1. Don’t overstep your mark. Every pilot knows his own ability and for a test flight to go well it’s important not to overstep this unless you’re absolutely sure that you can happily achieve the goal. It’s good to have a healthy level of caution, and perhaps a little apprehension, as this shows respect for the fact that test flying is not always straightforward. It’s important, however, that a modeller whose experience is entrenched in the four-channel modelling scene does not jump into test flying four-engine scale models or high speed jets without, at the very least, doing some homework or consulting other pilots.
2. Set throws, C of G and engine. Apart from an obvious structural, mechanical or electrical failure, control throws, engine settings or an incorrect centre of gravity are the main areas of variance that can see a test flight end in failure. The test pilot should take an active role in ensuring that these variables have been taken into account and properly set. When checking control throws, don’t overlook the advantage of having a switched low rate setting in case excessive control movement proves to be a factor. Models built from kits or plans will most likely have C of G and basic control throw information provided in the associated documentation and, clearly, you should not fly your model without checking it. The truth of the matter is that, of all the possible pitfalls, it’s a wildly incorrect C of G that is most likely to bring about a model’s early demise. Extra care should, therefore, be taken in the calculation of an unknown balance point. Having done this a test pilot’s reliance on an engine or expected motor run time is essential. Landing with power available during a test flight is always more sensible than having to cope with an emergency dead-stick.
3. Know the transmitter. In these days of electronic wizardry we are fortunate to have a high level of computer power available in our hands. With the ability to assign switches to particular functions it is essential that the test pilot knows his way around the transmitter. There is, after all, no point in flicking a toggle to put the wheels away, only to find that you’ve hit the engine kill switch! Most pilots will employ some transmitter commonality between models, however if you’re asked to test fly a model for someone, you can expect the Tx configuration to be quite different from your own. A responsible test pilot will discover what all the bells and whistles do while the model is safely on the floor. A range test should also be undertaken on any new installation and fail-safes should be proven before flight.
4. Have a plan and plan to use it. It’s entirely possible that a test flight may pass completely without incident, however it’s always best to have an action plan in mind before the flight than to hurriedly think of one while the model is in the air. If it’s your intent to get the model airborne, quickly trim, asses control throws and land after a few circuits, then stick to that unless your pre-flight expectations turn out to be completely incorrect and you find the model a dream to fly. Since the BMFA safety codes ask us to consider if the flight can safely be made, and point us toward considering eventualities and emergencies, do give this some thought before you commit yourself and the aeroplane. Perhaps the most obvious thing to consider here is the good old weather. There really is such a thing as a perfect day for a test flight and, conversely, there’s also such a thing as a bad one!
5. One step at a time. Some models require baby steps. Optimising a model set up in one test flight is rare. It’s sometimes necessary to test fly and adjust, then test fly again and again. You should just accept this and expect to take a few sessions to get the best out of a model. In some cases this is mandatory. Large models, for instance, have a strict flight testing regime that must be witnessed over and over by inspectors and others before the model is given a general clearance to fly. Baby steps might encompass stages such as assessing ground handling or trying some fast taxi runs before committing aviation.
6. Take note. Whether this be mentally or literally depends on the model, the duration of the test flight and what it is you hope to achieve from it. It may be that tweaks are easily remembered following the test flight of, say, a high wing trainer. However, more complex models may require many more changes. Multi-function gliders, for example, have complex electronic mixer arrays which are often reliant on the settings of another. Accordingly, a small increase in one may set you back on another. I have undertaken glider test flights of over an hour where in-flight programming and repeated testing has been required. In these instances, it’s very handy to have a mate at the slope edge jotting down notes of mixer percentages and travels.
7. Programming in flight. Not for everyone this, but very useful if you do implicitly know the workings of your transmitter and are confident enough to remove your gaze from the flying model. More often seen in slope soaring where the flight times are prolonged and the mixes are complex and far easier to achieve if you have launched / taken off with the model memory ‘open’ in a particular place. Elevator compensation under flap deployment is a very common mixer to set in flight, as is aileron differential.
8. Find the balance. During any test flight or test flying programme, the goal is to achieve an inherently safe and aerodynamically pure model with well-balanced control authority that is suited to the main user of the aircraft. As the foibles of the model become apparent the test pilot will take steps to address them with trims, mixers or other adjustments. Thrust lines, the centre of gravity, and most aerodynamic mix compensations are going to remain the same once set, but any RC model pilot will tell you that it is entirely possible to develop a ‘feel’ for the model back on the transmitter sticks. This will vary between individuals and the test pilot needs to consider how the model is to be flown when he is aspiring to optimise its set up.
9. To stall or not to stall? A particular bone of contention during test flights is the tendency for some pilots to assess the stall soon after take-off, come what may. There is no issue here if the model flies sweetly straight ‘off the board’, however I’ve seen people force a stall onto a model that’s already struggling with a bad C of G, excessive control throws or that’s reached the limit of the flight trims yet still needs more! The stall assessment is often best left until a later proving flight when all the other bugs have been eliminated or tamed to an acceptable level.
10. Unwritten rules. Like it or not, our society is becoming more and more litigious, indeed I’m already aware of action having been brought against test pilots following the demise of another person’s expensive model. As a result of this, it should be understood by all parties that a test pilot asked to undertake a test flight on behalf of another will do everything within his ability to ensure that it goes well. This highlights the need for said test pilot to ask pertinent questions before he undertakes a flight and to ensure that he completes his very own list of checks. However, it also follows that the test pilot will not take unnecessary risks with another person’s model by pushing it hard, flying recklessly, or showing off.
