Terror Free Takeoffs

In a one year sample of accidents in 1984 examined by the Bureau of Air Safety Investigation, about 7% of the total accidents in Australian registered fix wing and rotary wing aircraft occurred on takeoff. It is interesting, though, that the aircraft that do the most takeoffs- light training aircraft- have the third lowest number of accidents on takeoff and initial climb. For example, of 32 aeroplanes and helicopters in the sample that were damaged and destroyed in takeoff accidents, O were in Airliner/Commuter, 2 were Charter, 4 were Training, 6 in other air work, 13 Private/ Business flown and, due to the riskier operations, 9 were agricultural. Though in 1984 32 people were unfortunately killed in all aircraft accidents, only a small percentage of the total aircraft takeoff accidents were fatal which proves that the low speed accidents are more survivable- provided the aircraft is in control when contact with ground or obstructions is made.

There was a mechanical failure of the engine in about one tenth of takeoff or initial-climb accidents. In most cases this was charged to a lousy pre-flight inspection or inadequate maintenance. Still the risk of mechanical failure is relatively mild. As it is, the pilot remains the prime culprit.

However, the possibility of mechanical failure still exists. How can we minimise the risk caused by the engine quitting on takeoff?

A start would be to avoid intersection departures. If the crankshaft or something as important chose to break, you’d feel pretty dumb having started the run with a thousand metres behind your tail. Another good procedure which may save your life one day is, that if you have multiple runways, to take the most favorable one from the standpoint of terrain or built-up areas, weather conditions permitting.

A final procedure is to maximise your options in the early stages of the climb. On a VFR climb in a single, use the best rate-of-climb speed and maximum continuous power until you’re high enough to see some good places to go. In a retractable, consider using the transport category gear retraction method: positive rate of climb, gear up. This maximises climb. While some delay gear retraction until there is no longer enough runway left to land on, a little more height is worth the risk of a gear-up. The blue line and maximum continuous power until reaching 1,000 feet is hard to beat in a twin.

What about returning to the departure runway if the engine in a single fails? Steep turn at low altitude and minimum airspeed are dangerous. Even if you practice these . at altitude you miss the sensation of the uprushing ground, the possibility of wind shear, and the fact that once you’ve made your practice turn the runway will not be straight ahead. In practice you would be ready, too, whereas it would be a surprise in real life. Just lowering the nose to the correct speed to maintain the best glide straight ahead would be challenging, but doing this while rolling into a steep bank might be overwhelming. You’d have to have plenty of both speed and altitude to attempt a successful return and land. It would require more than a 180 to line up on a runway, any landing would be downwind and, if you got the aircraft turned but couldn’t get back to the runway, an off-runway downwind forced landing could have far more serious consequences than one into wind.

The only time this manoeuvre is safe to carry out is in a sailplane or a high performance military aircraft – but not in a Piper, as one unfortunate pilot found out at Bankstown; at about several hundred feet on takeoff, his engine quit and immediately began to return to the reciprocal runway, he may have just made it back intact and alive if he had more height to clear a factory roof. However, if he had stayed straight and in control into trees with a headwind, the statistics would probably have looked a whole lot better. The slightest difference in airspeed or wind can make a big change. If there is only a 1 O knot surface wind, you’d be increasing your groundspeed by about 20 knots. This is a high percentage at average light aircraft speeds. One way of knowing how much extra runway is used by change in airspeed or wind is by trying this method:

There is a 56% increase in landing distance.

Past that mystical mechanical failure that we all give most thought to, what can we do to eliminate as much as possible from the other 90% of takeoff or initial climb risk?

For starters, using preflight and pretakeoff checklists would help eliminate a whopping one third of accidents. An oil cap left off caused an aviator to abort a takeoff, one Skyhawk pilot attempted to lift off with full flap and didn’t make it. The simple act of raising the flaps would have fixed everything. A Charter pilot forgot to remove the control lock, another left the trim in the fully up position- result: pranged aircraft and serious injuries to all. An unlatched seat did the same to another. Attempting a takeoff with fuel off, on a dry tank with copious amounts of water are common errors. I could go on. The message is simply this; check, check and double check!

If an aircraft hasn’t been flown for a while, or if it has been out in the rain, be especially wary.

Don’t squirt and glance, either. If you use a sampler and there is enough water in the tank to fill the sampler, that fuel/water line between it won’t be there. If you are in a situation where you don’t know for sure the sample is completely fuel or water, simply place a few drops of good old H20 into it. If it sinks to the bottom, it shows the sample as all fuel, or the drops will vanish indicating it’s entirely water. It’s the cheapest insurance policy water can buy! Make sure the aircraft isn’t misfuelled with Avtur as well. More than one bird has been brought down by that colourless fuel.

In a couple of accidents over the years, where the runway appeared long enough and there was nothing identifiably wrong with the engine, witnesses reported that the engine sounds were smooth but relatively quiet. The checklist tells us to put the prop control full forward for maximum rpm, but there are few pilots around who haven’t missed this a least once. Given the right circumstance, once is all it takes. Even if you use the list and miss the prop, sluggish acceleration should be cause for suspicion.

A checklist and suspicion can also catch an improperly latched door, which usually starts making sounds before liftoff. All certified aeroplanes fly fine with the door unlatched and it’s too bad that people wreck them in late aborted takeoffs and hairy low level attempts to return and land quickly just because the door comes open. Another problem that appeared in the statistics was smoke in the cabin. The pilot made an abbreviated, low level circuit and consequently lost control in a steep turn. Everybody wants to get back quickly if there is smoke but they still need to fly a safe circuit. Still, the best place to abort is in the takeoff roll, where any unusual sound, smell or feeling should blow the whistle.

Another big shot can be taken at the takeoff or initial climb accidents by following the regulations and pilot’s operation handbook Many of the takeoff and initial climb accidents occurred with some combination of overloading, high density and short runways.

A lot of people use short runways – they add usefulness to many operations but the risks have to be managed carefully. Often there is no way to use a particular airport and the full payload of an aircraft in conjunction. As pilot in command of the aircraft, it is our ultimate responsibility to make these allowances – and we can only do that if we start with an accurate weight, strip length, temperature, slope and wind velocity and if we use the power setting, configuration and technique recommended by the manufacturer.

Further, a trick used by old hands is to nominate a ‘HOWGOZIT’ point – a reference by which they can judge how it’s going. By observing the performance of your aircraft at different weights and temps against a feature say 200 metres along the strip, you will get to recognise the range of speeds that you should expect passing that point. If you’re appreciably below the usual speed, then abort the takeoff while there is still plenty of room to stop.

Keeping a plane’s weight in the limits is crucial. If you hear anyone say that the bird won’t even fit another suitcase in it, write that person off as a future statistic. If a pilot takes that unnecessary risk, consider how sloppy the rest of his approach to flying might be. Some really ask for it. Like the one flying a Twin Comanche that developed engine trouble. The fuel selector could not be moved and the pilot landed. (One reason a fuel selector might not move is frozen water in the system. This would also impede or block fuel flow). Later, on the ground, the selector moved and the pilot drained the tanks until the engine started and developed power. While taxiing back out for takeoff, one engine quit. The passenger wisely deplaned, and then the pilot attempted a one engine takeoff. It didn’t work!

The other types of initial climb accidents are related to pilot technique. Such as steep pull up after takeoff with a stall and spin to the ground. But most are much simpler. Students lose directional control in crosswinds and there are a number of minor accidents involving flight instructors and students on the ‘go’ part of the touch and go. The plane gets away from the student and the instructor doesn’t quite arrive on the controls in time.

The takeoff safety picture is pretty clear and the good news is that the factors leading to most accidents can be avoided. Maintain engines to a high level and always fly to increase your options as rapidly as is possible on the initial climb. Always follow preflight and pre-takeoff checklists, with special attention to fuel and flaps. Beware of high density altitude and heavy loadings, calculate the required runway distance. Watch crosswinds, especially when there are trees or buildings between you and the wind. And remember that if something does happen, flying the aeroplane under control to the scene of the accident is the way to minimise personal injury.