Failure of high pressure fitting

Whilst no damage was sustained other than to the fitting and the whip check and no injury was sustained, the incident is believed to have had the potential for more serious consequences and was investigated as a high potential event.

A dive vessel had been heading for port after completing a phase of work. Both diving bells had been surfaced to allow access for maintenance work. The gas from the bells had been reclaimed into storage in the gas hold of the vessel. The dirty reclaim tubes were full after this exercise, so it was decided to clean this gas for use later as a dive mix.

The on-shift life support supervisor (LSS) entered the gas hold to set up the gas supplies for cleaning, which included putting the dirty gas tubes on-line to the cleaner at a supply pressure of 200 bar gauge. Having opened the supply to the cleaner, he was walking towards it when he heard a loud bang and a continuous rush of gas. On turning round, he could see that the supply hose from the dirty gas had blown off and gas was venting into the gas hold from the tubes. He closed off the dirty reclaim tubes and isolated the escaping gas.

On examination, it was noted that the Comex spin fitting that connected the gas whip to the supply manifold had blown off. The whip was undamaged. There was clear evidence of the track the female connection had made across the male connection as it parted. On examination of the fitting to ascertain the cause of the failure, it was noted that the gas whip fitting had blown off the supply connection.

Since there was no clear way of establishing if the other spin-fit fittings could be subjected to the same failure and since new fittings of the same type were not immediately available, the only safe course of action was to temporarily replace all the Comex spin-fit connections on the gas supply lines with JIC fittings. During the shift handover meetings immediately after the incident, all personnel were reminded to check that HP fittings and whip check cords were in good condition and correctly aligned.

It was concluded that the failure was likely caused by worn threads and daily use exerted on a 20-year old fitting that had been subjected to stretch as a normal part of daily use. It is believed the whip check cord failed due to the kinetic energy generated as the fitting blew, coupled with the fact that the whip check was not fully extended against the whip and had some slack in it. Corrosion may also have been a contributory cause of the failure.

The root causes were identified as being that:

• There had been no planned maintenance/inspection ‘routine’ for checking wear on the Comex spin-fit fittings.
  
  
• There had been uncertainty regarding how best to position whip check cords on the gas manifolds and how they should be rated to ensure they could cope with the energy that may be exerted upon them.