Data-Center Clean Agent: NFPA 2001 Novec, FM-200 and Room Integrity

Data-center server halls (400–2,000 m³) are not spaces you want to protect with water. NFPA 2001 defines the clean-agent options: FM-200 (HFC-227ea), Novec 1230 (FK-5-1-12), IG-541 (Inergen). Key requirements: the right design concentration, the 10-minute hold time and a passed door fan integrity test.

Clean-Agent Comparison

The Class A and Class C minimum design concentrations below are from NFPA 2001-2025 Table A.7.2.2.3(b) (at 21 °C); GWP values are from Table A.4.6. If the manufacturer's listed value is higher, it governs.

• FM-200 (HFC-227ea): 6.7% Class A, 7.0% Class C; zero ODP but GWP 3,350
• Novec 1230 (FK-5-1-12): 4.5% Class A and C; GWP <1, about 5-day atmospheric life
• IG-541 (Inergen): 34.2% Class A, 38.5% Class C (oxygen reduction), natural gases
• CO2: 34% — lethal to humans; not covered by NFPA 2001 (NFPA 12); not used in occupied spaces
• Choice: new projects favor Novec 1230

NFPA 2001 Design

• Concentration: observe the NOAEL/LOAEL limits in occupied spaces (Table 4.3.2.4(a))
• Discharge time: 10 s maximum for halocarbons (7.5.1.1); for inert gases 60 s for Class B fuels and 120 s for Class A surface fires and Class C (7.5.1.2) — a server room is Class A/C
• Hold time: at least 85% of the minimum design concentration held at the top of the protected contents for 10 min, or long enough for trained personnel to respond (7.4.1)
• Agent quantity: W = (V/s)·C/(100−C) (7.3.1); at 20 °C about 0.55 kg/m³ for FM-200 at 7.0% and about 0.66 kg/m³ for Novec 1230 at 4.5%
• Overpressure vent: required where discharge pressures would threaten the structural strength of the enclosure (7.1.8.1)

Room Integrity (Door Fan Test)

• Test: ISO 14520 and NFPA 2001-2025 Annex D (Enclosure Integrity Procedure)
• Door fan: leakage measured at a typical 10 Pa pressure difference
• Leakage area: ELA (equivalent leakage area) calculation
• Hold-time simulation: concentration decay curve; must not fall below the limit at protected height for 10 minutes
• Smoke pencil check: cable trays, panel bases and subfloor voids

Alternatives and Control Logic

• High-pressure water mist: main alternative (~15 micron droplets)
• VESDA: very-early smoke detection
• Oxygen reduction (ReduOx): permanent 15% O2 atmosphere as preventive protection
• Interlock matrix: HVAC dampers, UPS shutdown and door control
• Manual release: two-person confirmation and mechanical lock

Maintenance and Weight Monitoring

• Inspection: agent quantity and container pressure checked at least semiannually (NFPA 2001-2025 11.3)
• Acceptance limit: a halocarbon container losing more than 5% of agent, or more than 10% pressure (temperature-adjusted), is refilled or replaced (11.3.1); for inert gas the limit is 5% pressure loss (11.3.4). Load cells make continuous weight monitoring easy.
• Recharge: containers sent for refilling within 48 hours of discharge; keep a spare set
• Container retest: per the transport regulations (DOT in the US); a container past its requalification date must not be recharged without retesting (11.6.1); for halocarbon containers the retest may be a complete visual inspection (11.6.1.1)
• Evacuation: pre-discharge alarm and time delay sufficient for evacuation (9.7.1; typically 20 s in practice)

MEP Calc — 110+ Engineering Calculators

MEP Calc bundles 110+ engineering modules in one iOS app: 21 fire calculations (sprinkler, fire pump, FM-200 clean agent) plus heating, cooling, HVAC, plumbing, steam and natural gas. Download it on the App Store.

Download MEP Calc on the App Store

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. One of the few tools that unifies all three major fire protection standards.

Download SprinkCalc on the App Store