The standard is still silent on grouped or clustered (is my preferred phrase 
due to its connotation) obstructions.  That is changing for ESFR. The 2016 
edition has finally plagiarized that part of the FM criteria too.  It will say 
that items close together will be considered one obstruction unless the 
separation is at least 3 X the width of the adjacent item.  NADA on spray 
sprinklers so still silent. This is not a cluster question though the overall 
philosophy of Ron’s summary is still a driving factor.

The only criteria that exists (and it’s only if the obstruction is less than 
18” below the deflector).  As discussed in A.8.6.5.2.1.4, if an obstruction is 
so close (even if just a single light fixture) that you can’t spray on both 
sides of the obstructions, then treat it light a continuous obstruction.  IT 
also state that fixtures and architectural features to conceal the sprinkler 
obstructing the discharge is a no no.  

We (and the AHJs that review these plans) have to keep in mind that the 
objective is to control the fire (aka avoiding flash-over is the name of the 
game). I’m focusing on the occupancy hazard approach.  Storage is a separate 
beast that warrants addressing penetration by the discharge but this hallway 
really isn’t storage. 

Roland


Roland Huggins, PE - VP Engineering
American Fire Sprinkler Assn.       ---      Fire Sprinklers Saves Lives
Dallas, TX
http://www.firesprinkler.org <http://www.firesprinkler.org/>

NFPA 13 member caveat - this is not a formal interpretation for NFPA or any of 
its technical committees.



> On Aug 21, 2015, at 8:54 AM, rongreenman . <[email protected]> wrote:
> 
> This all sounds like engineering overthink to me. It's an obstruction less
> than 48" in dimension in at least one direction of the horizontal plane(s)
> it describes. If the horizontal plane of the deflectors are the required
> distance above the upper horizontal plane of the obstruction (be it duct,
> catwalk, cantilever, cloud, soffitt, etc.) then sprinklers are not required
> below per the prescriptive code that applies in this case.
> 
> I once asked AFSA for an informal interpretation of the following scenario:
> A large room with 47" by the length of the room ductwork running in
> parallel, each 1" apart, that essentially covered the entire room. This
> meant an inch from the wall to a 47" wide duct, and then another inch to
> another 47" wide duct and so on all across the room. The rest of the
> scenario met deflector distances, et. al. per 13. The conclusion, penned by
> either Roland or Phil Brown, was that this was perfectly acceptable under
> the rules of 13 and did not require sprinklers below the ductwork. Now this
> is a very unlikely scenario (I'd say less the a fraction of 1% chance in my
> made up statistics toy box), nor do I think it would be good practice if
> someone was to actually build something to the scenario,but the point is no
> sprinklers required below.
> 
> And then when we go to a hallway and storage in a hallway. My presumption
> is the lockers have doors and locks (hence storage locker) because people
> storing their goods there do not want other people taking said goods (I
> tend to put no longer wanted stuff out on the road with a sign that says:
> "Free. Or Best Offer." and it magically disappears, sometimes along with
> the sign, and that is irksome), so they're unlikely to leave their things
> in the hallway.
> 
> And it's five feet wide. If a head goes off I'm pretty confident there's
> going to be more water than the average person wants and the 36 inches (I
> believe that was the width of the obstruction–the length is irrelevant per
> the rule) of furiously burning methylethyldeath stored down the center of
> the corridor would still be contained within the parameters of a design
> area (1500 sqft at three feet wide would be a corridor 500 feet long or 300
> feet at five feet wide.

_______________________________________________
Sprinklerforum mailing list
[email protected]
http://lists.firesprinkler.org/listinfo.cgi/sprinklerforum-firesprinkler.org

Reply via email to