Safety

INQUEST TOUCHING THE DEATH OF ADAM KIRBY

On the 2 November 1999, the Coroner’s Inquest into the tragic death of Adam Kirby, one of our junior members, re-opened following an investigation by the Metropolitan Police and the British Model Flying Association, assisted by the Air Accident Investigation Branch and the Royal Air Force. The aims of the Court were to establish the facts of the case and to give its verdict. The Coroner explained he was not permitted in law to allocate blame nor give any recommendations. However, he went on to explain that he had the right to contact authorities once the Inquest was closed if he felt this would help prevent reoccurrence of an accident.

Witnesses were called from the Phoenix Model Flying Club, the Metropolitan Police, the Air Accident Investigation Branch and the British Model Flying Association. The Court established that Adam was killed by a Loaded Dice model aircraft that was out of control. The Jury returned a verdict of accidental death.

Any proceedings such as this also seek to establish the probable cause or causes of such an accident using witness’s statements, specialist reports and evidence given. Whilst it would be inappropriate to go into detail, the most likely but not absolute cause of the accident was the simultaneous operation of two transmitters on the frequency of the accident aircraft. Although the absolute cause was unknown, it was positively established the aircraft which was fitted with PCM radio entered "Failsafe - Hold" mode shortly after take off.

It then flew for approximately 20 seconds during which time it climbed to approximately 100 feet in a gentle left hand turn before departing from relatively stable flight in a diving 270 degree turn and impacting into pits. Despite shouted warnings by the pilot and several others, these were not heard by a small group of members in sufficient time to take avoiding action, including Adam. This was due to the warnings being masked by an engine being run in their immediate vicinity.

The impact resulted in Adam’s instant death.

Notwithstanding, it was explained to the Court that extraneous interference could not be absolutely ruled out although evidence, investigations carried out by the Metropolitan Police and an interference survey by a Royal Air Force team indicated this was unlikely. In addition, equipment failure could not be entirely eliminated as an intermittently faulty receiver crystal was identified. However, the evidence and tests by the Air Accident Investigation Branch indicated that this cause was highly unlikely and that the crystal had probably been damaged during the impact.

Tests by the Air Accident Investigation Branch also confirmed that if a transmitter’s antenna is retracted it has a significantly reduced effect over another transmitter on the same frequency provided the second transmitter’s antenna is extended.

Following the evidence, the Coroner asked if the British Model Flying Association planned to instigate actions to try and prevent reoccurrence of a similar accident. We undertook to review our recommendations on the following to determine if we could improve our already comprehensive safety code:

• The siting of pits in relation to the take off area.
• The practice of starting and running engines in the pits.
• Transmitter control when more than one transmitter is on the same frequency.
• The use of "Failsafe" in the "Hold" mode.
• A transmitter antenna retracted philosophy whilst not flying the aircraft.
• Improvements in warning systems should an aircraft go out of control.

Any possible improvements we are able to recommend will be disseminated to all members and clubs through the BMFA News, Club Bulletins and Handbook Addendums. We also hope to enlist the help of the model trade, the model media and CIAM to inform non-BMFA flyers and clubs.

What can you and your Club do to help the BMFA and your colleagues?

It is essential that all clubs and members follow their legal obligations under Air Navigation Order 56 in that "A person shall not recklessly or negligently cause or permit an aircraft to endanger persons or property". Guidance is given in Civil Air Procedure 658, Small (Model) Aircraft - A Guide to Safe Flying and is referred to throughout the Handbook whenever it adds to the BMFA safety code.

Therefore, it is essential for you and your Club to fly in accordance with the BMFA Safety Code in order to protect yourselves and others. If not, you may be expected to justify your decision should it be tested in law. There can be no excuse for any transgressions no matter how minor as they can all accumulate with terrifying consequences, as we have recently witnessed.

Every member is a safety officer in their own right. If you witness a transgression then do something about it and politely inform the transgressor of the error. In many cases, it is simply a lack of knowledge. It is not acceptable to turn away and disown your responsibilities and leave it for others to sort out. However, if needs be, have the courage to inform your committee so they may undertake further action if they see fit.

Ensure you have a strong committee that enforces Club Rules and is capable of taking punitive action if members continue to act irresponsibly. Ensure that newcomers to your Club are given safety training and are correctly supervised. Encourage the achievement and instructor schemes. Ensure all members are regularly updated on safety issues and they are not out of date.

There is bound to be the inevitable outcry from clubs and members who see such action as "taking the fun out of it" or "interfering with liberties". We have one answer; "Where is the fun or liberty for those touched by Adam’s death?"

PCM / FAIL-SAFES

Report from the BMFA Technical Council - 30th October 1999
Compiled by Andy Ellison. R/C Power Technical Committee.

On Sunday 14th March 1999 eleven year old Adam Kirby was struck by a radio controlled powered model aircraft and killed. The coroners inquest into the accident has concluded that a contributory cause in this tragic accident was the settings of the computerized transmitter based programmable failsafe.

In this instance the radio in use was transmitting in pulse code modulation mode (PCM) and the programmable failsafe was set to HOLD.

On Saturday 30th October, after a long study and information gathering exercise, restricted by the legalities surrounding a case of this type, the BMFA Technical Council sat to deliberate proposals put forward by myself relating to the use of PCM radio and its association with programmable radio failsafe devices. This was in order that the problems associated with its usage could be clarified and passed on to R/C fliers via this publication, the BMFA members handbook and by the instigation of competition rule changes. The following report is a synopsis of my findings.

Please bear in mind that the terminology relating to this topic varies greatly from one radio manufacturer to another, and so for the purpose of this safety notice I have generalised. The term HOLD refers to a scenario whereby the servos HOLD the position they were in just before the interference arrived. The term PRESET refers to the moving of the servos into pre-programmed positions which were defined by the operator before flying, and FAILSAFE is a generic term to cover both.

---------------------------------------------------------

Computerized radios are an ever increasing feature of the modern R/C flying club. Regrettably it is also the case that many of the users of such devices do not fully understand their operation nor indeed the content and facilities of the operational software contained within. This is particularly the case when PCM mode is selected with respect to the PRESET/HOLD options.

In my research I found that a disturbing number of average club fliers using these radios (around 80%) were unaware that selection of PCM mode brings with it the failsafe features of the program. Basically if you select PCM you automatically have a failsafe set to prevent the model from flying away.

The program has no respect for the weight of the model (fail-safes are mandatory on models over 7Kg dry weight) and modellers transmitting on PCM, when questioned on the settings of their failsafe, frequently answered that they did not have a failsafe on their model as it was under the 7Kg limit.

The unfortunate fact of the matter is that they did indeed have a failsafe operating on their model but did not realise it. Moreover, because they did not realise it, they had not programmed it, and in the event of signal loss, it would defer to the factory default settings, usually HOLD.

This of course includes the throttle servo. If this interference occurs at take-off for example and the throttle holds at an opened position with minimal deflection on the other control surfaces, the results can be disastrous.

If you use a typical Computerised Transmitter, or have members in your club who do, you should be aware of the following basic findings.

Almost all sets give an automatic PRESET/HOLD feature within the program as soon as PCM is selected. This failsafe becomes operational if interference occurs at the receiver of the model or the transmitter signal is lost. Control returns to the pilot when the interference ceases. If interference does not cease, the pilot will not regain control.

The failsafe software cannot be totally inhibited. It will either hold all (HOLD) or go to positions pre-programmed by the pilot (PRESET). Some sets allow a combination of this feature. i.e. some channels will go to a PRESET position and some controls will HOLD at their last position.

The default setting if nothing is programmed is to HOLD at last input position, including throttle. This will become operational after a pre-determined interval set by the pilot ( 1.0 seconds, 0.5 seconds or 0.25 seconds) or after a default interval if nothing is programmed. The interval is the length of time it takes, starting the moment interference occurs, until the servos assume their pre-set positions.

The failsafe does not work in the event of receiver ni-cad loss.

Programming of the PRESET feature is done by opening the memory of the computer, putting the sticks, switches and potentiometers in a selected position and then closing the computer memory. This can be quite difficult to follow even from the radio instruction manual.

If a model normally flown on normal modulation (PPM) is swapped to PCM by a change of receiver, the failsafe feature is again automatically selected even though none of the other settings in the transmitter are altered. There is no warning given to the pilot to remind him that he is now flying under the protection of PRESET/HOLD software.

---------------------------------------------------------

To illustrate the seriousness with which the BMFA are taking this matter, the Technical Council unanimously agreed a plan of action which has been reported to the coroners inquest and includes :

1.The forwarding of our findings to all known radio manufacturers highlighting the problem and requesting/advising that they :-

1. Negate the problem through the use of revised software i.e. inhibit the failsafe function so that it must be selected to turn it on, and modify the factory setting for the throttle function so that it does not default to "HOLD";  and

2. Publish in the operations manual for the Radio, clear and concise instructions highlighting the potential danger of an "un-programmed" Transmitter based, radio failsafe device.

2. The findings are passed to the CIAM/FAI with the expressed request that they distribute them to all National Aeromodelling Governing Bodies in a suitable publication.

3. The CIAM/FAI/BMFA instigate an emergency general rule change to ensure that any powered models flown under the protection of a Radio Control Failsafe device must ensure that said device brings the motor/engine to run at its lowest speed, (stopped in the case of electric powered models), and specifically not to HOLD the last position of the motor/engine control.

There are other factors which contributed to the sad death of Adam Kirby and these are covered elsewhere. The purpose of this safety notice is one of education and promotion of awareness.

It is the strong recommendation of the BMFA Technical Council that clubs and individual fliers alike take on board the points made above and assess them against their own operating system and philosophy as responsible users of this equipment and as model flyers to ensure that such an occurrence is does not happen again.

We recommend that particular care is taken by all individuals when operating on PCM or when swapping from PPM to PCM to ensure that the PRESET/HOLD feature is correctly programmed with throttle to tick-over (stopped in the case of electric power).

We also strongly recommend that all clubs incorporate a regime into their flying field rules that will ensure that all operational fail-safes in use on powered models (including "add on" devices used with PPM) must set the throttle to tick-over (stopped in the case of electric power) regardless of the other control operations governed by the failsafe and regardless of the size of the model.