The client, owner of the flat in question, had removed a 2m wide traditional brick chimney breast to gain space in the small rented flat. A structural steel ‘box’ frame had been installed to provide load transfer from chimneys above the flat and this steelwork had been clad with fire protecting board. Several uninsulated stainless steel flues had been used to connect traditional brick flues in flats below and above, and the floor and ceiling had been properly fire stopped around the new flues. The flues were designed to be natural air ventilators and were not designed to function as fire flues. So far so good. This construction had been boarded over with 2 layers of fire protecting board supported by hidden timber battens of small unknown size – I refer to this as the ‘partition’ here. The partition had then been furnished with a timber bookcase over the whole area of the wall. I believe the builder had not considered the risk of fire getting into the flue.
The building control (BC) authority objected to the partition because it did not possess one hour fire resistance normally required where services penetrate a compartment floor even though the flue was intended to be used for air ventilation. BC suggested a partition of fire board either side of light steel/timber studs but the client resented the loss of space taken up by a new thicker partition. It was at this stage that I got involved. I did not agree with the need for one hour separation; however I had independently realised there was a small fire risk present – if a fire in a flat below become developed, i.e. reached flashover, fire effluent could enter the thin steel flue and radiate at a temperature of say 800 degC back into the flat above and this could cause charring of the nearest small timber battens causing a local weakness and possible flaming in the cavity, and if the fire was prolonged (unlikely owing to likely early fire service intervention) then the partition might collapse.I put forward several remedial options and these were considered by the client and rejected for practical or cost reasons.The options included the following:
- adding small RHS steel sections on the outside (room side) of the existing fire board and timber construction fixed direct to the steel I-section box frame at top and bottom which support the chimney breast for upper storeys such that failure of the timber battens can occur without collapse of the fire board,
- inserting intumescent dampers at the foot of each flue i.e. at floor level of my client’s flat – a difficult job and disruptive
- filling the cavity with pelleted rock wool from floor to ceiling so that the timber battens would be partly insulated from a hot flue
- filling the cavity with aerated or lightweight concrete from floor to ceiling – the client’s idea
- doing nothing on the assumption that the existing risk would be acceptable if a fire engineering calculation was done which showed that the extent of timber charring would not lead to collapse of the fire board lining. Such a calculation would use the principles of BS 7974 Fire Safety Engineering codes but the cost of doing the calculations and the uncertainty of acceptance by BC was a limiting factor. Besides, assumptions would need to be made on the basis of poor information as to the relative positions of battens and flues; and the radiation configuration factors are complicated (the author has made such calculations in his work for International Standards Organisation TC 92 SC4 Fire Safety Engineering committee of which he is a member – see publications on www.cookeonfire.com website).
The preferred option was to remove the existing partition of fire board and timber battens and replace this with a proprietary ‘Promat 35mm Solid Partition’ which involved the use of a 20mm and 15mm layer of Superlux board fixed together at 300mm spacings and to the perimeter steel I-sections at 500mm centres. A 30 by 30 steel perimeter angle (top, bottom and sides) would be fixed direct to the face of the steel I-section members with steel fastenings. This would provide well in excess of the small amount of fire resistance needed – small because at any point in time only one flue would radiate and the steel perimeter angle could easily cope with local thermal radiation owing to the heat sink effect of the large steel I-section to which it was fixed. BC still insisted that one hr fire resistance was needed from this partition. I did not agree, but then discovered that the same Promat construction could also provide 60 min fire resistance, and this was accepted by BC.
This was a classic example of a small job in which the flue fire aspect had been overlooked by the contractor and to rectify involved considerable effort.
Client: Private leaseholder living outside London
Share