Fire engineering Brief of an office building in City centre

Staircases from 7/F to roof

• The width of door accessing staircases from 7/F to roof floor = 1050mm
• The width of staircases from 7/F to roof floor = 1500mm
• The height of staircases from 7/F to roof floor = 2500mm.
• Staircase from 7/F to roof floor between 2 landings has 15 risers. Height of each riser is 150mm.
• Treads width of staircase from 7/F to roof floor = 250mm.
• Staircases from 7/F to roof floor have3 landing are provided. There are top, middle and bottom landing.
• The size of top and bottom landing = 2.6m (length) X 3m (width).
• The size of middle landing = 1.5m (length) X 3m (width).

4. Occupant Characteristics

Building occupants are characterized by their number, their capability to react to emergencies, their degree of familiarity of the building, and their ability to fire actively act in fire engineering situations.

4.1 Population

Occupant numbers are 2,100 persons since there are 2,000 seats and maximum 100 staff to hold the assembly.

4.2 State

For the no assembly period, there is just a few staff. For setting up an assembly, there is maximum 100 staff whose hold the assembly. We are just focusing on the assembly carrying period because the situation is much more person than the other situation.

4.3 Building Familiarity

Occupants are expected to be familiar with the only one access and egress routes from the storey. There is also only an entrance (or exits) in an emergency4. Besides, the staffs whose hold the assembly has been well trained for assist occupant evacuation.

4.4 Emergency Training

Staffs are to be trained in early fire fighting and emergency response. Staffs are to be advised on the actions necessary on the activation of the emergency warning and intercommunication system.

5. Project Scope

There are inadequate staircases providing for means of egress from the storey according to MoE. It is mean that people have not adequate time to egress from the storey for just provided one staircase during the fire. Therefore, the project scope is to provide trial design strategy and analysis of alternate trial design which represents deviation from the prescriptive solutions. It is mean that people have adequate time to egress to a place of safety during the fire.

6. Project Goal

Project goals are generated in reaction to the specifics of each project such as building type, intended use, intended occupancy, and site location. For this case, a office building have a 2,520㎡assembly hall on seventh floor for 2,000 staff and maximum 100 staff to hold assembly to attend. The Goals is to provide adequate time to egress to a place of safety to minimize fire-related injuries and prevent under loss of life when fire.

7. Project Objective

The objectives are developed to define each goal qualitatively in a manner which can then be more easily quantified for the analyses. The objectives for the goal are:

1. Provide adequate time for those people not intimate with the first materials burning to reach a place of safety on below floor and then reach a place of safety by the means of egress on below floor in where without being overcome by the effects of fire and fire effluents.
2. Provide adequate time for those people outside the floor of fire origin to reach a place of safety on below floor and then reach a place of safety by the means of egress on below floor in where without being overcome by the effects of fire and fire effluents.
3. Limit fire spread and thermal effects to threaten those people.
4. Limit smoke spread to threaten those people.

8. Performance Criteria

For the purpose of goal, adequate evacuation time should be provided when fire. A timeline chart for evaluation of egress safety is shown as below (Figure 1).

Figure 1: Timeline chart for Evaluation of Egress Safety (Technical Proposal).

Where

td=Recognition Phase: time from being alerted by a cue to knowing there is an emergency fire (includes acts such as investigate)

tr=Coping Phase: time from knowing there is an emergency to beginning of escape (includes acts such as fight fire)

te=Esczpe/ Evacuation Phase: time from the end of the coping phase i.e. beginning escape (evacuation) to leaving the building

tf=Available Safe Egress Time (ASET)

Required safe Egress Time (RSET) = td+ tr + te 

Provided adequate evacuation time can be increased ASET or decreased RSET. The following performance criteria are considered to increase ASET:

1. Smoke producing and spread
2. Fire and Thermal Effects
3. Toxicity

The following performance criteria are considered to decrease RSET:

4. Fire Detection and Alarm Time
5. Pre-movement Time
6. Travel Time

9. Design Fire Scenarios and Design Fire

The design fire scenarios and design fire is developed from all of the possible scenarios. Design fire scenarios are descriptions of reasonable yet severe fire that could possibly threaten a building or its occupants. Several major assumptions are made with respect to narrowing the field of design fire scenarios:

1. The case for multiple simultaneous fire will not be examined
2. The case for a thoroughly executed arson will not be used
3. The suppression system will be assumed to be in working order and operation according to their respective designs.
4. Simultaneous occurrence of fire and other natural disasters such that passive and/or active systems are compromised will not occur.

9.1 A Fire Located in nearest position of the staircase of the floor (5m apart from front of staircase)

There are no fire load from staircase to front of 5m apart (Smoke zone 14) and sprinkler and smoke exhaust system is installed in the zone. A drencher system also is installed to separate all smoke zones. Therefore, it is reasonable to assume all material ignite in the smoke zone (Smoke zone 11) nearest the staircase to be the worst case of fire scenarios.

The wood frame with latex foam cushions chair is used. The detail of the chair is shown as Appendix D. The furniture heat release rates are show as below (Table 2):

Table 2: Furniture Heat Release Rates

There are 200 seats in the smoke zone 11 so the heat release rate (HHR) can be calculated.

The HHR of chair is 85kW for 11.2kg by table 2.

The weight of a chair = 20lb × 0.456kg/lb = 9.12kg

The heat release rate = 85kW × 9.12kg × 200 seat / 11.2kg = 13,843kW

Therefore, a 14,000kW is used in the design fire scenarios.

10. Trial Design and Evaluation

There are several trial designs for this study to increase ASET and decrease RSET.

10.1 Smoke Control

The hazard in fires is mainly come from smoke. Past fire records revealed that most life loss caused by smoke rather than direct heat burning. Therefore, smoke zone and smoke exhaust system is provided to control smoke at the storey 10m in height.

10.1.1 Smoke Zone

Smoke Zone has been applied in the storey. Details of the Smoke Zone are:

• Smoke zones should be separated by fixed smoke baffles or screens from adjacent smoke zones
• Baffles are required to extend from 2000mm below the ceiling height.
• Smoke baffles are required to be provided around the new voids to prevent smoke from spilling up the voids and affecting the exiting smoke hazard management system. Perforations are required to be provided near the baffles to allow smoke into the ceiling plenum to be exhausted.
• Where 14 smoke zones are required. Each zone should be capable of operating simultaneously. The smoke zones is shown as Appendix C
• Ceiling material should be of non combustible with properties that provide a piloted ignition temperature greater than 300℃.
• The fire sprinkler system should be zoned similar to the smoke zones, and vice verse, throughout the storey.

10.1.2 Smoke Exhaust System

A smoke exhaust system has been installed. Details of the smoke exhaust system are:

• Each smoke zone should install independent smoke exhaust apparatus.

• Smoke exhaust from the smoke zones shall be provided to achieve a minimum exhaustion rate. The concealed ceiling space should be utilized as the smoke reservoir which extends over the storey.
• Make-up should be maintained and available from the adjoining smoke zones which should switch to fire mode operation. Adjoining smoke zones may revert to smoke extraction upon activation of any of the above listed items in that smoke reservoir.
• Minimum down stand of the smoke baffle separating smoke zones should extend from the underside of the floor slab above in order to prevent smoke spillage into the adjoining smoke zone (depth to be determined via fire engineering analysis).
• General mechanical ventilation and air conditioning systems should be shut down within the smoke zone upon initiation of a general fire alarm. Hence mechanical and ventilation systems should operate in normal mode in zones of non fire origin, unless required to run in fire mode for smoke exhaust make up air.
• Extraction points should be evenly distributed each smoke zone, and consideration should be given to the effective opening size to prevent plug holing during operation of the smoke exhaust system.
• Where equipment located in the concealed space poses a restriction in allowing free flow of smoke migration into the ceiling space, special consideration should be given to positioning the extraction points.
• The activation of zone smoke detectors, located below ceiling, should start the smoke exhaust system within the affected zone.
• Sprinkler activation should start the smoke exhaust system in the smoke zone in which the fire sprinkler is located.
• The activation of sprinklers should start the operation of the smoke exhaust system.
• Non-affected zones should operate on 100% outside air to provide make up air for the effected zone.

10.1.3 Smoke Control Measures Evaluation

The smoke densities and depths of smoke layers can be created by software “CFAST”. Besides, the software “EXITT” also can be calculating the psychological impact of smoke, S, the following equation is used:

S=2 × OD × D / H

Where OD is the optical density of the smoke in the upper layer

D is the depth of the upper layer, and

H is the height of the ceiling.

By Fire Code Reform, “Fire Engineering Guidelines”, First Edition-March 1996, an acceptable criterion is that the smoke layer does not fall below 2.1m in height from floor. This means that occupants will not have to move though products of combustion (smoke) in making their escape. By those software, the design measures can be evaluated against performance criteria for failure or success.

10.2 Fire and Thermal Control

The impact of the thermal effects on occupants is driven by the effects of convective and radiative heating. The convective criteria is related to the temperature of the air that the occupants will be breathing, and the radiative criteria is based on exposure to either a hot gas layer or energy source. The criteria used in the analysis are based on the assumption that occupants cannot breathe “wet” air more than 100℃, and “dry” air at more than 220℃ from the SFPE handbook of fire protection Engineering, 2nd ed. “Toxicity Assessment of Combustion Productions”. These criteria were developed for skin burns, but may be used to determine general tenability, as damage to the respiratory tract has not been observed without coincident damage to the skin.

The criteria for tenability related to radiative effects assumes that the occupant will not be exposed to a heat flux greater than 2.5kW/m2. This value is derived from experiment conducted by Perkins etal using searchlights, and is suggested by Babrauskas for use in modeling applications.

For controlling fire and thermal effect, drencher system and automatic fire sprinkler system should be installed.

10.2.1 Drencher System

The drencher units are installed around each smoke zone dividing the storey into different fire prevention zones. The drencher system comprises a 300 cubic metre water tank, water pumps, control valves, piping, nozzles and inlets. The capacity of the water tank can maintain the system's operation for at least 30 minutes. To enhance the system's reliability, an emergency generator is provided as a secondary power source. In case of fire, the system will be actuated by the heat detectors serving the protected area. The pumps will then automatically start to supply water through the drencher control valve into the system piping. The affected zone will be isolated by a curtain of water supplied by the open-ended drencher nozzles to prevent the fire spreading to other unaffected areas.

The fire signal will be transmitted simultaneously to the fire communication centre at the fire station via a direct telephone link for immediate response.

10.2.2 Automatic Fire Sprinkler System

An automatic fire sprinkler system should be provided in each smoke zone.

The system should be designed and installed including the following:

• Sprinkler is activated by the heat detector.
• Sprinkler heads should be of fast response type.
• The sprinklers should be zoned in a similar way to the smoke zones
• When Sprinkler is activated, the smoke exhaust system may operate immediately.
• In case of fire, the Sprinkler is activated by the heat detector and controlling the temperature of the affected zone.

Automatic fast response sprinklers provide faster sprinkler operation and should be provided to enhance life safety. Statistics indicate that sprinklers are effective and have a reliability of operation 95% as stated in the Fire Safety Engineering Guidelines. Research also indicates that on average, fewer quick response sprinklers activate to control the fire than standard response sprinklers, thus reducing the demand on the water supply.

10.2.3 Fire and Thermal Effects Evaluation

To assist the design team in its efforts to quantify the effects of the design fires via fluid flow modeling, a Computational Fluid Dynamics (CFD) program developed by the National Institute of Standards and Technology is utilized. The Fire Dynamics Simulator (FDS) model is used to characterize low speed hot gas flows represented by the Navier-Stokes equations. Additional subroutines are used to ascertain the production and behavior of smoke as well as certain aspects of sprinkler actuation during typical fire events. The OpenGL graphics program, smokeview, is used to visualize the numeric output of the CFD model. Primarily steady state conditions are modeled using FSD. Appropriate documentation of the modeling software and its mathematical components can be obtained from NIST.

10.3 Toxicity

The tenability limits for which human can sustain with exposure to smoke is difficult to predict and varies on different people group. But the criterion for stated exposure can be listed as Table 3:

Table 3: Tenability criteria given in NFPA 130.

Untenable conditions are considered to have occurred when the hot smoke layer descends below 2.1 m in height from the floor. The International Fire Engineering Guidelines suggest that limiting conditions for all toxic products (asphyxiates and irritants) are unlikely to be expected for up to 30 minutes if the smoke optical density (OD) does not exceed 0.1 m-1.

Besides, toxicity criteria are based on the Fractional Equivalent Dose, where unity represents either impairment or death, depending on the relationship being used. Specifically, the analysis considers the presence of outlying toxic gases such as CO and HCN as applicable. Fire Dynamics Simulator is used to determine species concentrations in those areas immediately adjacent to the fire source in order to ascertain whether toxicity adversely impacts the ability of occupants to safely and reasonably evacuate. Percent of CO in the bloodstream using the following equation:

%COHb = (3.317×10-5)(ppm CO)1.036RMV(t)

Loss of consciousness occurs at approximately 34% COHb. Depending on the concentration of HCN in the air, two equation are used as follows:

Concentration 80-180ppm

(tlcn)(min) = (185-ppm HCN) / 4.4

Concentration >180ppm

(tlcn)(min) = exp(5.396－0.023×ppm HCN)

These equation represent the time to incapacitation based on HCN concentration in parts per million.

10.4 Reducing Fire Detection and Alarm Time

Detect and Alarm activation is considered to occur via the following means:

• activation of smoke detectors; or
• visual detection of the hot smoke layer by occupants; or
• activation of a sprinkler by heat detectors.

10.4.1 Smoke Detector Activation

Some design fire scenarios are based on the installation of smoke detectors throughout the complex to provide early fire detection and initiate early occupant warning/evacuation procedures. Activation time of smoke detectors is to be estimated using FPETool or FDS, where the smoke detector activation is considered to occur at a temperature change of 13℃ above ambient or at 0.222 m-1 (smoke detector sensitivity of 5% obscuration per metre (Ob/m). An alarm verification delay in the fire detection and alarm system of 20 seconds is included, which is the time delay between the detection of smoke and the sounding of an alarm signal.

10.4.2 Visual Detection

Fire detection via occupants sighting smoke is to be based on information document within the draft National Building Fire Safety Systems Code20, which suggests that visual fire detection by occupants could be said to occur when the smoke layer descends to 95% of the ceiling level. This is to be determined using results from the CFAST or FDS fire and smoke analysis. But there is no compartment in the floor and is a large rectangle shape space (10m height), so the fire or smoke can be sighted directly very fast. During assembly period, many staff is in the hall so the fire or smoke can be discovered very fast. If not an assembly period or just setup period, there are a few persons so evacuation time is required small. But the time delay is difficult to evaluation.

10.4.3 Sprinkler Activation

The activation time of both sprinklers and heat detectors are to be calculated using the “Sprinkler/Detector” program, which is one of the programs within the FPETOOL or FDS suite of programs.

10.5 Reducing Pre-movement Time

Pre-movement time typically applies only to areas remote from the room of fire origin where they may receive only a single cue to the presence of a fire and where those cues do not present an immediate threat to their health and safety. An example is where an occupant remote from the fire origin may smell smoke however would be unsure of its origin and may take investigative action or rationalize that it is ‘normal’, e.g. someone burning off outside.

In assessing the likely response of the occupants, the issue of pre-movement time must be addressed. In the case of occupants who are awake and in the vicinity of the fire, the decision to evacuate is likely to be a function of the perceived threat associated with the fire. If the fire is not perceived as threatening, then the occupants may decide not to evacuate. However, if the opposite is true, evacuation will begin almost at once. It is assumed that most of the public will associate flaming fires and black smoke with a threatening situation. Thus, in undertaking calculations of evacuation, this can be assumed to commence once a threat is perceived.

In the situation where the occupants are intimate with the development of the fire (area of fire origin), it is reasonable to suggest that occupant avoidance will be immediate, as they will be presented with multiple fire cues and would include:

• Visual - smoke and flames
• Tactile - heat radiated and converted from fire
• Audible - sound generated by burning materials or live voice of assembly staff
• Olfactory - smell of smoke and other combustion products

In the storey, there is no compartment in the floor and is a large rectangle shape space and the assembly is conducting during the fire. Therefore, the trained assembly staff may be given live voice alarm and instructed the occupant to escape. By BS DD240:1997, the pre-movement time is shown as below (Figure 4):

Figure 4

The pre-movement time is less than 2 minimums in an assembly time. No assembly time may not consider since less person in the storey. But the prescriptive requirement is assumed the worst case that is just installed simple alarm in office building. Therefore, the pre-movement time is more than 4 minimums. For the worst compare, 2 minimums is applied for the pre-movement time for assembly hall and 4 minimums is for the prescriptive requirement, so 2 minimums of pre-movement time can be saved.

10.6 Reducing Travel Time

The performance-based approach recommends the use of a hydraulic or effective width method to calculate travel times in a building. These times then are adjusted using efficiency factors to reflect the characteristics of the occupancy. With this approach standard travel speeds and occupant densities are used in calculations.

An alternative method to that is to reflect the occupancy characteristics in the movement modeling by using travel speeds and occupant densities, which are reflective of the particular occupancy being modeled.

The following is provided several modeling or method to be applied.

10.6.1 EVACNET4

The evacuation modeling of this development can be to undertaken by EVACNET4 due to the large number of building occupants and the complexity and number of alternative evacuation routes. For the modeling, the relationship between time necessary for occupants to egress, and the time to hazard is best described by a simple mathematical relationship. It is as follows:

th >= (td+ta+to+ti+2×tt) = tev

where th = time from ignition to conditions hazardous to life.

tev = evacuation time.

td = fire detector response time

ta = time from detection to occupant notification.

to = time from notification to occupant response.

ti = time for occupants to investigate fire, collect belongings, and fight fire.

tt = occupant travel time to place of safety.

10.6.2 Analysis by Togawa and the London Transport Board / S.J. Melinek and S. Booth

Based on the data from the investigations by them, the flow movement in buildings can be analyzed and provided the following formulae:

The maximum population M which can be evacuated to a staircase, assuming a permitted evacuation time of 2.5 min, is given by

M = 200 b + (18b+14b2) (n-1)

Where b is the staircase width in m; and

n is the number of storeys served by the staircase (In our situation, we take n=2 since we just study occupants to evacuate to below one storey. 2 storey is considered.)

If the population Q and the staircase width b are the same for each floor then the minimum evacuation time is the larger of T1 and Tn where

T1 = n Q / (N’b) + ts

Tn = Q / (N’b) + n ts

T1 ccoresponds to congestion on all floors and Tn to no congestion. Melinek and Booth suggested as typical values of N’ and ts 1.1 persons/sec/min and 16 sec.

10.6.3 Flow Models based on empirical studies of crowd movement developed by Jake Pauls

This model is based upon his extensive empirical studies of crowd movement on the stairs as well as the data about the mean egress flow as a function of stair width. In this context, he conducted several evacuation drills in high-rise office buildings and observed normal crowd movement in large public-assembly building. The speed–density relation data from a study by Pauls in uncontrolled total evacuations of tall office building are shown as below (Figure 5).

Figure 5: Relation between speed and density on stairs in uncontrolled total evacuation.

The model describes the following phenomena:

1. The usable portion of a stair width, i.e. the effective width of a stair begins approximately 150mm distance from a boundary wall or 88mm distance from the centerline of a graspable. Figure 6 is showing the measurement of the effective stair width in relation to walls and handrails.

Figure 6: Measurement of effective stair width in relation to wall and handrails.

2. The relation between mean evacuation flow and stair width is a linear function and not a step function as assumed in traditional models based on lanes of movement and units of exit width. The evacuation flow is directly proportional to the effective with of a stair. It is shown as below (Figure 7)

Figure 7: Predicted and observed total evacuation time for tall office building.

3. Mean evacuation flow is influenced in a nonlinear fashion by the total population per effective width of a stair.

Pauls provides the following equation for the evacuation flow (f) in persons per metre of effective stair width:

f = 0.206 p0.27

p is the evacuation population per metre of effective stair width. The total evacuation time (t) is given by

t = 0.68 + 0.081 p0.73

(Note the upper limit of 800 persons per meter of effective stair width)

This calculation method has been accepted for an appendix to the NFPA Life Safety Code, 1985 edition.

This method is most suitable for our situation. It is because the study have considered the crowd movement in large public-assembly building and conducted in high-rise office buildings. Besides, we assume maximum 2,100 persons at the storey including all staff holding the assembly and a 10m width staircase is provided.

The number of person per meter of effective stair width = 2100 / (10-0.3) = 217 < 800

Therefore, the above equations can be used.

11. Future Action

According to this report, we should select design performance criteria met or not. If not, we should modify the design objectives. If it is met, we should select the final design and conduct a Performance Based Design Report. Finally, the design documentation, specifications, drawing and O&M Manual should be provided.

12. Reference List

1. Wayne D. Moore, “National Fire Alarm Code Handbook” National Fire Protection Association, 1993.
2. D Canter, J Powell and K Booker, “Psychological aspects of informative fire warning systems”, Department of the Environment Building Research Establishment Fire Research Station.
3. Jake Pauls, “Movement of People,” The SFPE Handbook of Fire Protection Engineering, 2nd edition, Society of Fire Protection Engineers, Boston, Mass., 1995, Section 3, Chapter 13.
4. Arthur E. Cote and Jim L. Linville, “Fire Protection Handbook”, Seventeenth Edition, National Fire Protection Association, Quincy, Massachusetts.
5. Paul R. DeCicco, “Evacuation From Fires Volume II”, Baywood Publishing Company, Inc.
6. Ichiro Hagiwara and Takeyoshi Tanaka “Fire Safety Science – Proceedings of the 4th International Symposium - International Comparison of Fire Safety Provisions for Means of Escape”, Building Research Institute.
7. H.A. Donegan, A. J. Pollock and I. R. Taylor, “Fire Safety Science – Proceedings of the 4th International Symposium – Egress Complexity of a Building”, Department of Mathematics, University of Ulster.
8. Rita F. Fahy, “Fire Safety Science – Proceedings of the 4th International Symposium – Exit 89 – An Evacuation Model for High-Rise Buildings – Model Description and Example Applications”, Fire Analysis and Research Division, National Fire Protection Association.
9. Leong S. Poon, “Fire Safety Science – Proceedings of the 4th International Symposium – EvacSim: A Simulation Model of Occupants with Behavioural Attributes in Emergency Evacuation of High-Rise Building Fire”, The Centre for Environmental Safety and Risk Engineering (CESARE), Victoria University of Technology (VUT).
10. Hamish A. MacLennan “Fire Safety Science – Proceedings of the 1st International Symposium – Towards an Integrated Egress/Evacuation Model Using an Open System Approach”, School of Building Studies, The N.S.W. Institute of Technology.
11. Jonathan D. Sime “Fire Safety Science – Proceedings of the 1st International Symposium – Perceived Time Available: The Margin of Safety in Fires”, School of Architecture, Portsmouth Polytecthnic.
12. Ezel Kendik, “Fire Safety Science – Proceedings of the 1st International Symposium – Methods of Design for Means of Egress: Towards a Quantitative Comparison of National Code Requirements”, COBAU Ltd.
13. Daniel F. Gemeny, “3rd International Conference on Fire Research and Engineering – Performance-Based Design Analysis of A Shopping Mall”, Rolf Jensen & Associates, Inc.
14. Kazunori Harada, Naohiro Takeichi and Ai Sekizawa,“3rd International Conference on Fire Research and Engineering – Performance Evaluation Methods for Evacuation Safety and for Structural Fire Resistance”
15. R. F. Fahy and J. I. Sapochetti“3rd International Conference on Performance-Based Codes and Fire Safety Design Methods – Balancing Fire Prediction and Egress Prediction”.