Reduced material strengths
In the design of the repaired section of the structure it is necessary to take account of the reduced strength of the remaining concrete and reinforcement which may have suffered from the elevated temperatures during the fire. Ideally this reduction in strength should be based on the results of tests on samples taken from the sections of the fire damaged zone of the structure which are to remain after the repairs are complete. The concrete strength should be based on compression tests on concrete cores, see Section 2.3.5. Similarly the design strength of the steel reinforcement should be based on tensile tests on reinforcement samples taken from the fire zone, see Section 2.3.6. Due account may taken of the fact that the test results give an indication of the actual strength of the concrete and steel by making appropriate modifications to the material strength factors. Further information may be found in the Institution of Structural Engineers document Appraisal of existing structures(35) and the Highways Agency Assessment of concrete highway bridges and structures(36). 3.3.2
Residual strength factor
39
17th March 2008 Section 3.4.4 amended
As an alternative to the above, where there is some knowledge of the fire temperature, residual strength factors obtained from Figures 3 and 10 may be used. The factors are to be applied to the design stresses in the residual concrete and reinforcement. An average factor for the whole member may be obtained by considering separate layers in a cross-section. Stresses in compression, tension, shear, torsion and bond are to be reduced in this way. If no data exists regarding the original strength of the concrete it should be determined from cores taken from existing undamaged concrete. The strength of the reinforcement should be determined from samples taken from sections of the undamaged concrete structure. 3.3.3 Bond strength Provided the full cover to the existing reinforcement, along with any shear reinforcement, is reinstated during the repair, full bond between the existing reinforcement and surrounding repair material may be assumed and hence lap and anchorage lengths will not be compromised. Where existing bars have buckled (see Section 2.2.7 and Figure 11) they will no longer be adequately bonded to the concrete. In this situation, it will be necessary to remove sufficient concrete from behind the bars so that the repair material fully surrounds the reinforcement, to ensure full composite action. 3.3.4
Bar size and spacing
Bar spacing should be sufficient to ensure full compaction of the repair material, see Section 4.6.1. Bar spacing should also consider the direction from which the sprayed concrete or other material is to be applied. Some reduction in the spacing may be considered where there is access from several faces. Adequate compaction is vital to successful repair, and any deviations from the recommended minimum should be discussed with the repair contractor. 3.3.5
Shear reinforcement
Additional shear reinforcement should be anchored in such a way to enable it to function properly with the undamaged portions of the member concerned, see Section 4.6.2 and Figure B.10. 3.4
MEMBER DESIGN
3.4.1 General Member design should take into account the stress history of the remaining concrete and steel. Beams and slabs may be propped during the repair process and the undamaged concrete may therefore be assumed to be unstressed at the time of the repair. However the remaining concrete in columns and walls may be highly stressed at the time of the repair by loads from the structure above. In these cases careful consideration needs to be given to the distribution of loads between the remaining concrete and steel and the new repair materials. 3.4.2
Beams and slabs – bending
Where appropriate, beams and slabs may be repaired and strengthened to resist the applied bending moments by adding repair concrete and tension reinforcement. The added reinforcement must be sufficient to resist the redesign tensile forces, minus the permissible
40
17th March 2008 Section 3.4.4 amended
tensile force which can be resisted by the original tension reinforcement, taking into account the residual strength. The approach is illustrated by Examples 1 and 2 in Appendix B. 3.4.3
Beams – shear
Where appropriate, beams may be strengthened to resist the redesign shear forces by adding shear reinforcement. The added reinforcement must be sufficient to resist the redesign shear force minus the permissible shear forces, which can be resisted by the original shear reinforcement taking into account the residual strength for both the steel and the concrete. The approach is illustrated by Example 2 in Appendix B. 3.4.4
Columns
Columns may be strengthened to resist the redesign loads by adding reinforcement and repair concrete. The resultant section comprising the repair materials and the original concrete and reinforcement, taking account of the residual strengths, must be sufficient to resist the redesign axial load and bending moments. Example 3 in Appendix B shows the approach for an axially loaded column. Special consideration should be given to columns with a permanent lack of verticality and the resulting additional P-∆ effect should be included in the design based on measured imperfections. The design should also account of any potential permanent reduction of the stiffness of the concrete and reinforcement due to the fire. This becomes particularly important for a slender column, whose failure mode would be dominated by stability rather then strength. Designers should note the possible difficulties in adding longitudinal reinforcement due to overhead obstructions from upper floor slabs, beams, etc, particularly where bending is a dominant influence in the column design and it is necessary to provide adequate anchorage beyond the point at which the additional steel is no longer required. 3.4.5
Walls
The general principles of column strengthening may be applied to walls but there may be difficulty in threading new bars through the remaining floors. It is recommended that vertical wall reinforcement should be therefore be nominal, terminating above and below the floor slabs, and that the original, undamaged concrete acting together with the repair should accordingly be capable of achieving the desired strength as an unreinforced wall wherever possible. This recommendation will not apply to shear walls which may be heavily reinforced to resist lateral forces and for which special treatment to ensure maintenance of continuity of the steel reinforcement may be required in designing the repair. 3.5
DESIGN OUTPUT
3.6.1
Demolition and construction sequence drawings
The designer should prepare drawings which clearly show the extent of demolition of the fire damaged structure and should include details of any associated temporary propping which may be required in the temporary condition to enable the repair works to be carried out safely.
41
17th March 2008 Section 3.4.4 amended
In some instances it will be necessary to provide temporary propping beyond the extent of the damaged structure (e.g. where demolition could cause increased design forces in adjacent spans). Where critical to the design assumptions the designer should prepare construction sequence drawings showing the timing of the installation and removal of any temporary propping and the sequence of the repair operation. 3.5.2 Key plans Key plans should be prepared at each floor level showing the location of the repair work and the positioning of the detailed sections. 3.5.3 Design details Design details are required at each of the repair locations. These will be determined on the basis of testing (as described in Chapter 2) along with practical considerations such as the method of breaking out the damaged concrete and the subsequent method of repair (see Chapter 4). The details should include the following information: • The extent of breaking out of fire damaged concrete and removal of fire damaged reinforcement steel. • Requirements for preparation of concrete surfaces that are to receive repair concrete including any special requirements to prevent feathered edges. • Details of new steel reinforcement including lap length and splicing with original bars, mechanical anchorage, cover etc. • Any fabric reinforcement that may be required to hold the repair concrete in place in the temporary condition, including means of supporting the fabric and the required concrete cover. • The thickness and the properties of the repair concrete. Some typical repair details, using sprayed concrete, are shown in Section 4.9.7. Similar details will be used for hand-applied repair materials or concrete cast in formwork. 3.5.4
Specifications
In addition to the design drawings and details the designer should prepare detailed material and workmanship specifications for the repair work. This should include full information on the repair materials and include means for ensuring quality control. Information on suitable repair methods is given in Chapter 4. 3.5.5
Design calculations
The designer should prepare design calculations for the repair works. The calculations should clearly set out all assumptions used in preparing the repair information, including the following: • Material properties of the existing structure • Residual strength of fire damaged concrete and steel to remain in the repaired building • Material properties of repair concrete and new steel reinforcement • Design codes of practice 42
17th March 2008 Section 3.4.4 amended
• • • • • • •
Design loadings Fire resistance requirements for the repaired structure Durability requirements for the repaired structure Analysis method including any simplifying assumptions to be used for the frame analyses Load arrangements to be used in structural analyses Details of the frame analyses including design forces at the repair sections Member design based on these design forces.
3.5.6 Method statements Method statements giving full details of the proposed means of carrying out the repair works should be prepared by the repair sub-contractor and submitted to the design engineer for comment before commencing work on site. 3.6
LOAD TESTS
Quality control of the materials carried out in the repair will generally be by routine tests on the compressive strength of the repair concrete and tension tests on the steel reinforcement. Where there is any uncertainty about the material strength or the quality of the repairs a load test may be carried out on the repaired structure, although this is not generally considered to be routine. Where the circumstances are such that load testing is deemed to be necessary it should be carried out in accordance with current codes of practice, such as Section 9 of BS 8110 Part 2(37). Some general guidance on load testing is included in Institution of Structural Engineers’ Appraisal of existing structures(35).
43
17th March 2008 Section 3.4.4 amended
4
Repair methods
The methods used for the repair of fire-damaged concrete are no different from those used for the repair of concrete damaged by corrosion of the reinforcement, with advice available from various sources, including Concrete Society Technical Reports 26, Repair of concrete damaged by reinforcement corrosion(38) and 38, Patch repair of concrete subject to reinforcement corrosion(39). 4.1
GENERAL
For reinforced concrete, the main processes to be undertaken are as follows: • Removal of damaged or weakened concrete • Replacement of weakened reinforcement • Replacement of concrete both to reinstate the original form and to provide adequate structural capacity, durability and fire resistance. In some circumstances there may also be a requirement for the reinstatement of special finishes and appearance. Before finalising remedial works specifications the concrete should be thoroughly assessed to ensure that repairs and reinstatements address any inherent or pre-existing problems, such as low covers, excessive levels of chloride or depths of carbonation; Advice is available from Concrete Society Technical Report 54, Diagnosis of deterioration in concrete structures(40). In addition, the effects any contaminants derived from the fire on the long-term durability of the structure should be considered. The approach to the repair of prestressed structures needs particular care, as the repair work will need to ensure the full structural capacity of the element is maintained following the remedial works. Damaged unbonded post-tensioned tendons can, perhaps following installation of temporary tendons, be de-stressed, removed and replaced, with new tendons restressed once the concrete has been reinstated. However, the same approach cannot be adopted for pre-tensioned elements (or bonded post-tensioned elements) as the damaged tendons cannot be removed. Such elements will generally be demolished and replaced, although one approach could be to use pre-stressed externally bonded FRP plates, see Section 4.10. For some types of structures (e.g. car parks, bridges, or circular tanks) external tendons may be an appropriate solution. The chosen repair method or methods will depend on a number of factors, including the original concrete type and condition prior to the fire, the extent of fire damage, costs, ease of access and programme constraints. Fire damaged heritage structures will come with a host of additional factors linked to any listed status. The designer will have to decide upon a repair method and obtain necessary approvals before proceeding to the detailed design. However, in some cases the only viable option will be demolition and reconstruction; this is outside the scope of this report.
4.2
HEALTH AND SAFETY
All operatives should wear the appropriate personal protective equipment (i.e. safety helmet, gloves, boots etc) when carrying out repairs. When cleaning concrete surfaces (see Section 4.4) or breaking out concrete (see Section 4.5) steps should be taken to suppress dust and to dispose of waste material in a suitable manner. Repair materials must be used strictly in 44
17th March 2008 Section 3.4.4 amended
accordance with the manufacturers’ instructions and the Control of Substances Hazardous to Health (COSHH) Regulations(41). As indicated in Chapter 1, it is essential to consider the safety of the structure at all stages, from extinguishing the fire, through the assessment phases and throughout the repairs. Where necessary, beams and slabs should be propped, with temporary bracing. Phased breaking out may be required in some circumstances. Special consideration is needed if the structure is to be left to deteriorate further before repair works are carried out, with the obvious risk of collapse considered, but also the safety of third parties, from the risk of falling debris. During the course of the remedial works risks, such as falling concrete during breaking out, should be assessed and appropriate actions specified to mitigate any identified events that could arise. Safe access to the area being repaired should be provided for personnel. Repair materials and equipment should be stored in a suitable location, taking account of any additional loads they may apply to the weakened structure. 4.3
QUALITY CONTROL
All aspects of the repair of concrete structures should be in accordance with the various Parts of BS EN 1504, Products and systems for the repair and protection of concrete structures – Definitions, requirements, quality control and evaluation of conformity(42), which includes guidance on the quality control tests to be carried out. (Note that Part 9 of BS EN 1504, General principles for the use of products and systems, which was in draft form at the time of writing, specifically excludes the repair of concrete structures damaged by fire. The reasons for this exclusion are unclear; as indicated earlier in this Technical Report, the same principles should apply to the repair of any damaged structure.) Routine testing will be required for the materials used (e.g. cube tests to access the strength) and pull-off tests on completed areas to demonstrate adequate bond between the repair material and the original concrete. Specific quality control requirements for sprayed concrete are outlined in Section 4.9.1.
4.4
SURFACE CLEANING
The cleaning methods used to remove soot and smoke damage from reinforced concrete elements will be dependent upon the degree of discolouration, the intended finishes of the concrete surface and also the location of the damaged element. A lesser degree of cleaning may be required if the finished surface is to be hidden from view, such as behind a suspended ceiling, but more intense cleaning may be required if the concrete elements are to be left exposed and uncoated. The method of removal is generally pressure water jetting. In some areas this technique is inappropriate, such as occupied residential or commercial buildings. In this case power wire brushing with vacuuming is often used. Surface cleaning may be required prior to the commencement of any repair works to enable the clear identification of areas 4.5
BREAKING OUT
As indicated in Section 3.5 the extent of breaking out that is required will be determined by the assessment of the extent of the damage to the concrete and by practical considerations, such as the intended method of working. Before breaking out is undertaken, it is necessary to be certain that the reduction in structural section will not over-stress the member. In some
45
17th March 2008 Section 3.4.4 amended
cases it may be necessary to remove any heavy load (e.g. equipment or plant) supported by the member and reinstate it after repair. An alternative may be to prop the structure. Propping is essential, with full removal of the load, if a full structural repair is required, i.e. in cases where the new concrete or mortar is expected to carry its full share of the load on the repaired member. There are various methods of breaking out, ranging from a simple hammer and cold chisel for very small repairs to electrically or pneumatically powered breakers (see Figure 21) and hydro-demolition for larger areas (see Figure 22). Each of these methods carries with it various pros and cons with respect to Health and Safety, and Environmental Risks, which must be carefully considered on a site-specific basis, see Table 10. The use of small hand-held tools, drills and breakers are now limited, at least under the provisions of The Control of Vibration at Work Regulations(43), with the use of drills and breakers, and hydro-demolition, at least limited under the provisions of The Control of Noise at Work Regulations(44). Drills, breakers and hydro-demolition may also have environmental impacts such as disturbance to neighbours etc., and the production of dust and contaminated waste-water and spray.
Figure 21: Breaking out small area of concrete using hand-held equipment.
Figure 22: Breaking out using hydro-demolition.
46
17th March 2008 Section 3.4.4 amended
Table 10: Features of methods of breaking out concrete. Method Hammer and chisel
Electrically or pneumatically powered breakers
Hydro-demolition
Comment Slow and labour-intensive Generally only suitable for small areas of repair Physically exhausting and limited under the provisions of The Control of Vibration at Work Regulations 2005 Extremely noisy Physically exhausting and limited under the provisions of The Control of Vibration at Work Regulations 2005 and The Control of Noise at Work Regulations 2005 Labour-intensive, but reasonable productivity possible Heavy equipment may require independent support at the work face May cause cracking of the substrate May damage the reinforcement Potential environmental impacts such as disturbance to neighbours etc., and the production of dust Can lack the fine control and precision of electrically or pneumatically powered breakers, in conjunction with concrete saws/disc cutters Can be excessively noisy at close-quarters Physically exhausting and limited under the provisions of The Control of Vibration at Work Regulations 2005 and The Control of Noise at Work Regulations 2005 Very high level of productivity Cleans reinforcement Does not damage the reinforcement Potential environmental impacts such as disturbance to neighbours etc., and the production of contaminated waste-water and spray
The use of electrically or pneumatically powered breakers may cause damage to some substrates. Once the extent of an individual repair has been decided, the edges should be clearly marked on the surfaces. For breaking out using electrically or pneumatically powered breakers this is to facilitate the cutting of the edge, using a concrete saw or disc cutter, to a depth of at least 5–10mm, depending on the repair material being used and the cover to the reinforcement. To prevent cracking and debonding, of the repair materials, away from the sawn/cut edge, the edges of the repair are usually cut perpendicular to the face of the element; although it can be beneficial if the edges are undercut slightly so that the repair is keyed into the surface. The objectives of breaking out are to remove all the deteriorated concrete, to deepen the repair area so that it is a suitable size and shape to receive the repair material and to do this without damage to the concrete and reinforcement that are to remain in place. Repair dimensions and edges may therefore need to be modified during the breaking out process to ensure all deteriorated concrete is removed. The depth of breaking out is controlled by the depth of cover to the reinforcement at the repair; it is necessary to provide a clear distance behind bars so that the repair material can be easily compacted into the space and surround the bars. The minimum clear distance behind
47
17th March 2008 Section 3.4.4 amended
the bars will be dependant upon a number of factors, including the type of repair material and its maximum aggregate size, but for guidance a gap of 20mm is adequate for a patch repair using hand-applied mortar, or flowable materials, while 50mm is needed for large repairs being reinstated in concrete with a maximum aggregate size of 20mm. Once the bulk breaking out has been completed, the repair should be shaped to make reinstatement easier. Corners should be rounded slightly as it can be difficult to compact material into right-angled corners. Edges cut by saw or disc can have very smooth surfaces to which it is difficult to bond the repair. Therefore saw-cut edges could be roughened slightly by light grit blasting. 4.6
REINFORCEMENT
4.6.1
Bar size and spacing
It is recommended that the new reinforcement should be of the same size and at the same spacing as the original. This will simplify connection to the existing reinforcement. However, consideration should be given to the method of reinstatement of the concrete; the spacing of the reinforcement should be such that the material can be properly compacted around the bars. The Sprayed Concrete Association recommends that bars should have a maximum diameter of 25mm and be spaced at least 50mm apart (or four times the bar diameter, whichever is the greater) and at least three times the maximum bar diameter (or 40mm, whichever is the greater) away from the substrate(45). With two layers of reinforcement a better repair is produced if the second reinforcing layer is not placed until the first has been sprayed, to reduce the likelihood of voids behind the reinforcement. If smaller bar spacings are required, steps should be taken to ensure that the sprayed concrete contractor can still achieve adequate compaction around the reinforcement. 4.6.2
Connecting reinforcement
Lapping of reinforcement is the simplest method of connecting new reinforcement to the existing. However, sufficient concrete must be broken back to provide a full tension (or compression) anchorage length in accordance with the design code. Some types of coupler may provide a suitable method of connecting new reinforcing bars to old. For bars in compression, a simple sleeve and wedge coupler will be adequate, provided the ends of the bars are cut square. However, these must only be used where the bars are always in compression. There are various types of tension couplers used in new construction. Clearly those that require the ends of the bars to be threaded are not appropriate for repair. Swaged couplers are installed using a hydraulic press, and hence require sufficient working space around the bars. Couplers with shear bolts and serrated saddles can be used in comparatively confined locations. In all cases it is necessary to ensure that the cover to the coupler in the finished repair is at least equal to that specified for the reinforcement. Welding of reinforcement should generally be avoided. Where it is essential it must be undertaken by qualified personnel, working in accordance with an agreed quality procedure. Care must be taken that the heat from welding does not locally damage the concrete.
48
17th March 2008 Section 3.4.4 amended
Additional bars may be anchored into holes drilled into sound concrete to the required depth. The holes should be cleaned out and a suitable anchor resin (usually based on epoxies or polyesters) injected into the hole. The method of filling the holes with resin is critical as it is difficult to avoid entrapping pockets of air, which would weaken the connection. Trials, including pull-out tests, should be carried out at the start of the repair contract to demonstrate that materials and workmanship are capable of achieving the required standards. Pull-out tests on anchored bars should also be carried out during the course of the contract. 4.7
MORTAR
Mortar is generally used for relatively small repairs (typically less that 1m2) that are place by hand. The mortars themselves are usually proprietary products containing sand, cement (including Portland cement, ggbfs and micro-silica combinations), polymers, other minor constituents and, in some cases, fibres. The materials should be mixed and used strictly in accordance with the manufacturers’ instructions. Mortar is applied by trowel (as in Figure 23) or by gloved hand. It has to be placed onto either a soaked substrate or while the bonding aid, if used, is still tacky. The mortar has to be compacted against the substrate and around the reinforcement so that all crevices are filled and there are no voids. In deeper repairs the mortar has to be built up in layers. For structural repairs consideration should be given to compliance testing, including compressive strength (both minimum and maximum strengths should be considered, as many repair mortars achieve much higher strengths than substrate concretes, which could be significant with thinner repairs) and tensile bond strength, as well as qualitative assessments of compaction.
Figure 23: Hand-applied mortar repair.
4.8
FLOWABLE MICRO-CONCRETE AND CONCRETE
Repair material may be a proprietary flowable micro-concrete or a conventional concrete, depending on the size and shape of the repair and the complexity of the reinforcement. The techniques used in each case are essentially similar, requiring robust, water-tight formwork, designed to ensure full evacuation of any entrapped air.
49
17th March 2008 Section 3.4.4 amended
Flowable micro-concretes are self-levelling and self-compacting, whereas conventional concretes will generally require mechanical compaction. Such requirements should be carefully considered on a site-specific basis. Flowable micro-concrete and normal concrete repairs are carried out where large areas or large volumes of concrete have to be replaced and a formed finish is required. Flowable micro-concrete is generally used when the repairs are relatively thin or where reinforcement is congested. Normal concretes are generally used where the sections are similar to those found in conventional construction, though it is difficult to indicate a division between situations where flowable micro-concrete or normal concrete should be used. One consideration is access for vibration or the use of other means of compaction. Minimum thickness for flowable micro-concrete, which is placed without vibration, is around 50mm but will depend on reinforcement congestion. 4.9
SPRAYED CONCRETE
4.9.1 General A comprehensive description of all aspects of sprayed concrete technology, including equipment, construction practice, quality control and site safety considerations, is provided by Austin and Robins(46) . Concrete Society Technical Report 56, Construction and repair with wet-process sprayed concrete and mortar(45) provides a useful overview of construction and repair using modern wet-process sprayed concrete. EFNARC(47) provides a model specification for sprayed concrete, with guidelines published separately for the information of specifiers and contractors. Several major suppliers of sprayed concrete and fibres produce their own guidelines and detailed product support, and may be contacted as necessary. A European Standard, BS EN 14487(48) covers sprayed concrete and one for testing, EN 14488(49), is currently being developed; most parts of the latter have been published as British Standards. Sprayed concrete is produced by two processes, wet and dry. In both processes, mortar or concrete is conveyed through a hose and projected pneumatically at high velocity from a nozzle into the repair, see Figure 24. No other compaction is required but the surface can be finished by hand if this is required.
Figure 24: Sprayed concrete application.
50
17th March 2008 Section 3.4.4 amended
The EFNARC Specification gives guidance on the quality control that is required when using sprayed concrete. Test panels should be sprayed using the same materials, equipment, operator etc, from which cores or beams can be cut and tested to determine the compressive and flexural strengths respectively. Pull-off tests are used in trial areas to determine the bond between the sprayed concrete and the substrate. The advantages and disadvantages of sprayed concrete for repair work are shown in Table 11. The main advantages of the wet process over the dry process are better quality and consistency, lower material losses and a cleaner, less dusty working environment. Table 11: Advantages and disadvantages of sprayed concrete. Advantages No formwork required Rapid placing Dense, homogeneous, high quality concrete Reduction in access requirements Good bond to substrate and between layers Reduced thermal stresses when placed in several layers Suitable for use of soffits where it may be difficult to get concrete to flow into a shutter 4.9.2
Disadvantages Specialists required for design and application Variable concrete quality (mainly with dry process) High material costs (pre-bagged materials, wastage) Poor encasement behind dense concentrations of reinforcement
Health and Safety
The spraying process is potentially dangerous as it involves spraying dense particles at high velocity onto generally hard surfaces which, in the case of hand spraying, may be only a short distance from the operator (optimum nozzle to substrate distances are of the order of 1 to 1.5m maximum). To minimise the risks to operatives, modern high production application of sprayed concrete should be undertaken using remote controlled robotic sprays, which are available in a range of sizes; manufacturers should be contacted for details. Remote controlled spraying allows the operator to control the process from a position of safety. Generally, the wet mix process produces considerably less rebound and dust than the dry mix process. However, appropriate personal protective equipment will usually be required, including gloves, head protector, safety shoes, eye protectors and dust masks. Usually, design risk assessments will indicate that wet mix sprayed concrete applied by robotic sprays should be specified wherever possible in preference to hand applied dry mix sprayed concrete. If fibres are used in the wet mix process, they can be batched automatically at the batching plant or directly by hand into the mixer unit before discharge into the robotic spray. Normally, fibres are not considered to increase the hazard associated with sprayed concrete, provided that all necessary precautions are taken in accordance with an appropriate risk assessment. 4.9.3
Substrate preparation
As with other forms of concrete repair, thorough preparation of the substrate is important if a good bond is to be achieved. The surface should be cleaned with an oil-free air supply and wetted for a period before spraying. The surface should be damp but with no free water. If a bonding agent is used, care must be taken to ensure that it does not dry out prior to spraying. 51
17th March 2008 Section 3.4.4 amended
Any formwork used in spraying should be treated with a chemical release agent as emulsions or oils may be removed by the stream of concrete. 4.9.4
Layer thickness
The thickness of each layer will be determined by the amount of concrete that can be applied without sloughing off, which in turn is influenced by the position of the reinforcement, direction of application (vertical or overhead), mix design and constituents. Overhead work is generally applied in 25–50mm layers as thicker layers may sag or drop off. Thicknesses of up to 100mm or even 150mm can be achieved on vertical surfaces. When spraying a second layer, the preceding layer should be allowed to stiffen and then all loose material should be scraped or brushed away before the second layer is added. 4.9.5
Surface finishing
Many sprayed concrete surfaces are left in the as-sprayed condition. This is to avoid disturbing the material and leads to better bond with the substrate, strength and durability. Freshly sprayed concrete is quite stiff and working or trowelling the surface can cause drag tears or reduce bond. Light brushing with a soft brush about an hour after spraying can reduce the occurrence of shrinkage cracks and a textured finish can be obtained by trowel or float. The surface should not be sprayed with water or otherwise wetted before trowelling as this can lead to crazing of the surface. Flash coats (up to about 6mm thick) and finish coats (6– 25mm) of mortar or fine concrete are sometimes used to produce a more easily worked surface. 4.9.6 Curing As with all concrete repairs, wet curing is vital. Curing, in the form of hessian kept continuously wet, covered by polythene sheet and secured tightly against the surface at every edge, should be applied as soon as practicable after completion of spraying and kept in place for at least seven days. Alternatively, a sprayed curing membrane may be used. 4.9.7 Repair details Examples of typical details for different classes of repair to beams, columns and floor slabs using sprayed concrete are shown in Figures 25 to 27 respectively.
52
17th March 2008 Section 3.4.4 amended
[NOTE: Replace ‘gunite’ by ‘sprayed concrete’ and ‘mesh’ by ‘fabric’.] Figure 25: Sprayed concrete repairs to beams.
[NOTE: Replace ‘gunite’ by ‘sprayed concrete’ and ‘mesh’ by ‘fabric’.]
53
17th March 2008 Section 3.4.4 amended
Figure 26: Sprayed concrete repairs to columns.
[NOTE: Replace ‘gunite’ by ‘sprayed concrete’ and ‘mesh’ by ‘fabric’.] Figure 27: Sprayed concrete repairs to floor slabs.
4.10
RESINS
Resins have often been used for repairs to lightly spalled areas and, though they may perform quite satisfactorily in normal service, there is no comprehensive information on the performance of such repairs or that of the materials when subject to heat (in some cases resin repairs may have failed due to diurnal thermal cycling during hot summers) or an actual fire. These materials soften at relatively low temperatures (80°C). As a consequence, it is possible that some resin repairs may not provide adequate fire protection to reinforcement and may fail to retain structural adequacy in compression zones. Accordingly it is recommended that resin repairs be used only when either: a. Performance data can be supplied to show that the particular formulation has adequate fire resistance and retains its structural properties after the envisaged fire condition, or, b. The material is adequately fire protected by other materials and retains its structural properties at the expected fire temperatures at the relevant depth in the section, or, c. Loss of form of the material will not cause unacceptable loss of structural section or fire resistance. Resin repairs can exhibit compatibility problems with some substrates, leading to durability issues and premature failure. The designer should refer to manufacturers’ literature for details of the performance of the various materials available. A wide variety of materials exist and their specifications are liable to change. In general, such materials might be capable of providing good bond and compressive strengths and their flexural and tensile strengths may exceed those of concrete. However, the thermal expansion is considerably larger than that of concrete, which may be a point to be considered where the temperature range is large. For further information the designer should refer to Concrete Society Technical Report 26, Repair of concrete damaged by reinforcement corrosion(38).
54
17th March 2008 Section 3.4.4 amended
The materials are applied by hand and the full details of the methods of preparation and application should be supplied by the manufacturer. The work will closely follow other repair methods and the details given in Concrete Society Technical Report 26, will be applicable to fire damage repairs. 4.11
STRENGTHENING WITH FIBRE COMPOSITES
An alternative to providing additional steel reinforcement may be the use of fibre composite materials, generally known as FRPs, bonded to the surface using an epoxy adhesive. Details of the properties of the materials, typical applications, design approaches and installation are given in Concrete Society Technical Reports 55, Design guidance for strengthening concrete structures using fibre composite materials(50). Additional guidance is given in Technical Report 57, Strengthening concrete structures using fibre composite materials: acceptance, inspection and monitoring(51). Donnelly(52) describes the use of externally bonded fibre sheets as part of the repair of a fire-damaged building in Jakarta. Various types of materials are available. Thin (1–2mm thick) composite plates, generally consisting of carbon fibres in an epoxy resin, are suitable for bonding to the soffits of beams and slabs. Carbon, glass and aramid fibres are also available in the form of fabrics which can be bonded to a concrete surface to form a composite. The chief advantage of fabrics over plates is that they can be wrapped round curved surfaces, for example round columns to form a complete shell. (The corners of square or rectangular columns should be rounded somewhat to avoid damage to the fabrics.) This approach could be particularly suitable for cases where the links have been damaged. (TR 55 mentions that the newly-constructed columns of a building in Dublin were found to have insufficient links. They were strengthened by wrapping with carbon fibre sheet.) Wrapping can also be used to enhance the ultimate capacity of columns; by confining the concrete, both the ultimate stress and ultimate strain are increased. The surface to which the FRP plate is to be bonded should be sound and free of any loose material. The preparation of the surface should be such that the adhesive layer is of uniform thickness when the strengthening material is in place. Any steps in the surface should be removed and hollows filled with a suitable quick-setting repair mortar. Generally the flatness of the surface should be such that the gap under a 1m straight-edge does not exceed 5mm. The thickness of the adhesive layer is commonly between 1 and 5mm depending on the material, though thickening to 10mm may be acceptable. Overlapping, required when strengthening in two directions (e.g. for slabs), is not a problem because the material is thin, but care should be taken with the application process in the region of the overlaps. In the majority of applications the FRP plates are unstressed. However, the use of prestressed plates has been developing over recent years and a number of proprietary systems are now available commercially. There are many reasons for strengthening concrete structures using prestressed FRP including increasing live load capacity, reducing dead load deflections and regaining prestressed conditions in the concrete that may have been lost by damage to the original prestressing tendons. One disadvantage of using prestressed FRP plates is the need to provide fairly complex anchorages, which are fixed to the concrete surface. One disadvantage of the use of externally bonded material is the risk of fire (in which the adhesives will fail), vandalism or accidental damage, unless the laminate is protected, for example by means of a sprayed coating. However, in design, fire is considered as an accidental load. Thus the partial safety factors on materials are reduced and, more
55
17th March 2008 Section 3.4.4 amended
importantly, the partial safety factors on the dead and live loads are also reduced. Thus, in many cases (depending on the level of FRP strengthening) the fibre composite strengthening could fail completely without risking failure of the structure. Finally, FRP rods or narrow strips can be bonded into grooves cut into the concrete, a technique known as near-surface-mounted (NSM) reinforcement. This technique has the advantage that the FRP material is protected from damage by the surrounding concrete. In the UK it has been used to strengthen the top surface of slabs though it could be readily used for vertical elements and the soffits of slabs.
56
17th March 2008 Section 3.4.4 amended
References 1. CONCRETE SOCIETY, Assessment of fire-damaged concrete structures and repair by gunite, Technical Report 15, The Concrete Society, Camberley, 1978. 2. CONCRETE SOCIETY, Assessment and repair of fire-damaged concrete structures, Technical Report 33, The Concrete Society, Camberley, 1990. 3. BRITISH STANDARDS INSTITUTION, BS EN 1990. Eurocode 0: Basis of structural design, BSI, London, 2002. 4. BRITISH STANDARDS INSTITUTION, BS EN 1991. Eurocode 1: Actions on structures design, BSI, London, 2002. 5. BRITISH STANDARDS INSTITUTION, BS EN 1992. Eurocode 2: Design of concrete structures, Part 1-1: General rules and rules for buildings, Part 1-2: General rules – Structural fire design, BSI, London, 2004. 6. TOVEY, AK and CROOK, RN. Experience of fires in concrete structures, Concrete, Vol. 20, No. 8, August 1986, pp. 19–22. 7. TOVEY, AK and CROOK, RN. Experience of fires in concrete structures, Evaluation and repair of the damage to concrete, Special Publication SP 92, American Concrete Institute, Detroit, 1986, pp. 1–14. 8. THE STATIONERY OFFICE, The Regulatory Reform (Fire Safety) Order, Statutory Instrument 2005 No. 1541, Stationery Office, London, 2005. 9. INGHAM, JP. Assessment of fire-damaged concrete and masonry structures: The application of petrography, Proceedings of the 11th Euroseminar on Microscopy Applied to Building Materials, Porto, Portugal, 5–9 June 2007. 10. THE STATIONERY OFFICE. Construction (Design and Management) Regulations, SI 2007/320, The Stationery Office, London, 2007. 11. BRITISH STANDARDS INSTITUTION, BS 5975. Code of practice for falsework, BSI, London, 1996. 12. SCHNEIDER, U. Behaviour of concrete under thermal steady state and non-steady state conditions, Fire and Materials, 1976, pp. 103–115. 13. FRANSSEN, J-M. Plastic analysis of concrete structures subjected to fire, Proceedings of the workshop on Fire design of concrete structures: What now? What next?, Milan, 2004. 14. BAZANT, ZP and KAPLAN, M. Concrete at high temperatures: material properties and mathematical models, Longman Group, Harlow, 1996. 15. FÉDÉRATION INTERNATIONAL DU BÉTON, Fire design of concrete structures – materials, structures and modelling, State-of-art report, Bulletin 38, fib, Lausanne, Switzerland, 2007. 57
17th March 2008 Section 3.4.4 amended
16. INGHAM, JP and TARADA, F. Turning up the heat – full service fire safety engineering for concrete structures. Concrete, Vol. 41, No. 9, October 2007, pp. 27–30. 17. CHANA, P and PRICE, B. The Cardington fire test, Concrete, Vol. 37, No. 1, January 2003, pp. 28, 30–31. 18. DUNSTER, A. HAC concrete in the UK: assessment, durability management, maintenance and refurbishment, Special Digest 3, Building Research Establishment, Watford, November 2002. 19. STEVENS, RF. Contribution to discussion on: Steel reinforcement, by R I Lancaster, The Structural Concrete, Vol. 3, No. 4. July/August 1966, pp. 184–185. 20. HOLMES, M, ANCHOR, RD, COOK, GME and CROOK, RN. The effects of elevated temperature on the strength properties of reinforcing and prestressing steels, The Structural Engineer, Vol. 60B, No. 1, March 1982, pp. 7–13. 21. BRITISH STANDARDS INSTITUTION, BS 5268. Structural use of timber, Part 4: Fire resistance of timber structures, Section 4.1: Method of calculating fire resistance of timber members, BSI, London, 1978 (plus amendments). 22. FELICETTI, R. The drilling resistance test for the assessment of fire damaged concrete, Cement and Concrete Composites, Vol. 28, Issue 4, 2006, pp. 321–329. 23. BRITISH STANDARDS INSTITUTION. BS 6089. Guide to the assessment of concrete in existing structures, BSI, London, 1981. 24. BRITISH STANDARDS INSTITUTION, BS 1881. Testing concrete, Part 201: Guide to the use of non-destructive methods of test for hardened concrete, BSI, London, 1986. 25. BRITISH STANDARDS INSTITUTION, BS EN 12504. Testing concrete in structure, Part 1: Cored specimens - Taking, examining and testing in compression, Part 2: Nondestructive testing - Determination of rebound number, Part 3: Compressive strength of test specimens, Part 4: Determination of ultrasonic pulse velocity, BSI, London, Various dates. 26. AMERICAN SOCIETY FOR TESTING AND MATERIALS, ASTM C856, Standard practice for the petrographic examination of hardened concrete, ASTM, West Conshohocken, Pennsylvania, USA, 2004. 27. LIN, DF, WANG, HY and LUO, HL. Assessment of fire-damaged mortar using digital image process, Journal of Materials in Civil Engineering, Vol. 16, No. 4, 2004, pp. 383–386. 28. SHORT, NR, PURKISS, JA and GUISE, SE. Assessment of fire damaged concrete using colour image analysis, Construction and Building Materials, Vol. 15, No. 1, 2001, pp. 9–15. 29. PLACIDO, F. Thermoluminescence test for fire-damaged concrete, Magazine of Concrete Research, Vol. 32, No. 11, June 1980, pp. 112–116.
58
17th March 2008 Section 3.4.4 amended
30. CHEW, MYL. Assessing heated concrete and masonry with thermoluminescence, ACI Materials Journal, Vol. 90, No. 4, July-August 1993, pp. 319–322. 31. SMART, S. Concrete – the burning issue, Concrete, Vol. 33, No. 7, July/August 1999, pp. 30–34. 32. BRITISH STANDARDS INSTITUTION, BS EN 12390-3. Testing hardened concrete, Part 3: Compressive strength of test specimens, BSI, London, 2002. 33. HANDOO, SK, AGARWAL, S and AGARWAL, SK. Physicochemical, mineralogical and morphological characteristics of concrete exposed to elevated temperatures, Cement and Concrete Research, Vol. 32, Issue 7, 2001, pp. 1009–1018. 34. ALARCON-RUIZ, L, PLATRET, G, MASSIEU, E and EHRLACHER, A. The use of thermal analysis in assessing the effect of temperature on a cement paste. Cement and Concrete Research. Vol. 35, Issue 3, 2005, pp. 609–613. 35. INSTITUTION OF STRUCTURAL ENGINEERS. Appraisal of existing structures (Second Edition), The Institution of Structural Engineers, London, 1996. 36. HIGHWAYS AGENCY, BD 44. Design Manual for Roads and Bridges, Vol. 3: Highway structures inspection and maintenance, Section 4: Assessment, Part 15: The assessment of concrete highway bridges and structures, The Highways Agency, London, 1996. 37. BRITISH STANDARDS INSTITUTION, BS 8110. Structural use of concrete, Part 1: Code of practice for design and construction, Part 2: Code of practice for special circumstances, BSI, London 1997 and 1985. 38. CONCRETE SOCIETY, Repair of concrete damaged by reinforcement corrosion, Technical Report 26, The Concrete Society, Camberley, 1984. 39. CONCRETE SOCIETY, Patch repair of reinforced concrete subject to reinforcement corrosion: Model specification and method of measurement, Technical Report 38, The Concrete Society, Camberley, 1991. 40. CONCRETE SOCIETY, Diagnosis of deterioration in concrete structures: Identification of defects, evaluation and development of remedial action, Technical Report 54, The Concrete Society, Camberley, 2000. 41. THE STATIONERY OFFICE, The Control of Substances Hazardous to Health Regulations 2002. Statutory Instrument 2002 No. 2677, The Stationery Office, London, 2002. 42. BRITISH STANDARDS INSTITUTION, BS EN 1504. Products and systems for the repair and protection of concrete structures – Definitions, requirements, quality control and evaluation of conformity, Part 1: Definitions, Part 2: Surface protection systems for concrete, Part 3: Structural and non-structural repair, Part 4: Structural bonding, Part 5: Concrete injection, Part 6: Grouting to anchor reinforcement or to fill external voids, Part 7: Reinforcement corrosion protection, Part 8: Quality control and evaluation of conformity, Part 9: General principles for the use of products and systems, Part 10: Site application of products and systems and quality control of the works, BSI, London. [Note that Part 9 is still
59
17th March 2008 Section 3.4.4 amended
in draft form.] 43. THE STATIONERY OFFICE, The Control of Vibration at Work Regulations. Statutory Instrument 2005 No. 1093, The Stationery Office, London, 2005. 44. THE STATIONERY OFFICE, The Control of Noise at Work Regulations, Statutory Instrument 2005 No. 1643, The Stationery Office, London, 2005. 45. CONCRETE SOCIETY, Construction and repair with wet-process sprayed concrete and mortar, Technical Report 56, The Concrete Society, Camberley, 2002. 46. AUSTIN, SA and ROBINS, PJ, Sprayed Concrete: Properties, Design and Application, Whittles Publishing, Caithness, 1995. 47. EUROPEAN FEDERATION OF PRODUCERS AND APPLICATORS OF SPECIALIST PRODUCTS FOR STRUCTURES (EFNARC). European specification for sprayed concrete, Aldershot, 1996. 48. BRITISH STANDARDS INSTITUTION, BS EN 14487. Sprayed concrete, Part 1: Definitions, specification and conformity, Part 2: Execution, BSI, London, 2005 and 2006. 49. BRITISH STANDARDS INSTITUTION, BS EN 14488, Testing sprayed concrete, Part 1: Sampling fresh and hardened concrete, Part 2: Compressive strength of young sprayed concrete (Draft), Part 3: Flexural strengths (first peak, ultimate and residual) of fibre reinforced beam specimens, Part 4: Bond strength of cores by direct tension, Part 5: Determination of energy absorption capacity of fibre reinforced slab specimens, Part 6: Thickness of concrete on a substrate (Draft), Part 7: Fibre content of fibre reinforced concrete, BSI, London, 2005. 50. CONCRETE SOCIETY, Design guidance for strengthening concrete structures using fibre composite materials, Technical Report 55 (Second edition), The Concrete Society, Camberley, 2004. 51. CONCRETE SOCIETY, Strengthening concrete structures using fibre composite materials: acceptance, inspection and monitoring, Technical Report 57, The Concrete Society, Camberley, 2003. 52. DONNELLY, E. Keeping a wrap on it, Concrete, Vol. 40, No. 4, May 2006, pp. 33–34. 53. BERRY, R. Restored conference centre exhibits unique repairs, International Journal of Construction Maintenance and Repair, Vol. 5, No. 4, July/August 1991, pp. 36–39. 54. SRINIVASAN, P, CHELLAPPAN, A and ANNAMALAI, S. Pre-evaluation of fireaffected RC framed structure and its post-evaluation by load testing after repair; a case study, Journal of Structural Engineering, Vol. 33, No. 5, December 2006–January 2007, pp. 419– 427. 55. SIVAGNANAM, B. Damage assessment and rehabilitation of concrete structures: three case studies, Indian Concrete Journal, Vol. 76, No. 12, December 2002, pp. 764–770.
60
17th March 2008 Section 3.4.4 amended
56. NENE, RL and KAVLE, PS. Rehabilitation of a fire damaged structure, Proceedings of ACI International Conference Hong Kong, 1991, Evaluation and rehabilitation of concrete structures and innovations in design (Malhotra, VM, Ed.), ACI SP-128, Vol. II, American Concrete Institute, Detroit, 1992. 57. MORALES, EM. Rehabilitation of a fire damaged building, Proceedings of ACI International Conference Hong Kong, 1991, Evaluation and rehabilitation of concrete structures and innovations in design (Malhotra, VM, Ed.), ACI SP-128, Vol. II, American Concrete Institute, Detroit, 1992, pp. 1457–1472. 58. SMART, SAS. Fire damage assessment using non-destructive site techniques – Case studies, Concrete Solutions, Proceedings of the Second International Conference, St. Malo, France (Grantham, MG, Jaubertie, R and Lanos, C, Eds.), BRE Press, Watford, 2006, pp. 479–485. 59. DILEK, U. Evaluation of fire damage to a precast concrete structure; non-destructive, laboratory, and load testing, Proceedings of the American Society of Civil Engineers, Journal of Performance of Constructed Facilities, Vol. 19, No. 1, February 2005, pp. 42–48. 60. DILEK, U. Assessment of fire damage to a reinforced concrete structure during construction, Proceedings of the American Society of Civil Engineers, Journal of Performance of Constructed Facilities, Vol. 21, No. 4, July–August 2007, pp. 257–263. 61. TAERWE, L, POPPE, A-M, ANNEREL, E and VANDEVELDE, P. Fire damage assessment using non-destructive site techniques – Case studies, Concrete Solutions, Proceedings of the Second International Conference, St. Malo, France (Grantham, MG, Jaubertie, R and Lanos, C, Eds.), BRE Press, Watford, 2006, pp. 426–433. 62. PEKER, K and PEKMEZCI, B. Damage analysis for a fire exposed industrial building, Structural Engineering International, Vol. 13, No. 4, November 2003, pp. 245–248. 63. BOAM, K. and CROPPER, D. Midlands Links motorway viaducts; rehabilitation of a fire damaged structure, Proceedings of the Institution of Civil Engineers, Structures and Buildings, Vol. 104, No. 2, May 1994, pp. 111–123. 64. CABRITA NEVES, I, BRANCE, FA and VALENTE, JC. Effects of formwork fires in bridge construction, Concrete International, Vol. 19, No. 3, March 1997, pp. 41–46. 65. CALAVERA, RJ et al. Analysis and repair of a bridge damaged by fire in Grenada, Spain, Proceedings FIP Symposium, Budapest, Vol. 1, Environmental protection, strengthening and rehabilitation, May 1992, pp. 383–398. 66. BUSSELL, MN. The era of the proprietary reinforcing systems, Proceedings of the Institution of Civil Engineers, Structures and Buildings, Vol. 116, Nos. 3 and 4, August/November 1996, pp. 295–316.
61
17th March 2008 Section 3.4.4 amended
Appendix A
Case studies
While there are many published papers on the performance of concrete in fire and reports of large scale fire tests, there are few published papers that deal with the behaviour of actual concrete structures and the subsequent investigation. Tovey and Crook(6, 7) summarised the information gathered from over 100 fire-damaged structures. They concluded that, almost without exception, the structures performed well during and after the fire. Most of the structures were repaired and returned to service; when structures were demolished and replaced, it was generally for reasons other than the damage sustained during the fire. Some specific examples are given in the following sections, which have been divided into fires in buildings and fires affecting bridges. (It should be noted that fire is not generally a specific design consideration for bridges.) Several of the fires reported occurred during construction. In all but one case the structure was successfully repaired. The exception was a bridge in Birmingham where demolition and replacement was the preferred option as it caused less disruption to the road network. A1
FIRES IN BUILDINGS
Ingham(9) describes the damage caused by a large fire to the reinforced concrete frame of a ten storey building, which occurred during construction. The fire swept through the three upper storeys while the soffit formwork for the slabs was still in place. Although the damage was widespread, it was confined to the outer 5–30mm of the soffits and some columns. Tests on samples of the reinforcement showed that the steel had not been significantly affected, and hence the slabs could be economically repaired. The damaged concrete was removed by hydro-demolition or manually, and was reinstated using sprayed concrete. Berry(53) describes the repairs carried out following a fire in the basement of a conference centre. Five of the columns were so seriously affected that they it was necessary to prop the adjacent beams carrying the slab above. All damaged concrete was removed and the columns and beams recast using flowing concrete. The opportunity was taken to increase the dimensions of the columns and the column capitals to improve the durability and the structural performance. The most severely damaged areas of the soffit were strengthened using bonded steel plates. Deflection of the slab during the fire had caused cracks on the top surface, which were sealed with epoxy resin. Otherwise the floor above was unaffected by the fire and continued to be used for exhibitions throughout the duration of the repairs to the basement below. Srinivasan et al(54) gives details of the investigation and repair of an eight-storey reinforced concrete frame building damaged by a fire on the fourth floor. The condition of the concrete was assessed by UPV testing, rebound hammer and core sampling. Samples of the reinforcement were removed for tensile testing. Repairs were required to the slab, beams and columns. Apart from some grouting of cracks in the slab all the repair was carried out using sprayed concrete. The repaired slab was load tested to 1.25 times the design live load, and performed satisfactorily. Sivagnanam(55) describes the repair of a reinforced concrete framed building in a fertiliser plant in India that was severely damaged by fire. UPV testing was carried out on the damaged columns and some core testing. Samples of the reinforcement were removed and tested, indicating that the yield stress had been reduced by more than 50% in the worst affected 62
17th March 2008 Section 3.4.4 amended
areas. The damaged concrete was removed, additional reinforcement fixed and sprayed concrete applied to the affected areas. The structure was brought back into service after only three weeks. Nene and Kavle(56) report the effects of a major fire in a three storey reinforced concrete frame building in a chemical factory in India and its subsequent repair. From examination of the debris it was concluded that the temperature reached 700°C in the worst areas and about 300–400°C elsewhere. The Authors particularly noted the damage to the roof beams caused by the extreme differential temperatures – 700°C on the soffit with the top surface cooled to about 25°C by the fire-fighting water. The capacity of some columns was reduced by 60% while that of some of the beams was reduced by 50%. These reductions meant that it was necessary to prop the complete building prior to and during repair, which was carried out using sprayed concrete throughout. Morales(57) describes the results of a fire at an armed forces headquarters building in the Philippines, which had suffered two previous fires. The original concrete would appear to have been of low strength, additionally weakened by the fires. Following the third fire, damage to part of the structure was so severe that it had to be demolished and rebuilt. Elsewhere it was considered appropriate to repair the beams and columns. However, the existing design did not satisfy current seismic requirements. Reinforced concrete jackets were installed around the beams and columns, with additional seismic reinforcement, and the concrete cast in formwork. Smart(58) gives details of the damage caused by a vehicle fire in the lower ground floor parking area of a small block of flats. Two parking bays were affected by the fire, causing some damage to the precast units forming the slab above. Of the eight units, three required significant repair and strengthening; the remainder required only superficial repairs and/or cleaning and redecoration. In a second Case Study, a three storey timber framed structure on a reinforced concrete pile-supported raft slab caught fire during construction and burned down completely. The remains were left to smoulder for three days. However, much of the slab suffered little or no damage. In some limited areas, the surface of the concrete had scaled to a depth of about 38mm; the steel had 65mm cover and was deemed not to have been affected. The weakened concrete was removed to a depth of 55mm and reinstated. Dulik(59) describes work carried out following a vehicle fire in a multi-storey car park. The fire affected two precast prestressed double-tee units, made with lightweight aggregate concrete. Local to the fire the webs were cracked and spalled; elsewhere the concrete was discoloured. Ultrasonic pulse velocity testing was used to determine the extent of affected concrete. Cores were removed from damaged areas and cut into 25mm thick disks, which were used to determine the concrete strength, dynamic modulus of elasticity and air permeability index. Analysis of the concrete properties at small depth increments indicated the extent of damage. However, there remained concern about whether there had been relaxation of the prestressing tendons. As a result, a load test was carried out on the two affected doubletees, which were isolated from their neighbours. Testing, which was carried out before any repairs to the concrete, indicated that the prestressing steel had not been significantly affected. In a later paper Dilek(60) provides information on the investigation carried out on a basement wall that was damaged during construction when the asphalt-based waterproofing layer and the insulation material caught fire. The depth of the affected concrete was determined using an ultrasonic pulse velocity method. The dynamic modulus of elasticity and air permeability 63
17th March 2008 Section 3.4.4 amended
were determined on 25mm thick specimens cut from cores extracted from the wall. In addition cores were removed from two locations, sawn into short lengths and tested in compression to give average concrete strengths. Once the damaged concrete had been removed, to a depth of around 25mm, Windsor probe tests were carried out to check that the strength of the exposed concrete was satisfactory. The concrete was then reinstated and the waterproofing and insulation reinstalled. Taerwe et al(61) report the results of a fire test on a precast portal frame building. After the fire one of the 18m long pre-tensioned prestressed roof girders was removed. There was significant spalling of the concrete, which extended to the depth of the shear stirrups at some locations. The cover was reinstated using sprayed concrete and the beam was then load tested, and carried about 2.5 times its service load before one of the tendons failed. This was an adequate margin of safety, even though subsequent testing indicated that some of the tendons near the bottom corners of the beam had lost about 35% of their tensile strength. Peker and Pekmezci(62) give details of the investigation carried out on a four-storey industrial building that suffered a major chemical fire. Maximum temperatures were estimated to have been over 680°C and to have been maintained for much of the duration of the fire (3½ hours). Following a visual survey of the damaged structure, samples of reinforcement were tested and the properties of the concrete assessed from cores and UPV measurements. Finally a complete 3-D finite element analysis of the building was used to assess the load capacity. As a result of the investigations, it was concluded that repair of the building, although comparatively more expensive that demolition and rebuilding, would allow the plant to resume production more quickly. The Authors identified that the original design of the building, which was built in 1968, did not satisfy the current seismic design code. However, the paper does not indicate what steps were taken to upgrade the structure during the repair. Donnelly(52) gives brief details of a building in Jakarta that was damaged by a major fire. As part of the repair, carbon and glass fibre sheets were bonded externally to beams to provide additional flexural and shear capacity. Fire broke out in the concrete core ventilation shaft of an office block in London during the construction. The fire, which lasted two to three hours, swept through the seven story shaft causing a ‘chimney effect’. The concrete walls were built in C32/40 concrete with a nominal cover of 35mm and had two layers of reinforcement on each face; they were protected with class 1 fire rated insulation on the inside. Visual examination and subsequent petrographic and core tests revealed that the reinforcement was not exposed and hence its strength was not significantly affected. The explosive spalling resulting from the fire was removed and the damage made good with new concrete. The walls were reassessed in accordance with BS 8110. A2
FIRES UNDER BRIDGES
Boam and Cropper(63) give details of the damage cause by a petrol tanker that crashed into the sub-structure of part of the Midlands Link elevated motorway in Birmingham, igniting on impact. The fire resulted in damage to the deck and supports. Straightforward repair of the concrete structures was complicated by the fact that there was existing severe chloride contamination and some of the reinforcement detailing was not in accordance with current practice. In addition the structure was heavily trafficked, forming one of the key routes into and around Birmingham. Various options were considered for repair or demolition and
64
17th March 2008 Section 3.4.4 amended
reconstruction. However, the need the need to minimise disruption to the road network was paramount and the decision was taken to demolish and reconstruct the damaged piers and deck span. Cabrita Neves et al(64) report on the damage caused by fire in the formwork beneath the recently concreted span of a bridge in Portugal. The concrete was only 30 hours old when the fire occurred and hence it had a high water content, leading to major spalling. The falsework failed, and but the span did not collapse. Although the test results showed acceptable values for the concrete strength, the extent of cracking and deformation led to the decision to demolish and reconstruct the span. The adjacent span was 30 days old at the time of the fire. There was some spalling of the soffit, but the prestressing tendons were not affected, because of the large covers. Sprayed concrete was used to repair the damaged areas and the bridge finally checked with a load test. The same paper reports briefly on formwork fires on two other bridges. In one, the anchorage zones of the tendons (which had already been grouted) had to be demolished and reconstructed. Calavera Ruiz et al(65) give details of another formwork fire beneath a newly-constructed bridge in Spain. The concrete was about one month old and the post-tensioning tendons had been stressed and grouted when the fire broke out, damaging three spans of the bridge. The affected area was surveyed using UPV, and samples of concrete and reinforcement were removed for testing. It was assumed that the heat had increased the relaxation of the tendons by 15% but that, because of the cover and grouting, their properties were otherwise unaffected. The structure was reanalysed assuming a loss of 60mm of concrete in the affected areas and the increased relaxation, from which it was concluded that the performance of the bridge was not significantly altered. The damaged areas were reinstated using sprayed concrete.
65
17th March 2008 Section 3.4.4 amended
Appendix B
Worked examples
The design examples have been kept to simple member design to show the basic approach for determining the strength loss due to the effects of fire (see Chapter 3). A more rigorous analysis taking into account the temperature variation within and along the member, changes in frame or member stiffness and the effects of any likely redistribution may be required. In addition to the ultimate strength the effects of deflection and other serviceability conditions may have to be assessed. As indicated in Section 3.4.1, the remaining concrete in columns and walls may be highly stressed at the time of the repair, which may affect the distribution of loads between the new and old materials; this has not been considered in these examples. B1
INTRODUCTION
Fire damaged concrete varies in strength as a result of the temperatures attained during the fire. That section of concrete wholly within a range of 300–100°C is referred to as the intermediate zone and its strength is taken to be 0.85 times the original ambient strength. Alternative values will be obtained dependent on the actual temperature distribution. Concrete heated to less than 100°C is considered to have no appreciable change in strength. Redesign of members is much influenced by the relative thickness of these layers within the compression block and the residual strength obtained. Similarly the residual strength of reinforcement is related to its temperature attained in the fire. Considering the residual strength of both concrete and reinforcing bars it is necessary to re-analyse the member; where necessary new bars or concrete are added to restore strength. In these examples, which are based on sprayed concrete repairs, it is assumed that the temperature profile of the concrete, including the 300°C zone, has been established as described in Chapter 2, e.g. by cores, estimation of fire intensity and duration. The examples are based on the design methods given in Eurocodes 0, 1 and 2 (BS EN 1990(3), BS EN 1991(4) and BS EN 1992(5) respectively). The residual strength factors for concrete and steel come from Figure 3 and Figure 10. Unless otherwise stated, the clauses specified in the examples relate to Part 1-1 of BS EN 1992. The details of the building are shown in Figure 20. The characteristic strength of the steel has been taken as 500MPa through out the examples. For older structures the characteristic strength of the steel will be lower (see Appendix C) and mild steel may have been used for shear reinforcement and links in columns. A partial safety factor of 1.5 has been applied to the strength of the original concrete but, as indicated in Section 3.4.1, a lower factor could be used if the actual strength had been determined experimentally. Similarly a modified partial safety factor could be applied to the steel strength if the actual value is determined from tests on samples removed from undamaged parts of the structure. B2
EXAMPLE 1 – CONTINUOUS SLAB
One way spanning continuous floor slab (Panel 202: Figure 20)
66
17th March 2008 Section 3.4.4 amended
[NOTE: Transverse bars to be added within existing bars top and bottom.] Figure B1: Original slab profile. Dimensions
4.25m × 5m × 0.175m
Main reinforcement (Figure B2)
10mm dia. high yield bars at 225mm centres bottom – centre span 12mm dia. high yield bars at 175mm centres top – supports Cover – 20mm
The damage to the slab was assessed as being Class 2. From the cores it has been established that the depth of the pink/red zone is 15mm (Figure B2).
Figure B2: Damaged slab. Since the pink/red zone did not reach the steel, the steel did not reach temperatures above 300°C. Figure 10 indicates that the steel has suffered no permanent damage thus the slab will be satisfactory at the centre of the span but needs to be checked at the supports. Redesign Loading Variable (qk) Office building (including partitions)
Permanent (gk)
4.0kN/m2 1.0kN/m2 2.5kN/m2
Finishes
67
17th March 2008 Section 3.4.4 amended
Sprayed concrete (Extra cover)
0.48kN/m2
Concrete self weight
4.375kN/m2 5.0kN/m2
Total
7.355kN/m2
This scenario satisfies the requirements of clause 5.1.3(1)P in BS EN 1990 and the National Annex; therefore the simplified arrangement will be considered, i.e. all spans loaded. Combination of Actions FD
= 1.25Gk + 1.5Qk = 9.194 + 7.5 = 16.69kN/m2 (Expression 6.10b BS EN 1990 and National Annex)
Analysis and design Consider a one metre width FD
= 16.69 × 1.00 × 4.25 = 70.95kN
(a) Middle interior span Design bending moment MEd
= 0.063FL = 0.063 × 70.95 × 4.25 = 18.996kNm
Design resistance moment d = 175 – 20 – 5 = 150mm K =MEd/(bd2fck) =18.995 × 106/ (1000 × 1502 × 35) = 0.024
(Figure 3.5 in BS EN 1992-1-1)
z = 0.95d MRd = Asfydz = 349 × 0.87 × 500 × 142.5 × 10-6 = 21.64kNm This section is therefore shown to be adequate (NB. If the slab was not damaged at the supports no further checks would be necessary and repairs would be considered non-structural, cosmetic or for durability and fire resistance purposes only.) (b) First interior support Concrete residual strength
68
17th March 2008 Section 3.4.4 amended
Figure B3: Temperature profiles. Temperatures reached 300°C at 15mm (Figure B3) and 100°C at 85mm. The residual strengths are: Undamaged concrete Damaged concrete
= kcfcd = 1.00 × 35 = 35 = kcfcd = 0.7 × 35 = 24.5
MPa MPa
The fire affected the bottom of the slab and the pink/red zone extended into an area where the concrete is required to resist compression at the supports. It is therefore necessary to check the section at the supports. Design bending moment MEd = –0.086FL = –0.086 × 70.95 × 4.25 = 25.932kNm Design resistance moment In this calculation, as a first step, the section is to be considered as being the depth of the undamaged concrete. Reduced effective depth d = 175 – 50 – 20 – 6 = 99mm K = MEd/bd2 kcfck = 25.932 × 106/(1000 × 1342 × 35) = 0.013 (Figure 3.5 in BS EN 1992-1-1) z = d/2[1+ (1 – 3.53K)0.5] = 91.882mm < 0.95d = 94.05mm MRd = Asfydz = 646 × 0.87 × 500 × 91.88 × 10-6 = 25.832kNm Since this value is insufficient it is necessary to refine the calculation by using the damaged concrete. Try using all the damaged concrete down to the pink/red zone (300°C) with the minimum concrete residual strength (i.e. at 300°C). Reduced effective depth
69
17th March 2008 Section 3.4.4 amended
d = 175 – 20 – 15 – 6 = 134mm K = MEd/bd2 kcfck = 25.932 × 10 6/ (1000 × 1342 × 24.5) = 0.059 (Figure 3.5 in BS EN 1992-1-1) z = d/2[1+ (1 – 3.53K)0.5] = 126.62mm < 0.95d = 127.3mm MRd = Asfydz = 646 × 0.87 × 500 × 126.62 × 10-6 = 35.59kNm …….. OK This section has been shown to work The calculation could have been further refined determining the average loss in strength within the compression block with or without inclusion of the sprayed concrete if deemed necessary. In this example the actual compression block lies wholly above the 100ºC line and therefore a damage factor of 1.0 would be applicable in determining the resistance moment (Figure B3). Notes: 1. No check has been made for shear as this is not normally critical for solid slabs. 2. If the above calculation had not demonstrated that the proposed repair was adequate then a more detailed assessment could be undertaken by carrying out a rigorous analysis, re-considering loadings or strengthening the section. To restore the cover to the steel and provide for structural repairs at supports, sprayed concrete would normally be applied. If only minor damage exists at the support such that there is no significant loss of strength or durability, a number of alternative repairs might be considered. In this instance a large area of slab has been affected and a whole area is to be cut back to the reinforcement to remove all damaged concrete. The repair will consist of sprayed concrete, reinforced with a layer of fabric, which will be 6mm thick overall, spaced 12mm off the base of the concrete with 20mm cover. Thus 20mm of concrete will be removed and 38mm of sprayed concrete is to be added. The design must be checked for the new situation. (See Figure B4).
[NOTE: “gunite” to be replaced by “sprayed concrete” and dimension of 40 to be added for sprayed concrete thickness.] Figure B4: Repaired section. B3
EXAMPLE 2 – SIMPLY SUPPORTED TEE BEAM
Simply supported tee beam (Beam 141: Figure 20) 70
17th March 2008 Section 3.4.4 amended
Figure B5: Original beam profile. Dimensions
0.3m web × 5m length
Main reinforcement
3 × 25mm dia. high yield bars
Links
10mm dia. at 250mm centres
Minimum cover (all reinforcement) 25mm The damage to the beam was visually assessed as being class 3. The pink/red zone is shown in Figure B6 and the temperature profiles in Figure B7.
Figure B6: Damaged beam profile.
71
17th March 2008 Section 3.4.4 amended
Figure B7: Temperature profiles.
Fire damage Reinforcement: From Figure B6 it may be seen that the central bar is within the pink/red zone but still covered with concrete. The pink/red zone of 300°C temperature profile occurs at 60mm from the surface and steel temperatures are assessed to have reached 370°C. The depth of the pink/red zone at the arrises indicates that spalling occurred during the fire and thus it will be assumed that the two outer bars have been directly exposed to the fire. The average temperature reached in the bars was 700°C (Figure B8).
Figure B8: Temperature profile at corner. From Figure 10 the residual strengths in the bars are: Outer bars Inner bars
= ks fyd = 0.6 × 500 = 300 = ks fyd = 1.0 × 500 = 500
72
17th March 2008 Section 3.4.4 amended
The average fire residual steel strength for the main reinforcement is therefore: [(1 × 500) + (2 × 300)]/3 = 366.67MPa Concrete: All concrete in which the temperature has exceeded 300°C (i.e. to the depth of the pink/red zone) will be removed and reinstated with sprayed concrete. For the remaining concrete assess the residual concrete strength (see Figure B6) For the slab, temperatures reached 300°C at 35mm (Figures B6 and B7) and 100°C at 85mm. The residual strengths are: Undamaged concrete Damaged concrete
= kcfcd = 1.00 × 35 = 35 = kcfcd = 0.85 × 35 = 29.75
MPa MPa
Hence the average concrete residual strength in the compression flange is: [(90 × 35) + (50 × 29.75)] / (175 – 35) = 33.125MPa For the beam, temperatures reached 300°C at 40mm (Figures B6 and B7) and 100°C at 90mm. For the shear calculations the minimum residual strength should be used as opposed to the average residual strength of the concrete. In the retained damaged concrete zone the maximum temperature reached was 300ºC and hence the residual strength for concrete exposed to this temperature is used Concrete minimum residual strength = kcfcd = 0.7 × 35 = 24.5MPa Area of main steel provided = 1470mm2 Reduced tensile force
= 1470 × 0.87 × 366.67 × 10-3 = 470kN
Loading Original design ultimate load = 70.953kN/m Add 20mm slab sprayed concrete Add beam sprayed concrete Total = 3.10 × 1.25 New design ultimate load
= 1.90kN/m = 1.20kN/m = 3.875kN/m = 74.82kN/m
Analysis and design Redesign bending moment
= wL2/8 = 74.82 × 52 / 8 = 233.82kNm
73
(Expression 6.10b)
17th March 2008 Section 3.4.4 amended
Effective flange width beff = beff1 + beff2 + bw
(Expression 5.7)
beff1 = beff2 = (0.2b1 + 0.1l0)
(Expression 5.7a)
b1 = b2 = (4250 – 220)/2 = 2015mm l0 = l = 5000
(simply supported beam)
beff1 = (0.2 × 2015) + (0.1 × 5000) = 903 < 1000 < 2015
(Expression 5.7b)
beff = (2 × 903) + 220 = 2026mm, d = 478mm K = M/bd2kcfck = 233.82 × 106/ (2026 × 4782 × 33.125) = 0.0152 z = d/2[1 + (1 – 3.53K)0.5] = 471.83mm > 0.95d = 454.1mm x = 2.5(d – z) = 59.75 < 1.25hf Therefore the neutral axis is in the flange and the beam may be designed as a rectangular section. Required tensile force
= 233.82 ×103 / 454.1 = 514.92kN
Residual tensile force
= 469.77kN
Additional tensile force
= 45.15kN
Additional area of steel required
= (45.15 × 103) / (0.87 × 500) = 103.79mm2
Provide 2 × H10 as new bars (157mm2) Shear: Check the area of links required taking into account the reduction due to the fire damage. Check shear at distance d from face of support using Clause 6.2.1(8) of BS EN 19921-1. Shear force at d from face of support VEd Shear stress
= 74.82 (5/2 – 0.19 – 0.454) = 137.08kN = 137.08 × 103/ (0.9 × 220 × 478) = 1.45N/mm2
Maximum shear resistance VRd,max = αcwbwzvkcfcd/(cot θ + tan θ)
(Expression 6.9 and National Annex)
where αcw bw z
(Clause 6.2.3 and National Annex)
= = =
1 220mm as before 454.1mm as before 74
17th March 2008 Section 3.4.4 amended
v θ
= =
0.6[1 – fck/250] = 0.516 angle of inclination of strut
(Expression 6.6N)
VRd,max = 1 × 220 × 454.1 × 0.516 × (24.5/1.5)/(2.5 + 0.4) = 290kN…… OK Shear links: shear resistance with links VRd,s = (Asw/s) zfywd cotθ Thus
(Expression 6.8)
Asw/s > VEd/zfywd cotθ Asw/s > 137.08 × 103/ (454.1 × 0.87 × 500 × 2.5) = 0.278 or 278mm2/m
Minimum required Asw,min/s = ρw,minbw sinα
(Clause 9.2.2(5))
where ρw,min = 0.08fcu0.5/fyk = 0.08 × 24.50.5/500 = 0.00079 (Expression 9.4 and 9.6N and National Annex) bw = 220 α = angle between shear and longitudinal axis; for vertical reinforcement sin α = 1.0 Asw,min/s = 0.00079 × 220 × 1 = 0.174 or 174mm2/m …..Not Critical Maximum spacing of links smax = 0.75d = 358.5
(Clause 9.2.2(6))
Original provision 10mm dia. mild steel (2 legs) at 250mm spacing Asw/s = 157/0.25 = 628mm2/m There is no requirement for additional shear reinforcement. A small additional amount will be added however to support the added main bars and the nominal reinforcement within the repair concrete. Provide nominal links H8 at 400mm centres to locate added main bars. Since the original links are essentially adequate for redesign, the new links may be embedded in the side of the beam as shown in Figure B9 as they are principally being used to locate and support added main bars. Where such links are required to contribute to the shear strength they should be secured through the slab as shown in Figure B10 or alternatively held by anchor plates and nuts attached to threaded ends of the links or similar to provide anchorage into the compression zone. The final profile of the repaired beam is shown in Figure B11.
75
17th March 2008 Section 3.4.4 amended
[Note: Change 2T16 to 2H10. Change R6 to H8.] Figure B9: Support of added main bars.
Figure B10: Anchorage of shear links.
[Note: “gunite” to be replaced by “sprayed concrete”.] Figure B11: Repaired section.
76
17th March 2008 Section 3.4.4 amended
Note: Checks for shear between the web and the flange are required by BS EN 1992-1-1. For this scenario it was not critical and therefore the check has not been included here. Where significant damage exists to the flange and web junction this may be significant B4
EXAMPLE 3 – AXIALLY LOADED COLUMN
This example is only for an axially loaded column; in some cases there will be a need to consider additional bending moments induced by permanent displacement of the firedamaged column, see Section 3.4.4.
[Note; Change T25 to 25mm dia high yield bars.] Figure B12: Original column profile. Axially loaded column (column 22: Figure 20) Dimensions
380mm × 380mm
Main reinforcement (Figure A8) 8 × 25mm dia. high yield bars, links 10mm at 300mm centres. Minimum cover (to all reinforcement) 25mm The damage to the column was visually assessed as being class 2 (Figure B13).
Figure B13: Damaged column profile. 77
17th March 2008 Section 3.4.4 amended
Redesign Effective height of columns 3.9m (both directions) Le/b = 3900/380 = 10.26 Ultimate axial load = 4900kN Depth of pink/red zone as determined from tests is equal to 35mm at which the concrete is taken to have reached 300ºC. The temperature of the main corner reinforcement has not exceeded 425ºC (Figure B14).
Figure B14: Temperature profiles. The depth at which the concrete reached only 100ºC can be seen from Figures B15 to be approximately 75mm.
Figure B15: Temperature profile at corner. As it is not intended to supplement the main reinforcement, the sprayed column may be square in section. The dimensions of the repaired column may be established directly assuming the repaired section does not fundamentally affect the behaviour of the frame. Material residual strength
78
17th March 2008 Section 3.4.4 amended
Temperatures reached 300°C at 35mm (Figure B14) and 100°C at 75mm. The residual strengths are: Undamaged concrete (0–100ºC) = kcfcd = 1.00 × 35 = 35 Damaged concrete (100–300ºC) = kcfcd = 0.85 × 35 = 29.75
MPa MPa
Analysis and design Check the ultimate capacity of the residual column section NRd
= Ackcfcd + Askcfyd
(Expression 5.39)
Area of steel = 3930mm2 Area of undamaged concrete (T < 100°C) Ac = 2302 = 52,900mm2 Area of damaged concrete (100ºC < T< 300°C) Ac = 3102 – 2302 – 3930 = 39270mm2 NRd = (52900 × 35/1.5) + (39270 × 29.75/1.5) + (3930 × 500/1.15) = 3722kN The sprayed concrete must have a capacity of: 4900 – 3722 = 1178kN and hence the required area of sprayed concrete is: 1178 × 103 / (30/1.5) = 58906mm2 The length of the side of the sprayed column is: (58906 + 3102)0.5 = 394mm
Say 400mm
If this new size of side is impractical, it may be necessary to add reinforcement to supplement the main bars; in this case, consideration should be given to the difficulty of ensuring that the shell can act compositely. Alternatively, sprayed concrete with a higher characteristic strength may be specified. Provide fabric reinforcement in the repair, see Figure B16.
79
17th March 2008 Section 3.4.4 amended
[Note: “gunite” to be replaced by “sprayed concrete” and “mesh” by “fabric.] Figure B16: Repaired section.
80
17th March 2008 Section 3.4.4 amended
Appendix C
Historical information
Unless indicated otherwise, all the Codes and Standards in this Appendix were issued by the British Standards Institution, London. C1
DESIGN CODES
The first British design code for concrete structures was introduced in 1934. Subsequently new codes have been introduced, as indicated in Table C1. Invariably there was some overlap at each transition between an old and a new code. In addition codes are subject to revision before they are eventually replaced. Table C1: The development of design codes. Date
Design code
1948
CP 114, The structural use of normal reinforced concrete in buildings
1959
CP 115, The structural use of prestressed concrete in buildings
1965
CP 116, The structural use of precast concrete
1972
CP 110, Code of practice for the structural use of concrete
1973
1984
Technical Memorandum (Bridges) BE 1/73 Reinforced concrete highway structures* Technical Memorandum (Bridges) BE 2/73, Prestressed concrete highway structures* BS 5400: Part 4, Code of practice for design of concrete bridges
1985
BS 8110, Structural use of concrete
1990
BS 5502, Buildings and structures for agriculture
2004
BS EN 1992, Eurocode 2: Design of concrete structures
* Issued by the Department of Transport (now the Highways Agency). C2
SPECIFICATION AND STRENGTH OF HISTORIC CONCRETE
The strength of the concrete in historic structures is likely to be very variable. The only sure way of determining it is to carry out appropriate tests. However, if design information is available, guidance on the likely minimum concrete strength may be obtained from the Codes of Practice current at the time. The First Edition of Reinforced Concrete Designers’ Handbook by Charles Reynolds (published by Concrete Publications in 1932) proposes a series of concrete mixes A to F, with proportions ranging from 1:3:6 up to 1:1:2. The corresponding working stresses for the hardened concrete range from 400psi (400lb/sq. inch) up to 875psi (say 2.8–6.0MPa), though it is not clear what cube strengths these would equate to. The specification of concrete in accordance with CP 114, was on the basis of cement: fine aggregate: coarse aggregate ratios by volume, namely 1:1:2, 1:1½:3 and 1:2:4. The code gave minimum values for the 28-day cube strengths associated with the three Nominal Mixes as 4500, 3750 and 3000psi respectively (i.e. approximately 31, 26 and 21MPa). 81
17th March 2008 Section 3.4.4 amended
CP 115 specified minimum 28-day cube strengths of 6000psi (approximately 41MPa) for pretensioned concrete and 4500psi (approximately 31MPa) for post-tensioned concrete. The Metric version of CP 114 (in 1969) retained the three Nominal Mixes but added a Table of Standard Mixes. There were three specified 28-day cube strengths (21, 25.5 and 30MPa) and mix proportions were given for different aggregate sizes and slumps. The Ninth Edition of the Reinforced Concrete Designer’s Handbook by Reynolds and Steedman (published in 1981) presents the same information (for 19mm aggregate only) but defines the mixes in terms of the letters A, B and C; Standard Mix A is equivalent to the 21MPa mix in CP 114, Mix B is equivalent to 25.5MPa and Mix C is equivalent to 30MPa. The Handbook also includes Designed Mixes D and E, with 28-day cube strengths of 40 and 50MPa respectively. Design in CP 110 (and all subsequent Codes) was based on specified concrete grades. Minimum concrete grades were specified for different types of element: 20MPa for reinforced concrete, 30MPa for post-tensioned prestressed concrete and 40MPa for pre-tensioned prestressed concrete. C3
REINFORCEMENT
C3.1
Early reinforcement systems
In the early days of reinforced concrete (from the 1890s to 1920s), a number of proprietary reinforcement systems were used. Many of these were developed in France, Germany and USA and imported into the UK. Details can be found in Bussell(66), which also covers various patented floor systems. The importance of the bond between the reinforcement and the concrete was identified at an early stage. Most of the systems are recognisable as variations on modern bars, with a wide variety of surface characteristics. For example, the Hennebique System used plain round (mild steel) bars with flattened ‘fish tail’ ends for anchorage. A very unusual product was the Kahn bar, which consisted of a square section with two projecting strips on diagonally opposite corners. These were slit along short lengths and bent up to form shear reinforcement. In the USA and UK the system was adopted by the Trussed Concrete Steel Company, which became abbreviated to Truscon. C3.2
Standards and strengths
The specified yield stresses given in the Standards for reinforcement have varied over the years; some values are given in Table C2. Table C2: Reinforcement standards and associated strengths. Plain round mild steel bars BS 785: 1938 & BS 785: Part 1, 1967 BS 4449: 1969, 1978 & 1988 Cold worked deformed bars BS 1144: 1943 BS 1144: 1967
36,000psi (250MPa) for bars up to 1½ inches (38mm) 250MPa 70,000psi (493MPa) for bars less than 3/8 inch (10mm) 60,000psi (423MPa) for bars over 3/8 inch (10mm) 50,000psi (352MPa) for twin twisted bars 66,000psi (465MPa) for bars up to 5/8 inch (16mm)
82
17th March 2008 Section 3.4.4 amended 60,000psi (423MPa) for bars over 5/8 inch (16mm) 72,000psi (507MPa) for 72 Grade ribbed bars with rolling mark 460MPa for bars up to 16mm 425MPa for bars over 16mm
BS 4461: 1969 Hot rolled deformed bars BS 785: Part 1, 1967
60,000psi (423MPa) for medium tensile bars 70,000psi (493MPa) for high tensile bars 410MPa for all size 460MPa for bars up to 16mm 425MPa for bars over 16mm 500MPa for all bar sizes
BS 4449: 1969 BS 4449: 1978 BS 8666: 2005
C3.3
Detailing symbols
The symbols used on reinforcement drawings to indicate the type of reinforcement have been specified in various Standards over the years, see Table C3. Table C3: Detailing symbols. Code BS 1478:1948 BS 1478:1964 BS 1478:1967 BS 4466:1969 BS 4466:1981 BS 4466:1989
BS 8666: 2000
BS 8666: 2005
Symbols Not specified: many variants for square and twisted bars. MR = round mild HR = round high yield HS = square high yield M = round mild steel H = round area high yield S = square area high yield R = round mild steel Y = round area high yield R = Grade 250 T = Grade 460/425 type 2 R = Grade 250 plain T = Grade 460 type 2 S = Grade 460 stainless W = Grade 460 plain D = Grade 460 type 1 R = Grade 250 plain T = Grade 460 type 2 S = Grade 460 stainless W = Grade 460 plain D = Grade 460 type 1 X = Type of reinforcement not included in the above but having properties defined in the design or contract specification. H = Grade B500A, B500B or Grade B500C according to BS 4449: 2005* HA = Grade B500A according to BS 4449: 2005 (similarly HB and HC) S = Specified grade and type of ribbed stainless steel to BS 6744: 2001 X = Type of reinforcement not included in the above but having properties defined in the design or contract specification. * In accordance with the 2008 amendment, this only applies for diameters up to and including 12mm. For bar diameters above 12mm, H = Grade B500B or Grade B500C only.
83