Loss of three gyros on a DP-3 drill ship

Three gyro compasses connected to DP system had a ‘heading freeze’ within 5 minutes of time.

What happened?

Three DP System connected gyros heading froze within 5 minutes of time. With all three gyros out of service, drill ship’s DP Model took over. 34 minutes later all gyros had been rebooted and were online in the DP System. Gyros were of modern fiber optic style.

The max excursion of vessel position was 20 meters over 34 minutes. Calm weather was benign enough to allow accurate model control of drilling vessel for duration of 34 minutes. Drill ship was connected to well head while tripping into hole when incident occurred but emergency disconnect was not required.

What went wrong?

The investigation is ongoing which will require analysis of all three gyros in a laboratory by original manufacturer. Nothing can be stated conclusively until root cause analysis and investigation is complete. Drill ship had been in service less than a year at the time of the incident.

What were the causes of the incident?

As of the writing of this safety flash, the proximate cause is unknown. Initial review reveals observations that may or may not have contributed to this incident, in whole or part and is not intended to reflect negatively upon any vendor or operators approach to construction/design:

  • Commonality of three identical MAKE & MODEL gyros with common software, firmware, and hardware;
  • Common backup 24 volt power source among three gyros could have allowed voltage instability to affect all three gyros at the same moment in time; . Out of date firmware existed on 3 gyros; . Automatic feed of GPS inputs existed; . Manufacturer’s technical bulletins existed but it remains unclear if that is relevant; . Simultaneous initial power-up start time of internal gyro clocks in shipyard possible; . Unknown age of gyros installed in shipyard.

What lessons were learnt?

  • An active, approved & well-rehearsed Well Specific Operating Guideline (WSOG) contributed to the safe reaction of on-board staff;
  • Calm weather contributed to the effectiveness of the DP Model;
  • Open and clear communications are crucial to successful response and resolution; . Common mode faults should be interrupted thru use of different brands, different power connections, and thoroughly tested during a Failure Mode Effect and Analysis (FMEA) conducted by an independent third party;
  • Concerns identified during FMEA testing should be followed up by management & risk assessed according to IMCA/industry

Actions

Deep root cause analysis is to be conducted; and a step change to installed equipment is ongoing.