Accident Report

Synopses (SACAA 1264)

On Friday 5 April 2019, at 0509Z, a Dassault Aviation Falcon 900B, with registration markings ZS-DFJ, departed Robert Gabriel Mugabe International Airport (FVRG) in Harare, Zimbabwe, on an unscheduled charter flight to Victoria Falls International Airport (FVFA) in Zimbabwe. On-board the aircraft were two crew members and eight passengers.

The aircraft reached a cruising altitude of 38 000 feet (FL380) en-route to FVFA. At approximately 60 nautical miles (nm) from FVFA, the crew commenced with the descent. Whilst passing through FL250, the pilot-in-command, who was seated in the left seat, reported that his Electronic Horizontal Situation Indicator (EHSI), Electronic Attitude Direction Indicator (EADI) and Multi-Function Display (MFD) started to blink and then went blank. This was followed by various warning lights on the ‘master caution’ panel. Shortly after this, the First Officer (FO) noticed smoke entering the flight deck through the roof lining above the right-hand flight deck window. The flight crew immediately donned their oxygen masks and carried out the necessary checklists from the aircraft flight manual (AFM) regarding electrical smoke or fire.The FO declared an emergency and broadcast a Mayday call to FVFA tower frequency. The aircraft was given priority and cleared for a straight-in approach for Runway 30. The crew expedited the descent and whilst passing FL190, the smoke began to dissipate. When passing through FL140, the crew removed their oxygen masks and prepared the aircraft for the approach. The FVFA Aircraft Rescue and Firefighting (ARFF) services were ready for the aircraft and positioned themselves next to the runway.

The aircraft landed safely at 0602Z and taxied to the apron where the occupants disembarked normally. The aircraft sustained damage to a wiring loom that had burnt in the ceiling of the flight deck, as well as excessive heat damage to a ceiling panel above the wire loom. The operator had done a modification by installing an automatic dependent surveillance-broadcast (ADS-B) system on 30 August 2018 and that required an additional wire to be added to the loom rooting to the circuit breaker panel. The last maintenance conducted in this area of the burnt loom was during the last C check on 8 November 2018 which was a visual inspection. None of the occupants sustained any injuries during the incident sequence.

The investigation revealed that during the modification which required cable be threaded on to the wiring loom secured by a clamp, it is likely that when the wire was threaded through the clamp, the wire insulation was inadvertently damaged due to the clamp not being opened and the open damaged wire started to arc on the clamp causing high temperature which resulted on the smoke in the flight deck, multiple system failure.

Probable cause (SACAA)

During the modification which required the cable be threaded on to the wiring loom secured by a clamp, it is likely that when the wire was threaded through the clamp, the wire insulation was inadvertently damaged due to the clamp not being opened and the damaged wire started to arc on the clamp causing high temperature which resulted on the smoke in the flight deck, multiple system failure.

Contributory Factors (SACAA)

Wiring clamp not opened during the threading of the wire on to the wiring loom.

Comments by Charlie Marais

If there is one thing of which I am truly afraid, it would be fire in the cockpit, or for that matter anywhere on the aircraft, internal or external. The procedures to follow were hammered into me from day one, but as human nature tends to be, it sounded serious and sufferable, but it had no true meaning. What I mean is that through our make-up, or design, we learn to live with some fears neatly packed away in order to forge our way towards happiness. I, like many before me, have contemplated man’s ability to seek and claim happiness, even when the cloud has a sliver of a silver lining.

Sorry, I definitely sound like either a complete softy or probably slightly mad. Both, if you ask me, and scarcely reconcilable with a chopper pilot. I do fly multi-engined fixed wing aircraft, but flying airline types was never a privilege. This then begs the question as to why I am writing about an environment about which I do not know anything, but being a passenger.

As passengers we also have the same fears and being knowledgeable to the seriousness of fire on board, adds a little to compartmentalising scary stuff. On the other hand, I know and trust the crew in the front of the aircraft more than most others and resign easily to the fact that my life is in good hands, and if they screw it up, well, then tough stuff. So, when the crew in this case smelled and saw smoke in the cockpit, not big enough for a flight deck, they were not only facing some attention-grabbing light to warn them, it was literally in their faces. Smoke and fire are synonymous. This is where we do the first thing and for those who have never been there, it has nothing to do with brain functioning. At least not in any form or shape consciously. Once the automatic body functioning has run its course, normally extremely short and mostly void of physical matter, but smelly, the training kicks in. Doing the procedure is what is called for and doing the procedure was what they did. Very precise, very professional and definitely commendable from anyone’s point of view. I know that all pilots would now join in and say, ‘well, they did what they had to and which they were trained to do.’ You are correct and this is the message that I would like not just every pilot, but all pilots, of all shapes and wing forms, to hear and remember when the smoke hits the cockpit climate control, or fan. The crew then had to make a visual approach, without aids and that could probably have been problematic. I mean, when I follow the GPS, I as a rule do not pay particular attention to the precise position I am at. However, this crew knew their way around and I would bet that the passengers did not even know that something was smouldering. So, ten out of ten and more for this crew.

Then there is the small problem of why and how this particular situation was birthed in the first place. This is where I would like to digress from the report and give a few pointers to us who work with wires, looms and fiddle with electrics when we probably should not. Hold on, I am not referring to the qualified people who did this job of pulling another wire through the P-clamp, but to all of us who think, especially in the non-type certified class, what not to do and what to look out for, is for the birds. Maybe not for the birds, but there are a few catches in fiddling with wiring on an aircraft.

To get to what we could learn from this, I unfortunately must point out a few things we did not learn from the investigation. Just a few basics first. When we make conclusions, we must base them either on definite facts, or possible events, that could lead to a most probable outcome. Therefore, the report defined the cause as probable, but not definite. I imagine that due to the sheer volume in which the investigation could be referenced, that brevity was chosen to get to the heart of the issue. For us there are a few considerations which we must take into consideration when we look at our own aircraft, or when our less experienced craftsmen perform this duty. The following is what we can glean from the investigation report; that there was work done 151,3 flight hours before the incident, that a wire was added to the loom and the P-clamp was not loosened to thread this wire, that some arcing took place in the loom clamp vicinity and that it led to multiple circuit breakers to fail, associated systems failed due to circuit breaker activation, with no more voltage going through the wires the arcing stopped and the smoke dissipated. The finding was that the wire was damaged during fitment and that arcing took place as a result. What we do not know, is how the damage was caused and then how the deterioration to the final moment of arcing took place. Granted, we may never know the exact causes and sequences, but there are a few lessons to learn. However, first, let me suggest some crucial questions. What method did the artisan use to feed the extra wire through the clamp to become part of the loom? If the clamp was not very tight, there may have been some space to insert the wire without any trouble. We know the clamp was not loosened and therefore assume that there was enough space to add a wire. Normal practice would be to just push the wire through the clamp space, or to pull the wire through the clamp space with a massive ‘needle’, which is the same way a needle and thread is worked through material. This would force the opening and put the rest of the wires under more stress, but in many cases, this is actually possible without causing any undue disturbance of the wire loom. It is known that in the absence of this needle, some quick fitters would use a small screwdriver. Now the exponential possibility of insolation damage, not only to the wire, but also to the insulation material on the P-clamp, must be considered. Be reminded that arcing can only take place if two wires have no insolation between them, or a wire and a clamp where there is no insolation. How the job was done is not covered and should this question have been posed, not revealed by the party doing the job, as it would have been contrary procedures and many other factors and as such self-incriminating. The next possibility would have been that the wires had been exposed to some heat and that the insolation became brittle. When one then works in this region, the disturbance could have been probable in mostly all cases, especially if the loom was as tight as it should have been prior to fitting the extra wire. Which circuit breaker tripped first? We still have not asked about the vibrations in this area and once we combine some of the possibilities of 151,3 flying hours, is what it could take for a slip, known or not known, to manifest. Information is crucial and should be pursued to the nth degree. So many questions and so much destroyed by the localised arc-fire. There may or may not be hidden issues to add, but to not actually know the very source of the manifested effect, is to try and fly visual in pitch darkness. Note that testing and visual inspections were made after the job and no problems were found. This would mean that the incubation of the cause to the effect, was in an undetectable stage. Even a load value check confirmed no anomaly.

With there being so many questions, it does not mean that we cannot learn from this incident. If extra awareness is all you can learn from this incident, it is much better than nothing. Yes, I know that we always want to see the bigger picture, but when it comes to causality, only a magnifying glass can reveal the origin.

Recommendations

The first recommendation is that all pilots must take note that their best chance of surviving an abnormal event, is to follow the manufacturer’s published sequence of isolation of the cause to the effect and to just follow the SOP. In the same breath I would recommend that even when we use electronic navigation as an only resort, we have some idea where we are in order to point the nose in the correct direction. To those who work on aircraft electrical systems, especially home built and home serviced, to understand that a small mistake or oversight, can take many hours to manifest into something really dangerous. Use only acknowledged guidelines and think twice when working with wires. They tend to cause smoke in the fan. Many inspections are visual and technicians are trained to look for very specific tell-tale signs of trouble brewing. If you work on your own aircraft, please do sufficient research and do not hesitate to make use of professionals whose day job it is to work on electrical components. Many times, it is possible to find the most probable place where trouble started. The question should then follow of how these sequences of events were enabled at the very start. I understand that in turbine technology fracturing can be traced down to grain sized errors during the casting operation, but I am sure we do not have to go that far back. Wires must be the correct load bearing capacity and not old or an offcut. Internal damage to a wire is very easy when the wire is bent past a certain angle and even that which has been specified in guidance material. The rules are simple; know the rules, apply the rules according to accredited and acknowledged methods and never ever use a wrong tool or a short cut.

A Hazard is normally dormant and will only become activated when prompted to do so by other factors which may or may not be exposed. This is why HAZARD exposure in all phases of aviation is of such importance. Prevention is cheaper, less scary and more life preserving than cure, but prevention can only be applied when a Hazard has been identified and brainstormed for possible nasty effects.

Happy Hazard Hunting.