The simplest, and 90% sure fix is to leave the ITV open when filling the system, and filling slowly. Works only if the ITV is remote from the riser, and the higher the better.
Ed Vining On Wed, Mar 4, 2009 at 7:08 AM, Chris Cahill <[email protected]>wrote: > Air is not coming through the city mains. It's trapped when filling the > system. The air in the system is expanding and contracting with > temperature > and pressure changes. Not unlike a bladder tank on your well or an > anti-freeze system. > > You are underestimating the size of the air pocket. It's not one air > pocket > it's many like at the end of each branchline. We have a very large system > 3-4,000 gallons, can't remember exactly. ESFR with 3" bl and 6" mains, not > gridded. We needed to bleed all the air out for other reasons. We filled > it to 45 psi. 700 gallons of air remained. We knew this because we knew > how big the system was and had the exact amount of liquid available to fill > with no air. At 45psi we still had 700 gallons of liquid remaining. As we > bled the air all 700 gallons went into the system. Granted not a typical > system or scenario. > > Another example of proof air pockets are a false alarm problem was some > research I did into all the times the FD was called due to sprinkler > alarms. > Could be fire, could be breaks, could be unknown. I had a lot of reports > of > dispatch for water flow but nothing found on wet systems. Chiefs were > complaining about some buildings in particular. Some buildings would be > fine for years then all of a sudden we'd get alarms over the course of a > few > days or weeks. Then they would subside. I stumbled on a pattern being the > guy that issued sprinkler work permits. Many, many of the buildings with > "nothing found" reports had work done on the sprinkler system recently. > Air > was introduced into the system through the work and as city pressure > fluctuated it further compressed the air and caused false alarms. Setting > long delays helped but weren't always a cure. > > Another issue here was the Chief's reported they knew it would be a false > alarm if the street sweepers were out. They filled the water truck with > the > hydrants. This was an overnight operation usually. Filling the trucks > would cause pressure surging in the water system and thus the sprinkler > systems near by. Even had a couple Chiefs predict the time and date of the > next false alarm. Within ten minutes a week in advance is good in my book. > > > The buildings that were more problematic we'd have them change the check > valve. That cured many of the most common buildings. These valves were > leaking and never could lock in the surge pressure. > > One more anecdote. I'm out for the flow and tamper inspection on a new > building. I note before the AHJ gets there the water meter spins every > couple minutes and pressures rise with each surge. I set the delay to 45 > seconds. AHJ shows up and we do the test. I'm in the valve room he opens > the remote ITV. Alarm trips over 1:00. He wants to do it again because he > only accepts under 1:00. We argue about the code, the importance of a few > seconds from a sprinkler extinguished fire vs. the inconvenience and risk > to > the FD and the false alarm problem and show him the fluctuating gauge. See > I knew it took 55 seconds for the flow switch to even change position. > Again this was a big system gridded with lots of air. The GC calms me down > and says give him what he wants they'd fix it later after he leaves. Well > we set the delay for about 0. We come in around 58 seconds. AHJ passes. > I > tell the general it's going to false alarm (it's Friday at 3pm already) and > don't call us this weekend. Well about 20 minutes later we're BS'ing > outside with the AHJ and as sweet as can be the sprinkler alarm occurs. FD > is dispatched before anyone could intercept. Yes, I tell the AHJ I told > you > so and head home with a big smile. Happy hour has many connotations. Sure > enough 3 more alarms occurred over the weekend. Then there were no > problems > for 2 years until another contractor was in there doing TI work. > > I agree it sounds ridiculous with long flow switch delays and 4-10 gpm > required that air pockets and surging pressure would be a problem but it > was/is. Remember you as a sprinkler company may not see the full picture > because you are not called every time there is a false alarm. But talk to > the FD and they know the false alarms just don't know why or it can be > fixed. > > > Chris Cahill, P.E. > Fire Protection Engineer > Sentry Fire Protection, Inc. > > 763-658-4483 > 763-658-4921 fax > > Email: [email protected] > > Mail: P.O. Box 69 > Waverly, MN 55390 > > Location: 4439 Hwy 12 SW > Waverly, MN 55390 > -----Original Message----- > From: [email protected] > [mailto:[email protected]] On Behalf Of Jay Jay > Blocker > Sent: Tuesday, March 03, 2009 7:10 PM > To: [email protected] > Subject: Re: statistics > > Rodney i do not believe you are a nut im just not sure we are looking at > the > same equipment. Maybe no one else is admitting this is a problem because > it > may not be. I say that to say this. > Lets look at basic common science here. If you have a flow switch set at > 76 > seconds and an air pocket tripped it that would suggest that there was > enough air in one single pocket to not only move the paddle but hold it in > the activated position. THis would mean the velocity of the air would have > to have enough force to over power the spring and hold it for 76 seconds. > None moving air could not perform this action. So if there is a pocket of > air that can do this i do not believe moving the ITC would help. I would > have to work with the water dept. and find a way to prevent these air > pockets from entering the system. Maybe even check with businesses > near bye > the affected buildings and see if there domestic water has had any > problems. I dont see how air pockets would discriminate between systems > (honestly seems like the domestic would be more susseptible to this > problem). Im not saying your wrong i am just really lost as to how this > could happen. Have you tied changing out the flow switch? > > Jay > > > > > ________________________________ > From: Rodney Hamm <[email protected]> > To: [email protected] > Sent: Tuesday, March 3, 2009 3:26:47 PM > Subject: Re: statistics > > Jay, > > It is really not a "theory" at this point. I have found this problem at so > many locations and resolved the issue that I am absolutely convinced the > large trapped air pockets are causing the false alarms. > > I have placed my digital LPT Data Logger on the downstream side of wet > systems and recorded the system pressure on 30 second intervals for 7 day > periods. The graph shows the gradual pressure increases/decreases caused by > temperature changes and the exact time of the false fire alarms are always > indicated by a distinct pressure increase which lasts for more than 60 > seconds. This happens when the water supply increases at the same time as a > cold front (or significant drop in temperature). Public water supply > systems > typically fill gravity tanks at night and temperature changes usually drop > at night during Spring and Fall (this is typically the time of year these > false alarms are most common). I use the weather channel to determine the > greatest temperature drop on the night when the false fire alarm occurred. > > After I bleed as much air from the system as possible, the LPT Data Logger > verifies that the pressure fluctuations are much less severe and the false > fire alarms are a thing of the past. > > I have one customer where this problem occurred when there was a > temperature > drop and the jockey pump filled the system EVEN WITH THE DELAY ON THE FLOW > SWITCH SET AT 76 SECONDS! After we bled the air from the system, the > problem > was resolved. LPT Data Logger was also used on this system. Yes that is > correct same problem even with a fire pump......the jockey pump at this > facility typically requires 20-25 seconds to start/stop, but when the > underground main pressure is just right and the air pocket/temperature > issue > combines.......we had false fire alarms. I checked the electronic water > flow > switch 7-8 times and the delay was always between 70-80 seconds! > > In summary, air will expand and contract with temperature changes while > water will not. All system have some air, but systems with a significant > percentage/volume of air will allow water to flow into the system to > equalize the system pressure when the pressure downstream of the alarm > check > valve drops below the upstream pressure. When the time required to > replenish > the system exceeds the flow switch delay.....the fire alarm will activate. > It is even more common for systems which have pressure switches on the > retard chambers. > > The flow switch cycling issue is also a huge problem in some instances! > Refer to my previous post regarding the typical test method used by alarm > technicians. > > I sincerely hope this helps. I imagine some of the people on this forum > think I am a freak'n nut. I'm really not crazy! I will admit that I seem a > little obsessed with this issue, but I am getting tired of finding the > problem and none of my buddies in the Fire Sprinkler Industry want to admit > this is an issue. It has become a bit of a pet-peeve of mine I suppose. > > Take care, > > Rodney K. Hamm, P.E. > President/Owner > Falcon Fire Protection > Office (478) 953-1677 > Cell (478) 396-6988 > > -----Original Message----- > Date: Mon, 2 Mar 2009 17:07:39 -0800 (PST) > From: Jay Jay Blocker <[email protected]> > Subject: Re: statistics > To: [email protected] > Message-ID: <[email protected]> > Content-Type: text/plain; charset=iso-8859-1 > > Rodney could you please elaborate on your theory of the air pocket tripping > the flow switch. Several of us in my office are baffled. > > Jay > > > _______________________________________________ > Sprinklerforum mailing list > http://lists.firesprinkler.org/mailman/listinfo/sprinklerforum > For Technical Assistance, send an email to: [email protected] > > To Unsubscribe, send an email > to:[email protected]<to%[email protected]> > (Put the word unsubscribe in the subject field) > > > > > _______________________________________________ > Sprinklerforum mailing list > http://lists.firesprinkler.org/mailman/listinfo/sprinklerforum > For Technical Assistance, send an email to: [email protected] > > To Unsubscribe, send an email > to:[email protected]<to%[email protected]> > (Put the word unsubscribe in the subject field) > > _______________________________________________ > Sprinklerforum mailing list > http://lists.firesprinkler.org/mailman/listinfo/sprinklerforum > For Technical Assistance, send an email to: [email protected] > > To Unsubscribe, send an email > to:[email protected]<to%[email protected]> > (Put the word unsubscribe in the subject field) > -- Ed Vining 4819 John Muir Rd Martinez CA 94553 925-228-8792 _______________________________________________ Sprinklerforum mailing list http://lists.firesprinkler.org/mailman/listinfo/sprinklerforum For Technical Assistance, send an email to: [email protected] To Unsubscribe, send an email to:[email protected] (Put the word unsubscribe in the subject field)
