Note: this article is intended to be
a commentary on the major fire that destroyed 1,400 cars in a multi-storey car
park in Liverpool on December 31st. The author has referred to the scarce
informations currently available and the article has no pretension to be
exhaustive, but has the objective of analyzing the safety of an activity
similar to that of the garage of Liverpool, based on the latest Italian
firefighting standards and with the help of some fire simulations with CFD
software.
This is the English translation of an
article initially written for the Italian public.
Brief chronicle of the event
The story is already known and has aroused considerable wonder, especially
in Great Britain.
The fire that originated from a car parked on the third floor of the
structure, spread easily to the entire building that housed a multi-storey car
park, involving most of the cars present at that time.
The material damage is enormous because not only about 1,400 cars have been
destroyed, but the entire building is now unusable and will certainly be
demolished. In fact, some floors have collapsed and others have been seriously
damaged.
The environmental impact of the smnokes and the substances produced by the
fire was certainly heavy.
First comments
It is well known that cars in a parking lot are a possible source of fire
due to the electrical components and, to a lesser extent, the fuel contained in
some of them.
Likewise, it is well known that modern automobiles contain a greater
quantity of combustible materials than in the past, in terms of plastics,
insulation, expanded materials, electrical cables, tires ...
The open-type garages (with reference to the “open” definition valid in
Italy) have already been the subject of numerous studies, but above all limited
to the issue of fire resistance of the structures.
What in the case of Liverpool is perplexing is the progression of the fire,
which was not possible to contain and extinguish, not even thanks to the timely
intervention of the English firefighters.
The first inspections are under way by the insurance companies, using
drones as access to the building is currently also closed to the technicians.
There are still not many details available and certainly in the coming
weeks will be deepened the investigations and technical assessments, but from
the first photographs and the first comments issued by different professionals
or local Authorities, fire engineers can deduce some first considerations.
First assessments of the fire engineer
The following is an attempt to provide a plausible interpretation based on
early evidence and fire engineer experience in similar cases.
It is immediately evident that the Liverpool fire has been extended to the
entire building because the appropriate protective and containment measures
have been lacking.
From the photographs of the state of the building following the fire we can
deduce that:
• Each floor was not divided into fire compartments that could have reduced
the extent of the damage;
• All the floors of the parking lot were communicating and in any case they
were made such by the partial collapse of the floors;
• In the façade there were no blind parapets of adequate height, which
could have limited the spread of the fire on the upper floors through the
facades;
• The floors did not guarantee either the fire resistance (partially
collapsed) nor the flames and smokes protection, because they were partially
opened allowing the fire to rise up and down easily and dramatically (the
firefighters of the local fire hypothesize that temperatures of 800-1,000 ° C
have been reached);
• There was no automatic fire control system, for example of water type
with sprinkler heads, which would have significantly reduced the power of the
fire, containing it to the first or the first cars, and would have guaranteed
to the firefighters the possibility of intervene to proceed with complete
extinction.
Despite all this, there are no victims or injured, but consider that:
• the occupants had already parked the cars and had left to go to the
nearby Arena to attend an event;
• the parking lot was of open-type (large areas of natural ventilation on
the perimeter) and with open stalls.
In UK the event is provoking considerable clamor and controversy, also
because it follows the sadly famous fire of the Grenfell Tower, which caused
the death of 71 people (to this link the article dedicated to the tragedy
(Italian version): link )
Figure 3. Part of a collapsed
floor on the level below, which increased the spread of the fire between the
floors (MERSEYSIDE FIRE AND RESCUE SERVICE)
State of the art of fire regulations in Italy
Following the author intended to analyze the safety conditions and
regulatory compliance of a garage similar to that of Liverpool according to the
most recent Italian fire safety standards.
In Italy, the fire protection regulations of garages (but also in general
for other activities subject to them) are in a phase of profound renewal
because they are characterized by the progressive overcoming of the
prescriptive approach in favor of a performance-based approach.
In the issue n. 52 of 3 March 2017 of the “Gazzetta Ufficiale” was
published the Decree of 21 February 2017 on the "Approval of fire
prevention technical standards for garage activities".
Attached to the aforementioned decree is the (long awaited) new Vertical Technical
Rule (RTV) of fire prevention for garages.
The new “RTV” is an alternative to the current provisions (D.M. 1 February
1986) and is developed to modify and integrate the Italian Fire Prevention
Code.
The new vertical technical regulation applies to the car parking activities
with a total covered area of more than 300 m2, identified by the
number 75 in Annex I of the D.P.R. n. 151/2011, existing on the date of entry
into force of the D.M. February 21, 2017, or for those of new construction. It
can be applied to these activities as an alternative to the specific fire
prevention rules laid down in the decree of the Minister of the Interior of 1
February 1986 and the decree of the Minister of the Interior of 22 November
2002.
(at this link the article dedicated
to the new Fire Safety Technical Rule on car parkings (Italian version): link )
According to Italian fire regulations,
the Liverpool garage (or similar) would be defined as:
• public (SB type)
• open (it is assumed that the
natural ventilation surface is more than 15% of the floor area)
• (isolated or mixed, the
necessary information is missing)
• open spaces (not divided into
boxes)
• above ground
• multi-storey.
In relation to the surface of
the garage, which is estimated as about 25 m2 per car space, or
25x1.600 = about 40,000 m2, the RTV would classify it as AD (Area >
10,000 m2), as it is clear from the photos that several floors were
in fact a single fire compartment.
In relation to the height of the
activity, it is assumed that it is between 12 and 24 m, so the garage would
fall into the HB case.
The fire resistance according to
the RTV should be at least R30 (garage above ground).
Regarding the subdivision into
fire compartments, the RTO (Italian Fire Prevention Code) asks that each level
be separate REI / EI from to the others, unless it falls under the conditions
to apply the multi-compartment partitioning (not all conditions can be found in
Liverpool case).
The maximum area of the compartments
is established in Table S.3-4 of the Italian Code, and is equal to 8,000 m2.
With regard to fire control
measures, 2017 RTV would require a level
IV measure, that is an automatic type of protection (sprinkler system), which
was not present in the Liverpool garage.
Regarding the measures for the smoke and heat control, for the above ground
garage of Liverpool, the 2017 RTV would ask to apply the level of performance
II, that is the simple disposal of smoke and heat, then only a series of
openings of natural ventilation (appropriately distributed) with the function
of removing the products of combustion during the phase of the extinction of
the fire by rescuers. Instead it would not require a real smoke and heat
evacuation system that must ensure the maintenance of a smoke-free lower layer.
Summarizing, the recent Italian
legislation would have asked, for the Liverpool garage:
• R30 fire resistance (no data
are available to confirm that it was)
• REI30 separation of each level
(it can be deduced that it was not guaranteed)
• sprinkler system (its absence
is assured)
• natural ventilation for smoke and heat
disposal (present).
From this first analysis it is clear that the main problem of the activity
in question was the lack of any measure able to contain the fire, both horizontally
(sprinkler system and subdivision of each floor into compartments) and
vertically (protected vertical connections, REI floors, parapets of adequate
height and resistance to outdoor fire ...). The simple awareness of these
deficits could and should have been an alarm bell on the certainty of not being
able to contain nor, consequently, face and extinguish, a fire in this
structure. An on-site inspection would have been sufficient to carry out a risk
assessment on-line and therefore be able to propose the appropriate
compensatory measures.
A CFD modelling of Liverpool car parking fire
As anticipated in the
introduction, the technical and regulatory tools are now available to be able
to face a fire safety project with the performance-based method (the Italian
normative references for the case in question are the D.M. 3 August 2015, the
D.M. 21 February 2017).
In this brief study, the
performance-based approach has been applied, with informative and certainly not
forensic engineering intentions, in order to illustrate the modalities foreseen
by the Italian law for the examination of the fire safety performance levels of
the activity constituted by a garage like the one in Liverpool.
Therefore, some fire simulations
were performed with Pyrosim (FDS), a computational fluid dynamics modelling software.
Further technical references are
those listed in the bibliography, in particular:
• BRE (Building Research
Establishment), "Fires In Enclosed Car Parks" Research Project.
• Study Report "Car Park -
Fires Involving Modern Cars and Stacking Systems", P.C.R. Collier, BRANZ,
2011.
Fire Safety Engineering applied in accordance with Decree February 21st
2017 (Italian fire safety standard about car parking)
In order to model the fire scenarios
illustrated below, reference was made to the criteria that the 2017 RTV indicates
with regard to fire resistance design with a performance-based approach.
These criteria are applicable to the
case study as it falls under the conditions provided for in paragraph V.6.7.1:
• open garage, whose ventilation
openings constitute at least 50% of the surface of the facades;
• garage above ground and open space
type (no boxes).
Therefore, it was assumed, as an initiator event, the fire of a car that
subsequently involved the two cars parked at the sides, with HRR curves as per
tables V.6-5 and V.6-6 of the RTV.
Regarding the HRR curve used in the modeling, reference was made to the
technical literature of the sector, such as the research project "Fires In
Enclosed Car Parks" of BRE (Building Research Establishment) which were
then taken up by the Italian car parking fire safety standards .
Figure 5. Experimental and project HRR curves (BRE)
Figure 6. Illustration V.6-2 of the Italian car parking fire
safety standards
Not being a goal of this study the
verification of resistance to fire, however, were not considered the fire
scenarios provided for in paragraph V.6.7 of the RTV; instead it was assumed
that the fire involved three cars, for the sole purpose of illustrating the
different progression of the fire between the two scenarios:
• basic scenario, without sprinkler
system
• improvement design scenario, with
sprinkler system.
In fact, the complete reconstruction of the fire that took place in
Liverpool is beyond the scope of this publication.
Figure 7.
Location of the 3 cars affected by the fire and the measurement probes
Figure 8.
Sprinkler system settings in Pyrosim software
The sprinkler system settings were made on the basis of a classic standard system, with the following characteristics (reference UNI EN 12845):
• risk level: OH2
• k factor: 80 l / (min * atm1 / 2)
• working pressure: 0.56
bar
• design discharge density: 5 mm / min
• flow rate: about 60 l / min
• operating area: 144
square meters
• area of influence of each head: 12
sqm
• activation temperature: 79 ° C.
A
first simulation with FDS was carried out in order to identify the instant at
which the sprinkler system is activated with the settings listed above.
The
simulation showed that after 210 seconds from the start of the first car fire,
two sprinkler heads are activated simultaneously, corresponding to those positioned
closer to the fire.
Figure 9.
After 210 seconds, the first two sprinkler heads are activated.
At this point we refer directly to the Italian Fire Safety Code, which
in paragraph M.2.6.2 indicates the following.
Figure 10.
Illustration M.2-1 of the Italian Fire Safety Code
Therefore, two further CFD simulations have
been carried out with a duration of 1,800 seconds (30 minutes, corresponding to
the class of fire resistance):
1. basic scenario (similar conditions to the
fire in Liverpool): without sprinkler system, with HRR curve as in the 2017 RTV,
without any reduction of HRR;
2. improvement design scenario
(corresponding to the Liverpool parking project hypothetically developed in
accordance with the 2017 RTV): with sprinkler system, with HRR curve that
includes an initial phase as from 2017 RTV, then reduced to a constant value
equal to that achieved by the base curve at the instant of activation of the
sprinkler (considering also a margin of 30 seconds further), then at 210 + 30 =
240 seconds. At that moment the thermal power is: 1,400 kW.
The following figure shows both the HRR
curves set in FDS.
Figure 11.
Graph of the HRR curves of the two scenarios in FDS
Figure 12.
Basic scenario, smoke propagation
Figure 13. Basic
scenario, distribution of visibility along a vertical section
Figure 14.
Basic scenario, distribution of visibility along a vertical section
Figure 15. Basic
scenario, the isosurfaces of visibility provide an image of the smoke
propagation after 30 seconds
Figure 16. Basic
scenario, the isosurfaces of visibility provide an image of the smoke
propagation after 158 seconds
Complete modeling of fire scenarios with and without
sprinklers
After determining the instant of activation
of the sprinkler system, two complete simulations were performed, in the two
fire scenarios, with and without sprinklers.
The following figures show the comparison
graphs between the two fire scenarios, from which it can be deduced that:
• the Italian (but also European) standard
allows to set a HRR curve with very low power compared to the basic HRR curve,
from which derive much lower temperature values (and which would allow
considerable savings in terms of structural fire protection);
• the automatic fire extinguishing system
ensures the achievement of human safety performance levels for the entire
development of the event, at least in terms of temperature (it is assumed that
after 30 or 60 minutes of sprinkler duration, rescuers have intervened with the
hydrants);
• in this specific case the performance level
for visibility is not guaranteed, in neither of the two fire scenarios.
In the simulations carried out, which had
another purpose rather than following the same conditions and results found in
the real fire occurred in Liverpool, the flash-over conditions are not reached
(general fire in which all the combustible material present is triggered
simultaneously), at least in first 30 minutes. It should be noted that the
considerable natural aeration surface can exclude that a real flash-over
occurred, but rather a generalized fire (though with rapid development) that
recorded a progression without any stop, therefore the final consequences are
comparable to those of a fire with flash-over.
Definition of flash-over from “NFPA101:
Life Safety Code”: «A transition phase in the development of a contained fire
in which surfaces exposed to thermal radiation reach ignition temperature more
or less simultaneously and fire spreads rapidly throughout the space.»
Figure 17.
Comparison graph of the temperature values recorded above the fire
Figure 18.
Comparison graph of the temperature values recorded in the car lane
Figure 19.
Comparison graph of the visibility values recorded in the car lane
At the following link the complete eBook " Autorimesse.
Il Progetto antincendio. Norme tradizionali. Codice. Fire
Safety Engineering.”: link
At the following link the complete eBook “Antincendio.
Casi pratici di progettazione. Codice di prevenzione incendi - Regole tecniche
verticali - Fire Safety Engineering”: link
At the following link the article concerning the new
Italian Fire Safety Standards on Car Parkings: link
At the following link the article concernig the Grenfell tragedy: link
Bibliography
[1] Decreto 3 agosto 2015,
Approvazione di norme tecniche di prevenzione incendi, ai sensi dell'articolo
15 del decreto legislativo 8 marzo 2006, n. 139. (GU Serie Generale n.192 del 20-8-2015 -
Suppl. Ordinario n. 51).
[2] Decreto 21 febbraio 2017,
Approvazione di norme tecniche di prevenzione incendi per le attività di
autorimessa. GU Serie Generale n.52 del 3-3-2017.
[3] “Codice di prevenzione
incendi commentato.” F. Dattilo, C. Pulito. EPC Editore.
[4] BS 9999:2017. Fire safety in the design,
management and use of buildings – Code of Practice.
[5] Commissione per la sicurezza
delle costruzioni in acciaio in caso di incendio, Applicazione dell’ingegneria
della sicurezza antincendio alla progettazione strutturale di autorimesse
aperte, fuori terra ed a spazio aperto ai sensi del D.M. 21/02/2017
[6] BRE (Building Research Establishment),
Progetto di ricerca “Fires In Enclosed Car Parks”.
[7] Study Report “Car Park – Fires Involving
Modern Cars and Stacking Systems”, P.C.R. Collier, BRANZ, 2011.
[8]
eBook “Autorimesse: il progetto antincendio. Norme tradizionali, Codice, Fire Safety
Engineering.” Ing. Filippo Cosi. Wolters Kluwer.
[9] eBook “Antincendio. Casi
pratici di progettazione. Codice di prevenzione incendi - Regole tecniche
verticali - Fire Safety Engineering.” Ing. Filippo Cosi. Wolters Kluwer.
