Friction loss calculations for Hosemonster testing scenario

[ContractorDave]

Please feel free to correct this process wherever an error in methodology or assumption is made.

Electric VIL Fire pump data plate: Net 500 gpm @ 90 psi, 150% = 79 psi, max = 100 psi. Assume suction side is always 50 psi. The fire pump is in a mechanical space with a test header that requires 100' of 2 1/2" hose. We are using a 2 1/2" Hosemonster with a 1 3/4" orifice (we have the tables but the coefficient is .975).

Start fire pump by dropping pressure at the control sensing line. Pump is operating in churn at 150 psi. Take rpms and controller v/a readings etc., typical of all stages.

Now here is where the Hosemonster procedure becomes problematic when plotting curves. If we throttle the test header valve so that the discharge reads 140 psi expecting to get our 500 gpm, we're not actually going to see that at the pitot correct? Would we not have to account for the friction loss of the hose at 500 gpm and add that number to the psi we are attempting to throttle the discharge psi to in order to see an actual 500 gpm?

 

[SprinklerDesigner2]

You need to throttle the the test header so you read 32 psi on the pitot (517*0.975=504 gpm) and then take your discharge pressure reading.

It doesn't matter if you have 10 feet of hose or 400' of hose once you read 32 psi on the pitot you are flowing 500 gpm.

What you can expect and what you get are two different things.

 

[NJ1]

I was under the impression that NFPA 20 does not allows more than 50' of hose per hose valve. Please advise.

 

[SprinklerDesigner2]

No limits on length of hose as far as I know.

 

[FFP1]

Hose length does not matter.......as SD2 stated, the flow rate is the flow rate (based on the pitot tube reading at the end of the playpipe). The flow rate is established and controlled by the test header control valves; once you have obtained the desired flow rate, the volume of water flowing through the fire pump is a known data point........then you can obtain the remaining data and evaluate the performance and reliability of the fire pump for that particular test point.

 

[NJ1]

I confirmed that NFPA 20 does not have a maximun length hose requirement however I heard that in the past many times so I will look into this for our own sake.

 

[ContractorDave]

SD2, FPP1

In my example the curve is 0 @ 100; 90 @ 500; and 79 @ 750. If I have 1000' of hose as opposed to 50' of hose and I throttle my test header valve to get 32 psi on the pitot, will I not have a significantly higher discharge pressure at the fire pump due to friction loss in 1000' of hose vs 50?

Thx
Dave

 

[TravisMack]

Dave:

I have been on pump tests where we had to run about 100' of hose and we could not get the pump to pass. After arguing enough with guys on site about what could be wrong, I disconnected the hoses so we only had 25' of fire hose and did the best we could to protect landscaping. Fire pump passed. So, the length of hose did seem to matter.

Travis Mack
MFP Design, LLC

http://www.mfpdesign.com

 

[FFP1]

The hose length obviously creates a problem IF the friction loss through the fire hoses prevents adequate flow rates (based on the fire pump rating); however, there is no arbitrary maximum length allowed. The properly trained inspector needs to be well versed and experienced enough to decide how many lengths of hose can be used and determine the best/safest setup for the flow test.

Dave: The concise answer is NO. The fire pump performance is based on the given flow rate. You are obtaining your flow test data pressure readings AT the suction and discharge sides of the fire pump.....the hose length could be a factor if you were obtaining your pressure readings at a location other than the fire pump (but this would be a mistake due to improper training or lack of experience).

 

[TravisMack]

FFP:

That was our problem. The friction loss was preventing adequate flow rates. It was a unique project to say the least. There were multiple problems with the existing installation. This was just one of the obvious ones.

Travis Mack
MFP Design, LLC

http://www.mfpdesign.com

 

[ContractorDave]

FFP1 (and all)

I really appreciate your opinions here but I'm still not clear on the curve. If I do an annual fire pump flow and I want to plot the performance curve, the whole purpose is to compare it to the original bench curve or the data plate. Using play pipes off a test header that is right at the pump will give you accurate readings. But if the pump is working harder to achieve net flow due to a given length of hose, the discharge psi is going to be higher at the pump and the curve is not going to match without somehow accounting for friction loss of the hose- correct? (As well, the rpms and voltage readings will be different).

This seems rather puzzling to me in that I can't possibly be the first to have asked this question or had this problem. This is why in the original question I mentioned that perhaps I am making an error of assumption as well as possibly methodology.

Regards
Dave

 

[NJ1]

The reason why I brought that issue is because I contracted a Fire Pump Expert once to do a fire pump test. The pump was in a windowless basement and the minute he saw the set up he stated that the pump test could not be done as per code. When I asked him he mention to me that in order for us to even accomplished any water outside we would need about 150' length and that it was not allowed by the standard.
I asked again where can we find in any book such comment. I even looked last night in the Fire Protection Handbook 9th Edition. I found nothing.
Then we go back again to NFPA 25. Do I tell the customer that a test header must be installed to satisfy NFPA 25 annual flow test or should I charged him for the test and fail him anyways? At that point you must likely not going to reach 100%

 

[FFP1]

Dave: You need to think through exactly what you are confirming during the fire pump flow test. This will help you realize and understand WHY the length of hose does not matter as long as you can maintain consistent flow rates at the 100% & 150% test points. Remember that you are determining the differentiual pressure between the suction & discharge sides of the fire pump (among several other test criteria for each test point which a large percentage of fire pump flow test inspectors fail to documant). Maybe you can find someone in your area to have a face-to-face conversation regarding the fire pump flow test procedure details.........I could help you graspe the issue at hand, but it will be too time consuming and akward using the blog entries.

NJ1: Charging for the test and NOT performing the fire pump flow test would be wrong (my personal opinion). In some instances a suitable test header is the best option. I have determined solutions by "thinking outside of the box". One example: I have performed fire pump flow tests using hoses and playpipes fed by multiple fire hydrants which were downstream of the fire pump.

 

[pipesnpumps]

Forgive me for saying, but the concept here is really very simple.. I think this is being mis-understood or over thought.

the procedure is basically:
1) Measure flow at the hose monster (via pitot pressure).
2) Measure both suction pressure and discharge pressure of the pump while it is flowing the desired flowrate.

Hose length doesn't matter. If the hose is so long it causes enough pressure drop that the flow cannot be met, then use two of the longer hoses or shorten up the one hose.. If needed, use three hoses, or more. I've seen 8 used..

The maybe-not-so-obvious fact here is: if you hit the desired FLOW out of the hose monster(s) then you will ALSO by default have the desired PRESSURE differential (aka TDH or total dynamic head) across the pump. This is because the pump follows its supply curve.

Using a shorter hose would mean the test valves would be closed more.. For example, lets say 150 ft of hose and valve is open 75%. Change to 25 ft of hose and you have to adjust valve to 40% open to get the same flowrate.. In both cases the pump differential pressure is 90 psig.

Lightecho, I hope this helps answer your questions. I know it is hard to "see" this concept through our words. I suggest learning a bit about pump curves until it sinks in.

The pump MUST follow its supply curve, unless the laws of physics breaks down. And physics are hard to get around. This means if you know the flowrate (out of the monster) then by default you know the differential pressure it should be at. If the pressure is not the same as the time before then you know there is a problem, such as impeller corrosion or a clearance issue.. Thus is the purpose of the test.

 

[NJ1]

FFP1

Dont you think that performing a fire pump test via fire hydrants ant other remote locations could interfere with the result of the test. Feel me on this for one second.

I go do a fire pump test where the test header is right outside the wall. Obviously I get all the required numbers since there is no friction loss.

I do another fire pump test where I am employing remote fire hydrants but yet pump fails due to lots of friction loss and remote discharge

 

[SprinklerDesigner2]

"This seems rather puzzling to me in that I can't possibly be the first to have asked this question or had this problem. This is why in the original question I mentioned that perhaps I am making an error of assumption as well as possibly methodology."

Doesn't matter, no hose or 5,000' of hose the readings will be the same. The gauges are located at the pump not at the playpipe.

As far as the example with 150' of hose, or 250' or whatever, the problem doesn't lay with the length of hose but poor design.

NFPA #20 Table 2-20 gives minimums not absolutes. For a 1,000 gpm pump the minimum number of hose valves on a hose header is 4 but nowhere does it say we couldn't have 6 or 8 if needed. The trick here is to know when more than the minimum specified will be needed.

Example:

Given the following conditions would a four valve test header with four hoses be adequate to test @ 150%?

Assume all elevations are equal.

A 1,000 gpm @ 60 psi fire pump taking suction from a city water main 40 psi static, 30 psi residual flowing 1,200 gpm flowing. Flow test indicates 25 psi @ 1,500 gpm is available.

The pump suction line from city water main to pump suction flange is a total equivalent length of 250’ of 8” Class 150 DR18 which would result in 3.3 psi head loss flowing 1,500 gpm and on top of this we have a backflow assembly that produces another 4.0 psi. When flowing 1,500 gpm we can expect to have 17.7 psi at the pump suction flange.

Flowing 1,500 gpm we can expect the pump to deliver maybe 75% of its rated pressure which would be 45 psi which would produce 62.7 psi (could be as little as 56.7 psi if the pump delivers 65% of its rated pressure at 150% of its rated capacity) and you can see right off we’re going to have a problem with four 100’ hoses.

With four hoses each hose must flow 375 gpm to test at 150%.

18 psi will be required at a 1 3/4" playpipe to discharge 375 gpm.

http://www.okolonafire.org/docs and sops/Friction Loss Chart 1.pdf

Looks to me like we can expect a friction loss of around 33 gpm through the a 100' length of 2 1/2" hose.

Now we have the friction loss from the fire pump discharge flange to the pump test header itself. With equivalent fitting lengths (120' of 6" would be typical) I think we can assume we'd lose at least 11 psi and to this add another 5 psi for the angle hose valve.

18+33+11+5=67 psi.

We don't have 67 psi and if the pump performs at 65% we're not even close! Four hoses with 1 3/4" tips and fully opened hose valves and you're not going to get 1,500 gpm flowing.

Now imagine the playpipes at elevation 20' (8.6 psi) with 150' of hose and (another 17 psi on top of the 8.6) and it becomes laughable.

What you need to do is use five hoses

With five hoses and playpipes you need 300 gpm or a 12 psi pitot instead of 18psi and head loss through 100' of hose would be 21 psi instead of 33. Now you got a shot of making at least and if you want to make sure increase the discharge and hose header pipe to 8" using hose gate valves instead of angle valves and you'll probably make it.

Just because NFPA 20 says the minimum number of hoses required is 4 doesn't guarantee it will be adequate.

Go with 10 hoses discharging through a 1 1/4" tip. 150 gpm is 11 psi pitot and 150 gpm through 100' of 2 1/2" hose is 6 psi... change the hose header to 8" and you probably won't have a problem testing with 300' of hose.

That's my two cents worth anyway.

 

[FFP1]

I want to make sure Dave (Lightecho) knows that I have a strong desire help and I think his posts on this blog are insightful as well as pertinent. Dave obviously cares and has a strong desire to perform quality inspection/testing services. I am NOT trying to make anyone feel bad (in any way shape or form), but I will probably be judged or attacked for this post anyway:

I will simply state that most of the water-based fire protection system inspectors (including NICET III & NICET IV certified) that I have met over the past 20+ years are not properly trained to conduct fire pump flow tests or preaction/deluge system inspections. The inspectors are NOT to blame in most instances.........the Certificate of Competency holder and/or the company these inspectors work for send them out to perform inspection and testing activities prior to providing adequate training and ensuring they have the required experience. The inspectors have great intentions and they put their best effort towards the tasks, but they are simply not properly trained.

This blog is extremely helpful for our industry.......my concern is that inspection/testing activities for fire pumps, preaction systems and deluge systems are so much more involved than many of the other water-based fire protection systems. I would suggest formal training and more time trailing experienced/qualified inspectors to make sure lead inspectors are well versed and properly trained prior to assigning these more complicated/involved inspections. Being NICET III or IV certified means this person passed a written test, but does not equate to being properly trained and experienced.......more employees need to press their employers to provide formal training for areas of need.

Comments posted by SD2 & pipesnpumps are very helpful regarding this topic and their comments suggest they have a solid understanding that some others more than likely do not have to date. I am a little concerned when I witness an inspector who knows how to complete the steps 1,2,3 but does not really understand all of the important aspects of the system he/she is testing. In summary, I do not think it is possible to learn everything one needs to know about completing fire pump flow testing via an internet blog.

Please be kind if you decide to respond; I truly have good intentions and I hope to have a positive impact with this post.........just maybe one person out there will be motivated to pursue formal training or be humble enough to admit they would benefit from trailing a more experienced inspector in an attempt to become more informed!

For the record, I just hesitated prior to selecting the "Submit Post" button.

 

[SprinklerDesigner2]

FFP1,

We continue to see the same faces.

By virtue of wanting to spend the time it's readily apparent that everyone cares deeply about the industry and the performance of their job. Everyone here wants not just to do their best but to be the best.

But our industry has always been a "learn as you go" and given the number of systems and inspectors at any certification level there's simply to much work. It may not be the best but it's better for a system to be looked at by someone with 2 years experience than not looked at at all. In 5 years they'll know more than they did at 2 and at 20 they'll know more than they did at 10.

I've been laying out systems for 35 years and I'll never know it all but when you think about it aren't we all lucky to be in an industry where you can learn something new every day? I consider myself most fortunate that when I started the personal computer and computerized calculations was ten years away. For me it was all hand calculations but it served me well. I remember spending two weeks proving three K-Mart grids and it wasn't "drawing time" it was a worksheet, a pencil, a pocket calculator and tables you found in the back of Vikings data book. Where I rested the pencil on my middle finger had a callous you wouldn't believe. A good third of my time wasn't spent drawing on a drafting table it was dong hand calcs.

Picture this. It was 1979, Rubbermaid Corporation was building a new 200,000 sq. ft. warehouse 68' to the bottom of the roof steel with one of the first automated rack systems... the overhead was loops, rack sprinklers galore for rubber products... face sprinklers.... a Factory Mutual Job and I didn't have a computer not even a spread sheet. You learned some estimating tricks in pretty short order.

The point I am making is it took me years to gain the understanding I think I have. Years.

It is unfair to expect a Level II or III inspector to understand the calculations he could be called on to do as in performing a pump test. Those of us that have the experience owe it to our inspectors to teach them and it won't be quick because we didn't do it quick either... we took years.

But all the certified inspectors I've met are smart... they'll get it with our help and we owe them our help as others helped us along the way.

It is all time well spent and everyone in the company will benefit along the way. Do it, teach and mentor.

The inspector gig isn't easy. It's 95 degrees and humid in south Georgia right now and an inspector that spent a day inspecting a paper mill worked a whole lot harder than any of us layout technicians ever did. I've done it and it about kicks my butt.

 

[NJ1]

FFP1, SprinklerDesigner2

[SprinklerDesigner2]
You made this comment earlier-
Doesn't matter, no hose or 5,000' of hose the readings will be the same. The gauges are located at the pump not at the playpipe.

The readings are taken at the end of the hoses not at the pump. (I am not challenging your comment)

I just got back from Atlanta, Ga doing some consulting and I have to tell you that in my opinion there are some great inspectors out there NICET Level III however there is a
very limited amount of people that can provide such training.

The individual that I mention earlier is one of the most respected individuals performing fire pump test in the Tri-State area.

I do lots of NICET training and expert opinion. I also perform lots of fire pump test and I can ensure you that the more hoses you use the less pressure at the pitot you will have especially if you have to turn the hoses thru corridors and stair cases.

 

[ContractorDave]

FPP1

I have always appreciated your thought out comments. Please continue to press the submit post button. There's a fair list of people I have learned a lot from. SD2, Travis, Stookey and many more make this forum a valuable and interesting resource. You guys rock.

Dave