Extensive pressure shocks in fire water sprinkler system

[Tarkjell]

We have a firewater pump system where the piping has been ruptured when the pumps started. A "top up" vessel was then installed to eliminate the possibilities for water hammering to occur. This pressure dampening helped. However, pressure peaks are registered on the gauges which show peaks way over the operating pressure. My job is now to find the best solution to further smooth the pressure peaks and hence protect the piping system. The sprinkler system is the weakest part and consists of CuNi soldered piping. Does anyone have any idea how to go about solving the problem?

 

[BigInch]

Depends on the system equipment and piping configuration.
Sounds like you may not be using a jockey pump to maintain operating pressures and when the main unit kicks in to bring pressure up, its too much all at once. Jockey pumps may be one solution. Another would be to somehow slow your pump ramp time on startup, but not the best technique for firewater. Could be something more simple as just an orifice plate between the pump and the new vessel. You can try a number of possible solutions, but the best thing to do is to get a copy of some software that can model transient flows and learn how to use it and model your system. Its much easier and cheaper to find a solution using the software than it is knocking around a lot of pipe.
Have a look areound here,

http://www.aft.com

**********************
"Pumping systems account for nearly 20% of the world’s energy used by electric motors and 25% to 50% of the total electrical energy usage in certain industrial facilities." - DOE statistic (Note: Make that 99.99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[Artisi]

Who designed the system, yourselves or a fire pump company / engineer.

The system "standby" pressure should be above the system / pump operating pressure and the pump supply side should be fully primed and air-free meaning that start-up should be smooth and not subject to any major pressure spikes.

As discused by BigInch, is the system fitted with a jacking pump? if not you should look into fitting one in conjunction with a bladder tank which will maintain the sprinkler system in a pressurised ready state.

 

[msquared48]

How about using one or more pressure relief valves in the system to control the spiking?

Mike McCann
MMC Engineering

 

[BigInch]

Another possibility, but since that treats the symptoms rather than the disease, it can make things better or sometimes even worse. One thing for certain is that you need more equipment, maintenance, possibly a tank and reinjection or other disposal system, so trying to reduce hammer by other means first is generally preferrable.

**********************
"Pumping systems account for nearly 20% of the world’s energy used by electric motors and 25% to 50% of the total electrical energy usage in certain industrial facilities." - DOE statistic (Note: Make that 99.99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[chief]

Fit a high point air vent in your system and a pulsation damper in the pump discharge line

Offshore Engineering&Design

 

[Artisi]

There shouldn't be any air in the "system", a fire system should be fully primed with everything on 100% standby.

 

[MortenA]

A question:

Do you have high point more than 10 metres higher that your lowest "off take" and do you as BigInch asks use jockey pumps to maintain a normal elevated pressure and then start the main FW pumps on PALL? If so then your problem may be that you get a vapour pocket in these high places - and when pressure increases after the main FW pump starts then you see a large surge due to collasing vapour pockets in these places.

I would like to warn against air vents at these high point. If you have hydrant/monitors here 8e.g. the high point are risers to elevated hydrants/monitors) then you will have the problem come back if you open any of these monitors/hydrants.

Instead i would like to suggest non return valves in the risers (if the high point are such riser). I have done simulation of these systems a number of times and this is a common problem.

Best regards

Morten

 

[Tarkjell]

Thanks for valuable feed back, guys!

Jockey pumps are used in the system.

The piping system is extensive with fire ring mains serving hoses, hydrants, water canons, sprinklers, foam systems etc.

Elevation levels exceed 10 metres by far.

There are definitely air pockets in the system and air is probably also pushed in by the fire pumps as the air relief valves most likely can't release enough when the pumps start instantly.

100% air free priming of the system is not feasible. If the system actually hadn’t air pockets and vacuum was avoided (by starting the pumps before any water was released), then there wouldn’t be a problem like you are indicating Artisi.

Besides, air pockets may be good for cushioning pressure spikes.
I now know that it was a 10" riser that busted and that it happened at one of the higher elevations. An atmospheric tank was then installed to feed water into this riser as the water column drops in an attempt to avoid vacuum. When the large fire pumps starts then what I think happens is what Morten describes.

More than one phenomenon is probably occurring, pressure surges due to water-air-water flow and water hammer effects.

We will be making a simulation model. There is talk about a computer program called Pipenet. Just getting together enough information of what the whole system looks like is a big job. Morten, what simulation program have you used? And have your systems included a pressure tank for avoiding vapour pocket collapsing? And are your non return valves mounted further down in the riser than where the vacuum occurs preventing the water gushing back up the riser when the pumps start?

 

[msquared48]

Tell the secretaries to quit smoking under the sprinkler heads...

Mike McCann
MMC Engineering

 

[BigInch]

Air pockets don't act as cushions, they allow water coming under pressure to accelerate against virutally no resistance until they reduce their volume, the water stops suddenly with higher pressure spikes than you would get with no air pockets. The compressed air can and does quickly release its stored energy at any reduction of pressure and the velocity changes to the reverse direction.

That's one reason why you have air valves to get the air out of there.

If the system is as large as you imply, I doubt the jockey pumps have enough capacity to keep the pressure of the entire system up to a constant level.

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25% to 50% of the total electrical energy usage in certain industrial facilities." - DOE statistic (Note: Make that 99.99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[rconner]

I am just curious how thick is the 10" CuNi piping?

 

[Artisi]

"There are definitely air pockets in the system and air is probably also pushed in by the fire pumps as the air relief valves most likely can't release enough when the pumps start instantly. "

There shouldn't be any air in the fire pumps - they should be fully primed and air-free - if not you stand the chance that they will air-lock and fail to deliver when called into service.

 

[tkall]

I have used Surge from KY Pipe to analyze non-equilibrium flow in piping systems. It is not unreasonably complicated but it is very robust.
Tom
eliminite.com
squirtonsite.com

 

[786392]

I fully agree to and endorse Artisi
As what fire fighting reliability may we expect/Plan with air pockets in the system/pipelines.

Best Regards
Qalander(Chem)

 

[Artisi]

Lets put this problem into a scenario that everyone understands, would you tolerate air pockets in the hydraulic brake system of your motor car and be happy to tell everyone that a 100% airfree system is not feasible -- I think not, that's unless of course you don't value your life.

 

[786392]

Merely Splendid!

Best Regards
Qalander(Chem)

 

[MortenA]

I use Flowmaster and have done this type of simulations a number of times for FPSO's especially. Normally this forum is not for soliciting so i hesitate to forward my mail adress - but maybe we could make an exception?

The "air pockets" has in the cases i have investigated shown to be cavities. Letting in air can be a road to more troubles.

Best regards

Morten

 

[MortenA]

Artisi

Many systems has dry FW pumps.

FRAMO a big Norwegian supplier of FWP uses a two stage system with a submerged pump, a riser to the deck and a main pump. During start the air is released via a separator vessel (surge tank its called) and a control valve that first lets out air then maintains min flow. The FWRM is hoever filled with water on the FPSO's i looked at, and the pressure maintained by use of a jockey pump.

As Biginch says: When a deluge valve opens then the jockey dosnt have sufficient capacity to maintain pressure until the FWP has started. Pressure drops and now cavties may form at high point then once the pumps kicks in pressure increases again and the vapour pockets collapses causeing pressure spikes. Air valves/vacuum breaker) wont allways help because some high points may be at hydrants. When air is let into the risers the air remains there - then when somebody opens a hydrant first air rushes out - and then when the water collumn hits the outlet valve you get a water hammer effect because of the much higher capacity for air of the valve when comparing to water.

Best regard Morten

 

[786392]

At times manually air had to be bleeded at few hydrantsin such a circumstance; however the Prompt Fire Fighting capablity is somewhat in jeopardy
without adequate air venting arrangements indicated by Morten/BigInch

Best Regards
Qalander(Chem)