This is an interesting topic to be discussing today and I'm sorry I'm late to input.
Our Yankee Chapter meeting today was a tour of FM Global's RI Test Facility. We saw a 25' palletized rack (plastic cups in cardboard boxes) fire using control mode K=11.2 heads (stressed NOT per FM Guidelines) @ 50 PSI and watched the ring of heads go from 4 to 9 to 13 and on... Not exactly on this topic or thread but similarly interesting. You guys gotta see this if you come to Boston in June. I hope it's possible. Tom Duross GRS Bill, The number of rings that operates is a function of the occupancy. In a light hazard occupancy, a fire will not operate sprinklers byond the second ring unless something goes wrong. In more challenging occcupancies, you get into the third, or even fourth wring, even if nothing goes wrong. The curves for claa I-IV are a very special case. They were developed from testing in the late 60's and early 70's, when most existing sprinkler system were simple pipe schedule designs. It was deemed necessary to provide curves to allow as many existing systems to be evaluated as possible. A curve was developed from testing of class II commodity at a variety of densities. As would be expected, the number of sprinklers that operated went down as the density went up. No safety factor was applied to the curve. Each test was plotted as an unadjusted point. Two tests of class III commodities were performed, and a curve parallel to the class II curve was plotted through those two points. Two test were performed on a class IV commodity, and the curve was plotted the same way. The class I curve was drawn without any tests. It's not pretty, but there you have it! You will note that no protection requirements based on actual testing subsequent to the class I-V tests have involved curves. The reason that designs based on the curves have worked so well is that they were based on tests using K5.6 sprinklers and the design requirements were specified as densities and areas. As we have subsequently learned, discharge density is not a fundamentel measure of sprinkler performance. A 0.30 density is significantly more effective from a K8.0 sprinkler than from a K5.6, and far more effective from a K11.2 sprinkler. Economics have made the larger K-factors the normal choice. That's the safety factor we have now. Joe Brooks, Bill wrote: >I laid out a 11.4 x 11.4 pattern (130 sq ft) and drew three rings. 1st ring >included 4 sprinklers (520 sq ft). 2nd ring included 12 sprinklers (1520 sq >ft). 3rd ring included 24 sprinklers (3120 sq ft). The numbers are >comparable to NFPA ~1500 sq ft and the DoD ~ 3000 sq ft. > >The ring theory makes sense in this context. NFPA is looking for a two ring >design and DoD is looking for a three ring design? Was this purposeful? If >so it provides the basis for the 1500 sq ft minimum design area. > >Using this approach, the 900 sq ft design looks like it has a very small >margin of error. > >If we have this two ring NFPA design, what led to the development of the >area/density curves where density is decreased as design area is increased? > >Bill Brooks >Pittsburgh, PA > > >Bill, > >You are correct. Sometimes politics prevails. > >Joe > >Brooks, Bill wrote: > > > >>Joe, >> >>I thought I understood your train of thought until I started considering >> >> >the > > >>area reduction for quick response sprinklers. How many rings are there >> >> >with > > >>a 10 ft ceiling height QR design? 900 sq ft or 5 heads. >> >>Can I conclude the ring theory falls apart when this design approach is >>used? It looks like you don't have your 1 ring cushion in this case. >> >>Bill Brooks >>Pittsburgh, PA >> >> >>The idea that there is a huge difference between a 5-sprinkler fire and >>a 15- or 30-sprinkler fire seems logical on the surface, but if you know >>how sprinklers operate, you know they operate in rings. The heat that >>operates sprinklers spreads in a circle from the fire origin. For >>symetrically-spaced sprinkler system where the fire starts centered >>among 4 sprinklers, there are 4 sprinklers in the first ring around >>ignition, so there is virtually no difference between a 1-sprinkler >>fire and a 4-sprinkler fire. If those sprinklers don't control the fire, >>then any (and all) of the 12 sprinklers in the next ring can operate, >>meaning that there is very little difference betwen a 5-sprinkler fire >>and a 16 sprinkler fire. If you get into the next ring, then the number >>is 36 sprinklers. So, the 1-ring "safety factor" between 4 and 16 >>sprinklers is the same as the 1-ring safety factor between 16 and 36 >>sprinklers. A 1-ring "safety factor" is not overkill in my book! >> _______________________________________________ Sprinklerforum mailing list [email protected] http://lists.firesprinkler.org/mailman/listinfo/sprinklerforum
