A centrifugal fire pump's performance is qualified by three points on its curve: the churn point at no flow, the rated point at 100% of rated flow, and the overload point at 150% of rated flow. NFPA 20, Standard for the Installation of Stationary Pumps for Fire Protection, defines acceptance limits at each of those three points. The shape of a compliant curve must stay inside the envelope those limits define, not hit them exactly. Designing or reviewing a pump submittal without understanding that distinction — limits versus outputs — leads to over-pressured systems at the top end and under-pressured systems at the bottom.
The data used for every comparison is net pump pressure: the pressure the pump adds between its suction and discharge flanges. Raw discharge gauge pressure includes suction pressure and cannot be used directly. Conflating the two is the most common measurement error in fire pump curve checks, and it runs in the dangerous direction: a raw discharge reading appears higher than net pump pressure, making a failing rated-point look acceptable on paper.
The NFPA 20 Three-Point Acceptance Envelope
NFPA 20 Section 4.26 defines the performance criteria for centrifugal fire pumps around three standard operating points. At churn, with the discharge valve closed and flow equal to zero, the pump's net pressure must not exceed 140% of its rated pressure. At the rated point, 100% of rated flow, the pump must develop at least its rated net pressure. At the overload point, 150% of rated flow, the pump must hold at least 65% of its rated net pressure.
The hazard at each point runs in a specific direction. Churn pressure too high means the system is overpressured at shutoff; downstream piping, fittings, valves, and sprinkler heads see maximum static pressure at zero flow, and a churn above 140% of rated can exceed equipment pressure ratings. Rated-flow pressure too low means the pump does not deliver what is promised to the hydraulic calculation. Overload pressure too low means the pump cannot sustain flow when demand surges during a fire event, collapsing pressure at the most critical moment.
These percentages are limits on the shape of the curve, not design targets. A pump that produces exactly 140% at churn and exactly 65% at 150% flow is at the edge of compliance. Real pump curves sit inside the envelope, and a manufacturer submitting a curve that bumps against both limits on the same pump should be reviewed carefully. The NFPA 20 field acceptance test is the formal verification; the screening envelope is the first check.
Net Pump Pressure: The Only Acceptable Input
Net pump pressure is the difference between total discharge pressure and total suction pressure, measured at the pump flanges:
P_net = P_discharge_gauge + (elevation_discharge − elevation_suction) × unit_weight + P_velocity_head_discharge − P_suction_gauge
For most installations where gauge elevations are close, velocity heads are small, and suction pressure is available from a suction gauge reading, the simplified form applies:
P_net = P_discharge_gauge − P_suction_gauge
A pump with 120 psi on the discharge gauge and 30 psi on the suction gauge delivers 90 psi net. Comparing 120 psi to a 100 psi rated requirement would produce a false pass of 1.20 versus the 1.00 limit. Comparing the correct 90 psi produces a ratio of 0.90, which fails the rated-point requirement. That is not a subtle difference. Per NFPA 20 Section 14.2, all curve verification uses net pump pressure; the witnessed field acceptance test uses calibrated pressure gauges at both suction and discharge flanges with gauge elevation correction.
Head and pressure convert at the water-density basis used in fire protection calculations:
head (ft) = pressure (psi) × 2.307
head (m) = pressure (bar) × 10.197
These conversions allow the curve criteria to be applied in either pressure or head without change to the ratios; the percentages are dimensionless.
Acceptance Envelope Formulas
The three acceptance limits derive directly from the rated point:
P_churn_max = P_rated × 1.40 (maximum churn — ceiling at zero flow)
P_rated_req = P_rated × 1.00 (rated-point requirement)
Q_overload = Q_rated × 1.50 (overload flow — 150% of rated)
P_overload_min = P_rated × 0.65 (minimum at overload — floor)
Checking a measured curve means computing the ratio of each measured net pressure to the rated pressure and comparing to the limit:
churn ratio = P_churn_measured / P_rated → pass if ≤ 1.40
rated ratio = P_rated_measured / P_rated → pass if ≥ 1.00
overload ratio = P_overload_measured / P_rated → pass if ≥ 0.65
All three ratios must pass for the curve to comply with NFPA 20 Section 4.26 criteria. A pump that passes churn and overload but fails at the rated point does not comply, and only checking all three points reveals that failure.
Worked Example: 1,000 gpm / 100 psi Pump (US and Metric)
A listed centrifugal fire pump is rated at 1,000 gpm and 100 psi per UL 448. The witnessed NFPA 20 field acceptance test records the following net pump pressures: 130 psi at churn, 102 psi at rated flow (1,000 gpm), and 70 psi at 1,500 gpm.
Acceptance envelope:
| Point | Formula | Value |
|---|---|---|
| Max churn | 100 × 1.40 | 140 psi |
| Rated requirement | 100 × 1.00 | 100 psi |
| Overload flow | 1,000 × 1.50 | 1,500 gpm |
| Min at overload | 100 × 0.65 | 65 psi |
Curve check:
| Point | Measured | Ratio | Limit | Verdict | Margin |
|---|---|---|---|---|---|
| Churn | 130 psi | 1.30 | ≤ 1.40 | Pass | 10 psi below ceiling |
| Rated flow | 102 psi | 1.02 | ≥ 1.00 | Pass | 2 psi above requirement |
| 150% flow | 70 psi | 0.70 | ≥ 0.65 | Pass | 5 psi above floor |
All three points pass. The governing point is the rated flow with the smallest margin, 2 psi above the requirement.
Metric equivalent (same pump: 3,785 L/min, 6.90 bar):
| Point | Formula | Value |
|---|---|---|
| Max churn | 6.90 × 1.40 | 9.66 bar |
| Min at overload | 6.90 × 0.65 | 4.49 bar |
| Overload flow | 3,785 × 1.50 | 5,678 L/min |
Measured net pressures converted: 8.96 bar churn, 7.03 bar rated, 4.83 bar at 150% flow. Ratios 1.30, 1.02, 0.70 are identical; the dimensionless ratios are unit-independent.
Where the Curve Check Fails and What It Reveals
Four failure modes account for most fire pump curve problems in design review and acceptance testing.
Rated-point failure hidden by passing churn and overload. A pump is rated at 1,000 gpm and 100 psi. The test records: 130 psi churn (1.30, pass), 95 psi at rated flow (0.95, fail), 70 psi at 1,500 gpm (0.70, pass). Churn and overload both pass; the rated point misses by 5 psi. NFPA 20 Section 4.26 requires all three points. A check limited to two of the three points is not a valid curve verification.
Raw discharge used as net pressure. The discharge gauge reads 120 psi at rated flow. Suction pressure at that moment is 28 psi. Net pressure is 120 − 28 = 92 psi, a ratio of 0.92 against the 1.00 requirement. The raw gauge reading of 120 psi appeared healthy at 1.20; only the net reading reveals the failure. This error is more likely when suction conditions are favorable and the suction pressure is a large share of the total discharge pressure.
Churn failure at high suction pressure. On a pump fed from a pressurized city main at 80 psi, the suction-side pressure is high. The pump adds 140% of rated at churn, but the total system pressure at churn — suction plus net churn — can exceed the pressure rating of downstream components. NFPA 20 Section 5.2 and Annex A address the total system pressure consideration; the curve check here verifies only the net churn pressure against 140% of rated.
Speed mismatch between certified curve and test. The certified curve is at 1,760 rpm. The test runs at 1,700 rpm, a 3.4% speed reduction. Per the pump affinity laws, head scales with speed squared: the 100 psi rated pressure at 1,760 rpm becomes approximately 93 psi at 1,700 rpm. Comparing test pressures at 1,700 rpm to the 100 psi rated requirement from the 1,760 rpm curve without speed correction applies the wrong reference. NFPA 20 Section 14.2.7 requires that test speed be within ±1% of the rated speed for the acceptance test to be valid, or that readings be corrected to rated speed before comparison.
Application Boundaries
The three-point NFPA 20 check is a curve-shape verification, not a system sizing tool. Several conditions fall outside its scope.
This method checks the pump curve against the NFPA 20 acceptance envelope. It does not verify that the pump meets the specific system demand, which depends on the hydraulic calculation per NFPA 13, NFPA 14, or NFPA 15 as applicable. A pump that passes all three curve points can still be undersized for the system's required flow and residual pressure.
It does not assess suction conditions. NPSH available versus required, submergence, suction lift, pipe losses, and vortexing all affect whether the pump will operate on its published curve in the field. A pump that passes curve verification can still cavitate under the actual installation suction conditions, per ANSI/HI 9.6.1 for rotodynamic pump NPSH.
It does not evaluate the driver. Motor sizing per NEC 430 or diesel engine selection per NFPA 20 Chapter 12 governs whether the driver can sustain rated pump speed through the overload point. A pump that passes the curve check with the right driver can still stall or trip if the motor is undersized for the 150% flow power requirement, which is the pump's peak power demand per the standard centrifugal pump power curve shape.
The accepted NFPA 20 edition must be verified with the authority having jurisdiction. The percentage criteria (140%, 100%, 65%) are the values in widely adopted editions; verify that the AHJ-adopted edition matches before using any screening result in a design document or test report.
Fire Pump Performance Curve Calculator
Build the NFPA 20 acceptance envelope from rated flow and pressure (Define mode), or verify measured churn, rated-flow, and 150%-flow net pressures against the envelope (Check mode). Per-point ratios, margins, and governing-point identification in US (gpm, psi or ft) and Metric (L/min, bar or m).
Open Fire Pump Performance Curve CalculatorFAQ
What does a fire pump performance curve show?
It shows the net pressure a centrifugal fire pump produces across its flow range, from churn at no flow to the overload point at 150% of rated flow. NFPA 20 Section 4.26 standardizes the acceptable shape with three limits: churn no higher than 140% of rated pressure, rated pressure at rated flow, and at least 65% of rated pressure at 150% flow. The curve tells a designer or inspector what the pump actually delivers at each operating condition.
What is churn pressure on a fire pump and why does NFPA 20 cap it?
Churn pressure is the net pump pressure at zero flow, with the discharge valve closed. It is the maximum pressure anywhere on the pump's curve. NFPA 20 Section 4.26 caps it at 140% of rated pressure because downstream sprinkler piping, fittings, and valves see maximum static pressure at churn; a pump producing churn above 140% of rated can overpressure system components rated for the expected working pressure range. A churn ratio above 1.40 requires a review of the system pressure rating and the need for a pressure-relief valve.
Why must a fire pump deliver 65% of rated pressure at 150% flow?
A fire event can demand more water than the pump's rated flow, particularly in a large system with multiple sprinkler branches open. NFPA 20 sets the floor at 65% of rated pressure at 1.5 times rated flow so the pump continues to deliver useful pressure under high-demand conditions. If pressure falls below 65% of rated at that flow, the pump cannot serve the overload demand and the selection must be reconsidered. The 65% floor is not a performance target; it is the minimum that NFPA 20 permits.
Are 140% and 65% the pressures a fire pump produces?
No, and treating them as outputs rather than limits is the most common pump selection error. They are the ceiling and floor of the acceptance envelope. A pump that produces exactly 140% at churn and exactly 65% at 150% flow is at the compliance boundary; the actual curve from a well-selected pump sits comfortably inside the envelope. Assuming the pump produces exactly 140% at churn and selecting system components for exactly that pressure leaves no margin. The actual curve values come from the manufacturer's certified pump data sheet and UL 448 listing test results.
Does the three-point curve check replace the NFPA 20 field acceptance test?
No. The curve check is a screening and design-review tool for evaluating a proposed or certified curve before installation. The acceptance test that counts is the witnessed NFPA 20 field acceptance test, conducted at the installed pump with calibrated instruments, witnessed by the authority having jurisdiction, and evaluated against the certified curve and the adopted NFPA 20 edition. Per NFPA 20 Section 14.2, the field acceptance test flows the pump at churn, rated flow, and 150% of rated flow under actual installation conditions. A screening result is not a record document.
Related Calculators
Expansion tank sizing for closed-loop fire protection and standpipe systems: Expansion Tank Sizing Calculator.
Compressed air pipe sizing for pneumatic suppression system supply lines and standpipe air maintenance device supply: Compressed Air Pipe Sizing Calculator.
Condensate return line sizing for steam systems in buildings with fire suppression wet pipe systems above steam heating areas: Condensate Return Line Sizing Calculator.
Drain field sizing for pump room sump drains and testing discharge in fire pump rooms per NFPA 20 Section 5.13: Drain Field Sizing Calculator.