Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations GregLocock on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Fire Sprinkler water suppied from a pond

Status
Not open for further replies.

StoneCold

Chemical
Mar 11, 2003
992
I am considering creating a pond on our property to supply our fire sprinkler water. The design is about 4000gpm. If we don't create the pond and go with existing fire suppresson supply water we have to pipe it to our facility and then add a booster pump to get the pressure up. I think the piping/pump is going to run us about $600k. So I am thinking a pond may be the way to go. Storage tanks are also an option but the cost is about a wash with the piping system. Has anyone out there used the pond method? Would you do it again? One more little tid bit is that we are located in colorado so the pond will freeze on the top.

Regards
StoneCold
 
Replies continue below

Recommended for you

Also, as a side note from the sprinkler end, all of your pendent sprinklers have to be on return bends to eliminate possible build-up at the bottom of the drops.
 
Stone,

From an insurance point of view here are some comments:

1. Some insurance companies will not give you credit for the fire department if you are not within 1000’ of a public fire hydrant. You would be considered an unprotected risk, EVEN if you have an on site water supply. This can affect the rate you pay, so check with your insurance company.
2. I would install a diesel AND electric fire pump. Make sure you calculated the added flow for the fire hydrants when you size the fire pump.
3. How will the pond be filled when you have a drought? Well, etc.
4. The required water flow is based on how much the pond can hold minus the ice, silt build up, etc.. Go bigger then your calculations.
5.Remember to place the pumps at the lowest part of the pond to get maximum capacity.
6.Vertical turbine pumps will need to be used, a bit more then your typical fire pump.
 
Actually being in Colorado is even more reason to go with a pond over an on site storage tank. Heating costs, even with an insulated tank, can eat you alive and it's a utility bill that never stops.

I've done a number of these installations (maybe a dozen) over the years.

Would you like to see photo's?
 
SprinklerDesigner2
I would like to see the pictures if you have them. I am going to have to discuss this more with my insurance agency and others in the project. I think the cost is about a wash between the two systems if you compare the total life cycle costs so I may want to lean toward running a long pipe from the city supply so I don't have all the maintenance issues.

Thanks
Stonecold
 
Design is for 4,000 gpm so a 3,000 gpm vertical turbine should do just fine.

From the appendix of NFPA #20 1999 Edition
"A-2-3 - A stationary pump for fire protection should be selected in the range of operation from 90 percent to 150 percent of its rated capacity. The performance of the pump when applied at capacities over 140 percent of rated capacity can be adversely affected by the suction conditions. Application of the pump at capacities less than 90 percent of the rated capacity is not recommended."

A 3,000 gpm fire pump producing 4,000 gpm is operating at 133% which is about where you would want it.

A wild stab here at some dollar figures. An electric pump would probably run $40,000 with a diesel maybe running around $70,000. Not sure about these figures it has been a long time since I got prices on something this large so don't be surprised if I am off a bit.

The pumphouse, and I will provide photo's and details later this evening, will probably run between $60,000 and $100,000 and then you have electric (if you don't go with diesel which is the way I would go) and digging the pond.

I would be amazed if the total installation ran over $300,000 and if I had to make a stab at it I would guess total costs to run somewhere around $200,000 and $250,000 but remember construction costs can vary widely around the country.

The problem you would have with a booster pump from a municipal supply would be finding one that can deliver the amound of water you would need. A 3,000 gpm pump would need 4,500 gpm and by NFPA standards you couldn't pull a negative suction but EPA standards mostly limit your suction to not below 20 psi.

4,500 gpm @ 30 psi is a tall order for almost any municipal supply to handle.

Going with a pond and diesel pump has many advantages among those being.

1. On a wild guess a 3,000 gpm pump will require somewhere around a 300 to 40 HP and the electric to this could get unreasonably high. With a diesel I would think a 200 amp panel would take care of all the pumphouse needs.

2. You can set the diesel pump to automatically start when power is lost. The heat coming off an engine this large would keep the pump house warm and toasty.


 
"running a long pipe"

How long is long?

With 12" you are losing 16 psi per 1,000 feet so I would guess you are going to need at least a 14" supply pipe.

With 14" you are going to be losing around 7.5 psi per 1,000 feet and with 16" the loss would be around 4 psi.

Being a chemical plant the local water people will probably require an RPZ and those, I figure you will need 2 10" in parallel at a minimum, are going to be costly both in terms of dollars and maintenance.

RPZ's often foul creating a real mess.

That and two of them will cost about 10 psi in head loss. NFPA standards call for backflow preventors to be downstream the fire pump but most water purveyors require them to be upstream the pump and that is their call.

Not saying their aren't water purveyors that can deliver 4,500 gpm at 30 psi but they are few and you have to be in the right location.





 


SprinklerDesigner2 Do you have a code section for this one??????

" NFPA standards call for backflow preventors to be downstream the fire pump but most water purveyors require them to be upstream the pump and that is their call."
 
We have about 1500 ft run from the connnection point in the street to what I would call the beginning of our "loop" piping that runs around the facility.
We can get 4000gpm at 40psi at our connection point.
You seem like you know a lot about it. Do you work in the sprinkler design area, as in for a contractor? I have had some contact with fire protection engineers but they don't seem to know anything about the actual install.

Regards
StoneCold
 
SprinklerDesigner2 Do you have a code section for this one??????

" NFPA standards call for backflow preventers to be downstream the fire pump but most water purveyors require them to be upstream the pump and that is their call."

===================

Yes.

NFPA #20 Section 2-9.9

Section 2-9.9 Devices in Suction Piping.

The requirements for devices in suction piping shall be as follows.

(1)No device or assembly, including, but not limited to, backflow prevention devices or assemblies, that will stop, restrict the starting, or restrict the discharge of a fire pump or pump driver shall be installed in the suction piping.

Exception No. 1: Except as specified in 2-9.5.

Exception No. 2: Check valves and backflow prevention devices and assemblies shall be permitted where required by other NFPA standards or the authority having jurisdiction.

Exception No. 3: Flow control valves that are listed for fire pump service and that are suction pressure sensitive shall be permitted where the authority having jurisdiction requires positive pressure to be maintained on the suction piping.

(2) Suitable devices shall be permitted to be installed in the suction supply piping or stored water supply and arranged to activate an alarm if the pump suction pressure or water level falls below a predetermined minimum.

Here's Section 2-9.5 and a question.

2-9.5* Valves.

A listed outside screw and yoke (OS&Y) gate valve shall be installed in the suction pipe. No valve other than a listed OS&Y valve shall be installed in the suction pipe within 50 ft (16 m) of the pump suction flange.

==============

Let's say I build a pump room on the property line with the city water main just 10 feet away. The city is going to tap the line with a standard tapping valve and sleeve, or perhaps cut in a tee and install a roadway valve and box, then I run into the pump house with a total of 10 feet of horizontal pipe, a 6 foot riser piece and then begin my pump installation.

Unless I do some running around in the pump room it is obvious I won't have the required 50 feet between the roadway valve (not an O.S/&Y.) and my pump suction flange.

Now what? What say you FPE's?

=================

We can get 4000gpm at 40psi at our connection point.

You seem like you know a lot about it. Do you work in the sprinkler design area, as in for a contractor? I have had some contact with fire protection engineers but they don't seem to know anything about the actual install.

==================

Are you a lucky one at least you have something that can be worked with.

The question is how much residual pressure do you have at 4,500 gpm which is what you will use for a 3,000 gpm pump?

I have to guess it will be somewhat less then 40 maybe 38? Let's jump on 38 psi and use that.

Word of caution here. Whenever you try to calculate what you might have be generous to yourself giving yourself some wiggle room. Many a time I have done this and all to often the actual results don't come out to what my calculations show me I should have and it is never to my benefit.

You have equivalent fittings and I would figure 500 feet at least because not only do you have a tapping valve and sleeve but you will probably have some sort of manifold to a couple of RPZ's that, being a chemical plant, I am certain the water department will demand be used.

Make that more then 500 feet; let's talk about the physical tap.

A tapping sleeve/tee is not like your normal tee its equivalent length is staggering. For an 8" tap the hole drilled for the coupon is only slightly larger then 6" and the actual head loss will be much, much higher then a standard 8" tee even though many designers figure it as just a tee. It isn't.

Had trouble on a job once and I ended up concluding an 8" tapping sleeve had the equivalent friction loss of over 250 feet while a normal tee is assigned a value of only 45.5 feet. (Per Table 14.4.3.2 I used a C Value Multiplier of 1.33).

To be safe I would roughly use 2,250' of pipe and equivalent fitting length.

What you don't want to happen is to have your company expend all that money only to find out on the day of the fire pump acceptance test you don't have enough water. I've done a lot of mistakes in my days... running into beams, not seeing duct work; that sort of thing but if I have enough water I can make anything work it's when you don't have enough water things get ugly. And spending a half a million on something that doesn't work would have to be ugly.

At 2,250 feet you are going to lose somewhere around 9 psi while flowing 4,500 gpm.

From 38 psi we now have 29 psi at the end of the line.

Those RPZ backflow preventers are killers.

The literature is going to tell you the head loss through these devices are going to be between 7 and 10 psi but I am convinced the manufacturers lie. They all lie.

To get the UL sticker they must tweak a factory model then test it because every time I have take readings on one my readings are ALWAYS more then what the literature tells me it should be. Literature tells me at 800 gpm I should lose 6 psi and when I measure it, and I use calibrated test gauges not cheap sprinkler contractor gages, it is always 2, 3 or 4 psi higher then what it should be. I can't emphasize this enough. Be prepared for a 12 psi loss maybe more.

So take 12 psi from the 29 and you are down to 17 psi suction pressure at the pump when flowing 150% and you might have just failed.

In addressing the issue NFPA #20 says anything above zero is passing but whether they push it or not you also have to meet the Federal Clean Water Act requirements and one of the requirements is booster pumps taking suction off public water supplies can not draw the suction below 20 psi.

Take Georgia for example.

Let me get this off and I will finish up in the next post. My computer locks up once in a while.





 
We were talking about the 20 psi suction pressure that is required to be maintained on fire pumps when operating at 150%.

Yeah, I know NFPA #20 will accept any positive pressure but NFPA isn't concerned about public drinking facilities and the clean water act.

Here's what the State of Georgia has to say about it.

it's a snapshot of pages 62 and 63.

Looks like 20 psi to me.

I know the 20 psi requirement is the law in Ohio and Kentucky too. In Ohio all fire pumps taking suction from a public supply have to be equipped with a low suction shutdown device that cuts off the fire pump anytime suction pressure falls below 20 psi at the pump suction flange and when that baby kicks in before reaching 150% there is no doubt you just failed the acceptance test. Talk about ugly.

I don't know where to find it but I am willing to bet a steak dinner Colorado has the same 20 psi requirement somewhere. Anyone that took money from the federal government for water systems has it.

If your project was mine I wouldn't accept anything less then a 20" suction line wanting the friction loss in the pipe to be as close to negligible as possible figuring I would lose plenty through the backflow prevention devices.

And, if it were my project, I would add at least 50% to any head loss value the backflow prevention manufacturers are telling you what to expect. If they say 10 psi I would figure the real value would be at least 15 psi. If this is your project you don't want to get burned.

I'm a sprinkler designer working for a small sprinkler installation company. With the over 30 years experience I have I am overkill for the small company I work for but I like it. Never did well in large corporate environments I guess I don't play well with other people. I need my freedom more then I need the money.
 
SprinklerDesigner2:

About the 20 psi; is that at the public main, or the suction flange of the pump? I have had water purveyors and other AHJs take it both ways. For example, I have seen pump tests where the gauge at the suction flange gets down between 5 and 10 psi, but the gauge on the hydrant at the street (I always gauge those for pump tests as well) is reading 20+ psi. This is as you said above, due to friction loss in the lines and loss across the backflow preventers.

These tests have been accepted by the local water authority without question. I have seen other instances where the water authority in a different jurisdiction would not accept the same scenario. That was a real messy situation!

So, where is the 20 psi requirement? The pump suction flange? City main?

Have a good day!
 
"So, where is the 20 psi requirement? The pump suction flange? City main?"

It's at the pump suction flange.

Sure, I've taken a gage to the nearest fire hydrant too but the private fire main to the pump is still part of the public water supply even if it is on private property and privately owned.

In Ohio it is the pump suction flange and their inspector's are learning about it for what it is worth.

 
I wish this had an edit.

Anyway, look at what it says in Georgia again.


"the pressure in the suction line shall be maintained at or above 20 psi...." and that suction line extends from the city water main to the pump suction flange where it becomes the discharge line after the pump.

Even though many aren't aware of it I would bet the same wording is used everywhere in the US.
 
Here's a doctored screen shot from NFPA #20.


If you haven't done this sort of thing before what is going to surprise you most is how large the area of the intake screens will be. With a 3,000 gpm pump they are going to be huge!

How big? You will find your answer in NFPA #20 Section 2-9.8.

2-9.8* Suction Screening.
Where the water supply is obtained from an open source such as a pond or wet pit, the passage of materials that could clog the pump shall be obstructed. Double removable intake screens shall be provided at the suction intake. Below minimum water level these screens shall have an effective net area of openings of 1 in.2 (645 mm2) for each gpm (3.785 L/min) at 150 percent of rated pump capacity. Screens shall be so arranged that they can be cleaned or repaired without disturbing the suction pipe. A brass, copper, monel, stainless steel, or other equivalent corrosion-resistant metallic material wire screen of 1/2-in. (12.7-mm) mesh and No. 10 Brown & Sharpe (B. & S.) gauge wire shall be secured to a metal frame sliding vertically at the entrance to the intake. The overall area of this particular screen shall be 1.6 times the net screen opening area. (See screen details in Figure A-4-2.2.2.)

We need an effective open area of 1 sq. in. per gpm at 150% of rated pump capacity. For a 3,000 gpm pump we need 4,500 sq. inches.

If we use the No. 10 Brown & Sharpe (B. & S.) wire mesh we have to multiply the 4,500 sq. in. by 1.6 and we end up with 7,200 sq. in. or 50 sq. ft.

This 50 sq. ft. should be available at the ponds lowest pumping level not when it is filled.

The screen can be any shape... we can have 4'-0 wide by 12'-6” high or we can square it up to 7'-0” x 7'-0” (I know, only 49 sq. ft. but close enough) and in any event it's going to be bigger then most would expect.

Here's a preliminary drawing of a pump house I did for a customer last September.


This is for a baby pump compared to the Colorado big boy. 32 sq. ft. of screen for 20 sq. ft. of opening.

2,880 sq. in. this pump was going to be 2,000 gpm but I used a different stainless steel screen and that was all I needed to get my area.

How do you figure the lowest level of water? I never thought it being the bottom of the pond because some ponds, like Lake Superior, can get pretty large. If NFPA #13 (or whatever) required a 2 hour supply I figured the where that level would be with the pump operating at 150% after two hours but you are going to want to be conservative remembering those hot summer days when pond levels can drop from evaporation and lack of rain.

On this drawing you will also notice I placed the pump close to the side of the wet well. Try to get the pump bowl a couple inches away from the side of the shaft it breaks up vortices.

Here's a pump house showing the intake screens for a 2,000 GPM diesel engine driven vertical turbine fire pump.


Here's the same pump house with the pump being installed through the roof hatch.


Another angle at the intake screens.


Here's the inside of the pump house taken during installation.


Here's a different pump I think it is a 3,000 gpm pump. This is what you would have in Colorado.


During installation here's an I-Beam base. When it came to steel I never designed it as it was beyond my capability. Turned that part of the job over to a PE so weights and thrusts would be his problem. These pumps weigh a lot.


Here's a shot looking down at the screen. It's 4'-0” wide.


A shot of the non-exciting side. Use double doors so you can remove items easily and if you opt to go with a diesel engine driven pump make sure the doors are aligned with the engine so it can be removed with a fork lift.


This is how an intake should not look. Sad to say this house had two 2,000 gpm pumps one diesel the other an electric. This intake should have been designed for 6,000 gpm and looks to me like you could barely get a jockey pump to work. Sad, sad, sad.


Ponds do take a little care and algae killer helps a lot but taken care of you can have the best fishing hole in the county. :) This was one of my designs from 1982.


Here is the same installation showing what algae can do to the intake screens.


And finally the same pond a few weeks later after a much needed algae treatment.


Oh, and StoneCold , one more thing.

What about elevation? I didn't think about it before, being in the Tidewater area of southeast Georgia where everything is flat, but what is the elevation difference between where the pressure was measured and your proposed installation? Perhaps this was already calculated? Ten feet can make a big difference with 20 feet being a disaster if your site is higher then the waterline.

Night all!
 
About pond freezing over.

12" of ice is safe to skate on.

Shouldn't be a problem, I've done a few of these in Michigan were it gets cold too, just make sure the "lip" of the pump house intake extends below low water level and it should be just fine.

A few customers installed small airators because they were concerned but the most they ever got was just a skim of ice in the wet well.


 
Let's say I build a pump room on the property line with the city water main just 10 feet away. The city is going to tap the line with a standard tapping valve and sleeve, or perhaps cut in a tee and install a roadway valve and box, then I run into the pump house with a total of 10 feet of horizontal pipe, a 6 foot riser piece and then begin my pump installation.

Unless I do some running around in the pump room it is obvious I won't have the required 50 feet between the roadway valve (not an O.S/&Y.) and my pump suction flange.

Now what? What say you FPE's?


Would say that a roadway valve would be equal to a os&y

Exception No. 2: Check valves and backflow prevention devices and assemblies shall be permitted where required by other NFPA standards or the authority having jurisdiction.


So a backflow is allowed if required by the ahj
 
cdafd:

I seriously have to believe that a NRS Gate valve would be acceptable in the underground line for a pump. I believe the intent is simply not to have a butterfly valve in the suction piping. I have done the scenario you describe on a few cases and have never been questioned on it.

And we also put backflow preventers on the suction side of the pump all the time because of the water purveyor. You just have to get creative sometimes to get the 10 pipe diameters of piping between the BFP and pump suction flange, while not having a horizontal elbow in the plane of the pump.

T
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor