It’s the plumbing, guvnor

There is nothing new under the sun.

In the 1950s a Boeing B52 strategic bomber was lost during a long duration, high level sortie, and it was believed this was caused by fuel system icing. Of course it will never be proven because ice melts before you can check it out, but that was deemed the probable cause.

Following the January 2008 British Airways 777 accident on final approach to Heathrow when the engines failed to respond to demands for higher power after a long descent at low power, it has come as a relief that the UK Air Accident Investigation Branch has been able to issue a clarification of how icing can form in the plumbing of engine fuel systems, and how that can intermittently affect power delivery..

It’s a relief because - now the AAIB has been able to replicate the phenomenon in tests with Boeing and Rolls-Royce – we have more knowledge about the issue than we had before. What’s more we understand enough about it to be able make a recurrence much less likely by employing one of several straightforward procedures while we await system modification to make the accumulation of ice in fuel supply tubing a harmless phenomenon. Ice accumulation in the plumbing has probably always happened, and almost certainly in many different aeroplane engine types, but it has hardly ever caused a problem until recently.

The difference between existing fuel systems and the modified ones has to be that, when (not if) ice forms in them, it will never prevent the demanded power from being delivered.

Today’s long-haul aeroplanes are longer haul than they ever were before, so the cold-soaking of fuel has never been more thorough. At the same time engines, especially in the big twins, cruise at a smaller fraction of their potential power output than they ever did, with lower fuel flow rates lasting for longer. These are the basic combinations of circumstances that make modern aircraft more vulnerable than earlier ones to the accumulation of fuel/water ice slush in the plumbing.

The AAIB has made it quite clear in its latest interim statement that this is a generic problem that we have been able to ignore until now.

But although we have begun to understand its nature, we still don’t know what we don’t know. So we have to find out what we don’t know. The job has only just begun.