How will a cooker extractor fan affect detection time?

[wainy1]

Hi All
I would value your opinions on the following matter;
I have been consulted on a design for a studio flat within student halls of residence. The building consists of 3 floors; the flats are approximately 4.5m x 8.5m, self-contained, complete with bathroom, Kitchenette and sleeping area. The flat is served by a single entrance door.
The internal design of the flat varies from the guidance contain in 2.13 of Approved Doc B (internal planning of flat) with the bed being located at the furthest part of the room from the door and the cooking facilities being located between the bed and the entrance door. ADB suggests the cooking facilities are remote from the entrance door.
The design has also specified the smoke detection is located above the bed to reduce the incidence of unwanted activations of the alarm system.
Above the kitchenette hob an extract fan is provided (which I believe will have a minimum extract rate of 30 litres per second to comply with building regulations).

My question; is the delay in the smoke reaching the detector likely to have a significant effect on the activation time of the detector, should a fire occur on, or in the vicinity of the cooker when the extractor is running?

Many thanks in anticipation

[kurnal]

Is the smoke detector connected to a communal areas alarm or is it a part 6 alarm covering the studio only? Stay put or full evacuation? 

[Phoenix]

While waiting for the answer to kurnal's question I can say this.  We don't have to be too worried about the air movement created by things that 'suck' relatively small quantities of air such as this extractor.  Because extractor fans draw air from every possible direction around them, their sphere of influence (literally a sphere) is limited in scale.  They're not like fans that blow air, such as air conditioning units.  Things that blow do so in a single direction and their effect is felt much further from the device. 

Having said that, the extractor will probably have some effect on actuation time if the fire is right beneath it but it's difficult to quantify and virtually impossible to declare whether or not it will be significant.

I think a lot more people will be concerned about the layout of this flat, maybe believing that a chip pan fire, for example, might stop the occupant(s) being able to get to the door.  Maybe that's a valid point.  Maybe it depends on the precise layout.  Maybe, even if the location of the cooker is such that people will not be trapped by a chip pan fire, the fact that people cannot simply turn their backs on the fire and move away will mean that they are more tempted to try to pick up the burning pan, which is never a good thing to do.  Some designers install small localised suppression systems above the cookers in such layouts.

[wainy1]

Hi Kurnal

The fire alarm system installed has multisensors in each room and is a category L1, It is proposed to utilise simultaneous evacuation.

Hi Phoenix

Thank you for your reply
Although the non-compliant layout gives me some cause for concern, the rooms could feasibly be fitted with heat detectors and be in compliance with 5839 Part 1, smoke detection should provide sufficient early warning in the early stages of fire, to enable a person to evacuate past the fire.
 
My main concern is the combination of the non-compliant layout and the extractor fan (over one of the potential risks) will delay the activation of the detector until the fire has developed and escape past the fire is no longer possible.
From previous experience I am aware extractor fans have been used in sheltered housing schemes and similar premises to reduce the incidence of unwanted fire signals with some success, however this only works because it will delay activation of the alarm. 

The buildings are almost to the stage where they will be handed over from the developer and therefore any changes at this stage will be difficult without having definitive evidence.

Has anyone modelled a fire in a compartment where there is smoke extract directly over the fire? Any views or opinions would be welcome.

[John Webb]

As far as I am aware, the only smoke modelling using extracts over the fire has been for much larger fires in much larger buildings, such as shopping centres, warehouses and the like.

Assuming a 3m high ceiling, your flat is around 115 cubic metres (cm) in volume. Your fan is extracting 1.8cm per minutes, less than 2% of the volume. I would have thought the effect on detection would have been minimal - and while the extract might delay detection of a chip pan or other fire on the stove, it will also be removing a large part of the products of combustion, until such time as the fan is damaged by the heat.

The one question I would ask is "Where is the make-up air coming into the flat?" because that might well affect detection if a high-level vent into a room is being used.

[wainy1]

Thanks John

I assume the make up air for the extraction is provided from the gap under the entrance door or through the bathroom vent (however if this is also running it may result in an even longer delay until activation)

[kurnal]

I see a lot of this wainy1. Student accommodation is becoming a free for all in respect of Building Regs. I was brought in to look at one scheme that sounds similar to yours where they were having 40 unwanted fire alarms per week.

I would have a close look at the fire alarm configuration. Multi sensor heads can be configured in so many different ways, some such as the Gent Vigilon SQuads multi sensor detector sounders are brilliant and can be configured so smoke gives a local alarm in the room of origin and at security desk but heat gives a general alarm in the communal areas, some open protocol systems are not so clever. Whether you want to give the person in the room itself a warning of smoke is down to the policy of the accommodation provider - if you do - and most Unis do- then it needs a clever well thought through  C&E with multi sensors  or domestic smoke alarms in the studios so only 1 person is inconvenienced.  I have seen the vent fan linked to the cooker on switch, helps a bit to reduce unwanted alarms but does not eliminate them.

The cookers tend to come with rules, such as no frying, not sure if that can be made to stick though! In most cases you have to pass the cooker to get from the bed to the door and there isn't usually enough room to swing a cat. On the other hand travel distances are usually very short.

The worst cases I have found are in pod based studio units where the pods are brought onto site fully fitted out and stacked on top of each other many storeys high, all fully wired plumbed and vented, you end up with all services brought into communal areas for each pod, millions of connections for water, drainage, vent ducting, wiring, fire alarms and compartmentation none existent or shot to pieces and hidden above a suspended ceiling and under a chipboard floor. Just finished dealing with one such 6 storey block  with zero fire stopping and uninterrupted voids everywhere, and that one was given a completion certificate by a local authority building control. Ended up  having communal areas fully gutted and a refit of the fire alarm. Oh and a dry riser put in because it was "simply overlooked" first time round!

[wainy1]

Hi Kurnal
Thank you for your comments.
The University do indeed consider their duty of care extends to the occupant of the room. The approved inspector is happy the design complies with the Building Regs; however they offer no justification for the non-compliant design which is exacerbated by the cooker extraction.
Although my initial thoughts are that sufficient early warning will be given to the occupant of the room, I would like to see some evidence the cooker extraction will not delay the smoke from activating the detector until such a point where the fire has developed to such an extent to compromise the escape from the non-compliant layout. 

[Phoenix]

Hi wainy,

John has given a good idea of the effect of the extractors.  To supplement this I stuck some figures into some calculations.  This isn't really evidence but it's indicative of what might occur.  I have put your room into a simple fill time spreadsheet that I use and assumed a fast fire growth rate that might occur in the case of a chip pan.  The results indicate that your 0.03 cu m/sec of extraction will have very little impact on the overall quantity of smoke throughout the period of fire.  Most significantly, for determining detection time, the extractor will have very little impact right from the outset.

For example, at 10 seconds after ignition, the rate of smoke production will be 0.25 cu m/sec.  The extractor will be removing approximately one eighth of this amount.  There will be plenty left in the room for the smoke detector to detect. 

The model shows that at 30 seconds the rate of smoke production will be 0.59 cu m/sec and the smoke will be .24m deep, at 60 seconds the smoke will be 0.68m deep and at 120 seconds the smoke will be nearly down to the pan.  This seems intuitively about right so the simulation seems reasonably realistic and can therefore be trusted to some extent.

I ran the model with no extraction, for comparison, and it made virtually no difference to the results, even in the early seconds after ignition.

Better evidence can be achieved through CFD but I would suggest that the zone modelling that I have done is sufficient.

[wainy1]

Hi Phoenix
Thank you very much for your response and taking the trouble to calculate.
My initial thought were the extractor would have some effect but would not significantly delay activation of the alarm, to the point where escape is compromised. Ultimately I thought I may have to resort to using FDS to evaluate the design, but my familiarity is a little rusty so it would have taken a considerable amount of time
Is the spreadsheet you refer to a commercially available product?
Once again thanks.
Wainy1

[Phoenix]

Wainy, I adapted the spreadsheet from one I did for another job.  I'll send you a link and some guidance.

[wee brian]

Is an extract fan going to have more impact than an open window on a windy day - I think not.

so long as you dont put detectors imediately next to grilles and fans than its all a bit accademic.

[wainy1]

Hi Wee Brian, Thank you for your reply. I think it will depend to some extent on the location of the detector relative to the fire and window and of course wind speed and direction. for example; if the fire was directly under an open window on the leaward side of the building with the detector situated at the oposite end of the compartment and an adequate supply of make up air I think there may be a significant delay in the activation of the alarm. Do you think this scenario is possible?