How to Address Fire Pump with Poor Supply?

1/20/2026

We have a disagreement on a fire pump installation. Flow test shows 42 psi static, 10 psi residual at 475 gpm from an 8" municipal main.

The fire sprinkler contractor plotted the water supply curve and says it cannot support the specified 1,000 gpm @ 100 psi pump; they're refusing to certify the fire pump without a suction tank or another resolution.

The mechanical engineer argues the 8" main at 42 psi static should theoretically deliver way more than 1,000 gpm and questions whether the flow test results accurately represent what the pump will actually see at its suction connection.

The steep pressure drop (42 psi to 10 psi in only 475 gpm) seems abnormal for an 8" main. The area has aging infrastructure and known low-pressure issues.

Looking for the community's input on: (1) Is the fire sprinkler contractor correct to base their analysis on residual pressure under flow conditions, or should we be looking at static pressure as the mechanical engineer suggests?

(2) What's the correct NFPA 20 methodology for evaluating water supply adequacy for a fire pump?

(3) When you see such a steep pressure drop in a flow test, what are the typical causes and how would you approach resolving this?

What solutions have worked for others dealing with marginal municipal water supplies?

Thanks in advance.

Sent in anonymously for discussion. Click Title to View | Submit Your Question | Subscribe

17 Comments

Paul PInigis link

1/20/2026 08:14:53 am

1. There is a serious problem with the public water supply. Either an obstruction in the lines or a partially-closed valve.
2. The water supply (as depicted by the flow test) cannot support a fire pump.
3. The mechanical engineer is just plain wrong. The pipe diameter and static pressure do not provide the information required to assess the available water supply.
4. The required approach is to do a flow test and determine the characteristics of the available water supply and then incorporate that into the design.
5, You can add a water storage tank or find and fix the problem with the public water distribution system.

1/20/2026 08:19:07 am

Yikes.

NFPA 1, 2018 Chapter 4 "Flow Testing" §4.1.3 states "It is generally recommended that a minimum residual pressure of 20 psi should be maintained at hydrants when delivering fire flow. Fire department pumpers can be operated where hydrant pressures are less, but with difficulty."

Most of the jurisdictions in our area require a minimum public water supply residual pressure of 20psi (and I know of 1 that allows 10psi) for fire flow and to prevent cavitation which is a public health concern.

475 gpm @ 10psi is clearly not going to provide 1000gpm at either 20 psi or 10 psi (as a simple supply curve would indicate) and a tank would be required for this project.

Save yourself a ton of issues down the road and spend the money to hire a licensed professional fire protection engineer or delegate the design to a contract with a NICET certified engineering technician that knows what they're doing.

1/20/2026 08:25:29 am

These are the correct sections:

NFPA 1, 2018 §3.3.123 Fire Flow "The flow rate of a water supply, measured at 20psi residual pressure, that is available for fire fighting."

NFPA 291, 2016 Chapter 4 "Flow Testing" §4.1.3 states "It is generally recommended that a minimum residual pressure of 20 psi should be maintained at hydrants when delivering fire flow. Fire department pumpers can be operated where hydrant pressures are less, but with difficulty."

Jonathan Griffin

1/20/2026 08:19:09 am

The water flow test that you describe has real issues: the Minimum flow required for firefighter use is based upon a 20-PSI residual pressure, and usually more than 1,000 GPM.

A Fire Pump will only pull all of the available water from the municipal water supply and affect the community by depleting the water supply momentarily.

There are 2 components at play here. Usually, we are only concerned about augmenting available residual pressure to supply a riser system, and water has available GPM flow available (an 8" main should be delivering quite a bit here).

I suspect either a blockage, a water leak or MIC/corrosion choking the pipeline. Also, an intermediate control valve could be partially shut (Roadway box somewhere on the line).

Art G

1/20/2026 08:19:59 am

To see that much of a drop in an 8" main would be very rare but possible.

I've been designing sprinklers for nearly 50 years now and I don't think I have ever encountered an 8" main with the results of a flow test that low.

I would run another flow test with certified gauges and plenty of witnesses around. and if it produces the same result then a tank is the only way to go.

Some states, Georgia for example, require a minimum residual pressure of 20 psi at the pump suction when flowing 150% of rated capacity.

1/20/2026 08:20:01 am

Firstly, static pressure alone does almost nothing for us here. If the city has booster pumps we may see static pressure of 42 psi and a drop in residual pressure that is barely noticeable without opening many hydrants to as nfpa 291 suggests. Residual pressure and flow is the way that we are able to plot a supply curve- static pressure only is not going to help us diagnose this issue.
Secondly, nfpa 20 states that an acceptance test of a pump is adequate if the pump meets 100% of the demand of the most demanding system. If the system demand is 1000+gpm, the contractor is right that this will not happen. Nfpa 20 requires positive suction head and you will not even come close to maintaining that based on this flow test.
Thirdly, for such a steep drop I personally would perform another flow test and personally verify that these numbers are accurate. If they are, contact the municipality and ask why it's so bad. If these are the real numbers and are verified, the only immediate option i know of is a tank.

1/20/2026 08:23:12 am

Also want to add (just for fun) that it's scary the engineer says static is all that matters. As others have said, 'yikes'!

Derrick

1/20/2026 08:23:40 am

Unless you have another flow test that shows different...you have to have a pump and tank. If it only delivers 485 gpm @ 10 psi. It would never be able to deliver 1,000 gpm at a positive pressure to your building without additional capacity. I don't see how you can base the flow on static pressure only.

Glenn Berger

1/20/2026 08:26:21 am

Lots of questions presented -

1) I agree with the sprinkler contractor that a tank will be required. The mechanical engineer needs to be educated on the design of fire sprinkler systems.

2) You need to ensure that at least 100% of the pump flow rating is available at 20 psi. Yo do really need 150% of the pump flow rating to properly test the pump and to verify the sizing of the fire pump based upon system demands.

3) You need to understand the water system infrastructure to evaluate. Also obtain historical data to see if the test is an anomaly. Steep pressure drops do occur.

Anthony

1/20/2026 08:33:45 am

How to Address Fire Pump with Poor Supply? -- with money.

Given that flow test I can only see a tank in your future probably a break tank? This is a very poor flow test if you need 1000 gpm at 100 psi for your system. Best of luck!

Dan Wilder link

1/20/2026 08:39:31 am

Not to step on toes, but using a static pressure to determine water supply is wrong for fire protection, your contractor is correct in their efforts. Works great with plumbing fixtures...has no place in fire protection at the contractor level (IMO) specifically for this reason.

Our pump suppliers require a copy of the flow test as a good faith measure in identifying potential negative suction issues. The fact that the residual dropped to 10PSI already violates several AWWA and likely Municipal public health requirements if this is a true capture of the water supply. The pump doesn't understand (typically) to not go negative and since this supply is 1/2 of the capacity and 1/3 of the 150% for testing (yes, I understand there are some exceptions), this will cavitate the pump and possibly collapse the incoming water lines, or at a minimum start pulling out gaskets.

I would ask the water department for all water supply tests in the area for as far back as they have records. I would also look into other sprinkler systems in the area for any water supply info from the AHJ. This would establish a history of the water supply and identify if the test was the anomaly or if there is a larger issue.

Request if the water department is doing any work, service, maintenance, or if another contractor is working in the area to identify possible closed valves. Bring out a street key or request the water department exercise the street valves. You can count the number of turns opening/closing an OS&Y to see if they were left partially closed/opened (8"=27, 10"=33, 12"=39 - see below link)

As for resolutions, assuming the flow test is accurate and the pump is sized correctly and the hydraulics require this, a tank is the best course of action.

We have a couple areas in my neck of the woods with water supplies from the 1940-1950 era. Previous projects completed by us and others that have exceeded the water supply have received water tanks, sometimes break tanks sometimes stand alone. Not much you can do without a large-scale municipal upgrade of the system as a whole.

Operation - Bottom of Page 3 - This is pretty typical for most OS&Y valves.
https://www.muellercompany.com/sites/muellercompany.com/files/var/www/mueller-sites/public_html/sites/muellercompany.com/files/uploads/media/12661RWGVOperManRev.pdf

Stephen P Merrill

1/21/2026 11:28:58 am

Always- Always require a hydraulic gradient to be done prior to suggesting a pump or it size. The must be done in recently not results for 3 years ago.
Anyone that just uses Static Pressure to determine this needs to rethink it.

Dave Branson

1/20/2026 08:53:27 am

I have a similar situation on a private campus, with my flow test yielding 44 psig static, 27 psig residual and 650 gpm (15 psig) flowing. My remote load plus hose allowance is approx. 800 gpm for 90 minutes. I'm looking at a 91 kgal water storage tank and a 1,000 gpm fire pump to handle the remote load. The existing main should have enough pressure to refill the tank within 8 hours.

Jack G

1/20/2026 09:04:53 am

I think the water supply should be evaluated by performing other flow tests to determine if there really is an obstruction in the incoming line to the fire pump.
Also is it a dead end main?
Has anyone used a camera ( like a sewer camera ) to see if it is a dead end main clogged with sediment ?
What if the water supply is really bad ?
Is there an elevation change to the building?
I once performed a fix to a plant that the Prior contractor used a flow test half a Mile away and the elevation loss was 70 psi.
My first priority is what happens to the community when the pump sucks the main dry and cavitation occurs? I ll bet that could happen with the description above.
After evaluating , and found to be a bad supply, I would agree with the contractor.
As tanks take time, what do you do to protect the community ?
Take the pump off line—— post signage?
Provide an orifice plate on the discharge to control the plow to 20 psi ! ( temporarily )
A complete evaluation of the water supply is in order.

Dwight Havens

1/20/2026 12:19:34 pm

Good discussion. The available water supply needs to be fully evaluated and it's configuration understood. I have seen sprinkler systems designed to the available flow from the two nearest hydrants, only to find that the 6-inch main that completed the the top of the grid between the two hydrants was corroded to less than 3-inches in diameter. Problem was found during acceptance testing using the Main Drain for flow. In other words, the assumptions made from the flow test and the piping connecting the two hydrants did not reflect the realty of the reduction in diameter of the 6-inch main to which the system was initially connected. Of course, in this case, the reality of the main configuration was unknown until the investigation into the cause of the poor Main Drain test reveled it.

Ed Kramer

1/21/2026 10:24:33 am

Great questions and discussion.

Get all stakeholders together and conduct another flow test. If the initial flow test data is verified, you flat-out don't have the water supply for the pump. Maybe it's easily fixable (partially closed valves) or not (long dead-end main, main with an effective inside diameter MUCH less than what it should be due to sediment/corrosion). Either way, your options are to fix the water supply or install a storage tank.

Your sprinkler contractor is correct. And I'm not out to bash engineers, but his particular one is wrong about this particular thing - flow test residual pressure and flow are critical pieces of data.

Jose Antonio Garcia

1/21/2026 11:09:26 am

I believe the key element that hasn’t been mentioned in the comments is this: What is the actual demand of the sprinkler system?
In my opinion, that is the crucial question. For example, if the system demand is 500 GPM (or less), the results from the hydrant flow test will allow the installed fire pump to deliver that flow rate without the suction pressure dropping to 0 psi (or negative).
On the other hand, if the demand exceeds 600–650 GPM, the municipal main supply would not be sufficient, and the pump would not be able to deliver its rated capacity without suction problems.
So, my main question would be: What is the calculated hydraulic demand of the sprinkler system (in GPM)?

How to Address Fire Pump with Poor Supply?

Leave a Reply.

ALL-ACCESS

SUBSCRIBE

COMMUNITY

YOUR POST

PE EXAM

FILTERS

ARCHIVES

PE PREP SERIES