Roof exhaust fans and ESFR's

[WallyFP]

Looking for guidance on how to deal with powered exhaust fans located in warehouse roofs, commonly protected by supression mode heads. Can't find anything in NFPA (other than testing is based on them not being present). FM Global used to allow sprinkler heads to be installed in the center of fans up to 4.5-ft. diameters, but that option no longer exists (in FMDS 2-0). Thoughts on shutting down with general fire alarm (water flow, pull station, etc) - similar to what is done for HVLS fans? Anybody ever take airflow measurements near AS heads to see if velocities of 5-ft./sec. exist?

Thanks for your time.

 

[cdafd]

Shut the fans down upon water flow alarm

 

[WallyFP]

Thanks cdafd. That's my thought as well but I can't find any reference to doing this for exhaust fans (only HVLS) in FM or NFPA. Am I missing it?

 

[cdafd]

Cannot give you a reference, waiting for Scott to reply,

But anything on the sprinkler head data sheet??? Plus code cannot address every issue

 

[cdafd]

Do not have access to Nfpa till Tuesday but if you do, you might check

http://www.nfpa.org/aboutthecodes/AboutTheCodes.asp?DocNum=204&cookie_test=1

Nfpa 204m

 

[Sdpaddler50]

Wally, this topic was previously covered in FM 2-2, but it was placed under 2-0, along with some big changes a couple years ago. The main concern is the supply fans, pushing air at high velocities, across the heads, and delaying activation in the event of a fire. 5 ft/sec is the max you should have. I was on a lift last year, in a 40 ft ESFR warehouse, and measured the air flow velocity at 16 ft/sec, under a 6 ft diameter, powered, supply vent - thats pretty high. Tieing these to shut down upon waterflow, is not a good approach, because you already have a fire raging by then, and a key component of ESFR, is early activation. We therefore went with sheet metal barriers, the same size as the vent, hung below it, to diffuse the velocity to a lower value (below 5, when i went back up there to measure) - which was achieved. ESFR heads are then placed under the barrier, because you now have an obstruction to the heads above. Exahust vents are not as bad, and we were able to just put heads insdde the vent opening, with no barrier. 2-0 has details, and it can be a bit complicated, so you will have to read it about 3 x to understand it - at least i had to. Regards, Bill.

 

[Sdpaddler50]

One more, and not sure if you, or the client will like it. You can use high sensitivity detectors (the expensive ones)around the vent openings, or line detection (detectowire) within the racks below. Unlike a water flow alarm interlock, this type of detection is very rapid, so the fans would shut down very early in the fire growth stage. We did this in a 1M sq ft warehouse about 5 years ago, and it was not cheap, so sprinklers are probably your better option. 2-0 has more details if you are interested.

 

[WallyFP]

Thanks Bill! Yes, I have read 2-0 many times and it is still confusing to me in some areas. It talks about ridge vents and exhuast vents in the same section (2.2.1.7.2). It wasn't clear to me if that included powered exhaust vents as 2.2.1.8 also mentions "ceiling level air discharge vents". In 2-2, they used to limit fan size to 4.5-ft. diameter and said stick an ESFR head in the middle and be done with it. I like that concept.

Regarding the intake/supply fans, the 5-ft./sec. guideline, as I understand it, is measured at the sprinkler heads. If you have a 5-ft. fan and heads spaced 2.5-ft. away on either side, I wonder if the air flow is an issue at those heads. It seems that if you put a false ceiling as descrived in 2.2.1.8(a), then that would push the airflow horizontally towards the heads? The heads are 12 to 18 inches below the roof deck, how far down was the barrier in your case? Does the false ceiling/deflector impact the supply fan performance relative to the overall warehouse air flow.

Thanks again, I apprecaite the expertise. Data Sheet 8-9 sure did simplify things from a design standpoint .... but 2-0 had the opposite effect!

 

[Sdpaddler50]

Hi Wally, Ridge type are referring to a gravity type vent. You should be in the 2218 area which is geared toward motor driven fans. I agree, i used to love the 4.5' rule as well. Problem is, many bldgs are getting very large fans these days, so 6 ft and larger is gettng common, with high air V's. If you have a max 4.5 ft diam power vent, and you have a head dead in the middle of it, i would think you would be ok, but obviously i dont know the particulars, so i will leave the final decision to you.

Yes, the 5ft/sec is at the heads. That is where i took measurements, near the heads, when i was on the lift. I can tell you when i got hit with that 16 ft/sec, 40 ft up, its a bit unverving. Yes, the false ceiling will push it horizontally, out toward the heads. But, keep in mind, before it was going straight down, in one solid column, blasting the heads. By adding the barrier, the air flow is diffused out all four sides, so it decreases. i verified this when i went back up to measusre it after the barriers went in. I can calc water, but dont pls dont ask me about air flows so i will leave that to others who know more than me on the air flow subject.

The heads in this case were about 14 in down from the roof deck. The barriers are about 2'6" below the deck. This place gets hot in the summer, and i have not heard any complaints about interrupting the air flow to date.

 

[cdafd]

One more, and not sure if you, or the client will like it. You can use high sensitivity detectors (the expensive ones)around the vent openings, or line detection (detectowire) within the racks below. Unlike a water flow alarm interlock, this type of detection is very rapid, so the fans would shut down very early in the fire growth stage


Now that you said that did see that set up with detection at the exhaust, many moons ago, been awhile since did a good warehouse job

 

[stookeyfpe]

The IFC does not allow automatic operation. They must be manually activated for all the reasons discussed. The fans need to be controlled by the firefighters and only after fire suppression is achieved.

 

[cdafd]

Scott , it looks like they are talking about comfort ventilation fans, not smoke removal

Does IFC address comfort ventilation fans??? That might be running 24 hours a day

 

[stookeyfpe]

CDA

HVAC is different than smoke removal fans. These are two completely different systems.

 

[cdafd]

But I have seen a couple of times where a warehouse has had large ventalation fans stuck in the roof, for "ventalation comfort", which if running could act like smoke removal fans.

So my question is does any standard address these types of fans, if say they run 24 hrs a day and you have a fire ????

Have seen them shut down either on sprinkler flow or some type of detection

But does any standard require it? Or just common since to do it???

 

[LCREP]

CD,

NFPA 13, 2013 Ed coming out in a few weeks will address HVLS High Velocity Low Speed fans such as these

Link

video Link

They need to shut down via water flow or detection regardless if ESFR, area density, etc sprinkler design.

Here VESDA and Big Ass Fans partnered to provide a solution

Link

****************************************
Fire Sprinklers Save Firefighters’ Lives Too!

 

[Sdpaddler50]

LCREP, The HVLS (big ass is one mfg) are a different animal than i was referring to. I was referring to large fans within a shroud, that are at, or above the roof line. As others noted, these are HVAC related, not actual smoke removal fans that are turned on by manually by the FD from the fire control room when they arrive on scene. Fire testing has shown these to be an issue with ESFR for reasons previously noted. If a fire starts below one, or even nearby, the delay of the ESFR heads can be delayed. Tieing these in to the water flow alarm is therefore not quick enough, because the train has already left the station and the rack fire is past the incipient growth stage. That is why either barriers, and ESFR heads are needed to supplement this, or, HSSD detectors (the expensive flame detectors for example) are needed as I noted. Keep in mind, this is based on full scale fire testing by guys in white coats that have PHD after their names, not my personal opinions.

Regarding the HVLS, testing a few years ago showed these to not be that bif of a deal. Using a water flow switch or simple smoke detector to shut these down is fine. Either way, its best not to place these other high pile storage, but if you interlock as noted, it is ok. I have dealt with these alot, and most plants dont install them over storage anyway. They are normally in processing areas, to keep workers cool in the summer, and these areas normally have low combustible loading.

 

[WallyFP]

Thanks all for your input. Bill, I agree that the mechanical ventilation fans mounted in the roof (used for warehouse cooling), especially the intake/supply fans should be shut down prior to the sprinkler operation. FM 2-0 is the only standard that I'm aware of that provides guidance on this. The 5-ft./sec air flow rule at the sprinler heads can help in the evaluation. I see these fans in most all new large warehouses in the south. They don't always have supply fans, but most have exhaust fans in the roof deck. I would imagine that as we progress more towards suppression mode head designs, NFPA/IFC will likley address these as had to be done with HVLS fans. As a side note, I believe NFPA 13-2013 will recognize 6 head storage designs as FM 8-9 currently does. This makes the fast resopnse of the AS heads even more critical - thus the impact of ventilation will probably receive more attention. Just my thoughts. Hope you all have a good Labor Day!

 

[Duwe6]

Might want to ask the AHJ. I'll bet that the Fire Dept will want the exhaust fans still running, so they won't have to take time to cut vents in the roof[which is a fairly dangerous piece of work]. And, of course, all supply and circ fans to shit off at water flow.