Report!on PPV Trials
at Oxford Road, Preston, U.K.
10 –14 January, 2000
Richard
Stott B.Eng (Hons),
Assistant
Divisional Officer
Lancashire Fire and Rescue
Service, U.K.
1.
INTRODUCTION
1.1
Ventilation
Ventilation as a
fire-fighting tactic has been used since time immemorial.
Hence it is nothing new. Venting
(the creation of exhaust/inlet ports for the products of combustion) a fire
has been done with caution, following much training and consequently has been
instrumental in reducing property damage and increasing comfort levels for
fire-fighters at virtually every fire. The
maxim within the Fire Service has been ‘vent, vent early, but only when safe
to do so’. Hence it has always
been recognised that caution is the better part of valour but that used
correctly, venting is a key tool in the fire-fighter’s toolbox.
Positive pressure ventilation is a method of using high volumetric flow rate
fans to increase the pressure within a compartment so that the products of
combustion are encouraged to exit the compartment in order to reach a region
of lower pressure. Generally speaking, the mobile fan unit is placed
approximately 3 metres from the doorway, i.e. outside the building, and the
air is directed in through the doorway. Positive pressure ventilation can be
viewed as simply an advanced use of natural ventilation techniques.
Positive pressure ventilation is widely utilised in the USA and the concept
was first introduced there many years ago. The constructional methods used for
buildings in USA differ widely to those of the UK, so it is difficult to
justify the direct application of the ventilation techniques developed there.
However, many local authority Fire Services in Great Britain now use positive
pressure ventilation in the post-fire scenario. Most however, are wary
regarding the use of positive pressure ventilation prior to the fire being
extinguished. This has been
termed ‘aggressive positive pressure ventilation’.
In fact, at the time of writing, only one of the fifty-seven local
authority brigades has routinely adopted this approach.
In
contrast to the above paragraph, it has been decided to utilise the terms
offensive and defensive in this report, to parallel the Home Office and LFRS,
with the following definitions(1);
Defensive – ventilating away from the fire, or after the fire is out.
Offensive -
ventilating close to the fire to have a direct effect on the fire itself.
From hereon, positive pressure ventilation will be referred to as ‘ppv’.
With regard to domestic ppv use, as it is very unlikely in a single domestic
dwelling that a fan will be used away from the fire, the term defensive should
be considered to mean use of ppv once the fire has been extinguished.
1.2
The
Trials
During the week
of 10 – 14 January 2000, Lancashire Fire and Rescue Service, supported by
the Department of Built Environment, University of Central Lancashire,
undertook a series of ppv trials at Oxford Road, Preston.
The main objectives of the trials were as follows: -
·
To
examine the possible tactical advantages of offensive ppv
·
To
improve knowledge and skills in the use of defensive ppv
·
To
further assess fan selection in terms of fan size
·
To
produce a comprehensive report, (including stills and video footage), to
assist in future training packages and the further development of operational
procedures.
·
To
provide realistic, live fire observation for personnel involved.
·
To
provide comprehensive data for use in risk assessments.
The trials were originally planned to take place in two-storey properties but
were relocated, due to uncontrollable circumstances to single-storey
properties.
Plate
1. Oxford Road, Avenham
Each
fire experiment room was furnished with the type of materials that would be
expected with a real incident i.e. carpets, curtains, sofas, bedding etc.
Consequently, the trials were possibly the most realistic, in terms of fire
loading, to have been carried out to date in the U.K.
The
trials took place over five days. The chosen approach was to carry out each
scenario twice, once using standard fire fighting techniques, (defensive) and
once using offensive positive pressure ventilation.
Although much
quantitative data was gathered, the nature of fire, with its many variables,
dictates that qualitative data is very important. Consequently, each volunteer
has been canvassed for his or her experiences during the trials. This
information is included within the report.
This report!is a précis of a more detailed and technical engineering
analysis, which is available if so required.
1.3
Positive Pressure Ventilation
Extensive research has been undertaken regarding the subject of positive
pressure ventilation and its applications for the Fire Service. Lancashire
Fire and Rescue Service and several other Brigades have conducted ‘in
house’ experiments to determine the effectiveness of positive pressure
ventilation post fire, as a result, the use and safe application at incidents
is well documented.
Supporters
of the use of ppv offensively claim that it has a number of advantages, such
as:
·
The enhancement of fire-fighter safety
due to the improved atmosphere that positive pressure ventilation can create.
·
The rapid improvement of conditions
remote from the fire-affected compartment will enhance the chances of survival
for persons trapped in a building on fire.
·
Fire-fighting operations can be
expedited as the improved conditions enable the fire attack to take place
quicker.
·
The
correct use of positive pressure ventilation prevents firespread beyond the
fire compartment.
The
literature review undertaken for this project demonstrates that although
research into positive pressure ventilation has been fragmented, the overall
picture is one of support!for the technique. There is a distinct lack of
formal academic evidence to support!this however and consequently many
sceptical Chief Fire Officers, this being demonstrated by the reticence of UK
Fire Brigades in offensive positive pressure ventilation use. Most of the
reports reviewed indicate the need for further research.
Two Fire Brigades have already undertaken offensive ppv experiments, Tyne and
Wear Metropolitan Fire Brigade and Greater Manchester County Fire Brigade.
The results of the two sets of experiments undertaken by Tyne &
Wear indicate that the use of ppv does not seriously intensify fire spread (2).
In contrast to this the report!of offensive ppv experiments in Greater
Manchester indicate a rapid acceleration of the fires involved on operation of
the fans (3).
Clearly from the differing results produced by two large Metropolitan Brigades
the use of offensive ppv warrants further investigation and research. It is
universally accepted that without proper training and correct tactical
deployment at incidents, ppv could prove potentially fatal for fire fighters
and members of the public.
When assessing the use of positive pressure ventilation there is a tendency to
forget that pressurisation has been incorporated within building designs for
several decades. Heating ventilation and air conditioning systems can be
designed so that, in the event of a fire, they can be used as a smoke control
system. Separate heat and smoke control systems are also designed for the
larger building and for certain industrial applications, therefore the concept
of ppv and fire application is not a new one.
1.4 Previous U.K.
Research -- Overview
A
major concern amongst fire officers and the research establishment in this
country is the fact that ppv may prove ineffective when used offensively,
possibly even dangerous. Ventilation techniques for domestic dwellings and
small industrial units in the UK have relied on the available openings of the
building involved in fire, windows, doors, or occasionally roof lights are
used when appropriate to exhaust the contaminants. The problem foreseen
initially with ppv in UK construction buildings is the lack of available or
adequate exhaust points to compensate for the amount of air entering the
building from the ventilation units. Flashover,
backdraught and firespread were all perceived problems associated with the
fans in early assessments.
Early
in 1994, Warrington fire research consultants undertook a survey of
ventilation when used as a fire-fighting tactic; the report!from the survey
was published later that year. The report!concluded that further research
needed to be undertaken on the subject of ppv ventilation since little or no
research existed in the UK. As a result of the Warrington report!the Fire
Experimental Unit at Moreton in Marsh undertook a further research project,
this work to be backed up by a survey of the expertise available in the USA
and other countries (4).
The research results indicated that ppv to be particularly suited to fire
attack in small domestic dwellings however, caution was expressed regarding
its use (ppv) due to the ‘significant amount of additional training’ that
would be required.
Two trials of note were conducted by the Fire Experimental Unit using the Fire
Service College’s facilities; the first using fires on the ground floor and
the second experiment consisted of fires in a basement.
The
results of these experiments demonstrated the need for a systematic
ventilation procedure when using ppv. It was found that any compartment where
an outside window was not opened was effectively sealed by the ppv, preventing
any smoke clearance. Where ppv was not used the compartment cleared gradually,
but this did not happen with ppv. The main conclusion from this experiment is
that if using ppv for smoke clearance, rooms should be cleared one at a time
and once a room is clear of smoke, the windows should then be closed. The
report!further concludes that these results were due to the limited flow from
the portable ventilation units and that ppv was not as effective in clearing
rooms with upwind windows.
Another very important point to be concluded from the Experimental Unit’s
experiments was the fact that ppv can easily be overcome by a strong wind
therefore its use is limited to certain scenarios.
A comprehensive research document was produced by J G Rimen (5),
which researched the use of positive pressure ventilation in domestic
properties. The document describes a series of trials conducted in still air
conditions and a further series of trials conducted in a four bedroom detached
house. J.G Rimen had listened to the views of Fire Officers and others in the
building industry and acknowledged the sceptics who regarded increasing the
oxygen content to fires via ppv as dangerous.
A series of trials were undertaken in a ‘mock’ detached dwelling, the
primary objective being to determine basic rules for the use of ppv at
incidents. The results obtained during the trials were recorded using a data
logger and the results subsequently analysed.
The report!concluded that an inlet / outlet ratio of 2 / 1 is the ideal ratio
for the use of ppv and this is what Brigades should aim for. If this ratio is
unable to be obtained then it would be still advantageous to have the inlet
opening larger than the outlet opening. Experiments undertaken also suggest an
optimum distance of 1.0m from the inlet opening for the fan and that it was
not necessary to seal the opening with the fan, this last conclusion
contradicts current working practises which have been adopted from other
research. Another conclusion from the report!indicates that the most efficient
method of using the ppv fan is when it is placed so that air is projected
horizontally into the building.
The research conclusions indicate that ppv has the capability of rapidly
improving the environment at fires, but can also make things worse and that it
is virtually impossible to predict with any certainty the effect of the fan in
a given circumstance. This particular conclusion from the report!was used to
back up the evidence of those individuals opposed to the adoption of ppv for
fire service operations.
Others adopted a totally opposing opinion within the UK fire service for ppv
application as a result of J.G Rimen’s report. The report!states that
‘each fire situation, and specifically whether or not to deploy ppv, would
need to be considered on its particular merits’, in other words risk
assessment. Those in favour of ppv use argue that a ventilation fan is just
another piece of equipment available to the incident commander and with
adequate training and correct incident assessment, it (ppv) is safe to use.
Fire
Authorities in the UK have acknowledged the training requirement for ppv; two
metropolitan Brigades have conducted fire experiments to assess the training
implications and effectiveness of ppv.
Tyne
and Wear Metropolitan Fire Brigade acquired the use of two derelict
semi-detached properties; these were subsequently used for tactical
ventilation trials in real fire situations during November 1996. The
properties used were identical in construction and layout, which enabled fires
to be repeated and compared.
Cribs
were placed in various rooms within the houses and holes drilled in the walls
at various heights to facilitate thermocouples. Temperature readings for all
the exercises were recorded and used to compare the effect of ppv in different
locations of the houses. Wooden pallets, straw, diesel and paraffin produced
the heat and smoke for the experiments with maximum temperatures reaching
around 800°C.
The
experiments from Tyne and Wear are summarised as follows (6):
·
Experiments carried out indicated ppv
would overcome wind strengths of 8-10m/s, the experiments carried out by the
experimental unit at Moreton showed ppv to be ineffective against a 2m/s wind.
·
Ppv did not seriously intensify or
spread a fire however some localised flaming did occur.
·
Ppv improves visibility in a short time
scale.
·
Ppv reduces temperatures dramatically
in a short period of after application.
·
Effective use of ppv improves the
working environment and reduces the risk of flashover.
The
Tyne and Wear report!recommends the defensive use of ppv and appears positive
regarding the use of ppv offensively.
On 1st May 1998 Greater Manchester Fire Service carried out a
series of tests using disused terraced properties. The objectives for the
tests were to document and develop the behaviour of fire using offensive
positive pressure ventilation, the results to be used for the formulation of
Brigade training policy. The tests were carried out with assistance from the
University of Central Lancashire, who produced the report(7) from
which this information is drawn.
Three main tests were carried out; a bedroom fire, with a vent made prior to
application of the fan, a bedroom fire without a vent, a lounge fire with vent
made after application of the fan.
Equipment
from the University Of Central Lancashire was used to measure temperatures,
gas compositions and fire growth in a similar manner to the Tyne and Wear
experiments. Video cameras erected outside the buildings observed the smoke
movement from the exhaust points whilst a thermal imaging camera was used to
show heat transfer and fire spread internally.
With these tests the fire was not confined to a crib and sufficient materials
were present to support!flashover. The fire load consisted of upholstered
furniture and chipboard that were arranged to simulate the furniture and
fittings of a typical room.
During all the tests, the application of PPV produced higher temperatures
within the fire compartment after an initial decrease. It should be mentioned,
however, that two of the tests were deliberately arranged so that ppv was used
incorrectly.
In the first test, the compartment (internal) door was open until it was
considered the fire had almost reached the fully developed stage, and then the
door was closed. The graph indicated that the fire had become strongly
ventilation controlled by these actions, dropping from a temperature of 7000c
to just over 200. The room was then vented, by breaking the window and opening
the internal door, prior to applying the ppv fan.
The
same criticism can be applied to each of the tests carried out i.e. it cannot
be concluded that ppv accelerated the fire any more than natural ventilation
alone would achieve.
During
all the tests a visible ‘flashback’ was reported on application of ppv.
The timing of this event for each test is unclear and no evidence is apparent
on the graphs. No reference had been made to this effect before and
consequently it is an interesting area for consideration.
There
are a number of possible reasons for this effect:
 | The fire may be undergoing
transition from growth to fully developed, this transition is rapid; the
fire wants a large volume of oxygen which results in lengthened flames.
The fire nearly always becomes ventilation controlled at this stage.
|
| The fire compartment
pressure due to buoyancy is in excess of that in the doorway.
|
| There is too restrictive a
vent to outside; the fire had nowhere else to go.
|
| The fire was affected by
wind from outside.
|
The
data from Tyne and Wear does highlight a small localised increase in
temperatures after ppv application, but does not mention the possibility of a
‘flashback’ on initial application, the report!does acknowledge the
potential danger of offensive use.
In
July 1998, Tyne and Wear Fire Brigade carried out a further set of trials at
the Fire Service College. The trials were conducted in conjunction with the
Fire Experimental Unit of the Home Office. The brief for these tests was to
establish the effects of ppv on a casualty situated between the fire and the
vent, and the effect of ppv on fire spread, two issues perceived at the time
to be holding ppv use back.
The
report(8) of the trials, produced by the Fire Research and
Development Group, concludes; ‘ppv produced no measurable difference in fire
spread from that observed under natural ventilation alone. However
ventilation, whether natural or forced, did appear to increase fire spread.’
This
was an interesting conclusion which again conflicts with the findings from
GMC.
Note:
the performance of positive pressure ventilation fans has improved
rapidly since their inception. Many of the early trials used fans that were of
low output and consequently some results may be misleading.
1.
Preston Fire Trials
2.1
The Properties
The
properties used for the trials were a connected block of single-storey, two and
three bedroom flats, built circa 1970. The block was scheduled for demolition
the week following the trials. The properties had been donated for use by
Collingwood Housing Association.
The
construction of the properties was mainly of concrete, which proved ideal for
repeatable experiments.
Plan
drawings of the properties are included in Appendix A
2.2
Equipment and Technical Apparatus
Appliances
and general firefighting equipment was provided by Lancashire Fire and Rescue
Service.
Positive
pressure ventilation fans were supplied by FSE Ltd. The company supplied a range
of fans for use but for consistency the same fan, (18” Tempest) was used
throughout the trials.
Technical
measuring apparatus was supplied and controlled by the University of Central
Lancashire.
Each
experiment house was instrumented using thermocouples (temperature measurement),
and a gas analysis probe. Thermal imaging equipment was also utilised inside and
outside the properties. A digital camera was used to create a photographic
record of events.
Wind
velocity for each experiment was measured with a Vane Anemometer. The wind
conditions were gusting light throughout the week, consequently the wind was
considered to have had a negligible effect and are ignored in the analysis
below.
3
Experiments
3.1 Overview
The
experiments were to take place over three days, Tuesday, Wednesday and Thursday.
Each experiment was repeated using both defensive tactics and offensive tactics.
It had been decided that time was available to carry out three experiments per
day. Consequently, the third experiment each day was set aside to enable a
repeat experiment to be carried out in light of what had happened on the first
two experiments.
Due
to a very high level of interest shown by the media, it was decided that the
final day, Friday, should be considered as a press day to allow demonstrations
and interviews to take place. The reason for this was to ensure that the
experiments could take place Tuesday, Wednesday and Thursday without
interruption. Some experimenting however did take place on the Friday and is
included in this project.
The
experiments are described in the order that they took place. Included in the
description are the graphs drawn from thermocouple and gas data, and the visual
observations of the fire-fighters involved.
3.2 Events
of 10th January 2000
The
Monday of the experiment week was given over to preparation. The venue was
visited for the purpose of carrying out a final assessment of the site. The
following tasks were carried out:
·
A final health and safety risk
assessment check
·
An initial inspection by the technical
support!team
·
Collection of furniture from storage
·
Preparation of the first experiment
house for technical apparatus
·
Collection of fire-fighting equipment
3.3 Events of 11th January 2000
3.3.1
Experiment 1A
Scenario
– defensive positive pressure ventilation
Fire
crews arrive to find a well-developed front room (lounge) fire. The objective of
the fire crews is to enter the premises through the rear door wearing breathing
apparatus, and extinguish the fire using standard operating procedure.
Objective:
To
develop a well-established single room fire and to measure the temperatures in
and around the fire compartment. To measure the temperatures and gas levels in a
bedroom adjacent to the fire compartment. To establish the effects of defensive
positive pressure ventilation fire-fighting tactics on such a fire.
A
fire was started in a sofa in the lounge. The fire loading in the lounge was
similar to that which would be expected in a real situation. Once the fire was
lit, the fire was allowed to develop until such time as was considered necessary
before fire crews were instructed to enter and tackle the blaze. Within the
house, only the door to the lounge and the door to the bedroom were open. All
external doors and windows were closed.
Parameters
|
T0
|
Entry
|
Attack
|
Vented
|
Fan
on
|
Vent(m2)
Inlet
Outlet
|
|
150c
|
210sec
|
240sec
|
380sec
|
400sec
|
1.6
|
0.62
|
Results/Analysis
Refer
to graphs 1A1 – 1A5
Fire
compartment temperatures peaked at around 10000c and then decayed
rapidly, probably due to the fire becoming ventilation controlled. The fire
intensified slightly when the rear door was opened. On application of water the
temperatures within the fire compartment reduced only minimally, hovering around
5000c. Temperatures within the bedroom reached more than 5000c.
Compartment temperatures only dropped rapidly when the fan was applied.
The
thermocouple tree indicates that the neutral plane, (base of smoke layer), was
somewhere in the region of 900-1200mm from floor level at the most intense
period of burning.
The
gas analysis shows that the atmosphere would not have sustained life until
ventilation commenced. CO levels (measured in the bedroom at bed height),
reached 28000 parts per million – the highest levels reached during these
experiments. Once the fan was applied the CO levels dropped rapidly and the O2
levels increased from 10% to 20% within 30 seconds.
Visual
Observation
Flames
were apparent in the early stages, but appeared to be subdued by the smoke
levels. On venting, there was a large quantity of grey smoke. The rate of
exhaustion of this smoke increased markedly on application of the fan.
Observation
of the post-fire scene showed that there was much combustible material remaining
in the fire compartment.
3.3.2
Experiment 1B
Scenario
– offensive positive pressure ventilation
Fire
crews arrive to find a well-developed front room (lounge) fire. The objective of
the fire crews was to enter the premises through the rear door wearing breathing
apparatus, and extinguish the fire using offensive positive pressure ventilation
tactics.
Objective:
To
develop a well-established single room fire and to measure the temperatures in
and around the fire compartment. To measure the temperatures and gas levels in a
bedroom adjacent to the fire compartment. To establish the effects of offensive
positive pressure ventilation fire-fighting tactics on such a fire, and compare
the results to those of the previous experiment.
A
fire was started in a sofa in the lounge. The fire loading in the lounge was
similar to that which would be expected in a real situation. Once the fire was
lit, the fire was allowed to develop until such time as was considered necessary
before the fire compartment was vented and the fan was applied via the rear
door. Fire crews were then instructed to enter and tackle the blaze. Within the
house, only the door to the lounge and the door to the bedroom were open. All
external doors and windows were closed.
Parameters
|
T0
|
Vented
|
Fan
on
|
Entry
|
Attack
|
Vent(m2)
Inlet
Outlet
|
|
160c
|
230sec
|
240sec
|
260sec
|
290sec
|
1.6
|
0.47
|
Results/Analysis
Refer
to graphs 1B1 – 1B5
Again
the fire appears to have developed rapidly although temperatures were lower,
peaking at around 7500c, than in Experiment 1A. The fire was in decay
when fire-fighting operations began. No fire growth occurred on venting or
applying the fan. The fire has appeared to die down on its own accord, possibly
due to the lack of ventilation causing a smothering effect, the fire was
probably smouldering and the smoke mixture in the compartment was possibly very
fuel-rich, although when the fan was operated there wasn’t sufficient energy
in the fire to create ignition. It is possible that backdraught conditions were
achieved here and if the fan had been applied moments earlier then perhaps a
backdraught would have occurred.
Some
cooling was apparent on application of the fan, particularly at the probes on
the entrance to the bedroom and on the thermocouple tree on the internal fire
compartment doorframe. The neutral plane appears to have not reached a level
quite as low as test 1A, reaching somewhere between 1200 – 1500mm from the
floor.
The
gas analysis shows a similar profile to the previous experiment, with a dramatic
improvement of conditions on application of the fan. CO levels peaked at
approximately 15000 parts per million. The improvement in O2 levels
was slower than in the first experiment.
Visual
Observations
The
fire fighting crew reported smoke coming back down the hallway, towards the fan.
This may be the cause of the slower improvement in gas levels. It was thought
that this might be due to the exhaust vent being to small, creating an
overpressure in the compartment. The vent ratio was 3.4:1 in favour of the
inlet.
3.3.3
Experiment 1C
Scenario
– offensive positive pressure ventilation
The
previous experiment, 1B, was repeated this time using a larger vent size. The
reason for this was to see if the vent restriction was the cause of the smoke
returning down the corridor.
Parameters
|
T0
|
Vented
|
Fan
on
|
Entry
|
Attack
|
Vent(m2)
Inlet
Outlet
|
|
160c
|
230sec
|
240sec
|
260sec
|
295sec
|
1.6
|
1.0
|
Results/Analysis
Refer
to graphs 1C1 – 1C5
A
similar fire profile was achieved although this time the fire intensified on
venting and application of the fan. The intensification in terms of temperature
was only in the region of 750c. The rate of growth due to this was,
although observed for only a short period, no greater, perhaps even less, than
the original growth rate prior to venting
The
neutral plane came somewhere between 600 – 900mm, lower than in the previous
test. On application of the fan, the neutral plane appears to have risen,
perhaps as high as 1500mm.
The
gas analysis showed a quicker improvement in conditions. The vent ratio for this
test was 1.6:1,
Visual
Observation
No
report!was made of smoke or flame in the corridor. The conditions on the
approach to the fire compartment were cooler and very much clearer than in the
previous experiment. The fire fighters were very impressed with the operation.
3.4 Events of 12th
January
3.4.1
Experiment 2A
Scenario
– defensive positive pressure ventilation
Fire
crews arrive to find a well-developed bedroom fire. The objective of the fire
crews is to enter the premises through the rear door wearing breathing
apparatus, and extinguish the fire using standard operating procedure.
Objective:
To
develop a well-established single room fire and to measure the temperatures in
and around the fire compartment. To measure the temperatures and gas levels in
the hallway adjacent to the fire compartment. To establish the effects of
defensive positive pressure ventilation fire-fighting tactics on such a fire.
A
fire was started in bedding materials in the bedroom. The fire loading in the
bedroom was similar to that which would be expected in a real situation. Once
the fire was lit, the fire was allowed to develop until such time as was
considered necessary before firecrews were instructed to enter and tackle the
blaze. Within the house, only the door to the lounge and the door to the bedroom
were open. All external doors and windows were closed. Temperature probes were
fitted to the vent.
Parameters
|
T0
|
Entry
|
Attack
|
Vented
|
Fan
on
|
Vent(m2)
Inlet
Outlet
|
|
170c
|
1125sec
|
1155sec
|
1180sec
|
1195sec
|
1.6
|
0.67
|
Results/Analysis
Refer
to graphs 2A1 – 2A5
Comparatively
low temperatures were achieved with this fire (3600c). Due to the
nature of the combustible materials the fire growth rate was much slower than
that achieved on the previous day.
The
neutral plane was clearly exhibited on the graphs, being between 900 – 1200mm
from the floor. The high temperatures recorded at the vent are thought to be due
to localised burning observed at the window frame by the fire fighters.
Low
gas levels were also exhibited. The combustible materials were predominantly
bedding and the mattress of the bed.
Visual
Observations
This
experiment was visually observed from the hallway by the author using Breathing
Apparatus. The application of the fan, even when the fire was extinguished, had
a dramatic effect on conditions. There was an instant cooling effect, a fire
compartment immediately after a fire attack is very hot and humid. There was
also a dramatic improvement in visibility; ppv created a far more comfortable
environment for the fire fighters.
3.4.2
Experiment 2B
Scenario
– offensive positive pressure ventilation
Fire
crews arrive to find a well-developed bedroom fire. The objective of the fire
crews was to enter the premises through the rear door wearing breathing
apparatus, and extinguish the fire using offensive positive pressure ventilation
tactics.
Objective:
To
develop a well-established single room fire and to measure the temperatures in
and around the fire compartment. To measure the temperatures and gas levels in
the hallway adjacent to the fire compartment. To establish the effects of
offensive positive pressure ventilation fire-fighting tactics on such a fire,
and compare the results to those of the previous experiment.
A
fire was started in bedding materials in the bedroom. The fire loading in the
bedroom was similar to that which would be expected in a real situation. Once
the fire was lit, the fire was allowed to develop until such time as was
considered necessary before the fire compartment was vented and the fan was
applied via the rear door. Fire crews were then instructed to enter and tackle
the blaze. Within the house, only the door to the lounge and the door to the
bedroom were open. All external doors and windows were closed.
Parameters
|
T0
|
Vented
|
Fan
on
|
Entry
|
Attack
|
Vent(m2)
Inlet
Outlet
|
|
170c
|
1715sec
|
1725sec
|
1745sec
|
1775sec
|
1.6
|
0.67
|
Results/Analysis
Refer
to graphs 2B1 – 2B4
The
fan appears to have been applied as the fire growth rate was accelerating and
does not appear to have increased this rate noticeably. There was a rapid
decrease in the hallway temperatures on application of the fan, although this
cooling rate was arrested momentarily. Maximum compartment temperatures reached
8000c.
The
thermocouple tree shows a similar result to that of earlier experiments, in that
the neutral plane rose. In this case the rise was very clear, from between 900
– 1200mm to between 1200 – 1500mm on application of the fan.
Surprisingly
low gas levels were recorded again in this fire, but the decrease in gas levels
followed the same profile as the hallway temperatures. Vent temperatures dropped
at some probes and increased at others, with no clear pattern.
Visual
Observation
This
experiment was visually observed from the hallway by the author. Conditions in
the hallway prior to positive pressure ventilation were hot and very
smoke-logged. On application of the fan the corridor became cooler and clear
very quickly. This made the approach to the fire compartment comparatively
straightforward.
At
some point between the fan being used and the fire being attacked, flames were
observed coming out of the fire compartment and across the hall at head height.
Indications of fire spread outside the fire compartment were identified due to
this and the graphs also record data that could be attributed to such an event.
3.4.3
Experiment 2C
Scenario
– offensive positive pressure ventilation
The
previous experiment, 2B, was repeated this time using a larger vent size.
Parameters
|
T0
|
Vented
|
Fan
on
|
Entry
|
Attack
|
Vent(m2)
Inlet
Outlet
|
|
170c
|
575sec
|
585sec
|
730sec
|
800sec
|
1.6
|
0.89
|
Results/Analysis
Refer
to graphs 2C1 – 2C4
The
fire demonstrated a similar development curve as experiment 2B. The fire appears
to have reach steady state, possibly following the onset of flashover. There was
no significant growth rate increase on application of the fan, which would be
expected if the fire were at this stage fully developed. Maximum temperatures
reached 8000 c, consistent with flashover.
The
thermocouple tree showed a steady neutral plane, even after application of the
fan, again, to be expected with a fully developed fire. Hallway temperatures
dropped significantly (1500c) but then appeared to begin to increase.
Gas
analysis was inconsistent, and difficult to quantify, but it appears that the
fan reduced gas levels initially, but then the gas levels began to increase
again.
Visual
Observation
Observers
inside the building stated that the ‘flashback’ effect of the previous
experiment had not been in evidence and that conditions in the hallway had
improved considerably on application of the fan. Observers from outside at the
vent witnessed a large amount of flame and smoke issue shortly after the fan was
applied.
3.5
Events of 13th January
Experiment 3A
Scenario
– offensive positive pressure ventilation
Fire
crews arrive to find a fire in the middle bedroom. They use offensive positive
pressure ventilation tactics but ventilate the wrong compartment.
Objective:
To
develop a well-established single room fire and to measure the temperatures in
and around the fire compartment. To measure the temperatures and gas levels in
the hallway between the fire compartment and the compartment where the vent is
situated. To establish the effects of offensive positive pressure ventilation
fire-fighting tactics on such a fire to that has been incorrectly vented.
Parameters
|
T0
|
Vented
|
Fan
on
|
Entry
|
Attack
|
Vent(m2)
Inlet
Outlet
|
|
160c
|
195sec
|
210sec
|
430sec
|
455sec
|
1.6
|
0.67
|
Results/Analysis
Refer
to graphs 3A1 – 3A4
The
application of the fan appears to have slightly reduced the growth rate of the
fire, although the fire has continued to grow after the fan was applied.
In
the hallway, the temperature has risen on application of the fan, as expected. A
similar effect was exhibited at the vent, although the effects are perhaps more
exaggerated.
The
thermocouple tree shows an initial drop in neutral plane level, prior to the
neutral plane quickly re-establishing its original level.
The
gas levels were adversely affected by the fan, increasing dramatically on
application.
Visual
Observation
A
char stick (narrow stick of timber), was placed in the hallway for this test, at
ceiling height, to help gauge any potential fire spread. Following the fire, the
stick was found to be severely charred, indicating direct flame impingement from
the fire or self-ignition due to temperatures it was exposed to.
The
increase in smoke issuing from the vent was clearly evident when the fan was
applied.
3.5.2
Experiment 3B
Scenario
– offensive positive pressure ventilation
Fire
crews arrive to find a fire in the middle bedroom. They use offensive positive
pressure ventilation tactics this time ventilate the correct compartment. This
scenario was used in an attempt to use positive pressure ventilation offensively
to its best effect, using the lessons learned during the week. The scenario was
similar enough to the previous experiment to allow reasonable comparisons to be
drawn.
Objectives:
To
develop a well-established single room fire and to measure the temperatures in
and around the fire compartment. To measure the temperatures and gas levels in
the hallway adjacent to the fire compartment. To establish the effects of
offensive positive pressure ventilation fire-fighting tactics on such a fire.
Parameters
|
T0
|
Vented
|
Fan
on
|
Entry
|
Attack
|
Vent(m2)
|
|
160c
|
500sec
|
515sec
|
560sec
|
590sec
|
1.6
|
0.5
|
Results/Analysis
Refer
to graphs 3B1 – 3B4
The
fire appears to have peaked at around 5500c, and then reduced
slightly, steadying at around 4000c. This is possibly due to the fire
becoming ventilation controlled. The fan, although generating an increase in
temperatures did not appear to increase the growth rate in the fire compartment
to any greater rate than it had exhibited prior to peaking.
Hallway
temperatures increased initially with the fan but then dropped rapidly prior to
fire attack.
Gas
levels dropped dramatically on fan use. Again, some of the vent probes showed a
drop in temperature whilst others increased.
Visual
Observation
Exterior
observation showed a dramatic increase in the volume of smoke and flames when
the fan was applied. Again, fire-fighters observed vastly improved conditions
inside the house.
3.5.3
Experiment 3C
A
experiment was carried out to assess the effectiveness of various fans (18”
and 21” Tempest) in the smoke clearance of a staircase enclosure.
The
staircase was smoke logged by burning sofa cushions. It was then vented at the
head and the time each fan took to clear the staircase of smoke was taken. The
experiment was measured visually only and consequently is of limited value,
however it was agreed by all observers that the 21” fan was obviously the most
effective for this purpose.
An
interesting point to note was that the smoke was cleared very quickly with both
fans, demonstrating the usefulness of ppv in the post-fire situation.
3.6
Events of 14th January
As
previously stated, the day was set aside for the press, and several
demonstrations were carried out, utilising offensive ppv tactics. It is worthy
of note that each of these demonstrations were carried out successfully.
In
the afternoon, time became available to run some tests of the effect of ppv on
chip-pan fires. Instrumentation was unavailable for these tests, however events
were recorded from inside the houses on thermal video.
The
basis of these tests was to demonstrate the effectiveness of ppv by using a fan
to ‘hold-back’ the effects of a chip-pan fire when water is applied. This
was arranged using a cup attached to a stick passed through a hole drilled in
the kitchen wall.
The
fan was started, and allowed to pressurise the house. The chip-pan was alight at
this stage. At a given signal the water was poured in resulting in the
well-documented result. When the fan was used, the flash flames from the kitchen
were prevented from entering the house further and the majority were exhausted
through the open kitchen window.
This
was an unscientific, but nevertheless impressive demonstration for all who
observed. The results can be viewed clearly on the thermal images, which are in
the process of being transferred to standard videotape, along with the video
footage of all the experiments of the week.
4.
Experiment Analysis
4.1
Introduction
The
analysis is presented separately for each day of the tests. Conclusions drawn
from the analysis are included in section 5.
Throughout
this analysis, consideration of effects of wind are not considered due to the
fact that the experiments were conducted in light, gusting winds all week and it
is believed that consequently the wind had little, if any, detrimental effect on
the results.
4.2
Analysis/Synthesis of 11th January Experiments
The
absence of fire growth on application of positive pressure ventilation on the
second experiment appears, when looking at the experiments as a whole, to be a
rare event. It is considered that this fire was strongly ventilation controlled,
but had subdued to such an extent that further ventilation had little effect.
This was not the case in the first fire, which appears to have self-extinguished
-- the reasoning behind this assumption is that the application of water
had no cooling effect. There was also plenty of combustible material remaining
in the post-fire compartment, as the plate below testifies.
Plate
2 Experiment 1A – Post-fire
compartment
With
the second fire, the vent was the smallest, and in the highest position. An
alternative explanation to the fact that no growth was exhibited is that the
vent was too small, and too remote from the burning combustibles to have had a
direct affect on the fire. The third test of the first day produced a very
similar curve to the second, but with a larger vent. This fire had intensified
on venting and application of the fan. Thus, if this is the case, the fan had no
detrimental effect and the growth has to be attributed to the vent alone. This
correlates with the analysis of the Manchester results and the discussion in the
literature review.
Fire
crews reported cooler and clearer conditions in the approach to the fire
compartment in the second experiment than in the first experiment. The smoke
that was observed moving back towards the entry point was probably due to the
exhaust vent being restrictive. This, combined with the fact that the fire was
still burning probably was the cause of the slower increase in O2
levels.
Plate
3 Showing Vent for Experiment 1B
This
effect was not noted in the third experiment, where the vent was larger.
Thus, the restrictive vent, although creating no fire growth, rendered
the use of offensive ppv less effective. It is also possible that the adverse
smoke flow patterns generated by the smaller vent and fan could lead to fire
spread remote from the fire compartment. The plate below shows the difference in
effect created by the larger vent size for experiment 1C.
Plate
4 Showing Effect of Larger Vent
Size (Experiment 1C)
It
is interesting to note that similar fire profiles were achieved on all three
experiments even though the fire load materials were different. Although the
temperatures and smoke production rates were slightly different, the experiments
appear to have produced meaningful results.
4.3
Analysis of 12th January Experiments
The
second day’s experiments were focused on the bedroom scenario. All three fires
were very much slower in developing, but all three accelerated rapidly at a
given point, interestingly when the compartment temperature had reached 250 –
3000c.
The
second experiment, 2B, was in rapid growth when the fan was applied. The fire
was possibly approaching flashover at this point. This fire produced the
‘phenomenon’ of flashback, with flames projecting back into the hallway from
the fire compartment. This phenomenon, which was reported in the Manchester
tests, was discussed in the literature review. It was postulated that the fire
had become momentarily ventilation controlled, due to the restricted size of the
vent, and flames were seeking oxygen as the fire was becoming fully-developed.
It
may be that the flames were allowed to exit the fire compartment, against the
pressure generated by the fan, due to some inconsistency in the pressure planes
at the doorway to the compartment.
The
thermocouple tree graph shows a dramatic drop in the temperature of the
thermocouple at 1200mm, which is below the height that the flames were seen
issuing. This ties in with the theory above, and that the fires rapid growth
suddenly required a much greater rate of oxygen uptake
It
is therefore suggested that the flashback phenomenon that has been associated
with ppv use marks the transition of a growing fire to a fully-developed,
ventilation limited fire. As a flashback was not reported for any of the other
experiments, consequently its occurrence in this case might well be due to the
undersized vent.
It
is possible that in the absence of ppv this would lead to rapid fire spread.
Again,
application of the fan did not increase the burning rate significantly.
Temperatures remote from the fire compartment were improved with positive
pressure ventilation, although indications that incorrect use could worsen the
situation were noted. Experiments 2B and 2C appear to suggest that the vent size
is the important consideration, as the fire, in under ventilated conditions will
be forced to seek oxygen from the inlet, creating a situation where the fire
moves towards the fire-fighters. This also indicates that a casualty, situated
between the fire and the vent, may be adversely affected by offensive positive
pressure ventilation use.
4.4
Analysis/Synthesis of 13th January Experiments
The
first fire was a deliberately created scenario where the wrong compartment was
ventilated. The effect of the fan on the fire compartment was to create an
initial decrease in temperatures, followed by growth. The fire appears to have
grown, but at a slower rate than prior to application of the fan. Temperatures
and gas levels both increased quite dramatically in areas remote from the fire
compartment.
The
neutral plane appears to have risen distinctly when the fan was applied, from
between 600 – 900mm to between 1200 – 1500mm. This was a momentary effect
however; the plane appeared to re-establish its original level. The plane then
seems to behave erratically, perhaps suggesting turbulence in the fire
compartment.
The
behaviour of this fire on the application of ppv, from a practical point of
view, had been expected. Pressurising a fire compartment to a greater pressure
than that of the area of the vent is obviously likely to encourage the products
of combustion to migrate towards the area of lower pressure. The effects may
have been worse, had the fire compartment not self-ventilated shortly after the
fan was applied. This however, raised an interesting point; that the glazing to
a fire compartment will be in a weakened state and pressurisation from a fan
could generate self-ventilation, which would generally be beneficial.
Plate5
Experiment 3A, shortly after self-ventilation of the fire compartment.
Note: the actual vent is the window to the right of the fire compartment.
The
use of the fan when the wrong compartment is vented seems to have a potentially
dangerous effect. Both temperatures and gas levels were increased remote from
the fire compartment. There were indications that fire would have spread beyond
the confines of the fire compartment had the fire not been extinguished at that
stage. The graphs indicate that the actions of the fire fighters using the ppv
fan probably changed the atmosphere within the building from one of marginal
tenability to one where survival was unlikely.
The
second experiment, 3B, was a straightforward attempt to utilise ppv to its best
effect, with knowledge of the location of the fire. Temperatures had become
steady at around 5000c prior to application of the fan and venting,
suggesting ventilation control. Again, on application of the fan temperatures
indicate the fire began to grow, and again the growth rate was no greater,
possibly lesser than, the original rate.
Plate6
Experiment 3B, Immediately after Venting.
Visual
observation indicated that the fire became fully developed, the position of the
neutral plane was unclear; the thermocouple tree was positioned at the door but
this was only a small room, the thermocouples possibly being affected by direct
radiant feedback.
Plate7
Experiment 3B, on application of the fan.
Hallway
temperatures dropped rapidly on application of the fan, as did the gas levels in
the corridor.
In
essence, this fire behaved as had now become expected in that the fire
intensified, and the conditions remote from the fire compartment were generally
improved.
4.5
Post-fire use
Experiment
3C, and each of the experiments were post-fire ppv was employed, indicates that
the post-fire use of positive pressure ventilation is a very effective method of
cooling and clearing a property of smoke. Although no comparison was made
between the use of a fan and natural ventilation, it is clear, from visual
observation and discussion with fire fighters, that the use of a fan is very
advantageous for smoke clearance in the absence of very favourable wind
conditions.
4.6
Questionnaire
A
questionnaire was circulated to all volunteers that participated in the trials.
Of
the volunteers who attended the trials and returned the questionnaire, there was
a strong consensus of opinion. All respondents were fully confident of the
effectiveness of ppv. Some of the volunteers had been sceptical of its use,
particularly offensively, and had attended in order to formulate their own
opinion. All stated that they would be confident to use ppv offensively in
circumstances similar to those of the tests, and consequently that further tests
are needed on multi-storey premises.
The
same general points were reiterated throughout. Backdraught, firespread and
communications were felt to be the key areas. These issues are addressed below.
5.0 Conclusions
and Recommendations
The
conclusions below can only directly apply to situations similar to the ones
covered by these experiments (single-storey domestic premises), although in some
instances, assumptions may be drawn regarding the application of ppv to other
circumstances.
5.1 Firespread --
Internal
Generally
speaking, the fire growth rate was not greatly increased by ppv. In fact it is
postulated that the use of a fan creates no greater rate than the use of natural
ventilation. There is likely to be, in certain circumstances, an increase in the
burning rate but this is due to the airflow causing horizontal fire spread
across individual burning items. Any domestic fire where offensive ppv is used
would be attacked shortly after the fans application so the rise in temperature
etc., in the fire compartment would be short-lived.
Providing
the vent size is gauged correctly, and the fire compartment is not breached, the
risk of enhanced fire spread appears to be minimal.
However,
Experiment 2B gave an indication that if the compartment is breached, there is
an opportunity for the fire to break out of the compartment, if the breach leads
to an area of lower pressure.
Should
the fire compartment be breached, then it is likely that ppv would create a
potentially dangerous situation, generating spread to other parts of the
building.
It would be
possible in future experiments to artificially breach the compartment and
consequently examine this possibility. This is therefore recommended as an
important area for further research.
5.2
Firespread – External
The
application of the fan appeared to observers to generate larger flames than
would ordinarily be expected from the exhaust vent. No conclusive evidence is
available of this as in many tests flashover conditions were taking place and
there is no means of ascertaining the difference caused by the fan, if any.
It
is standard practice that where a vent is created it is protected. It is perhaps
more critical that this practice is not forgotten where ppv is used.
Thus the
protection of vents should be part of the operating procedure for the use of ppv
offensively.
5.3 Fire
Development
Generally
speaking, the fire growth rate was not increased by ppv during these
experiments. In fact it is postulated that the use of a fan creates no greater
rise in fire growth rate than the use of natural ventilation, other than some
local growth due to airflow.
Once
a fire becomes fuel-controlled, an increase in vent size becomes unimportant.
The fire in effect has sufficient oxygen. This was evident in the experiments
above as where the fire did intensify on application of the fan and venting, the
growth rate was no greater than that at which the fire had been growing prior to
becoming ventilation controlled.
This
is a very important issue, as it appears to be the stumbling block for ppv in
this country. The hesitance to adopt ppv for offensive use comes mainly from the
fear of intensifying an already dangerous situation. If it can be shown
consistently, in more controlled conditions, that ppv does not intensify the
fire any more than standard ventilation of the compartment, it is more likely to
be readily accepted.
In
most cases, ventilation caused the fire to resume growth. However, any domestic
fire where offensive ppv is used would be attacked shortly after the fans
application so the rise in temperature etc., in the fire compartment would be
short-lived.
5.4
Vent size
This
is an important area for consideration as it proved to be the largest
influencing factor on the behaviour of the fire and its products. The widely
held belief that the optimum vent size should be 2:1, inlet to outlet, appears
to have been borne out by these experiments.
The
‘flashback’ reported by GMC Fire Service was witnessed during Experiment 2B.
It appears to have been caused primarily by the exhaust vent being too small in
relation to inlet. Where circumstances are such that the above ratio of 2:1
cannot be achieved, it has been suggested that the flashback could be prevented
by reducing the speed of the fan and thereby the airflow.
This
is an interesting event and it is unclear whether the flashback was caused by
the fire searching for oxygen or by an area of low pressure in the region above
the compartment door.
It is therefore
recommended that the above issue be researched in future experiments.
5.5 Vent choice
This
issue has already been mentioned but is worthy of separate discussion. In the
scenario of experiment 3B, it was clear that the wrong vent choice caused severe
problems for the fire fighters, and potentially for casualties. It is therefore
considered straightforward to state that, given a similar scenario, the affected
compartment must be vented.
This,
of course, is not always a simple matter as it is often not clear as to which
room of a house is alight. Venting
the wrong compartment could lead to the spread of combustion products to areas
unaffected by the fire. It appears that a casualty, or fire fighters, may be
adversely affected if the vent choice is wrong.
Thus it is
recommended that vent choice be seen as a critical area for the training and
command and control of fire fighters.
5.6 Fan
positioning
No
positioning experiments were carried out however the fan performed well
throughout the week without any special consideration being given to the
placement of the fan. From this it appears that positioning is not critical –
probably due to the fan used being of a high capacity in relation to the size of
the property.
Experiments to
establish the optimum position would be relatively simple to arrange and
necessitates individual experiments for each type of fan purchased.
5.7
Command and Control
This
is a critical area and can only be addressed by training. The theory is
relatively simple. Locate the fire compartment, vent it and apply the fan.
However in the potential confusion of a fireground this is difficult to achieve.
Effective communications are crucial, along with a well-trained and disciplined
team.
Even
at these experiments, with the noise of the scene, achieving the coordinated
application of ppv and venting was difficult. Hence ppv demands a heavy
requirement for training.
5.8
Defensive (post-fire) ppv
All
participants were aware of the benefits of Post-fire ppv use in terms of smoke
clearance.
The
trials indicate that there is little danger or detriment when post-fire ppv is
used (domestic), providing it is used correctly. The signs in some tests that
ppv can intensify a fire perhaps raise a concern over the possibility that a
fire, not properly extinguished, may regenerate. Consequently it is imperative
that the fire is extinguished completely or smouldering materials are covered
with a jet prior to application of the fan.
It is concluded
that post-fire use of ppv is very effective for smoke clearance and provides
little risk providing the fire is satisfactorily extinguished.
5.9
Fire fighter safety
There
is no doubt that when ppv is carried out successfully the benefit in terms of
fire fighter safety is great. Fire fighters can access the fire in clear, cool
conditions, obviously removing many of the inherent dangers of entering,
searching for and attacking a fire.
It
is probable that the danger from flashover and backdraught are also removed from
fire fighting, as it is likely that offensive ppv tactics will ensure that such
phenomenon will have occurred (if they were going to occur) as a result of fan
use before fire fighters enter the building.
It is recommended
that further research be carried out into the effect of ppv on conditions
favourable for backdraught.
5.10
Casualty survival
In
circumstances were a fire is reasonably well developed, it is likely that a
casualty within the fire compartment will be beyond recovery. The fire may
intensify on application of ppv but fire fighters attacking the fire earlier,
due to the improved conditions, may mitigate this effect.
It
has been seen that ppv can improve the tenability of the parts of a building
remote from the fire compartment.
The
danger to a casualty not located in the fire compartment comes primarily from
the products of combustion. Ppv has been seen to rapidly improve gas levels. As
such, in many instances the atmosphere in the location of the casualty will
become tenable almost as soon as the fan is applied, and the time taken in
search and rescue is therefore less crucial.
Due
to the improved conditions, a search and rescue team may be able to affect a
faster rescue.
5.11
Wind Effects
During
the trials the wind appears to have had no adverse affect on the use of ppv. In
general the wind speeds were considered to be gusting light throughout the
trials. Wind speeds were recorded gusting between 1 – 4 m/s.
5.12
General conclusion
It
is clear from these experiments that ppv, when used correctly, can be extremely
beneficial to fire-fighting. That
said, it is not a simple task to ensure correct use, as there are many variables
to consider.
Much knowledge has
been gained about the effects of ppv, however, its use cannot be progressed
until such time as trials using two-storey properties have been carried out.
To this end, it is
strongly recommended that a further series be arranged.
Training
is the next area to be addressed once further trials have taken place. Training
has been identified as a critical issue for ppv but unfortunately the technique
demands a large amount of training input.
Consequently,
it would be extremely beneficial if the models created by Lancashire fire
fighters were validated during these future experiments, as the value as a
research and training tool of such models is enormous. Once reasonable
validation has been achieved, there would be no requirement for further
full-scale trials, which are unfortunately very heavily resource orientated.
