George,
The 5 systems I referred to were wet, side fed tree systems in a saw-tooth
roof design; not gridded systems. I have many many many more real world
examples where this problem exists or existed. Gridded systems have inherent
issues due to the inevitable trapped air pockets (i.e. the requirement for
175 psi PRVs); I will state unequivocally that the air pocket issue with
gridded systems is much more significant when the ITC is at the riser or
located somewhere on the system other than the most remote corner (i.e. fed
from the near main). All I am stating is that the practice of installing the
ITC at the riser is a poor practice, this approach should not be allowed by
NFPA and that many installers/designers/FPE's that I have met are completely
unaware of the inherent problems.
The most frustrating situation is with gridded systems on sloped ceilings
where the ITC is located at the riser or on the far end of the near main.
Sometimes the installing contractor uses the auxiliary drain on the far main
as the ITC, but they place the drain connection on the closest corner of the
far main rather than at the farthest corner of the far main (from the
riser). These systems (on sloped ceilings) have significant problems with
trapped air even when the ITC is located at the opposite corner of the
system.
Your comment: "I think separating the two items- air bleed-off and ITC- will
also enable you to not be stopped by those who like the ITC at the riser for
convenience, economy, because their riser manifold design has it there"
makes good sense. This statement highlights my frustration........the
reasoning behind the opposition is based on "convenience, economy, because
their riser manifold design has it there" and to maximize profit for the
installing contractors. NFPA 13 should never have been changed to allow this
approach because it is a poor design from and ENGINEERING standpoint.
The PRV's will resolve the excessive pressure issue, but will not resolve
the trapped air pockets, false alarms, flow switch cycling and/or rust
accumulation. Unless the high point air bleed-off connections at the high
points are automatic, this approach does not make much sense. Many or most
of the high point bleed-off connections will be located in inaccessible or
at least very difficult to access locations; therefore, manually operated
bleed-off valves are not feasible.
In most instances, when the ITC is located at the most remote point, a
significant percentage of the air is pulled through the system and
discharged during the filling activities or subsequent alarm tests. In
systems with significant elevation differences, it might be prudent to pipe
the ITC from the high point of the system rather than the actual most remote
point so that the trapped air pockets can be pulled through the system and
directly discharged through the ITC.
Jay Blocker emailed the following:
As far as the comment of "They simple bypass the alarm or set the delays so
high that flow switch cycling becomes a problem during alarm testing."
I don't have my book with me but if memory serves me correctly we have a
very high amount of time in which to receive the alarm. If you have found a
situation where the alarm initiating devices have been bypassed the location
of the ITC is a mood point and I would be contacting an AHJ.
The location of an ITC is pointless if the alarm can never be generated.
Again I am confused as to how an air pocket would trip the flow switch so
any insight into this would be helpful.
Jay
I would respond by stating the alarm technicians typically do not flow water
when they test the electronic water flow switches. The alarm company gets
called by the building owner to resolve false fire alarms. They usually
increase the delay to prevent false fire alarms. {BTW, NFPA requires the
electronic alarm switch to activate within 90 seconds of water flow through
one sprinkler.} When expansion/contraction of the trapped air pockets causes
the false alarms, this approach sometimes does not work (depending on the
fluctuations in the public water supply and the size of the trapped air
pocket) even when they set the delays for 60-75 seconds. When this approach
does prevent false alarms, the alarm technician and the building owner think
the problem has been properly resolved because the flow switch activates
just fine when the alarm technician manually holds down the flow switch
"arm" attached to the flapper.
With significant trapped air pockets, the flow switch will usually cycle
during water flow (due to the trapped air pockets as explained by Mike
Cabral in his email). Whether this happens during an alarm test or a true
fire emergency, the flow switch will constantly reset rather than activate
the fire alarm due to the 30-75 second delay setting. As a result, the
electronic water flow switch does not activate. PROBLEM!
IN MANY INSTANCES, THE SPRINKLER CONTRACTORS NEVER EVEN GET A CALL. The
alarm companies and building owners deal with these issues all of the time
and many sprinkler contractors are completely unaware.
These problems become more apparent during Spring, early Summer and Fall
when the temperature changes are more severe. Combine temperature change
with public water supply increases at night (to fill gravity tanks) and you
have inconsistent false alarms which are difficult to figure out!
Rodney K. Hamm, P.E.
President/Owner
Falcon Fire Protection
Office (478) 953-1677
Cell (478) 396-6988
Date: Sun, 1 Mar 2009 17:58:54 -0500
From: "George Church" <[email protected]>
Subject: RE: statistics
To: <[email protected]>
Message-ID: <[email protected]>
Content-Type: text/plain; charset="US-ASCII"
Well, if you want to meet someone with influence and pull with the
development of NFPA #13, simply grab a copy of it and turn to the last page,
and look in the mirror. You'll find a FORM FOR PROPOSALS ON NFPA TECHNICAL
COMMITTEE DOCUMENTS that enables you to influence the development of NFPA
#13.
You may wish to introduce the requirement for air bleed-offs at high points,
because otherwise you'll need to introduce a requirement for ITCs to be
piped off the highest point and in many cases, that's the ridge far away
from the exterior wall you'll want it piped to. I think separating the two
items- air blled-off and ITC- will also enable you to not be stopped by
those who like the ITC at the riser for convenience, economy, because their
riser manifold design has it there, whatever. While you may use it as an air
bleed-off, the ITC's function IS to test the alarm device. Hence the
suggestion to enable two items to address two different tasks.
I'm not a metallurgist, nor would many watch one on TV, so I won't address
the claims you've made. I believe there was discussion many moons ago on the
Forum WRT the NZ lack of MIC versus USA experience, and the difference in
ITC Location was offered up as a possible reason.
I understand PRVs will be required on all wet systems shortly, so one of the
problems is already being addressed; if those were gridded systems, as I'd
expect in a >1M SF facility, then the code already addressed the problem of
high pressure and you applied it- PRVs.
Rather than list all reasons in one proposal, I'd suggest breaking it into
twain and providing technical substantiation for each.
Congrats on going public with your complaints about ITCs, and I encourage
you to take the next step so your thoughts will be considered by the
committee.
glc
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Rodney Hamm
Sent: Sunday, March 01, 2009 5:38 PM
To: [email protected]
Subject: RE: statistics
If anyone has any pull or influence with the development of NFPA 13........
Please discuss the idiotic approach of installing Inspector's Test
Connections at the sprinkler riser (~12 in. downstream of the flow switch).
This has been considered "acceptable" by NFPA 13 since 2002 and is quickly
becoming standard practice in the great state of Georgia. There are several
significant reasons why this is a bad idea AND from an engineering
standpoint this is a very poor design. The inevitable trapped air in the
system causes or can potentially cause a number of problems: Excessive
system pressure when temperatures increase, false water flow alarms due to
air pocket expansion/contraction, failed water flow alarms due to cycling
flow switches, increased rust on the interior surfaces of sprinkler piping,
etc.
>From what I have seen, most contractors or sprinkler designers who
primarily
install or design new systems have no idea this approach causes problems for
the building owner and/or fire protection system inspectors. I can confirm
these are common problems at several of the facilities I have visited over
the past few years. In many instances, the only effective solution is to
install a suitably located Inspector's Test Connection or bleed the air
pockets using sprinklers at the highest elevations. We have installed air
bleed off connections for numerous systems simply to allow trapped air to be
removed from the high point. The problem is access at some facilities after
the solid ceilings or other obstructions have been introduced.
In my humble opinion, Inspector's Test Connections should be installed at
the most remote point of each system. The connection point (prior to being
piped to within 6 ft. of the floor level) should be at the highest elevation
of the sprinkler system piping whenever possible. Auxiliary drains or low
point drains should NOT be considered Inspector's Test Connections.
I would assume it is safe to say anyone who has enough experience and
personal dealings with inspections and/or problem solving (after the design
phase and initial installation project has been completed) realizes that the
Inspector's Test Connection serves more purposes than just testing the water
flow alarms. {I have heard "the sole purpose of the ITC is to test the alarm
device" so many times that I get a smile every time someone spouts it.} I
simply disagree with the relatively recent change in NFPA 13 which allows
this approach. I strongly believe the practice of installing the Inspector's
Test Connection a few feet downstream of the water flow switch should be
considered unacceptable.
In many instances, the fire alarm companies and building owners have to deal
with the issues caused by this poor design. They rarely if ever realize the
true source or reason for the intermittent false alarms. They simple bypass
the alarm or set the delays so high that flow switch cycling becomes a
problem during alarm testing.
We see systems with >175 psi on a regular basis caused by this
problem.......some systems even peg the 300 psi gauges. On of my customers
had 5 separate incidents where sprinklers developed leaks and/or popped open
due to excess system pressure during Spring & early Summer of 2007. We added
pressure relief valves in August of 2007 for all of their wet systems and
this resolved the problem. This is a >1 million sq. ft. facility with
several wet systems. False fire alarms and excess air pressure are very
common problems when the ITC is not located at the most remote and/or
highest elevation of the system piping. I could provide many more examples,
but I think I have made my point.
Black steel, moisture and trapped air allows more rust to develop on the
interior surfaces of the sprinkler piping than if the majority of the
sprinkler piping is full of water. Therefore, it would be safe to conclude
that trapped air in the sprinkler systems will shorten the life of the
sprinkler system piping.
The fact that someone or some group convinced the NFPA committee to change
the standard and the fact that the approach saves time and money during
installation does not make it a good engineering design!
There are probably a few instances where placing the ITC a few feet
downstream of the flow switch should be considered acceptable due to limited
or no feasible alternatives; however, these instances should be the last
resort alternative rather than standard practice.
Just one man's perspective!! I wonder if anyone on this forum agrees with me
on this subject.
Rodney K. Hamm, P.E.
President/Owner
Falcon Fire Protection
Office (478) 953-1677
Cell (478) 396-6988
------------------------------
Date: Sun, 1 Mar 2009 11:53:24 -0500
From: Mike Cabral <[email protected]>
Subject: RE: statistics
To: [email protected]
Message-ID: <20090301115324.sekkv.1683217.r...@mp20>
Content-Type: text/plain; charset=utf-8
A long time ago when I was a young man I did a lot of work on old systems in
industrial buildings in New England. . . old textile mills mostly. I wish I
had a few $ for every piece of 3/4" pipe I changed to 1" on these old pipe
schedule systems.
A high vent was almost always present with instructions on filling the
system posted on the riser. The location of the vent or vents was stated and
the order or the procedure for operating them was often posted. Some times
the instructions would advise you to open the vent while filling other times
the vent was opened after the system was under pressure.
If you followed the instructions you went home for the day and that was
that. If you didn't you could expect a call sometime through the night
because water would move as the air migrated through the system and trip the
alarm.
These systems were old in the late 1960's and early 1970's. . . willing to
bet most of them are still in service. Venting the air keeps trapped air out
of the mains and cross mains which would eliminate the air water interface
which if present invites corrosion.
Sometimes there is something to learn from accent history
Mike Cabral
---- George Church <[email protected]> wrote:
> Minimize false trips and air for MIC.
>
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]] On Behalf Of Forest
Wilson
> Sent: Sunday, March 01, 2009 10:43 AM
> To: [email protected]
> Subject: Re: statistics
>
> What advantage is there to venting the air? Other than response time?
> Sent from my Verizon Wireless BlackBerry
>
> -----Original Message-----
> From: "George Church" <[email protected]>
>
> Date: Sun, 1 Mar 2009 10:12:12
> To: <[email protected]>
> Subject: RE: statistics
>
>
> Nothing fancy, a 1" ball valve; or an ITC-type set up if you want to get
> fancy and run it out, we don't always.
>
> Potter has an auto air release you could also use.
>
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]] On Behalf Of Ergo
> Sent: Sunday, March 01, 2009 10:02 AM
> To: [email protected]
> Subject: Re: statistics
>
> George,
>
> You have a way to bleed the air off of the system at the high point? This
> sounds like a good idea. Cold you explain some more on how to make this
> happen?
>
> Thanks,
>
>
> Ergo
_______________________________________________
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)
