EFFECT OF ENCLOSURE ON HEAT RELEASE RATE IN FURNITURE FIRES

摘要

The heat release rate (HRR) in large scale room fire tests is affected by the enclosure. Thernpotential effects include radiative feedback from the hot gas layer, the heated walls and ceiling, andrnobstruction of sufficient supply of fresh air into the test room. In very large spaces the effect isrninsignificant, but in practice enclosures are usually small enough so that the increase of the HRR isrnnot negligible. This paper describes a review of experimental studies that measured the effect forrnupholstered furniture. A summary of the review is presented in the next few paragraphs.rnThe first pertinent study was conducted in the early 1980s and involved two furniture items tested in arnfurniture calorimeter and in a room1. The results were somewhat inconclusive. The next study,rninvolving mattresses and other furnishings, suggested that the room caused a 15-30% increase in peakrnHRR2. A later study found that room size (within a relatively small range) has little effect on heatrndevelopment if the HRR is below 600 kW~3. The work involved ten sets of chairs, some of which wererntested in the standard 2.4 × 3.6 × 2.4-m room using a gas burner, while some were tested in a furniturerncalorimeter (with the same burner or with an equivalent paper ignition source) and some others wererntested in a room of a different size, with both ignition sources. The study involved few actualrncomparisons between room and furniture calorimeter. The lack of effect of the room size was clearrnbut the enclosure effects were less clear, in spite of the published conclusions.rnAnother key study conducted in the first half of the 1990s looked at peak HRRs in a standard roomrnand in a furniture calorimeter for 27 items with peak HRRs ranging up to 2363 kW. On average thernpeak HRR in the room calorimeter seemed to be approximately 25% higher than in the furniturerncalorimeter. This is based on a linear regression analysis of the data after removing fourrnapparent outliers. The linear fit has an R~2 of 0.98 and a standard error of 159 kW.rnAn analysis of the aforementioned studies published in 1995 found significant effects atrnlower HRRs and proposed a threshold below which enclosure effects are negligible of 460rnkW~4. As a result, the perceived wisdom since the mid-1990s was that room walls have an effectrn(increasing measured heat release) but only if the HRR exceeds 460 kW.rnA recent study conducted for the US National Institute of Justice (NIJ) investigated models tornestimate the burning rate of upholstered furniture. As part of that study, a detailed analysis was madernof literature results. The analysis shows that the effect of compartment walls can be found at HRRs asrnlow as 200 kW and that the effects increase with HRR. However, even this would mean that typicalrnregulatory (or code) requirements would be unaffected, since most pass/fail criteria are in the range ofrn80-200 kW.rnThe NIJ study also included a few tests both in a compartment and in a furniture calorimeter.rnUnfortunately all those tests involved relatively low peak HRRs and no notable compartment effectsrnwere seen. Thus, the conclusions of the analysis of the literature studies remain valid, to a largernextent.