Skip to content
From Fire Safety Academy Wiki
Revision as of 16:09, 9 November 2026 by Peter (talk | contribs) (Correct errors identified in the editorial review of 9 November 2026)
Contents
  1. 1Definition and engine types
  2. 2Applications
  3. 3Fire hazards
  4. 4Installation safeguards
  5. 5Fire protection
  6. 6Estimating a fuel spill fire
  7. 7Fire experience and risk data
  8. 8See also
  9. 9Notes
  10. 10References

A stationary combustion engine is an engine, such as a reciprocating engine or a gas turbine, that is installed in one location as permanent equipment to drive a generator, pump or compressor.[1] The most common use is a generator set that produces electricity for normal or emergency service, while larger engines drive pumps and compressors in the petrochemical and gas pipeline industries.[1] Such an engine brings fuel, lubricating oil, hot exhaust parts and arcing electrical parts into one room, and in rare cases the engine itself can disintegrate.[2] In the United States, NFPA 37, Standard for the Installation and Use of Stationary Combustion Engines and Gas Turbines, and local codes cover their installation, as summarised in the 2008 handbook.[2] The typical safeguards are separation from combustible construction, controlled fuel supplies with spill containment, automatic shutdown devices and, for larger installations, automatic fire suppression.[3][4] The likely size of a fuel spill fire can be estimated with the pool fire methods used in fire safety engineering, and the result depends strongly on whether the spill is confined.[5]

Definition and engine types

Biggins (2008) defines stationary combustion engines as prime movers that are installed for use in a location as permanent equipment. The term covers internal combustion engines, external combustion engines, gas turbine engines, rotary engines and free piston engines.[1] A prime mover is the machine whose mechanical shaft output drives, for example, a generator.[6] Many codes and standards also treat a portable engine as stationary when it stays connected in the same location for one week or more.[1]

The engines fall into two main groups: reciprocating engines and gas turbine engines.[1] Reciprocating engines range from small portable gasoline engines to very large diesel engines for power generation. In an Otto cycle engine a spark plug ignites a fuel and air mixture. In a diesel cycle engine the air is compressed until its temperature reaches the ignition temperature of the injected fuel oil.[1][2] The hazards of all reciprocating engines are similar.[2]

Gas turbine engines are usually larger, with stationary units ranging from about 75 to 75,000 kW (100 to 100,000 hp).[2] The first gas turbine patent dates from 1884, but turbines became economically viable only around 1940.[2] In an open cycle turbine the working fluid passes through the engine once, and in a closed cycle turbine it is recycled continuously. Most gas turbines in use when the handbook was written were of the open cycle design.[2]

Applications

The most frequent application in buildings is the engine-driven generator for emergency and standby power. Such a system consists of a prime mover, a governor that keeps the speed nearly constant, a generator, transfer equipment, and distribution and protective equipment.[7] The prime mover can be a reciprocating engine fuelled by gasoline, diesel fuel, or natural or synthetic gas. Gas turbines are also used, mainly for larger outputs.[7] Chisholm (2008) notes that gasoline and gas engines cost less than diesel engines, but the gasoline engine and its fuel store are more susceptible to fire and explosion.[7][6] A natural gas engine also depends on an off-site utility supply that may fail together with the normal power, so the authority having jurisdiction may require an on-site fuel supply.[6]

For emergency service in the United States, the National Electrical Code (NEC), as summarised by Chisholm (2008), requires the prime mover to start automatically and reach full speed and power within 10 seconds. It also requires an on-site fuel supply for at least 2 hours of full output.[6] Water-cooled engines, generally those above 50 kW, may not be cooled with the municipal water supply, and a starting battery needs an automatic charger that is independent of the prime mover.[6] Gas turbines are much smaller than reciprocating engines of the same output and need no cooling water. They are, however, slow to reach full speed (more than 10 seconds). For such a prime mover the NEC, as summarised by Chisholm (2008), permits an auxiliary power supply to carry the load within 10 seconds, provided it can do so until the prime mover takes over.[6]

Large stationary engines are also central to power station fire protection. In US power plants they are usually called combustion turbines, and units of 10 to 40 MW were traditionally used for peak demand. Most of this peaking capacity was installed in the decade after the 1965 Northeast blackout.[8] In 1999 there were 2366 combustion turbine units in operation in the United States, a 17 percent increase over 1998.[9]

Fire hazards

Biggins (2008) lists five fire hazards of stationary combustion engines. The first is the combustible liquids and flammable gases used as fuel, hydraulic fluid and lubricant, and in some cases the material being pumped or compressed. The second is the temperature of exposed engine parts, particularly exhaust manifolds and pipes. The third is arcing of electrical components such as starters, generators, distributors and magnetos, which can ignite flammable vapours. The fourth is combustible material stored in the engine room or used in the installation, such as air filters. The fifth is engine disintegration.[2]

Engine drawing with five numbered callouts for fuel and oil, hot exhaust, electrical arcing, stored combustibles and disintegration
The five fire hazards of a stationary combustion engine.

Hot exhaust parts are a direct ignition source. Exhaust manifolds, mufflers and exhaust piping can exceed 538 °C (1000 °F), and fires have started where too little clearance was left between exhaust piping and combustible walls, ceilings or roofs.[3] Combustible dust and flyings that settle on exhaust parts can also ignite, but only when conditions allow prolonged contact with the hot surface. Good housekeeping lowers this chance.[2]

The liquids in and around the engine form the second group of hazards. Gasoline is the most hazardous common liquid fuel because of its low flashpoint of −43 °C (−45 °F).[3] Lubricating and hydraulic systems generally use combustible oil, although some larger gas turbines use a synthetic fire-resistant fluid with a higher flashpoint, which reduces the chance that a spill ignites on hot engine parts.[3] In large reciprocating engines an oil mist is often present in the crankcase. If it ignites, a combustion explosion can occur inside the crankcase, so many such engines have crankcase explosion vents, and in other cases the crankcase is kept under a nonflammable atmosphere.[3]

Backfires through gas and air mixers and carburettor air intakes result from engine malfunction and can ignite grease and oil deposits on and around the engine.[4] Total mechanical failure of a reciprocating engine is very rare. Gas turbines are at greater risk because of their very high rotational speeds, and stationary turbines are usually built to contain the compressor, turbine blades and rotor parts if they fail.[3][4] A diesel engine in a hazardous location has a further hazard: if flammable vapour reaches the air intake, the engine can keep "running on" with the surrounding atmosphere as its fuel. For that reason such an engine should be stopped by isolating both the fuel and the combustion air supplies.[4]

Installation safeguards

In the United States, NFPA 37 and local codes give minimum requirements for engine installation, engine room compartmentation, fuel supply piping, room ventilation and engine exhaust systems, as described in the 2008 handbook.[2] Engines may be located outdoors, inside structures or on roofs. The choice of location should consider the distance from combustible walls and ceilings, openings in exterior walls and the supply of air for combustion and ventilation. Engine room compartmentation should follow the applicable standards and local codes, which may set minimum fire resistance ratings.[2] Combustible materials should not be stored in engine rooms or in the compartments of a gas turbine, and engines should be isolated from places where combustible dust or flyings are produced.[2]

Exhaust systems are designed for the expected flue gas temperature and pressure. They need protection against damage from the ignition of unburned fuel, low points with drainage for condensate and, where vibration or thermal expansion could break the pipe, a flexible connector at the engine.[3] Biggins (2008) gives a minimum clearance of 229 mm (9 in.) between the exhaust system and combustible materials. A wall thimble or similar measure is needed where the exhaust passes through combustible walls or roofs, and the exhaust should end outside at a point where hot gases and sparks cannot reach combustible material.[3]

Schematic of an emergency engine generator room with fuel tanks, exhaust, transfer switch and ventilation openings
Typical engine generator installation for emergency power.

The electrical installation should comply with NFPA 70, the National Electrical Code. An engine room is not classified as a hazardous location solely because of the engine fuel. Where flammable liquids or gases are pumped or compressed, arcing engine components are safeguarded by flashover shielding, flame arresters, purging or ventilation.[3] Flame arresting equipment is also placed at gas and air mixers and carburettor air intakes in such installations, to stop backfires from igniting the surroundings.[4]

Liquid fuel installations follow the principles of a fuel oil system, with diesel fuel and gasoline as the most widely used fuels. NFPA 37, as summarised in the 2008 handbook, limits gasoline in buildings to an engine-mounted tank of at most 95 L (25 gal). Where more fuel is needed indoors, a diesel engine is generally used.[3] Fuel is supplied by pumps, except for gravity feed from an engine-mounted tank. Tanks filled by pumps should have an overflow line, a high-level alarm and a high-level automatic shutoff, and diking or other spill containment should be considered for tanks inside structures or above ground outdoors.[3]

In World Trade Center Building 7 (WTC 7), New York City regulations allowed one day tank of up to 1.04 m³ (275 gal) per floor. One tenant therefore used a pressurised fuel distribution system instead of a second day tank.[10]

Gas-fuelled engines are installed in accordance with codes such as NFPA 54, National Fuel Gas Code, and NFPA 58, Liquefied Petroleum Gas Code. Propane is heavier than air and collects in low areas, so the supply and exhaust openings of the ventilation need care.[3] A gas pressure regulator inside a structure should vent to outdoors at least 1.5 m (5 ft) from the structure, or have a listed vent limiting device. The gas piping should have a shutoff valve remote from the engine, preferably outside the building. Each engine should also shut off its gas supply automatically if it stops for any reason.[3]

Automatic speed governors control the disintegration hazard, and an overspeed shutdown device should be fitted to large engines above 75 kW (100 hp).[4] Engines above 7.5 kW (10 hp) should shut down on high jacket water temperature, high cylinder temperature or low lubricating oil temperature, as the handbook prints it, and gas turbines also on high exhaust temperature.[4] The starting sequence of a gas turbine includes a purge cycle long enough to leave a nonflammable atmosphere in the turbine and exhaust system before ignition.[4] Larger installations should have a remote emergency shutdown of the engine and of fuel and lubricating oil pumps that the engine does not drive. Inlet air filters on larger gas turbines should, where possible, be made of materials that do not burn freely.[4]

Instructions for normal starting, stopping, operation and routine maintenance should be posted near the equipment, and emergency shutdown procedures should be developed and kept readily accessible. These procedures typically cover the remote shutdown stations, the shutdown of the lubricating and hydraulic oil pumps and the closing of the fuel shutoff valve. Where fuel valves are not clearly identified, a diagram of their location is posted near the engine, and operating staff must be trained in these procedures.[4]

Fire protection

Portable fire extinguishers are provided in accordance with the applicable standards. Larger installations, especially those with an appreciable amount of lubricating or hydraulic oil or with liquid fuel, should have automatic fire suppression. Sprinkler systems are the usual protection for reciprocating engines inside structures.[4] For gas turbines the options listed by Biggins (2008) are preaction water spray, water mist, and gaseous carbon dioxide and clean agent systems, with dry chemical sometimes used for exhaust compartments. Interlocking the suppression system with automatic closing of the fuel valves helps to limit the size of a fire.[4]

Combustion turbines are usually bought as a package, and the seller rather than the buyer normally chooses the suppression system. According to Sheridan (2008), most were initially protected with carbon dioxide or halon, and more recent installations with FM-200. Fear of thermal shock was the main reason for avoiding water.[9] Thermal shock here means rapid cooling of the hot turbine casing, which contracts and can damage the rotating blades.[11] An Electric Power Research Institute study of 24 events in which water reached hot turbine-generator parts found no significant damage, and no cases of thermal shock had been reported since.[8]

Water mist has since been developed for turbine enclosures. Mawhinney (2008) describes a turbine enclosure as a machinery space in which the fire source is limited to the turbine and its fuel lines. Fuel line breaks can cause spray fires, three-dimensional spill fires and pool fires in the bilge under the turbine.[11] The FM Global test protocol includes a plate cooling test, and its results feed a finite element analysis of the clearance between casing and blades. Listings for turbine enclosures up to 260 m³ had been obtained by 2008.[11]

Water storage is based on the FM Global minimum of 10 minutes or the NFPA 750 requirement for two complete discharges (20 minutes) for pre-engineered systems.[11][12] The FM Global approval standard, Class 5560 (May 2005), has separate fire test appendices for combustion turbines up to 80 m³ (2825 ft³), up to 260 m³ (9175 ft³) and above 260 m³.[13]

The SFPE Handbook adds that a machinery room with a total compartment system may have so high a ceiling that specific equipment must also be surrounded by local application water mist nozzles. Examples are diesel engines, flammable liquid or vapour compressors and large turbines on an open floor.[14] Pre-engineered systems are tested in a mock-up of a turbine or diesel engine enclosure with set ceiling height, obstructions, ventilation openings and fire scenarios.[15]

Estimating a fuel spill fire

A fuel release is one of the main fire scenarios for an engine installation, and its likely size can be estimated with pool fire methods. For a liquid pool, the maximum heat release rate is the heat release rate per unit area of the fuel multiplied by the burning surface area:[5]

Q˙m=q˙″A

Symbol Description Unit
Q˙m Maximum heat release rate of the liquid fuel fire kW
q˙″ Heat release rate per unit area of the fuel kW/m²
A Exposed surface area of the pool or spill m²

The value of q˙″ depends on the fuel and on confinement. For diesel, Budnick et al. (2008) give 1950 kW/m² for a confined pool and 390 kW/m² for an unconfined spill.[5] An unconfined spill spreads freely and burns at a lower rate per unit area, about one fifth of the confined value. On concrete the unconfined depth is about 2.8 mm or less, and deeper pools may be treated as confined.[5] The duration of the fire can be found from the fuel mass and heat of combustion, or as the pool depth divided by the regression rate, the speed at which the liquid surface falls.[5]

Budnick et al. (2008) work through a utility room with a backup diesel generator. Its 400 L fuel tank stands in a curbed area of 1.4 m × 1.4 m, 0.21 m deep. The curb has an area of about 2.0 m², and a 0.2 m deep pool fire is assumed after a pipe or tank rupture.[16] With the confined value, the fire reaches 1950 kW/m² × 2.0 m² = 3900 kW, or about 3.9 MW.[17] A plywood workbench 1.3 m from the curb would ignite after about 186 seconds. The sprinklers, spaced 4.5 m × 4.5 m under a 5 m ceiling, were estimated to operate after 48 seconds. Fire spread beyond the pool was therefore judged unlikely.[17][18]

Time line from 0 to 200 s with sprinkler activation at 48 s and plywood ignition at 186 s
Time line of the diesel generator room example: sprinkler activation at 48 s, plywood ignition at 186 s.

The same numbers show why spill containment matters. If the full 400 L spread unconfined over concrete at 2.8 mm depth, it would cover about 143 m². At 390 kW/m² this gives about 56 MW, roughly fourteen times the confined fire, unless the ventilation of the room limits the fire first.[5] For the duration, the National Institute of Standards and Technology (NIST) used a surface regression rate of about 4 mm/min for a large open pool of diesel fuel.[19] At that rate the 0.2 m deep curbed pool would burn for about 50 minutes, a rough estimate because the rate was given for a larger pool.[19]

Bar chart of heat release rate for a 400 L diesel spill: 3.9 MW confined in a 2 m2 curb versus about 56 MW unconfined
Heat release rate of a 400 L diesel spill, confined versus unconfined.

A US nuclear plant example gives a smaller case. In NUREG-1805, 11.4 L (3 gal) of diesel fuel spills into a diked area of 0.093 m² (1 ft²) in a standby diesel generator room. The fire gives about 186 kW for 42 minutes, with flames 1.4 to 1.5 m high.[20] Both examples assume that the whole pool is involved at once, without a growth phase, and the NUREG-1805 case also assumes burning in the open.[20][16]

Fire experience and risk data

Fire risk analysis of US nuclear power plants treats diesel generators as a separate ignition source group. NUREG/CR-6850 gives a generic fire frequency of 2.1 × 10−2 per reactor year for the diesel generators of a unit. Of these fires, 84 percent are classed as oil fires and 16 percent as electrical fires.[21] Each diesel generator is counted together with its subsystems, such as the air start compressors, air receiver, batteries and fuel storage and delivery system.[22] No frequency was estimated for gas turbine emergency generators, because too little was known about how many were in use.[23]

The WTC 7 investigation gave a large-scale example of engine fuel systems in a high-rise building. Its 5th floor housed two 900 kW diesel generators for the building and nine more generators for a tenant, supplied from large tanks below the loading dock.[24][10] NIST simulated a diesel fire fed by a ruptured fuel line and found that the gases above the nine water-cooled generators would exceed their operating limit of about 100 °C within an hour. Without the generators the fuel pumps would have lost power after an hour or two. The fire could then not have lasted the seven hours between the collapses of WTC 1 and WTC 7.[19] In this scenario a fire near the engines would have disabled the emergency power they were installed to provide.[19]

The handbook requirements describe US practice around 2007 and 2008, and the edition of NFPA 37 is not named. Fuel cells, which produce electricity without combustion, share the handbook chapter but are treated under fuel cell.[1]

See also

Notes

  1. ↑ 1.0 1.1 1.2 1.3 1.4 1.5 1.6 Biggins (2008), p. 8-105.
  2. ↑ 2.00 2.01 2.02 2.03 2.04 2.05 2.06 2.07 2.08 2.09 2.10 2.11 Biggins (2008), p. 8-106.
  3. ↑ 3.00 3.01 3.02 3.03 3.04 3.05 3.06 3.07 3.08 3.09 3.10 3.11 3.12 Biggins (2008), p. 8-107.
  4. ↑ 4.00 4.01 4.02 4.03 4.04 4.05 4.06 4.07 4.08 4.09 4.10 4.11 Biggins (2008), p. 8-108.
  5. ↑ 5.0 5.1 5.2 5.3 5.4 5.5 Budnick et al. (2008), p. 3-149.
  6. ↑ 6.0 6.1 6.2 6.3 6.4 6.5 Chisholm (2008), p. 10-6.
  7. ↑ 7.0 7.1 7.2 Chisholm (2008), p. 10-5.
  8. ↑ 8.0 8.1 Sheridan (2008), p. 9-185.
  9. ↑ 9.0 9.1 Sheridan (2008), p. 9-186.
  10. ↑ 10.0 10.1 McAllister et al. (2008), p. 61. The report converts 275 gal to 1.25 m³, which corresponds to imperial gallons; 275 US gal is 1.04 m³.
  11. ↑ 11.0 11.1 11.2 11.3 Mawhinney (2008), p. 16-137.
  12. ↑ Mawhinney (2008), p. 16-138.
  13. ↑ Mawhinney (2008), p. 16-155.
  14. ↑ Mawhinney and Back (2002), p. 4-328.
  15. ↑ Mawhinney and Back (2002), p. 4-329.
  16. ↑ 16.0 16.1 Budnick et al. (2008), p. 3-163.
  17. ↑ 17.0 17.1 Budnick et al. (2008), p. 3-164.
  18. ↑ Budnick et al. (2008), p. 3-165.
  19. ↑ 19.0 19.1 19.2 19.3 McAllister et al. (2008), p. 369.
  20. ↑ 20.0 20.1 Iqbal and Salley (2004), pp. 3-27–3-28.
  21. ↑ Nowlen et al. (2005b), p. 6-3.
  22. ↑ Nowlen et al. (2005b), pp. 6-15–6-16.
  23. ↑ Nowlen et al. (2005b), p. 6-8.
  24. ↑ McAllister et al. (2008), p. 49.

References

  • Biggins, J. B. (2008). Stationary combustion engines and fuel cells. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 8-105–8-111). National Fire Protection Association.
  • Budnick, E. K., Hunt, S. P., & Wright, M. T. (2008). Closed form enclosure fire calculations. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 3-145–3-166). National Fire Protection Association.
  • Chisholm, D., Sr. (2008). Emergency and standby power supplies. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 10-3–10-8). National Fire Protection Association.
  • Iqbal, N., & Salley, M. H. (2004). Fire Dynamics Tools (FDTs): Quantitative fire hazard analysis methods for the U.S. Nuclear Regulatory Commission Fire Protection Inspection Program (NUREG-1805). U.S. Nuclear Regulatory Commission, Office of Nuclear Reactor Regulation.
  • Mawhinney, J. R. (2008). Water mist fire suppression systems. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 16-135–16-180). National Fire Protection Association.
  • Mawhinney, J. R., & Back, G. G., III. (2002). Water mist fire suppression systems. In P. J. DiNenno, D. Drysdale, C. L. Beyler, W. D. Walton, R. L. P. Custer, J. R. Hall Jr., & J. M. Watts Jr. (Eds.), SFPE handbook of fire protection engineering (3rd ed., pp. 4-311–4-337). National Fire Protection Association; Society of Fire Protection Engineers.
  • McAllister, T. P., Gann, R. G., Averill, J. D., Gross, J. L., Grosshandler, W. L., Lawson, J. R., McGrattan, K. B., Pitts, W. M., Prasad, K. R., Sadek, F. H., & Nelson, H. E. (2008). Structural fire response and probable collapse sequence of World Trade Center Building 7 (NIST NCSTAR 1-9, Vols. 1–2). National Institute of Standards and Technology.
  • Nowlen, S. P., Najafi, B., Wyant, F. J., Forester, J., Kazarians, M., Kolaczkowski, A., Joglar, F., Funk, D., Hannaman, G., & Anoba, R. (2005b). EPRI/NRC-RES fire PRA methodology for nuclear power facilities: Vol. 2. Detailed methodology (NUREG/CR-6850; EPRI 1011989). Electric Power Research Institute; U.S. Nuclear Regulatory Commission, Office of Nuclear Regulatory Research.
  • Sheridan, D. J. (2008). Electric generating plants. In A. E. Cote, C. C. Grant, J. R. Hall Jr., & R. E. Solomon (Eds.), Fire protection handbook (20th ed., pp. 9-179–9-189). National Fire Protection Association.