On a data sheet the K-factor looks like one line of a catalogue entry. In practice it is the decision that sets the pump head, the tank volume and the pipe sizes. If the wrong K is chosen and nobody notices at design stage, it comes back on site either as extra cost or as inadequate protection. This note walks through the arithmetic, the typical failure patterns and a worked example.
The core relationship: Q = K√P
Sprinkler flow equals the K-factor multiplied by the square root of the pressure at the head. In SI units Q is in litres per minute, P in bar and K in L/min/bar0.5. Rearranged, the required pressure is P = (Q/K)². Because the relationship is squared, a modest change in K produces a large change in pressure: increasing K by 44% roughly halves the pressure needed for the same flow.
Common K-factors and their US equivalents: K80 ≈ K5.6, K115 ≈ K8.0, K160 ≈ K11.2, K200 ≈ K14.0, K240 ≈ K16.8, K360 ≈ K25.2 (the conversion factor is about 14.4). Unit confusion is a failure mode in its own right: entering a US K5.6 into an SI calculation as 5.6 makes the result meaningless. For a quick sanity check use the fire protection unit converter.
Worked example: one density, three K-factors
Take a design density of 12.2 mm/min (0.30 gpm/ft²) and 9 m² of coverage per sprinkler. The minimum flow from the most remote sprinkler is:
Q = 12.2 × 9 ≈ 110 L/min
| K (L/min/bar0.5) | Required end-head pressure P = (Q/K)² | Comment |
|---|---|---|
| K80 | (110/80)² ≈ 1.89 bar (27.4 psi) | Pump and pipework sized for the higher pressure |
| K115 | (110/115)² ≈ 0.91 bar (13.3 psi) | End-head pressure almost halves |
| K160 | (110/160)² ≈ 0.47 bar (6.9 psi) | The standard's minimum end-head pressure governs; flow at that pressure is slightly above what is needed |
The difference between K80 and K115 is about 1 bar at the remote end. With a 232 m² (2,500 ft²) design area, roughly 26 sprinklers operate and the sprinkler demand is at least about 2,900 L/min (770 gpm). An extra 1 bar at that flow is around 4.7 kW of hydraulic power; at 70% pump efficiency that is roughly 7 kW at the driver and can push the selection up a pump or motor size. With a diesel driver the difference carries through to fuel consumption and fuel tank volume as well.
Note: These figures illustrate the method. The minimum end-head pressure, design area and density depend on the standard you apply (NFPA 13, EN 12845, FM Global) and on the edition in force. Confirm them for your project against the standard and with the authority having jurisdiction.
Pattern 1: K115 in the calculation, K80 on site
The most common case is a procurement substitution of a head with a different K described as "equivalent". If the calculation was run with K115 and the pump chosen accordingly, there is 0.91 bar at the remote end. At that pressure a K80 head delivers only 80 × √0.91 ≈ 76 L/min, which over 9 m² is about 8.5 mm/min. A system running about 30% below its design density will not be caught at acceptance, because sprinklers are not opened during testing. It shows up only in a fire or in a careful as-built review.
Pattern 2: K80 in the calculation, K115 on site
The reverse happens on the basis that "a bigger K is safer", and it is the more insidious problem. If the calculation used K80, the system is sized to deliver 1.89 bar at the remote end. At that pressure a K115 head discharges 115 × √1.89 ≈ 158 L/min, 44% more water per head. Overdischarge grows at heads close to the riser, total flow rises, the operating point slides right along the pump curve and pressure falls. The pressure reaching the most remote head can end up below the calculated value.
The tank calculation breaks as well. For a 60-minute duration, 2,900 L/min needs about 174 m³; had every head delivered 158 L/min, demand would exceed 4,100 L/min and the tank would need more than 247 m³. In reality the supply cannot sustain that flow, but that simply means the tank will run dry sooner than calculated. Check tank volumes with the fire water tank sizing calculator.
Pattern 3: K mismatch with ESFR and CMSA
For storage sprinklers the K-factor is the design. ESFR and CMSA design rests on tables and product listings that tie together K-factor, ceiling height, storage height and commodity. As a result:
- Changing K changes the design pressure. Using K360 instead of K200 at the same pressure raises flow by 360/200 = 1.8, an 80% increase. As ESFR design is usually based on a set number of the most demanding heads operating together, total demand rises in the same proportion.
- Not every K is listed for every ceiling height. A K accepted at one height may not be permitted under a taller roof. Screen the case with the ESFR suitability checker, then confirm against the product listing.
- No mixing of technologies in one area. ESFR is not mixed with CMSA or standard spray heads within the same design area; response times and discharge characteristics differ.
- Spares must match the K. Heads in the spare sprinkler cabinet must match the installed heads in K-factor, temperature rating and response type.
The differences between the two technologies are set out in ESFR vs CMSA, and the FM Global side in the DS 8-9 ESFR notes.
Pattern 4: Change of use and retrofits
When space designed as offices becomes storage, the required density rises. At 20 mm/min over 9 m², each head must deliver 180 L/min. That needs (180/80)² ≈ 5.1 bar with K80, about 1.27 bar with K160 and about 0.81 bar with K200. Pushing an existing K80 network beyond 5 bar at the end head is rarely practical, and moving to a larger K sets off a chain of changes:
- Larger-K heads usually have a larger thread size, so fittings and nipples change.
- Flows go up, so branch lines and mains are recalculated and often upsized.
- Layout and coverage are rechecked against the storage rules; some storage designs restrict small-K heads anyway.
- Water supply, pump and tank are re-verified for the new demand.
For that reason "just swap the heads" is seldom enough after a change of use. The NFPA 13 hazard classification tool is a good starting point for re-establishing the hazard.
Checklist for choosing the right K
| Check | Question |
|---|---|
| Density and area | Was end-head flow calculated as Q = density × coverage area? |
| Minimum pressure | Does P = (Q/K)² fall below the standard's minimum end-head pressure? If so, was flow recalculated at the minimum pressure? |
| Units | Have SI and US K-factors been kept apart? |
| Listing | Is the head listed for this hazard, ceiling height and storage arrangement? |
| Procurement | Does the K on the approved materials list match the head delivered to site? |
| As-built | Has every head change been fed back into the hydraulic calculation? |
| Spares cabinet | Do spare heads match the system in K, temperature and response? |
Pressure–flow tables by K-factor are on the K80 and K115 pages, and the selection logic is covered in sprinkler K-factor selection.
Frequently Asked Questions
Is a larger K-factor always the safer choice?
No. A larger K discharges more water at the same pressure. If the calculation used a smaller K, that overdischarge raises total demand, shifts the pump operating point and can lower pressure at the most remote head. Any K change must be fed back into the hydraulic calculation.
How can I tell whether a different K was installed on site?
Check the sprinkler identification number (SIN) and manufacturer code on the frame against the catalogue. In an as-built review, compare the SINs on site with the approved materials list for each area.
Which units should I use in Q = K√P?
In SI, Q is in L/min, P in bar and K in L/min/bar^0.5. In US units Q is in gpm and P in psi, and the K value is about 14.4 times smaller. Never mix the two in one calculation.
Can I replace an ESFR head with a larger-K ESFR?
Only if the new head is listed for that ceiling height, storage arrangement and commodity, and the hydraulic calculation, pump and tank have been re-verified for the new demand. At the same pressure, going from K200 to K360 raises flow by about 80%.
SprinkCalc — Fire Sprinkler Design Across Three Standards
SprinkCalc covers hazard classification, design density and area, K-factor selection, water demand and hydraulic calculations for NFPA 13, FM Global and BS EN 12845 in a single iOS app, and exports a professional PDF report.
Download SprinkCalc on the App StoreNFPA 13 (current edition) · BS EN 12845:2015+A1:2019 · FM Global DS 8-9, DS 2-0 · Manufacturer listing and catalogue data. The numerical examples illustrate the method; confirm project values against the standard in force and with the authority having jurisdiction. The findings here are typical defect patterns, not an account of events at any particular site.