In the absence of available guidance on how local override switches should operate with a BRE smoke shaft, Howard Morgan and Paul Harris consider how they should be designed.
Smoke shafts based on the building research establishment (bre)
study of fire-fighting shafts* can be used to reduce the pressure in a corridor or lobby, by taking advantage of wind suction at the roof (the shafts must rise through the building to the roof for this to be a factor) as well as of any buoyancy in the smoke (see figure 1).
The space open to the shaft – in most cases a corridor or lobby connected to the shaft usually by an Automatically Opening Vent triggered by smoke detection – is not itself protected from the ingress of smoke from an adjoining fire room, or from smoke generated by a fire in the ventilated space itself. The space immediately adjacent to the ventilated lobby or corridor (usually a staircase forming a part of the means of escape from the building) is protected as a result of the relatively lower pressure in the lobby or corridor.
Section 7 of the BRE Project Report 79204 suggests that a manual override should be provided to the Automatically Opening Vents (AOVs), and that an override to open and close the stairwell ceiling ventilation also be provided. There is no guidance in the public domain on the operation of the manual override for the AOVs (e.g. whether these should be ‘open only’, ‘shut only’, or ‘open and shut’) and no guidance on the exact position of the override for the stair vent. It is known that a number of approval bodies have been developing their own recommendations, although these are not readily available in the public domain.
Recently, IFC has considered the question of how local manual over-ride switches can be compatible with the operation of a BRE smoke shaft system. The AOVs at the head of the stairs, and at the head of the smoke shaft, both need to open when smoke is detected (or when a break-glass switch is operated) in any corridor or lobby served by that smoke shaft in order to protect those stairs. The AOV at the head of the stair should, in addition, open on detection of smoke in the stairwell. The corridor/shaft vent between the corridor or lobby and the smoke shaft must only open at the fire storey i.e. usually the corridor/lobby first affected. Opening other flow paths into the shaft on other storeys will reduce the pressure drop in the smoke-affected corridor, so increasing the chance of smoke entering the stairwell. It follows that detection of smoke in a corridor or lobby should not only open the AOV to the shaft on that floor, but should also automatically lock-out the opening of any other shaft vents on other storeys. This implies that the local manual switch should not have local control, but should only send a signal to the control panel.
Both the head of stair and head of shaft AOVs can safely fail open and the BRE system would still work, so it would be acceptable to have a local override switch to open either vent. In the event of fire, neither of them should close, because this would negate the functioning of the system – it’s better to leave closure to a central operation from the control panel. This implies that only an ‘open’ switch should be considered for these two vents.
If the corridor/shaft vent has an ‘open’ switch locally, it will be useful to trigger initial operation on the fire storey, but only as a part of the entire system coming into action. An AOV open from a corridor to the shaft – without either vents at the head of shaft or the head of stairs operating as well – can achieve nothing useful in terms of smoke control. A local break-glass switch opening the nearby vent between shaft and corridor/lobby will be useful in the event of the smoke detector not operating, by sending its signal to the control panel which in turn initiates the full pattern of open and closed vents needed.
In summary, there does not appear any situation where giving local control to AOVs independently of the control panel is not potentially harmful to the proper functioning of the system.
Where there is key-operated control of all vents available on the control panel for use by people familiar with the building, we conclude that it is safer not to have local manual control other than via the break-glass option signalling to the control panel.
This has led us to a more general conclusion, namely that it will often be safer for the only override switches to be on the control panel, and for them to be operated by people familiar with the building.
H.P.Morgan BSc, PhD, C Phys, MInstP, CEng, FIFireE, MBEng and P N Harris BSc, M B Eng, MIFireE, International Fire Consultants Ltd
* BRE Project Report 79204 Smoke Shafts protecting Fire-Fighting Shafts; their performance and design BRE 2002 (PR79204) Harrison and Miles].