2010 requirement for Relief valve on all Wet systems 3

[Cidona]

The 2010 edition of NFPA 13 has section 7.1.2.1, which states 'Relief valves will be required on all wet pipe systems that do not have auxilary air reservoirs. Previously, they were only required on gridded systems. The size of the relief valve was increased to 1/2" and the relief valve now needs to be listed'.
I thought I heard that the reason for the relief valve on all wet systems was because there was something in there that was going to require that the air be expelled from the system when filling. And that therefore due to the lack of air in the system to act as a cushion for the expansion and contracton with temperture flucuation, that the relief valve was now required.
However, I can't see anything in '10 requiring any provisions for removing air from the system. 8.17.4.2.4 allows the inspectors test anywhere after the WFS, so they're not trying to enforce it that way. Anyone see anything more in this regards?

 

[cdafd]

this was my guess, because I have seen systems set there and build pressure:::::

13-117 Log #18 AUT-SSI Final Action: Reject
(7.1.2.1)
_______________________________________________________________

SUBSTANTIATION: In general, it is poor engineering practice to trap a
volume of water behind a valve indefinitely without providing a means of
reliving pressure increases due to thermal expansion. I have encountered one
instance and heard of others where non-gridded wet pipe systems reached
pressures of 225 to 250 psi due to thermal expansion on hot summer afternoons
in warehouses that were not air conditioned. (In the case I investigated, the
supply pressure on the public main was stable at 75 psi.) This places undue
stress on components and causes premature failures. While non-gridded
systems do tend to trap air pockets that cushion such thermal expansion
initially, these air pockets can not be relied upon to remain in place indefinitely
since, over time, the air will dissolve into the water in the system. The
dissolved air in the water does not meaningfully increase the compressibility of
the water. The small relief valve needed to correct this problem costs little and
is not difficult to install.
COMMITTEE MEETING

http://www.nfpa.org/assets/files/PDF/ROP/13-A2006-ROP.pdf

 

[Cidona]

Thank you for the post cdafd.
Interesting I have not seen the ROP before (struggle to keep up on the ones that make it through). It seems that the proposal above was looking for an exception to the requirement for the relief valve if the building's temperatures were going to be kept below 100F. It was shot down, but what's interesting is in the 2nd last sentence in the explaination of why it was shot down, where it says 'With the efforts underway to remove trapped air in all wet pipe systems to avoid MIC problems ...'
That's the part I'm looking for! What efforts underway are they discussing? I would imagine in order for a proposal to be shot down based on something, that that something must be more than just a 'plan' they hope to incorporate in the future. i.e. That there is a requirement to remove the trapped air.
Thanks again for the link!

 

[SprinklerDesigner2]

This is a good change, I applaud it.

Many times I have encountered a wet tree system and due to heating or whatever the pressure gauges were pegged out and on a few heads had to be replaced..

It is scary.

 

[FFP1]

Whomever wrote the "substantiation" comments is simply wrong.

I have proven, documented and resolved the excessive system pressure problem more than 30 times in my career. I have witnessed and diagnosed the problem more times than I can remember. The problem is absolutely caused by significant volumes of air trapped in the closed sprinkler systems. I have used digital pressure recorders (pressure readings every 30 seconds over 7 day periods) to prove the problem. Gridded systems are worse than tree fed systems, but the problem is common, in Georgia anyway, regardless of the system type (elevation differences which allow air to be trapped in a significant percentage of the system piping IS the common denominator). One gridded system fed by a gravity tank {60 psi static} had system pressures in excess of 500 PSI in the afternoon periods! We bled most of the air from the system and we have not seen pressures above 65 psi to this day; two years later.

I have attached a copy of the chart from this system for reference. We bled air from 16 upright sprinklers (one from each of the grid branch line) for almost 90 minutes until all 16 sprinklers were dripping.

Inspector's Test Connections at the riser and ignorant/lazy sprinkler contractors who do not bleed most of the air from the system piping are most definitely the source of the problem. ITC at the riser (rather than at the end of the system through the high point of the system) is simply poor engineering design. The requirement for a PRV on all wet systems is a RESULT of the relatively recent trend of placing the ITC at the riser and the fact that many contractors do not remove most of the air from the systems any more (like the good old days)!!

You might be skeptical and you might even think I am wrong, but I assure you that I am not incorrect in this matter.

People who claim the excessive sprinkler system pressure problem resulting from temperature changes is caused by the thermal expansion of water are wrong..........while this is technically possible, it is only actually possible if the systems are 100% water & the thermal expansion for water does not increase exponentially at temperatures below 150F. Neither is true for almost all sprinkler systems.

 

[MGXFP]

Water expands such a tiny amount over the environmentally relevant temperature range that its laughable to think its responsible for pressure build up over time.

Thermal expansion of water - the delta density of water over this temperature range would easily be accomdated by the slack in a grooved coupling/fitting.

 

[Cidona]

I was unable to open the file FPP1 had linked. Can someone please let me know if they are able to open so I know whether there's a setting that needs fixing on my end or if there was a problem with the posting of the file. Thanks.
I had come across some calculations on this in the past and the verdict derived that while the volume of air increases and decreases vastly greater than water over the range of temperatures that a wet system could be subject to, the fact that the volume of water increases with temperature more than steel, and the fact that water unlike air is uncompressible, that the increases in water volume, though slight in comparison to air, would be more prone to mechanical failure of the piping if there was no air in the system to act as a cushion in this situation.
If I find it I will pass on the info, though the paper only represented one man's option (though seemed very comprehensive), and as the with a lot of scientific theories there would often be others than can present a different view, present different calculations and show the opposite. With the intent of having a meaningful discussion in this regards, I want to mention some documentation that would seem to indicate that the concern with pressure is more in the absence of air than excess air.
(a) While not in NFPA, Section 15.6 of EN12845 (Fire Protection Association, LPC Rules for Automatic Sprinkler Installation, European Standard), it says 'Pipework which is completely full of water, may be damaged by the increase in pressure due to temperature rises. If complete venting of air in an installation is likely to occur, e.g. in the case of a gridded layout with flushing connections at the extremities, consideration shall be given to the fitting of pressure relief valve'.
(b) In looking at the ROP A2009 (see link below), Section 13-75 log 406, and 13-76 Log 435 (in regards to the Pressure Relief valve section 7.1.2.1 of NFPA 13), also seems to the panel's concern to require relief valve was due to 'growing popularity of venting trapped air from metallic wet pipe systems to inhibit corrosion activity increases the importance of protecting all wet pipe systems from over-pressurization'.
(c) In looking at ROP A2009 Section 13-78 Log 407, the submitter is proposing a means to remove air from wet systems. The panel seemed to agree with the submitter's 'desire' to remove air from the system, but rejected it, suggesting that the submitter resubmit addressing 5 concerns the panel had on the verbiage. However, the desire to remove the air from the system is again stated as to limit corrosion rather than to address the potential for excess pressure build up due to excess air.
(d)The exception to providing a relief valve (Wet and Antifreeze), is to provide an air reservoir. So it would seem the committee feels that air can act as a cushion. It does not mandate (yet, seems pending the revision on ROP 2009 Section 13-78 Log 407 next cycle), the removal of air from the system.
I do not doubt that excess air in the system can drive system pressure to the levels you speak of. I can easily enough visualize the air pockets contracting in the colder temperature, allowing more water past the check valve, and then the system pressure increasing significantly the next day when the air pockets increase in volume when it warms up the next day. I can also see a system with excess air causing false flow alarms. And I can see air in the system increasing corrosion conditions. So there are lots of good reasons to vent the system, and installing a relief valve is not that big a deal. However it seems at the moment there is no requirement to remove the air from the system.
I know there has been considerable discussion within the industry and on this forum in regards to the location of ITC. In the 2007 edition of the NFPA 13 Handbook there seemed to be a contradiction in that one section is telling you 'ideally this is located at the highest most remote branch line’, and another section discouraging it because it allows fresh oxygen into the system whenever an inspectors test was used.
I thought that they were going to be requiring a means to vent the system of air while allowing the ITC to be located anywhere downstream of the WFS. Which would seem to keep everyone happy. I guess if the Section 13-78 Log 407 of ROP A2009 gets revised with the recommendations of the panel in next cycle will allow everyone to sleep save. Maybe?.

http://www.nfpa.org/assets/files/PDF/ROP/13-A2009-ROP.pdf

 

[FFP1]

The file name had a "#" in the name.....maybe that was the problem??

 

[Cidona]

OMG!!! That is scary. And I bet when you were first picking out the model of logging you were figuring there'll be loads of flex with the 0-520 psi model. It just made it
Well theories be damned. If you come accross something like that you're going to need alot of conviencing to think otherwise. I'm very surprised that the variences in pressure are happening in such a shory period. I was expecting to see the pressure build and fall more gradually over the course of the day/night. What would you attribute the sudden contraction (reduction in pressure) to?
Also, when you mentioned that you bleed air from 16 sprinklers for 90mins, please tell me you meant 90 seconds, which even then is in excess of the 60 required for dry systems.
Which brings up another point in regards to the need to remove air from the system, delay in putting the wet stuff on the red stuff.
Thank you for the graph FPP1! Never saw anything like that before.

 

[Cidona]

Interesting, I see the annex to section 8.16.4.2.2 discusses venting of system. Problem with that is (a) it is in the annex and therefore technically unenforcable, and (b) In secion 8.16.4.2.2 within the main body of the Standard, it is only addressing 'Where water supplies or environmental conditions are known to have unusual corrisive properties...' and therefore is not applicable to your everyday system.
The annex seems to provide the verbage, etc that was requested by the panel in regards to ROP A2009 Section 13-78 Log 407. Kinda surprised that it wasn't just incorporated to be applied to all wet systems.

 

[MGXFP]

Now the graph FPP1 posted is very interesting. I wonder where the location was of the test and the environmental conditions. A change in pressure of 450 psi within a 24 hour span of time seems quite large to be taken into account by the ideal gas law alone. The only condition I can envision is that of a desert with high high temps and low low temps.

However I do not have all the facts and data but this is a fascinating case.

What a great forum and thank you FPP1 for the info!

 

[FFP1]

I need to clarify a few facts.

We identified the problem and used a scissor lift to loosen 16 individual sprinklers on the high side of the system branchlines until we could hear the air hissing. The air bled from the system piping for >90 minutes before all 16 sprinklers were dripping (i.e. it required that long to remove the large volume of air from the system). The branchlines for this system are elevated using riser nipples ~24 in. above the near main and ~24 in. above the far main. The low point drain & ITC could not possibly remove the trapped air for this system. When this system is drained and filled, the air removal must be completed manually due to the design of the system.

This is a VERY common problem during Spring & Fall (sometimes summer as well) in Georgia. The weather temperatures during this test in April varied between 38-42F at night to 78-85F during the afternoon. The pressure increases & decreases followed the temperature fluctuations exactly as one would expect; gradual increase and gradual decrease as the temperature changed. You cannot see the pressure gauge change, but if you document the pressure and walk away for 30 minutes, the pressure will be different when you return.

Most systems create a small drip or leak well before the system pressure reaches 500 psi. I have one customer who had 6 sprinklers POP on three different systems over a two month period before they finally let me bleed the air from 5 of the 19 wet sprinkler systems (only one of the 5 systems affected by the "trapped air problem" used a gridded design!), but all had elevation changes which allowed significant trapped air pockets.

I have many more convincing examples, but I am not going to spend the time trying to describe them on this forum. Have an open mind, remember in general what I am saying, look for the problem in the field, verify for yourself whether or not I am right or wrong AND hopefully one day enough people will realize the truth that the NFPA committee will take the appropriate action to resolve the problem by changing the applicable NFPA requirements!

I will go to my grave knowing trapped air creates this problem (as well as SEVERAL other SERIOUS problems) in sprinkler systems..........I have resolved this issue way too many times simply by removing the trapped air.

I hope the time I just spent on this topic proves to be helpful to someone out there!

 

[TravisMack]

Wouldn't it make sense just to put one of these on the high point of the system?

http://www.acfirepump.com/files/FP2.7_AARV_301.pdf

Then, you still keep the ITC at the riser to facilitate testing of the flow switch.

Travis Mack
MFP Design, LLC

http://www.mfpdesign.com

 

[FFP1]

TMACK has a great idea..............this would be a huge improvement for many/most systems. No brainer really, but it must be a "shall" item in NFPA 13 to make sure everyone complies. A manual isolation valve should be required at the base of the automatic air release valve to allow servicing or replacement WITHOUT draining the entire sprinkler system!

The only other obvious problem is what do we require for systems which have multiple high points (i.e. grid systems which have near & far mains lower than all of the branch lines)?

 

[DRWeig]

Interesting topic!

From an operations side of things, would the air relief valve possibly vent enough at a time to jiggle my flow switch and summon the firefighters?

We get significant pressure fluctuations on our supply side (small town water supply with big industrial users). Two or three times a year we get alarmed -- have caught the city pressure as culprit with a recorder.

What say?

Good on y'all,

Goober Dave

 

[FFP1]

DRWeig:

Trapped air pockets can definitely cause false fire alarms (especially during Spring and Fall)........most sprinkler systems do not have automatic air release valves.

FYI.......if your suction supply fluctuates AND the alarm check valve on your sprinkler system does not hold properly (i.e. bad rubber seat), this combination can and will cause false fire alarms.

Most building owners contact the fire alarm company when these problems surface - most of the time this approach is a waste of time & money. You need an experienced and reputable water-based fire protection contractor to evaluate and resolve this problem.

 

[DRWeig]

Thanks for the info, FFP1!

Good on ya,

Goober Dave

 

[Cidona]

There seems there are a few products on the market for the auto venting. Some seem to come with a small pan to collect small amounts of water that get discharged during normal operation. There are also lots of cool options and features with the supervision/automatic closing of the vent shut off valve.
Travis's looks to be considerably cheaper and has a rating to 300psi where most the others seem to only go to 175.
Has anyone actually come across a spec requiring auto vents on a system?
In general these auto vents do seem to hold a lot of promise though will only have partial benefit if not fully embraced by the Standard and the industry.
Would seem in order for these to be utilized on a system, that their location(s) should be marked on a sign by the riser, similar to dry system low point drains to ensure proper ITM and so when serviced, the person servicing can tell wheather there is an auto ent on the system or if they have to manually vent somehow.
As far as the gridded system FFP1 was mentioning with riser nipples up from the main to branch lines, there is probably no perfect answer. With how software, etc is advancing I could see a scenario in the future where fluid delivery programs will be able to show you how much air is left in the system when you fill the system. Moving the auto vent location would show the different amounts of trapped air. You give the program the supply pressures, expected temperatures and it tells you if you need additional auto vents
Until then I guess the best would be to try visualizing the filling given the layout of the system in relation to the supply, balanced with where you can physically locate the auto vent in the field, and with proper ITM in accordance with NFPA 25 should be considerably better than the norm today. Since the pressure relief valve is mandated now anyway the excess pressure would not be an issue, but would be best to prevent the other negatives discussed with excess air in the systems.

http://www.fpscmi.com/pdfs/Ejector AAV-8.5x11_2009-11-05.pdf

http://www.pottersignal.com/Sprinkler/pdf/5401135-M_PAAR.pdf

 

[Cidona]

I wonder if the following could be considered in accordance with the NFPA requirement (or intent) for relief valve. Or if not, should it be?
Say you had a 3 story building that you were providing floor control assemblies on each floor. No check valve on the floor control assemblies as there is no FHVs. Would you not be able to provide a single pressure relief valve in the valve room to remove excess pressure from all the systems?

 

[sprink131371]

One of the first things I learned as a young helper, was that it was VERY IMPORTANT to relieve all of the trapped air through the inspectors test while filling the system. Not only to stop water hammer, but to cut down on false alarms. I don't recall a mention of increased pressure, but after installing many systems in the heat of Georgia, I have no doubt that the trapped air is a problem, (I had one system testing at 220 psi and at 295 psi within an hour).
Things are not now as they used to be and it seems you find more and more poor design, & sloppy installation. Even the old sprinkler man saying "whatever it takes" doesn't seem to apply anymore (to most, there is always the exception).
I feel the pressure relief installed on all wet systems is a grand idea. Plus it eliminates the "guess" work as to whether one is required for a certain type of system.

If we all agreed, we would be Communists. - Me