Fire pump testing - how might one do the attached example with a PRV? 3

[ContractorDave]

Hello

I'm doing ITMs on a mine site. There's a fire pump module / trailer with two 2000 gpm pumps fed from tank supplies. I drew up a rather corny looking iso with one system represented. There's a flow meter test loop on the system but there's also a PRV loop that does not have isolation valves. The PRV's are the Watts 116FM. Attached is the iso and hopefully I can attach the pdf I found on line for the PRV but it sure doesn't tell you much.

Does anyone have any thoughts on how to proceed under these circumstances? How might you lock down that PRV so it doesn't affect your flow rates or how might you otherwise account for it?

Thx
Dave

 

[NJ1]

Very simple lightecho

I will remove them and plug off the outlet in order to proceed with the test. You will have to add some extra labor to do so.

 

[NJ1]

I apologize but I spoke tooo soon without looking at the attachments
This could get complicated. Thinking of it you may have to remove it and install filler piece of pipe during the test. After test is completed factor in the pressure lost from the PRV's to the pump curve.

 

[Sprinklerfitter]

New here and perhaps I don't understand the question, but the pressure relief valves would have been designed into the original pump specifications.
Are these pumps diesel? If so see 4.18.1.1 NFPA 20 2011"Where a diesel engine fire pump is installed and where a total of 121 percent of the net rated shutoff (churn)pressure plus the maximum static suction pressure, adjusted for elevation, exceeds the pressure for which the system components are rated, a pressure relief valve shall be installed."
As per the shutoff valves please see 4.18.9 NFPA 20 2011 Shutoff Valve. A shutoff valve shall not be installed in the relief valve supply or discharge piping.
In my opinion don't worry about the PRV's, test pumps as normal.

 

[ContractorDave]

Ah yes, Sprinklerfitter - but I must worry about the PRV, for something is amiss! The rating on the fire pump is 117.5 @ churn, 104@ 2000 gpm, and 83.5 @ 3000 gpm. Actual test results (and I use Ashcroft 1082's, good for .25% accuracy and calibrated annually): With a tank supply of 19.5 psi.
111 @ churn
104 required to see 2000 and
46! required to see 3000. Unless the impeller is totally snafu-ed, I have a different problem. The only other possible culprit would seem to be the PRV.

Now this particular fire pump is a diesel, but it's part of a fire pump-house package that has an electric fire pump with similar specs tied into the same test header - and it too has it's own PRV (same one) that bypasses the test header. Though the electric fire pump flow results were not nearly as bad, they were still far enough off as to cause concern.

So these PRV's. They are Watts 116FM 'fire pump relief valves'. I had little luck finding documentation on them and my Watts dealer is still trying to source some more information from the manufacturer for me. But they are 6" devices so will obviously relieve a lot of volume (gpm) as well as pressure and this probably accounts for the fire pump having to run all the way down to 46 psi to finally squeak 3000 gpm through the leg that has the flow meter on it. The rest of the pressure must be going through the PRV and back to the tank. I see no other explanation. And I'm not certain as to why it's been installed this way then unless we are experiencing failures of the PRV's themselves and normally this wouldn't happen (probably the case but again I don't have the installation and maintenance sheets for these yet so I am not going to dick with them until then).

I did manage to get an opinion from "somebody" at Watts (the technical expert was in a meeting) who suggested that I just tighten the adjusting screw right down for the test and then put it back to it's original settings after ..... this does not seem quite right to me.

Regards
Dave

 

[LCREP]

The Watts guy is correct, back down on the nut that operates the valve and take it out of service. Take your pump readings and then reset it to open at a higher pressure say 175 psi, it looks like it is set way to low. See NFPA 25, 2011 below

8.3.3.3* For installations having a pressure relief valve, the
operation of the relief valve shall be closely observed during
each flow condition to determine whether the pump discharge
pressure exceeds the normal operating pressure of the
system components.

A.8.3.3.3 A pressure relief valve that opens during a flow condition
is discharging water that is not measured by the recording
device(s). It can be necessary to temporarily close the pressure
relief valve to achieve favorable pump test results. At the
conclusion of the pump test, the pressure relief valve must be
readjusted to relieve pressures in excess of the normal operating
pressure of the system components.
If the pressure relief valve is open during the flowing conditions
due to the fact that the pressure is too high for the
components in the fire protection system, the discharge control
valve should be closed prior to closing the pressure relief
valve to make sure that the fire protection system is not overpressurized.
After the test, the valve must be opened again.

8.3.3.3.1* The pressure relief valve shall also be observed during
each flow condition to determine whether the pressure
relief valve closes at the proper pressure.

A.8.3.3.3.1 A pressure relief valve that is open during a flow
condition will affect test results.

8.3.3.3.2 The pressure relief valve shall be closed during flow
conditions if necessary to achieve minimum rated characteristics
for the pump and reset to normal position at the conclusion
of the pump test

****************************************
Fire Sprinklers Save Firefighters’ Lives Too!

 

[ContractorDave]

I keep coming back here and little wonder. What a great site with some great people, or as the side ad says in brief; 'What a wonderful resource'.

Thanks NJ and Sprinklerfitter. But one star to LCREP (again).

Regards
Dave

 

[LCREP]

See the link below on page 15 for

FIGURE PV20CB
DIRECT RELIEF CONTROL

Look for START-UP/ADJUSTMENT bottom of the page. Look at the diagram to the right. Look for "ADJUSTMENT SCREW" and "NUT
ADJUSTMENT SCREW" Loosen the "NUT ADJUSTMENT SCREW" then move the
ADJUSTMENT SCREW"

1 - Turn the screw Clockwise (IN)
to increase the set-point you are
controlling.
2 - Turn the screw Counterclockwise
(OUT) to decrease the set-point you
are controlling.

****************************************
Fire Sprinklers Save Firefighters’ Lives Too!

 

[LCREP]

Dave,

Glad I could help, miss adjusted relief valves is a too common problem and so simple to adjust IF you know how to do it, now you know how!

****************************************
Fire Sprinklers Save Firefighters’ Lives Too!

 

[Sprinklerfitter]

Good information here, posts like this are why I joined!

As a person who has also seen many issues with pressure relief valves I wish to clarify my statement with regards to not worrying about these devices during a performance test:

When performing an annual pump performance test should you as the inspector have to adjust or close off a pressure relief valve to perform a pump test or would their affect on said test be considered a deficiency? will you or someone else be present to do the same during a fire? What did the original pump acceptance test indicate, and what if anything has changed since then to reflect this change?

I feel that this type of problem is outside the scope of a performance test and should be addressed under a service contract, my reasoning for this includes the following concerns:

Who specified these PRV's? Why do they not discharge directly to the atmosphere as recommended by NFPA 20? Was the flow of 3000 GPM by these devices considered when specified? (Issue addressed in NFPA 20 2011 A.4.18.7) Is this large flow of water by these devices as indicated on the included ISO creating a vacuum on the dischage side of the PRV affecting it's performance? perhaps this is why the issue is most reflected at the larger flows, it would also be an indication as to why the electric pump is not affected as dramatically as I would assume that it is not flowing as much water.

Sorry for the long post I only wish to vent my concerns with what seems to be a growing problem with improperly designed/specified systems were the inspector must bare the burden of responsibility.

 

[ContractorDave]

Hello Sprinklerfitter

Long posts are good if there's meat to them.

LCREP addresses one of your comments; NFPA 25 does indeed specify in 8.3 that you may have to deal with the presence of a PRV while doing fire pump flow testing. The purpose of this flow test is to make a comparison to the manufacturers benchmarks to ensure there has been no degradation in performance and therefore the PRV needs to be taken out of the equation. I agree with you though that in a fire scenario, if the PRV is going to effect the performance of the pump, I'm wondering how the PRV was taken into account when spec'ing the pump?

In regards to PRV on the electric pump, I'm not sure why it's even there.
I don't believe it needs to be there, therefore shouldn't be as it's just something else that can fail.

As for the water from the PRV going back to the tanks and not discharging to 'atmosphere', they'd have to be piped up the side of a 30' tank some 75' away in another building across cold weather so there were probably some exceptions made. And being in an environmentally sensitive area we're not allowed to dump it back into a lake or open pit.

Regards
Dave

 

[LCREP]

Dave,

See NFPA 25, 2011 ED below regarding meeting the sprinkler demands. This is new for the 2011 ed.

8.3.5.7 The pump performance shall be evaluated using the
unadjusted flow rates and pressures to ensure the pump can
supply the system demand as supplied by the owner.

So you would need to calculate the pump discharge pressure and flow to determine IF it is meeting the "system demand". So IF the PRV is opening too soon, OR staying on too long and is NOT meeting the sprinkler demand, then you need to adjust the PRV.

The PRV was added to the diesel fire pump in case the overspeed electronics does not work. The need for a PRV has been in and out of NFPA 20 at least once that I know of. As to why it is on the electric pump perhaps someone did not have a clue as to why it was needed and just added it to BOTH pumps. I would set the electric pump PRV at 175 psi and move on.

Discharging the PDV back to the tank is done all the time to save water.

Hope this helps.

Do you know how to plot a combined pump curve? If not just ask.

Tom

****************************************
Fire Sprinklers Save Firefighters’ Lives Too!

 

[ContractorDave]

LCREP

From what I understand, the intention of 8.3.5.7 is to specifically address the practice of adjusting flow rates in regards to affinity laws and motor rpms so that the test curve better matches the manufacturers bench curve. I believe this was addressed in Russell Flemings recent NFSA on-line seminar in regards to the changes in 25 / 2011.

The rest of your post makes perfect sense.

I know how 'I' plot curves. But nothing interests me more than seeing how others do it. It's amazing how much one can learn by seeing something done from anothers point of view.

So yes! Please do.

Much appreciated

Regards
Dave

 

[Sprinklerfitter]

Good info Gentleman,

Still my concern lies with the original installation. The 2010 fire pump handbook is a great source to clarify some of the code requirements, here's one that applies here:

NFPA 2010 20HB "4.18.1.2* Pressure relief valves shall be used only where specifically permitted by this standard. Paragraph 4.18.1.2 states that main relief valves should only be used where specifically permitted by this standard. No section in this standard specifically permits the use of a main pressure relief valve. The use of a main pressure relief valve on an electrical fire pump is considered to be poor design and should be avoided. Several methods are available to cope with excessive pressures such as the following:

1. A break tank (see Section 4.31)
2. A variable speed pressure-limiting control device (see11.2.4.3)
3. Other pressure-regulating devices downstream of the fire pump

While the above applies mostly to the electric pump with regards to the diesel I also question the need for the PRV, if we follow the rules set forth in NFPA for calculating when a PRV is required I come up with the following based on information from an earlier post:

Max churn 117.5x1.21(overspeed)=142.175PSI
Static pressure (which if tank levels are properly monitored should be fairly constant)19.5PSI
Total maximum system pressure during an overspeed condition should still be only 161.675PSI or well within system ratings.

If the test results submitted are net pump head pressures then it would seem that the PRV's are opening the most at approx 65.5 PSI (46+19.5)it would be expected in normal operating conditions that they would only open at the greater points on the curve, in this case I,m not sure that closing the valves will help

Discharging the water back to the tank is definetly common practice, but must also be done with care as the turbulance/air pockets created can introduce more problems into the system

Thanks again

Jamey

 

[ContractorDave]

Nice bit of information yourself Jamey. I agree with you in regards to both pumps.

I have that Hand book. I should be reading it more obviously but as I grow older my interests are getting broader and I find it difficult to squeeze everything in. A friend of mine just convinced me to go for my pilots license. That'll side track me for a while!

Know one ever learns everything by themselves of course. And that's why this forum has been so valuable to me. Throw a question out there and if it's interesting, you'll get all sorts of intelligent answers and every now and then a fun debate to boot.

Regards
Dave

 

[LCREP]

Dave,

I do not agree with Russ, the purpose of 8.3.5.7 was to make sure with the pump operating it can meet the sprinkler/standpipe system demand for the building. No use having a fire pump operating and it can not meet the demand. We sometimes see fire protection demands not being met because of water supplies that deteriorate over time, closed valve(s), RPZ failure, pump performance issues, etc. The insurance industry has done a combined curve at the annual fire pump tests for many years. This section was put into NFPA 25 to insure the building has an adequate water supply with the fire pump operating.

Take a look at NFPA 25, 2011, page 77 in Annex A figure A8.3.4.3(1)(b) bottom of the page. Note the clear circle at the right bottom of the graph "system design#2(if available). As you can see the intent of the NFPA 25 committee was to compare the performance of the pump to the system demand.

Regarding how to plot a fire pump test and a combined graph for folks who many not know how to see this web site:

http://www.inspector911.com/checklist/firepump/index.html

look for "Reference Materials" then click on the link "how to graph fire pump test results"

Hope this helps.

Tom

****************************************
Fire Sprinklers Save Firefighters’ Lives Too!

 

[ContractorDave]

Interesting Tom, and thanks very much for the link.
Your argument is sound and comes with the pr-requisite citations.

This begs a question though: a situation I run into almost daily are systems where you don't know the system demand. In most cases one would be able to 'assume' certain things and look this up in say 13 / table 11.2.3.1.1 - but we all know that assumptions in our line of work has an inherent amount of risk tied to it.

Taking my current situation, I am doing inspections at a mine site. There is no hydraulic data plate. Although there is a fair amount of documentation on site for the sprinkler systems, the client is unable to find anything for the Process Plant. The fire pump package I have been talking about in this post feed the process plant.

I can eliminate the PRVs from the equation to come up with a flow test that says whether or not the fire pump performance is matching factory curves. I'm not sure if I agree with an inspector having to confirm that the results meet system demand. The original curve had to meet system demand after all. If an inspector verifies this (or finds otherwise), why is the onus being put on the inspector to do what should have been done to begin with?

Isn't this similar to situations where part of a sprinkler system has been incorrectly installed - and hence it is beyond the scope of NFPA 25?

I haven't perused the full document you link to above and may change my mind after reading it or after further conversations here, but not at the moment.

Regards
Dave

 

[LCREP]

Dave,

8.3.5.7 The pump performance shall be evaluated using the unadjusted flow rates and pressures to ensure the pump can supply the system demand as supplied by the owner.

Key words "as supplied by the owner", if it is not supplied , then you can not do it!

We do it, insurance point of view, by calculating the system to determine the demand when the hydraulic placard is not available.

You are not required as long as the "owner" does not have it.



****************************************
Fire Sprinklers Save Firefighters’ Lives Too!

 

[ContractorDave]

Tom

Ok, after reading the combined curve article here are my thoughts. Attached is a partial view of my curve sheet for reference.

I input my MFG spec's and the last flow test results if available into my test sheet (my page 5) along with the required peripheral data such as voltages, rpms, water temp, etc depending on type of pump - but I also input the suction pressure and it automatically deducts this from the results and plots it on page 6 for me. As in my previous post I am comparing the fire pump performance to the MFG data plate. I have to assume that the correct fire pump was chosen regardless of the water supplies. It's the main drain test that is used (or traditionally has been used) to keep tabs on the water supplies if supplied by municipal sources for example. And of course a problem with water supplies in a tank fed system would show up as a problem in the test flow results.

Now at one time, NFPA 25 had been a little more specific in saying that 25 applies to systems that were properly installed in accordance with the applicable standards. Over the last two issues (2008 and 2011) they have leaned away from this specific wording but I believe the intent is still there. 2002 was more clear in this regard in Chapter 1 section 1.1 Scope where it read in part "Where a system has not been installed in accordance with generally accepted practices, the corrective action is beyond the scope of this standard...."

This means any such problem is not a deficiency as per NFPA 25, but a deficiency as per the applicable installation standard. In doing inspections, I am following NFPA 25 to the letter because I am not a Fire Protection Engineer. In saying this, I will most certainly point out to the client if I see something which appears to be an issue in regards to the installation or water supplies, but only in-so-far as that I recommend they have an engineer assess the situation.

Hence, why would I use the combined curve? Does it not speak more to the design of the system than the performance of the fire pump I am inspecting?

Regards
Dave

 

[ContractorDave]

Hi Tom

Would you mind my posting over at I911 in regards to this and using part of your comments above? I'll link you to the new topic.

Dave