The Swedish Accident Investigation Board recently published its report RO 2006:03, describing investigation of the fire at the Borgholm Hotel on 9th April 2004. It refers to the recommendation that the National Board of Housing, Building and Planning should press for the introduction of requirements for fire gastight doors to hotel rooms. Previous issues of BrandPosten have discussed and pointed out the misunderstanding that fire-resistant structures are not necessarily smoketight. This is worth repeating. It is very probable that smoketight doors should be required in many buildings: not just hotels, but also such as parts of health care facilities, old people’s homes, prisons etc. Methods for testing and classification exist.
Fire doors are normally classified in respect of their resistance to fire, with class E and class EI ratings. Certain additions can be applied to these ratings, such as C if the door has an automatic closer. The new European classification system now also includes a special classification for smoketightness of doors, indicated by S. Two different classes are available: Sa for smoketightness at room temperature, and Sm for smoketightness at 200 °C. This means that there are now two types of doors for ‘fire separation’: one which is fire-separating, and one which is smoke-separating. In certain cases, a particular design of door may possess both properties, but not always.
Testing smoketightness performance uses a particular type of furnace, in which both temperature and pressure can be controlled. As both the temperature and the rate of temperature rise are considerably lower than in conventional fire-testing furnaces, it is possible to use a simpler design of furnace with electric heating. The leakage is measured using a flow meter that records the quantity of air that is necessary in order to maintain a particular pressure in the furnace. In order to ensure that the measurement provides a correct value, the air flow against an airtight wall is also measured. This provides a means of compensating for any leakage in the furnace itself.
The door to be tested is fitted in a wall which is in turn fitted in a steel frame that can be secured against the furnace. In principle, the door must be fitted in the same type of wall as that in which it is intended to be fitted in practice, i.e. a stud wall clad with gypsum board, or a rigid wall of (for example) concrete. Before performing the actual smoketightness test, the door must be conditioned by opening and closing it a number of times: 25 cycles for most doors. This conditioning is described in a special standard, EN 14 600.
After preconditioning and measuring the closing forces of doors fitted with closers, the entire frame with the wall and door is fitted to the furnace. For testing at room temperature, it is only the pressure in the furnace and the air flow into or out of it that are recorded. If the door is also to be tested at 200 °C, the temperature is increased linearly to this value over a period of 30 minutes, after which the leakage flow is measured with positive pressures in the furnace of 10, 25 and 50 Pa. These measurements must be made within a period of ten minutes after reaching the temperature of 200 °C.
Testing for class Sa – i.e. at room temperature – leakage beneath the door is disregarded. This means that any gaps between the door and the threshold are sealed. The leakage is presented as the flow divided by the periphery of the door, excluding the length of the threshold. Testing is carried out in both directions, with pressure differences of 10 Pa and 25 Pa across the door. The two directions can be tested on the same occasion, by maintaining a positive pressure in the furnace for one direction, and a negative pressure in the furnace for the other direction. Classification standard EN 13501-2 specifies a maximum permissible flow of 3 m³/h per metre of periphery.
If the door is to be tested for class Sm, it must first undergo the class Sa test, after which two tests are carried out at 200 °C, one with the door mounted so that it opens inwards, and one with it mounted so that it opens outwards the furnace. Positive furnace pressures of 10, 25 and 50 Pa are used for this test. The leakage results in this case are presented as total leakage, i.e. as the flow past the door. The requirement here is that the flow past the door does not exceed 20 m³/h for single doors, or 30 m³/h for double doors.