Repeated Boiler Flame Failure After a Fuel Changeover
An 18-year-old product tanker has a vertical cylindrical auxiliary boiler rated at 4.5 t/h, burning HFO 380 cSt at 15 bar to supply fuel heating and accommodation services. On departure from a port where the ship had been burning MGO for two days, the second engineer carries out the changeover to HFO on the boiler burner. Within two hours the boiler trips on flame failure three times. At the burner inlet the viscosity controller reads 132 °C, the fuel pressure swings between 13 and 16 bar, and the atomising steam pressure is 6.5 bar against a normal 9 bar. The flame scanner window is discoloured. The fuel heater is steam heated and the ship's steam demand is high because cargo heating has just started. The third trip happens with the boiler on UMS; the duty engineer finds fuel vapour in the furnace and discovers that the purge timer had been shorted out during the previous restart.
1. What are the probable causes of the repeated flame failures, and in what order would you investigate them?
Read the worked answer
The pattern points to a fuel atomisation problem compounded by a control failure, not to an electrical fault in the burner management system.
- Fuel temperature and viscosity (primary). 380 cSt fuel needs roughly 140 °C at the burner to reach about 15 cSt, the viscosity at which a pressure-jet atomiser produces a stable spray. At 132 °C the viscosity is appreciably higher, the spray is coarse, ignition is delayed and the flame is long and unstable - exactly the condition in which a flame scanner loses sight of the flame. The shortfall is credible because cargo heating has just started and is taking steam from the fuel heater.
- Atomising steam pressure (primary). 6.5 bar instead of 9 bar gives poor break-up of the fuel into droplets. Check the reducing valve, the line strainer and whether cargo heating is drawing down the auxiliary steam pressure.
- Fuel pressure swing 13–16 bar. Look for a choked burner filter, a worn fuel pump, or air in the line after the changeover. A swinging pressure means a swinging spray.
- Changeover procedure. Changing from MGO to HFO quickly, or before the heater has brought the fuel up to temperature, sends cold heavy fuel to the burner and is a classic cause of this exact sequence.
- Flame scanner. The discoloured window attenuates the signal and can cause a spurious trip even with a good flame, but cleaning it alone would not explain the pressure and steam anomalies, so treat it as a secondary contributor.
Investigate in that order: temperature and viscosity first, then atomising steam, then fuel pressure, then the changeover record, then the scanner.
2. What immediate actions restore steam safely, and what must you refuse to do?
Read the worked answer
Refuse absolutely: do not short the purge timer, do not bypass or jumper the flame failure device, and do not attempt to relight a furnace that may hold unburnt fuel. The furnace found full of vapour is proof that the earlier purges were inadequate, and a furnace explosion is a far greater danger than the loss of steam.
Sequence:
- Treat the boiler as out of service, inform the bridge and the cargo watch, and warn that cargo heating will be lost or reduced so the cargo owner can be told.
- Trip the fuel, shut the burner manual valve, and keep the forced draught fan running to purge. The purge must run for its full designed period - typically 5 minutes at the specified air flow, several furnace volumes - before any relight.
- Restore the fuel temperature to about 140 °C and the atomising steam to 9 bar before relighting. If either cannot be achieved quickly, run the boiler on MGO from a separate service tank: the ship has it on board and it removes the viscosity problem immediately.
- Clean or renew the flame scanner window and prove the scanner responds to a test flame from the correct sighting angle.
- Check the burner nozzle for carbon, renew the fuel filter element, and confirm the fuel pressure is steady.
- Light off under local supervision after the full pre-purge, with the boiler on manual control until the flame is proven stable for at least 30 minutes.
- If cargo heating is critical, reduce the heating rate or bring a second boiler on line rather than pushing a marginal boiler.
3. Apart from the burner itself, what procedural and planned maintenance failures does this incident reveal, and what would you do about them?
Read the worked answer
The equipment faults are the smaller part of the finding. The procedure and the safety culture failed:
- There is no written, risk-assessed changeover procedure for the boiler, and the changeover was carried out in the middle of a new cargo-heating demand instead of before it.
- The purge timer was shorted during an earlier restart. That is a deliberate defeat of a safety interlock, a reportable safety violation, an SMS non-conformity and a likely port state detention item. It must be recorded and dealt with under the company's procedures, not quietly corrected, and the person responsible must understand why it is treated as a serious matter.
- Flame scanner cleaning and testing are not on the PMS with a defined interval and acceptance check.
- Low atomising steam pressure and low fuel temperature are neither alarmed nor trended; the viscosity controller may be out of calibration and should be checked against a sample.
- The boiler was left on UMS with a known unstable flame, and the alarm was not attended promptly.
- No burner nozzle or spare parts record exists, so the condition of the nozzle was unknown.
Actions: raise a non-conformity in the SMS, brief every engineer on the changeover and purge requirements, add the burner assembly, scanner and steam reducing valve to the planned maintenance routine with defined intervals, and review the UMS alarm response during cargo operations.