[
Following previous research into the general effectiveness of residential sprinklers, BRE was commissioned to study the performance of concealed and recessed sprinklers. Kelvin Annable summarises the project and its conclusions.
The majority of fire casualties in the UK each year occur in residential and domestic accommodation. Residential sprinklers continue to be the subject of much debate in the UK. On the one hand, they offer a potential means of saving lives, preventing injuries and reducing property damage, but on the other hand they do represent a significant investment.
Where a sprinkler system has been installed in residential and domestic premises, concealed sprinklers are common and are chosen mainly for aesthetic reasons. However, their fire performance in the UK is unproven. Concern has been expressed about these sprinkler types as there are some potential problems with their use, especially for life safety applications.
British Standard BS 9251 (systems) and British Standard Draft for Development (DD 252 components) have been published but some aspects of assessing the performance of recessed and concealed sprinkler products are not addressed. The Office of the Deputy Prime Minister (now Communities and Local Government – CLG) therefore commissioned BRE to conduct research into these types of sprinklers following a previous project (see panel) on the effectiveness of residential sprinklers which concentrated on pendent type residential sprinklers.
The overall aim of this project was to investigate the suitability of concealed and recessed pattern sprinklers for use in residential premises, particularly concerning their effectiveness and maintainability. The specific objectives were to:
– carry out an experimental programme of stylised and realistic fires to assess the effectiveness of concealed and recessed pattern sprinklers in fire suppression, in particular for life safety in the room of fire origin
– develop and evaluate a robust, reliable and credible UK test method for rating the thermal sensitivity of these types of products, and
– review maintainability issues.
A steering group which included representatives from government regulators, the residential sprinkler industry, the housing industry and the fire service met twice during the course of the work programme. Members of the group provided general advice and review on all aspects of the project methodolgy and specific advice and information on residential sprinkler products.
Stylised fires
The previous study established a suitable benchmark fire test for UK conditions to support the further development of BS DD 252, and assessed the performance of pendent residential sprinklers in fire test conditions. Fire performance is one of a large number of tests that a residential sprinkler head is subjected to when being assessed against DD 252; other tests include horizontal and vertical water distribution, thermal sensitivity, and mechanical and ageing tests.
The DD 252 fire test essentially involves burning a stylised, representative fuel package of simulated furniture and wall and ceiling linings arranged inside a residential ‘room’ with door openings and an installed sprinkler system. The room contains two sprinkler heads of the same type spaced 4m apart, and a ‘dummy’ sprinkler located near the doorway at the far end of the test room, as the test requires that such ‘neighbouring’ heads should not be activated.
The sprinkler system performance is determined by its ability to control the fire for a period after sprinkler operation. This is defined by the gas temperatures being limited to specified values, and the dummy sprinkler head not operating. Residential sprinkler models (approved in the US) were reviewed and those considered most appropriate for further examination were selected for testing.
For the latest research, six residential sprinkler models – one pendent, one recessed and four concealed – were selected for the stylised fires. The concealed sprinklers comprised: flat and domed concealer plates, vented and non-vented recess cups, solder and glass bulb thermal elements and drop down and static deflectors.
A series of 12 stylised fires was conducted, based on DD 252 and BS 9251, and on findings of the previous study. The main conclusions of the stylised fires for the conditions studied were that the sprinklers examined met all the acceptance criteria of the DD 252 fire test, with two exceptions:
– The recessed sprinkler in the corner test at maximum recess distance exceeded one of the temperature criteria just after sprinkler operation but subsequently controlled the fire.
– One of the concealed sprinklers, at maximum recess distance with unfavourable frame orientation, did not control the fire. The test was prematurely terminated to protect the experimental facility. This failure was due to a combination of maximum recess distance and the alignment of the frame arms and concealer plate retainer lug which caused a detrimental effect on water distribution in the location of the fire.
Realistic fires
A series of ten fires, using real fuel arrays representative of domestic and residential premises, was conducted inside a compartmented experimental facility. It essentially comprised residential rooms with a timber structure supporting plasterboard walls, and a calcium silicate board ceiling. It was configured to provide alternative fire rooms at either end each measuring 4m by 4m, connected to a middle room measuring 3.8m by 4m by open internal doorways. The internal heights of the rooms were 2.5m and the effective enclosure volume was approximately 118 m3. Smoke alarms were located inside the room of fire origin and were replaced after every fire.
Two fire scenarios were conducted, with sprinklers and with the door of the room of fire origin open. Two types of lounge fire were investigated, each with nominally the same fuel arrangement but with different items used for the primary ignition source, as follows:
– ignition by nightlight under front corner of television (television fire)
– ignition by crumpled newspapers underneath a magazine rack under the table directly beneath sprinkler (table fire).
The television fires were relatively slow-growing, smoky, and relatively shielded from the sprinkler head. The table fires were rapidly-growing, flaming and shielded from the sprinkler, with the table placed directly beneath the sprinkler. One pendent and three concealed residential sprinkler models were examined.
Sprinklers installed to BS 9251 are not designed to extinguish fires but to control them. The effectiveness of the residential sprinklers was primarily assessed, particularly in the room of origin, by their ability to control toxicity, temperature effects and visibility effects during a period of 30 minutes from ignition. Fractional effective dose (FED) calculations were performed using the results of each test in order to assess the tenability, primarily in the room of fire origin and connected spaces. The calculations were based on carbon monoxide, carbon dioxide and oxygen concentrations (at 1.0m and 1.6m above the floor), gas temperatures (at various heights) and optical density per metre (at 1.0m and 1.6m above floor level).
The water flow rate of the single operating sprinkler was either 60 l/min, or the manufacturer’s recommended minimum if greater than 60 l/min (as in the previous study). Using results from the previous study for the conditions studied and for a duration of 30 minutes from ignition, the main general conclusions relating to the room of origin in the realistic fires are:
– Concealed and pendent sprinklers significantly reduced the effect of convected heat from the fire. However, these sprinklers did not observably improve visibility.
– Television fires: concealed and pendent sprinklers generally greatly improved conditions in the room of fire origin, maintained tenable conditions in terms of toxic effects, reduced the effects of convected heat, but produced no observed improvement in visibility.
– Table fires: for all the pendent and concealed sprinklered and unsprinklered fires, the conditions became un-survivable in terms of toxic effects. Concealed and pendent sprinklers greatly reduced the effects of convected heat but had no observed improvement in visibility.
– For all the unsprinklered fires in the previous study, the conditions became un-survivable/lethal.
– For all the unsprinklered fires in the previous study, the first tenability criteria to be reached was visibility, followed by convected heat, then toxicity effects.
– The life safety benefit of fitting smoke alarms was clearly demonstrated. Smoke alarms, fitted in the room of fire origin responded in 31% to 57% of the time required by sprinklers, and well before conditions had become life threatening. Smoke alarms, fitted in adjacent spaces with an open doorway, responded in 43% to 77% of the time required by sprinklers and well before conditions had become life threatening.
Thermal sensitivity
The purpose of this stage of the project was to develop a suitable thermal sensitivity test for concealed and recessed residential sprinklers. During the project, a new test was specified and developed, based on previous experimental work by Factory Mutual Research Corporation in the USA, using a modified mounting plate and frame with appropriate instrumentation. One pendent, one recessed and five concealed sprinkler models were used during wind tunnel testing.
In EN 12259-1, there are four categories of thermal response of sprinklers: quick, special, standard response A, and standard response B. BS 9251 and DD 252 specify ‘quick response’ in terms of thermal sensitivity rating for residential sprinklers. The main conclusions for the thermal sensitivity work for the conditions studied were as follows:
– A suitable thermal sensitivity test for evaluation of concealed sprinklers has been established for consideration in the development of residential sprinkler standards.
– Higher pressure differentials between the tunnel and the plenum box were found to be beneficial in terms of concealed sprinkler thermal sensitivity.
– The pendent sprinkler resulted in a thermal sensitivity rating of ‘quick response’ under favourable conditions.
– The recessed sprinkler resulted in a thermal sensitivity rating of ‘quick response’ under favourable conditions, and resulted in ‘special response’ in unfavourable conditions.
– None of the concealed sprinklers achieved a thermal sensitivity rating of ‘quick response’, even in the most favourable conditions. Although the concealed sprinkler thermal sensitive elements were ‘quick response’ rated, the operation time of the sprinklers, after operation of the concealer plate, was not comparable to the quick response pendent sprinkler tested. But based on the overall performance of the concealed sprinklers tested in the realistic fire tests, the concealed sprinklers were shown to produce results comparable to ‘quick response’ pendent sprinklers in terms of tenability conditions within the room of fire origin. On this basis, concealed sprinklers could be considered for use in some life safety applications. However, as with the design of any fire safety system, due consideration should be given to the likelihood of occurrence of potential fire types and taken into account in the design process. For example, ‘quick response’ pendent sprinklers may be beneficial for situations in which a slow growing or shielded fire hazard has been identified.
– The issue of whether the thermal sensitivity ratings determined in this study for concealed and recessed residential sprinklers are suitable for life safety applications requires further consideration.
Design, installation and maintenance
The purpose of this part of the project was to review maintenance issues with concealed and recessed pattern sprinklers. The main conclusions of the review were that there are potential problems relating to the design, installation and maintenance of concealed sprinklers, such as:
– The maximum recess distance, alignment of the concealer plate retainer lugs, frame arm/deflector attachment pins orientation, proximity to high level obstructions and a combination of these can be detrimental to the water distribution of the concealed sprinkler.
– Unfavourable frame arm orientation, maximum recess distance and unfavourable concealer plate position, and a combination of these, can delay the thermal response of concealed sprinklers.
– Water in the concealer plate, blocked vent holes, glued and painted concealer plates can delay the thermal response of concealed sprinklers.
These potential problems are avoidable, however, through awareness and improved installation practices.
Conclusion
In the experimental test programme, the concealed and recessed pattern products evaluated demonstrated comparable results to pendent sprinklers, in terms of their benefits to life safety in the room of fire origin. A test method for rating thermal sensitivity of concealed and recessed sprinkler products has been developed. Problems with the installation and maintainability of concealed sprinklers – which can affect their operational performance – were identified, though these are avoidable. Overall, the test work has also demonstrated that the use of properly installed and maintained sprinklers can potentially have a significant beneficial impact on life safety
and property protection in residential premises.
Kelvin Annable is senior consultant at BRE Fire & Security. The author would like to acknowledge the contribution of Dr Corinne Williams to this article.
PREVIOUS FINDINGS
The project on concealed and recessed sprinklers follows a previous study into the wider effectiveness of sprinklers in residential premises, also undertaken by BRE. The project consisted of a pilot study, a cost benefit analysis, benchmark performance tests and an experimental programme. The benchmark tests were carried out to support the development of DD251 (Sprinkler systems for residential and domestic occupancies), and DD252 (Components for residential sprinkler systems – specification and test methods).
The main findings of the project were:
– For the majority of experimental scenarios, residential sprinklers proved effective in potentially reducing casualties in the room of fire origin and connected spaces. But slow growing and shielded fires can be a problem for them.
– Residential sprinklers are probably cost-effective for residential care homes, and for tall blocks of flats (11 storeys plus).
– Residential sprinklers are not cost-effective for other dwellings and in order to become so, high risk buildings need to be targeted and assessed. In order to be cost-effective for a broader range of dwellings, installations and maintenance costs must be minimal and/or trade-offs may be needed to reduce costs indirectly.
The report concluded that further work was needed on concealed sprinklers and on the cost-benefits of sprinklers as a compensatory feature.
For further information on both projects visit: www.communities.gov.uk